Skip to main content

Full text of "A Biogeographic Assessment of the Samoan Archipelago"

See other formats


A BlOGEOGRAPHIC ASSESSMENT 
OF THE SAMOAN ARCHIPELAGO 




Matthew Kendall and Matthew Poti (Editors) 

Prepared by NOAA/NOS/NCCOS/CCMA Biogeography Branch 

with Support from NOAA's Office of National Marine Sanctuaries 

and Coral Reef Conservation Program 

NOAA Technical Memorandum NOS NCCOS 132 







"*»^r*'f d * 



Citation for the entire Document: 

Kendall, M.S. and M. Poti (eds.), 2011. A Biogeographic Assessment of the Samoan Archipelago. NOAA Tech- 
nical Memorandum NOS NCCOS 132. Silver Spring, MD. 229 pp. 

Example citation for an individual chapter (example of Chapter 3: Currents and Larval Connectivity): 

Kendall, M.S., M. Poti, T. Wynne, B. Kinlan, L. Bauer. 2011. Ocean Currents and Larval Transport Among Is- 
lands and Shallow Seamounts of the Samoan Archipelago and Adjacent Island Nations. In: Kendall, M.S. and 
M. Poti (eds.), 2011. A Biogeographic Assessment of the Samoan Archipelago. NOAA Technical Memorandum 
NOS NCCOS 132. Silver Spring, MD. 229 pp. 



A Biogeographic Assessment of the 
Samoan Archipelago 



Biogeography Branch 

Center for Coastal Monitoring and Assessment (CCMA) 

NOAA/NOS/NCCOS 

1305 East-West Highway, SSMC-4, N/SCI 1 

Silver Spring, MD 20910 

NOAA Technical Memorandum NOS NCCOS 132 

July 2011 

Editors: 

Matthew S. Kendall 
Matthew Poti 



The graphic wrapped on the cover was created by Kang Sevao of American Samoa and used with permis- 
sion here. This tapa design symbolizes the creation of land and sea. The processes of island formation and 
reef growth are central to understanding biogeography of the Samoan Archipelago. 



Mention of trade names or commercial products does not constitute endorsement or recommendation for 
their use by the United States Government. 




United States Department 
of Commerce 

Gary Locke 
Secretary 



National Oceanic and 
Atmospheric Administration 

Jane Lubchenco 
Administrator 



National Ocean Service 



David Kennedy 
Assistant Administrator 



ABOUT THIS DOCUMENT 

This assessment represents the continuation of ongoing partnerships between NOAA's National Centers 
for Coastal Ocean Science (NCCOS), Center for Coastal Monitoring and Assessment (CCMA), Biogeogra- 
phy Branch and the Office of National Marine Sanctuaries (ONMS) and Coral Reef Conservation Program 
(CRCP). The Biogeography Branch has applied a biogeographical approach to inform the management of 
marine resources within both coral reefs and National Marine Sanctuaries since 1998. To date, nine ONMS 
sites and most of the coral reef ecosystems in US states and territories have had some level of biogeographic 
characterization or mapping completed through these partnerships. 

The results of this ecological characterization are available via website. For more information on this project 
and those in other ONMS and reef ecosystem locations please visit the Biogeography Branch webpage at 
http://ccma.nos.noaa.gov/about/biogeography/ or direct questions and comments to: 

Chris Caldow, Biogeography Branch Chief 

National Oceanic and Atmospheric Administration 

1305 East West Highway 

SSMC 4, N/SCI-1 

Silver Spring, MD 20910 

Phone:(301)713-3028 

Email: Chris.Caldow@noaa.gov 




ACKNOWLEDGEMENTS 



EXECUTIVE SUMMARY 



This particular work was jointly funded by ONMS, CRCP, and CCMA and was conducted in consultation with 
local scientists and managers. The assessment would not have been possible were it not for the joint funding 
commitment and generous in-kind contributions from regional partners including suggestions during project 
inception, sharing of key datasets, contributions of co-authors, and review of draft analyses and documents. 
Mark Monaco led initial development of the project. Jamie Higgins formatted and organized the figures, 
tables, images, and text into this document. 



This report examines the marine biogeography 
of the Samoan Archipelago (-14° S latitude 
along the international date-line) with a focus 
on regional ocean climate, connectivity among 
islands due to larval transport, distributions of 
reef fish and coral communities, and the extent 
of existing marine protected areas. Manage- 
ment decisions and prior assessments in the 
archipelago have typically been split along the 
international political boundary between the 
islands of Samoa and those of American Sa- 
moa despite their close proximity and shared 
resources. A key goal in this assessment was 
to compile data from both jurisdictions and to 
conduct the characterization across the entire 
archipelago. The report builds upon earlier as- 
sessments by re-analyzing and interpreting 
many pre-existing datasets, adding more re- 
cent biogeographic data sources, and by com- 
bining earlier findings into a multidisciplinary 
summary of marine biogeography. 




Image 1. Acoral reef in the National Park of American Samoa near Vatia. 
Photo credit: Matt Kendall, NOAA Biogeography. 



The assessment is divided into 5 chapters and supporting appendices. Each chapter was written and re- 
viewed in collaboration with subject matter specialists and local experts. In Chapter 1 , a short introduction to 
the overall scope and approach of the report is provided. In Chapter 2, regional ocean climate is character- 
ized using remote sensing datasets and discussed in the context of local observations. In Chapter 3, regional 
ocean currents and transport of coral and fish larvae are investigated among the islands of the archipelago 
and surrounding island nations. In Chapter 4, distinct reef fish and coral communities across the archipelago 
are quantified on the basis of overall biodiversity, abundance, and community structure. In Chapter 5, the 
existing network of MPAs in American Samoa is evaluated based on the habitats, reef fish, and coral com- 
munities that are encompassed. Appendices provide analytical details omitted from some chapters for brevity 
as well as supplemental datasets needed as inputs for the main chapters in the assessment. Appendices 
include an inventory of regional seamounts, a description of shore to shelf edge benthic maps produced for 
Tutuila, analytical details of reef fish and coral datasets, and supplemental information on the many marine 
protected areas in American Samoa. 

The main objectives and some key findings of each chapter are as follows: 

Chapter 1: Introduction 

• Objectives were to introduce the physical setting of the archipelago and describe the scope and main 
components of the biogeographic assessment. 

Chapter 2: Oceanography of the Samoan Archipelago 

• Objectives were to summarize regional atmospheric and oceanographic conditions as well as trends in 
winds, waves, currents, sea surface temperature, chlorophyll, and sea surface height anomalies, and 
discuss potential influences they may have on biogeography of Samoan reef ecosystems. 

• Ocean conditions in the region are characterized by small seasonal fluctuations and often much larger 
multiyear fluctuations in response to broad climatic cycles such as the Southern Oscillation/El Nino. The 
major source of variability is seasonal for winds, waves, and sea surface temperature whereas chlorophyll 
and sea surface height are affected more by interannual processes. 



• Nearly all aspects of ocean climate for the archipelago vary more significantly by latitude than by longi- 
tude such that all islands except Swains, -400 km to the north of the archipelago, experience very similar 
conditions. 

• The archipelago has relatively more stable oceanic conditions compared to latitudes to the north and 
south. 

• Key climate related changes include gradual sea level rise as well as periodic low sea level events cor- 
responding to El Nino and also rising surface water temperatures and the threat of coral bleaching. 

Chapter 3: Ocean currents and larval transport among islands and shallow seamounts of the Samoan 
Archipelago and adjacent island nations. 

• Objectives were to describe regional ocean currents, identify key sources and destinations of coral and 
fish larvae for each island, and understand the influence of various combinations of larval life history char- 
acteristics on larval connections. 

• Major surface currents identified around the archipelago were the meandering westward flow of the South 
Equatorial Current (SEC) directly across the archipelago (13-19° S), the eastward flowing South Equato- 
rial Counter Current (SECC) (8-12° S) that seasonally (October -April) bifurcates the surface flow of the 
SEC, and a regularly occurring eddy south of the archipelago centered at -16° S and 172° W. 

• A wide range of larval longevities (10 to 100 days), mortality rates (3-46 % daily mortality), and settlement 
zones (9 to 36 km from islands) were investigated. 

• Major sources of larvae in the region are likely to be the large islands of Samoa, which contribute over 
twice as many larvae as the smaller islands of American Samoa. 

• Current transport is primarily westward along the archipelago such that each island tends to seed its natal 
reefs (especially with short-lived larvae) and island neighbors to the west (especially with long-lived lar- 
vae). In addition, the north coasts of Samoa may seed the islands of American Samoa via the feedback 
loops connecting the SECC with the SEC for organisms with long larval durations. 

• Current orientations and the long distance from upstream islands suggest the archipelago is heavily de- 
pendent on internal sources of larvae to sustain reef populations. Predicted connections among islands 
suggest potential benefits to coordinated management of marine resources and conservation planning 
between Samoa and American Samoa. 

Chapter 4: Biogeographic assessment of fish and coral communities of the Samoan Archipelago. 

• Objectives were to identify geographic patterns of hotspots, breakpoints, and spatial trends in reef fish 
and coral communities among and within islands of the archipelago. 

• Analysis focused on six variables: coral cover, coral diversity, coral community structure, fish biomass, 
fish diversity, and fish community structure. 

• Results from 8 studies were combined to determine regions with high, medium, and low values for each 
variable. 

• 30 distinct biogeographic regions with distinct patterns in one or more variables were identified across the 
archipelago. 

• 51 regional hotspots with relatively high values for particular fish or coral variables were identified. 

• Regions that were hotspots for several variables were northern, northeastern, and southern Savai'i, 
Swains Island, Ofu and Olosega Islands, Aunu'u and the eastern tip of Tutuila, southwestern Tutuila, and 
the Fagamalo area of northwestern Tutuila. 

• Regions that were not hotspots for any variables investigated included the Apolima Strait between Upolu 
and Savai'i, the north coast of Upolu, and parts of the northwest coast of Tutuila. 

• Regions that were identified as having unique reef fish and/or coral communities included southern 
Savai'i, Pago Pago Harbor, Aunu'u, Rose Atoll, and Swains Island. 



Chapter 5: The existing network of marine protected areas in American Samoa. 

• Objectives were to characterize the reef fishes, corals, habitats, and other key features of each existing 
MPA, evaluate the distribution of MPA sites in the context of the biogeographic regions and ecological 
hotspots defined in Chapter 4, summarize the area of reef ecosystem that is currently protected, by bot- 
tom type and reef type, and identify potentially important areas not currently in the network. Unlike the 



other chapters, analysis was restricted to American Samoa due to the lack of needed input datasets for 
Samoa. Creation of benthic maps and GIS datasets of MPA boundaries and regulations should be a prior- 
ity for conservation planning and resource management in Samoa. 

There are 23 MPAs in American Samoa managed by Territorial, Federal, or combined authorities. 
Only 8% of the potential coral reef ecosystem (defined as bottom regions less than 150 m deep) in Ameri- 
can Samoa is within existing MPAs. Only 3% has complete no-take restrictions. 

Fourteen of the twenty ecologically distinct biogeographic regions identified around American Samoa 
include at least one MPA, leaving only six with no representation in the present MPA network. 
High-value regions (those that were hotspots for 3 reef fish/coral variables) represented in the existing 
MPA network include southwestern Tutuila, the Fagamalo area, and Ofu and Olosega Islands. High-value 
regions lacking an MPA in the network include Aunu'u, the eastern tip of Tutuila, and Swains Island. 
Regions not currently represented in the existing MPA network that have been identified as having unique 
reef fish and/or coral communities include only Swains Island and Aunu'u. 

A comprehensive and coordinated MPA network strategy based on the findings of this study and other 
information is needed to define and accomplish conservation and resource management goals across the 
entire archipelago. 



EXECUTIVE SUMMARY TRANSLATED IN SAMOAN 

TALA 'OTO'OTO 

Translation provided by Veronika Mata'utia Mortenson. O lenei ripoti ua saunia faapitoa e su'esu'e ai le 
atoatoa o le olaga faa-natura o meaola o le sami i totonu o le atu-Samoa (-14° S o le ekueta, latalata i le 
laina o loo sui ai aso o le lalolagi) ma e patino lenei lomiga i tulaga o le tau, fesoota'iga ma isi atu-motu ona 
o feoa'iga o fua o i'a, faasoasoaina o nofoaga o i'a, a'au ma 'amu, faapea ma le tulaga o loo iai nei nofoaga 
faasao o le sami. O Amerika Samoa ma Samoa Tuto'atasi o ni motu e tu lalata ma e masani i le fefa'asoa'i 
o le tamaoaiga faa-natura, peita'i, o faai'uga ma ni isi o su'esu'ega na faatino i totonu o le atu-Samoa ua le 
maua iai se maliega ona o le ese'esega o faiga-nuu ma faiga-malo. O le faamoemoe maualuga o lenei ripoti 
o le tuufaatasia lea o ni faamaumauga ma faaiuga e uiga i le ola faa-natura i Amerika Samoa ma Samoa. O 
le oto'otoina o lenei lomiga na tuufaatasia mai i le anoano o su'esu'ega ua mae'a ona fa'asalalau, faapea le 
taumafaiga e tuu atu iai ma ni isi o faamatalaga lata mai e uiga i nofoaga o meaola o le sami.. 

O su'esu'ega o loo i totonu o lenei Ripoti ua mafai ona vaevaeina i ni mataupu se 5, faapea ma ni isi o faama- 
talaga e sapasapai ai mataupu ta'itasi. O mataupu uma ua faamauina, na tusia ma iloiloina e ni isi ua atamai 
ma iai agavaa faapitoa i le aotelega o lenei tusiga. I le Mataupu 1, o le upu tomua e faailoa ai galuega ma 
tulaga i loo i totonu o le ripoti. O le Mataupu 2 o loo faamatala ai fesuisuiaiga o le tau, mai molimau a tagata 
lautele faapea ai ma ni isi o su'esu'ega faa-saienisi. I le Mataupu 3, o le gasologa o le au ma le femalagaa'i o 
fua o 'amu ma i'a i le va o Samoa ma isi atu-motu. I le Mataupu 4, o le aofa'iga o i'a-a'au ua lauiloa ma ituaiga 
'amu ese'ese ua mafai ona faavasega i le anoanoa'i ma le felanulanua'i o ituaiga ese'ese. I le Mataupu 5, 
o le iloiloina o le fesootaiga o gataifale faasao i Amerika Samoa e afua mai lea i le mata'ituina o ituaiga i'a 
ese'ese ma ituaiga 'amu o loo maua ai. O isi Faamatalaga i le pito i tua o le Ripoti o loo auili'ili atu ai ni isi 
o mataupu taua. O loo aofia ai foi ma le faitau aofa'i o tama'i mauga o loo i le alititai, faapea faamatalaga o 
le faafanuaina o le talafatai e oo atu i le alititai, ma ni isi o faamatalaga e uiga i gataifale faasao i totonu o 
Amerika Samoa. 

O le taula'iga ma vaega faapitoa o mataupu ta'itasi: 



Mataupul: Upu Tomua 

• O ta'iala, na faaogaina e faalauiloa ai tulaga o atu-motu ma sa faapea ona faaoga fo'i e faamatala ai le 
faasoasoaina o meaola i lea motu. 



Mataupu2: Su'esu'ega o le Sami i le atu-Samoa. 

• Na faaogaina ta'iala e 'oto'oto ai tulaga o le 'ea ma le sami, faapea ma le tulaga o matagi, o galu, o le tafe o 
le au, vevela ma le loloto o le sami - e tau sa'ili ai fesoota'iga o nei mea uma i a'au ma 'amu o le atu-Samoa. 

• O le tulaga ole aga'i i luga ma lalo o le suasami ua mafai nei ona maitauina ona o le vevela o le kelope i le 
El Nino. O le mafua'aga 'autu o lenei luga lalo e oso masina pe faa-vaimasina mo matagi, galu ma le vevela 
o le suasami, a o le a'afia o le maualuga masani o le suasami e oso faa-vaitausaga. 

• Se'i vagana ai le motu o Swains, -400 km i matu o le atu-Samoa, e toeitiiti lava tutusa tulaga o le tau i le 
sami o atu-motu uma e tu lalata i le ekueta. 

• O le atu-Samoa e tutusa lelei tulaga o le sami i le itu i matG ma saute o le ekueta. 

• O suiga ua tula'i mai i fesuiaiga o le tau e aofia ai le si'isi'i malie i luga o le maualuga masani o le suasami 
faapea ai ma le maualalo tele ona o le El Nino. 

Mataupu 3: O au o le sami ma le femalagaa'i o fua-o-i'a ma figota i le va o motu faapea tama'i mauga 
ua ola i le alititai. 

• O le 'autG o lenei mataupu o le faamatalaina lea o le ala o au o le sami, faailo mafua'aga taua o taunu'uga 
o le femalagaa'i o 'amu ma fua-o-i'a i le va o motu ta'itasi, ma le malamalama'aga i le feso'otaiga o meaola 
uma o le sami. 

• O au o le sami i totonu o atu-motu ua mafai ona faamau i tusitusiga e aofia ai le South Equatorial Current 
(SEC) ma e tafe aga'i i sisifo, o le South Equatorial Counter Current (SECC) (8-12° S) e tafe aga'i i sasa'e 
mai le masina o Oketopa - Aperila ma e vaelua e lenei au le tafe a le SEC, ma ei totonu o le -16° i Saute 
ma 172° i Sisifo. 

• O le tele ma le umi o le olaga o fua o i'a (10 i le 100 aso), o le saoasaoa o le faatama'ia o fua-o-i'a (3-46 % 
i le aso), ma mea o loo ofaga ai (9 i le 36 km mai motu ta'itasi) sa mafai ona su'esu'eina. 

• O le anoanoa'i o fua-o-i'a o loo masalomia e tele i motu tetele o Samoa Tuto'atasi, ma fa'aluaina i lo motu 
laiti o Amerika Samoa. 

• O le malaga a fua-o-i'a e tele na aga'i atu i le itu i sisifo ma e 'umi lo latou ola i lo fua-o-i'a e aga'i atu i sasa'e. 
Ma o le isi, o le tafa-tai i matu o Samoa Tuto'atasi ua mafai ona faafailele ai fua-o-i'a ona toe taamilo mai 
lea ma aga'i mai i Amerika Samoa. 

• Talu ai le mamao o motu, e iai le talitonuga e ao ona faamoemoe le atu-Samoa i fua-o-i'a nei aua le 
faatupula'ia ma le faaleleia o a'au. O lenei fesootaiga i le atu-Samoa e aoga tele i le galulue fa'atasi o 
Samoa-na-lua i le faia lea o faaiuga ma ni maliega 'autasi - aua le puipuia atili o le tamaoaiga a Samoa 
Tuto'atasi ma Amerika Samoa. 

Mataupu 4: Su'esu'ega o ituaiga i'a ma 'amu ese'ese i le atu-Samoa. 

• O sini o nei su'esu'ega o le faailo lea o le anoanoa'i o 'auala ese'ese o loo lamatia ai le olaga o i'a ma 'amu 
i tafatai ma totonu o le atu-Samoa. 

• O nei su'esu'ega na faapito i ni vaega se ono: aofa'iga o 'amu, ituaiga 'amu ese'ese, anoano o 'amu 
ese'ese i se nofoaga e tasi, o le aofa'iga o le mamafa o i'a i se nofoaga e tasi, ituaiga i'a ese'ese, ma le 
ituaiga nofoaga e ofaga ai i'a. 

• O taunuuga o su'esu'ega e 8 na tuufaatasi e tau sa'ili ai poo fea vaega e maualuga, feololo ma maualalo. 

• I totonu o le atu-Samoa, e 30 ni nofoaga na mafai ona faailoa ona o le ma'oti o le itua'iga e tasi pe sili atu 
fo'i. 

• E 51 ni vaega vevela o le sami mai maugamu o loo mauluga ai le faitau aofa'i ma le aoga o i'a ma 'amu. 

• O vaega vevela o le sami sa tele ai meaola ese'ese o matu, matu-i-sasa'e, ma saute o Savai'i, Motu o 
Swains, Ofu ma Olosega, Aunu'u ma itu i sasa'e o Tutuila, faapea le itu i saute-sisifo, ma Fagamalo. 

• O isi nofoaga e le'i agava'a i vaega vevela o le sami e aofia ai le ava i Apolima i le va o Upolu ma Savai'i, le 
talafatai i matu o le motu o Upolu, ma ni vaega o le gataifale i matu-i-sasa'e o Tutuila. 

• O ni isi nofoaga sa mafai ona iloa ai le tutasi o i'a-a'au ma 'amu ese'ese o Savai'i, Pago Pago, Aunu'u ma 
le Motu o Swains. 

Mataupu 5: O le feso'ota'iga o nofoaga faasao o le gataifale i Amerika Samoa. 

• O le manulauti, o le faavasegaina lea o ituaiga i'a, 'amu, le mea o loo ofaga ai, ma isi faailo ua taua mo 
Gataifale Faasao - i le faatusatusaina lea i eria o loo 'oa i le tamaoaiga e pei ona faailoa atu i le Mataupu 4. 



E le pei o isi Mataupu, sa tapula'a su'esu'ega i Amerika Samoa ona e le'i iai ni faamaumauga mai Samoa. 

O le fausiaina o ni ata ma faafanua o le alititai e tatau ona tapena iai Samoa, faapea ma ni faamaumauga 

(GIS) e iloa ai tonu tua'oi ma tulafono i le puipuiina o le tamaoaiga. 

E 23 ni gataifale faasao o loo iai nei i Amerika Samoa, ma o loo pulea e le Teritori, Malo Feterale, ma isi 

pulega. 

Na o le 8% o a'au (e faauiga o nofoaga o loo i le loloto e 150 m) i Amerika Samoa ei totonu o gataifale 

faasao. E na o le 3% e le mafai ona fagotaina. 

Mai i le sefulufa o nofoaga faapitoa ua mafai ona tamau i faamaumauga i Amerika Samoa, o loo iai lava se 

Gataifale Faasao se tasi i totonu o lenei sefulufa, ma e ono e le o maua iai ni faamaumauga i le fesootaiga 

o gataifale faasao o loo iai nei. 

O nofoaga o loo maualuga le oa o puna'oa (le vaega sa 'avea ma tulaga vevela o le sami mo i'a ma 'amu 

ese'ese) i totonu o Gataifale Faasao e aofia ai le itu i saute-sisifo o Tutuila, Fagamalo, ma Ofu & Olosega. O 

nofoaga o loo maualuga ai puna'oa e le o mafai ona tuu i totonu o le fesootaiga o gataifale faasao a Amerika 

Samoa e aofia ai Aunu'u, le itu i sasa'e o Tutuila faapea ai ma le Motu o Swains. 

O nofoaga o loo tutasi ma tele ai le tamaoaiga, ma e le o iai i totonu o le li'o o Gataifale Faasao e aofia ai 

na'o le Motu o Swains ma Aunu'u. 

O loo mana'omia ni faamaumauga maumaututG e uiga i le fesootaiga o Gataifale Faasao, ina ia mafai ona 

aoga i le uiga ma ni taunuuga lelei o le faasao ma puipui atili o le tamaoaiga faa-natura i le atu-Samoa. 



TABLE OF CONTENTS 

CHAPTER 1 INTRODUCTION TO THE BIOGEOGRAPHIC ASSESSMENT 1 

CHAPTER 2 OCEANOGRAPHY OF THE SAMOAN ARCHIPELAGO 3 

INTRODUCTION 3 

DATA AND METHODS 3 

RESULTS 4 

CONCLUSIONS 22 

ACKNOWLEDGEMENTS 24 

REFERENCES 25 

CHAPTER 3: OCEAN CURRENTS AND LARVAL TRANSPORT AMONG ISLANDS AND SHALLOW 

SEAMOUNTS OF THE SAMOAN ARCHIPELAGO AND ADJACENT ISLAND NATIONS 27 

INTRODUCTION 27 

METHODS 28 

RESULTS AND DISCUSSION 35 

CONCLUSIONS 90 

ACKNOWLEDGEMENTS 92 

REFERENCES 93 

CHAPTER 4: BIOGEOGRAPHIC ASSESSMENT OF FISH AND CORAL COMMUNITIES 

97 

INTRODUCTION 97 

METHODS 98 

RESULTS 104 

CONCLUSIONS 114 

ACKNOWLEDGEMENTS 118 

REFERENCES 119 

CHAPTER 5: THE EXISTING NETWORK OF MARINE PROTECTED AREAS IN AMERICAN SAMOA ...123 

INTRODUCTION 123 

METHODS 125 

RESULTS: BENTHIC HABITATS OF AMERICAN SAMOA 129 

RESULTS: SITE CHARACTERIZATIONS 130 

RESULTS: M PA NETWORK ANALYSES 174 

CONCLUSIONS 180 

ACKNOWLEDGEMENTS 182 

REFERENCES 183 

APPENDIX A: SEAMOUNTS WITHIN THE EXCLUSIVE ECONOMIC ZONES OF SAMOA AND 

AMERICAN SAMOA 189 

APPENDIX B: SHORELINE TO SHELF EDGE BENTHIC MAPS OF TUTUILA, AMERICAN SAMOA 197 

APPENDIX C: FISH AND CORAL DATA PLOTS 205 

APPENDIX D: KEY ATTRIBUTES AND ACTIVITIES OF MPAS IN THE EXISTING NETWORK OF 

MPAS IN AMERICAN SAMOA AS OF JANUARY 2011 225 



List of Tables 

Table 2.1. Original data sources 4 

Table 3.1. Full factorial ANOVA on the effects of region, season, and ENSO status on median drifter heading 36 

Table 3.2. Full factorial ANOVA on the effects of region, season, and ENSO status on mean drifter speed 36 

Table 3.3. Islands, seamounts, or island groups used as source locations in simulations of regional larval 

connectivity, their corresponding potential reef area (0-150 m shelf), number of virtual larvae 

used in modeling, and the percentage of the simulated larval pool contributed by each source 42 

Table 4.1. List of datasets and variables used in the analysis 100 

Table 4.2. Assigned breakpoints between low and medium (L -^ M) and medium and high (M — ► H) values for 

each of the fish and coral variables by dataset 102 

Table 4.3. Hotspot analysis summary table 113 

Table 5.1. Existing MPAs in American Samoa as of January 2011 126 

Table 5.2. Biogeographic regions, ecological hotspots, and overlap with existing MPAs 176 

Table 5.3. Potential reef ecosystem area (km 2 ) by benthic structure type for existing MPAs 179 

Table A.1. Locations and morphological characteristics of seamounts within the EEZ of American Samoa and 

Samoa 193 

Table D.1. General description of MPA implementation and management 225 

Table D.2. Conservation focus and management practices 226 

Table D.3. Biological and socio-economic monitoring/assessment and community involvement 227 

Table D.4. Current and future projects 228 



List of Figures 


Figure 


1.1. 


Figure 


2.1. 


Figure 


2.2. 


Figure 


2.3. 


Figure 


2.4. 


Figure 


2.5. 


Figure 


2.6. 


Figure 


2.7. 


Figure 


2.8. 


Figure 


2.9. 


Figure 


2.10. 


Figure 


2.11. 


Figure 


2.12. 


Figure 


2.13. 


Figure 


2.14. 


Figure 


2.15. 


Figure 


2.16. 


Figure 


2.17. 


Figure 


2.18. 


Figure 


2.19. 



Figure 2.20. 



Figure 3.1. 



Figure 


3.2. 


Figure 


3.3. 


Figure 


3.4. 


Figure 


3.5. 


Figure 


3.6. 


Figure 


3.7. 


Figure 


3.8. 


Figure 


3.9a. 


Figure 


3.9b. 


Figure 


3.10a. 


Figure 


3.10b 


Figure 


3.11. 


Figure 


3.12. 


Figure 


3.13. 


Figure 


3.14. 


Figure 


3.15. 


Figure 


3.16. 


Figure 


3.17. 


Figure 


3.18. 


Figure 


3.19. 


Figure 


3.20. 


Figure 


3.21. 


Figure 


3.22. 



Figure 3.23. 



Samoan Archipelago study region 1 

Wind direction measured by the QuikSCAT satellite 5 

Path and intensity of cyclones passing through the EEZs of Samoa or American Samoa 

from 2000-2007 6 

Major surface currents of the Southern Pacific Ocean adapted from Tomczak and Godfrey (2003) 7 

Sea surface temperature data from CoRTAD presented as an average annual cycle 8 

Sea surface temperatures from CoRTAD 9 

Sea surface temperature and anomaly values from CoRTAD for the years 1985 to 2006 10 

Sea surface temperature anomalies from CoRTAD during the month of July for the years 1997 to 2005.... 11 

Sea surface temperature anomaly plots for warmest water month 12 

Bleaching alert time-series from NOAA Coral Reef Watch 13 

Chlorophyll concentration estimated from SeaWiFS and presented as an average annual cycle 14 

Chlorophyll concentration estimated from the SeaWiFS satellite 15 

Chlorophyll concentrations estimated from SeaWiFS for the years 1997 to 2005 16 

Chlorophyll anomalies estimated from SeaWiFS for June 2002 and April 2005 17 

Sea level values for Pago Pago, American Samoa from 1948 to 2008 17 

Sea surface height anomalies from AVISO are presented as an average annual cycle 18 

Sea surface height anomalies from AVISO 19 

Sea surface height anomalies from AVISO for the years 1993 to 2006 20 

Sea surface height anomalies from AVISO during the month of March for the years 1993 to 2006 21 

Linear regression of maximum monthly sea level deviation versus SOI during El Nino events 

since 1954 22 

Sea Surface Temperature from CoRTAD at the intersection of 170° W longitude (the approximate 
boundary between Samoa and American Samoa) and 0°, 5° S, 10° S, 15° S, 20° S, 25° S, 

and 30° S latitude respectively 23 

Samoan Archipelago and surrounding islands depicted by the 9 km grid cells of the HYCOM 

hydrodynamic model 28 

Example of regional current vectors 30 

Tonga Trench Eddy 30 

Example of typical drifter paths. January 2007 drifters shown 31 

Box plots of median current headings by region, season, and ENSO conditions based on drifter data... 35 

Median current headings by season and region based on drifter data 36 

Box plots of median current speeds by region, season, and ENSO conditions based on drifter data 37 

Mean gross and net displacement of drifters after 10, 20, 30, 50, and 100 days 37 

Gross displacement of individual drifters as a frequency histogram after 10, 20 ,30, 50 and 

100 days at large 38 

Net displacement of individual drifters as a frequency histogram after 10, 20 ,30, 50 and 

100 days at large 39 

Surface current patterns of the Samoan EEZs and surrounding region for October through April 40 

Surface current patterns of the Samoan EEZs and surrounding region for May through September 40 

Proportion of virtual larvae in the study area that were started from each island, seamount, 

or island group 41 

Cumulative connectivity, 2004-2008 43 

Transport of virtual larvae from Swains Island by model year 45 

Position of virtual larvae from southern Savai'i for all model years by PLD 46 

External larval supply and local larval retention at Savai'i-South as a function of PLD 

and mortality rate 47 

Destinations (and sources) of simulated larvae originating from (arriving at) Savai'i-South 

for low, medium, and high larval mortality rates 48 

Position of virtual larvae from northern Savai'i for all model years by PLD 50 

External larval supply and local larval retention at Savai'i-North as a function of PLD and mortality rate.... 51 
Destinations (and sources) of simulated larvae originating from (arriving at) Savai'i-North for low, 

medium, and high larval mortality rates 52 

Position of virtual larvae from southern Upolu for all model years by PLD 54 

External larval supply and local larval retention at Upolu-South as a function of PLD and mortality rate. ...55 
Destinations (and sources) of simulated larvae originating from (arriving at) Upolu-South for low, 

medium, and high larval mortality rates 56 

Position of virtual larvae from northern Upolu for all model years by PLD 58 



List of Figures (cont.) 

Figure 3.24. External larval supply and local larval retention at Upolu-North as a function of PLD and mortality rate 59 

Figure 3.25. Destinations (and sources) of simulated larvae originating from (arriving at) Upolu-North for low, 

medium, and high larval mortality rates 60 

Figure 3.26. Position of virtual larvae from Tutuila for all model years by PLD 62 

Figure 3.27. External larval supply and local larval retention at Tutuila as a function of PLD and mortality rate 63 

Figure 3.28. Destinations (and sources) of simulated larvae originating from (arriving at) Tutuila for low, 

medium, and high larval mortality rates 64 

Figure 3.29. Position of virtual larvae from South Bank for all model years by PLD 66 

Figure 3.30. External larval supply and local larval retention at South Bank as a function of PLD and mortality rate 67 

Figure 3.31. Destinations (and sources) of simulated larvae originating from (arriving at) South Bank for low, 

medium, and high larval mortality rates 68 

Figure 3.32. Position of virtual larvae from East Bank for all model years by PLD 70 

Figure 3.33. External larval supply and local larval retention at East Bank as a function of PLD and mortality rate. ... 71 
Figure 3.34. Destinations (and sources) of simulated larvae originating from (arriving at) East Bank for low, 

medium, and high larval mortality rates 72 

Figure 3.35. Position of virtual larvae from Northeast Bank for all model years by PLD 74 

Figure 3.36. External larval supply and local larval retention at Northeast Bank as a function of PLD and 

mortality rate 75 

Figure 3.37. Destinations (and sources) of simulated larvae originating from (arriving at) Northeast Bank 

for low, medium, and high larval mortality rates 76 

Figure 3.38. Position of virtual larvae from Manu'a for all model years by PLD 78 

Figure 3.39. External larval supply and local larval retention at Manu'a as a function of PLD and mortality rate 79 

Figure 3.40. Destinations (and sources) of simulated larvae originating from (arriving at) Manu'a for low, 

medium, and high larval mortality rates 80 

Figure 3.41. Position of virtual larvae from Rose Atoll for all model years by PLD 82 

Figure 3.42. External larval supply and local larval retention at Rose Atoll as a function of PLD and mortality rate. ... 83 
Figure 3.43. Destinations (and sources) of simulated larvae originating from (arriving at) Rose Atoll for low, 

medium, and high larval mortality rates 84 

Figure 3.44. Position of virtual larvae from Swains Island for all model years by PLD 86 

Figure 3.45. External larval supply and local larval retention at Swains Island as a function of PLD and 

mortality rate 87 

Figure 3.46. Destinations (and sources) of simulated larvae originating from (arriving at) Swains Island for low, 

medium, and high larval mortality rates 88 

Figure 4.1. Datasets and corresponding survey sites included in the analyses 99 

Figure 4.2. Biogeographic regions (Bioregions) assigned based on analyses offish and coral data 105 

Figure 4.3. Coral cover at survey sites across Samoa and American Samoa 106 

Figure 4.4. Coral richness at survey sites across Samoa and American Samoa 107 

Figure 4.5. Summary of Bioregions sharing similar coral communities and those with unique coral 

communities as identified from the MDS analyses 108 

Figure 4.6. Fish biomass at survey sites across Samoa and American Samoa 109 

Figure 4.7. Fish richness at survey sites across Samoa and American Samoa 111 

Figure 4.8. Summary of Bioregions sharing similar fish communities and those with unique fish communities 

as identified from the MDS analyses 112 

Figure 4.9. Fish and coral hotspots by Bioregion 114 

Figure 5.1. Existing MPAs in American Samoa as of January 2011 127 

Figure 5.2. Density of pigs (pigs/km 2 ) in piggeries by watershed for Tutuila and Manu'a watersheds 127 

Figure 5.3. (a) Proportion of mapped benthic structure types in American Samoa overall, (b) Proportion of 

coral reef and hardbottom in each reef zone 129 

Figure 5.4. Benthic habitat (by structure type) and fish and coral survey data within Alega Private 

Marine Reserve 130 

Figure 5.5. (a) Proportion of benthic structure types in Alega Private Marine Reserve, (b) Proportion of coral 

reef and hardbottom in each reef zone 131 

Figure 5.6. Fish and coral data collected in Alega Private Marine Reserve 131 

Figure 5.7. Benthic habitat (by structure type) and fish and coral survey data within the Alofau CFMP reserve 132 

Figure 5.8. (a) Proportion of benthic structure types in the Alofau CFMP reserve, (b) Proportion of coral reef 

and hardbottom in each reef zone 133 

Figure 5.9. Fish and coral data collected in the Alofau CFMP reserve 133 

Figure 5.10. Benthic habitat (by structure type) and fish and coral survey data within the Amanave CFMP reserve 134 



List of Figures (cont.) 

Figure 5.11 . (a) Proportion of benthic structure types in the Amanave CFMP reserve, (b) Proportion of coral 

reef and hardbottom in each reef zone 135 

Figure 5.12. Fish and coral data collected in the Amanave CFMP reserve 135 

Figure 5.13. Benthic habitat (by structure type) and fish and coral survey data within the Amaua and 

Auto CFMP reserve 136 

Figure 5.14. (a) Proportion of benthic structure types in the Amaua and Auto CFMP reserve, (b) Proportion 

of coral reef and hardbottom in each reef zone 137 

Figure 5.15. Fish and coral data collected in the Amaua and Auto CFMP reserve 137 

Figure 5.16. Benthic habitat (by structure type) within the Aoa CFMP reserve 138 

Figure 5.17. (a) Proportion of benthic structure types in the Aoa CFMP reserve, (b) Proportion of coral reef 

and hardbottom in each reef zone 139 

Figure 5.18. Benthic habitat (by structure type) within the Aua CFMP reserve 140 

Figure 5.19. (a) Proportion of benthic structure types in the Aua CFMP reserve, (b) Proportion of coral reef 

and hardbottom in each reef zone 141 

Figure 5.20. Benthic habitat (by structure type) and fish and coral survey data within the Fagamalo CFMP 

reserve 142 

Figure 5.21. (a) Proportion of benthic structure types in the Fagamalo CFMP reserve, (b) Proportion of coral 

reef and hardbottom in each reef zone 143 

Figure 5.22. Fish and coral data collected in the Fagamalo CFMP reserve 143 

Figure 5.23. Benthic habitat (by structure type) and fish and coral survey data within the Fagamalo No-Take MPA 144 

Figure 5.24. (a) Proportion of benthic structure types in the Fagamalo No-Take MPA. (b) Proportion of coral 

reef and hardbottom in each reef zone 145 

Figure 5.25. Comparison offish and coral data collected in the Fagamalo No-Take MPA to data from all 

of American Samoa 145 

Figure 5.26. Benthic habitat (by structure type) within the Leone Pala SMA 146 

Figure 5.27. Proportion of benthic structure types in the Leone Pala SMA 147 

Figure 5.28. Benthic habitat (by structure type) and fish and coral survey data within the Masausi CFMP reserve 148 

Figure 5.29. (a) Proportion of benthic structure types in the Masausi CFMP reserve, (b) Proportion of 

coral reef and hardbottom in each reef zone 149 

Figure 5.30. Fish and coral data collected in the Masausi CFMP reserve 149 

Figure 5.31. Benthic habitat (by structure type) and fish and coral survey data within the Matu'u and 

Faganeanea CFMP reserve 150 

Figure 5.32. (a) Proportion of benthic structure types in the Matu'u and Faganeanea CFMP reserve. 

(b) Proportion of coral reef and hardbottom in each reef zone 151 

Figure 5.33. Fish and coral data collected in the Matu'u and Faganeanea CFMP reserve 151 

Figure 5.34. Benthic habitat (by structure type) and fish and coral survey data within the Nu'uuli Pala SMA 152 

Figure 5.35. (a) Proportion of benthic structure types in the Nu'uuli Pala SMA. (b) Proportion of coral reef and 

hardbottom in each reef zone 153 

Figure 5.36. Benthic habitat (by structure type) and fish and coral survey data within the Ofu Vaoto Marine Park.... 154 
Figure 5.37. (a) Proportion of benthic structure types in the Ofu Vaoto Marine Park, (b) Proportion of coral reef 

and hardbottom in each reef zone 155 

Figure 5.38. Fish and coral data collected in the Ofu Vaoto Marine Park 155 

Figure 5.39. Benthic habitat (by structure type) within the Pago Pago Harbor SMA 156 

Figure 5.40. (a) Proportion of benthic structure types in the Pago Pago Harbor SMA. (b) Proportion of coral 

reef and hardbottom in each reef zone 157 

Figure 5.41. Benthic habitat (by structure type) and fish and coral survey data within the Poloa CFMP reserve 158 

Figure 5.42. (a) Proportion of benthic structure types in the Poloa CFMP reserve, (b) Proportion of coral reef 

and hardbottom in each reef zone 159 

Figure 5.43. Fish and coral data collected in the Poloa CFMP reserve 159 

Figure 5.44. Benthic habitat (by structure type) within the Sailele CFMP reserve 160 

Figure 5.45. (a) Proportion of benthic structure types in the Sailele CFMP reserve, (b) Proportion of coral reef 

and hardbottom in each reef zone 161 

Figure 5.46. Benthic habitat (by structure type) and fish and coral survey data within the Vatia CFMP reserve 162 

Figure 5.47. (a) Proportion of benthic structure types in the Vatia CFMP reserve, (b) Proportion of coral reef 

and hardbottom in each reef zone 163 

Figure 5.48. Fish and coral data collected in the Vatia CFMP reserve 163 

Figure 5.49. Benthic habitat (by structure type) and fish and coral survey data within Fagatele Bay NMS 164 

Figure 5.50. (a) Proportion of benthic structure types in Fagatele Bay NMS. (b) Proportion of coral reef and 

hardbottom in each reef zone 165 



List of Figures 
Figure 5.51. 
Figure 5.52. 

Figure 5.53. 

Figure 5.54. 

Figure 5.55. 
Figure 5.56. 

Figure 5.57. 

Figure 5.58. 

Figure 5.59. 

Figure 5.60. 

Figure 5.61. 
Figure 5.62. 

Figure 5.63. 

Figure 5.64. 

Figure 5.65. 

Figure 5.66. 



Figure 5.67. 
Figure A.1. 
Figure A.2. 

Figure A.3. 
Figure A.4. 
Figure A.5. 
Figure A.6. 
Figure A.7. 
Figure A.8. 
Figure B.1. 
Figures C. 1-31 

Figure C.32. 
Figure C.33. 
Figure C.34. 



(cont.) 

Comparison offish and coral data collected in Fagatele Bay NMS to data from all of American Samoa. . 165 
Benthic habitat (by structure type) and fish and coral survey data within the Ofu Unit of the 

National Park 166 

(a) Proportion of benthic structure types in the Ofu Unit of the National Park, (b) Proportion of 

coral reef and hardbottom in each reef zone 167 

Comparison offish and coral data collected in the Ofu Unit of the National Park to data from all 

of American Samoa 167 

Benthic habitat (by structure type) and fish and coral survey data within the Ta'u Unit of the National Park. . 168 
(a) Proportion of benthic structure types in the Ta'u Unit of the National Park, (b) Proportion of coral 

reef and hardbottom in each reef zone 169 

Comparison offish and coral data collected in the Ta'u Unit of the National Park to data from all 

of American Samoa 169 

Benthic habitat (by structure type) and fish and coral survey data within the Tutuila Unit of 

the National Park 170 

(a) Proportion of benthic structure types in the Tutuila Unit of the National Park, (b) Proportion 

of coral reef and hardbottom in each reef zone 171 

Comparison offish and coral data collected in the Tutuila Unit of the National Park to data from 

all of American Samoa 171 

Mapped benthic habitat (by structure type) and fish and coral survey data within Rose Atoll MNM 172 

(a) Proportion of mapped benthic structure types in Rose Atoll MNM. (b) Proportion of coral reef 

and hardbottom in each reef zone 173 

Comparison offish and coral data collected in the Rose Atoll MNM to data from all of 

American Samoa 173 

(a) Proportion of the total potential reef ecosystem area around American Samoa by benthic 
structure type for the entire suite of existing MPAs and for the rest of American Samoa, (b) Proportion 

of the total potential reef ecosystem with no-take restrictions and with other fishing restrictions 174 

(a) Proportion of coral reef habitat by benthic structure type for the entire suite of existing MPAs 
and for the rest of American Samoa, (b) Proportion of coral reef habitat with no-take restrictions 

and with other fishing restrictions 175 

Distribution of existing MPAs relative to the locations of significant ecological features, including 
Bioregions that are hotspots for three fish/coral variables (Chapter 4) and the mesophotic coral 

banks surrounding Tutuila (Appendix B) 177 

Proportion of potential reef ecosystem area by benthic structure type for each existing MPA 178 

Seamounts of the Samoan Exclusive Economic Zones 189 

Frequency distribution of seamounts within the Samoan and American Samoan EEZ based on 

a) depth of seamounttop and b) height 190 

Vailulu'u Seamount, the active hotspot for the Samoan Island Chain 191 

Papatua Guyot (South Bank) 191 

Tulaga (East Bank) 192 

Muli Guyot (NE Bank) 192 

Pasco Seamount 192 

Toafilemu Seamount 192 

Cross section of Zones 198 

. Bar graphs depicting the distribution of percent coral cover (C.1-8), coral richness (C.9-15), 

fish biomass (C. 16-23), and fish richness (C. 24-31) for each of the studies 205 

MDS plots based on coral community data for sites in each of the American Samoa studies 221 

MDS plots based on fish community data for sites in each of the American Samoa studies 222 

MDS plots based on coral and fish community data for sites in each of the Samoa studies 223 



Introduction to the Biogeographic Assessment 

Matthew S. Kendall 1 




Figure 1.1. Samoan Archipelago study region. 



This report provides an assessment 
of the marine biogeography of the 
Samoan Archipelago with a focus on 
oceanography, reef fish, and coral 
communities. Biogeography examines 
the distribution of biota and their habi- 
tats as well as the environmental fac- 
tors that have shaped them. Biogeo- 
graphic characterizations are among 
the basic information inputs required 
not only for making informed manage- 
ment decisions but also building public 
support for them. 

The Samoan Archipelago lies in the 
South Pacific Ocean along -14° S 
latitude at the international date-line 
(Figure 1). The archipelago is com- 
prised of a chain of volcanic islands, 
seamounts, and coral atolls and is di- 
vided into two countries: Samoa and 
American Samoa. The much larger 
islands of Savai'i and Upolu comprise 

most of the independent nation of Samoa, formerly called Western Samoa. American Samoa (a Territory of 
the United States) is made up of the comparatively medium sized island of Tutuila, the smaller islands of the 
Manu'a group, and the two small, remote coral atolls of Swains Island and Rose Atoll that are not derived 
from the same volcanic hotspot as the rest of the island chain. 

Many prior assessments have touched on the biogeography of either Samoa or American Samoa and are 
cited throughout this document. The present report builds upon these earlier assessments by combining and 
re-analyzing their original datasets, adding more recent biogeographic data sources, and by combining and 
re-interpreting their individual findings into a multidisciplinary summary of marine biogeography. 

Despite their close proximity and shared resources, management decisions and prior assessments in the 
region have typically been split along the international political boundary between Samoa and American 
Samoa. In contrast, a key goal in this assessment was to compile data from both areas and to conduct the 
characterization across the entire archipelago. Results of the assessment are intended partly to support the 
"2 Samoa's Initiative", a recent cooperative agreement between the two jurisdictions that seeks to foster 
improved collaboration, coordination, and information exchange on natural resource management and other 
topics. The Governments of Samoa and American Samoa should be contacted directly for more information 
on the current status of this unfolding initiative. 

Of note, much of the data used in this assessment was collected prior to the September 2009 tsunami that 
devastated some shallow water and low lying segments of the archipelago. Most parts of this assessment 
however, were conducted at a broad analysis scale and the types of data used were not highly sensitive to 
this significant and anomalous natural disturbance. For more information on tsunami impacts, interested 
readers are directed to specific studies that were conducted to evaluate the extent and severity of damage 
due to that event. 



NOAA/NOS/NCCOS/CCMA Biogeography Branch 



A key application intended for the report is to pro- 
vide guidance in the ongoing development of a 
network of Marine Protected Areas (MPA) in the 
Samoan Archipelago. The region is already home 
to a diversity of MPAs implemented at various 
levels of government from individual villages and 
communities to federally protected areas of inter- 
national significance. Many of the different MPAs 
in the network were created through independent 
processes for different objectives but each con- 
tributes to the mosaic of marine resource man- 
agement in the region. Understanding what fish, 
coral, and habitat resources this diverse network 
of MPAs collectively encompasses is a key ob- 
jective of this work and is critical for understand- 
ing the scope of current protection and thought- 
fully designing additional network elements. , „ r _ , 

J ° ° Image 2. Reef fish sold in a local grocery. 

Photo: Matt Kendall, NOAA Biogeography. 
As a result of discussions with project partners in 

the design phase of the assessment, this report 

focuses on corals and reef fish, transport of their larvae, and the reef habitats where they live. Additional as- 
pects of biogeography that are not included in this assessment but are important to the region and Samoan 
culture include sea birds, cetaceans, deep coral habitats, and pelagic fish communities to name but a few. 
Including these resources was beyond the scope of our assessment although they have been investigated in 
several individual studies that should be consulted for more information. 




The assessment is divided into 5 chapters with supporting appendices. Each chapter was based on compila- 
tion of multiple pre-existing datasets, original analysis, and discussion that has not been previously published. 
Each chapter was written or reviewed in collaboration with subject matter specialists and local experts. Here 
in Chapter 1, the overall scope and approach of the report is introduced. In Chapter 2, regional ocean climate 
is characterized including wind and wave climate, sea surface temperature, primary productivity, and sea 
level fluctuations. The focus is on the spatial and temporal patterns and trends in ocean climate that may 
affect marine biogeography. In Chapter 3, regional ocean currents and transport of coral and fish larvae are 
investigated among the islands of the archipelago as well as the surrounding island nations. The degree of 
self seeding versus dependency on outside sources offish and coral larvae for maintaining each islands reef 
ecosystem is quantified. Major and secondary sources of larvae for each island are discussed in terms of 
resilience of reefs to disturbance. In Chapter 4, the reef fish and coral communities of the archipelago are 
quantified on the basis of overall biodiversity, abundance, and community structure. Biogeographic trends, 
breakpoints, and hotspots are identified among and within each of the islands in the archipelago. In Chapter 
5, we summarize the existing network of MPAs in American Samoa based on their habitats, reef fish, and 
coral communities. Presently protected features are compared to regional resources, and remaining gaps 
in resource protection are highlighted. Appendices include analytical details omitted from some chapters for 
brevity as well as important secondary analytical products needed as inputs for the main chapters in the as- 
sessment. This includes an inventory and summary of regional seamounts needed for the larval connectivity 
chapter (Chapter 3), analytical details of the reef fish and coral datasets (Chapter 4), a description of the 
shore to shelf edge benthic maps created and used for the MPA network analysis (Chapter 5), and supple- 
mental information on the many marine protected areas in American Samoa (Chapter 5). 



Oceanography of the Samoan Archipelago 

Doug Pirhalla 1 , Varis Ransi 1 , Matthew S. Kendall 2 and Doug Fenner 3 




Image 3. A close-up look at a diverse benthic community. 
Photo credit: Matt Kendall, NOAA, Biogeography Branch. 



INTRODUCTION 

The biogeography and health of coral reef 
ecosystems in the Samoan Archipelago are 
shaped in part by the oceanographic condi- 
tions and processes of the equatorial South 
Pacific. Larvae that reach the archipelago are 
carried to the region on ocean currents and 
those organisms that arrive and thrive must be 
adapted to the climatic conditions that charac- 
terize the region including temperature, winds, 
waves, nutrients, tides, sea level, and other 
factors. Once established, reef ecosystems 
can be stressed and modified by a wide range 
of climate-related phenomena such as elevat- 
ed ocean temperatures, sea level fluctuations, 
and ocean acidification. Many oceanographic 
and atmospheric processes affecting Samoan 
reefs are presently in flux due to global climate 
change (Chase and Veitayaki 1992, Timmer- 
man et al. 1999, US EPA 2007, Young 2007, 
Barshis et al. 2010). This chapter provides a 
summary of regional atmospheric and oceanographic conditions and trends including winds, waves, cur- 
rents, sea surface temperature, chlorophyll, and sea surface height anomalies, and discusses potential influ- 
ences they may have on Samoan reef ecosystems. 

Climate Background 

The climate of the Samoan Archipelago is characterized by year-round mild air temperatures, high humidity, 
persistent easterly or northeasterly trade winds, and infrequent but severe cyclonic storms. Mean daily air 
temperature varies between 22°C and 30°C (SPSLCMP 2007). The islands are noted for high annual rainfall 
that averages >3,000 mm (120 inches) per year but varies locally depending on topography (http://www7. 
ncdc.noaa.gov/CDO/cdo). Maximum rainfall occurs in the austral summer (December-February) where it 
can exceed 300 mm/month. In winter (June-August), rainfall is 30% lower at approximately 200 mm/month. 

DATA AND METHODS 

A diversity of satellite sensors has provided estimates of oceanic and atmospheric variables at global scales 
for the last few decades. These satellite-based datasets and other supporting information were used to de- 
scribe the typical seasonal fluctuations, inter-annual variability, long-term trends, and anomalous events of 
importance to coral reef ecosystems in the Samoan Archipelago. Oceanographic variables in this assess- 
ment include winds, waves, ocean circulation, sea surface temperature, chlorophyll, and sea surface height 
anomalies. For each variable, the assessment provides: 1) a brief description of the remote sensing and 
other data that were analyzed, 2) a broad-scale overview of the major ocean features and processes at work 
while highlighting the position of the Samoa and American Samoa Exclusive Economic Zones (EEZs), 3) a 
finer-scale description of the seasonal patterns for each variable comparing ocean measurements close to 
the islands of Savai'i (172.66 W, 14.26 S) and Tutuila (170.2 W, 14.26 S) respectively (for these analyses, 
ocean characteristics were extracted for an area of 80 km 2 at the same latitude excluding land and shallow 
water areas) and throughout the American Samoa EEZ (only the American Samoa EEZ is included for sim- 
plicity since it encompasses the conditions experienced in the Samoan EEZ), 4) a time series of available 



1 NOAA/NOS/NCCOS/CCMA Coastal Oceanographic Assessment Status and Trends Branch 

2 NOAA/NOS/NCCOS/CCMA Biogeography Branch 

3 American Samoa/Department of Marine and Wildlife Resources 



data showing multi-year trends in climate patterns, and 5) a description of the frequency and intensity of 
anomalous conditions that are of particular relevance to coral reef ecosystems. Ocean pixels in the satellite 
data that were contaminated with land or shallow water signatures were excluded in analyses. Preliminary 
analysis revealed that for most variables, monthly averages or plots for every other month were suitable to 
convey seasonal patterns. Where annual cycles are plotted, monthly means are averaged across all years 
of data. For example, there were 21 years of sea surface temperature (SST) data. Average SST for January 
in all 21 years was averaged to create a composite seasonal cycle. This enables identification of typical sea- 
sonal patterns but can obscure important short-term phenomena. Such extreme values that have occurred 
in specific years or months are highlighted in separate plots. Original remote sensing data used in this study 
are freely available and should be downloaded from original sources (Table 2.1). 

Table 2.1. Original data sources. 



PRODUCT TYPE 






DATA SOURCE 






TIME SPATIAL TEMPORAL IIMITQ 
FRAME RESOLUTION SUMMARY unmj> 




QuikSCAT Sea Surface Winds 


Remote Sensing Systems 


1999-2007 


25 km 


Weekly/Monthly 


m/s, degrees from 
north 


Jason-1, Topex/Poseidon, 
ERS-1/2ENVISAT Geostrophic 
Surface Currents 


AVISO, SSALTO/DUACS & 
CNES 


1992-2006 


1/3° grids 


Weekly/Monthly 


cm/s degrees from 
north 


Pathfinder SST and SST 
Anomaly (CoRTAD)) 


NOAA/NESDIS/NODC 


1985-2006 


4 km 


Weekly/Monthly 


°c 


GOES-1 0/11 SST and 
SST fronts 


NESDIS/NODC/STAR 


2000-2007 


4 km 


Daily/Monthly 


°c 


SeaWiFS Ocean Color 
Chlorophyll and Anomalies 


NASA 


1997-2007 


1 km 


Daily/Monthly 


M9L-1, 
Steradian -1 


Jason-1, Topex/Poseidon, 
ERS-1/2ENVISAT Sea Surface 
Height Anomolies 


AVISO, SSALTO/DUACS & 
CNES 


1992-2006 


1/4° grids 


Weekly/Monthly 


cm 



RESULTS 
Wind 

Magnitude and direction of winds near the ocean surface are measured by the QuikSCAT satellite's micro- 
wave scatterometer (http://www.remss.com/qscat/qscat_description.html). Weekly and monthly averaged 
data are available at a 25 km spatial resolution for a 7-year period (July 1 999 to September 2007). Data were 
used to discern the broad-scale atmospheric circulation features in the South Pacific, place the Samoan 
EEZs into regional context, and to depict prevailing wind patterns within the archipelago over a typical annual 
cycle. 

The region is dominated by the Trade Winds, a persistent atmospheric system where surface winds blow 
from the northeast to the southwest (yellow-green colors; Figure 2.1). Trade Winds are typically stronger in 
winter (July) than in summer (Merrill 1989). A major atmospheric feature affecting the Samoan climate is the 
South Pacific Convergence Zone (SPCZ) where Trade Winds converge at the surface (Figure 2.1). To the 
north of the convergence zone winds are generally southwestward. To the south of the convergence zone 
winds are generally westward/northwestward. This area of convergence results in heightened rainfall, espe- 
cially during summer months (December-February). The SPCZ undergoes shifts in position and intensity on 
both a seasonal and interannual basis. The SPCZ crosses over the Samoan Archipelago twice a year (Alory 
and Delcroix 1999). It is most clearly established over the Samoan Archipelago during the summer months 
(December - February) whereas in winter (June - August) the zone shifts slightly northward resulting in 
stronger winds and lower rainfall (Alory and Delcroix 1999). 

Interannual and decadal-scale variability of winds and many other aspects of climate within the Samoan 
Archipelago are associated with the El Nino and Southern Oscillation (ENSO) phenomenon (Alory and Del- 
croix 1999, Halpin et al. 2004) (see CPC website: http://www.cpc.noaa.gov). The Southern Oscillation is the 
change in atmospheric pressure between the eastern and the western regions of the South Pacific (Chowd- 
hury et al. 2007). The Southern Oscillation Index (SOI) measures the strength of the oscillation and is com- 






!*_ ■■■- 




ifiQF 



\*|.i 



in: 



\?\\k\ 



irtl;^ 



m ■> 



U&VA 



\\ 



<& 




*■ 





Wind Direction (Inwards] 



S 



W 



"SI 



Figure 2.1. Wind direction measured by the QuikSCAT satellite. Monthly averages are based on the years from 1999 to 2007. Key 
atmospheric features and wind vectors (black arrows) are labeled in the plot for January at upper left. Odd numbered months are 
displayed. EEZs of Samoa and American Samoa are outlined in the center of each map. 

puted from the difference in atmospheric pressure at Tahiti and Darwin, Australia. Sustained negative values 
of the SOI often indicate El Nino episodes which are characterized by a decrease in strength of the Trade 
Winds (Luick 2000) and warmer surface waters in the equatorial Pacific (Vecchi and Wittenberg 2010). This 
shifts the SPCZ to the north and coincides with higher winds in the Samoan region (Alory and Delcroix 1999). 
Positive SOI values indicate La Nina episodes where equatorial Trade Winds are strengthened. A time series 
of SOI values is provided for reference alongside plots of several variables in the assessment including SST, 
sea surface height anomaly (SSHA), and chlorophyll to demonstrate its relationship with ocean climate in the 
Samoan Archipelago. 



Cyclonic storms (also called tropical storms, hurricanes or typhoons elsewhere) are infrequent but severe 
departures from the typical wind climate described above. The Samoan EEZs lie along the eastern edge of 
a region conducive to development of cyclonic storms in the South Pacific (Craig 2009). Six cyclones have 
struck or passed near the Samoan Archipelago in the past 30 years including 2 recent and very powerful 
Category 5 storms with sustained winds over 155 mph (Figure 2.2). In 2004, the eye of Heta passed south 
of the archipelago coming within 150 km of Savai'i (Fenner et al. 2008) creating a 0.3 m storm surge (SPSL- 
CMP 2007) and variable damage to Samoan reefs (Tausa and Samuelu 2004). In 2005, Olaf passed through 
the middle of the Samoan EEZs from northwest to southeast going almost directly over the Manu'a Islands 
where it caused substantial damage to both terrestrial and marine resources. 

Waves 

The wave climate of the Samoan Archipelago has been characterized extensively through Waverider buoys 
(Barstow and Haug 1994), wave and tide recorders (Brainard and others 2008), models such as NOAA 
Wavewatch III (Tolman 2010), and satellite altimetry (Barstow and Haug 1994). Wave power exposures are 



typically highest on the eastern and southern facing coasts of Samoan islands but can vary seasonally and 
among years (Barstow and Haug 1994). The wave climate can be split into two main components, short 
period (-2-10 seconds) "wind seas" that result from local forces such as the easterly Trade Winds versus 
long period (-10-20 seconds) "ocean swells" that originate from storms many of which are far south of the 
archipelago (Barstow and Haug 1994). Ocean swell from the south and wave power in general are highest 
during May-September (2-3 m wave height is common) with the increased intensity of the Trade Winds and 
frequency of swell producing storms at higher latitudes (Barstow and Haug 1994, Brainard et al. 2008). No- 
vember through March is a period often characterized by shorter period waves, lower wave heights (-2 m), 
and more variable directionality (Brainard et al. 2008). Although correlations with the SOI are somewhat ir- 
regular as with other variables, there is some evidence that El Nino conditions increase wave height (Barstow 
and Haug 1994). In contrast to the typical seasonal and interannual patterns, anomalous wave events occur 
due to tsunamis (Roeber et al 2010), the passage of cyclones (Militello et al 2003) (e.g. >8 m wave heights 
were recorded during Cyclone Ofa in 1990 and Heta in 2004) and even storms in the North Pacific which can 
cause unusually large swells on the relatively more calm northern coasts of the islands (Barstow and Haug 
1994, Brainard and others 2008). 



****** * 




£',£ 




/ 



s 



^Gtr 




f«S 



'Ct 




***** 



fl» 



Hurricanes 

2000-2010 
Category 




» fUtan 



■fig U 

/ 

f 

f 

t 
\ 

\ 




\ 



L_ 




Ocean circulation 

At the broadest scale, the Samoan Archipelago lies along the northern edge of the South Pacific Gyre, a 
series of connected ocean currents with a counter-clockwise flow (Alory and Delcroix 1999, Tomczak and 
Godfrey 2003, McClain et al. 2004, Craig 2009) (Figure 2.3). At a regional scale, there are 2 major surface 
currents affecting the archipelago (Qiu and Chen 2004): (1) the westward flowing South Equatorial Current 
(SEC), and (2) the eastward flowing South Equatorial Counter Current (SECC) (Figure 2.3). The intensity of 
these currents in Samoan waters is variable among seasons and years. 




Figure 2.2. Path and intensity of cyclones passing through the EEZs of Samoa or American Samoa from 2000-2007. 



Figure 2.3. Major surface currents of the Southern Pacific Ocean adapted from Tomczak and Godfrey (2003). EEZs of Samoa and 
American Samoa are outlined in the center of the map. 




Current patterns are major influences on larval 
transport and connectivity among islands in the 
Samoan Archipelago and adjacent island na- 
tions (Treml et al. 2008). Finer-scale patterns in 
currents and implications for larval transport will 
be discussed in greater detail in Chapter 3. 



'mmm 




Ocean temperature 

Sea surface temperature (SST) data are col- 
lected globally by the NOAA/NASA Advanced 
Very High Resolution Radiometer (AVHRR) 
Oceans Pathfinder Program, which measures 
and reprocesses sea surface temperature, 
global cloud cover, vegetation cover and other 
variables. These data are the basis for the Cor- 
al Reef Temperature Anomaly Database, which 
has produced weekly average SST estimates 
for a 20 year period (January 1985 to Decem- 
ber 2005) at a resolution of 4 km (Selig 2008). 
Monthly averaged data for the entire time period 
were used to discern the broad SST patterns in 
the South Pacific, place the SST of the Samoan 
Archipelago into context, depict changes in SST 
within the Samoan EEZs over an average an- 
nual cycle, and identify anomalous or unusually 

high or low SST events in the waters of the archipelago. Continuous water temperature data have recently 
been recorded by data loggers deployed at several near shore locations around American Samoa by NOAA's 
Coral Reef Ecosystem Division. These data are discussed in detail by Brainard and others (2008) and pro- 
vide an important record of localized temperature variability. 

At the edge of the equatorial Pacific warm water pool, the entire Samoan Archipelago experiences relatively 
high and stable ocean temperatures throughout a typical annual cycle (Figure 2.4). Average SST ranges ap- 
proximately 2° C from a low of 27.2° C in August to a high of 29.5° C in March. Maximum SST occurs three 
months behind the maximum sunlight intensity, which indicates that SST increases as long as the intensity 
of sunlight is higher than its mean annual value (Alory and Delcroix 1999). Regional maps of monthly mean 



SST reveal gradual seasonal patterns (Figure 2.5). On average there is a -1° C SST range latitudinally in 
the American Samoa EEZ in any given season such that waters around Swains Island are 0.5 - 1 ° C warmer 
than those around the rest of the Samoan Islands. There is minimal longitudinal variation in SST. Sea sur- 
face temperature fronts that frequently occur at higher latitudes and are associated with enhanced biological 
productivity (Polovina et al. 2001) are essentially absent from the Samoan EEZs (Figure 2.5). 

The 21 year time series of available SST and temperature anomaly data revealed both seasonal and more 
irregular patterns (Figure 2.6). Overall trends in SST within the American Samoa EEZ exhibit an increase 
of -1° C from 1985 through 2006 (p < 0.0003, R 2 = 0.05, SST = 28.1 + 0.0023*month). All years since the 
major El Nino of 1997-1998 showed generally positive SST anomalies in the Samoan Archipelago, indicating 
warmer than average conditions. 



Image 4. Bleached acropora. Photo: D. Fenner, ASDMWR. 





liCrt i?Dt lnr i rii 






V 



^m 



MO W 

I 



^^m 



Jan 



Feb 



Mar 



Apr May 



Jun 



Jul 



Aug Sep 



Oct 



Nov 



Dec 




-American Samoa EEZ 



-Tutuila 



Savaii 



Figure 2.4. Sea surface temperature data from CoRTAD presented as an average annual cycle. Monthly averages are based on 
data from 1985 to 2006. Colors denote average values for the EEZ of American Samoa (red), waters adjacent to Tutuila (blue), and 
Savai'i (green). 



StfP Surf ncfl tfln^atrafluf ■ (Cp 

Figure 2.5. Sea surface temperatures from CoRTAD. Monthly averages are based on the years 1985 to 2006. EEZs of Samoa and 
American Samoa are outlined in the center of each map. 




Southern Oscillation 



| ^A^^vyw^Y^^^v^ 



i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i | i i i 
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 




Sea Surface Temperature 




Figure 2.6. Sea surface temperature and anomaly values from CoRTAD for the years 1985 to 2006. Values are monthly averages 
for the EEZ of American Samoa. Southern Oscillation Index (SOI) values for the same time period are from NOAA/NWS. El Nino 
conditions are represented in dark blue with strong negative SOI values. La Nina conditions are represented by orange with strong 
positive values. 

The seasonal SST range of 1-3° C is evident in all years. The specific range and temperature extremes in any 
given year are affected in part by the Southern Oscillation (Alory and Delcroix 1999). This is shown by focus- 
ing on winter (July) temperature anomalies, when the effects of El Nino are most pronounced in the region. 
During strong El Nino years significant warming of ocean temperatures occurs in the equatorial central/east- 
ern Pacific (Chowdhury et al. 2007) and generally cooler SST conditions occur in the Samoan Archipelago 
(Alory and Delcroix 1999, Fenner et al. 2008). Temperature minima generally occur during peak negative 
SOI values but begin several months prior (Alory and Delcroix 1 999)(Figure 2.6). For example, July 1 987 and 
1997 featured persistent SOI values of less than -2, indicative of strong El Nino conditions. SST within the 
Samoan EEZs during this time was 26.5°C, approximately 1 °C cooler than average during both years (Figure 
2.7). With one notable exception (1998), during La Nina conditions (e.g. 1989, 1999, 2000, 2001) the July 
anomalies show cooler SST along the equator but little change from expected values in the Samoan EEZs. 
Inspection of other SOI patterns reveals that the SST/atmospheric interactions are complex and do not yield 
perfect correlations. Of note is the strong negative SST anomaly in the equatorial region in July 1998 that is 
coupled with strong negative values in the Samoan Archipelago (Figure 2.7), a pattern not seen in the other 
19 years of available data. This strong negative Samoan SST is a potential lag effect from the 1997 El Nino, 
may be related to the rapid shift from El Nino to La Nina conditions during 1 998, and highlights the complexity 
and uncertainty of the effects of climate oscillations (SPSLCMP 2007). 

Water temperatures can also become too high for the corals on reefs in the Samoan Archipelago. Herma- 
typic, or reef building corals, require warm tropical water, however when ocean temperatures are higher than 
1° C above the highest temperature expected in the summer, corals can become stressed (Glynn and D'Croz 
1990). This temperature is called the "bleaching threshold" and if this elevated temperature persists for long 
enough or temperatures are especially high for even a short period of time, corals will expel their symbiotic 
algae (zooxanthellae) and appear white. Three recent major coral bleaching events have been documented 
in American Samoa (Craig 2009) with a severe event in 1994 (Goreau and Hayes 1994) and additional wide- 
spread events documented in the summers of 2002 (Fisk and Birkeland 2002) and 2003 (Fenner et al. 2008). 

Because the length of time the water temperatures are elevated plays a role in coral stress and bleaching, 
a metric called Degree Heating Weeks (DHW) is often used to highlight peak periods. It is a weekly metric 





^-E 






3TO3- 



£> 



L^ 



1004 



<£> 



2002 



.^ - 



2005 






S*d Surface Temperature Anurnal y (C) 



a 
_l_ 



Figure 2.7. Sea surface temperature anomalies from CoRTAD during the month of July for the years 1997 to 2005. EEZs of Samoa 
and American Samoa are outlined in the center of each map. 

calculated based on the number of the previous 12 weeks when the temperature exceeded the bleaching 
threshold as well as the number of degrees the temperature is above the bleaching threshold. Based on 
research conducted at NOAA's Coral Reef Watch (http://coralreefwatch.noaa.gov/), when the thermal stress 
reaches a value of 4 DHW, significant coral bleaching is likely. When thermal stress is 8 DHW or higher, 
widespread bleaching and mortality from the thermal stress is likely. 

Using the Coral Reef Temperature Anomaly Database (CoRTAD), DHW was calculated for 1985-2006 for 
waters near Savai'i and Tutuila (Figure 2.8). Elevated DHW (>4 degree C) occurred at irregular intervals with 



Southern Oscillation 



o 



£ -4- 
O 



^A^v*^\ 




I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 



Degree Heating Weeks 



4' 
3H 



o 





JJ U l. iJ U 




85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 



Savaii 
Tutuila 



Figure 2.8. Sea surface temperature anomaly plots for warmest water month. 



stronger peaks observed since 1990. Elevated DHW values generally occurred in Fall (April-May). Thermal 
stress exceeded 4 DHW in Savai'i and/or Tutuila in 1991, 1994, 2001, 2002, and 2003 with 3 of these years 
corresponding to documented coral bleaching events (Goreau and Hayes 1994, Fisk and Birkeland 2002, 
Craig 2009, Fenner et al. 2008). Unlike the SST anomaly time series, there appears to be minimal relation- 
ship between SOI and DHW in this region. Regional snapshots during peak DHW events in the time series 
reveal considerable variability in their spatial extent. A latitudinal but patchy band of increased DHW is evi- 
dent in the region at these times. 

NOAA Coral Reef Watch monitors bleaching conditions at a different site than those considered here and 
is based on waters off Ofu in the Manu'a Islands of American Samoa (Ofu virtual monitoring station- based 
on SST measured in a 50 km pixel centered at 14° S, 170° W). From 2000 to 2009 bleaching watches were 
issued by NOAA in all months between November and June with the core Summer/Fall months of January 
through May experiencing a watch at some time during nearly all nine years of data. Only in 2008 were SSTs 
consistently low and no bleaching watches were issued. Only one year during the nine years of data did SST 
at the Ofu site rise above the bleaching threshold. This occurred at times during three consecutive months 
from January to March of 2003. During this period, NOAA Coral Reef Watch issued 3 bleaching warnings and 
tracked a period of several weeks from late March through mid April where temperatures at Ofu surpassed 
the 4 DHW threshold for which significant coral bleaching is likely. Differences in DHW between the Ofu vir- 
tual monitoring site and the Savai'i and Tutuila sites examined here are likely due to the localized variability 
in water temperature that can occur in the regions (e.g. Figure 2.9). 

Overall, these data suggest that the coral reefs of the Samoan Archipelago have been subjected to thermal 
stress conditions in -1/3 of the last 15 years. An important caveat to interpretation however, is that localized 
water temperatures and other factors, primarily including shallow depths, wind conditions, incident light, and 
low circulation, may produce bleaching conditions in particular lagoons and reef flats that is not predicted by 
the satellite based approach used here. In fact, most of the in situ temperature loggers deployed on reefs 
around Tutuila and Swains Island recorded sustained temperatures 0.5 to 1° C higher than those based on 
satellites (Brainard and others 2008). In an extreme example, on one day in 2005 satellite measurements by 
NOAA Coral Reef Watch indicated an average SST of 30° C for an area that included an in situ temperature 




2/28/99 



4/1/01 



* a, fv 





■*£! 


5*. ' 


- m . 






■££ 






m 



4/2S/0 



5/4/03 



- 

■W i p -^ 9- f f.f" 



!5b 



V egree Heating Week (degree C) 



Figure 2.9. Bleaching alert time-series from NOAA Coral Reef Watch. 



logger which recorded a value of 34.9° C (Fagaitua, American Samoa) (Fenner et al. 2008). In contrast, in 
situ temperature loggers around Ta'u recorded values typically 1° C lower than those based on satellite- 
derived surface estimates (Brainard and others 2008). This indicates that localized bleaching events are not 
always well predicted or detected by satellite based monitoring. Annual coral bleaching from 2004-2008 has 
been documented in two lagoon pools near the airport on Tutuila (Fenner et al. 2008). Even for widespread 
bleaching events, the severity can vary widely across islands in the Samoan Archipelago (Fisk and Birkeland 
2002) with some corals and localities able to withstand thermal conditions typically associated with bleaching 
(Craig etal. 2001). 

Chlorophyll 

The ocean color dataset from the Sea-viewing Wide Field-of-View Sensor (SeaWiFS) satellite provided a 
ten-year (September 1997 to October 2007) dataset of estimated chlorophyll concentration, at 9-km spatial 
resolution. Chlorophyll a is the dominant pigment in marine photosynthetic organisms and measuring its con- 




Image 5. Red algal bloom in Pago Pago Harbor. 
Photo: D. Fenner , ASDMWR. 

centration in ocean waters provides one measure of nutrient input to surface waters and subsequent biologi- 
cal productivity. Chlorophyll a concentration, referred to simply as chlorophyll in this report, can be estimated 
using SeaWiFS color sensors. Monthly averaged data for this period were used to discern broad chlorophyll 
patterns in the South Pacific, place the Samoan Archipelago into context, depict changes in chlorophyll within 
the Samoan EEZs over an average annual cycle, and identify unusually high or low chlorophyll events in the 
waters of the archipelago. 

The entire archipelago shows low chlorophyll levels all year with very limited but discernable seasonal vari- 
ability (Figure 2.10) (Dandonneau et al. 2004, McClain et al. 2004). Chlorophyll averages range 0.02 to 
0.03 |jg/L from a low of 0.05 |jg/L in January to a high near 0.08 |jg/L in July. The archipelago lies at the 
northwestern edge of a distinct region of minimal oceanic productivity associated with the South Pacific Gyre 
(Figure 2.11) (Dandonneau et al. 2004, McClain et al. 2004), a region recently shown to be expanding at a 
rate of 1 .4% per year (Polovina et al. 2008). Slight chlorophyll increases are evident just north of the EEZ at 
approximately 8° S and to the southwest near the Islands of Fiji. 




Jan 



Feb 



Mar 



Apr May 



Jun 



Jul 



Aug Sep 



Oct 



Nov 



Dec 



-American Samoa EEZ 



-Tutuila 



Savaii 



Figure 2.10. Chlorophyll concentration estimated from SeaWiFS and presented as an average annual cycle. Monthly averages are 
based on the years 1998 to 2007. Colors denote average values for the EEZ of American Samoa (red), waters adjacent to Tutuila 
(blue), and Savai'i (green). 



The 10-year time series shows two short-lived episodic increases in chlorophyll (Figure 2.12). During fall- 
winter (April-June) of 2002 and 2005, subtle chlorophyll increases of 0.12 |jg/L to 0.2 |jg/L were evident in 
ocean waters near Savai'i and Tutuila, respectively but not in the ocean around the other smaller islands of 
the archipelago (Figure 2.13). Minimal change was evident when chlorophyll concentration was averaged 
for the entire EEZ of American Samoa during the same months. The limited spatial extent of these events is 



SeaWiFS Chlarcphy 1 1 {1 S97-2W7) Jan 

ISO E 17DE IBfT iraft 3BQW HITW UD M 



Mai 



■ w J _ " n i i w "D w i« w imi w md yy 





v 



Nov 




Chlorophyll (ugL) 

JL 

Figure 2.11. Chlorophyll concentration estimated from the SeaWiFS satellite. Monthly averages are based on the years 1998 to 
2007. Odd numbered months are displayed. EEZs of Samoa and American Samoa are outlined in the center of each map. 



Southern Oscillation 




Chlorophyll 



0.24— r 




r~-ooooooooo50NONON 



— American Samoa EEZ 

— Tutulia 

— Savaii 



Figure 2.12. Chlorophyll concentrations estimated from SeaWiFS for the years 1997 to 2005. Values are monthly averages for the 
EEZ of American Samoa (black), waters adjacent to Tutuila (blue), and Savai'i (red). Southern Oscillation Index (SOI) values for the 
same time period are from NOAA/NWS. El Nino conditions are represented in dark blue with strong negative SOI values. La Nina 
conditions are represented by orange with strong positive values. 

highlighted in regional plots for June 2002 and April 2005 (Figure 2.13). These episodic events are gener- 
ally not associated with large-scale climatic phenomenon. It is speculated that these chlorophyll anomalies 
are the result of increased eddy activity inside the EEZ (Barber et al. 1996, Foley et al. 1997, Strutton et al. 
2001). Enhanced westward circulation of the SEC on the north side of the EEZ and eastward circulation of 
the SECC during the fall-winter (March-June) time frame could promote eddies and meanders (see Chapter 
3). The resulting mixing could result in areas of localized nutrient enrichment and heightened chlorophyll 
(Domokos et al. 2007). It is also possible that topographically induced upwelling around the larger islands of 
the archipelago could cause the elevated chlorophyll signature (Brainard and others 2008). April is a time of 
high flow for the SECC, which flows directly across the Samoan Archipelago during some years and can be 
deflected by the larger islands. Another possibility is that rainfall runoff and associated terrestrial and human 
nutrient inputs at these times elevated the chlorophyll levels around the larger islands in the archipelago that 
have greater land area and higher human populations (Brainard and others 2008). 

Water samples collected within 1-2 km around the islands and atolls of American Samoa during February/ 
March 2006 were analyzed for chlorophyll and nutrient concentrations by NOAACRED (Brainard and others 
2008). These nearshore samples typically showed an order of magnitude higher concentration of chlorophyll 
than those measured by satellite farther offshore reported here. Nearshore water samples for the largest and 
most populated island Tutuila, showed the highest and most variable concentrations of chlorophyll (average 
of 0.7 MQ/L) with lower values for Ofu, Olosega, and Ta'u (averages of 0.3-0.4 MQ/L) and lower still values for 
Rose and Swains atolls (0.2-0.25 |jg/L). 

Despite some measureable seasonality and episodic, but very small, spikes in chlorophyll concentration, 
the oceanic waters of Samoan EEZs are nutrient poor and have low biological productivity year round. This 
results in Clearwater, deep light penetration, and conditions suitable for growth of the coral reef ecosystems 
that characterize the region. 



Sea surface height anomalies 

Anomalies in sea surface height are best under- 
stood in the context of tides and sea level trends. 
A positive trend in mean sea level of 2.07 mm/ 
year ± 0.90 (95% CI) is evident at Pago Pago 
from 1948 to the present (http://tidesandcurrents. 
noaa.gov/sltrends) (Figure 2.14). Sea level is 
also monitored in the region by the South Pacific 
Sea Level and Climate Monitoring Project (SPSL- 
CMP), which operates a SEAFRAME (Sea Level 
Fine Resolution Acoustic Measuring Equipment) 
gauge which measures sea level in Apia, Samoa. 
Data from this gauge shows a similar if not higher 
rate of sea level rise (4.9 mm/year) for the period 
1993-2007 although a longer time series of data 
is needed to establish a more reliable estimate at 
this site (SPSLCMP 2007). Tides in the archipela- 
go consist of two highs and lows daily with a mean 
range of 2.51 ft as measured at Pago Pago (http:// 
tidesandcurrents.noaa.gov/). Seismic events and 
the associated tsunami signals are also recorded 
on these instruments. 

The Archiving, Validation and Interpretation of 
Satellite Data in Oceanography (AVISO) Program 
merges sea surface height data from the Topex/ 
Poseidon (T/P), Jason-1/2, ERS-1/2 and ENVISAT 
satellites. Sea Surface Height Anomaly (SSHA) 
refers to vertical deviations from expected mean 
sea level. To calculate SSHAs, a map of estimat- 
ed mean sea level is created from T/P data for the 

Southern Oscillation 







, . §b 



r, 



June 2002 



'*%<& ... - -*>. . "*-"**r 









~ 



April 2005 



Chlorophyll Anomaly Iws/L) 



-Pi"-: 



Figure 2.13. Chlorophyll anomalies estimated from SeaWiFS for 
June 2002 and April 2005. EEZs of Samoa and American Samoa 
are outlined in the center of each map. 



<d 0- 



co 



Htfy 



<tKyf«^ 



I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 

1951 1956 1961 1966 1971 1976 1981 1986 1991 1996 2001 2006 

Sea Level 



1400 




I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 
1951 1956 1961 1966 1971 1976 1981 1986 1991 



I I I I I 
1996 



2001 



I I I I I 
2006 



Figure 2.14. Sea level values for Pago Pago, American Samoa from 1948 to 2008. Values are monthly averages. Data are from U. 
Hawaii Sea Level Center/National Oceanographic Data Center Joint Archive for Sea Level. Southern Oscillation Index (SOI) values 
for the same time period are from NOAA/NWS. El Nino conditions are represented in dark blue with strong negative SOI values. La 
Nina conditions are represented by orange with strong positive values. 




Image 6. Exposed corals on a reef flat, southern Tutuila. Photo: D. Fenner, ASDMWR. 

period January 1993 to December 1999 at a global scale. All deviations in sea surface height are presented 
in reference to mean sea level during this period (http://www.aviso.oceanobs.com/fileadmin/documents/data/ 
tools/hdbk_duacs.pdf). Observed deviations from this mean sea level were plotted and evaluated in several 
ways for the study region. A 1 5 year dataset (1 992-2006) of monthly mean SSHAs was obtained from AVISO 
and used to discern broad patterns of sea level fluctuation in the South Pacific, place SSHAs of the Samoan 
EEZs into context, depict seasonal and inter-annual patterns of SSHAs in the Samoan EEZs, and identify 
unusual or extreme observations of SSHAs and discuss their relevance to coral reef ecosystems. 

The entire Samoan Archipelago experiences similar changes in SSHA during a typical annual cycle. Anoma- 
lies are highest in winter (May-August) and lowest in summer (December-January) and have a range of only 
~4 cm (Figure 2.15). Note that all SSHAs in Figure 2.16 are positive and expressed relative to the global 
mean sea level calculated for 1993-1999. Maps of SSHA averaged by month reveal somewhat predictable 
spatial variations in sea level in the region. Elevated sea surface height anomalies on the northern side of 
the EEZ around Swains Islands (~ 5°S) are noted in March (circled red; Figure 2.16). A general southward 
shift of this anomaly across the archipelago in the band of elevated heights can be seen from March to May. 



< 

x 








Jan Feb Mar Apr May 



Jun 



Jul 



Aug Sep 



Oct 



Nov Dec 



-American Samoa EEZ 



•Tutuila 



Savaii 



Figure 2.15. Sea surface height anomalies from AVISO are presented as an average annual cycle. Monthly averages are based on 
the years 1993 to 2006. Colors denote average values for the EEZ of American Samoa (red), waters adjacent to Tutuila (blue), and 
Savai'i (green). Note that the scale is relative to the global mean sea level for the period 1993 to 1999. 



m 


ISO SSWft (1 B*2-»CT} jgn 

AC iJfl'E l#C WW 1WW 1MW UDW 




■ 1 


» * A 




^ 




m » *-*i A - ■' 






■■ 


www • _T_^_ ft m^ 


j_ 


.1 


* »^T ■* r V -. . 





May 








* • 





Stp 




c 


I 





Mgv 



C#rjrcd* naih»n 







Si? Svrfrc* HrtjpM Angmaly |cm) 



i 



Figure 2.16. Sea surface height anomalies from AVISO. Monthly averages are based on the years 1993 to 2006. Odd numbered 
months are displayed. EEZs of Samoa and American Samoa are outlined in the center of each map. 

These elevated heights represent the signature of the SEC and SECC and their corresponding seasonal 
shifts (Domokos et al. 2007). These SSHAs and currents dissipate and become less intense and defined 
from winter through spring (also see Currents Section). 

When monthly values are not averaged across years and instead are shown as a 15 year time series, sev- 
eral key patterns emerge. SSHAs in the Samoan Archipelago have a significant response to the Southern 
Oscillation (Alory and Delcroix 1 999). A major negative height anomaly occurred in the Samoan EEZs during 
the strong El Nino of 1998 (SPSLCMP 2007). Sea surface heights in the EEZ averaged 25 cm below normal 




Southern Oscillation 



~ 4- 



■S -2- 



— i ^f^rl 



= -4-S 

o _6^: 



Vv^M*^^^ 



20 



| I I I ( I I I | I I I | I I I | I I I I M I | I I I I ( I I ] I I I | I I I | 

93 94 95 96 97 98 99 00 01 02 03 

Sea Surface Height Anomaly 



ITT 



! I I I I I I 



T 

04 05 



TT 



06 



10- 
_20- 

<-10- 

c/) 
c/) 

-20 



-30 




VsAA 




I ' r F ' ' I r ' I ' r 1 ' ' F 

93 94 95 96 97 98 



99 



1 i ■ r r ■ ■ r ■ 

00 01 02 



" r T" r 

03 



04 



05 



r " r T 

06 



T - T 



— EEZ 



Figure 2.17. Sea surface height anomalies from AVISO for the years 1993 to 2006. Values are monthly averages for the EEZ of 
American Samoa. Southern Oscillation Index (SOI) values for the same time period are from NOAA/NWS. El Nino conditions are 
represented in dark blue with strong negative SOI values. La Nina conditions are represented by orange with strong positive values. 

during March -April, 1998 (Figure 2.17). This event was recorded by sea level gauges widely in the South Pa- 
cific (SPSLCMP 2007). Another noteworthy event correlated with an El Nino occurred during March of 2005, 
where vertical sea surface heights dropped 10 cm below mean sea level in the Samoan Archipelago. The 
very broad spatial extent of these negative height anomalies, well beyond the Samoan EEZs, is evident in 
SSHA plots for March 1998 (and see SPSLCMP 2007) and 2005 in comparison to the same month in other 
years (circled blue; Figure 2.18). An east-west band of lower SSHA, at its largest extent in March, affected 
the South Pacific from the central Cook Islands to the Solomon Islands, a distance of ~3, 000km. The area 
where sea level is most affected by El Nino corresponds closely with the SPCZ (Figure 2.1), because of 
changes in the strength and position of the Trade Winds and the SEC (Figure 2.3) (SPSLCMP 2007). Long- 
term sea level data for Pago Pago, American Samoa confirmed the SSHA events depicted in satellite altim- 
etry in 1998 and 2005 but also revealed sea level drops in 1954, 1958, 1966, 1973, 1978, 1983, 1987, and 
1992 (Figure 2.14). All of these low sea level anomalies correspond to documented El Nino events (Chowd- 
hury et al. 2007). Based on this time series, low water conditions can be expected in the Samoan EEZs every 
4-8 years with the frequency of the anomalies being directly related to the timing of negative SOI values with 
greatest height anomaly values occurring roughly eight months behind the SOI (Alory and Delcroix 1999). 
To determine if the magnitude of sea level drop was correlated with the strength of the El Nino (SOI) a lin- 
ear regression of SSHA versus SOI was conducted. For every documented El Nino event since 1954, the 
lowest SOI (standardized) value observed was obtained from the NOAA's Climate Prediction Center (http:// 
www.cpc.noaa.gov/). For each El Nino event, the corresponding maximum deviation from monthly mean sea 
level observed at Pago Pago (station 1770000) was obtained from NOAA's Center for Operational Oceano- 
graphic Products and Services (http://tidesandcurrents.noaa.gov). The results of the regression indicate a 
significant positive relationship between magnitude of negative sea level anomalies and El Nino strength (p 
< 0.001, R2=0.63, |sea level deviation|=0.034 + 0.081*SOI) (Figure 2.19), emphasizing the strong ocean- 
atmospheric interconnections for the region. This corroborates a lag correlation analysis that found a value 
of 0.58 between SOI and height anomalies at eight months following SOI minima (Alory and Delcroix 1999). 

The deviations from mean sea level reported here may at first seem to be of such low magnitude that they 
have little significance to coral reef ecosystems. The maximum deviation observed was for sea level to be 
-30 cm lower than the expected monthly average. In fact however, when coupled with the right environmen- 




1995 



1996 .,, 



\ '" . „ 5b 

r 



1997 



A 






5b 






1999 



— i5b 



*k 



c& 



2000 



i 



** 



2001 



5b 






2002 



# 



, ,5b 



fv ■ 



2003 



ifiS" 



* 




4¥ 




2005 



?'■** m 



2006 



fc 






5b 



5ea Surface Height Anomaly Temporal Deviation (cm) 




o 

i 



Figure 2.18. Sea surface height anomalies from AVISO during the month of March for the years 1993 to 2006. EEZs of Samoa and 
American Samoa are outlined in the center of each map. 

tal conditions, these low sea level events can have a major impact on coral communities of reef flats. Reef 
flats make up an extensive component of the coral ecosystems around most of the islands in the Samoan 
Archipelago. The vertical growth limit of corals on reef flats is closely tied to the height of low tides. Coral 
colonies in the reef flat zone have flattened tops that clearly demarcate water depths suitable for growth and 
survival during average SSH conditions. Coral colonies and branches within the same colony are observed 



to achieve a remarkably uniform vertical growth of within 
1-2 cm. An anomalous drop in sea level 30 cm below the 
lowest tide that the reef flat corals have established as 
their growth limit has indeed resulted in exposure of cor- 
als and fatal conditions for large areas of the reef flat. 



0.3 



X 0.2 



0.1 




SOI*-l 

Figure 2.19. Linear regression of maximum monthly sea 
level deviation versus SOI during El Nino events since 
1954. p < 0.001, R2=0.63, |sea level deviation|=0.034 + 
0.081*|SOI| 



A massive reef flat mortality event during which up to 84% 
of corals died due to sea surface height anomalies and the 
corresponding "extreme low tides" was documented by 
researchers in American Samoa in 1998 (Alison Green, 
unpublished data). Locally, Samoans refer to such events 
as "kaimasa", a term related to the odor from decaying 
coral. For an exposure event to occur, a low tide must 
be present at the same time as a low sea surface height 
anomaly. Factors reducing sea surface height document- 
ed by this report include inter-annual events associated 
with documented El Nino cases (~-30 cm magnitude) and 
seasonal patterns (~-4 cm magnitude). Severity and ex- 
tent of the event next depends on additional factors most 

likely related to incident light, winds, and waves. High solar radiation as measured by low cloud cover and 
high sun angle heats and desiccates corals exposed during periods of low sea surface height. Calm seas 
prevent wave swash from regularly covering, moistening, and cooling exposed reef flats. When all, or some 
critical number and intensity of these conditions are in phase, reef flat exposure and coral mortality could 
result. Satellite images presented here indicate that low sea surface height events can affect vast areas of 
the Pacific (Figure 2.18). Understanding and predicting the periodicity, extent, and severity of such events 
should be a focus of future research to inform coastal planning for reef flat habitats. 



CONCLUSIONS 

The islands and atolls of Samoa and Ameri- 
can Samoa are characterized by small sea- 
sonal fluctuations in ocean conditions and 
often much larger multiyear fluctuations in 
response to larger climatic cycles such as 
ENSO. The major source of variability is sea- 
sonal for winds, waves, and SST whereas 
chlorophyll and sea surface height are affect- 
ed more by interannual processes (Alory and 
Delcroix 1 999). Nearly all aspects of ocean cli- 
mate for the Samoan Archipelago vary much 
more significantly by latitude than by longi- 
tude. As expected, most variables examined 
here demonstrate that Samoa and American 
Samoa lie in essentially identical ocean cli- 
mates. Swains Island, located in the northern 
extremity of the American Samoa EEZ and 
not derived from the same volcanic hot spot 
as the Samoan island chain, is affected by 
slightly different oceanic or atmospheric fea- 
tures than the rest of the archipelago depend- 
ing on the year. This is influenced primarily 
by the latitudinal shifts in features such as 
the South Pacific Convergence Zone, South 
Equatorial Current, and the South Equatorial 
Counter Current. 




Image 7. Flat topped coral head. 
Photo credit: M. Anderson. 



Overall, the Samoan Archipelago lies in a region 
with relatively stable oceanic conditions compared 
to areas to the north and south. The Archipelago is 
generally unaffected by the more dynamic ocean 
fluctuations such as subtropical sea surface temper- 
ature fronts that occur farther south and equatorial 
upwelling to the north. This is demonstrated through 
time series plots of SST at the intersection of 1 70° W 
longitude (the approximate boundary between Sa- 
moa and American Samoa) and 0°, 5° S, 10° S, 15° 
S, 20° S, 25° S, and 30° S latitude (Figure 2.20). 
The Samoan EEZs lie roughly between 10 and 17° 
S latitude. The SST in this region has a much smaller 
range of values, fewer extreme fluctuations, and less 
interannual variability than SST observed at higher 
or lower latitudes. This relative stability of the Sa- 
moan coral reef environment is worth considering in 
the context of reef resiliency. The relative constancy 
of conditions may be among the factors that have 
allowed the very long-term survival and growth of 
some of the largest individual hermatypic coral colo- 
nies in the world such as those located off the island 
of Ta'u at the eastern extremity of the archipelago. 
Some Porites colonies are up to 41 m in circumfer- 
ence and 500-1,000 years old (Brainard and others 
2008, Brown et al. 2009). It is important to continue 
monitoring the oceanic characteristics of the region 
in response to global climate change (Chase and 
Veitayaki 1992, US EPA 2007, Vecchi and Soden 
2007, Young 2007, Barshis et al. 2010). The low 
chlorophyll and biological productivity values associ- 
ated with the South Pacific Gyre have recently been 
shown to be expanding at a rate of 1.4% per year 
(Polovina et al. 2008). ENSO activity and charac- 
teristics will continue to affect Samoan sea levels, 
the eastward expansion of the warm water pool, and 
long-term precipitation patterns in the region (Vec- 
chi and Wittenberg 2010). Sea level rise has obvious 
implications to a human population already crowd- 
ed into a narrow and often low lying coastal zone 
(Chase and Veitayaki 1992, Coral Reef Advisory 
Group 2007). Increasing water temperatures pose 
a serious threat to corals already living near their 
thermal tolerance (Craig et al. 2001, Barshis et al. 
2010). Given that the reefs of the archipelago have 
developed in a region with relatively stable condi- 
tions, oceanic anomalies or trends exacerbated by 
climate change may have greater affects on Samo- 
an reefs than in regions adapted to such perturba- 
tions (US EPA 2007, Barshis etal. 2010). 



0° (equator) 



32 

30 ; 
28: 

26 
24 

32 

30 ; 

28: 
26 
24 J 

32 

30: 

28 

26i 

24] 

32: 

30 ; 
28 
26 
24; 



5°S 



10°S 



15°S 



20° S 



25° S 



32 
30: 
28 
267 

24; 

32 
30 
28 
26 
24 



30° S 



1985 




1990 



1995 



2000 



2005 



Figure 2.20. Sea Surface Temperature from CoRTAD at the 
intersection of 170° W longitude (the approximate boundary be- 
tween Samoa and American Samoa) and 0°, 5° S, 1 0° S, 1 5° S, 
20° S, 25° S, and 30° S latitude respectively. 



ACKNOWLEDGEMENTS 



REFERENCES 



The authors wish to thank the following individuals for contributing valuable discussions, data, general infor- 
mation and review comments to drafts of this material: Kelley Anderson (The Climate Foundation), Russell 
Brainard (NOAA/NMFS/PIFSC/CRED), Jamison Gove (NOAA/NMFS/PIFSC/CRED), Alison Green (The Na- 
ture Conservancy), Clare Shelton (Coral Fellow, Coral Reef Advisory Group, American Samoa Department of 
Commerce), Oliver Vetter (NOAA/NMFS/PIFSC/CRED), and Phil Wiles (Oceanographer, American Samoa 
EPA). 



Alory, G. and T Delcroix. 1 999. Climatic variability in the vicinity of Wallis, Futuna, and Samoa islands (1 3°-1 5° S, 1 80°- 
170° W). Oceanologica Acta 22: 249-263. 

Barber, R.T., M.P. Sanderson, ST. Lindley, F. Chai, J. Newton, C.C. Trees, D.G. Foley, and F.P. Chavez. 1996. Primary 
productivity and its regulation in the equatorial Pacific during and following the 1991-92 El Nino. Deep-Sea Research II 
43: 933-970. 

Barshis, D.J., J.H. Stillman, R.D. Gates, R.J. Toonen, L.W. Smith, and C. Birkeland. 2010. Protein expression and ge- 
netic structure of the coral Pontes lobata in an environmentally extreme Samoan back reef: does host genotype limit 
phenotypic plasticity? Molecular Ecology 19: 1705-1720. 

Barstow, S.F. and O. Haug. 1994. The wave climate of Western Samoa. SOPAC Technical Report 204. 34 pp. 

Brainard R., and 25 others. 2008. Coral reef ecosystem monitoring report for American Samoa: 2002-2006. NOAA Spe- 
cial Report NMFS PIFSC. 472 pp. 

Brown, D.P, L. Basch, D. Barshis, Z. Forsman, D. Fenner, and J. Goldberg. 2009. American Samoa's island of giants: 
massive Pontes colonies at Ta'u island. Coral Reefs 28: 735. 

Chase, R. and J. Veitayaki. 1992. Implications of Climate Change and Sea Level Rise for Western Samoa. United Na- 
tions Environmental Project. Apia, Western Samoa. 43 pp. 

Chowdhury, M.R., PS. Chu, and T.A. Schroeder. 2007. ENSO and seasonal sea-level variability: A diagnostic discus- 
sion for the U.S. affiliated Pacific islands. Theoretical and Applied Climatology. 88: 213-224. 

Coral Reef Advisory Group. 2007. Potential Climate Change Impacts to American Samoa. February 12, 2007. American 
Samoa Governors Coral Reef Advisory Group. 8 pp. 

Craig, P., C. Birkland, and S. Belliveau. 2001. High temperatures tolerated by a diverse assemblage of shallow-water 
corals in American Samoa. Coral Reefs 20: 185-189. 

Craig, P. (editor). 2009. Natural history guide to American Samoa. 3rd Edition. National Park of American Samoa, De- 
partment of Marine and Wildlife Resources, and American Samoa Community College. Pago Pago, American Samoa. 
131 pp. 

Domokos, R., M.P. Seki, J.J. Polovina, and D.R. Hawn. 2007. Oceanographic investigation of the American Samoa 
albacore (Thunnus alalunga) habitat and longline fishing grounds. Fisheries Oceanography 16: 555-572. 

Dandonneau, Y, PY. Deschamps, J.M. Nicolas, H. Loisel, J. Blanchot, Y Montel, F Thieuleux, and G. Becu. 2004. 
Seasonal and interannual variability of ocean color and composition of phytoplankton communities in the North Atlantic, 
equatorial Pacific and South Pacific. Deep-Sea Research II 51: 303-318. 

Fenner, D., M. Speicher, S. Gulick, and 35 others. 2008. The State of Coral Reef Ecosystems of American Samoa. Pp. 
307-351 . In: The State of Coral Reef Ecosystems of the United States and Pacific Freely Associated States: 2008. Wad- 
dell, J.E. and A.M. Clarke (eds.). NOAA Technical Memorandum NOS NCCOS 73. NOAA/NCCOS Center for Coastal 
Monitoring and Assessment. Biogeography Team. Silver Spring, MD. 569 pp. 

Fisk, D. and C. Birkeland. 2002. Status of coral communities on the volcanic islands of American Samoa. Department 
of Marine and Wildlife Resources, Government of American Samoa. Pago Pago, American Samoa. 135 pp. 

Foley, D.G., T.D. Dickey, M.J. McPhaden, R.R. Bidigare, M.R. Lewis, R.T Barber, ST. Lindley, C. Garside, D.V. Manov, 
and J.D. McNeil. 1997. Longwaves and primary productivity variations in the equatorial Pacific at 0, 140 W. Deep Sea 
Research II 44: 1801-1826. 

Glynn, PW. and L. D'Croz. 1990. Experimental evidence for high temperature stress as the cause of El Nino coincident 
coral mortality. Coral Reefs 8: 181-191. 

Goreau, T.J. and R. Hayes. 1 994. Survey of coral reef bleaching in the South Central Pacific during 1 994: Report to the 
International Coral Reef Initiative. Global Coral Reef Alliance. Chappaqua, New York. 201 pp. 




Halpin, P.M., P.T. Strub, W.T. Peterson, and T.M. Baumgartner. 2004. An overview of interactions among oceanography, 
marine ecosystems, climatic and human disruptions along the eastern margins of the Pacific Ocean. Revista Chilena 
de Historia Natural 77: 371-409. 

Luick, J. 2000. Seasonal and interannual sea levels in the western Equatorial Pacific from Topex/Poseidon. Journal of 
Climate 13: 672-676. 

McClain, C.R., S.R. Signorini, and J.R. Christian. 2004. Subtropical gyre variability observed by ocean-color satellites. 
Deep-Sea Research Part II 51: 281-301. 

Merrill, J.T. 1989. Atmospheric long range transport to the Pacific Ocean. Chemical Oceanography 10: 15-50. 

Militello, A., N.W. Scheffner, and E.F. Thompson. 2003. Hurricane - induced stage-frequency relationships for the Terri- 
tory of American Samoa. Coastal and Hydraulics Laboratory. Technical Report CHL-98-33 Revised, 226 pp. 

Polovina, J.J., E. Howell, D.R. Kobayashi, and M.P Seki. 2001. The transition zone chlorophyll front, a dynamic global 
feature defining migration and forage habitat for marine resources. Progress in Oceanography 29:1670-1685. 

Polovina, J.J., E. Howell, and M. Abecassis. 2008. Ocean's least productive waters are expanding. Geophysical Re- 
search Letters. 35:L03618, doi:10.1029/2007GL031745, 2008. 

Qiu, B. and S. Chen. 2004. Seasonal modulations in the eddy field of the South Pacific Ocean. Journal of Physical 
Oceanography 34: 1515-1527. 

Roeber, V., Y. Yamazaki, and K.F. Cheung. 2010. Resonance and impact of the 2009 Samoa tsunami around Tutuila, 
American Samoa. Geophysical Research Letters 37:L21604 doi: 10.1 029/201 0GL04441 9. 

Selig, E.R. 2008. The Coral Reef Temperature Anomaly Database (CoRTAD) - Global, 4 km, Sea Surface Temperature 
and Related Thermal Stress Metrics for 1985-2005 (NODC Accession 0044419): University of North Carolina (UNC) - 
Chapel Hill. NOAA National Oceanographic Data Center, Silver Spring, Maryland. 

SPSLCMP (South Pacific Sea Level and Climate Monitoring Project). 2007. Pacific Country Report on Sea Level and 
Climate: Their Present State, Samoa. South Pacific Sea Level and Climate Monitoring Project, http://www.bom.gov.au/ 
pacificsealevel/ 36 pp. 

Strutton, P.G., J. P. Ryan, and F.P Chaves. 2001. Enhanced chlorophyll associated with tropical instability waves in the 
equatorial Pacific. Geophysical Research Letters 28: 2005-2008. 

Tausa, N. and J. Samuelu. 2004. Summary report on current status of coral reefs in Samoa after Cyclone Heta. Fisher- 
ies Division, Ministry of Agriculture and Fisheries. Apia, Samoa 5 pp. 

Timmermann, A., J. Oberhuber, A. Bacher, M. Esch, M. Latif, and E. Roeckner. 1999. Increased El Nino frequency in a 
climate model forced by future greenhouse warming. Nature 398: 694-697. 

Tolman, H.L. 2010. WAVEWATCH III (R) development best practices Ver. 0.1 . NOAA/ NWS / NCEP / MMAB Technical 
Note 286, 19 pp. 

Tomczak, M. and J.S. Godfrey. 2003. Regional Oceanography: an Introduction. 2nd improved edition. Daya Publishing 
House, Delhi. 390 pp. 

Treml, E.A., PA. Halpin, D.L. Urban, and L.F. Pratson. 2008. Modeling population connectivity by ocean currents, a 
graph theoretic approach for marine conservation. Landscape Ecology 23: 19-36. 

US EPA (United States Environmental Protection Agency). 2007. Climate change and interacting stressors: Implications 
for coral reef management in American Samoa. Global Change Research Program, National Center for Environmental 
Assessment, Washington DC; EPA/600/R-07/069 http://www.epa.gov/ncea 61 pp. 

Vecchi, G.A. and B.J. Soden. 2007. Global Warming and the Weakening of the Tropical Circulation. Journal of Climate 
20:4316-4340. 

Vecchi, G.A. and AT. Wittenberg. 2010. El Nino and our future climate: where do we stand? Wiley Interdisciplinary Re- 
views: Climate Change. DOI 10.1002/wcc.33. 

Young, W.J. 2007. Climate risk profile for Samoa. Samoa Meteorology Division. 26 pp. 




Ocean Currents and Larval Transport Among Islands and Shallow 
Seamounts of the Samoan Archipelago and Adjacent Island Nations 

Matthew S. Kendall 1 , Matthew Poti 2 , Timothy Wynne 3 , Brian P. Kinlan 2 and Laurie B. Bauer 2 

INTRODUCTION 

The biogeography of coral reef ecosystems in 
the Samoan Archipelago is shaped in part by 
the ocean currents which carry the eggs and 
larvae of marine biota to and from the islands 
and seamounts in the region (Craig and Brain- 
ard 2008). Many of the marine organisms that 
inhabit the coral reef ecosystems of the region 
posses a pelagic larval phase. This includes 
bony fish, broadcast spawning corals, giant 
clams, palolo worms, crown-of-thorns starfish 
and a diversity of other fauna that are sub- 
ject to transport by ocean currents for at least 
some portion of their larval life. The connectiv- 
ity among island populations that results from 
larval transport is important because it means 
that the ecology, conservation, and manage- 
ment of each place in the Samoan Archipelago 
is intricately linked to and dependent on deci- 
sions made at other locations (Gaines et al. 
2007, Christie et al. 2010). Even when man- 
agement efforts are focused on particular sites within the archipelago, information on connectivity patterns 
via larval exchange is necessary to achieve management and conservation planning goals (Gaines et al. 
2007, Cowen and Sponaugle 2009, Costello et al. 2010). This chapter provides a detailed characterization of 
ocean currents in and around the archipelago and discusses their potential influence on important sources of 
larvae for maintaining Samoan reef ecosystems, and the contribution of Samoan reefs to population replen- 
ishment throughout the regional ecosystem. 

There has been a recent proliferation of studies on reef connectivity and MPA resilience (Almany et al. 2009, 
Jones et al. 2009, McCook et al. 2009, Steneck et al. 2009, Sale et al. 2010). It has become clear that plan- 
ning a regional network of MPAs that are resilient to disturbance, whether natural or anthropogenic, is depen- 
dent upon an understanding of larval transport (Gaines et al. 2003, Shanks et al. 2003, Botsford et al. 2009, 
Planes et al. 2009). Sufficient larval sources must be protected and spaced appropriately such that network 
sites can successfully repopulate between disturbance events. In addition, the fates of larvae produced at 
network sites should be considered to understand the role of protected areas in maintaining the broader 
ecosystem (Botsford et al. 2009, Christie et al. 2010, Costello et al. 2010). 

A variety of factors can affect the transport of larvae among islands. Most obviously it is necessary to under- 
stand the speed, direction, and seasonality of the ocean currents by which larvae are transported. It is also 
necessary to understand how aspects of the larvae themselves can affect their transport. Size of source 
populations, timing of spawning, duration of the larval period, daily mortality rates, sensory and swimming 
capabilities, and even random chance arising from the turbulent nature of ocean flows can all affect the 
probability that larvae will be transported from a source island to a particular destination (Siegel et al. 2008, 
Cowen and Sponaugle 2009). 



Image 8. Mass recruitment event of Ctenochaetus striatus. 
Photo credit: Peter Craig, NPS. 



1 NOAA/NOS/NCCOS/CCMA Biogeography Branch 

2 NOAA/NOS/NCCOS/CCMA/Biogeography Branch and Consolidated Safety Services, Inc., 



Fairfax, VA, under NOAA 



Contract No. DG133C07NC0616 
3 NOAA/NOS/NCCOS/CCMA Coastal Oceanographic Assessment Status and Trends Branch 



The goals of this chapter were to: 

1. Quantify and describe regional ocean currents in the Samoan Archipelago, 

2. Model the potential transport pathways of virtual larvae among island sources throughout the region, 

3. Identify key sources and destinations of larvae for each island, and, 

4. Quantify the influence of various combinations of larval life history characteristics (e.g. larval longevity, 
daily mortality rate) on those connections. 

To achieve these goals, we have combined observational data on ocean currents with simulations of larval 
dispersal in a three-dimensional regional ocean model. 

METHODS 

The study region was centered on the islands and seamounts of the Samoan Archipelago but also included 
surrounding island nations of Tokelau, western Cook Islands, Niue, Tonga, and Wallis Island to understand 
regional connectivity (Figure 3.1). Two primary techniques were used to quantify currents and larval con- 
nections among islands and shallow seamounts: observational ocean current data from passive drifters and 
transport simulations of virtual larvae from a three-dimensional hydrodynamic model. 






B- 



fe" 



-, 



1 



t 



frll'1-MJ 



'- Fi*icujLi 

tori 



I 




*#¥■, ■ 



MbrtiHMBtnh 





Jcc-.i 






MM -4MB 

■ KH1 -BOH 
I »B*H 



T...-.>ii" , .| 



I 






!»I 



■■ 



u* 



— P — 

IB'ff 



■ ?i * 



tow 



Figure 3.1. Samoan Archipelago and surrounding islands depicted by the 9 km grid cells of the HYCOM hydrodynamic model. 
Dotted lines denote separate islands, seamounts, or island groups clustered for analysis. Red lines denote larval sources from 
American Samoa. Green lines denote sources from Samoa. Black lines denote all other sources. 



The NOAA Global Drifter Program uses satellites to track an extensive array of passively drifting drogues de- 
ployed at 15 m depth (NOAA Coral Reef Ecosystem Division [CRED], NOAA Global Drifter Program [GDP]- 
Surface Drifter Program 2010). Drifter position, speed, and heading are recorded every six hours and provide 
a detailed record of actual surface currents. These data were used to describe patterns of surface currents, 
ground truth the current vectors of the hydrodynamic model, and select realistic parameters for modeling 
larval transport. 

The Hybrid Coordinate Oceanographic Model (HYCOM) is a three dimensional hydrodynamic model (Bleck 
and Boudra 1981, Bleck and Benjamin 1993, Halliwell et al. 1998, Christie et al. 2010) with a horizontal 
resolution of 1/12 degree (approximately 9 by 9 km grid cell size), a 1 day time step, and is available for 
the period 2004-2009 in the study area. The surface/mixed layer of HYCOM (surface to -10 m depth) was 
used to map the currents in the study area and to model the movement of passive particles or "virtual larvae" 
of corals, invertebrates and reef fish. Current vector data were downloaded from the HYCOM consortium 
(http://opendap.org/download). The 9 km model resolution is not sufficient to capture localized eddy and con- 
vection currents very close to shore (Swearer et al. 1999, Harlan et al. 2002) and we therefore used it only to 
evaluate broader scales of larval transport among islands. Islands are represented in HYCOM as grid cells 
with null current vectors (black cells in Figure 3.1). These null cells were used as a land mask and blocked 
larval movement. 



The drifter data and hydrodynamic model were first used to describe the currents in and around the study 
area and then to model interisland connectivity of virtual coral and fish larvae with realistic parameters. Our 
general approach was to evaluate a range of model parameters that have the potential to affect dispersal 
(Leis 2007) rather than specifying model parameters for the life history and behaviors of a particular spe- 
cies. We instead examined how a wide range of combinations of larval durations, precompetency periods, 
swimming capabilities, and mortality rates reported in the scientific literature may impact the connectivity of 
Samoan reef ecosystems. 

HYCOM Validation 

Modeled currents from HYCOM were validated by comparing them with drifter data on daily and monthly tim- 
escales. To evaluate daily current vectors, latitudinal and longitudinal drifter velocities were compared to the 
corresponding model velocities using linear regression at 100 randomly chosen drifter dates/positions during 
a randomly chosen model year (2004) (e.g. Rudorff et al. 2009). To evaluate the modeled current vectors on 
longer timescales, average monthly current vectors for every model year were plotted and visually compared 
to the tracks of drifters present in the study area (216 total) during the corresponding month/year. 

General Current Patterns 

To describe seasonal and interannual patterns in ocean currents in the region, average HYCOM monthly 
current vectors for 2004 to 2009 were plotted at -36 km resolution. Monthly vector plots at this spatial reso- 
lution were suitable for this purpose and showed gradual transitions in the major current patterns. Modeled 
current vectors (e.g. Figures 3.2-3.3) were visually compared to drifter paths (e.g. Figure 3.4) during the 
same month/year to qualitatively check model accuracy. In addition, drifter data were analyzed separately to 
determine the influence of season, El Nino/Southern Oscillation (ENSO), and region within the study area on 
current heading, speed and potential transport distance. For this analysis, drifter data were categorized into 
several groupings, 1) winter (June-August) or summer (December-March) seasons with transition months 
excluded for simplicity, 2) El Nino (2005), La Nina (2008), or neutral years (2004, 2006, and 2007) based 
on the Southern Oscillation Index, and 3) by position in the study area relative to the typical position of the 
major currents described for the region. The South Equatorial Current occurs all year throughout the region 
but is interrupted in the latitudes between 9° S and 12° S by the South Equatorial Counter Current during the 
summer. The influence of these variables on drifter heading, speed, and transport distance were evaluated 
using ANOVA, multiple means comparisons, and histograms to understand the significant differences and 
sources of variability in currents among seasons, years, and positions in the study area. Median bearing and 
mean speed were calculated for each drifter by season, ENSO status, and region (South Equatorial Current 
or South Equatorial Counter Current) and used as response variables in the ANOVAs. The low number of 
drifters observations that fell completely within a given season, region, and ENSO year prevented use of the 



10S - 




15S - 



20S - 



I 



^ 



I 



^ 



i 






"* ^ v i, y -» 
^ * * -> -> 



^ 



i 



^^* 



v " V t. * " It M* * I 

r* \v^-^-» ^ ******** v\l * * « * * * * ****; /»**** 






* a " " ^ ^ * ' 

* «. * U? IS t^ te- 4r 







175W 



170W 



165W 



160W 



Figure 3.2. Example of regional current vectors. Average surface current vectors from HYCOM for 
February 2004. The South Equatorial Counter Current is evident as eastward vectors between 10 
and 12 S. 



10S -*' 



20S - 



I 



I 



,»** 



&- «• V k-4. 



<~, 4r4r4^<r- <r * 







^V & V If <r~ 



^^^> «- <- < 



_«-^" kr *r & te 

&r & u u it 



& *r 









4- <- * * *"^- IS V IS 

' V **" ^ IS t! & Z 

" r & V te le V 
****** 
m P if * t * 4. 

■ * 4: v ^ *■ ^ 4, 









**>*^^*^**7 1*171: ^ y * « ^ ^ - - -- - ~ * -- - * fc ' 






if if 



is is is 



if IS * It * 

ir if 4r is k 

or *r er 2Z^j<r 4r ^is if 
IS is 



is *• 



^<r- 

'e- ir 
i? IS~ 



IS IS 



It It 



, * it 



is \s 

^ It H g It 




-is- * * t k \\ ± > ^„ < e 4r ^^ <t ^ 

15S -v",V.v\W.\v.» 






•<r <- 



- ir * 

* * * * ^r 

* T « * . 

, ^ * A| * * 

* \ ^ 

r . " ^ K r * * :, 
" ^ \ K v 1 t. t. 

t« ^ •*- if if if is 



* *i 







•4 4 2*^ 
V V « " 



t*- ir- ir &r <r- •* * *t is ir <- ^~ ^ «■ 
* * lC le * ? * *f ^4><-f.?. <■ t M f" 

a ^ j 4 f M M H * 4^1 

A^V|^|^^^rv^-»-J^^.->>^A tVt A-r *^^y <^irHaii.i(.lf 
1 *-«-^/^»'»-> ->^^.-^ *•*-».»*■» < * * h < V^ *l v •tA-.^-tr^-u * 

<r « ► >| ^ J < » /» ->^^^^.^>-> -^ "* * ► A «• *. <p» * PV^I y*«ir«-«* f •< M 
if«-T=-<A-sk> 5 f 7f^ -» ~^ -»-> -» -4 V if * **" <-<-<- <-<•« ^ ^ ^ ^ •*• 1! 



V fc v • 






If «> A -» 
*» ^ 3f T 



1 I r 

175W 



1 I r 
170W 



1 I r 
165W 



160W 



1 m/s 



Figure 3.3. Tonga Trench Eddy. This clockwise eddy formed in September through December of 
2006, 2008, and 2009. Average current vectors for April 2009. 





1 1 1 1 I 1 1 1 1 1 1 1 1 1 1 1 


1 1 1 1 
Drifter Bearing 






North 
West • East 






South 


10°S- 


<# •. ..•- ' l \ 


6 










v, 


15°S- 














• 


s> 









20°S- 


T 1 1 1 ■— 1 1 1 1 1 1 1 1 1 1 1 1 


1 1 1 1 



160°W 

Figure 3.4. Example of typical drifter paths. January 2007 drifters shown (NOAA Global Drifter Program). 

multi-way ANOVA analysis for transport distance. Instead, gross and net transport distances were calculated 
for all drifters in the study area that were at large for 10, 20, 30, 50, and 100 days respectively, and plotted 
as means and histograms. Net displacement is simply the straight-line distance between the start and end 
points of a drifter after a specific number of days whereas gross displacement is the total length of the drifters 
path (sums of all path segments recorded every six hours). 

Larval Sources 

Virtual larvae were started at each of 20 island groups and shallow seamounts in the study area (Figure 3.1 ; 
Appendix A). Islands and seamounts very close together were aggregated into larger groups especially at 
the edges of the study region to simplify and focus presentation of the results on the Samoan Archipelago. 
Only islands in the northern extent of the Tonga chain and eastern portion of the Wallis chain were included 
as larval sources since preliminary analysis indicated that most larvae from farther out would quickly leave 
the study area westward toward Fiji. 

Larval production was scaled to the area of the insular shelf (or seamount) with potential coral reef habitat. 
This was defined as the area from shore to the 150 m isobath and was based on the approximate depth lim- 
its of both photic and meso-photic reef communities in the area (Mesophotic Coral Ecosystems 2010, Bare 
et al. 2010). Initially, a set of 10,000 larvae were placed randomly within each coastal grid cell in HYCOM. 
This number of larvae was then adjusted based on the proportion of the cell with bottom depths between 
and 1 50 m. For example, a cell comprised of 50% land, 20% water deeper than 1 50 m, and 30% water with 
depth between and 150 m, was assigned a total of 3000 larvae (30% of 10,000) (Figure 3.1). This provided 
a very large but computationally reasonable number of virtual larvae scaled to potential area of reef habitat 
that could be "spawned" or started moving in simulations by HYCOM currents on any date specified. 



Fish and corals do not spawn at random positions on the reef as modeled here and, in the case of some 
fish species, move to areas where currents sweep eggs and larvae rapidly away from coasts to avoid reef 
based predators. For example, many fish species exhibit a vertical "spawning rush" away from reefs toward 
the surface and position themselves in habitats conducive to broadcast spawning. Mature surgeon fish in 
American Samoa are most abundant on points and headlands where strong currents are found (Ochavillo et 
al. 2011) and spawning can often be observed in reef channels where water flows in a seaward direction as it 
drains the reef flat (Craig 1998). Such localized currents are not included at the scale of HYCOM and should 
be studied using models with greater spatial resolution. 

Mass spawning events and start dates of larval transport 

Many coral species and other organisms in Independent and American Samoa and nearby Fiji have been ob- 
served to spawn 5-7 nights after the full moons in either late October or early November (Mildner 1 987, Itano 
and Buckley 1988, Mildner 1991, Mundy and Green 1996). The presence of coral spawning slicks is used 
as a cue for harvest of the edible Palolo worm, Eunice viridis, (Craig 2009) which is also known to engage in 
a synchronous spawning event and has a well documented lunar timing in the Samoan Archipelago (Mundy 
and Green 1 996). This annual mass spawning event for coral was the focus of our simulations. Starting dates 
for each model year were identified as six days after the first full moon to occur after October 1 2. These dates 
were 3 November 2004, 23 October 2005, 11 November 2006, 1 November 2007, and 20 October 2008. It 
is recognized that some spawning can occur across several days or even following successive full moons in 
October and November. However, preliminary evaluations of the model indicated that transport patterns did 
not differ substantially when start dates were on successive days or even separated by as much as a month 
among various phases of the moon, a finding similar to other studies (James et al. 2002) and consistent with 
time-scales of variation in ocean current patterns within the region. 

In addition to the October-November mass spawning event, year round spawning has been observed for 
several fish species in American Samoa (Craig 1998). A locally important fisheries species, the surgeonfish 
Acanthurus lineatus, has peak spawning in November-January (Craig et al. 1997) which would result in simi- 
lar transport patterns to those reported here. However, to evaluate the inter-island connectivity associated 
with year-round spawning, additional start dates and modeling would be required, especially for islands in the 
region of the strongly seasonal South Equatorial Counter Current described below. 

Larval Transport and Model Uncertainty 

Virtual larvae began at random locations within each coastal grid cell and were moved in the direction and 
distance specified by the corresponding current vectors from HYCOM for that date and grid cell. The Gen- 
eral NOAA Operational Modeling Environment (GNOME v 1 .3.0) was used to track larvae for all simulations. 
Diffusion or random variability in larval paths originating from the same location is an important aspect of 
connectivity studies (Polovina et al. 1999, Cowen et al. 2000, Siegel et al. 2003, Kobayashi 2006, Chiswell 
and Booth 2008, Treml et al. 2008, Rudorff et al. 2009). GNOME provides both a deterministic "best guess" 
calculation of larval path that assumes no error in current vectors and also enables a controlled amount of 
random variability to be applied to vectors at each time step for a more stochastic path. Adding random vari- 
ability or uncertainty to larval transport results in a cloud of potential pathways even when larvae all start 
at the same date and grid cell. The cloud has variable density with more larvae occurring along paths with 
higher probability of occurrence but also regions with fewer larvae that indicate less probable larval paths. 

To identify an appropriate level of diffusion/uncertainty, actual drifter paths tracked by satellite were com- 
pared to the paths predicted by HYCOM for virtual larvae originating at the same date and location as the 
drifters. Only drifter segments that began near islands (within -30 km) were used for this analysis because 
the goal was to evaluate variability of transport paths originating from these features. A total of 58 drifters (n 
= 1 3 NOAA CRED, n = 45 NOAA GDP) met this criterion. The start date and position of drifters from a subset 
of model years were loaded into HYCOM as starting points for particle drift. To identify an appropriate level 
of variability in particle drift, separate model runs using 1000 larvae were performed using a range of random 
error values (10-50%) in current vectors. Model paths were compared to each drifter path for general agree- 
ment while recognizing that a given drifter represents only one possible track out of a wide potential distribu- 
tion that reflects variation in drift. Using only 10% uncertainty rarely produced a probability cloud of the 1000 



larvae that included the drifter paths. Using 
50% uncertainty nearly always encompassed 
drifter tracks and provided reasonable clouds 
of larval pathways which highlighted more 
likely tracks (those occurring more frequently 
had greater density) while also depicting less 
likely, but still possible, pathways. All subse- 
quent model runs were conducted using 50% 
random error in current vectors at each time 
step to reflect this realistic level of randomness 
in larval transport. 

Precompetency 

Spawned gametes, fertilized eggs and young 
larvae cannot immediately settle even if they 
encounter suitable habitat and instead must 
spend some time developing as plankton. Pre- 
competency is the term used to describe the 
planktonic phase prior to achieving a body form 
capable of settlement. For many reef fish spe- 
cies, settlement is documented over a range of 

larval ages (e.g. settlement marks recorded on otoliths at 14 to 21 days old). For a wide variety of reef fish 
species for which reasonable sample sizes are available it is evident that individuals begin to settle once 
60-90% of their maximum larval lifespan (see next section: Pelagic Larval Duration) has elapsed (calculated 
from values in Victor 1986, Thresher et al. 1989, Wellington and Victor 1989, and Junker et al. 2006). Pre- 
competency periods for coral larvae are less known and appear somewhat more variable (Harrison et al. 
1 984, Wilson and Harrison 1 998, Miller and Mundy 2003, Graham et al. 2008, Jones et al. 2009). To simulate 
this developmental period, virtual larvae in the present study were prevented from settlement until a minimum 
of 60% of their Pelagic Larval Duration (see next section) was completed. 




Image 9. A guttatus spawning aggregations at dusk. 
Photo credit: Peter Craig, NPS. 



Pelagic Larval Duration 

Pelagic larval duration (PLD) is defined as the period of development spent in the water column during which 
larvae are susceptible to transport by ocean currents. For many species, larvae simply die in the plankton at 
the end of their PLD if they lack a suitable settlement habitat or energy source. PLD is quite varied among 
coral reef organisms even within the same genus and can be hours, days, weeks, or months (e.g. Bonhom- 
me and Planes 2000, Blanco-Martin 2006, Junker et al. 2006, Graham et al. 2008). Within species there can 
be considerable variability in PLD as well (Wilson and Harrison 1998, McCormick 1999, Junker et al. 2006) 
with influences such as water temperature and availability of suitable settlement habitat (McCormick and 
Molony 1995, Munday et al. 2009). In addition, larvae can shorten or lengthen their competent time in the 
plankton through various mechanisms such as delaying or partly reversing metamorphosis until a suitable 
habitat is encountered (McCormick 1999, Richmond 1985), directed swimming faster than ambient currents 
for some portion of their larval phase (Leis and Carson-Ewart 2003), and entrainment into coastal eddies to 
avoid extensive offshore transport (Swearer et al. 1999, Harlan et al. 2002, Paris et al. 2007). 



Also of note, climate change and the associated rise in sea surface temperatures may accelerate larval 
development of many species, cause earlier reef seeking behaviors, and even increase larval swimming 
efficiency (Wilson and Harrison 1998, Munday et al. 2009). All of these factors may serve to shorten PLDs 
overall, making it important to simulate connectivity over a range of PLD values and shift predictions toward 
potentially shorter dispersal distances. 

Rather than modeling particular species with specific life history parameters, we used a wide range of PLDs. 
This enables a variety of species and their corresponding PLDs (and changes to PLD due to factors such as 
climate change) to be considered. We evaluated PLDs of 10, 20, 30, 50, and 100 days which encompasses 
the range of PLDs expected for a wide variety of the fish and coral species of the Samoan Archipelago and 




Image 10. A. triostegus and A. guttatus spawning aggregations at 
dusk. Photo credit: Peter Craig, NPS. 



adjacent island nations (e.g. summary tables 
in Bonhomme and Planes 2000, Blanco-Martin 
2006, Graham et al. 2008, Jones et al. 2009). 
Longer planktonic longevity is possible (e.g. 
>200 days for some coral species, Graham 
et al. 2008); however, preliminary analyses 
revealed that many tracked particles had left 
the study area after 100 days, and modeled 
trajectories probably have reduced accuracy 
for such long periods. Also of note, some spe- 
cies have very short larval period (minutes or 
hours) and are never subjected to the interis- 
land currents studied here. 

Buffer for 'Settlement Zones' 

Fish larvae can sense the reef from some dis- 
tance away and perform behaviors that can 
help them reach desirable settlement habitats 
including vertical migrations into current fields 
moving toward reefs or simply out-swimming 

the ambient currents they are embedded in (Leis 2002, 2006, Gerlach et al. 2007, Leis 2007,). For as much 
as 50% of their larval phase, some fish larvae may be capable of sustained directional swimming that is suffi- 
cient to overcome their treatment as merely passive particles in ocean currents (Leis and Carson-Ewart 2003, 
Fisher 2005, Leis 2006). Larval fish can probably sense odor plumes from appropriate settlement habitat at 
distances of several kilometers (Atema et al. 2002, Leis 2006) and orient their movements horizontally or 
vertically in the water column to increase their chances of reaching settlement habitat. Recent studies show 
that some reef fish larvae can swim impressive distances on the range of 10-50 km (Atema et al. 2002, Leis 
2002), although this swimming could not be intentionally directed toward a reef beyond the sensory zone in 
which the larvae could detect the reef. Although the precise distance at which fish larvae can begin to orient 
towards reefs and the effectiveness of this orientation against ambient currents are topics of active debate, 
it is clear that larvae need not rely exclusively on passive transport in currents to arrive precisely at a distant 
island and instead need simply to come within a "settlement zone" sufficiently close such that they can sense 
and swim to the settlement habitat. Consequently, recent researchers have used buffers around islands to 
represent this "settlement zone" ranging from 1 to >100 km depending on the species under investigation 
(Lugo-Fernandez et al. 2001 , James et al. 2002, Cowen et al. 2006, Chiswell and Booth 2008). 

In this study, we investigated a range of potential settlement zone distances including 9 km (the resolution 
of the coastal grid cells in HYCOM), 18 km, and 36 km to accommodate a wide spectrum of organisms and 
potential sensory and swimming capabilities. A buffer of each distance was calculated using the maximum 
depth contour for mesophotic reefs around all islands and shallow seamounts and was designated as a 
settlement zone. If a larvae passed into an island's settlement zone after its precompetency period it was 
considered to have successfully settled at that island. Preliminary analysis revealed that, while settlement 
rates were somewhat affected by buffer distance, the spatial patterns of connectivity were relatively unaf- 
fected. We therefore display results for only the 18 km settlement zone in this report. 

Mortality 

Larval mortality has a significant effect on successful transport especially over long distances and lengthy 
time periods. Daily mortality rates are affected by environmental conditions and can vary significantly. Cowen 
et al. (2000) reported a wide range of mortality estimates for fish larvae of 42 species ranging from 3 to 46% 
per day with a mode of 18%. At these rates it is clear that most larvae will die prior to reaching settlement 
habitat, especially for small islands or seamounts that don't produce many larvae to begin with. For example, 
for every 1000 larvae produced at a source, after 20 days in the plankton, approximately 544 would remain 
if daily mortality were 3%, 19 would remain if mortality were 18%, and less than 1 would remain if mortality 
were 46%. 



In the present study, we investigated the influence of a range of daily mortality rates including 3%, 18%, and 
46% to reflect the spectrum of known values (Cowen et al. 2000). For each daily mortality rate, these per- 
centages of the larval population were randomly selected and removed from the larvae remaining at each 
daily time step. 

Calculating Connectivity 

Results are first summarized for the entire study area as connectivity matrices that display the island/sea- 
mount sources and destinations of all the virtual larvae tracked in the study. Islands in the southern part of 
the Tonga and western Wallis Island chains were not modeled as larval sources but were still shown in the 
connectivity matrices as destinations. For each PLD and mortality rate, we counted the number of simulated 
larvae released at each source location that travelled to each of the possible destination locations, cumulated 
over all 5 model years. Matrix cells denote the proportion of larvae from each source (rows) that arrived at a 
given destination (columns). Rows thus sum to a number <= 100%. The fraction of larvae that are lost (that 
do not successfully settle at any one of the destinations in our study) is equal to 1 00% minus the sum of each 
row. Columns can sum to >100%, because it is possible for a high proportion of the larvae produced at sev- 
eral sources to travel to the same destination. Separate matrices are provided for each combination of PLD 
and daily mortality rate. Note that connectivity calculated in this way depicts the pattern of larval transport 
pathways, without considering the variation in number of larvae produced at each source that could occur as 
a result of large differences in the size of reef populations. 

Each of the islands and shallow seamounts within the Samoan Archipelago was examined in detail as a 
larval source and destination. Each island/seamount was characterized individually using a combination of 
plots of larval distribution by PLD, three dimensional graphs that display larval retention and how reliant each 
location is on outside larval sources for each PLD and daily mortality rate, and bar graphs showing larval 
sources and destinations by location. 

RESULTS AND DISCUSSION 



HYCOM Validation 

Current vectors from HYCOM showed 
good correspondence to actual drifter 
paths on both short (daily) and long 
(monthly) timescales. Drifter and mod- 
el velocities at 100 randomly chosen 
dates/positions were positively cor- 
related in both the longitudinal (p < 
0.0001) and latitudinal (p < 0.0001) 
directions based on linear regression. 
Monthly current vectors from HYCOM 
showed good general correspondence 
to drifter paths in visual comparisons. 
These findings indicate that the mod- 
eled current vectors from HYCOM are 
a reasonable representation of actual 
transport processes in the study area. 

Drifter heading, speed and transport 
distance 

The multiway ANOVAs and histogram 
analyses of drifter data indicated that 
current headings are significantly af- 
fected by season and region within 
the study area (Figures 3.5-3.6, Table 
3.1). Overall transport throughout the 



300 



West 



250 



200 



South 



150 



100 



East 



E 
N 


L 
N 


n 
a 


E 
N 


L 
N 


n 
a 


E 
N 


L 
N 


n 
a 


E 
N 


L 
N 


n 
a 


summer 


winter 


summer 


winter 


SEC 


SECC 



Figure 3.5. Box plots of median current headings by region, season, and ENSO 
conditions based on drifter data. EN = El Nino, LN = La Nina, na = neutral ENSO, 
SEC = South Equatorial Current, SECC = South Equatorial Counter Current. 



region was westward except in the region 
between -8° S and 12° S during summer, 
wherein drifter motion was nearly opposite 
with most heading ESE. This marks the re- 
gion of the South Equatorial Counter Cur- 
rent. ENSO was not a significant influence 
on drifter headings and contributed only a 
minor and indirect effect in that headings 
were more variable in La Nina and neu- 
tral ENSO years relative to the El Nino 
year. Current speeds were largely uni- 
form throughout the study region at 20 to 
30 cm/s with drifter speed not influenced 
significantly by season, region, or ENSO 
alone (Table 3.2, Figure 3.7). 

ENSO did not have a major effect on drift- 
ers, however it should be noted that only 
one El Nino year occurred in the analysis 
period. Ideally, many El Nino years could be 
included in the analysis and drifters among 
years would have greater independence. 
Unfortunately, the analysis was limited to 
the Southern Oscillation conditions that oc- 
curred during the time drifters have 
been deployed. Additional years 
of drifter data and model runs with 
varying ENSO conditions would be 
necessary to more fully understand 
the effects of ENSO on current pat- 
terns. 



Gross transport distance (path 
length of drifters) was positively 
and linearly related to time at large 
whereas net displacement distanc- 
es were 40-70% shorter and leveled 
off between 50 and 100 days as 
currents looped many drifters back 
closer to their starting points at lon- 
ger time scales (Figure 3.8). These 
results are consistent with the diffu- 
sive nature of dispersal by turbulent 
ocean currents (Okubo 1980). As a 
reference for dimensions in consid- 
ering the scale of larval connectivity 
in this region, it is useful to note that 
400 km is the approximate length 
of the sides of a triangle with ver- 
tices at Rose Atoll, the western tip 
of Savai'i, and Swains Island and 
includes all islands of the Samoan 
Archipelago. After -30 days, mean 
net transport of drifters was over 




SEC winter (269°) 



SECC position winter 



SECC summer (154°) 



SEC summer (225.5°) 



Figure 3.6. Median current headings by season and region based on drifter data. 



Table 3.1. Full factorial ANOVAon the effects of region, season, and ENSO status on 
median drifter heading. Significant effects are labeled according to the level of significance. 



Variables 


Df 


Sum Sq. 


Mean Sq. 


F value 


p value 


region 


1 


134,317 


134,317 


92.4 


<2.2e-16*** 


season 


1 


421,055 


421,055 


289.7 


<2.2e-16*** 


ENSO 


2 


2,182 


1,091 


0.8 


0.473 


region:season 


1 


41,411 


41,411 


28.5 


2.0e-07 *** 


region:ENSO 


2 


13,818 


6,909 


4.8 


0.009** 


season:ENSO 


2 


5,274 


2,637 


1.8 


0.165 


region:season:ENSO 


2 


612 


306 


0.2 


0.810 


Residuals 


266 


386,599 


1,453 







Signif. codes: p < 0.001 (***); p < 0.01 (**); and p < 0.05 (*) 
Residual standard error: 38.12 on 266 degrees of freedom 
Multiple R-squared: 0.62, Adjusted R-squared: 0.6 
F-statistic: 38.7 on 11 and 266 DF, p-value: < 2.2e-16 

Table 3.2. Full factorial ANOVA on the effects of region, season, and ENSO status on 
mean drifter speed. Significant effects are labeled according to the level of significance. 



Variables 


Df 


Sum Sq. 


Mean Sq. 


F value 


p value 


region 


1 


0.5 


0.51 


0.01 


0.914 


season 


1 


36.1 


36.13 


0.8 


0.366 


ENSO 


2 


160.0 


80.02 


1.8 


0.164 


region:season 


1 


6.3 


6.29 


0.1 


0.705 


region:ENSO 


2 


35.0 


17.48 


0.4 


0.672 


season:ENSO 


2 


857.7 


428.83 


9.7 


8.3e-05*** 


region:season:ENSO 


2 


459.7 


229.84 


5.2 


0.006** 


Residuals 


267 


11,765.2 


44.06 







400 km and gross transport of some indi- 
vidual drifters was over 1000 km indicating 
ample opportunity for connections among 
islands for larvae with a PLD > 30 days. 
In addition, histograms were used to char- 
acterize the spread of transport distances 
associated with each duration of drift. As 
expected, the longer drifters were at large, 
the greater the range and more extreme 
the tails in distribution of gross distance 
traveled (Figure 3.9a). For net distance 
however, there was little difference in dis- 
tance traveled between 50 and 100 days 
because many drifters actually looped back 
on currents or eddies and ended up not far 
from their starting points despite being at 
large for twice as long (Figure 3.9b). The 
difference in net and gross transport must 
be interpreted with care because the maxi- 
mum net transport distance possible was 
limited by the size of the study area (~ 1 300 
km wide) whereas gross distances essen- 
tially could be infinite. 




E 
N 


L 
N 


n 
a 


E 
N 


L 
N 


n 
a 


E 
N 


L 
N 


n 
a 


E 
N 


L 
N 


n 
a 


summer 


winter 


summer 


winter 


SEC 


SECC 



Figure 3.7. Box plots of median current speeds by region, season, and 
ENSO conditions based on drifter data. EN = El Nino, LN = La Nina, na = 
neutral ENSO, SEC = South Equatorial Current, SECC = South Equatorial 
Counter Current. 



Quantitative Description of Regional 
Ocean Currents 

At the broadest scale, the Samoan Archipel- 
ago lies along the northern edge of the South 

Pacific Gyre, a series of connected ocean currents with a counter-clockwise flow that spans the Pacific basin (Alory 
and Delcroix 1999, Tomczak and Godfrey 2003, Craig 2009) (Chapter 2 Figure 2.3). Based on analysis of the 
modeled current vectors and drifter tracks, 4 major surface currents or eddies were identified that affect the 
archipelago (Figure 3.10a-b). From north to south these are: 1) the westward flowing South Equatorial Cur- 
rent at the northern edge of the study area, 2) the eastward flowing South Equatorial Counter Current that 
seasonally bifurcates the surface flow of the South Equatorial Current, 3) a westward but meandering flow of 
the South Equatorial Current directly across the 
archipelago, and 4) a regularly occurring eddy 
south of the archipelago, hereafter referred to as 
the Tonga Trench Eddy due to its consistent posi- 
tion over this geologic feature. The heading, posi- 
tion, and strength, as well as yearly and seasonal 
variability for each of these four major current 
features in the study area are described below. 



200Q 



1500 



Signif. codes: p < 0.001 (***); p < 0.01 (**); and p < 0.05 (*) 
Residual standard error: 6.64 on 267 degrees of freedom 
Multiple R-squared: 0.11, Adjusted R-squared: 0.08 
F-statistic: 3.209 on 11 and 267 DF, p-value: 0.0004 



South Equatorial Current 
(north of the South Equatorial Counter Current) 
The northern edge of the South Pacific Gyre is 
the westward flowing South Equatorial Current 
(SEC). The SEC is visible as westward or south- 
southwestward vectors along the northern edge 
of the study area during most seasons and years 
(Figures 3.3 and 3.10). Typical velocity based on 
drifter data is -25 cm/s. Although its latitude is 
variable among seasons and years, this compo- 
nent of the SEC seldom extends far below -9° S 



1000 



™ 



-NttC WSpiiMflWflt 




« 



M 



100 



Figure 3.8. Mean gross and net displacement of drifters after 10, 20, 
30, 50, and 100 days. 




SIM 



m 



«% 



ffl% 



IK 



10% 



» 50 fry* 

Figure 3.9a. Gross displacement of individual drifters as a frequency histogram after 10, 20 ,30, 50 and 100 days at large. 



3» 



M* 




Zfff 



20% 



Figure 3.9b. Net displacement of individual drifters as a frequency histogram after 1 0, 20 ,30, 50 and 1 00 days at large. Grey shading 
depicts net transport distances greater than the maximum possibility as determined by the size of the study area (~1 ,300 km wide). 



or into the EEZ of American Samoa or Samoa (Kessler and Taft 1 987). Flow has been characterized as stron- 
gest from March to July and weakest from October to February (Kessler and Taft 1987), although this was 
not necessarily evident in the study region in recent years. A narrow region between -6° S and 9° S of mild 
eddy formation (relative to the southern half of the study area)(Qui and Chen 2004, Domokos et al. 2007) is 
apparent between the SEC and the opposite flowing South Equatorial Counter Current described next. 

South Equatorial Counter Current 

Embedded on the SEC at latitudes between -8° S and 12° S lies the eastward flowing South Equatorial 
Counter Current (SECC) (Kessler and Taft 1 987, Chen and Qiu, 2004, Domokos et al. 2007), the most clearly 
visualized current feature in the region (Figures 3.2, 3.4, and 3.10). The SECC is a shallow current that exists 
above the main thermocline at -200 m depth (Kessler and Taft 1 987, Chen and Qui 2004). Below the thermo- 
cline the SEC continues a weak westward flow (Kessler and Taft 1987). Swains Island in the northern EEZ of 
American Samoa often lies in the middle of the SECC current field. The eastern end of this current generally 



curls south between ~1 60° and 1 70° W and ultimately joins the southern component of the SEC headed west 
across the Samoan Archipelago (Chen and Qui 2004). The SECC is well developed in most years between 
October and April with peak speed and width observed during January and February in this region. Typical 
summer velocities based on drifters range from 22 to 30 cm/s and current heading is east-southeast (Figures 
3.5-3.7). The current is often dissipated or absent in May through September (Chen and Qui 2004, Qui and 
Chen 2004), as confirmed by meandering or even westward drifter tracks and lower drifter speeds of 1 9 to 25 
cm/s during these winter months (Figures 3.6 and 3.7). The signature of this current can however, be present 
yearlong as was observed in 2005, an El Nino year. Interestingly, the SECC was absent throughout 1982 
and 1984, neutral ENSO years that bounded the strong El Nino conditions of 1983 (Kessler and Taft 1987). 
Also of note, the SECC was also virtually absent throughout 2009, a period of La Nina conditions. These 
observations highlight the irregular correlations between ENSO and SECC and the need for additional study 
during various ENSO conditions. 




-> 



Tokelau ^ ) 



Tokelau 



-> 



-> 



•rial Cou 



Swains 



hEqu a to ri a I Co u"igjt e r C u rre m^ 



<- 



Wallis 





^< 

^" 0a ^rch/ pe/ag0 ^oseAtoN 



Figure 3.10a. Surface current patterns of the Samoan EEZs and surrounding region for October through April. The position of curled 
current vectors and meanders are highly variable and denote general patterns only. Patterns are based on data from HYCOM (2004- 
2009) and NOAA's Global Drifter Program (n=216). 




Figure 3.10b. Surface current patterns of the Samoan EEZs and surrounding region for May through September. The position of 
curled current vectors and meanders are highly variable and denote general patterns only. Patterns are based on data from HYCOM 
(2004-2009) and NOAA's Global Drifter Program (n=216). 



South Equatorial Current (south of the South Equatorial Counter Current) 

The SEC continues its westward flow south of the SECC between -13° S and 19° S, including all the islands 
of the Samoan Archipelago and the southern half of the study region. In contrast to the northern component 
of the SEC described above, the SEC in this region is characterized by many irregular meanders and ed- 
dies (Domokos et al. 2007). This overall westward but meandering flow pattern is apparent in current vectors 
for all years and seasons (Figures 3.2 and 3.10) and is confirmed by the overall westward tracks of drifters 
(mean heading of 225° in summer to 269° in winter) (Figures 3.4 - 3.6). Typical velocities are -25 cm/s. 
Much irregular eddy activity is apparent between the opposite flowing SECC and this component of the SEC 
(Domokos et al. 2007). 

Tonga Trench Eddy 

The last regularly occurring feature of note is a clockwise eddy (negative vorticity) centered at 16° S and 
172° W, a location south of the Samoan Islands and positioned approximately over the Tonga Trench (Figure 
3.3). The eddy was most common in September through December in 2006, 2008, and 2009, all of which 
correspond to mild or moderate La Nina conditions. Beginning in September 2008, the eddy was present 
throughout 2009. This feature is persistent and not to be confused with the relatively short-lived (-weekly) 
eddies and dynamics recently investigated by Domokos et al. (2007). 






E. a-.il ML \\lMd 

0J\ 



Pjhn.-rLliktJlEuJl 

in 



MM 




Figure 3.11. Proportion of virtual larvae in the study area that we re started from each island, seamount, or island group. Red shading 
denotes larval sources from American Samoa. Green shading denotes sources from Samoa. 



Larval Sources Table 3.3. Islands, seamounts, or island groups used as source locations in simula- 

Based on DOtential reef area the tions °^ re 9' onal larval connectivity, their corresponding potential reef area (0-150 m 
^ ' shelf), number of virtual larvae used in modeling, and the percentage of the simulated 

major sources Of larvae in the Study larval pool contributed by each source. Green shading denotes islands of Samoa. Red 
region are likely to be Upolu (-25% denotes islands and seamounts of American Samoa, 
of the total larvae), Savai'i (-15%), 
and the Wallis Island group (-15%), 
although many of the larvae origi- 
nating from this more diffuse group 
are quickly transported westward 
out of the study area (Figure 3.11, 
Table 3.3). Altogether the islands of 
Samoa and American Samoa were 
the source of over half the virtual lar- 
vae tracked in the study. 

Larval Connectivity: Overall Patterns 

The connectivity matrices reveal 
broad patterns of overall larval trans- 
port and several island groups with 
strong internal connectivity (Figure 
3.12). Beginning at the origin of the 
connectivity matrices it is clear that 
Wallis, Tonga, and the many small 
islands and seamounts associated 
with them are strongly interconnect- 
ed but contribute few larvae to the 
Samoan Archipelago or elsewhere 
in the study region. Their position in 
the western or southwestern region 
of the study area results in most of 
the larvae from them being gradu- 
ally transported westward out of the 

study region in the direction of Fiji. In fact, this is the reason that some of the islands in these groups were 
not included as sources in the analysis. The island nation south of American Samoa, Niue and its associated 
seamounts, is largely isolated from the other islands in the study region. This is due to Niue's long distance 
from both upstream and downstream islands. Connections with its nearest downstream neighbor, Tonga may 
have been stronger were it not for the frequent development of the Tonga Trench Eddy between these two 
island groups which entrained many larvae until their PLD elapsed. Upolu and Savai'i in Samoa are highly 
interconnected and also have a large export of larvae to the islands and seamounts to the west such as Wal- 
lis, Niuafo'ou (Tonga), and Tafahi (Tonga). Samoa exports a smaller proportion of its larvae eastward against 
the SEC toward American Samoa. The islands and seamounts of American Samoa are also internally con- 
nected. The pattern of transport is such that most of the larvae either settle at the source or are transported to 
destinations downstream to the west, including the islands of Samoa, but often not as far as Wallis Island and 
its associated seamounts and not in very high proportions. Swains Island and Tokelau are relatively isolated 
from the other islands in the study area and are the sources of only relatively minor larval contributions to the 
Samoan Archipelago via the return loop of the SECC. Islands and seamounts in the Cook Islands group (Su- 
warrow Atoll, Palmerston Atoll, and Pukapuka Atoll and Nassau) are largely isolated, even from each other, 
and show few larval connections with other sites considered here despite their generally upstream position 
in the SEC. These islands were simply too far away and produced too few larvae to be a significant larval 
source for even the Samoan Archipeago, the next islands downstream. 

Overall, the spatial pattern of connectivity was controlled by the SEC with islands to the east providing larvae 
to both themselves and islands to the west. Despite expectations that the eastward flowing SECC and its 
feedback loop to the SEC would carry larvae from sites such as Swains Island back along the Samoan 



Site 


Potential 

Reef Area 

(km 2 ) 


No. Larvae 
Spawned 


Percent 

Larvae 

Spawned 


Wallis and Seamounts 


993 


132,722 


16.2 


Niuafo'ou and Seamounts 


34 


4,606 


0.6 


Pasco and Toafilemu Seamounts 


279 


37,067 


4.5 


Tafahi and Seamounts 


400 


54,704 


6.7 


Niue 


285 


40,304 


4.9 


Savai'i South 


300 


39,793 


4.9 


Savai'i North 


600 


80,919 


9.9 


Upolu South 


571 


76,831 


9.4 


Upolu North 


949 


127,661 


15.6 


Tutuila 


576 


77,873 


9.5 


South Bank (Papatua Guyot) 


33 


4,481 


0.6 


East Bank (Tulaga Seamount) 


3 


448 


0.1 


Northeast Bank (Muli Guyot) 


10 


1,335 


0.2 


Manu'a Islands 


186 


25,182 


3.1 


Rose Atoll 


12 


1,564 


0.2 


Swains Island 


16 


2,133 


0.3 


Tokelau 


350 


45,644 


5.6 


Pukapuka and Nassau 


187 


24,507 


3.0 


Suwarrow Atoll 


203 


26,796 


3.3 


Palmerston Atoll 


90 


12,440 


1.5 


Total 


6,077 


817,010 


100.00 



MwTGmy-3Ntf J 



Mcrfpfif^ J0fc if J 



iMartirftrraff&tf 1 




if) 

CD 

-I— • 
CO 

o 

_C 

if) 
CD 
_Q 
CO 



_c 

TD 
CO 
_C 
C/) 

d 
o 

CO 
"-+- • 

en 
CD 






CD 
if) 

8 

o 
en 



CD 

in 

CO 
CD 



CD 
CO 

_co 

■D 
CD 

-i— • 

_co 

3 



c 
o 



o 

CO 



if) 

B 
"co 
o 

TD 

c 



CO 

o 

if) 

O 
O 

o 

00 

o 

8- 



O 

O CD 

™& 
.£« 

> o 

8 § 

° o 

CD "^ 

> CD 

O . 



CM 



CO £ 



O) 



uunos 



CD 

3 C 

o>_co 
17 to 



Archipelago, such connectivity patterns were not seen in the simulations. Swains is simply too far north in the 
SECC current field and there are simply too few larvae that can survive the long transport time necessary for 
current loops to carry them back to settlement sites. 

An important caveat to interpretation of matrix elements for islands close to the edges of the study area is that 
virtual larvae that hit the edge of the hydrodynamic model were lost to further transport. In the real world, ed- 
dies or current reversals may have eventually looped larvae back into the study area for possible settlement 
for larvae with longer PLDs. The importance of these lost trajectories is probably minor due to the compound- 
ing of daily larval mortality. Islands at the edge of our study extent may also play important larval source or 
destination roles with adjacent islands just outside of our study region that were not characterized here (e.g. 
Fiji). For these reasons matrix elements for islands near the edge of the study area must be interpreted only 
in the context of the study extent. Findings for islands in the core of the study area, American Samoa and 
Samoa are the most robust. 



Also important is that, whereas we tracked >800,000 virtual larvae, the real larval output of these marine 
systems is orders of magnitude higher. Blank cells in our matrices denote cases in which zero virtual larvae 
connected a source and destination, but had we used more virtual larvae in the simulation or in a real world 
spawning event involving many more larvae, some may have actually made the connection. For this reason, 
blank cells in our simulations should not be viewed as impossible connections but instead thought of as rela- 
tively less probable. 

The proportion of larvae traveling to a destination will be a function not only of source population locations, 
size, and ocean current patterns, but also of the size of the destination (specifically, the size of the destination 
island and its surrounding settlement zone). Since this is an inherent feature of the geography of the region, 
we do not standardize the connectivity values for size of the destination area, although this may be desirable 
for some population modeling purposes. 

Influence of PLD and Daily Mortality Rate 

Longer PLDs had three main effects on inter-island connectivity (Figure 3.12). First, the proportion of self 
seeding (fraction of larvae produced at a source location that settled at the same location) was reduced over- 
all since larvae were not competent to settle until they had been transported farther from sources. Second, 
connectivity with islands farther downstream increased noticeably after PLDs of 1 days. This was especially 
noticeable in seeding of Wallis, Niuafo'ou (Tonga), Tafahi (Tonga), and the nearby seamounts with larvae 
from American Samoa. Third, larvae with PLDs of 50 or 1 00 days could be transported nearly any place in the 
study area although in very low abundance provided that the mortality rate was low enough. This widespread 
potential for transport at long PLDs suggests that the low amount of connectivity needed to prevent species 
divergence by genetic drift is easily possible throughout the study area. This is especially true considering 
that our study reflects cumulative connectivity over only a five year period, which is short relative to the gen- 
eration time of many species in Samoan reef ecosystems. This widespread transport could also promote 
rapid (re-) colonization of unoccupied reefs, although the likelihood of colonization actually occurring when 
larvae arrive in low density is influenced by a variety of life history features (Kinlan et al. 2005). Widespread 
transport of organisms with longer PLDs could be important in understanding and predicting resilience to 
disturbance and responses to climate change of this regional ecosystem. 

The fate of long-lived larvae is highly dependent on mortality rates. For low to moderate daily mortality rates 
at PLDs up to 30 days, the islands involved in predicted larval exchange changed little, but mortality and PLD 
did affect the strength of the connection (cell color changed but pattern of empty cells in the matrix did not) 
(Figure 3.12). In contrast, higher mortality rates affected both the strength of the connections (cell color) as 
well as the spatial pattern of island connections (many more blank cells), especially at longer PLDs. At high 
levels of mortality only those islands that were large larval sources (e.g. Savai'i, Upolu, Wallis) or that were 
close together had any measureable connectivity. For the very longest PLDs and medium to high mortality 
rates considered, no larvae settled successfully at any islands (Figure 3.12; three matrices in lower right cor- 
ner). They all simply died before the end of these very long PLDs. This highlights the importance of better 
information on larval mortality, particularly for species with long PLDs. 



MHo'uMyihp 4 







Influence of Interannual Variability 

There was little difference in the drifter or 
current data among years, a pattern that 
was borne out in the simulations of larval 
transport. The one exception was in 2008, 
a year in which the SECC failed to devel- 
op. This had only a small effect on trans- 
port of larvae for most islands in the study 
region except for Swains, which lies in the 
middle of this current field. In most model 
years, larvae from Swains were quickly car- 
ried eastward in the SECC (Figure 3.13). In 
contrast, in 2008 the SEC persisted in its 
westward transport throughout the region 
and entrained Swains larvae in the opposite 
direction. 

Larval Connectivity: Islands/Seamounts 
within the Samoan Archipelago 

Each of the islands and shallow seamounts 
in the Samoan Archipelago are character- 
ized separately in the following sections. 
Moving from west to east along the archi- 
pelago and ending with Swains Island, each 
site is the focus of a detailed set of analyses 
to characterize its role as a larval source 
and destination in the region. The north and 
south shores of Savai'i and Upolu are each 

characterized separately due to their large size and slightly different patterns of larval connectivity. The 
Manu'a Islands (Ofu, Olosega, and Ta'u) are combined due to their small size and close proximity relative to 
the scale of the hydrodynamic model. 






p?l J, A 


^^^^^— ^^^^^- VJn ■!■■■ 


Hi ?M 940 


T» 1 M» 


| ■ aw * am ■ &x& 


2Q0T ■ £0-1X9 ■ Lend] 



iH-tt 



Figure 3.13. Transport of virtual larvae from Swains Island by model year. 
All PLDs shown as same color for each year. 



SAVAI'I (SOUTH COAST) 







- ■ - 






~f F » 



1 ' v l o 






o 



■^ 



% 



© 



O 



lrs-LV 



170" ft 



IS K Hortilrty ttey"' 



1G& W 



®0 



*" 



bo s °o 



fe 




cg§s 



® 



a 



- 1 — 



% 



© 







i. H 3h- W 



SM 2£C 



r.:;o 



TM 



iKjBnr*ie*4 
VQGQ 



"I.-- A 



•PLD -ID PLD-20 PLD -30 


PLD =50 • PLD-1DD 


I Land ] 18km wfllameni zones 





Figure 3.14. Position of virtual larvae from southern Savai'i for all model years by PLD. Upper and lower plots denote 3% and 18% 
daily mortality respectively. 



Destinations 

Southern Savai'i has a moderate area of potential reef compared to other islands and was the source of 5% 
of the larvae for the region (Figure 3.11, Table 3.3). Most larvae from Savai'i's south coast are transported to 
the southwest via the SEC and have reached Wallis and the northern islands and seamounts of the Tonga 
group (Niuafo'ou, Tafahi, etc.) after only 10 days (Figure 3.14). A smaller number of larvae are passed north- 
ward and then entrained into the eastward flowing SECC which carries them between Swains Island and the 
rest of the Samoan Archipelago after 30 to 50 days. 



a; 

E 



lfHK* 



B0% 



B0% 



1 

5 2p% 




PLD 



(days) 



3% 



1007c 



90% 



w 



60% 



W% 



20% 



0% 






10 




50 

p LDWaysj 



16% 



0^* 



Figure 3.15. External larval supply and local larval retention at Savai'i-South as a function of PLD and mortality rate. Top panel: 
Percent of simulated larvae settling at this site that were produced at other sites. Bottom panel: Percent of simulated larvae produced 
at this site that return here. 



A. Low 0a\ ly Mortality (1%) 



Destinations 



Lii. 



Max. PLO (days} 
3D 

^^1 100 




Sources 



.._ w -b k «llljj . 



a ii 



B. Medium Daily Mortality (1B%) 

a 100% • 
I 50% 



$ 



LU 






a if 



O- W 



Destinations 



# 0% - 1 * j*--*^« 



i, 



ii^ 



Sources 



it 



lCv . 




C High daily Mortality (40%) 
J 100% 

i stwt 

in 
Q 

* 0% - -— - 



I. 



De5lin3tioni5 




Spluw^ 




Figure 3.16. Destinations (and sources) of simulated larvae originating from (arriving at) Savai'i-South for low, medium, and high 
larval mortality rates. Shading of labels indicates core island groups: red, American Samoa; green, Samoa. ND= no data. 



Nearly 25% of the larvae spawned from southern Savai'i are retained there for the low mortality scenario and 
10 day PLD (Figure 3.15). Much smaller fractions of local production are retained for longer PLDs and higher 
mortality. As PLD is lengthened, an increasingly large fraction of the settling larvae end up at destinations 
to the west such as Wallis and the islands and seamounts of northern Tonga (Figure 3.16). Higher mortality 
rates result in only a small proportion of the larvae produced at southern Savai'i successfully returning there 
(Figure 3.15), but very small proportions successfully settle anyplace else either (Figure 3.16). This highlights 
the point that most larvae produced at Savai'i die without reaching suitable settlement habitat. 



Sources 

Southern Savai'i is reliant on outside larval sources for a relatively consistent -60-80% of its arriving larvae 
depending on PLD and daily mortality rate (Figure 3.15). Although many larvae that reach southern Savai'i 
come from southern Savai'i, large fractions also arrive from the northern and southern coasts of Upolu (Figure 
3.16). Notably, southern Savai'i receives over half of its larval supply from Upolu regardless of mortality rate 
and PLD. Were these major larval sources to be disturbed, recovery at Southern Savai'i could be relatively 
slow and reliant on the much smaller larval sources from Tutuila and the other islands of American Samoa. 



SAVAI'I (NORTH COAST) 




to 



■ o 



o 



?■•• 



% 



"9. 




.-*■• 



, ® 



© 



175-lV 



I TO" W 




1ES W 



'^ 



© 



O 



— i — 



% 



© 







1 



174-W 



i?5 Kfl 



r.:vO 



?M 



iKiidffAIHfl- 



•PLD -ID PLD-2Q PLD - 30 PLD = 5D »FLD=1M 

■ Land Q 1 8^ m wtiiflmem z&ne& 



Figure 3.17. Position of virtual larvae from northern Savai'i for all model years by PLD. Upper and lower plots denote 3% and 18% 
daily mortality respectively. 



Destinations 

Northern Savai'i has a large potential reef area and was among the larger larval sources in the study (Figure 
3.11, Table 3.3). Transport of larvae from Savai'i's north coast is similar to the south shore described previ- 
ously except the distribution is shifted northward (Figure 3.17). More larvae are transported westward toward 
Wallis and fewer are carried toward the southern part of the Tonga Chain compared to the larvae originating 
from the south shore. Also, many more larvae are entrained into the SECC and reach Swains after PLDs of 
20 to 50 days depending on the year. 



* 



3 




50 



3* 
IOC 46% j^*" 



Figure 3.18. External larval supply and local larval retention at Savai'i-North as a function of PLD and mortality rate. Top panel: Per- 
cent of simulated larvae settling at this site that were produced at other sites. Bottom panel: Percent of simulated larvae produced at 
this site that return here. 



M?K 



A. Low Daily Mortality (3%) 



I>crtir.™>n3 




ii 




Sources 




B. Medium Dally Martatity (1B%) 




DriF.^inaiiffns 



ii 




■Sources 



3 



a. i- i- 




C. High Duly Mortality \ *C%) 



lOOtt 




L>es(irt3lions 



j 



n 




ND 

s sis I* 



Sources 
I 

. i 



■/> a &* 2 i c 

i s s t J i 



^ 



DC 



|2i*jJ 



5 T. 



e 



3 J 



Figure 3.19. Destinations (and sources) of simulated larvae originating from (arriving at) Savai'i-North for low, medium, and high 
larval mortality rates. Shading of labels indicates core island groups: red, American Samoa; green, Samoa. ND= no data. 



Over 40% of the larvae spawned off northern Savai'i end up settling there for the low mortality and 10 day 
PLD scenarios (Figure 3.18). Progressively smaller fractions of local production are returned there for longer 
PLDs and higher mortality rates. Much smaller fractions of larvae settle elsewhere in Samoa and at sites to 
the west as compared to Southern Savai'i (Figure 3.19). 

Sources 

Northern Savai'i is reliant on outside larval sources for 40-70% of its arriving larvae depending on PLD and 
daily mortality rate (Figure 3.18). An increasing fraction of larvae come from elsewhere at longer PLDs. 
Larvae that reach northern Savai'i come primarily from the north sides of Savai'i and Upolu and to a lesser 
degree the south sides of these islands for all PLDs (Figure 3.19). Were the major larval sources disturbed, 
recovery of northern Savai'i would be reliant on the much smaller larval sources from Tutuila and the other 
islands of American Samoa. 



UPOLU (SOUTH COAST) 




tM KO 



KlkVTlGIMfl- 

10CO 



Ifff W 



•PLD -1(3 PLD -20 H_D-30 PLD = $D *PLD-10G 

■ Land Q 13km wtlismem zw& 



Destinations 

Upolu's south coast is a large source of larvae for the region (Figure 3.11 , Table 3.3). Most larvae are trans- 
ported westward or southwestward in the SEC (Figure 3.20). Some larvae slip northward between Upolu and 
Savai'i or around Savai'i and are entrained in the SECC which can ultimately send them eastward between 
Swains Island to the north and the rest of the Samoan Archipelago to the south after 20-50 days. 

Over 20% of larvae spawned at southern Upolu are retained there for the low mortality and 10 day PLD sce- 
nario (Figure 3.21 ). Much smaller fractions of local production are returned there for longer PLDs and higher 



s 

I 



100% 



so% 



60% 



g 40% 



« 20% 



0% 




50 



1B% 



100 4S% m^ 

0* 





100% 

eo% 

E tiTVK. 






C 40% 




3 








0<% 











SO 



3% 
18% j*J 
100 46% yttfP** 

0^ 



Figure 3.20. Position of virtual larvae from southern Upolu for all model years by PLD. Upper and lower plots denote 3% and 18% 
daily mortality respectively. 



Figure 3.21. External larval supply and local larval retention at Upolu-South as a function of PLD and mortality rate. Top panel: Per- 
cent of simulated larvae settling at this site that were produced at other sites. Bottom panel: Percent of simulated larvae produced at 
this site that return here. 



A, Low Daily Mortality {3%} 



Designations 



SR% 



D% 



lG0*/= 



50% 



0% 



J I -Lll ...J-Lfc. 



.iHiilLi-^^- _._ _ 




Sourr-fi^ 



mortality. A large proportion of the larvae from Upolu's south coast settle back on the Islands of Samoa espe- 
cially for short PLDs (Figure 3.22). Wallis and its neighboring seamounts as well as Niuafo'ou are significant 
destinations for PLDs of 20-50 days even under scenarios with moderate mortality. 

Sources 

Southern Upolu is reliant on outside larval sources for 15-70% of its arriving larvae depending on PLD and 
daily mortality rate (Figure 3.21). At short PLDs the area is seeded primarily from local production whereas 
larvae with longer PLDs arrive from outside sources. The fraction of larvae arriving from northern Upolu and 
Tutuila increases with PLD for scenarios with low or moderate mortality and demonstrates the importance of 
these larval sources for southern Upolu (Figure 3.22). 



I i 



„_ ND _ 

p m 4» 

nZ 4} H □ Z 

m q. t- ^ 

Z 




B. Medium Daily Mortality 418%) 
„ 100% 



Deslmfltioii& 



50% 



D% 



JLiiL 



Sources 




C. High Daily Mortality (46% ) 
a t«% 

■* .1 - 



Dominations 



a 
* o% 



it i . - 



z 



00% 



Sources 



50"tt 



=* 0% 




fl j £ f 



S S « | 

* 5 



MS 



:: L n 



£ 



Figure 3.22. Destinations (and sources) of simulated larvae originating from (arriving at) Upolu-South for low, medium, and high 
larval mortality rates. Shading of labels indicates core island groups: red, American Samoa; green, Samoa. ND= no data. 



UPOLU (NORTH COAST) 













irs-iv 



T 
170" W 



UHMDrtilitirctay' 1 



•.:■/- 



©q 



n.- 



o 
'-Ju 



o 







© 







ft 




~~ r — 



^> 



© 



© 



i*fr" W 



d f?5 z*a 



WQ 



J&G 



1. 000 



~x\- .■- 



•PLD -ID PLD-2Q H.D-30 PLD=5P • FLD-1M 
■ Lsnd □ 18km wttlefmsril zones 



Figure 3.23. Position of virtual larvae from northern Upolu for all model years by PLD. Upper and lower plots denote 3% and 18% 
daily mortality respectively. 



Destinations 

The northern coast of Upolu is the largest source of larvae in the Samoan Archipelago with more poten- 
tial reef area than all the islands and seamounts of American Samoa combined (Figure 3.11, Table 3.3). 
The pattern of dispersal is similar to the southern coast described previously but the distribution is shifted 
northward with more larvae entrained into the SECC (Figure 3.23). Most larvae are trajected westward or 
southwestward in the SEC and have reached Wallis and northern Tonga after only 10 days. Larvae entrained 
in the SECC are ultimately sent eastward between Swains Island to the north and the rest of the Samoan 
Archipelago to the south after 20-50 days. Some in the 50 day range have reached as far as Pukapuka and 
Nassau in the Cook Islands. 




50 



100 46% y^ 1 



1QD% 



sdh 




ecm 



4CrtS 



20% 



SO 



Figure 3.24. External larval supply and local larval retention at Upolu-North as a function of PLD and mortality rate. Top panel: Per- 
cent of simulated larvae settling at this site that were produced at other sites. Bottom panel: Percent of simulated larvae produced at 
this site that return here. 



A. Low Daily Morality {3%} 
p 100% 
I SOvi 

s 

$ 0% 



Deslmations 



uJl 




| 50% 

# 0% -rw— -* — ND 



Jl 



Sources. 



i i 






3« ^ w "*■ Li 
5 m £L I- h- 






S £ j2 J ~ J I 




B. Medium Daily Mortality f1<%) 



Deslmations 




„ li 






1QD% 



Soli-: ■■--.- 



50% 
0% 




C. High Daily Mortality (46%) 

^ 1M% 

I 50% 
I 

* 0% 



Destinations 



~A 



I 






100% 



50% 



# 0% 



Sources 



JJB 



„ 




s -; |? » i « 2 || || 5 s - j * 



E 



j 1*1 

i ftl" £ 5 «! * tfl 



U) 3 3 



J" 






Figure 3.25. Destinations (and sources) of simulated larvae originating from (arriving at) Upolu-North for low, medium, and high 
larval mortality rates. Shading of labels indicates core island groups: red, American Samoa; green, Samoa. ND= no data. 



Over 40% of the larvae from Upolu's north coast return there for the short PLD and low mortality scenarios 
(Figure 3.24). Smaller fractions of local production are returned to northern Upolu for longer PLDs and higher 
mortality scenarios. A large proportion of larvae from northern Upolu settle at Savai'i for all mortality sce- 
narios (Figure 3.25). 



Sources 

Northern Upolu is reliant on outside larval sources for only 10-35% of its arriving larvae depending on PLD 
and daily mortality rate (Figure 3.24). The site is among the least reliant on outside larval sources since most 
of the successfully settling larvae here are produced locally for PLDs up to 50 days and scenarios with low or 
moderate mortality. The next highest source of larvae for northern Upolu is Tutuila although the proportion of 
larval supply is relatively minor compared to self-seeding (Figure 3.25). Were northern Upolu's larval produc- 
tion to be disturbed, it would be reliant on the smaller sources of Tutuila and the other islands of American 
Samoa for recovery. 



TUTUILA 




l?ft"W 



1^ w 



-,.K ^ 



D 1M ?*Q 



f:"C 



JM 



iKitdfflblHft 
10CO 



•PLD -10 PLD = 20 PLD =30 


PLD =50 • PLD-1DG 


H Lsnd Q 18km soniameni zones 





Figure 3.26. Position of virtual larvae from Tutuila for all model years by PLD. Upper and lower plots denote 3% and 18% daily mor- 
tality respectively. 



Destinations 

Tutuila has a large potential reef area relative to the other islands and seamounts in American Samoa but is 
comparatively small relative to Savai'i and Upolu as a source of larvae to the region (Figure 3.11, Table 3.3). 
Transport of larvae from Tutuila is westward in the SEC and splits into two groups along the north and south 
shores of Samoa (Figure 3.26). At PLDs of 10 days, larvae have just passed western Savai'i and begun to 
reach Tafahi in the northern end of the Tonga chain. By PLDs of 20-50 days larvae are well into Wallis and 
its neighboring seamounts farther west. Many larvae trajected south of Tutuila are entrained in the Tonga 
Trench Eddy. Larvae trajected north of Tutuila that are entrained into the SECC are quickly swept to the east 
and pass south of Swains Island with little settlement occurring there. 



100% 




50 



3% 

Q#1 




£0 



Figure 3.27. External larval supply and local larval retention at Tutuila as a function of PLD and mortality rate. Top panel: Percent 
of simulated larvae settling at this site that were produced at other sites. Bottom panel: Percent of simulated larvae produced at this 
site that return here. 



A. Low Daily Mortality (S%> 

~ 100% 

60% 

0% J - 1 — — ■ -- 



Dealinalions 



Jl 



Max PLD 


aaj-*: 






10 












20 






30 






SD 




100 



Sources 




B. Medium Daily Mortality (18%) 



□eslinadons 



ao* 



07= 



i 




. ^ . ^ - ._■ -l! J 1 J I L fc— ^_ 



ND 



i 



■£ i S| S> S v> z w z = 

£ .5 5 £ g z 1 ■■ a .3 I 



i. ; * £ 



I I lie £ S 



C. High Dally Mortality (46%) 



DeslinaEionrs 





,i 



L 




Sources 



JO. 



I 



I I II 



£ 




Figure 3.28. Destinations (and sources) of simulated larvae originating from (arriving at) Tutuila for low, medium, and high larval 
mortality rates. Shading of labels indicates core island groups: red, American Samoa; green, Samoa. ND= no data. 



Nearly 40% of the larvae spawned at Tutuila settle back there for the 1 day PLD and low mortality scenario. 
Much smaller fractions of local production are returned there for longer PLDs and higher mortality scenarios 
(Figure 3.27). Larvae that are exported settle primarily at northern Upolu, and to a lesser extent southern 
Upolu, northern Savai'i, and Wallis and the northern Tongan Islands for longer PLDs (Figure 3.28). 



Sources 

Tutuila is reliant on outside larval sources for 10-65% of its arriving larvae depending on PLD and daily mor- 
tality rate (Figure 3.27). Outside sources become proportionally more important at longer PLDs, especially 
when mortality is low. Larvae that reach Tutuila come primarily from Tutuila especially at short PLDs (Figure 
3.28). The Manu'a Islands, Upolu, and even Savai'i are the most notable sources of outside larvae and would 
be important to Tutuila's recovery if local larval production were disrupted. 



SOUTH BANK 




irt iv 



~& 



IBS' W 



1ft % Mertfty diy 



ID 
r.' 



S© 



0° O 
CD 



^ 







© 



— r — 



Sb 



— r — 



© 



G 



i?ar w 



1H J5D 



Mi 



7W 



iKikaricl^-s 



•PLD -ID PLD = 20 H_D=30 PLD=5D *PLD-1DG 
■ Land Q 18km setttetwrn zcm&s 



Figure 3.29. Position of virtual larvae from South Bank for all model years by PLD. Upper and lower plots denote 3% and 18% daily 
mortality respectively. 



Destinations 

South Bank (Papatua Guyot in Seamount Catalog, Koppers et al. 2010) is a relatively small guyot south of 
Tutuila and provides a very small contribution to the total larval pool of the region (Figure 3.11, Table 3.3). 
Most larvae are transported south and west from this site with very few slipping north between islands of 
the Samoan Archipelago (Figure 3.29). Transport is either westward in the SEC between Savai'i and Tafahi 
(Tonga), or southward into the Tonga Trench Eddy. 



I BOtt 
1 

* eOTfc 
8 

5 20% 
0% 



^ 



1 



5 



100V, 

aoH 

60% 
40tt 



20V, 



0% 







50 



Figure 3.30. External larval supply and local larval retention at South Bank as a function of PLD and mortality rate. Top panel: Percent 
of simulated larvae settling at this site that were produced at other sites. Bottom panel: Percent of simulated larvae produced at this 
site that return here. 



A. Low On i ly Mortality (1%) 




Deslinalions 



A j^iu A *_i!fcJ-LLl_ 




100% 



Sources 



50% 



OK HD -. ... , 

U IS SJ - 

3 [5 o. H h- 



B. Medium DaHy Mortality (18%) 



100% 




Destinations 




. u Jil. Li 



"■ ™~ ,r r- 






100% 






Sources 



jl j III b t _ ... 



Ill 

I Jail | 1 



LU 



3 * 5 w 

E 2 ° * 

i § i 1 



C. High Daily Mortality {46%} 




B * * 



to 



Deslmaiiorvfi 



ILL 



ICO'Si 



Sources 



50% 

0% -— - 



ND 



fa 8 5 

3 m D. K i- * 



J . II 



SSBJZrtZSc^ 



5 * ? g j J 

l|:ass|l|f||i 
S i - 1 « $ 3 ^°I| 



« 



Figure 3.31. Destinations (and sources) of simulated larvae originating from (arriving at) South Bank for low, medium, and high 
larval mortality rates. Shading of labels indicates core island groups: red, American Samoa; green, Samoa. ND= no data. 



In contrast to the other sites discussed so far, a very low proportion of larvae from South Bank are retained 
locally for any PLD or mortality scenario (Figure 3.30). Tutuila and Upolu receive a significant proportion of 
larvae in the 10-50 PLD ranges (Figure 3.31). Tafahi (Tonga), Wallis, and other sites near them receive a 
noticeable proportion of the larvae originating from South Bank at longer PLDs. Of note, recent field surveys 
of South Bank indicate that the seamount is relatively uncolonized by corals and reef fish (R. Brainard, NOAA 
CRED and D. Fenner, American Samoa DMWR pers. comm.), which could indicate recruitment limitation and 
is consistent with the chronically low larval supply predicted by our model runs. It also suggests that South 
Bank is even less of a larval source than estimated here based purely on the potential reef area inferred from 
bathymetry. 



Sources 

Unlike most other sites discussed thus far, South Bank is reliant on outside larval sources for a very large 
proportion, 80-95%, of its arriving larvae regardless of PLD or daily mortality rate (Figure 3.30). The major 
sources of larvae for South Bank are Tutuila and Manu'a (Figure 3.31). Were these primary sources dis- 
turbed, South Bank would rely primarily on Upolu as a larval source. 



EAST BANK 



1ft 



i« 



3 tt mortality dfty ' 



TT 



®0 



0_ 



CD" 



^ 







® 



lTE'LY 



m m 



2- 






fe 



It % Mortality day' 1 



O 

0° o 

CD 



TT 



© 



■o^i 




o 



© 



1?& _ W 



— r — 
it w 



% 











1 
WW 



Destinations 

East Bank (Tulaga seamount in Seamount Catalog, Koppers et al. 2010) is the shallow crest of a submerged 
ridge extending east from Tutuila. Due to its small reef area it is a very small contributor to the regional larval 
pool in this study (Figure 3.11, Table 3.3). Many larvae from East Bank are trajected westward in the SEC 
along the north and south shores of the Samoan Archipelago (Figure 3.32). Others are transported south- 
ward toward Niue but few arrive due to the long transport distance, larval mortality, and the low number of 
starting larvae spawned in the simulation. 



% 



© 



G 



1 



1H 2» 



530 



T5Q 



I KikHTwLn 
TXO 



•PLD -10 PLD = 20 H_D - 30 PLD =50 » PLD- 100 

■ Lard Q 18km Htfiiamem ewim 




20 sa 



10 



too 4e%'^^ 




so 



18* 






Figure 3.32. Position of virtual larvae from East Bank for all model years by PLD. Upper and lower plots denote 3% and 18% daily 
mortality respectively. 



Figure 3.33. External larval supply and local larval retention at East Bank as a function of PLD and mortality rate. Top panel: Percent 
of simulated larvae settling at this site that were produced at other sites. Bottom panel: Percent of simulated larvae produced at this 
site that return here. 



A. Low Dally Mortality (3%> 

" 100% 

C% J -I -^ -J -* ~ 



De&linalions 



— m j l J*i tli. Li ^ . j — 




Source* 



# 0% 



5 ■= * lit 




3 = 




B. Medium Daily Mortality (18%) 

£ 1W% 
£0% 



Destinations 



# 0% 



- - 



I 




Sources 



MD 

8 ■§, % £ 



.1 _ J 



II 



_l 



I s ii i 

5 1 £ i= iS 



mi i^^g 



111 



3 3 = = I I * 



C. High Daily Mortality (46%) 




Destinations 



3 



i_ W 

a e 

s5 



.ill 




Sources 



i 



§ i 



. 



f 



& 5 a * « z i 
II? 



15 



£ 5 

Xi m 



2 * * * 

* &■ m E 

fir fft 



Figure 3.34. Destinations (and sources) of simulated larvae originating from (arriving at) East Bank for low, medium, and high 
larval mortality rates. Shading of labels indicates core island groups: red, American Samoa; green, Samoa. ND= no data. 



East Bank has virtually no retention of locally produced larvae (Figure 3.33). A high proportion of larvae settle 
at South Bank at the 1 day PLD, Tutuila and Upolu at 20 to 30 day PLDs and islands farther west for longer 
PLDs (Figure 3.34). Note that longer PLDs, high mortality, and few larvae to begin with result in few strong 
connections west of Savai'i despite prevailing currents. 

Sources 

Similar to South Bank, East Bank is reliant on outside larval sources for a very large proportion, 95-100%, 
of its arriving larvae regardless of PLD or daily mortality rate (Figure 3.33). The majority of larvae settling at 
East Bank are from Manu'a for PLDs of 1 0-30 days (Figure 3.34). A large fraction of successful settlers come 
from Tutuila at 20 to 30 day PLDs especially at higher mortality rates given the relatively low number of larvae 
starting from Manu'a. Were these larval sources to be disturbed, East Bank would rely primarily on Upolu as 
a (much smaller) larval source. A small but measureable proportion of larvae are from Tokelau, well to the 
north, for the 100 day PLD with low mortality rate. 



NORTHEAST BANK 









fe1 



3 % mortality eby ' 



T 



®0 



a 



Qjn» 



DO ° ^b .^g^ 



® 




irt- tv 



i 

170" W 



v? 

* 



i- 



LA 



18 W Mortality day 



o 

Do ° ^ 



TT 



©0 



o 



CD 



# 



o 



o 




© 



o 



1?& _ W 



— r — 



% 



© 







liS" W 



<% 



© 



o 



— r — 



1H K& 



WO 



TM 



I Kitofr-E!'?' j 
1000 



•PLD -ID PLD = 20 PLD ^30 PLD = 50 • PLD-1M 
■ Land Q 13km sefllsmeni zanes 



Destinations 

Northeast Bank (Muli Guyot in Seamount Catalog, Koppers et al. 2010) is a small seamount between Tutuila 
and Manu'a. Due to its small size Northeast Bank is a very minor contributor to the total larval pool of the re- 
gion (Figure 3.11 , Table 3.3). Most larvae from Northeast Bank are trajected westward along the north side of 
the Samoan Archipelago (Figure 3.35). Many are quickly entrained in the eastward flowing SECC well south 
of Swains Island where they are largely dispersed and die in the open ocean between American Samoa and 
the Cook Islands. Many are also trajected south and expire in the region of the Tonga Trench Eddy prior to 
reaching Niue. 



| BOW 

3 

S 

£ eo% 

£ 40% 

« 20% 

0% 



10 20 




50 



3% 

100 4G* f^ 



1B% ^ 



100% 

ao% 



5 oov f 
I 

I 10% 

20% 

0% 




60 






Figure 3.35. Position of virtual larvae from Northeast Bank for all model years by PLD. Upper and lower plots denote 3% and 18% 
daily mortality respectively. 



Figure 3.36. External larval supply and local larval retention at Northeast Bank as a function of PLD and mortality rate. Top panel: 
Percent of simulated larvae settling at this site that were produced at other sites. Bottom panel: Percent of simulated larvae produced 
at this site that return here. 



A. Low Oally Mortality 45%) 




Designations 



J ,-J _1l.»L _J _ 



^.^^tnlLL.^ * 





Sources 



*_J .l 



* * 11 is* 




Northeast Bank has virtually no retention of locally produced larvae (Figure 3.36). For short PLDs of 10-30 
days, most larvae from Northeast Bank end up at South Bank, Tutuila, and the north coasts of Upolu and 
Savai'i (Figure 3.37). Longer PLDs of 30-50 days result in a more even dispersal of larvae spread along the 
islands farther west all the way to Wallis and its associated seamounts. At high mortality rates, the few larvae 
starting from Northeast Bank are largely dead after the 10 day PLD and no connections among islands were 
detected. 

Sources 

Like the other seamounts of American Samoa, Northeast Bank is reliant on outside larval sources for 90- 
1 00% of its arriving larvae regardless of PLD or daily mortality rate (Figure 3.36). The majority of larvae with a 
10 day PLD arrive at Northeast Bank from the Manu'a Islands (Figure 3.37). Interestingly, a significant source 
of larvae in the 30-50 day PLD range is northern Upolu. A large group of larvae from 2006 were quickly en- 
trained in the SECC and arrived at Northeast Bank after 18-30 days. A small but measureable proportion of 
larvae are from Tokelau in the 100 day PLD with low mortality rate. 



B. Medium Daily Mortality (18%) 



□eslinaiions 




_i_ 




Sources 



ND 



J 



ll 



i 



I 



81 a H « z m i 3 



^ m m rq 






C. High Daily Mortality l46%l 



DeslinaGionrs 



=? 



| s 



3 



1 



03- 2 

i i 







1 1 



100*. 



so% 



0% 



Sources 



M2_ 

nsi a g » a! » ± 



ji M 



S > ?■ 



1 

I" £ S 

zi 4 






i 



■*= =" S c 

* I i I 

Or US 



Figure 3.37. Destinations (and sources) of simulated larvae originating from (arriving at) Northeast Bank for low, medium, and high 
larval mortality rates. Shading of labels indicates core island groups: red, American Samoa; green, Samoa. ND= no data. 



MANU'A ISLANDS 



"T" 



... 






3 % Mortality (toy- 1 



D 

CD 
o ° 



0© 











- 



O 



&: o. 



9 







© 



— i — 



17»'W 



170- W 



It K Hortilrty day" 1 



®<7 







0° O 



4fl 



o 



CD 

j 




o 




• ■ 



O 



© 



G 



— r — 



\ 



© 







— i — 



D 1M Hfl 



MM 



?M 



iKjidfTAIHi- 
1.000 



•PLD -ID PLD -20 PLD =30 


PLD =50 •PLD-1EH 


H Land ] 18km seniameni zones 





Figure 3.38. Position of virtual larvae from Manu'a for all model years by PLD. Upper and lower plots denote 3% and 18% daily 
mortality respectively. 



Destinations 

The Manu'a Islands of Ta'u, Ofu, and Olosega represent the last moderately sized source of larvae in the Sa- 
moan Archipelago (Figure 3.11, Table 3.3). Larvae from Manu'a split into two groups, one along the northern 
side of the Samoan Archipelago and another that is cast southwestward into the region of the Tonga Trench 
Eddy (Figure 3.38). Many in the northern trajectory are entrained in the SEC and transported well south of 
Swains where they mostly expire in the open ocean between the Samoan Archipelago and the Cook Islands. 
Over 20% of the larvae spawned at Manu'a are retained there for the 1 day PLD and low mortality scenario 
(Figure 3.39). Much smaller fractions of local production are retained locally for longer PLDs and higher mor- 







3% 



Figure 3.39. External larval supply and local larval retention at Manu'a as a function of PLD and mortality rate. Top panel: Percent 
of simulated larvae settling at this site that were produced at other sites. Bottom panel: Percent of simulated larvae produced at this 
site that return here. 




D9sHinalions 



_A . _i_l. N .1 ll T .J 



L 



Max. PLD (days} 
If 10 

1 20 

30 

SD 
100 



r 




p Sourcas 

5 100% 

I soft 

* OH — . -MB -j_j.jI.jI . - _ IIL ... . _ .J *__■__ 

H £ * 



ll 



_? .2 ~ 



If 



3 3 



i 1 



lu 



^ a. 
to 



B. Medium Dally Mortality (1S%) 




Destinations 



-e S ill kl Lm 



L 




Sources 



ND 



J _l 



_»._l ___ 






12 I I n 

M tf 3 3 



5 



C. High Daily Mortality (46%) 
? 10O% 

0% 



Destinations 



- h 



..... I 



Sources 




Figure 3.40. Destinations (and sources) of simulated larvae originating from (arriving at) Manu'a for low, medium, and high larval 
mortality rates. Shading of labels indicates core island groups: red, American Samoa; green, Samoa. ND= no data. 



tality scenarios. Larvae settle along the archipelago on the seamounts, Tutuila, and the islands of Samoa 
in similar proportions (Figure 3.40). Larvae are then spread farther westward reaching Wallis and nearby 
islands and seamounts in the 50-100 day range in the low to moderate mortality scenarios. 



Sources 

The Manu'a Islands are reliant on outside larval sources for a wide range, 0-90%, of their arriving larvae de- 
pending on PLD and daily mortality rate (Figure 3.39). Nearly all larvae with a 100 day PLD are from outside 
sources whereas a high proportion of self seeding occurs for larvae in the 10-30 day PLD range. For longer 
PLDs, sources include Tutuila and Samoa with larvae entrained in the eastward flowing SECC and then fed 
back south along the Samoan Archipelago including Manu'a in the 50-1 00 day range (Figure 3.40). Small but 
measureable contributions to the larvae arriving at Manu'a in the low to moderate mortality rate scenarios are 
from Tokelau to the north and Pukapuka and Nassau, and Suwarrow to the east. 



ROSE ATOLL 






i 


170' \ft 


1 






IS % Hwulhy day ' 


G 






« 




©Q 






*- 












O 


O 


% 




jr. 


CD 






© 


u- 


0° fi 


0. .,;. f 

O 
® 




■3 


1A 


V 


o 

1 


1 





l?ft"W 



i^o- w 



■l -■ A 



& 8i?5 2ifl 



K0 



7M 



iKj^lfAIHft 
1000 



•PLD -10 PLD = 20 H_D - 30 PLD =50 «PLD=10G 

■ Lard Q 13km wniamem z&ne& 



Figure 3.41. Position of virtual larvae from Rose Atoll for all model years by PLD. Upper and lower plots denote 3% and 18% daily 
mortality respectively. 



Destinations 

Rose Atoll (Motu O Manu or Muliava by many locally) is a small island at the eastern tip of the Samoan Archi- 
pelago. Due to its small size Rose is a very minor contributor to the total larval pool of the region (Figure 3.11, 
Table 3.3). Larvae from Rose split into two groups depending on the year, one is transported north for partial 
entrainment into the SECC and the other group is moved southwestward in the SEC and Tonga Trench Eddy 
(Figure 3.41). For such a small, isolated source, there is a moderate amount of retention of locally produced 
larvae for the short PLD, low mortality scenario (Figure 3.42). For the short PLDs of 10-20 days, larvae from 
Rose Atoll settle primarily locally or at the nearby Manu'a Islands (Figure 3.43). Longer PLDs of 30-1 00 days 




SO 



3 ™ 

100 *&*> Mfi&W 
0** 



Figure 3.42. External larval supply and local larval retention at Rose Atoll as a function of PLD and mortality rate. Top panel: Percent 
of simulated larvae settling at this site that were produced at other sites. Bottom panel: Percent of simulated larvae produced at this 
site that return here. 



Max. PLC (day&> 



A. Low Daily Mortality (3%) 
| tDO% 



LUssIinattwis 



0* .*■ _l _j* 4. j b _.d 



..A 



H10 

■ 20 

30 

50 

^M 100 



h -1 * P 1 . i H II ■■ , % 

ro K #* ■■- ^ -r -Jc \m 3 3 _H *■ J5 JS c ■□ ni 

s I a 1- 1- 51 n^ £au s e i 



I 50% 



B. Medium Daily Mortality (16%) 




Dcslmatiwis 



I,, 

1 £. 

2 5 



J^_ 



n s: 



I 1 




sources 



ND 



A. III. . 



S 



S? «i «z 
C. High Dally Mortality f 46% f 



i 1 
■■■J 



3 



« 



11 

u 

a 



I 5 



Deslmalioris 






j 



100% 




Sources 



NP 



1? ^ ft ^ 
■= 5 s * 

£ id EL *- 






l 



™ M 3 s 1 3 z 



c ^ 



£L tfj 



Figure 3.43. Destinations (and sources) of simulated larvae originating from (arriving at) Rose Atoll for low, medium, and high 
larval mortality rates. Shading of labels indicates core island groups: red, American Samoa; green, Samoa. ND= no data. 



cast small proportions of larvae to all the seamounts and islands to the west via the SEC and can even reach 
Wallis in the scenario with low mortality. 

Sources 

Similar to Manu'a, Rose Atoll is reliant on outside larval sources for a very wide range of larvae recruits, with 
0-100% of its larvae arriving from elsewhere depending on PLD and daily mortality rate (Figure 3.42). Nearly 
all larvae with 50-100 day PLDs are from outside sources, whereas Rose Atoll is heavily reliant on local larval 
production for PLDs of 1 0-20 days. For these short PLDs, Rose receives few larvae from elsewhere and then 
only in scenarios with low mortality. Pukapuka to the east and Tokelau to the north contribute measureable 
proportions of larvae to Rose but only in low to moderate mortality scenarios and for PLDs of 30-50 days 
(Figure 3.43). This highlights the isolation of Rose from the rest of the islands in the region since it is so far 
downstream in the SEC from the small larval sources provided by the Cook Islands and so far upstream rela- 
tive to the large sources of larvae produced elsewhere in the Samoan Chain. 



Of note, the locally used name "Muliava" can be translated from Samoan to English as "end of the current" 
This could refer to its position at one end of the Samoan Archipelago at the upstream end of the SEC. 



SWAINS ISLAND 






3 % Utility dfry n 



o 



aw»n 



10 



CD 




%SE> o 



o 







® 



© 



© 



5 

IBS' W 



iwn win 

16 ",Moftnl:^tf3y : GT 



H 



CD 
°o3 



®<3 



^o 



n 






£§§3 




® 



— r — 
i^- w 



«% 



© 







— 1 — 



ua-" w 



125 2W 



MO 



JM 



iKibffAIHfl- 
t.QW 



•PLD-10 PLD = 2Q PLD =30 PLD = 5D »FLD=1M 

■ Land Q 18km wftismem zone* 



Figure 3.44. Position of virtual larvae from Swains Island for all model years by PLD. Upper and lower plots denote 3% and 18% 
daily mortality respectively. 



Destinations 

Swains Island is an isolated atoll in the northern part of the American Samoa EEZ and lies in a region of very 
different ocean currents relative to the Samoan Archipelago. The position of these currents isolates Swains 
even more as a larval source or destination than suggested by distance alone. Due to its small size, it is not a 
major larval source to the region (Figure 3.11, Table 3.3). Swains lies in the center of the SECC in most years 
(2004-2007) and consequently most of the larvae spawned there are entrained in currents flowing eastward 
toward Pukapuka and Nassau in the Cook Islands (Figure 3.44). 




3% 
100 46% ^^ 

Figure 3.45. External larval supply and local larval retention at Swains Island as a function of PLD and mortality rate.Top panel: 
Percent of simulated larvae settling at this site that were produced at other sites. Bottom panel: Percent of simulated larvae produced 
at this site that return here. 



A. Low Dally Mortality (3%) 

50% 
0% 



Donations 



.1 L L J _J _^ __ J_ _J — J» nl .-* JL 




Sources 



| 100% 



0% Ik .- --_ MD 



14 I I. 



I, 







.2 a s i ^ t ? 

TS ^ 3 w £ £ 



■g 




in 






6. Medium Daily Mortality |18%) 



& 



Deslinalions 



100% 



] 

I so% 



2" 



07, 



lJ 



.1 




. I 



MO 



j s a% B r if 



C. High Daily Mortality (46%) 



|S 



Sources 



I J 

vi z ■ 

« = = js 



Desltnaliiorts 




100% 



S0% 



* ow 



Sources 



100% 



so% 



* 0% 



& 3 



I 



J5 



m * 



•j> z -^ 



5 



J! £ 



je x ? m 

ills 

5 s- § 3 s 

3 « 



S rS 1 I 



*- i= E 

l N 

Or US 



Figure 3.46. Destinations (and sources) of simulated larvae originating from (arriving at) Swains Island for low, medium, and high 
larval mortality rates. Shading of labels indicates core island groups: red, American Samoa; green, Samoa. ND= no data. 



Virtually no larvae are retained locally at Swains for any PLD or mortality scenario (Figure 3.45). Due to the 
long distances that must be traveled and the low number of larvae starting from a small site like Swains, only 
larvae in the 20-50 day PLD can range make it to Pukapuka and Nassau successfully and only in the low to 
moderate mortality scenarios (Figure 3.46). In 2008 however, a dramatically different pattern emerged with 
the eastward flowing SECC not forming at all, a condition that affected Swains more than any other island 
considered. In 2008, all the larvae from Swains can be observed moving westward in the SEC with many 
reaching Pasco and Taufilemu seamounts after 20 days, and Wallis and nearby seamounts by 50-100 days 
(Figures 3.13, 3.44, 3.46). 



Sources 

Swains Island is reliant on outside larval sources for virtually 100% of its arriving larvae regardless of PLD 
or daily mortality rate (Figure 3.45). Tokelau is the main source of larvae for Swains Island and practically 
the only source for larvae with a 10 day PLD (Figure 3.46). Swains receives many of its larvae in the 20-100 
day PLD range from the major larvae producers of Samoa and even Wallis and its seamounts. A significant 
proportion of the larvae from these sources are entrained in the eastward flowing SECC and due to the large 
numbers of larvae involved and position farther north and west relative to American Samoa, some larvae can 
arrive at far off Swains even before the moderate mortality rate eliminates them all. In contrast, none of the 
islands of American Samoa are a significant source of recruits for Swains. Most of the larvae entrained in the 
SECC from these islands are swept eastward towards the Cook Islands prior to reaching Swains. 



CONCLUSIONS 

Our results indicate a high but variable degree of inter-island connectivity in the Samoan Archipelago and 
surrounding region, with substantial larval retention at most locations, but also substantial larval export and 
some degree of dependency of any individual reef on outside sources even at the shortest PLDs considered. 
The overall picture is of an inter-connected system in which no single location operates in isolation. Although 
some locations are isolated in the sense that they are not important sources for other reefs, these same loca- 
tions are in general dependent on larval arrival from external sources (e.g. Swains). Conversely, some sites 
that are not particularly common destinations can be significant sources to other reefs (e.g. Rose). These 
findings have important implications for the conservation and management of Samoan reef ecosystems 
(Craig and Brainard 2008). 

Most sites in the central portion of the study region are broadly connected to a wide variety of other sites as 
both sources and destinations, indicating that turbulent diffusion plays a significant role in spreading larvae 
widely despite the strong mean flow patterns in this region. In fact, for longer PLDs and, when mortality is 
sufficiently low, virtually all sites in the region are interconnected by at least some small rate of larval ex- 
change (Figure 3.12). However, the predominant patterns of current flow are clearly evident in patterns of 
connectivity, particularly at shorter PLDs and high mortality rates. 

Current flow, and consequently larval transport, is primarily westward along the Samoan Archipelago via the 
SEC. In general, larvae produced at any given location in the Samoan Archipelago tend to seed their natal 
reefs and island neighbors to the west. An overall effect of this directional tendency is that the islands of 
American Samoa export much larval production to Samoa especially for organisms with shorter PLDs of 10- 
30 days. That would be the extent of the connectivity pattern, and probably is for much of the year, were it not 
for the seasonally strong SECC. The north coasts of Samoa are far enough west and north that many larvae 
produced there are entrained in the east flowing SECC and can ultimately settle along the islands of Ameri- 
can Samoa via the feedback loops connecting the SECC with the SEC at -165-170° W longitude. These 
currents are often well developed throughout the PLDs of organisms spawned in late October or November 
as simulated here. These feedback currents make Samoa an important source of larvae for American Samoa 
especially for organisms with PLDs of 30-1 00 days. The connections established by this larval conveyor belt, 
while based on organisms with different PLDs, demonstrate the potential benefits of coordinated manage- 
ment of marine resources and conservation planning between Samoa and American Samoa. The orientation 
of the dominant currents and the long distances downstream from any other large source of larvae suggest 
that much of the Samoan Archipelago is largely dependent on internal sources of larvae transported among 
islands and should be managed accordingly. This is reflected in the interconnected "core" of the connectivity 
matrices, especially evident at short PLDs and high mortality rates (Figure 3.12). 

The islands of Samoa are probably the major source of larvae for the region overall given their very large 
potential reef area. Moving east along the Samoan Archipelago the islands are smaller, have less potential 
reef area, and therefore smaller potential spawning populations. Tutuila and the Manu'a Islands of American 
Samoa have moderate levels of larval production relative to Upolu and Savai'i. Swains Island and the sea- 
mounts of American Samoa are almost entirely dependent on larvae from elsewhere and, due to their very 
small size, contribute relatively little to the larval pool of the Archipelago. 



Despite the potential for circular transport in the current loops connecting the SEC and SECC as noted 
above, the simulations demonstrate that Swains Island is too far north in the SECC current field to function 
in this way and is thus largely disconnected from the rest of the Samoan Archipelago. Most larvae from Ar- 
chipelago sources entrained in the SEC-SECC current system are quickly swept south of Swains and either 
loop back into American Samoa or expire in the open ocean between Rose Atoll and the Cook Islands at the 
end of their PLD. Only the large larval sources on the northern sides of Upolu and Savai'i reach Swains in the 
SECC under the right conditions. Larvae from Swains are largely lost to the open ocean and are not entrained 
quickly enough into the feedback loops connecting the SECC to the SEC for highly successful settlement in 
the Samoan Archipelago. It is important to note that the SECC is a strong, but highly seasonal current. Were 
spawning dates to occur at a different time of year, connectivity patterns in the region of the SECC would be 



quite different. Additional simulation start dates are 
needed to evaluate larval transport during the rest 
of the year. 

The larval sources upstream in the SEC from 
the Samoan Archipelago are in the Cook Islands 
group. These sites are quite far upstream and not 
well connected to islands in the Samoan chain. Or- 
ganisms with short PLDs fail to reach Samoa from 
Cook Islands sources due to the long distance that 
must be travelled. Even organisms with long PLDs 
from the Cook Islands are probably not very likely 
to reach Samoa due to larval mortality, the length 
of time it takes to get to the Samoan Archipelago, 
the turbulent diffusion that spreads larvae widely in 
the ocean, and because the Cook Islands are small 
larval sources to begin with. 






Image 11. Satellite imagery of tiny Rose Atoll, approximately 3.6 
kilometers across, an isolated end-point along the SEC. 
Image Provided By: NCCOS. 



It is important to remember that larval dispersal is 
an inherently stochastic process because it is driv- 
en by current fields of a turbulent ocean (Siegel et 
al. 2008). We have focused on cumulative patterns 

of connectivity over a recent and representative 5 year period, but individual dispersal events are impossible 
to predict, and the patterns we have described here represent an accumulation of connectivity over a par- 
ticular time period. There is always the possibility that a single, rare, anomalous current pattern will result in 
unusual patterns of larval dispersal that deviate from those predicted here. Such events could transport large 
numbers of larvae even from small sources, if they are timed just right, and could connect cells that are blank 
in our simulated connectivity matrices under the right circumstances. Robustness to the inherent stochastic- 
ity and variability in connectivity is another attractive feature of well-designed MPA networks. 

The patterns of connectivity documented here have important implications for the resilience of reef ecosys- 
tems to disturbance events, whether anthropogenic or natural. Resilience refers to the rate of recovery of 
a population, community, or ecosystem following disturbances that could include storms, fishing, pollution, 
bleaching, predator outbreaks (e.g. crown-of-thorns starfish) or disease. Sites such as Rose Atoll and Swains 
Island are far upstream from any large larval sources and thus may be among the slowest to recover follow- 
ing a disturbance due to a lack of recruits. One reason these two locations have a biogeographically distinct 
community structure relative to the rest of the archipelago (Tribolletetal. 2010, Williams etal. 2010, Chapter 
4) may be a preponderance of either species with no pelagic dispersal, or species with very long PLDs that 
are capable of making the trip from distant sources. For these two reasons (probable slower recovery poten- 
tial following disturbance and biogeographic uniqueness/isolation) these sites are worthy of consideration for 
special protection status. In contrast to the isolation of Rose and Swains, Samoa is an important source of 
larvae for itself and the entire region. Recovery from localized disturbance elsewhere in the archipelago may 
depend on larvae from Upolu and Savai'i which make them important to consider when devising a resilient 
regional MPA network. Were these two sites disturbed, recovery would probably be slow due to their high self 
seeding and would depend primarily on the relatively more modest larval sources from Tutuila and Manu'a in 
American Samoa. The highly inter-connected pattern of larval connectivity in the Samoan Archipelago sug- 
gests that a regional planning effort, aimed at the design of an integrated network of marine conservation and 
management areas, would be more likely to achieve management and conservation goals than any effort 
undertaken at a single location without considering linkages with other sites. When connectivity is high, vari- 
able, and asymmetric among locations, as is the case in this region, integrated spatial planning can improve 
realization of conservation, fisheries, and economic goals and can help to identify "win-win" strategies that 
reduce multiple use conflicts and provide benefits to multiple stakeholder groups (Gaines et al. 2007, Cowen 
and Sponaugle 2009, Costello et al. 2010). 



ACKNOWLEDGEMENTS 



Peter Craig (National Park Service) led the charge on understanding interisland connectivity in the Sa- 
moan Archipelago for many years and inspired much of the present study. Doug Fenner (American Samoa 
DMWR) provided insightful comments during development of the approach and review of drafts of the report. 
Don Kobayashi (NOAA/NMFS/PIFSC) has provided models for the study of connectivity in the Pacific and 
provided helpful comments on this work. Phil Wiles (AS EPA) reviewed a draft of this work. Many thanks to 
the NCCOS IT staff and our office neighbors for facilitating the use of the dozen computers needed to run the 
hydrodynamic model simulations. 



REFERENCES 

Alory, G. and T Delcroix. 1999. Climatic variability in the vicinity of Wallis, Futuna, and Samoa islands (13°- 
1 5° S, 1 80°-1 70° W). Oceanologica Acta 22: 249-263. 

Almany, G.R., S.R. Connolly, D.D. Heath, J.D. Hogan, G.R Jones, L.J. McCook, M. Mills, R.L. Pressey, and D.H. Wil- 
liamson. 2009. Connectivity, biodiversity conservation and the design of marine reserve networks for coral reefs. Coral 
Reefs 28: 339-351. 

Atema, J., M.J. Kingsford, and G. Gerlach. 2002. Larval fish could use odour for detection, retention and orientation to 
reefs. Marine Ecology Progress Series 241: 151-160. 

Bare, A.Y., K.L. Grimshaw, J.J. Rooney, M.G. Sabater, D. Fenner, and B. Carrol. 2010. Mesophotic communities of the 
insular shelf at Tutuila, American Samoa. Coral Reefs 29: 369-377. 

Blanco-Martin, B. 2006. Dispersal of coral larvae: a modeling perspective on its determinants and implications. Doctoral 
Dissertation. School of Marine Biology and Aquaculture, James Cook University. 281 pp. 

Bleck, R. and D. Boudra. 1981. Initial testing of a numerical ocean circulation model using a hybrid (quasi-isopycnic) 
vertical coordinate. Journal of Physical Oceanography 11: 755-770. 

Bleck, R. and S. Benjamin. 1993. Regional weather prediction with a model combining terrain-following and isentropic 
coordinates. Part I: Model description. Monthly Weather Review 121: 1770-1785. 

Bonhomme, F. and S. Planes. 2000. Some evolutionary arguments about what maintains the pelagic interval in reef 
fishes. Environmental Biology of Fishes 59: 365-383. 

Botsford, L.W., J.W. White, M.A. Coffroth, C.B. Paris, S. Planes, T.L. Shearer, S.R. Thorrold, and G.P Jones. 2009. Con- 
nectivity and resilience of coral reef metapopulations in marine protected areas: matching empirical efforts to predictive 
needs. Coral Reefs 28: 327-337. 

Chen, S. and B. Qiu. 2004. Seasonal variability of the South Equatorial Countercurrent, Journal of Geophysical Re- 
search 109, C08003, doi:10.1029/2003JC002243. 

Chiswell, S.M. and J.D. Booth. 2008. Sources and sinks of larval settlement in Jasus edwardsii around New Zealand: 
Where do larvae come from and where do they go? Marine Ecology Progress Series 354: 201-217. 

Christie, M.R., B.N. Tissot, M.A. Albins, J. P. Beets, Y. Jia, D.M. Ortiz, S.E. Thompson, and M.A. Hixon. 2010. Lar- 
val Connectivity in an Effective Network of Marine Protected Areas, PLos ONE 5(12),:e15715, doi:10.1371/journal. 
pone.0015715. 

Costello, C, A. Rassweiler, D.A. Siegel, G. De Leo, F Micheli, and A. Rosenberg. 2010. The value of spatial information 
in MPA network design. Proceedings of the National Academy of Sciences of the United States of America. In press. 

Cowen, R.K., K.M.M. Lwiza, S. Sponaugle, C.B. Paris, and D.B. Olson. 2000. Connectivity of marine populations: Open 
or closed? Science 287: 857-859. 

Cowen, R.K., C.B. Paris, and A. Srinivasan. 2006. Scaling of connectivity in marine populations. Science 311: 522-527. 

Cowen, R.K. and S. Sponaugle. 2009. Larval Dispersal and Marine Population Connectivity. Annual Review of Marine 
Science 1:443-466. 

Craig, PC. 1998. Temporal spawning patterns of several surgeonfishes and wrasses in American Samoa. Pacific Sci- 
ence 1 : 35-39. 



Craig, PC, J.H. Choat, L.M. Axe, and S. Saucerman. 1997. Population biology and harvest of the coral reef surgeonfish 
Acanthurus lineatus in American Samoa. Fishery Bulletin 95: 680-693. 

Craig, P. and R. Brainard. 2008. Connectivity among coral reef fish populations in the remote Samoan Archipelago: 
metapopulation concept and implications. Pp 120-130. In: Kilarski SandAEverson (Eds.). Proceedings of the American 
Samoa Coral Reef Fishery Workshop (Oct. 2008). U.S. Dept. of Commerce. NOAATech. Memo. NMFS - F/SPO 114, 
143 pp. 



Craig, RC. (editor). 2009. Natural history guide to American Samoa. 3rd Edition. National Park of American Samoa, De- 
partment of Marine and Wildlife Resources, and American Samoa Community College. Pago Pago, American Samoa. 
131 pp. 

Domokos, R., M.P Seki, J.J .Polovina, and D.R. Hawn. 2007. Oceanographic investigation of the American Samoa 
albacore (Thunnus alalunga) habitat and longline fishing grounds. Fisheries Oceanography 16: 555-572. 

Fisher, R. 2005. Swimming speeds of larval coral reef fishes: impacts on self-recruitment and dispersal. Marine Ecology 
Progress Series 285: 223-232. 

Gaines, S.D., B. Gaylord, and J.L. Largier. 2003. Avoiding current oversights in marine reserve design. Ecological Ap- 
plications 13 Supplement:S32-S46. 

Gaines, S.D., B. Gaylord, L.R. Gerber, A. Hastings, and B.P Kinlan. 2007. Connecting places: The ecological conse- 
quences of dispersal in the sea. Oceanography 20: 90-99. 

Gerlach, G., J. Atema, M.J. Kingsford, K.P Black, and V. Miller-Sims. 2007. Smelling home can prevent dispersal of reef 
fish larvae. Proceedings of the National Academy of Sciences 104:. 858-863. 

Graham, E.M., A.H. Baird, and S.R. Connolly. 2008. Survival dynamics of scleractinian coral larvae and implications for 
dispersal. Coral Reefs 27: 529-539. 

Halliwell, G., R. Bleck, and E. Chassignet. 1998. Atlantic Ocean simulations performed using a new hybrid-coordinate 
ocean model. EOS, Trans. AGU, Fall 1998 AGU meeting. 

Harlan, J.A., S.E. Swearer, R.R. Leben, and C.A. Fox. 2002. Surface circulation in a Caribbean island wake. Continental 
Shelf Research 22: 417-434. 

Harrison, PL., R.C. Babcock, G.D. Bull, J.K. Oliver, C.C. Wallace, and B.L. Willis. 1984. Mass spawning in tropical reef 
corals. Science 223: 1 1 86-1 1 89. 

Itano, D. and T. Buckley. 1988. Observations of the mass spawning of corals and Palolo (Eunice viridis) in American 
Samoa. American Samoa Government, Department of Marine and Wildlife Resources. Report 1 1 . 1 3 pp. 

James, M.K., PR. Armsworth, L.B. Mason, and L. Bode. 2002. The structure of reef fish metapopulations: modeling 
larval dispersal and retention patterns. Proceedings of the Royal Society of London B 269: 2079-2086. 

Jones, G.P, G.R. Almany, G.R. Russ, PF. Sale, R.S. Steneck, M.J.H. van Oppen, and B.L. Willis. 2009. Larval reten- 
tion and connectivity among populations of corals and reef fishes: history, advances and challenges. Coral Reefs 28: 
307-325. 

Junker, M., L. Wantiez, and D. Ponton. 2006. Flexibility in size and age at settlement of coral reef fish: spatial and tem- 
poral variations in Wallis Islands (South Central Pacific). Aquatic Living Resources 19: 339-348. 

Kessler, W.S. and B.A. Taft. 1987. Dynamic heights and zonal geostrophic transports in the Central Tropical Pacific dur- 
ing 1979-84. Journal of Physical Oceanography 17: 97-122. 

Kinlan, B.P, S.D. Gaines, and S.E. Lester. 2005. Propagule dispersal and the scales of marine community process. 
Diversity and Distributions 11: 139-148. 

Kobayashi, D.R. 2006. Colonization of the Hawaiian Archipelago via Johnston Atoll: a characterization of oceanographic 
transport corridors for pelagic larvae using computer simulation. Coral Reefs 25: 407-417. 



Koppers, A.A.P, H. Staudigel, and R. Minnett. 2010. Seamount catalog: Seamount morphology, maps, and data files. 
Oceanography 12: 37. 

Leis, J.M. 2002. Pacific coral-reef fishes: the implications of behavior and ecology of larvae for biodiversity and conser- 
vation, and a reassessment of the open population paradigm. Environmental Biology of Fishes 65: 199-208. 

Leis, J.M. 2006. Are larvae of demersal fishes plankton or nekton? Advances in Marine Biology 51 : 59-141 . 

Leis, J.M. 2007. Behavior as input for modeling dispersal offish larvae: behavior, biogeography, hydrodynamics, ontog- 
eny, physiology and phylogeny meet hydrography. Marine Ecology Progress Series 347: 185-193. 



Leis, J.M. and B. Carson-Ewart. 2003. Orientation of pelagic larvae of coral-reef fishes in the ocean. Marine Ecology 
Progress Series 252: 239-253. 

Lugo-Fernandez, A., K.J. P. Deslarzes, J.M. Price, G.S. Boland, and M.V. Morin. 2001. Inferring probably dispersal of 
Flower Garden Banks coral larvae (Gulf of Mexico) using observed and simulated drifter trajectories. Continental Shelf 
Research 21: 47-67. 

McCook, L.J., G.R. Almany, M.L. Berumen, J.C. Day, A.L. Green, G.P Jones, L.M. Leis, S. Planes, G.R. Russ, PF. Sale, 
and S.R. Thorrold. 2009. Management under uncertainty: guide-lines for incorporating connectivity into the protection of 
coral reefs. Coral Reefs 28: 353-366. 

McCormick, M.I. 1999. Delayed metamorphosis of a tropical reef fish (Acanthurus triostegus): a field experiment. Marine 
Ecology Progress Series 176: 25-38. 

McCormick, M.I. and B.W. Molony. 1995. Influence of water temperature during the larval stage on size, age, and body 
condition of a tropical reef fish at settlement. Marine Ecology Progress Series 118: 59-68. 

Mesophotic Coral Ecosystems. 2010. A research cooperative between the NOAA Center for Sponsored Coastal Ocean 
Research, Perry Institute of Marine Science, and the Centre for Marine Studies at the University of Queensland. Web- 
site accessed March 2010, http://www.mesophotic.org. 

Mildner, S. 1987. Investigation into coral reproduction on the fringing reefs of Western Samoa. Research Report No. 2. 
December 1987. Samoa, Fisheries Division, Ministry of Agriculture, Forests, Fisheries, and Meteorology. 7 pp. 

Mildner, S. 1991. Aspects of the reproductive biology of selected scleractinian corals on Western Samoa and Fijian 
reefs. Thesis, James Cook University of North Queensland. 118 pp. 

Miller, K. and C. Mundy. 2003. Rapid settlement in broadcast spawning corals: implications for larval dispersal. Coral 
Reefs 22: 99-106. 

Munday, PL., J.M. Leis, L.M. Lough, C.B. Paris, M.J. Kinsford, M.L. Berumen, and J. Lambrechts. 2009. Climate 
change and coral reef connectivity. Coral Reefs 28: 379-395. 

Mundy, C. and A. Green. 1 996. Spawning observations of corals and other invertebrates in American Samoa. American 
Samoa Government, Department of Marine and Wildlife Resources Report. 12 pp. 

NOAA Global Drifter Program. 2010. http://www.aoml.noaa.gov/phod/dac/gdp.html. 

Ochavillo, D., S. Tofaeono, M. Sabater, and E.L. Trip. 2011. Population structure of Ctenochaetus striatus (Acanthuri- 
dae) in Tutuila, American Samoa: The use of size-at-age data in multi-scale population size surveys. Fisheries Re- 
search 107: 14-21. 

Okubo, A. 1980. Diffusion and ecological problems: mathematical models. Springer- Verlag, Berlin. 

Paris, C.B., L.M. Cherubin, and R.K. Cowen. 2007. Surfing, spinning, or diving from reef to reef: effects on population 
connectivity. Marine Ecology Progress Series 347: 285-300. 

Planes, S., G.P. Jones, and S.R. Thorrold. 2009. Larval dispersal connects fish populations in a network of marine pro- 
tected areas. Proceedings of the National Academy of Sciences 106: 5693-5697. 

Polovina, J.L., P. Kleiber, and D.R. Kobayashi. 1999. Application of TOPEX-POSEIDON satellite altimetry to simulate 
transport dynamics of larvae of spiny lobster, Panulirus marginatus, in the Northwestern Hawaiian Islands, 1993-1996. 
Fisheries Bulletin 97: 132-143. 



Qiu, B. and S. Chen. 2004. Seasonal modulations in the eddy field of the South Pacific Ocean. Journal of Physical 
Oceanography 34: 1515-1527. 

Richmond, R.H. 1985. Reversible metamorphosis in coral planula larvae. Marine Ecology Progress Series 22: 181-185. 

Rudorff, C.A., J.A. Lorenzzetti, D.F.M. Gherardi, and J.E. Lins-Oliveira. 2009. Modeling spiny lobster larval dispersion in 
the Tropical Atlantic. Fisheries Research 96: 206-215. 



Sale, P.F., H. Van Lavienren, M.C. Ablan Lagman, J. Atema, M. Butler, C. Fauvelot, J.D. Hogan, G.R Jones, K.C. Linde- 
man, C.B. Paris, R. Steneck, and H.L. Stewart. 2010. Preserving Reef connectivity: A handbook for marine protected 
area managers. Connectivity Working Group, Coral Reef Targeted Research & Capacity Building for Management 
Program, UNU-INWEH. 79 pp. 

Shanks, A.L., B.A. Grantham, and M.H. Carr. 2003. Propagule dispersal distance and the size and spacing of marine 
reserves. Ecological Applications 13 Supplement: S159-S169. 

Siegel, D.A., B.P Kinlan, B. Gaylord, and S.D. Gaines. 2003. Lagrangian descriptions of marine larval dispersion. Ma- 
rine Ecology Progress Series 260: 83-96. 

Siegel, D.A., S. Mitarai, C.J. Costello, S.D. Gaines, B.E. Kendall, R.R. Warner, and K.B. Winters. 2008. The stochastic 
nature of larval connectivity among nearshore marine populations. Proceedings of the National Academy of Sciences of 
the United States of America 105: 8974-8979. 

Steneck, R.S., C.B. Paris, S.N.Arnold, M.C. Ablan-Lagman, A.C. Alcala, M.J. Butler, L.J. McCook, G.R. Russ, and PR 
Sale. 2009. Thinking and managing outside the box: coalescing connectivity networks to build region-wide resilience in 
coral reef ecosystems. Coral Reefs 28: 367-378. 

Swearer, S.E., J.E. Caselle, D.W. Lea, and R.R. Warner. 1999. Larval retention and recruitment in an island population 
of a coral-reef fish. Nature 402: 799-802. 

Thresher, R.E., PL. Colin, and L.J. Bell. 1989. Planktonic duration, distribution and population structure of Western and 
Central Pacific Damselfishes (Pomacentridae). Copeia 1989(2): 420-434. 

Tomczak, M. and J.S. Godfrey. 2003. Regional Oceanography: an Introduction. 2nd improved edition. Daya Publishing 
House, Delhi. 390 pp. 

Treml. E.A., PA. Halpin, D.L. Urban, and L.R Pratson. 2008. Modeling population connectivity by ocean currents, a 
graph theoretic approach for marine conservation. Landscape Ecology 23: 19-36. 

Tribollet, A.D., T Schils, and PS. Vroom. 2010. Spatio-temporal variability in macroalgal assemblages of American Sa- 
moa. Phycologia 49(6): 574-591 . 

Victor, B.C. 1986. Duration of the planktonic larval stage for one hundred species of Pacific and Atlantic wrasses (family 
Labridae). Marine Biology 90: 317-326. 

Wellington, G.M. and B.C. Victor. 1989. Planktonic larval duration of one hundred species of Pacific and Atlantic dam- 
selfishes (Pomacentridae). Marine Biology 101: 557-567. 

Williams, I.D., B.L. Richards, S.A. Sandin, J.K. Baum, R.E. Schroeder, M.O. Nadon, B. Zgliczynski, P. Craig, J.L. Mcll- 
wain, and R.E. Brainard. 2010. Differences in Reef Fish Assemblages between Populated and Remote Reefs Spanning 
Multiple Archipelagos Across the Central and Western Pacific. Journal of Marine Biology 2011: 14pp. Article ID 826234, 
doi:10.1155/2011/826234. 



Wilson, J.R. and PL. Harrison. 1998. Settlement-competancy periods of larvae of three species of scleractinian corals. 
Marine Biology 131: 339-345. 



Biogeographic Assessment of Fish and Coral Communities 

of the Samoan Archipelago 

Matthew S. Kendall 1 , Matthew Poti 2 , Ben Carroll 3 , Doug Fenner 3 , Alison Green 4 , Lucy Jacob 3 , Joyce Samuelu Ah Leong 5 , 

Brian P. Kinlan 2 , Ivor D. Williams 6 , Jill Zamzow 6 



INTRODUCTION 

Reef fish and corals are two of the most iconic and 
locally important components of the marine ecosys- 
tem in the Samoan Archipelago. These organisms 
provide a wealth of aesthetic, cultural, and economic 
opportunities to island residents and visitors (Craig 
2009, Sabater 2010). Coral reefs of the archipelago 
fringe the steep sided islands and atolls forming a 
diversity of structures including lagoons, reef flats, 
slopes, pinnacles, and banks (NOAANCCOS 2005, 
Brainard and others 2008, Bare et al. 2010). The 
rich biodiversity of corals comprising these struc- 
tures with their various encrusting, massive, and 
branching morphologies form the physical founda- 
tion of the reef and thereby provide a home for most 
other organisms in the reef ecosystem. Reef fish in 
turn have evolved sizes, colors, and shapes to fill 
every habitat and occupation on the reef. 




Image 12. Pair of long nosed filefish in American Samoa. 
Photo credit: Kevin Leno. NOAA/CRED. 



There are multiple scales at which the marine biogeography of the Samoan Archipelago may be described. At the 
broadest scale, the entire archipelago has been placed into a global context as a unit in the "central Polynesia" 
ecoregional province within the "eastern Indo-Pacific" realm as defined by Spalding et al. (2007) and has a biodiver- 
sity determined by its location on the diversity gradient between the high at the "Coral Triangle" in the Philippines, 
Indonesia, northern New Guinea and the Solomon Islands, and the low at the Pacific Americas (Veron 2000, Veron 
et al. 2009). The present study focuses at finer scales on biogeographic patterns offish and coral among and within 
the islands of American Samoa and Samoa. 



Coral and fish communities are not evenly distributed throughout the Samoan Archipelago. Island age (e.g. distance 
from volcanic hotspot), size, geomorphology, reef structure, oceanographic climate (Chapter 2), position in ocean 
currents (Chapter 3), habitats, wave exposure, human impacts, and other factors have shaped the distribution of reef 
fish and coral among and within the islands (e.g. Green 1996, 2002, Craig et al. 2005, Whaylen and Fenner 2005, 
Sabater and Tofaeono 2006, 2007, Birkeland et al. 2008, Brainard and others 2008, Fenner 2008, Fenner et al. 
2008, Samuelu and Sapatu 2008, Craig 2009, Fenner 2009 a b, Houk et al. 201 0, Carroll 201 0, Williams et al. 201 1 , 
Ochavillo et al. 201 1 ). Basic physiography alone can be used to broadly divide the archipelago from west to east into 
relatively larger high islands with several broad reef flats and shallow lagoon areas (Savai'i and Upolu), a moderately 
sized high island with relatively narrow fringing reefs as well as submerged bank reef formations (Tutuila), smaller 
high islands with fringing reefs and steep shelf slopes (Manu'a Islands of Ofu, Olosega, and Ta'u), and the small, 
low-lying and geologically separate atolls of Rose (Muliava) which lies to the east of the Samoan volcanic hotspot, 
and Swains Island, which lies -400 km to the north and may share geologic origins with the Tokelau Island group. 
The purpose of this chapter of the characterization was to identify geographic patterns, spatial trends, and relatively 
high values or "hotspots" of coral and fish distribution among and around these islands. Documenting biogeographic 
patterns of these foundational resources is a first step in devising informed monitoring, management, conservation, 
and sustainable use strategies (Oram 2008, Conservation International et al. 2010). 

1 NOAA/NOS/NCCOS/CCMA/Biogeography Branch 

2 NOAA/NOS/NCCOS/CCMA/Biogeography Branch and Consolidated Safety Services, Inc., Fairfax, VA, under NOAA Contract No. DG133C07NC0616 

3 American Samoa/Department of Marine and Wildlife Resources 

4 The Nature Conservancy, Asia Pacific Resource Center 

5 Samoa/Ministry of Agriculture and Fisheries/Fisheries Division 

6 NOAA/NMFS/PIFSC/Coral Reef Ecosystem Division 



This assessment combines data from many pre-existing studies into a more robust characterization of the 
reef communities of Samoa and American Samoa that none of the studies could have achieved alone. Al- 
though there are challenges inherent in normalizing and combining results from many studies, this approach 
maximizes use of available information, provides the broadest possible geographic scope, and reduces 
sensitivity of the findings to the biases associated with any one dataset. We took the approach of using more 
datasets for greater geographic coverage and information density at the expense of taxonomic resolution. 
The assessment focused on six general groups of variables: percent cover of live coral, morphological vari- 
ety of corals, community structure or relative abundance of corals, biomass of reef fish, variety of reef fish, 
and community structure or relative abundance of reef fish. A wide diversity of additional measures of reef 
ecosystem conditions are possible, however these six variable groups are collected by most researchers, are 
simple to calculate and interpret, can offer relatively comparable data even when moderately different survey 
methods are used, and characterize some of the most important aspects of reef ecosystems for scientists 
and managers. 

Specifically, our objectives were to: 

1 ) Combine multiple studies of coral and reef fish using normalized data into an analysis of the recent status 
of the six key variable groups. 

2) Assign relatively high, medium, or low values to study sites within the archipelago for each of the coral 
and fish variables and plot their positions around each island. 

3) Identify geographic patterns of hotspots, breakpoints, and spatial trends in the coral and fish variables 
among and within islands of the archipelago. 



METHODS 

The analysis was restricted to data sets that, 1 ) included a broad component of the coral or fish communities 
(i.e. were not restricted to a single taxon or trophic group), 2) had sites spread widely among islands or ex- 
tensively around one of the larger islands, 3) were recent (less than -10 years old), and 4) utilized a relatively 
un-biased approach to site selection that enabled broad geographic inference (e.g. random stratified design). 
Many studies were not included in this assessment primarily because they lacked a broad distribution of sites 
and therefore lacked the widespread geographic scope of inference sought in the characterization. Eight 
studies met the criteria above and are included in the analysis (Figure 4.1, Table 4.1). 

Typical reef morphology differs significantly between Samoa and American Samoa. Zonation of Samoan 
reefs generally consists of a much wider reef flat and shallow lagoon area relative to the narrower fringing 
reef flats that predominate around American Samoa (Green 1996) such that reef flats and shallow lagoons 
comprise a much greater percentage of the total reef area in Samoa. The sampling designs of many reef 
studies in these two jurisdictions reflect this difference in dominant structure in that a majority of studies 
around American Samoa focus on reef slopes (fore reef) whereas most around Samoa focus on shallow 
lagoons. Consequently, the scope of inference for our assessment differs between these jurisdictions and 
reflects the reef zones where most monitoring stud- 
ies have taken place. 

The three coral variables and three fish variables for 
our analysis were selected primarily on the basis of 
their widespread use, ease of comparison across 
studies, and effectiveness in quantifying the status of 
coral and reef fish around the Samoan Archipelago 
(see "Included Datasets" side bar). Percent cover of 
live coral is among the simplest measures recorded 
in coral reef science and provides an estimate of the 
areal extent of live coral habitat at a given site. Im- 
pacts often reduce coral cover, so sites having high 
values are often considered higher quality or less 
impacted reefs (e.g. Nystrom et al. 2008, Cheal et 
al. 2010, but see Vroom 2011). Percent cover of live 
coral was available from all 8 of the studies in our 
analysis (Table 4.1). Coral diversity, or the variety 'mage 13. Diver collecting fish and coral data in American Samoa. 
y V' a "'^ ■*■«;- UUIQI uivciony, ^' u.^ vanciy Photo credit: NOAA/CRED. 




Savai'i and Upolu, Samoa 




Swains Island, American Samoa 

1/ 




0.5 1 



171.09° W 171.08° W 171.07° W 171.06° W 



Tutuila, American Samoa 



r-4 




■ ■■' '■ ^ w t' 



Ofu and Olosega, American Samoa 




*" • i\ 



^ Kilometers 



2 4 







169.65° W 1 


69.6° W 


Ta'u, American Samoa 


1° 

• 
o 

meters 


•• 

8-. J 


1 i 


M* A 


^ I Kilo 

2 4 





170.8 


W 


170.7° W 


Legend 








Samoa 






American Samoa 


o SFR 






o MPABR • ASEPA 


• GCRMN 






o KRS • REA(2010) 
• CRSR o REA(2006) 



Rose Atoll, American Samoa 




Figure 4.1. Datasets and corresponding survey sites included in the analyses. 

of corals on a reef, is another measurement commonly used to describe reef communities. High values are 
often considered to indicate higher quality reefs that may be more resilient to some stressors (Nystrom et 
al. 2008), and there are reports that human impacts reduce coral diversity (Edinger et al. 1998, Houk and 
Musberger 2008). As a measure of coral diversity, we calculated the number of coral genera or morphologies 
(depending on data recorded by each particular study) observed at each site. This metric will hereafter be 
referred to as coral richness and was available from 7 of the 8 studies used in our analysis. Next, reefs may 
have similar values of coral cover and richness but actually be comprised of completely different species or 
species groups. The relative abundances of coral genera or morphologies at each site were therefore used 
to distinguish similarities and differences among sites in community composition and to identify those sites 
with unique communities. Community structure data was available from 7 of the studies used in our analysis. 

A similar suite of 3 variables was used to evaluate reef fish communities. A common measure of the quality of 
reef fish assemblages is fish abundance or biomass per unit of area. Higher values, meaning more or larger 
fish in an area, are often considered to indicate higher quality or less impacted reefs (e.g. Friedlander et al. 
2002, Cheal et al. 201 0). All 8 of the studies in our analysis provided either fish abundance, fish biomass, or 
both. We used biomass per unit area surveyed when available (7 of the 8 studies), and hereafter refer to this 
metric as fish biomass. As a measure of reef fish diversity, we calculated the number of species or species 
groups of reef fish observed at each site. This metric will be referred to as fish richness and was available 
from all 8 of the studies used in our analysis. The relative abundances (biomass measures when available) 
of reef fish species or species groups were used to identify sites with unique or similar reef fish community 
structures. Reef fish community structure was available from 7 of the 8 studies used in our analysis. 



Table 4.1. List of datasets and variables used in the analysis. Y denotes that the variable was included in the analyses whereas NA 
indicates the variable was either not recorded or was unavailable for analysis. 



Study 


Coral Variables 


Fish Variables 


Coral 
Cover 


Coral 
Richness 


Coral 
Community 


Fish 
Biomass 


Fish 
Richness 


Fish 
Community 


American Samoa Enviromental Protection 
Agency 3 


Y 


Y 


Y 


Y 


Y 


Y 


Coral Reef Status Report bc 


Y 


Y 


Y 


Y 


Y 


Y 


Global Coral Reef Monitoring Network 01 


Y 


Y 


Y 


Y 


Y 


Y 


Key Reef Species 6 


Y 


NA 


NA 


Y 


Y 


Y 


Marine Protected Area Bioreconaissance f 


Y 


Y 


Y 


Y 


Y 


NA 


Rapid Ecological Assessment 9 ^ 


Y 


Y 


Y 


Y 


Y 


Y 


Samoan Fish Reserves 1 


Y 


Y 


Y 


Y 


Y 


Y 


Territorial Coral Reef Monitoring Program^ 


Y 


Y 


Y 


Y 


Y 


Y 



a from Houk and Musberger 2008; b from Green 1996; coral cover, fish biomass, and fish richness data were from 1994-5 surveys 
of American Samoa; c from Green 2002; coral richness, coral community, and fish community data were from 2002 surveys of 
Tutuila and Manu'a; d from Samoan Ministry of Agriculture and Fisheries, Fisheries Division and Wilkinson 2008; e from Sabater 
and Tofaeono 2006; f from Oram 2008; g from Brainard and others 2008 and Williams et al. 2011 ; h coral cover, fish biomass, fish 
richness, and fish community data were available from 2010 surveys; coral richness and coral community data were available 
from 2006 surveys; ' from Samoan Ministry of Agriculture and Fisheries, Fisheries Division; j from American Samoa Department of 
Marine and Wildlife Resources and Whaylen and Fenner 2005. 



Included Datasets: American Samoa 

American Samoa Environmental Protection Agency 

Seventeen sites around Tutuila have been monitored approximately every two years since 2003 by the American Samoa Environ- 
mental Protection Agency (hereafter ASEPA) (Houk and Musburger 2008). Sites were selected to assess pollution impacts associ- 
ated with watersheds of varying size and human population. Surveys were conducted at -10 m depth on homogenous habitat of 
the reef slope near stream discharges. Three replicate transects of bottom cover were conducted at each site using a 50 m tape 
and video. Percent coral cover was quantified using randomly selected points and the video data. Fish communities were quanti- 
fied at each site using 5 replicate stationary point counts (Bohnsack and Bannerot 1986). Only those fish >20 cm long and those 
exploited in fisheries were surveyed. Data for 16 sites (Figure 4. 1) from 2007 and 2008, the most recent years available, were used 
for this analysis (Houk and Musburger 2008). From these surveys, all six key variables were calculated (Table 4. 1). 

Coral Reef Status Report 

A resurvey of sites in Samoa and American Samoa was recently evaluated for a Coral Reef Status Report (hereafter CRSR) 
(Green 1996, 2002). In this study, 28 sites around Tutuila and the Manu'a Islands, 6 sites around Rose Atoll, and 2 sites at Swains 
Island were surveyed in 1994-95 using visual census techniques (Figure 4.1). Sites at Tutuila and Manu'a were resurveyed in 
2002. Seven sites around Upolu in Samoa were surveyed in 1994-95 however these data were not used in this study due to in- 
complete spatial coverage around only one island and a focus on a different reef zone relative to the more spatially comprehensive 
Samoan studies. Sites were selected to have broad distribution around the islands of American Samoa in a range of physical set- 
tings. At each site, 3-5 replicate 50 m transects were surveyed at~10m depth on the reef slope. Percent coral cover and colony 
morphology was quantified at three positions every 2 m along the transects. Fish communities were quantified along a 50 by 3 m 
belt (Green 2002). All diumally active, non-cryptic reef fish were recorded to the species level. Each individual fish was counted 
and a length estimate made. Data from 1994-95, the oldest in the study, were used in the analysis of coral cover, fish biomass, and 
fish richness since it encompassed the broadest spatial coverage including Swains and Rose Atolls. Coral community data from 
1994-95 categorized corals into only 4 morphological groups, which limited the ability to resolve differences among sites based 
on coral richness and community structure. Also, fish abundance by species or species group at each site was not available in the 
report based on the 1994-95 data (Green 1996). Therefore, 2002 data for American Samoa was used for coral richness and coral 
and fish community based analyses (Table 4. 1). 

Key Reef Species Program 

The Key Reef Species Program (hereafter KRS) is conducted by the American Samoa Department of Marine and Wildlife Re- 
sources (DMWR). Twenty four sites have been monitored annually around Tutuila using transects conducted at -10 m depth on 
the reef slope. Sites were selected based on wave exposure and coastal region and were well distributed around Tutuila. Four 
replicate video transects of bottom cover were conducted at each site using a 30 m tape and used to calculate percent coral cover. 
Fish communities were quantified using 3-4 replicate 30 by 5 m transects at each site (Sabater and Tofaeono 2006). This program 
only monitors fish that are targeted locally as a food source. Data from 2006, the most recent year made available for this analysis 
included 19 sites (Figure 4. 1). From these surveys, 4 of the 6 key variables could be calculated (Table 4. 1). 



MPA Biological Reconnaissance Assessment 

To support the development of a network of "no-take" MPAs, fish and coral surveys were conducted around Tutuila by the MPA 
Program of the American Samoa DMWR (hereafter MPABR) (Oram 2008). A total of 26 survey sites were spread within 14 regions 
selected based on literature review and scientific opinion of the best potential locations for no-take MPAs (Figure 4. 1). Surveys 
used a semi-quantitative scoring system for a number of coral and fish variables conducted during roving dives up the reef slope 
(Oram 2008). Each survey site was divided into eight five-minute observation stations beginning at the deeper part of the reef slope 
and progressing shallower (Oram 2008, Lucy Jacob DMWR pers. comm .). Data were collected in 2006-2008. From these surveys 
5 of the 6 key variables could be calculated (Table 4. 1). 

Rapid Ecological Assessment 

The Rapid Ecological Assessment (hereafter RE A) is one component of the monitoring conducted by NOAA's Coral Reef Ecosys- 
tem Division (CRED). Sites have been monitored every two years around all islands of American Samoa since 2002. Prior to 2008, 
sites were selected at 10-15 m depth on the reef slope primarily to be representative of reef conditions and management settings 
around the islands. Two replicate surveys of bottom cover were conducted at each site using a 25 m tape and recording the cover 
type at 0.5 m intervals (Brainard and others 2008). Beginning in 2008, but more comprehensively in 2010, sampling effort was dis- 
tributed based on the areas of three depth strata (0-6, 6-18 and 18-30 m). Sites were randomly selected a minimum of 100 m apart 
and fish communities were quantified using a stationary point count (Bohnsack and Bannerot 1986) with ~ two to four replicates 
at each site. Data from 2010 surveys at 241 sites were available for coral cover, fish biomass, fish richness, and fish community 
structure. Coral richness and community structure data were not available from the 2010 survey at the time of this analysis and so 
data from the most recent year available (2006 at 56 sites) were used for those two variables (Figure 4. 1, Table 4. 1). 

Territorial Coral Reef Monitoring Program 

The Territorial Coral Reef Monitoring Program (hereafter TCRMP) is conducted by the American Samoa DMWR (Whaylen and 
Fenner 2005, Fenner 2008, 2009a, b, Carroll 2010). Twelve sites (Figure 4.1) have been monitored annually around Tutuila using 
transects conducted at~10m depth on the reef slope. Sites were selected to achieve an equitable distribution around Tutuila and 
to represent various wave exposure and human impact levels. Benthic surveys were conducted using four replicate transects at 
each site using a 50 m tape. Coral cover by species was recorded to the lowest taxonomic group possible at 0.5 m intervals. All di- 
umally active, non-cryptic reef fish were recorded to species level using belt transects. At each site, 6 replicate 30 m long transects 
were conducted with several passes and widths being used to sample different groups. The first pass (15 m wide) sampled larger, 
more mobile species (e.g. sharks, snapper, jacks, large grouper), the second pass sampled parrotfish (10 m wide), the third pass 
surgeonfish(5 m wide), and remaining species are sampled on the fourth pass (5 m wide). Each individual fish was counted and a 
length estimate made. Data from 2006-2008 (fish) and 2008 (coral), the most recent years available, were used for this analysis. 
From these surveys, all six key variables were calculated (Table 4. 1). 

Included Datasets: Samoa 

Global Coral Reef Monitoring Network 

Eight permanent sites have been monitored around Samoa since 2002 as part of the Global Coral Reef Monitoring Network 
(hereafter GCRMN) (Samuelu and Sapatu 2008, Wilkinson 2008). The sites were selected to have a broad distribution around 
the islands and to be representative of Samoa following GCRMN protocols. Sites were located in the shallow lagoon and reef flat 
habitats around Upolu and Savai'i in 2-5 m depth. At each site, divers survey fish and corals using repeated passes along a 50 by 
2 m transect. First, a set of indicator fish species selected by GCRMN are tallied followed by invertebrates. Next substrate type is 
recorded every two meters along the 50 m transect by 3 divers, one directly above the transect tape and also at 1 m on both sides 
of the transect tape. Data were collected around Savai'i and Upolu from 2002-2010 (Figure 4.1). We used the most recent data 
available for each of the 8 sites in our analysis. From the available data, all 6 key variables could be calculated (Table 4. 1). 

Fish Reserves Monitoring 

Fish reserves are monitored by Samoa's Ministry of Agricul- 
ture and Fisheries, Fisheries Division as part of the techni- 
cal assistance provided to the Community Based Fisheries 
Management Program (hereafter referred to as Samoan, 
Fisheries, Reserves or SFR) (King and Faasili 1998). Cur- 
rently there are 54 fish reserves of variable size (average 
of -75,000 m 2 ). These comprise <1% of the total reef area 
of Samoa and are typically located in the broad reef flat or 
shallow lagoon areas at depths ranging from 2-10 m. The ex- 
act location and size for the fish reserves are proprietary for 
each village and regulations range widely including poten- 
tial rules such as no-take zones, seasonal closures, meth- 
ods restrictions, or size limits (Johannes 2002). Preliminary 
analyses indicated that reserves are providing a random ef- 
fect and are unlikely to introduce any consistent bias in the 
results. This is possibly due to the high variability in size and 
regulations among the reserves (Samuelu 2003, J Samuelu 
Ah Leong pers. comm.). Fish and coral data are recorded at 
5 replicate 50 m transects that are randomly placed within 
each reserve. Methods are similar to those used by GCRMN 
described previously and require multiple passes over the 
transect. Food fish are recorded to species level and all oth- 
ers are tallied at the family level. Data were collected around 
Savai'i and Upolu from 2003-2010. We used the most recent 
data available for each of the sites in our analysis (Figure 
4. 1). From the available data, all 6 key variables could be 
calculated (Table 4. 1). 




Image 14. Snorkler collecting fish and coral data in Samoa. 
Photo credit: Joyce Samuelu Ah Leong, MAF/FD. 



Analysis of Coral Cover, Coral Richness, 
Fish Biomass, and Fish Richness 

It was not possible to simply pool site values for 
each variable from all the datasets into a single 
analysis due to three main issues. First, studies 
in American Samoa were on reef slopes where- 
as those in Samoa were on reef flats and shallow 
lagoons making direct quantitative comparisons 
inappropriate. Second, even within a jurisdiction, 
data collection methods differed among studies 
resulting in incompatible values even when the 
underlying variable being measured was the 
same (e.g. stationary point counts vs. transects 
of multiple dimensions for reef fish). Last, stud- 
ies also quantified different aspects of the coral 
and fish community that were not directly com- 
parable (e.g. coral richness measured at the ge- 
nus level vs. by morphologic group, assessment 
of only food fish vs. day active fish vs. all fish seen). Therefore, a wide range of standardization, scoring, 
and scaling approaches were explored to transform the raw data among the diverse studies into comparable 
values. Results are reported separately by jurisdiction. 




Image 15. A high coral cover, low diversity reef in American Samoa. 
Photo credit: Matt Kendall, NOAA Biogeography. 



We devised a standardized approach to classify values of each variable at each site as high, medium, or low 
relative to other sites surveyed in the archipelago with the same study methods. Sites were only scored rela- 
tive to each other within the same study to avoid incompatibility issues among datasets. For each individual 
study and variable, site values were calculated from raw data (averaging over replicates where necessary) 
and ordered from highest to lowest. We then used the Natural Breaks function in ArcMap version 9.3 to 
identify two class breaks in the distribution of each variable for every study separately. The Natural Breaks 
algorithm chooses class breaks to maximize similarity of values within classes and maximize differences 
among classes, effectively setting boundaries where there are relatively big jumps in data values. The very 
general summary variables and analytical approach that we used were generally insensitive to highly skewed 
data for individual species. However, class breaks were reviewed individually to ensure that anomalous or 
extreme observations for a particular species did not bias the results (e.g. mass recruitment events in March/ 
April [Craig et al. 1997, Green 2002]). The two class breaks were used to assign site values of each variable 
as high, medium, or low relative to all the sites surveyed within a given study (Table 4.2, Appendix C; Figures 
C.1-C.31). This summarized site values in a consistent but qualitative scale for the variables percent coral 
cover, coral richness, fish biomass, and fish richness respectively. It is important to note that cut off values for 



Table 4.2. Assigned breakpoints between low and medium (L — > M) and medium and high (M — > H) values for each of the fish and 
coral variables by dataset. Breakpoints were assigned based on natural breaks in the data (see Appendix A). 



Study 


Coral Cover 


Coral Richness 


Fish Biomass 


Fish Richness 


L^M 


M^H 


L^M 


M^H 


L^M 


M^H 


L^M 


M^H 


ASEPA 


18% 


33% 


8 a 


11 a 


5034 g 


14008 g 


6 b 


13 b 


CRSR 


16% 


31 % 


4 c 


5 C 


375 kg/ha 


682 kg/ha 


102 d 


148 d 


GCRMN 


7% 


44% 


4 c 


7 C 


12 kg/trans. 


29 kg/trans. 


4 b 


6 b 


KRS 


26% 


40% 


NA e 


NA e 


84 kg/km 2 


184 kg/km 2 


25 b 


34 b 


MPABR 


NA f 


41 % 


269 


429 


25g 


33g 


529 


679 


REA 


15% 


35% 


11 a 


18 a 


24 g/m 2 


60 g/m 2 


24 b 


31 b 


SFR 


11 % 


38% 


3 C 


6 C 


5 kg/trans. 


18 kg/trans. 


4 b 


7 b 


TCRMP 


23% 


42% 


7 b 


12 b 


55 g/m 2 


72 g/m 2 


15 b 


19 b 



a number of genera per transect; b number of species per transect; c number of morphologies per transect; d number of species per 750 m 2 (area sur- 
veyed at each site); e no coral richness data for KRS; f no MPABR coral cover data values classified as low; g custom scoring scale (see Oram 2008) 



the high, medium, and low categories varied widely among studies due primarily to differences in methodol- 
ogy and units of the data recorded. It is suggested that readers examine Table 4.2 and the Figures C.1-C.31 
in Appendix C where quantitative cutoffs are shown and then refer to the corresponding description of each 
study to understand the expected range of high, medium, and low values that can result given each particular 
methodology. 

Analysis of Coral and Fish Community Structure 

To identify sites with similar coral and fish assemblages we performed a series of non-metric multi-dimen- 
sional scaling (MDS) analyses for each study using PRIMER version 6 (Clarke and Gorley 2006). MDS 
provided a plot of survey sites for each dataset based on their relative similarity to each other. Sites closer 
together in chart space have more similar communities to each other than sites plotted farther apart (Clarke 
1993, Legendreand Legendre 1998). Separate MDS plots were created for each dataset for coral communi- 
ty structure and fish community structure. The raw data for coral community analysis was percent coral cover 
by genus or morphological group depending on the study. The raw data for analyses of fish communities 
consisted offish biomass by species or species groups. From these values, the Bray-Curtis coefficient was 
calculated among all pairs of sites to measure community similarity. The Bray-Curtis similarity is commonly 
used in studies of ecological communities and emphasizes shared patterns in species abundances rather 
than simply the presence/absence of species (Clarke 1993, McCune and Grace 2002). Sites were then plot- 
ted in two-dimensional chart space (MDS plots) based on these similarity values such that dissimilar sites are 
far apart and similar sites are grouped close together. 

To support the qualitative interpretation of the MDS plots, we also explored potential groupings of sites in 
each dataset according to their fish and coral assemblages using hierarchical clustering (Clarke 1993, Leg- 
endre and Legendre 1998). Differences among clusters and among biogeographic regions (see below) were 
explored using an analysis of similarities (ANOSIM) test (Clarke 1993, Legendre and Legendre 1998, Clarke 
and Gorley 2006). ANOSIM produces a statistic (R), analogous to a correlation coefficient, that measures the 
association between pre-defined groups (e.g. biogeographic regions) and MDS patterns. Ap-value indicat- 
ing the statistical significance of the R statistic is also provided. Once MDS and cluster analysis were com- 
pleted for all datasets (Appendix C; Figures C.32-C.34), the results and plots were visually compared among 
the datasets for consistent patterns in site groupings. Results are highlighted where two or more datasets 
showed consistent patterns, groups of sites exhibited similar fish or coral communities, or sites had unique 
community composition. 



Identification of Biogeographic Patterns 

All sites were mapped according to their corresponding high, medium, or low classification for each vari- 
able. Site classifications were summarized at multiple spatial scales to facilitate comparisons among islands 
and to place sites into their regional context. For coral cover, coral richness, fish biomass, and fish richness 
respectively, the proportion of sites categorized as high, medium, and low were summarized in pie charts 
hierarchically for 1) Samoa and American Samoa, 2) for each island or island group (Savai'i, Upolu, Tutuila, 
Manu'a, Swains, and Rose Atoll, and 3) at the finest scale, along biogeographically distinct segments of 
coast. Due to the variable density of survey sites among regions, summary charts were scaled by approxi- 
mate reef length, using shoreline length as a proxy, to account for unequal sample sizes. For this reason, 
results within biogeographic regions are presented as the proportion of survey sites within each category 
(high, medium, or low) whereas results summarized across multiple biogeographic regions are weighted 
averages of the proportions for each region, with weights given by the length of shoreline. Regions with no 
surveys were excluded from summaries. For every analysis scale, the number of studies and number of sites 
comprising a given pie chart is provided. These values provide a measure of the relative confidence of the 
results with higher values representing more studies/sites and therefore a more robust analysis. 

Biogeographically distinct regions were identified through simultaneous consideration of two factors. First, 
each island was visually examined for spatial patterns in the high, medium, and low values of the survey sites 
with the goal of identifying clusters of similar values. Second, prominent features of coastal geomorphology 
(e.g. points, banks, bays, exposure, and even specific villages) were identified on either side of the clusters 
with the goal of defining the physical boundaries of each distinct region. This process was conducted for all 



six variables such that adjacent regions differ in their relative proportions of high, medium, and low values or 
coral and fish communities for at least one variable. The end result was that biogeographic regions, hereafter 
called "Bioregions", with distinct reef fish and coral communities were identified as defined above. 

Identification of Biogeographic Hotspots 

Bioregions with an especially large proportion of high site values compared to the study area as a whole can 
be considered ecologically important areas worthy of special monitoring or management considerations. For 
this study, such ecological "hotspots" were first identified for each variable individually (coral cover, coral rich- 
ness, fish abundance, and fish richness) and then across multiple variables since monitoring and manage- 
ment importance is often heightened for an area when hotspots co-occur for multiple variables. 

The term "hotspot" has been used widely to describe concentrations of high value sites using a diversity of 
approaches, scales, and variables (e.g. total number of species, threatened species, and/or endemics) and 
must be clearly defined. To identify hotspots for this study, we calculated the proportion of sites classified as 
high for each variable within each Bioregion. Any Bioregion with a proportion of high values greater than the 
proportion of high values in the entire jurisdiction (Samoa or American Samoa respectively) was considered 
to be a hotspot for the indicated variable. This was done for each variable individually and then hotspot re- 
sults were tallied across the four variables to determine the number of variables contributing to each Biore- 
gion's hotspot status. 

In addition, to aid in interpretation of hotspot values for each jurisdiction, the probability that the proportion 
of high value sites from any particular hotspot could have arisen by random chance was estimated using the 
statistical method of resampling. In this analysis, for each Bioregion identified as a hotspot we took 1*10 6 
random samples of n sites from the entire pool of survey sites within a given jurisdiction, where n is the num- 
ber of sites in the Bioregion, and calculated the proportion of sites classified as high in each random sample. 
The number of times the proportion of high value sites was greater than or equal to the actual observed 
proportion for the Bioregion was divided by 1*1 6 to provide a p-value that expressed the probability that the 
observed proportion (or greater) of high sites could have arisen by random chance. Lower p-values denote 
observations considered less likely to have occurred through random chance. For example, if the observed 
proportion of high sites was met or exceeded in 1 00,000 of the 1 ,000,000 random draws it could be assumed 
that the observed pattern could occur merely by chance only 10% of the time (p = 0.1). This analysis was 
not used to assign a formal significance level but rather as an aid to interpreting the observed proportions. 
Hotspots with high p-values should be interpreted more cautiously than those with lower values. 

RESULTS 

Distribution Of Survey Effort 

The number of studies and individual survey sites are summarized by variable for each jurisdiction, among 
islands, and according to biogeographic breakpoints. We identified 30 biogeographic regions (Bioregions) 
based on the six variables considered (Figure 4.2). Because results for Bioregions with more studies and 
higher numbers of survey sites are more robust than those with fewer, those Bioregions with few studies 
or sites will be presented in mapped results but discussed only sparingly due to the comparatively reduced 
confidence in the results. 



Of the eight datasets suitable for the study, six took place around American Samoa compared to only two 
around Samoa despite its much larger potential reef area (Figure 4.1). By far the greatest number of studies 
and survey sites occurred around Tutuila for all variables making results for that island the most robust in 
the assessment. For example, an average of 14 survey sites from 4 studies included coral cover data within 
Bioregions around Tutuila whereas much larger Bioregions around Samoa were represented by an aver- 
age of only 6 sites from 2 studies. The high density of points around Tutuila facilitated a much more detailed 
breakout of biogeographically distinct regions (n = 15) compared to the other islands. The Manu'a group was 
split into three regions, Ofu/Olosega, eastern Ta'u, and western Ta'u. Many more biogeographically distinct 
regions probably exist around Upolu and Savai'i than were identified here (n = 6 and 5 respectively) but could 
not be detected due to the limited number of surveys around those islands. The distribution of values at sites 
around Swains and Rose Atolls were spatially uniform for most variables on the reef slopes (exception was 



Number labels seperated by dashed 
lines indicate assigned bioregions based 
on analysis of fish and coral data. 



Swains Island 
16 



Tutuila 



Fagamalo 

/ 14 ; 



15 / ffflvli 




Cape Matatula 



10 



<'9; 



Cape Taputapu "' 



_Vw ^/j 




::::; v 



4 Pago Pago 6 

Harbor Fagaitua 
Bay 



Sail Rock 
Point o 




25 50 
1 



Fagaloa Bay 



/ \ Aleipata 

25 Falealili 27 % Islands 

Manono Is. 
and Apolima 
Strait 



I Kilometers 
200 



Manu'a Islands 



18 

Ofu/Olosega -in 



20 



Rose Atoll 
17 



168° W 



Figure 4.2. Biogeographic regions (Bioregions) assigned based on analyses offish and coral data. Islands and key geographic fea- 
tures such as villages, bays, or points that are referred to in the text are labeled. 

for sites inside versus outside the lagoon at Rose) and these islands were generally too small to warrant 
further biogeographic breakdown offish and coral patterns based on the variables we considered. 

Despite the relatively intense sampling around Tutuila, it should be emphasized that the scope of inference 
for Tutuila from this analysis is largely limited to the reef slope where the vast majority of survey effort took 
place. Only -5% of survey effort was spent on bank reefs around the island, and those surveys were almost 
exclusively on Taema and Nafanua Banks. There is a very large shelf area around Tutuila with many bank 
and pinnacle reef formations (Bare et al. 2010, Appendix B) that are poorly known relative to the reef slopes 
and are beyond the scope of this analysis. Most of the area of these banks lies much deeper than 1 m, the 
depth at which many studies used here were focused, and is below the depths of safe diving for extended 
survey work. Some surveys have been conducted on reef flats; however, those data were spatially limited 
and therefore not used in this assessment. Similarly, since survey effort around Samoa is focused landward 
of the reef crest, the scope of inference for that region is largely limited to the reef flat and lagoon reef zones 
and results are discussed separately for the two jurisdictions. 

Percent Coral Cover: Samoa 

Percent coral cover for Samoa overall was rated as high for over 40% of the coast (Figure 4.3a). Results 
summarized by island revealed that a much larger proportion of Savai'i (-60% of coastline) was rated as 
having high coral cover compared to Upolu (-30%). Biogeographic patterns of coral cover revealed north/ 
south patterns of coral cover that differed by island. The north and northeast facing coasts of Savai'i possess 
a large proportion of sites with high coral cover (Figure 4.3b). In contrast, Upolu has more moderate and vari- 



Samoa American Samoa 

2(61) 6(339) 

£ A 

/ \ / I \\ 

Savai'i Upolu Tutuila Manu'a Swains Rose 

2(23) 2(38) 6(215) 2(59) 2(26) 2(39) 



d 



Swains Island 



Legend 
Coral Cover (by site) Coral Cover (by region) 

Low^--"^^ High 



o Low 
o Medium 
• High 



)W / ^"^^ 

05 









■' ■ 



4i^ i££jsflfe>l ,' ; ^ =:: £x*'- \ > 













\ 



4(9) 6(16) 



2(20) 



1(6) 



5(12) 



2(7) w 



® 



Manu'a Islands 



(DO 

2(18) 1(9) 



25 50 



I Kilometers 
200 



© 



1 

173° W 



1 

171°W 



1 

169° W 



Figure 4.3. Coral cover at survey sites across Samoa and American Samoa. Sites and pie charts are coded as high, medium, or low 
coral cover values, (a) Proportions of high, medium, and low values by jurisdiction and by island. (b,c) Proportions of high, medium, 
and low values for individual Bioregions. Number labels represent the number of studies and sites (in parentheses) comprising each 
pie chart. 

able values with areas of low coral cover along the north and west coasts, especially in the Manono Island/ 
Apolima Strait area and between Apia and Fagaloa Bay (Bioregions 25 and 29). 

Percent Coral Cover: American Samoa 

American Samoa overall had only 22% of the coast rated as having high coral cover and the rest split ap- 
proximately evenly between the medium and low categories (Figure 4.3a). Results summarized by island 
revealed that Swains Island had high coral cover (-55% of sites) whereas a relatively large proportion of 
some islands such as Tutuila (-50% of coastline) and Rose Atoll (-75% of sites) had low coral cover. Spatial 
patterns of coral cover within islands or island groups revealed highly variable values even among adjacent 
segments of coast. Tutuila has some areas with a very large proportion of high coral cover sites (e.g. SW 
coast from Cape Taputapu to Sail Rock Point [Bioregions 1 and 2], coast east of Fagamalo Village [Bioregion 
14], northern coast including Matalia/Cockscomb Point [Bioregion 12], and southeastern regions including 
Aunu'u and the eastern tip of the island and Fagaitua Bay [Bioregions 8, 10, and 6]) separated by distinct 
areas with relatively low values (i.e. NW coast offshore from Fagali'i and Fagasa villages [Bioregions 13 and 
15], coastlines including and extending away from Pago Pago Harbor and the airport [Bioregions 3 and 4]) 
(Figure 4.3c). The Manu'a Islands showed perceptible biogeographic differences as well. The east side of 
Ta'u (Bioregion 20) possessed a large proportion of sites with high coral cover whereas western Ta'u (Bio- 
region 19) and Ofu/Olosega (Bioregion 18) possessed relatively lower values (Figure 4.3b). Swains Island 
(Bioregion 16) had generally high values of coral cover whereas Rose Atoll (Bioregion 17) possessed gener- 
ally low values. 



Samoa American Samoa 

2(61) 5(137) 

PA 

Savai'i Upolu Tutuila Manu'a Swains Rose 

2(23) 2(38) 5(88) 2(30) 1(8) 1(11) 



Swains Island 



Legend 
Coral Richness (by site) Coral Richness (by region) 

Low .^--^^ High 



o Low 
o Medium 
• High 



<5 



.4 




(2) . (3 , 

V_^ 5 (10) 



# ,. <r± X ^C5® 



z:: ^"" \ \ 

V (3 



2(7) 
4(7) 4(9) 



3'e> 

2(7) v_y 



«c 



2(17) 



Manu'a Islands 

«3 



25 50 



I Kilometers 
200 



1 

173° W 



1 

172° W 



1 

171°W 



1 

169° W 



Figure 4.4. Coral richness at survey sites across Samoa and American Samoa. Sites and pie charts are coded as high, medium, or 
low coral richness values, (a) Proportions of high, medium, and low values by jurisdiction and by island. (b,c) Proportions of high, 
medium, and low values for individual Bioregions. Number labels represent the number of studies and sites (in parentheses) compris- 
ing each pie chart. 

Coral Richness: Samoa 

Coral richness for Samoa overall was rated as high for -35% of the coastline (Figure 4.4a). Results sum- 
marized by island revealed that a very large proportion of Savai'i (50% of the coastline) was rated as having 
high coral richness compared to Upolu (<25%). Biogeographic patterns of coral richness within islands or 
island groups revealed a few notable spatial trends. Savai'i possesses a small proportion of sites with low 
coral richness relative to Upolu (Figures 4.4a-b). The proportion of sites with high coral richness generally 
declines eastward in Samoa. Compared to Savai'i, Upolu has somewhat more moderate and variable values 
with concentrated areas of low coral richness in the Manono Island/Apolima Strait area and between Apia 
and Fagaloa Bay (Bioregions 25 and 29). 

Coral Richness: American Samoa 

American Samoa overall had 23% of the coast rated as having high coral richness and the rest split ap- 
proximately evenly between the medium and low categories (Figure 4.4a). Results summarized by island 
revealed that the Manu'a group had high coral richness (-60% of coastline) whereas a relatively large pro- 
portion of sites at other islands, especially Swains (100% of sites), had lower values. Spatial patterns within 
islands or island groups revealed a north/south pattern for Tutuila but also some highly variable patterns even 
among adjacent segments of coast. Areas around Tutuila with relatively high coral richness (-25% of sites) 
include the SW coast from Cape Taputapu to Sail Rock Point (Bioregions 1 and 2), the northeastern coast 
from Masefao Bay to Matatula Point (Bioregion 11), southeastern regions including around Aunu'u Island and 
Fagaitua Bay (Bioregions 8 and 6 respectively), and even Pago Pago Harbor (Bioregion 5) which had mostly 



sites with low coral richness (>75%) but a few in the high richness category (Figure 4.4c). The rest of the 
regions around Tutuila were characterized by low or moderate values. The Manu'a group showed perceptible 
biogeographic differences as well. The east side of Ta'u (Bioregion 20) had only sites with moderate or high 
coral richness whereas western Ta'u (Bioregion 19) and Ofu/Olosega (Bioregion 18) were more variable and 
possessed relatively more high richness sites but also some with low values (Figure 4.4b). All of the coral 
richness values around Swains Island were low (Bioregion 16). Rose Atoll possessed only moderate and low 
values (Bioregion 17). 

Coral Community Structure: Samoa 

Sparse data and extensive overlap of sites in MDS plots limited the interpretation of results for coral com- 
munity structure in Bioregions around Samoa. However, both datasets (GCRMN, SFR) revealed consistent 
patterns of overlap among sites around southern Upolu (Bioregions 26 and 27) and northern Savai'i (21 and 
24) as well as a separate and unique coral community in sites on southern Savai'i (Bioregion 23). 

Coral Community Structure: American Samoa 

The MDS analyses for each study in American Samoa revealed extensive overlap in coral communities 
among sites and otherwise biogeographically distinct regions (Appendix C; Figures C.32 and C.34). Only two 
studies showed significant differences among Bioregions in the global ANOSIM (MPABR, R = 0.327 and p = 
0.002; REA, R = 0.566 and p = 0.001), and only 2-4 statistically different groups could be identified for some 
datasets through cluster analysis. Despite the overall finding of high overlap, coral communities of a few Bio- 
regions showed consistent patterns of similarity or uniqueness among datasets (Figure 4.5). Sites in Pago 
Pago Harbor (Bioregion 5) consistently showed a unique coral assemblage among datasets (ASEPA, REA, 






Legend 



Multi-headed arrows connect 
regions with similar coral 
communities 



Single-headed arrows 
indicate regions with 
unique coral communities 



Number labels indicate assigned 
bioregions. 



Swains Island 







8; 



A 



Manu'a Islands 



: 
19:20 



\ 



25 50 



I Kilometers 
200 



Rose Atoll 
17 



168° W 



Figure 4.5. Summary of Bioregions sharing similar coral communities and those with unique coral communities as identified from 
the MDS analyses. 



TCRMP). Note that unique sites are typically thought of in a positive sense, but in this case may indicate a 
uniquely unhealthy coral community because this Bioregion is heavily impacted by human activities (Ped- 
ersen Planning Consultants 2000). Sites between Cape Taputapu and Sail Rock Point (Bioregions 1 and 2) 
and also those around Aunu'u often plotted with the same group among datasets (ASEPA, CRSR, MPABR, 
REA, and TCRMP). Parts of NE (Bioregion 11) and NW Tutuila (Bioregion 14) showed similarities in coral 
community structure in two datasets (CRSR, TCRMP) as did the north/central coast including Matalia/Cocks- 
comb Point (Bioregion 12) and Fagaitua Bay (Bioregion 6) (CRSR, MPABR) although these areas were 
represented by few sites. Most sites in east and west Ta'u (Bioregions 20 and 19) and to a lesser degree 
Ofu/Olosega (Bioregion 18) were similar to each other (CRSR, REA). Rose Atoll (Bioregion 17) and Swains 
Island (Bioregion 16) sites plotted separately in relatively compact groups at the periphery of the MDS plot 
(surveyed in REA data only). This indicates a somewhat unique and homogeneous coral community as would 
be expected for each of these two small and isolated island Bioregions, a pattern also found in analysis of 
their algal communities (Tribollet et al. 2010). 

Fish Biomass: Samoa 

Fish biomass in surveys around Samoa overall showed that only -10% of coastlines were classified as hav- 
ing high biomass with the remainder divided approximately evenly between the low and medium categories 
(Figure 4.6a). A relatively large proportion of Savai'i, -22% of coastline, had high fish biomass whereas 
none of Upolu's coast was classified as high. Biogeographic patterns offish biomass within islands revealed 
that the northern coasts of Savai'i between Falealupo Village and Apolima Strait possess a large propor- 



Samoa American Samoa 

2 (52) 6 (344) 





/ 1 // \\ 

Savai'i Upolu Tutuila Manu'a Swains Rose 

2(18) 2(34) 6(215) 2(63) 2(26) 2(40) 

(M) ©ex^e) 



/^"^^ Swains 

03* 



Legend 
Fish Biomass (by site) Fish Biomass (by zone) 

Low / ---^^ High 



o Low 
o Medium 
• High 



05 



k 






(D 




oars ■■'' 



1(3) 



0(0) 



,-" Jl 



v\&>Y, i 



Upolu 



2(2) 



2(34) 



Manu'a Islands 



©0 




25 50 



I Kilometers 
200 







1 

173° W 



1 

172° W 



1 

171°W 



1 

169° W 



Figure 4.6. Fish biomass at survey sites across Samoa and American Samoa. Sites and pie charts are coded as high, medium, or 
low biomass values, (a) Proportions of high, medium, and low values by jurisdiction and by island. (b,c) Proportions of high, medium, 
and low values for individual Bioregions. Number labels represent the number of studies and sites (in parentheses) comprising each 
pie chart. 




Image 16. Halfspotted hawkfish on a reef with low coral cover in Ameri- 
can Samoa. Photo credit: Kevin Lino, NOAA/CRED. 



tion of sites with high fish biomass whereas 
Upolu has large regions of low biomass in the 
Manono Island/Apolima Strait area (Bioregion 
25), between Apia and Fagaloa Bay (Bioregion 
29), and on the southeastern coast between 
Falealili and Aleipata Islands (Bioregion 27) 
(Figure 4.6b). 

Fish Biomass: American Samoa 

Fish biomass in surveys around American Sa- 
moa overall showed that -10% of coastlines 
were classified as having high biomass with 
the remainder divided approximately evenly 
between the low and medium categories (Fig- 
ure 4.6a). Patterns among islands in Ameri- 
can Samoa were relatively uniform although 
Swains and Rose Atolls had a relatively greater 
proportion of sites with high biomass as would 
be expected for these two remote islands (Wil- 
liams et al. 2011). Tutuila had a greater pro- 
portion of low rated coastline compared to other islands. Biogeographic patterns within islands or island 
groups revealed that Tutuila has a more uniform distribution of biomass values relative to other variables. 
Areas around Tutuila with relatively high fish biomass (>20% of sites) include the eastern tip (Bioregion 10), 
Aunu'u (Bioregion 8), Fagaitua Bay (Bioregion 6), and Taema and Nafanua Banks (Bioregion 7; Figure 4.6c). 
The rest of Tutuila was dominated by low or moderate values which comprised >90% of the sites in those 
regions. The Manu'a group showed less pronounced biogeographic differences with Ofu/Olosega (Bioregion 

18) possessing a relatively large proportion of high biomass sites relative to Ta'u (Bioregions 19-20; Figure 
4.6b). Swains (Bioregion 16, -15% of values in the high biomass category) and Rose (Bioregion 17, -10% of 
values in the high category) possessed generally similar proportions in fish biomass categories. 

Fish Richness: Samoa 

Coastlines for Samoa overall were approximately evenly divided among high, medium, and low fish rich- 
ness values (Figure 4.7a). In contrast to other variables, Upolu had a greater proportion of sites classified as 
high richness and fewer classified as low richness relative to Savai'i. Biogeographic patterns within islands 
revealed that Savai'i possessed fewer high values and a large proportion of low value sites for fish richness 
on its northern coasts than were seen for other variables (Figure 4.7b). Upolu had greater proportions of high 
site values on the southern coasts than were seen in other variables. Areas of low richness were however, 
again found in the Manono Island/Apolima Strait area and eastward past Apia to Fagaloa Bay (Bioregions 
25, 29, and 30). 

Fish Richness: American Samoa 

American Samoa overall had 22% of the coast rated as having high fish richness and nearly 50% classified 
as moderate. Patterns among islands in American Samoa were more uniform with typically -25% of sites 
classified as high and -50% classified as medium fish richness. The exception was Rose Atoll which had 
very few sites classified as high and nearly half the remaining sites classified as having low richness. Bio- 
geographic patterns of fish richness within islands or island groups revealed highly variable patterns even 
among adjacent segments of coast. Tutuila especially had a more variable distribution of richness values 
around the island relative to some other variables. Notable locations around Tutuila with -40-50% of sites 
having high fish richness were around Aunu'u and the bank to the east (Bioregions 8 and 9), Fagatele and 
Larsen Bays (Bioregion 2), and along the NW coast east of Fagamalo village (Bioregion 14) (Figure 4.7c). 
The Manu'a group showed less pronounced biogeographic differences with the west half of Ta'u (Bioregion 

19) possessing a greater proportion of sites with high fish richness (-40%) than Ofu/Olosega (Bioregion 18) 
and the east half of Ta'u (-20% of sites) (Bioregion 20). Swains (Bioregion 1 6) had -25% of richness values 



Samoa American Samoa 

2 (52) 6 (347) 

£ A 

/I / I \\ 

Savai'i Upolu Tutuila Manu'a Swains Rose 
2(18) 2(34) 6(218) 2(63) 2(26) 2(40) 

0O £0(3(2) 



a- 



Swains Island 



Legend 
Fish Richness (by site) Fish Richness (by zone) 

Low^--^^ High 



o Low 
o Medium 
• High 



)w^-— ^^ Hi 

<5 







2(6) 

Savai'i 9o 



Tutuila 



6(19) 





® 









V"17 ,. 







' 



2(6) 



!L 






3 

2(2) / /""^ ' **° 48 * 

0$ 



e 



Manu'a Islands 



00 



25 50 



I Kilometers 
200 







Figure 4.7. Fish richness at survey sites across Samoa and American Samoa. Sites and pie charts are coded as high, medium, 
or low fish richness values, (a) Proportions of high, medium, and low values by jurisdiction and by island. (b,c) Proportions of high, 
medium, and low values for individual Bioregions. Number labels represent the number of studies and sites (in parentheses) com- 
prising each pie chart. 

in the high category and only a small proportion in the low category whereas Rose (Bioregion 17) had only 
-5% of values in the high category and nearly 50% in the low category. 

Fish Community Structure: Samoa 

For Samoa, the MDS analysis of fish community structure was limited by only 2 datasets and extensive 
overlap among sites. Despite this, both datasets (GCRMN, SFR) again revealed consistent patterns of over- 
lap among sites around southern Upolu (Bioregions 26 and 27) and northern Savai'i (21 and 24), a finding 
similar to the coral community analysis. 

Fish Community Structure: American Samoa 

Even more so than was observed with the coral community analysis, there was a great deal of overlap and 
similarity in fish communities among sites and Bioregions within each study in MDS plots for American Sa- 
moa (Appendix C; Figures C.33-C.34). Only three studies showed significant differences among Bioregions 
in the global ANOSIM (KRS, R = 0.375 and p = 0.003; REA, R = 0.372 and p = 0.001 ; SFR, R = 0.197 and p = 
0.003), and only 2-3 statistically different groups could be identified for some datasets through cluster analy- 
sis. Despite the overall finding of high overlap among sites, fish communities of a few Bioregions showed 
consistent patterns of similarity or uniqueness among datasets (Figure 4.8). Sites around Aunu'u (Bioregion 
8) showed a unique fish assemblage in three datasets (KRS, REA, TCRMP). Sites between Cape Taputapu 
and Sail Rock Point (Bioregions 1 and 2) generally plotted in the same group among three datasets as well 
(ASEPA, REA, TCRMP). Sites along the north/central coast including Matalia/Cockscomb Point (Bioregion 




25 50 



26 / 


27 






Tutuila 


Manu'a Islands 




<r 




! 














■ Kilometers 






DO 150 




200 





Figure 4.8. Summary of Bioregions sharing similar fish communities and those with unique fish communities as identified from the 
MDS analyses. 

12) and Fagaitua Bay (Bioregion 6) also showed some regularly occurring similarities among sites (ASEPA, 
TCRMP), a finding similar to the analysis of coral communities although based on two different datasets. 
There were other notable results within particular studies (ASEPA showed Pago Pago Harbor as unique, 
REA showed a distinct fish community at Swains, KRS showed separation of northern versus southern sites 
around Tutuila, CRSR showed separation of fish communities at eastern and western Ta'u), but these pat- 
terns were not confirmed across multiple datasets. 

Biogeographic Hotspots 

Biogeographic patterns in coral and fish variables were evident at several spatial scales. Comparing among 
Samoan islands, Savai'i had consistently high values for multiple fish and coral variables. The exception was 
for fish richness, for which Savai'i possessed a large proportion of low scoring coastline. There were notable 
locations in American Samoa at the island or island group level for single variables. Manu'a had many high 
values for coral richness and Swains had many high values for coral cover. Notable locations with a marked 
proportion of low values at the island or island group scale included Swains for coral richness, Upolu for fish 
biomass, and Rose Atoll for coral cover. Note that in some cases, such low values may be "normal" for these 
locations and are not to be considered as a derogatory finding. Rose Atoll, for example, has a high cover of 
crustose coralline algae, a variable not presented here, and represents a unique area that contributes to the 
overall diversity and health of the archipelago (Vroom 2011). 



In the hotspot analysis at the scale of Bioregions, 51 hotspots were identified among the four variables and 
12 of those had a very low (<10%) probability of occurring by random chance (Table 4.3, Figure 4.9). Of the 



Table 4.3. Hotspot analysis summary table. The proportion of sites categorized as 'high' is given for each variable and Bioregion. 
Bioregions with a greater proportion of high sites than calculated for the jurisdiction overall (in italics) are defined as hotspots and 
highlighted in green. Adjacent p-values indicate the probability of each hotspot occurring merely by chance (* denotes values <10%). 



CO 

o 
E 

CO 

CO 

c 

CO 

o 

CD 

E 

< 



CO 

o 
E 

CO 

if) 





Coral Cover 


Coral Richness 


Fish Biomass 


Fish Rich 


mess 






Proportion High 
Overall = 0.22 


Proportion High 
Overall = 0.23 


Proportion High 
Overall = 0.10 


Proportion High 
Overall = 0.22 




Bioregion 


Proportion 

High for 

Bioregion 


p-value 


Proportion 

High for 

Bioregion 


p-value 


Proportion 
High for 
Bioregion p-value 


Proportion 

High for 

Bioregion 


p-value 


Hotspot 
variables 


1 


0.67* 


0.000 


0.22 


0.11 0.679 


0.29 


0.219 


3 


2 


0.42 


0.147 


0.40 


0.203 


0.08 


0.38 


0.147 


3 


3 


0.17 


0.50 


0.423 


0.00 


0.17 


1 


4 


0.00 


0.00 


0.04 


0.25 


0.462 


1 


5 


0.00 


0.22 


0.11 0.680 


0.11 


1 


6 


0.31 


0.351 


0.20 


0.19 0.296 


0.13 


2 


7 


0.25 


0.564 


0.00 


0.25* 0.081 


0.20 


2 


8 


0.33 


0.385 


0.20 


0.33* 0.082 


0.56* 


0.032 


3 


9 


0.00 


n/a 


0.00 


0.40 


0.314 


1 


10 


0.50 


0.252 


n/a 


0.25 0.398 


0.25 


0.639 


3 


11 


0.11 


0.25 


0.610 


0.06 


0.17 


1 


12 


0.42* 


0.070 


0.09 


0.00 


0.11 


1 


13 


0.00 


0.00 


0.06 


0.00 





14 


0.43 


0.232 


0.00 


0.14 0.589 


0.43 


0.193 


3 


15 


0.00 


0.00 


0.08 


0.15 





16 


0.54* 


0.001 


0.00 


0.19 0.190 




0.365 


3 


17 


0.05 


0.00 


0.13 0.528 


0.10 


1 


18 


0.13 


0.59* 


0.002 


0.18 0.213 


0.29 


0.218 


3 


19 


0.06 


0.75* 


0.003 


0.06 


0.39* 


0.088 


2 


20 


0.44 


0.156 


0.40 


0.348 


0.09 


0.18 


2 




Proportion High 
Overall = 0.43 


Proportion High 
Overall = 0.35 


Proportion High 
Overall = 0.09 


Proportion High 
Overall = 0.31 




Bioregion 


Proportion 

High for 

Bioregion 


p-value 


Proportion 

High for 

Bioregion 


p-value 


Proportion 
High for 
Bioregion p-value 


Proportion 

High for 

Bioregion 


p-value 


Hotspot 
variables 


21 


0.91* 


0.003 


0.55* 


0.092 


0.22 0.279 


0.11 


3 


22 


0.00 


0.67 


0.231 


n/a 


n/a 


1 


23 


1.00 


0.130 


0.50 


0.526 


0.00 


0.50 


0.437 


3 


24 


0.67 


0.129 


0.50 


0.277 


0.67* 0.002 


0.17 


3 


25 


0.00 


0.00 


0.00 


0.29 





26 


0.29 


0.43 


0.379 


0.00 


0.50 


0.169 


2 


27 


0.21 


0.14 


0.00 


0.33 


0.352 


1 


28 


1.00 


0.360 


0.00 


0.00 




0.249 


2 


29 


0.00 


0.17 


0.00 


0.00 





30 


0.00 


0.33 


0.00 


0.00 






30 Bioregions, none were identified as a hotspot for all four variables considered in the analysis. Ten of the 
Bioregions were hotspots for three variables. This included the SW coast of Tutuila from Cape Taputapu to 
Larsen Bay (Bioregions 1 and 2), the eastern tip of Tutuila (Bioregion 10), the northwestern coast of Tutuila 
east of Fagamalo (Bioregion 14), Swains Island (Bioregion 16), Ofu/Olosega Islands (Bioregion 18), and the 
north, northeast, and south facing coasts of Savai'i (Bioregions 23, 21, and 24). It should be noted however, 
that none of these three-variable hotspots had a high degree of certainty (p<0.10 in the re-sampling analysis) 



Legend 

Coral Cover 
Fish Biomass 



Coral Richness 
Fish Richness 



[Green shading) indicates the region 
is a hotspot for the variable. 



[Grey shading] indicates that the region 
lacks data for the variable. 



Number labels indicate assigned 
bioregions. 



Swains Island 
16 



21 

CC CR 




24 



22 



Savai'i 
I' 



23 



CC 


CR 


FB 


FR 


... 





30 



CC 


CR 


FB 


FR 



29 



CC 


CR 


FB 


FR 



Upolu 



28 






CC 


CR 


FB 


FR 



26 



CC 


CR 


FB 


FR 



27 



25 







Manu'a Islands 


CC 


CR 


18 

19 




FB 


FR 


20 






CC 


CR 


CC 


CR 




FB 


FR 


FB 


FR 



I Kilometers 
200 



Rose Atoll 
17 



168° W 



Figure 4.9. Fish and coral hotspots by Bioregion. 

for all three variables. Considering only hotspots that were highly robust to chance observations (p<0.10), 
three Bioregions stood out as hotspots for multiple variables: Aunu'u (Bioregion 8, for fish biomass and rich- 
ness), western Ta'u (Bioregion 19, for coral and fish richness), and the northern coast of Savai'i (Bioregion 
21 for coral cover and richness). 

Also of note, of the 30 Bioregions, only 5 were not considered a hotspot for any variable. These "coolspots" 
included 3 regions along the north and western coast of Upolu from Manono Island/Apolima Strait eastward 
past Apia to Fagaloa Bay (Bioregions 25, 29, and 30), and two regions on the north coast of Tutuila including 
the NW coast between Cape Taputapu and Fagamalo Village (Bioregion 15) and the north central coast of 
Tutuila around Fagasa Village (Bioregion 13). Note that, except for the small watershed directly around Fa- 
gasa Bay, these last two coolspots straddle a hotspot for 3 variables (Bioregion 14) and occur along relatively 
less densely populated coast compared to the rest of Tutuila. Additional coolspots probably exist around 
Samoa but could not be identified due to the low density of surveys. All the other Bioregions were considered 
hotspots for at least one or two variables. 



CONCLUSIONS 

Reef fish and corals are distributed unevenly around the islands of the Samoan Archipelago. Among the most 
notable hotspots for fish and corals using the variables considered here were northern Savai'i, parts of north- 
western and southwestern Tutuila, the eastern tip of Tutuila, the Manu'a Island group, and Swains Island, 
although many other regions were identified as important for particular variables. Smaller hotspots at the 
scale of individual sites were evident as well (e.g. a site with high coral cover surrounded by many low cover 
sites) but were not the focus of this study and should be the subject of separate, finer-scale analyses. The 



biogeographic hotspots and breakpoints identified here may be useful for several purposes including: plac- 
ing the existing network of marine protected areas (MPA) and marine managed areas (MMA) into regional 
context for the variables included here, prioritizing Bioregions requiring more detailed study, and supporting 
review of overall natural resources monitoring practices throughout the region. 

It is important to note that this is not a study of reef resiliency and these results alone should not be used as 
the basis for MPA network design. Nor should these results be interpreted to suggest that only places identi- 
fied as hotspots in this analysis are biologically significant and worthy of conservation (e.g. Rose Atoll). The 
relative importance of each variable studied here and targeted in future assessments will vary based on the 
objectives of a particular management or conservation application (Roberts et al. 2003, Wilson et al. 2009) 
and should be specified in a process beyond the scope of this document. An effort focused on protecting bio- 
diversity might focus on coral and fish richness hotspots, collecting more datasets that identify fish and coral 
to the species level, and using tools such as rarefaction curves to identify combinations of sites that efficiently 
represent the widest variety of species and communities (Beger et al. 2003). Alternatively, if the goal was to 
protect large larval sources for seeding of unprotected areas (e.g. to enhance sustainability or yield of fisher- 
ies), one might focus on hotspots of coral cover or fish biomass (Murray et al. 1999, Ochavillo et al. 2011) 
that overlap with source origins (Chapter 3). Once MPA/MMA network objectives are clearly identified, the 
general ecological variables presented throughout this report (Chapters 2-4) could be appropriately weighted 
and combined, more focused analyses conducted, additional key datasets collected, and a variety of MPA/ 
MMA design scenarios applied to identify combinations of management strategies and areas to achieve 
those objectives (e.g. Kendall et al. 2008, Watts et al. 2009). 

The objective of this assessment was principally to identify biogeographic patterns of a few variables rather 
than to determine explanatory processes behind them. Reef type, larval supply, wave exposure, ocean 
climate, water quality, herbivore abundance, community processes, and various human pressures such as 
fishing, development, pollution and other factors no doubt interact at each site to produce the bioregional 
patterns identified here. The sparsely populated islands of Manu'a and Savai'i had generally higher values 
for most variables than the more densely populated Islands of Tutuila and Upolu. Within Tutuila and Upolu, 
highest population density areas such as around Pago Pago Harbor and Apia had many of the lowest values. 
These patterns are consistent with anthropogenic effects on fish communities correlated with human popula- 
tion density as noted for other Pacific Islands (Williams et al. 2011). 



Some prior researchers have broken the coast of Tutuila into four sectors (NE, NW, SE, and SW) based pri- 
marily on seasonal patterns of wave exposure (e.g. Mundy 1996, Green 1996, 2002, Sabater and Tofaeono 
2006). The edges of these sectors correspond 
well to the boundaries of some Bioregions iden- 
tified here (1 and 15, 12 and 13, 3 and 4, and 
4 and 10) although the combined use of many 
datasets enabled the discrimination of many ad- 
ditional distinct regions of the coast for Tutuila 
based on the variables that were considered. In 
fact, it is apparent from our study using the com- 
bined results of many datasets that even adjacent 
Bioregions with apparently similar environmental 
conditions can have very different coral and fish 
assemblages. Along the north shores of Tutuila 
and Upolu, for example, lie regions with outward- 
ly similar environmental characteristics but very 
different values for the coral and fish variables 
considered here. Disentangling the many influ- 
ences shaping regional biogeographic patterns 
will be an important next step for research. This 

could be undertaken with the datasets gathered 

....... . £ .. . .-r- ,. r .. Image 1 7. Reef with high coral cover in Samoa . 

in this study through further stratification of sites Pho f credit . Joyce Sa * uelu Ah Leong MAF/FD 





Image 18. Reticulated Dascyllus sheltering on a branching coral. 
Photo credit: Matt Kendall, Biogeography Branch. 



by such factors as reef type, watershed in- 
fluences, and exposure regime and the use 
of correlation or MDS based analysis (e.g. 
Houk et al. 2010, Ochavillo et al. 2011). 
This is a key step for identifying those fac- 
tors that can be managed through agency, 
village, or MPA/MMA actions, identifying re- 
silient reefs, and predicting how areas may 
respond over time to management (Nys- 
trometal. 2008). 

In general, the fish variables showed a 
more equitable distribution of values at all 
scales of analysis relative to the coral vari- 
ables. There were less extreme changes 
in the spatial distribution offish abundance 
and richness values and fewer differences 
in fish communities between adjacent Bio- 
regions. Even the SFR dataset, based in 
village fish reserves around Samoa, did not 
appear to have consistently biased the results offish biomass toward higher values. This could be due to the 
mobility of adult fish and their potential for relatively rapid redistribution in response to fishing pressure, natu- 
ral disturbance events, or other density dependent factors. Corals in contrast, may only become redistributed 
during the larval phase. At least for harvested species, targeted fishing pressure alone may act to smooth out 
the hotspots for fish density or biomass (Williams et al. 2011). 

The processes of larval transport documented in Chapter 3 are probably at least partly responsible for some 
of the observed biogeographic patterns among islands in the fish and coral variables documented here. The 
analysis of larval connectivity used shelf-area within ~9 km grid cells to set the number of potential larvae 
around each source island as a simplifying assumption. However, the results here demonstrate that there 
can actually be considerable variability in reef condition and therefore presumably spawning potential at finer 
scales. Although the coarse spatial scale of the hydrodynamic model limits its application to interpretation of 
the interisland-scale data presented in this chapter, some consistent patterns are evident. 

Swains Island, an atoll that did not originate at the Samoan volcanic hotspot and may be geologically part 
of the Tokelau island group, lies in a somewhat different ocean climate (Chapter 2). Swains Island was also 
clearly the most physically isolated Bioregion based on currents and larval connectivity (Chapter 3) and, as 
would be expected, also had a very unique and isolated fish and coral community based on MDS analysis 
(and see Tribollet et al. 2010 for algal analysis). Rose Atoll, also a small island isolated from large upstream 
sources of fish and coral larvae, had relatively low values for coral and fish richness. This low biodiversity 
is consistent with predictions from Island Biogeography Theory (MacArthur and Wilson 1967) for small tar- 
get islands a great distance from larval sources. In addition, east to west trends along the archipelago of 
increasing values for coral cover, richness, and even fish biomass are aligned with the prevailing current 
in the region (South Equatorial Current, Chapter 3). Within Samoa, Upolu generally has lower values than 
the downstream island of Savai'i. Within American Samoa, Rose has lower fish and coral richness than the 
downstream islands of Manu'a and Tutuila. These patterns are also consistent with Island Biogeography 
Theory in that larger downstream islands provide big settlement targets for fish and coral larvae spawned to 
the east and then carried to destinations westward along the archipelago in the South Equatorial Current. As 
noted above, however, there is much more at work shaping the reef communities than currents. For example, 
studies at finer scales around Tutuila have measured effects from more localized processes such as fishing, 
coastal development and poor water quality (e.g. Houk et al. 201 0, Williams et al. 201 1 ). Reef zonation and 
the much larger reef flat and lagoon area of potential juvenile habitat around Samoa as well as the greater 
diversity of habitat types there may also play a role in enhancing some fish populations (Green 1996, Adams 



et al. 2006). Archipelago-wide benthic maps produced using a consistent scale and classification scheme are 
a critical information need to support an analysis of habitat differences. 



There are several key caveats to interpretation of these findings. First, based on the high, medium, and 
low scoring system, all analyses and results are inherently expressed only relative to the suite of available 
data in the archipelago. This scale is not defined relative to reef conditions globally or even more widely in 
the south Pacific region. Were additional data collected at much higher or lower quality sites for any of the 
individual studies, or if very different islands outside the study region were to be included, the classification 
of "high," "medium" and "low" categories would have been redistributed. For example, if an island group 
with many severely impacted sites of lower reef quality had been included, the values of all Samoan reefs 
would have been scored "relatively" higher. Also, because data from American Samoa is from reef slopes 
and data for Samoa is from lagoons, the scope of inference differs between these jurisdictions and compari- 
son of "high" values between them is not possible. Also, although results are based heavily on recent data 
to reflect current status, catastrophic events and major environmental shifts may alter the distributions of 
even these very general variables. Of note however, is the observation that datasets used here show very 
consistent and robust patterns despite occurring over a decade marked by several hurricanes and bleach- 
ing events (Chapter 2) and even older data show patterns consistent with those described here (e.g. Mundy 
1996). Next, although the variables included in the analysis were among those considered to be important 
in describing reef conditions, only six variables were analyzed and they are based on very general aspects 
of the marine community. Distribution and abundance of particular species or groups of concern should be 
addressed in separate studies (e.g. parrotfish as in Page 1998, algae as in Tribollet et al. 2010, or surgeon- 
fish as in Ochavillo et al. 2011) and may result in the perception of different Bioregions (smaller, larger, or 
with different breakpoints). Last, the great disparity in the concentration of survey effort among islands and 
the comparatively low numbers of surveys around Upolu and Savai'i resulted in Bioregions of variable size. 
Archipelago-wide sampling using a randomized design, consistent methodology, more detailed taxonomic 
information, and more equitable distribution of survey effort stratified based on reef area and zonation (e.g. 
lagoon, reef slope) is a critical information need. Such a monitoring design and coordinated effort is neces- 
sary to understand the shared marine resources across the archipelago and to make informed management 
decisions cooperatively among regional management entities. The ongoing lack of archipelago-wide moni- 
toring data collected with a consistent methodology and sampling design will hinder attempts at coordinated 
management between Samoa and American Samoa. 



ACKNOWLEDGEMENTS 



REFERENCES 



All "meta analysis" investigations rely on information from many studies. The acknowledgements section of 
each individual study used here should be consulted for a complete list of the contributions to each investiga- 
tion. Several people made specific contributions to this chapter. Peter Houk provided original data from AS 
EPA monitoring as well as helpful suggestions on an early draft of the chapter. Marlowe Sabater, former Chief 
of Fisheries with AS DMWR, generously provided original datasets and suggestions during the inception and 
design phases of the study. Risa Oram designed the survey method and collected much of the raw data for 
the Bioreconnaissance Assessments for the no-take MPA Program at DMWR. Paul Anderson facilitated data 
acquisition in Samoa with local contacts. Larry Claflin of the NOAA's Biogeography Branch provided a helpful 
review of draft material. 



Adams, A.J. , C.R Dahlgren, G.T Kellison, M.S. Kendall, C.A. Layman, J.A. Ley, I. Nagelkerken, and J.E. Serafy. 2006. 
Nursery function of tropical back-reef systems. Marine Ecology Progress Series 318: 287-301. 

Bare, A.Y., A.L. Grimshaw, J.R. Rooney, M.G. Sabater, D. Fenner, and B. Carroll. 2010. Mesophotic communities of the 
insular shelf at Tutuila, American Samoa. Coral Reefs 29: 369-377. 

Beger, M., G.R Jones, and PL. Munday. 2003. Conservation of coral reef biodiversity: a comparison of reserve selection 
procedures for corals and fishes. Biological Conservation 111: 53-62. 

Birkeland, C, P. Craig, D. Fenner, L.W. Smith, W.E. Kiene, and B. Riegl. 2008. Chapter 20: Geologic setting and eco- 
logical functioning of coral reefs in American Samoa. Pages 741-765 in B. Riegl and R. Dodge, Coral Reefs of the USA, 
Springer. 

Bohnsack, J.A. and S.P Bannerot. 1986. A stationary visual census technique for quantitatively assessing community 
structure of coral reef fishes. NOAA Technical Report NMFS 41. 15 pp. 

Brainard, R., and 25 others. 2008. Coral reef ecosystem monitoring report for American Samoa: 2002-2006. NOAA 
Special Report NMFS PIFSC. 472 pp. 

Carroll, B.P 2010. Coral Reef Fish Monitoring Report: 2006 - 2008. American Samoa Coral Reef Monitoring Program. 
Dept. of Marine and Wildlife Resources, Biological Report Series, No: 1002. 

Cheal, A.J., M. Aaron MacNeil, E. Cripps, M.J. Emslie, M. Jonker, B. Schaffelke, and H. Sweatman. 2010. Coral-mac- 
roalgal phase shifts or reef resilience: links with diversity and functional roles of herbivorous fishes on the Great Barrier 
Reef. Coral Reefs 29: 1005-1015. 

Clarke, K.R. 1993. Non-parametric multivariate analyses of changes in community structure. Australian Journal of Ecol- 
ogy 18: 117-143. 

Clarke, K.R. and R.N. Gorley. 2006. PRIMER v6: User Manual/Tutorial. PRIMER-E, Plymouth. 

Conservation International- Pacific Islands Programme, Ministry of Natural Resources and Environment, Secretariat of 
the Pacific Regional Environmental Programme. 2010. Priority sties of conservation in Samoa: Key biodiversity areas. 
Apia, Samoa. 32 pp. 

Craig, PC, J.H. Choat, L.M. Axe, and S. Saucerman. 1997. Population biology and harvest of the coral reef surgeonfish 
Acanthurus lineatus in American Samoa. Fishery Bulletin 95: 680-693. 

Craig, PC, G. DiDonato, D. Fenner, and C Hawkins. 2005. The state of coral reef ecosystems of American Samoa, pp. 
312-337. In Waddell, J. (ed.), 2005. The status of coral reef ecosystems of the US and Pacific freely associated states. 
NOAA Tech Report NOS NCCOS 11. 522 pp. 

Craig, P. (editor). 2009. Natural history guide to American Samoa. 3rd Edition. National Park of American Samoa, De- 
partment of Marine and Wildlife Resources, and American Samoa Community College. Pago Pago, American Samoa. 
131 pp. 

Edinger, E.N., J. Jompa, G.V. Limmon, W. Widjatmoko, and M.J. Risk. 1998. Reef degradation and coral biodiversity in 
Indonesia: effects of land-based pollution, destructive fishing practices, and changes overtime. Marine Pollution Bulletin 
36:617-630. 

Fenner, D. 2008. Annual Report for 2006 of the Territorial Coral Reef Monitoring Program for American Samoa, Benthic 
Section. Dept. of Marine & Wildlife Resources, American Samoa. 

Fenner, D., et al. 2008. The state of coral reef ecosystems of American Samoa. P. 307-351 in J.E. Waddell and A.M. 
Clarke (eds.), The State of Coral Reef Ecosystems of the United States and Pacific Freely Associated States: 2008. 
NOAA Technical Memorandum NOS NCCOS 73. NOAA/NCCOS Center for Coastal Monitoring and Assessment's Bio- 
geography Team. Silver Spring, MD. 569 pp. 




Fenner, D. 2009a. Annual Report for 2008 of the Territorial Coral Reef Monitoring Program for American Samoa, Benthic 
Section. Dept. of Marine & Wildlife Resources, American Samoa. 

Fenner D. 2009b. Annual Report for 2007 of the Territorial Coral Reef Monitoring Program for American Samoa, Benthic 
Section. Dept. of Marine & Wildlife Resources, American Samoa. 

Friedlander, A.F. and E.E. DeMartini. 2002. Contrasts in density, size, and biomass of reef fishes between the north- 
western and the main Hawaiian Islands: the effects of fishing down apex predators. Marine Ecology Progress Series 
230: 253-264. 

Green, A. 1 996. Status of the coral reefs of the Samoan Archipelago. American Samoa Department of Marine and Wild- 
life Resources. Pago Pago, American Samoa. 126 pp. 

Green, A. 2002. Status of coral reefs on the main volcanic islands of American Samoa: A resurvey of long term monitor- 
ing sites. Report Prepared for the American Samoa Government, Department of Marine and Wildlife Resources. Pago 
Pago, American Samoa. 86 pp. 

Houk, P. and C. Musberger. 2008. Assessing the effects of non-point source pollution on American Samoa's coral reef 
communities. Report to American Samoa Environmental Protection Agency, 38 pp. 

Houk, P., C. Musberger, and P. Wiles. 2010. Water quality and herbivory interactively drive coral reef recovery patterns 
in American Samoa. Plos One 5:1-9 e13913. 

Johannes, R.E. 2002. The renaissance of community-based marine resource management in Oceania. Annual Review 
of Ecology and Systematics 33: 317-340. 

Kendall, M.S., K.A. Eschelbach, G. McFall, J. Sullivan, and L. Bauer. 2008. MPA design using sliding windows: Case 
study designating a research area. Ocean & Coastal Management 51: 815-825. 

King, M. and U. Faasili. 1998. A network of small, community-owned village fish reserves in Samoa. Parks 8: 11-16. 

Legendre, P. and L. Legendre. 1998. Numerical Ecology. Second English Edition. Elsevier, Amsterdam. 

MacArthur, R.H. and E.O.Wilson. 1967. The Theory of Island Biogeography. Princeton, N.J.: Princeton University Press. 

McCune, B. and J.B. Grace. 2002. Analysis of ecological communities. MjM Software Design, Gleneden Beach, OR. 

Mundy, C. 1996. A quantitative survey of the corals of American Samoa. University of Queensland, Australia. Report 
Prepared for the American Samoa Government, Department of Marine and Wildlife Resources. Pago Pago, American 
Samoa. 25 pp. 

Murray, S.N., R.F. Ambrose, J.A. Bohnsack, L.W. Botsford, M.H. Carr, G.E. Davis, PK. Dayton, D. Gotshall, D.R. 
Gunderson, M.A. Hixon, J. Lubchenco, M. Mangel, A. MacCall, D.A. McArdle, J.C. Ogden, J. Roughgarden, R.M. Starr, 
M.J. Tegner, and M.M. Yoklavich. 1999. No-take reserve networks: sustaining fishery populations and marine ecosys- 
tems. Fisheries 24: 11-25. 

NOAA, NCCOS (National Centers for Coastal Ocean Science). 2005. Shallow-water benthic habitats of American Sa- 
moa, Guam, and the Commonwealth of the Northern Mariana Islands (CD ROM). NOAA Technical Memorandum NOS 
NCCOS 8, Biogeography Team, Silver Spring, Maryland. 

Nystrom, M., N.A.J. Graham, J. Lokrantz, and A.V. Norstrom. 2008. Capturing the cornerstones of coral reef resilience: 
linking theory to practice. Coral Reefs 27:795-809. 

Ochavillo, D., S. Tafaeono, M. Sabater, and E.L. Trip. 2011. Population structure of Ctenochaetus striatus (Acanthuri- 
dae) in Tutuila, American Samoa: The use of size-at-age data in multi-scale population size surveys. Fisheries Re- 
search 107: 14-21. 

Oram, R.G. 2008. Marine protected area master plan: A manual to guide the establishment and management of no- 
take marine protected areas. American Samoa Government, Department of Marine and Wildlife Resources. Biological 
Report Series 2008-01. January 2008. Pago Pago, American Samoa. 64 pp. 

Page, M. 1998. The biology, community structure, growth and artisanal catch of parrotfishes of American Samoa. Ameri- 
can Samoa Government, Department of Marine and Wildlife Resources. Pago Pago, American Samoa. 52 pp. 



Pedersen Planning Consultants. 2000. American Samoa Watershed Protection Plan. Prepared for American Samoa 
Environmental Protection Agency and Coastal Zone Management Program. Pago Pago, American Samoa. Volumes 
1-4. 

Roberts, C, S. Andelman, G. Branch, R. Bustamante, J.C. Castilla, J. Dugan, B. Halpern, H. Leslie, K. Lafferty, J. 
Lubchenco, D. McArdle, H. Possingham, M. Ruckelshaus, and R.R. Warner. 2003. Ecological criteria for evaluating 
candidate sites for marine reserves. Ecological Applications 13: S199-214. 

Sabater, M. 2010. American Samoa Archipelagic Fishery Ecosystem Report. American Samoa Government, Depart- 
ment of Marine and Wildlife Resources. Biological Report Series No: 1001. Pago Pago, American Samoa. 48 pp. 

Sabater, M. and S. Tofaeono. 2006. Spatial variation in biomass, abundance and species composition of "key reef spe- 
cies" in American Samoa. American Samoa Government, Department of Marine and Wildlife Resources. Pago Pago, 
American Samoa. 54 pp. 

Sabater, M.G. and S. Tofaeono. 2007. Effects of scale and benthic composition on biomass and trophic group distribu- 
tion of reef fishes in American Samoa. Pacific Science 61 : 503-520. 

Samuelu, J. 2003. Village fish reserve report for Vaisala (re-survey). August 2003. Assessment, Research and Manage- 
ment Section Fisheries Division. Ministry of Agriculture, Forestry, Fisheries, and Meteorology. 6 pp. 

Samuelu, J. and M Sapatu. 2008. Status of coral reefs in Samoa 2008. Global Coral Reef Monitoring Network 2008 
Report. Fisheries Division, Ministry of Agriculture and Fisheries. Apia, Samoa. 33 pp. 

Spalding, M.D., H.E. Fox, G.R.Allen, N. Davidson, Z.A. Ferafia, M. Finlayson, B.S. Halpern, M.A. Jorge, A. Lombana, 
S.A. Lourie, K.D. Martin, E. McManus, J. Molnar, C.A. Recchia, and J. Robertson. 2007. Marine ecoregions of the world: 
A bioregionalization of coastal and shelf areas. Bioscience 57: 573-583. 

Tribollet, A.D., T Schils, and PS. Vroom. 2010. Spatio-temporal variability in macroalgal assemblages of American 
Samoa. Phycologia 49: 574-591. 

Veron, J.E.N. 2000. Corals of the World, Vol. 1-3. Australian Institute of Marine Science and CRR Ltd. 

Veron, J.E.N., L.M. DeVantier, E. Turak, A.L. Green, S. Kininmonth, M. Stafford-Smith, and N. Peterson. 2009. Delineat- 
ing the coral triangle. Galaxea, Journal of Coral Reef Studies 11: 91-100. 

Vroom, PS. 2011. "Coral Dominance": A dangerous ecosystem misnomer? Journal of Marine Biology 2011: Article ID 
164127,8 pp. 

Watts, M.E., I.R. Ball, R.R. Stewart, C.J. Klein, K.A. Wilson, C. Steinback, R. Lourival, L. Kircher, and H.P Possingham. 
2009. Marxan with Zones: Software for optimal conservation based land- and sea-use zoning. Environmental Modeling 
& Assessment 24: 1513-1521. 

Whaylen, L. and D. Fenner. 2005. The American Samoa Coral Reef Monitoring Program. Report Prepared for the De- 
partment of Marine and Wildlife Resources and Coral Reef Advisory Group, American Samoa. March 2005. Pago Pago, 
American Samoa. 34 pp. + appendices. 

Wilkinson, C. 2008. Status of coral reefs of the world: 2008. Global Coral Reef Monitoring Network and Reef and Rain- 
forest Research Centre, Townsville, Australia, 296 p. 



Williams, I.D., B.L. Richards, S.A. Sandin, J.K. Baum, R.E. Schroeder, M.O. Nadom, B. Zgliczynski, P. Craig, J.L. Mc- 
llwain, and R.E. Brainard. 2011. Differences in reef fish assemblages between populated and remote reefs spanning 
multiple archipelagos across the Central and Western Pacific. Journal of Marine Biology 2011: Article ID 826234, 14 
pages. Doi:10.1155/2011/826234. 

Wilson, K.A., J. Carwardine, and H.P. Possingham. 2009. Setting Conservation Priorities. Annals of the New York Acad- 
emy of Sciences. 1162: 237-364. 



; 



I- 



m nt^oiviftiLJ 



^^^^^^^^^^^M^^^^_ 


K.5i 


^Hu^^^l ^H " 


U'fi 



Image 19. Fagatele Bay National Marine Sanctuary sign. 
Photo credit: Matt Kendall, NOAA Biogeography. 



The Existing Network of Marine Protected Areas in American Samoa 

Matthew Poti 1 , Matthew S. Kendall 2 , Gene Brighouse 3 , Tim Clark 4 , Kevin Grant 3 , Lucy Jacob 5 , Alice Lawrence 5 , Mike Reynolds 4 and 

Selaina Vaitautolu 5 
INTRODUCTION 

Marine Protected Areas and Marine Managed 
Areas (hereafter referred to collectively as 
MPAs) are considered key tools for maintaining 
sustainable reef ecosystems. By limiting or pro- 
moting particular resource uses and activities in 
different areas and raising awareness issues on 
reef sustainability within MPAs, managers can 
promote long term resiliency. Multiple local and 
federal agencies have eagerly embraced MPA 
concepts in Samoa and American Samoa with 
a diversity of MPAs now in place across the ar- 
chipelago from the village and local community 
level to national protected areas and those with 
international significance. Many of the different 
MPAs in the network were created through inde- 
pendent processes and therefore have different 
objectives, have been in existence for different 
lengths of time, have a wide range of sizes and 
protection regulations, and have different man- 
agement authorities. Each contributes to the di- 
verse mosaic of marine resource management in the region (See Text Box: Summary of MPA Programs). 

Understanding the variety offish, coral, and habitat resources that this multifaceted network of MPAs encom- 
passes is critical for assessing the scope of current protection and thoughtfully designing additional network 
elements. Here we seek to summarize what aspects of the coral reef ecosystem are protected by MPAs indi- 
vidually, through brief summaries of each MPA, and then collectively, through analysis of the combined area 
encompassed by all MPAs. Based on the available datasets used to broadly characterize the biogeography of 
the region in the previous chapters and appendices of this assessment, key concepts of MPA network design 
including biogeographic representation and replication will be addressed. Representation is the idea that at 
least part of each distinct biogeographic region should be included in a 'complete' network of MPAs. Replication 
is the idea that there should be more than one MPA in each distinct biogeographic region. Replication spreads 
protection within each region thereby reducing the risk to the network that is associated with localized degrada- 
tion at any one site. 

In this chapter of the assessment we focus our analysis only on the MPAs of American Samoa. While Samoa is 
a key part of the MPA landscape in the archipelago as demonstrated in Chapters 3 and 4, two key datasets are 
in need of further development. First, benthic maps similar in spatial scope and categorical detail to those avail- 
able in American Samoa are needed to inventory the protected habitats of Samoa. Second, MPA boundaries in 
Samoa must be made available for analysis, but at present many are proprietary at the village level as part of 
the Community Based Fisheries Management Program (King and Faasili 1998, Samuelu 2003). 

The objectives of this chapter were to: 

1) Characterize the reef fishes, corals, habitats, and other key features of each existing MPA relative to all of 
American Samoa. 

2) Evaluate the distribution of MPA sites in the context of the biogeographic regions and ecological hotspots 
defined in Chapter 4 and identify key areas not currently in the network. 

3) Summarize the area of reef ecosystem, by bottom type and reef type, that is currently protected relative to 
American Samoa overall. 

1 NOAA/NOS/NCCOS/CCMA/Biogeography Branch and Consolidated Safety Services, Inc., Fairfax, VA, under NOAA 
Contract No. DG133C07NC0616 

2 NOAA/NOS/NCCOS/CCMA Biogeography Branch 

3 Fagatele Bay National Marine Sanctuary 

4 National Park of American Samoa 

5 American Samoa/Department of Marine and Wildlife Resources 



SUMMARY OF MPA PROGRAMS 

There are several agencies involved in MPA management and planning in American Samoa. Here we provide a brief summary of 
these programs and their objectives in American Samoa. They are separated into those that are exclusively Territorial in manage- 
ment authority and those that are co-managed by Territorial and Federal agencies. 

Territorial MPAs 

Department of Commerce Special Management Areas (SMAs) 

The American Samoa Government authorized the American Samoa Coastal Management Program through the American Samoa 
Coastal Management Act of 1990 (ASCA § 24.0503) to designate, as Special Management Areas (SMAs), places that "possess 
unique and irreplaceable habitat, products or materials, offer beneficial functions or affect the cultural values or quality of life 
significant to the general population of the Territory and fa'a Samoa" (ASAC § 26.0221). Three such places - Leone Pala, Nu'uuli 
Pala, and Pago Pago Harbor- have been designated as SMAs as of January 2011. Although no formal management plans exist 
for these SMAs, projects within these areas must comply with standards described in ASAC § 26.0221. 

Department of Marine and Wildlife Resources Community-Based Fisheries Management Program (CFMP) 
The American Samoa Government created the Community-Based Fisheries Management Program within the Department of 
Marine and Wildlife Resources (DMWR) in 2001 so that the "historical, cultural, and natural resources" of American Samoa and 
its marine environment would be "protected, managed, controlled, and preserved for the benefit of all people of the Territory and 
future generations" (ASAC § 24. 10). The CFMP promotes sustainable management of marine resources and enhances fisheries 
stocks through mechanisms such as seasonal closures and fishing restrictions within designated reserves, as agreed upon by 
village leaders and DMWR (ASAC § 24. 10). As of January 2011, eleven CFMP reserves were in existence around Tutuila. These 
reserves are sometimes referred to as village marine protected areas (VMPAs). Fishing restrictions within reserves may include 
prohibiting destructive fishing methods (e.g. use of bleach, poison, or dynamite), use of scuba gear and nets, the breaking up of 
corals, and fishing by outsiders. The number of CFMP reserves, their boundaries, and regulations are as of January 2011 but are 
regularly modified to meet local needs. 

Department of Marine and Wildlife Resources No-Take Marine Protected Area Program 

The American Samoa Government established the No-Take MPA Program within DMWR in 2006 in response to Governor Tauese 
Sunia's recommendation to protect 20% of American Samoa's coral reefs as no-take MPAs (Sunia 2000). The goal of the No-Take 
MPA Program is to "ensure protection of unique, various, and diverse coral reef habitat and spawning stocks" through the creation 
of a network of no-take areas (Sunia 2000, Oram 2008). The No-Take MPA Program is currently using the authority under the 
Community-Based Fisheries Management Program (under ASAC § 24. 1001) to enforce no-take regulations. 

Other Territorial MPAs 

Two additional MPAs are present in American Samoa but are not part of the formal programs listed above. One is a private reserve 
established in 1985 atAlega Bay by a local restaurant owner, Tisa Fa'amuli. This reserve is hereafter referred to as Alega Private 
Marine Reserve. The other is a small marine park adjacent to the Ofu unit of the National Park that was established by territorial 
legislation in 1994 to protect the "unique coral reef wildlife habitat while enabling the public to enjoy the natural beauty of the site" 
(ASCA § 18.0214). At present time the Ofu Vaoto Territorial Marine Park has no enforcement, monitoring, or management plan. 

Federal or Federal/Territorial Co-Managed MPAs 

National Marine Sanctuaries 

The Office of National Marine Sanctuaries (ON MS) is authorized by the National Marine Sanctuaries Act (NMSA, 1972 with sub- 
sequent amendments) to designate and protect areas of the marine environment with "special national significance" due to their 
"conservation, recreational, ecological, historical, scientific, cultural, archeological, educational, or aesthetic qualities" as national 
marine sanctuaries (16 U.S.C. 1431 et seq.). The Sanctuaries Program is intended to "improve the conservation, understanding, 
management, and wise and sustainable use of marine resources", to "enhance public awareness, understanding, and apprecia- 
tion of the marine environment", and to "maintain for future generations the habitat, and ecological services, of the natural as- 
semblage of living resources that inhabit these areas" (16 U.S.C. 1431 et seq.). In American Samoa, the Fagatele Bay National 
Marine Sanctuary (FBNMS) was designated in 1986 and is co-managed with the Territorial Government through the American 
Samoa Department of Commerce (15 C.F.R. 922.100-104). Additional potential areas were brought to the attention of ONMS 
via public meetings in 2009. A Site Selection Working Group of the Sanctuary Advisory Council evaluated each of the suggested 
areas using NMSA criteria to determine if they possess qualities of national significance worthy of sanctuary designation. Also, 
per Presidential Proclamation 8337, the marine areas of the Rose Atoll Marine National Monument shall be added to FBNMS in 
accordance with the NMSA (16 U.S.C. 1431 et seq., Proclamation No. 8337). 

National Parks 

The National Park Service (NPS) was created in 1916 to "conserve the scenery and the natural and historic objects and the wild 
life therein and to provide for the enjoyment of the same in such manner and by such means as will leave them unimpaired for 
the enjoyment of future generations" (16 U.S.C. 1). Under the direction of Congress, the NPS conducted a feasibility study in 
1986-87 to identify areas of significant natural and cultural resources in American Samoa and to assess the suitability of these 
areas for inclusion in a national park (NPS 1988). Through this process and consultation with village leaders and the Govern- 
ment of American Samoa, the NPS identified two areas (north-central Tutuila from Vatia Bay to Fagasa Bay and the south-central 
portion of Ta'u) that best met the criteria for inclusion in a national park. Additional areas, including the south coast of Ofu, were 
suggested as possible future additions (NPS 1988). Under recommendation of the NPS, the National Park of American Samoa 
(NPSA) was designated by Congress in 1988 to "preserve and protect the tropical forest and archaeological and cultural re- 
sources of American Samoa, and of associated reefs, to maintain the habitat of flying foxes, preserve the ecological balance of 
the Samoan tropical forest, and, consistent with the preservation of these resources, to provide for the enjoyment of the unique 
resources of the Samoan tropical forest by visitors from around the world" (16 U.S.C. 410qq). The NPSA currently consists of 3 
separate units - the areas on Tutuila and Ta'u identified by the feasibility study and the south coast of Ofu (NPS 1997). In 2002 
Congress authorized the addition of portions of the islands of Ofu and Olosega to the NPSA (16 U.S.C. 410qq-1). Formal estab- 
lishment of these additions awaits approval of a lease with the local villages. The NPSA is managed by the NPS in consultation 
with the territorial DMWR and the individual villages. Management of the NPSA maintains traditional Samoan customs and allows 
subsistence fishing by native American Samoans using traditional tools and methods in accordance with rules established by the 
NPS and village leaders. 



National Wildlife Refuges and Marine National Monuments 

The United States Fish and Wildlife Service (USFWS) mission is, working with others, to conserve, protect and enhance fish, 
wildlife, and plants and their habitats for the continuing benefit of the American people. In 1966 Congress authorized the USFWS 
through the National Wildlife Refuge System Administration Act (1966, with subsequent amendments) to "administer a national 
network of lands and waters for the conservation, management, and where appropriate, restoration of the fish, wildlife, and plant 
resources and their habitats within the United States for the benefit of present and future generations of Americans" (16 U.S.C. 
668dd-668ee). In American Samoa, Rose Atoll National Wildlife Refuge (NWR) was established in 1973 through a cooperative 
agreement between the American Samoa Government and the USFWS (RANWR 1974). Rose Atoll NWR has been closed to the 
public since its establishment to protect the fish and wildlife in the refuge. 

In 2009 Rose Atoll Marine National Monument (MNM), which includes the Rose Atoll NWR was established by Presidential 
Proclamation 8337 to protect objects of historic and scientific interest under the authority of the Antiquities Act of 1906 (16 U.S.C. 
431). The NWR is managed exclusively by USFWS, but management of the MNM is more complex. The Proclamation gave 
the Department of Interior (USFWS) management responsibility for the MNM in consultation with the Department of Commerce 
(NOAA). However, NOAA was given management responsibility for fisheries outside of the NWR, and the Secretary of Commerce 
was tasked with initiating the process of adding the marine areas of the MNM to Fagatele Bay National Marine Sanctuary. 



METHODS 

Inventory of existing MPAs 

Working with local MPA practitioners, the American Samoa Coastal Zone Management Program, and Island 
GIS User Group, we obtained boundary maps (GIS shapefiles) and implementation documents for the 23 
MPAs existing in American Samoa as of January 2011. This included eleven Community-Based Fisheries 
Management Program (CFMP) Reserves, one No-Take MPA, one Marine National Monument (MNM), one 
National Wildlife Refuge (NWR), one National Marine Sanctuary (NMS), three National Park units, one pri- 
vate marine reserve, three Special Management Areas (SMAs), and one Territorial Marine Park (Figure 5.1, 
Table 5.1 , Appendix D). 



Existing MPAs 

Community-Based Fisheries Management Program (CFMP) Reserves 

] Fagamalo No-Take MPA 
] AS DOC Special Management Areas (SMAs) 
_| National Park of American Samoa (NPSA) Units 
_| Fagatele Bay National Marine Sanctuary (NMS) 

Alega Private Marine Reserve 

Rose Atoll Marine National Monument (MNM) 
] Rose Atoll National Wildlife Refuge (NWR) 
_| Ofu Vaoto Marine Park 



Swains Island 



N 

A 



Rose Atoll MNM 



n?i 



171°W 170°W 



. Rose Atoll 
iT NWR 

Rose Atoll 



169°W 



168°W 167°W 



Tutuila and Aunu'u 



NPSA Tutuila Unit 



Fagamalo No-Take MPA 
Fagamalo 




\ Vatia Sailele 

Masausi /\ oa 



Pago Pago Amaua 

Harbor SMA &Auto 

Matu'u & 
Faganeanea 

Nu'uuli Pala SMA 



Masausi Aoa 

lega I Alofau 



Leone Pala SMA 



Fagatele Bay NMS 



Manu'a Islands 



I Nl 



PSA Ofu Unit 



Ofu Vaoto 
Marine Park 



R 



NPSA Ta'u Unit 



5 10 



Figure 5.1 . Existing MPAs in American Samoa as of January 201 1 . 



Table 5.1. Existing MPAs in American Samoa as of January 2011. 



ffiff CTS 


Level of 
Government 




Management 
Authority 


3d 




Community-Based Fisheries 
Management Program 


Territorial 


DMWR, villages 


11: Alofau, Amanave, Amaua & Auto, Aoa, Aua, 
Fagamalo, Masausi, Matu'u & Faganeanea, Po- 
loa, Sailele, Vatia 


Marine National Monuments 


Federal 


NOAA, USFWS 


1: Rose Atoll 


National Marine Sanctuaries 


Federal/Territorial 
Co-Managed 


NOAA, ASDOC 


1 : Fagatele Bay 


National Park of American 
Samoa 


Federal 


ASNPS 


3: Ofu, Ta'u, Tutuila 


National Wildlife Refuge 
System 


Federal 


USFWS 


1: Rose Atoll 


No-Take MPA Program 


Territorial 


DMWR 


1: Fagamalo 


Private Marine Reserves 


Private 


Alega village 


1 : Alega Bay 


Special Management Areas 


Territorial 


ASCMP, villages 


3: Leone Pala, Nu'uuli Pala, Pago Pago Harbor 


Territorial Marine Parks 


Territorial 


DPR, DMWR 


1:Ofu 



These boundaries were then reviewed, modified as necessary in the GIS, and confirmed for accuracy by 
their corresponding management authorities. 

For each MPA, we created a site profile that summarizes key information focused on the MPA's biogeograph- 
ic setting. For each 2-page profile, we first provide an overview that includes a site map and short description 
of MPA size, location, implementation date, and rationale. We also identify general characteristics of adjacent 
lands that may impact the marine environment including size and condition of watersheds, population density, 
erosion and runoff potential, and notable human use impacts (e.g. major sources of pollution). This informa- 
tion was obtained from the American Samoa Watershed Protection Plan prepared for the American Samoa 
Environmental Protection Agency in 2000 (Pedersen Planning Consultants 2000a-c). In addition, because 
pigs are a major source of nearshore pollution affecting coral reef ecosystems, the density of domestic pigs 
in watersheds adjacent to each MPA is noted. Pig density is described using four categories (high = >50 pigs/ 
km 2 , medium = 12-50 pigs/km 2 , low = <12 pigs/km 2 , and zero) assigned to watershed data from the ASEPA 
Piggery Compliance Program (ASEPA Piggery Compliance Program 2011) using the natural breaks function 
in ArcGIS (Figure 5.2). In addition, key natural resource regulations for each MPA are listed, specifically those 
that pertain to fishing or the ecological reasons for establishing the site. Original designation documents for 
each site should be consulted for a complete list of regulations. A more comprehensive description of each 
individual MPA including implementation, purpose, management practices, fishing regulations, biological and 
socio-economic monitoring, community involvement, and current and future projects, is provided in Appendix D. 

The main focus of each site profile is on the reef ecosystem habitats, reef fish, and coral communities pro- 
tected within each MPA. Boundary maps of each MPA were overlaid upon recently completed benthic maps 
of American Samoa (Appendix B, NOAA NCCOS 2005). Boundaries were used to clip portions of habitat 
polygons inside each MPA. Benthic maps for American Samoa categorize bottom features on the basis of 
2 attributes: 1) "structure" which refers to predominant physical composition of the feature and includes 15 
mutually exclusive bottom types such as patch reef, pavement, and sand, and 2) "zone" which refers to each 
feature's position on the insular shelf and includes 13 mutually exclusive categories such as lagoon, reef 
crest, fore reef (locally referred to as reef slope), and bank/shelf (Appendix B, NOAA NCCOS 2005). We 
summarized the areas within each MPA by structure type using pie charts and compared the proportions of 
benthic habitats within each MPA to those of American Samoa overall. A hierarchical approach was taken 
wherein the relative proportions of all coral reef and hardbottom structures are discussed, followed by those 
structure types representative of only the highest quality reef habitats. These include aggregate reef, patch 
reef, aggregated patch reefs, and spur and groove which all typically have high structural rugosity and often 
possess high coral cover and relatively more abundant and diverse fish communities compared to other 
hardbottom types. These four bottom types are hereafter referred to collectively as "coral reef habitats". 



high 



140 



120 



100 



80 



medium 



low 



zero 



E 
"55 



en 

§ 60 
Q 

b. 



40 



20 



■■■■■■■-- 



Watershed 

Figure 5.2. Density of pigs (pigs/km 2 ) in piggeries by watershed for Tutuila and Manu'a watersheds. Watersheds were classified as 
having high, medium, or low pig density using the natural breaks function in ArcGIS. 

Similarly, we summarized the zonation of the coral reef and hardbottom structures within each MPA in pie 
charts and compared the proportions of these reef zones to American Samoa overall. All zones are provided 
but description focused on reef flats, due to the importance of this habitat for village use by gleaners, and 
the fore reef, a high-value reef zone which has the greatest diversity of reef fish and is the focus of most reef 
monitoring around American Samoa. 

We also evaluated general reef fish and coral variables at each MPA compared to American Samoa overall. 
These variables were the same as those considered in Chapter 4: coral cover, coral richness, fish biomass, 
and fish richness classified into high, medium, and low categories (see Chapter 4 for a description of survey 
data and classification methods). Our goal was only to describe each MPA relative to the rest of American 
Samoa; therefore, only datasets with many, widely spread sites around Tutuila or the other islands of Ameri- 
can Samoa were used. Many MPAs are also characterized more individually with customized studies and 
methodology, but those studies did not enable island-wide comparison due to differences in site selection, 
methodology, or timing of surveys. Consequently, such studies are not included in the analysis but are noted 
in each profile for those interested in more detailed site characterization. 

Survey sites within each MPA were categorized as high, medium, or low for fish and coral variables, plotted 
on maps with the MPA boundaries, and summarized in pie charts. For MPAs with four or more survey sites, 
the proportions of high, medium, and low values within each MPA were compared to the proportions for 
American Samoa overall (see Chapter 4) using pie charts. Survey results for MPAs with too few sites to make 
sound comparisons are provided but are not compared to American Samoa overall. In addition, the spatial 



distribution of survey sites within MPAs was evaluated and key locations where greater effort is required are 
noted. 

While it would be useful to compile species lists and cumulative numbers of species observed for each MPA, 
our profiles did not include this information for two main reasons. First, because of the very different survey 
methods used among studies and lack of consistent species level information, creating rarefaction curves 
was not possible. Second, the very different levels of survey effort among the MPAs have resulted in severe 
inconsistency in total area surveyed (e.g. an MPA with 30 surveys inside it will have a much larger species list 
than one with only a few surveys). As a result of these limitations, our analysis was focused on more general 
summary variables described above for evaluating sites. 

Last, for each MPA we identified the biogeographic region (hereafter "Bioregion") in which it lies based on the 
archipelago-wide analysis of fish and coral data in Chapter 4. Also noted is the "hotspot" status for the reef 
fish and coral variables analyzed in Chapter 4 and any similarities between the fish and coral communities in 
the Bioregion of the MPA and those in other Bioregions. Key results from Chapter 3 on potential sources and 
destinations for coral and fish larvae are also noted. 



How much of American Samoa is protected in the MPA Network? 

To evaluate the proportion of the total area of potential reef ecosystem in American Samoa protected by 
MPAs, we used a pie chart to summarize the total area within MPAs versus the area outside. Potential reef 
ecosystem was defined as areas shallower than 150 m, which approximates the depth limit for photic and 
mesophotic reef communities in the region (Bare et al. 201 0, Mesophotic Coral Ecosystems 201 0, Appendix 
B). For most MPAs this is the same value as the total area since they only encompass regions shallower than 
150 m deep. Areas were categorized by structure type. For simplicity, some map categories were aggregated 
into major groups. These were coral reef habitats (aggregate reef, patch reef, aggregated patch reefs, spur 
and groove), other hardbottom types (pavement, pavement with patch reefs, pavement with sand channels, 
reef rubble, rock/boulder), and unconsolidated substrates (mud, sand with scattered coral/rock, sand). We 
repeated this comparison using only the coral reef category to examine how much coral reef habitat is pro- 
tected relative to the total area of coral reef habitat around American Samoa. Along with these comparisons 
we also provided charts showing the proportions of potential reef ecosystem and coral reef habitat with no- 
take restrictions and with other fishing restrictions. 

Which biogeographic regions and ecological hotspots are represented in the MPA network? 

The coastline of American Samoa can be divided into 20 ecologically distinct biogeographic regions (termed 
"Bioregions") based on the distribution of reef fish and corals (Chapter 4). Thirty-six ecological hotspots 
among these 20 Bioregions have been defined relative to American Samoa overall for each of four variables: 
coral cover (hotspot in n = 10 Bioregions), coral richness (n = 6), fish biomass (n = 10), and fish richness (n 
=10). Boundaries of the existing MPAs were overlaid onto the Bioregions and ecological hotspots to deter- 
mine which were already represented in the MPA network and which lacked an MPA and may be considered 
as gaps in coverage. 

Size and regulatory comparisons among MPAs 

MPAs in American Samoa have a wide range of sizes. We compared the sizes among MPAs by scaling the 
size of the habitat pie chart for each MPA relative to the total area of potential reef ecosystem within it. Scal- 
ing the size of pie charts in this manner allowed us to evaluate the relative contributions of each MPA to the 
overall network and also to compare the proportions of benthic habitats among the MPAs while taking into 
consideration the total area protected. This is significant because an MPA that has a high proportion of reef 
habitats but that is very small may actually protect a smaller reef area than an MPA that has a lower propor- 
tion of reef habitats but a much larger overall area. In addition, we identified which MPAs or parts of MPAs 
provide the strongest level of protection, complete no-take. 



RESULTS: BENTHIC HABITATS OF AMERICAN SAMOA 

Coral reef and hardbottom 



structures together com- 
prise -30% of the almost 
400 km 2 of mapped ben- 
thic habitat around Ameri- 
can Samoa (Figure 5.3a). 
Coral reef structures cover 
almost twice as much area 
as hardbottom structures, 
with aggregated patch 
reefs, spur and groove, 
and aggregate reef cover- 
ing -7%, -6%, and -5% 
of the area, respective- 
ly. However, the major- 
ity (-60%) of the mapped 
benthic habitat around 
American Samoa is al- 
gal plain in the bank/shelf 
zone. Nearly half of the 
coral reef and hardbottom 
around American Samoa 
is found in the bank/shelf 
zone, -20% is in the fore 
reef, and -10% is in each 
of the reef flat and bank/ 
shelf escarpment zones 
(Figure 5.3b, see Appendix 
B Figure B.1 for cross sec- 
tion of zones). 



(a) Total Mapped Area by 
Benthic Structure Type 



(b) Zonation of 

Coral Reef and Hardbottom 




Reef Flat 
(9%) 



Fore Reef 
(22%) 




Bank/Shelf 
(49%) 



Bank/Shelf 

Escarpment 

(10%) 



59% 



* small percentages of back reef, 
bank/shelf basin, pinnacle, and 
reef crest not shown 



-394 km 2 



-29% of total area is coral reef and hardbottom 



-> -115 km 2 



Legend 

Benthic Structure Types 

Aggregate Reef 



Pavement 



^£, Rock/Boulder 



| Aggregated Patch Reefs | jj 1 *^ 



Emergent Vegetation 



Individual Patch Reef 



Spur and Groove 



a 



Pavement with 
Sand Channels 

Reef Rubble 



Algal Plain 



Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Figure 5.3. (a) Proportion of mapped benthic structure types in American Samoa overall, (b) 
Proportion of coral reef and hardbottom in each reef zone. Structure types or zones representing 
<1% of the total mapped area are not shown. 



The zonation of coral reef and hardbottom structures varies with shelf geomorphology among the islands of 
American Samoa. The progression of reef zones from shoreline to reef slope is similar for Tutuila and the 
Manu'a Islands. However, the bank/shelf around Tutuila extends much farther from the shoreline than it does 
around the Manu'a Islands and includes pinnacle and bank/shelf basin zones not found on the other islands. 
As a result of the narrower shelf, a greater percentage of coral reef and hardbottom is in the reef flat zone 
around the Manu'a Islands compared to Tutuila. The two steep-sided atolls in American Samoa, Swains 
Island and Rose Atoll, are also fundamentally different features. At Rose Atoll, almost two-thirds of the coral 
reef and hardbottom is in the back reef, whereas -10% is in each of the reef crest, fore reef, and bank/ 
shelf zones. The coral reef and hardbottom at Swains Island in contrast is mostly in the reef flat, with lesser 
amounts in the reef crest and fore reef and none in the completely enclosed lagoon area. 



RESULTS: SITE CHARACTERIZATIONS 

Territorial MPAs 

Alega Private Marine Reserve 

Overview 

Alega Private Marine Reserve is located in the southeast of Tutuila in Alega Bay and extends from Vaiola 
Point to Tifa Point (Figure 5.4). It was initiated by Tisa Fa'amuli in 1 985 to protect the coral reef ecosystem in 
Alega Bay from overfishing and other destructive practices. By maintaining a low level of subsistence fishing, 




Alega Private Marine Reserve Boundary 



Legend 

Benthic Structure Types 

Aggregate Reef 



Pavement 



Aggregated Patch Reefs | J ^"^ 



Individual Patch Reef 
Spur and Groove 



Pavement with 
Sand Channels 



Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
I Mud Un 



Sand with 
Scattered Coral/Rock 



Sand 
Artificial 
known 



Fish/Coral Data 

O Survey Sites 

Site Values for each Variable 

Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.4. Benthic habitat (by structure type) and fish and coral survey data within Alega Private Marine Reserve. Coral cover, coral 
richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shading), or 
low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from ASEPA, KRS, 
and REA. 



the reserve allows for sustainable use of the marine resources in the reserve by the village community now 
and in future generations. The reserve fronts a -1.3 km 2 watershed in minimally impacted condition with low 
human population density. There are no domestic pigs reported in the watershed. In addition to natural sedi- 
mentation caused by highly erosive soils on the steep slopes of the watershed, nearshore waters may also 
have been slightly impacted by urban runoffs and insufficiently treated wastewater. Only subsistence fishing 
with traditional methods by village members is allowed within the reserve. Commercial fishing and fishing by 
outsiders are prohibited within Alega Private Marine Reserve. 



Habitat Composition, Reef Fish, 
and Coral Communities 
This small MPA is dominated by 
coral reef and hardbottom struc- 
tures which together comprise 
-88% of the area within the re- 
serve (Figure 5.5a). Coral reef 
structures comprise -41% of the 
area and include aggregate reef 
(-37%) and patch reef (-4%). 
In addition, pavement covers 
-44% of the area. In compari- 
son, these three structure types 
comprise less than 15% of 
American Samoa overall. About 
50% and -42% of the coral reef 
and hardbottom in the reserve 
are in the reef flat and fore reef 
zones, respectively, compared 
to only -9% and -22% around 
American Samoa (Figure 5.5b). 



(a) Total Area by 

Benthic Structure Type 



(b) Zonation of 

Coral Reef and Hardbottom 



12% 




Bank/Shelf 



37% 



44% 




Reef Flat 
(50%) 



Fore Reef 
(42% 



Reef Crest 
(3%) 



-88% of total area is coral reef and hardbottom 
-0.15 km 2 ► -0.13 km 2 

Figure 5.5. (a) Proportion of benthic structure types in Alega Private Marine Reserve, 
(b) Proportion of coral reef and hardbottom in each reef zone. Structure types or zones 
representing <1% of the total area are not shown. 



Only 3 surveys were located within Ale- 
ga Private Marine Reserve. Coral data 
at these sites includes one medium and 
two low values for cover and one medi- 
um value for richness. Fish data includes 
one low and two medium values for both 
biomass and richness (Figure 5.6). The 
small sample size greatly limits the scope 
of these findings and does not allow com- 
parisons with American Samoa overall. 
Additional, more widely spread surveys 
are needed to more fully characterize the 
reef fish and coral communities within this 
MPA. 



Coral Cover Coral Richness Fish Biomass Fish Richness 

3(3) 1(1) 3(3) 3(3) 







Alega 
Private 
Marine 
Reserve 



Figure 5.6. Fish and coral data collected in Alega Private Marine Reserve. 
Pie charts depict the proportions of high (red), medium (pink), and low (white) 
values for coral cover, coral richness, fish biomass, and fish richness. Number 
labels represent the number of studies and sites (in parentheses) comprising 
each pie chart. 



Biogeographic Characteristics 

Alega Private Marine Reserve is a small part of a biogeographic region that is a hotspot for fish richness 

(Bioregion 4, Chapter 4). 



Alofau CFMP Reserve 

Overview 

The village of Alofau is located in SE Tutuila on the eastern side of Fagaitua Bay. The Alofau CFMP reserve 
(Figure 5.7) was established in 2001 to "conserve the marine resources in the ocean and on the village reef" 
(ASDMWR 2002a). The -0.3 km 2 reserve extends north to south from Asasama Point at the boundary with 
Pagai village to Uea Point on Cape Fogausa with a seaward boundary that includes the entire reef area (AS- 
DMWR 2002a). It fronts the eastern end of a -4.9 km 2 watershed in intermediately impacted condition with 
moderate human population density and a medium density of pigs. In addition to natural sedimentation caused 
by highly erosive soils on the steep slopes of the watershed, nearshore waters have also been impacted by 
urban runoffs and insufficiently treated wastewater. Fishing is prohibited within the reserve with the excep- 




Legend 

Benthic Structure Types 

Aggregate Reef 



Pavement 



Aggregated Patch Reefs | jf™^ 



Individual Patch Reef 



Spur and Groove 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
I Emergent Vegetation 
Algal Plain 

I Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 

Site Values for each Variable 

Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.7. Benthic habitat (by structure type) and fish and coral survey data within the Alofau CFMP reserve. Coral cover, coral rich- 
ness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shading), or low 
(white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from ASEPA and REA. 



tion of occasional Saturday openings for subsistence fishing. Destructive fishing methods, including the use 
of bleach, poisons, and dynamite, are banned and fishing by outsiders is also prohibited (ASDMWR 2002a). 



(a) Total Area by 

Benthic Structure Type 



(b) Zonation of 

Coral Reef and Hardbottom 



23% 




Channel 
(3% 



Lagoon 
(19%) 



16% 



Fore Reef 
(17% 



Reef Crest 
(9%) 



34% 




Reef Flat 
(52%) 



-0.32 km 2 



~94%> of total area is coral reef and hardbottom 



-> -0.30 km 2 



Habitat Composition, Reef Fish, and 
Coral Communities 
The benthic habitat within the Alo- 
fau CFMP reserve is dominated by 
coral reef and hardbottom structures, 
which together comprise -95% of the 
area within the reserve (Figure 5.8a). 
Coral reef structures comprise -37% 
of the area and include aggregate 
reef (-19%), aggregated patch reefs 
(-16%), and spur and groove (~3%). 
In comparison, these three structure 
types comprise only -18% of the 
mapped benthic habitat around Ameri- 
can Samoa. About 52% and -17% of 
the coral reef and hardbottom in the 
reserve are in the reef flat and fore 
reef zones, respectively, compared 
to -9% and -22% around American 
Samoa (Figure 5.8b). Also of note, an- 
other -20% of the coral reef and hard- 
bottom structures are in the lagoon. 

Only two fish and coral surveys were lo- 
cated within the Alofau CFMP reserve. 
Coral data at these sites includes two 
low values for cover and one medium 
value for richness. Fish data includes 
one low and one high value for both 
biomass and richness (Figure 5.9). 
The small sample size greatly limits the 
scope of these findings and does not al- 
low comparisons with American Samoa 
overall. Additional, more widely spread 
surveys are needed to more fully char- 
acterize the reef fish and coral commu- 
nities within this MPA. 

Biogeographic Characteristics 

The Alofau CFMP reserve lies in a biogeographic region (Bioregion 6, Chapter 4) that is a hotspot for coral 
cover and fish biomass. The region's fish and coral communities are similar to those around north-central 
Tutuila, where the Tutuila unit of the National Park and the Vatia CFMP reserve are located. 

Additional References 

Orcutt 1993, Andrews 2004, Musburger 2004, Houk 2010 



Figure 5.8. (a) Proportion of benthic structure types in the Alofau CFMP reserve, 
(b) Proportion of coral reef and hardbottom in each reef zone. Structure types or 
zones representing <1% of the total area are not shown. 



Coral Cover Coral Richness Fish Biomass Fish Richness 



2(2) 



1(1) 



2(2) 



2(2) 



Alofau 
CFMP 
Reserve 





J 3 



Figure 5.9. Fish and coral data collected in the Alofau CFMP reserve. Pie 
charts depict the proportions of high (red), medium (pink), and low (white) 
values for coral cover, coral richness, fish biomass, and fish richness. Number 
labels represent the number of studies and sites (in parentheses) comprising 
each pie chart. 



Amanave CFMP Reserve 

Overview 

The village of Amanave is found on the western tip ofTutuila. The Amanave CFMP reserve (Figure 5.10) was 
established in 2009 to ensure the availability of the resources in the reserve for the villagers today and in 
the future. The -0.3 km 2 reserve extends offshore approximately 50 yards between the boundary with Poloa 
village and the boundary with Fa'ilolo village. It fronts a ~1 .0 km 2 watershed in intermediately impacted condi- 
tion with moderate human population density and medium density of pigs. In addition to natural sedimenta- 
tion caused by highly erosive soils on the steep slopes of the watershed, nearshore waters have also been 
impacted by urban runoffs and insufficiently treated wastewater. The reserve is closed to all commercial and 
recreational fishing apart from when it is opened for subsistence fishing one month every year. 




Amanave CFMP Reserve Boundary 



0.25 

Legend 

Benthic Structure Types 



Aggregate Reef 



0.5 



I Pavement 



Aggregated Patch Reefs g| "e^ 



Individual Patch Reef 
Spur and Groove 



m 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 

Site Values for each Variable: 

Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.10. Benthic habitat (by structure type) and fish and coral survey data within the Amanave CFMP reserve. Coral cover, coral 
richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shading), or 
low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from KRS and REA. 



Habitat Composition, Reef Fish, 
and Coral Communities 
Coral reef and hardbottom struc- 
tures together comprise -97% 
of the area within the Amanave 
CFMP reserve (Figure 5.11a). 
Coral reef structures, primar- 
ily spur and groove, comprise 
-22% of the area. Also of note, 
almost half of the benthic habitat 
within the reserve is covered by 
reef rubble. In comparison, spur 
and groove and reef rubble cov- 
er -6% and -2%, respectively, of 
American Samoa overall. About 
44% and -20% of the coral reef 
and hardbottom in the reserve 
are in the reef flat and fore reef 
zones, respectively, compared 
to -9% and -22% around Ameri- 
can Samoa (Figure 5.11b). 



(a) Total Area by 

Benthic Structure Type 



(b) Zonation of 

Coral Reef and Hardbottom 



47% 




Channel 
(2%) 



Bank/Shelf 
(17% 



Fore Reef 
(20%) 




Reef Flat 
(44%) 



Reef Crest 
(16%) 



-0.34 km 2 



-97% of total area is coral reef and hardbottom 



-> -0.33 km 2 



Figure 5.11. (a) Proportion of benthic structure types in the Amanave CFMP reserve, 
(b) Proportion of coral reef and hardbottom in each reef zone. Structure types or zones 
representing <1% of the total area are not shown. 



Coral Cover Coral Richness Fish Biomass Fish Richness 



2(2) 



1(1) 



2(2) 



2(2) 



Amanave 

CFMP 

Reserve 



j 






Only two fish and coral surveys were lo- 
cated within or just outside the Amanave 
CFMP reserve and these were both lo- 
cated near the eastern end of the re- 
serve. Coral data at these sites includes 
one medium and one high value for cover 
and one high value for richness. Fish data 
includes one low and one medium value 
for biomass and two medium values for 
richness (Figure 5.12). The small sam- 
ple size greatly limits the scope of these 
findings and does not allow comparisons 
with American Samoa overall. Additional, 
more widely spread surveys are needed 
to more fully characterize the reef fish and 
coral communities within this MPA. 

Biogeographic Characteristics 

The Amanave CFMP reserve lies in a distinct biogeographic region (Bioregion 1, Chapter 4) that is a regional 
hotspot for coral cover as well as fish biomass and richness. The region's fish and coral communities are 
representative of southwestern Tutuila. 

Additional References 

Randall and Devaney 1974, Orcutt 1993 



Figure 5.12. Fish and coral data collected in the Amanave CFMP reserve. 
Pie charts depict the proportions of high (red), medium (pink), and low (white) 
values for coral cover, coral richness, fish biomass, and fish richness. Number 
labels represent the number of studies and sites (in parentheses) comprising 
each pie chart. 



Amaua and Auto CFMP Reserve 

Overview 

The villages of Amaua and Auto are located in SE Tutuila on the western side of Fagaitua Bay. In response to 
concerns over declines in fish and shellfish populations from overfishing, the Amaua and Auto CFMP reserve 
(Figure 5.13) was established in 2003 to "manage, protect, and preserve the fish, shellfish, and the coastal 
area of the village of Amaua and Auto" (ASDMWR 2003a). The -0.4 km 2 reserve extends from the western 
boundary of Auto to the eastern boundary of Amaua with a seaward boundary ranging from 250 yards to the 
edge of the reef area (ASDMWR 2003a). It fronts the western end of a -4.9 km 2 watershed in intermediately 
impacted condition with a moderate human population density and medium density of pigs. In the part of 
the watershed fronted by the reserve there is moderate to high potential for runoff and erosion because of 




J Amaua & Auto CFMP Reserve Boundary ^ 



0.25 

Legend 

Benthic Structure Types 

Aggregate Reef 



Pavement 



| Aggregated Patch Reefs | J »£££% 



Individual Patch Reef 
Spur and Groove 



^ 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 
Site Values for each Variable; 
Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.13. Benthic habitat (by structure type) and fish and coral survey data within the Amaua and Auto CFMP reserve. Coral 
cover, coral richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink 
shading), or low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from 
MPABR and TCRMP. 



the highly erosive soils and steep slopes. Nearshore waters are also impacted to a lesser extent by urban 
runoffs and insufficiently treated wastewater. The reserve is closed to all commercial and recreational fishing 
apart from when it is opened for subsistence fishing at certain times of the year. Destructive fishing methods, 
including the use of bleach and poisons, are banned (ASDMWR 2003a). 



Habitat Composition, Reef Fish, 
and Coral Communities 
The Amaua and Auto CFMP 
reserve is dominated by coral 
reef and hardbottom structures, 
which together comprise -95% 
of the area within the reserve 
(Figure 5.14a). Coral reef struc- 
tures, primarily aggregate reef, 
comprise -22% of the area. Also, 
pavement covers -50% of the 
area. In comparison, aggregate 
reef and pavement comprise 
-5% and -7%, respectively, of 
American Samoa overall. About 
63% and -24% of the coral reef 
and hardbottom in the reserve 
are in the reef flat and fore reef 
zones, respectively, compared to 
-9% and -22% around American 
Samoa (Figure 5.14b). 



(a) Total Area by 

Benthic Structure Type 



(b) Zonation of 

Coral Reef and Hardbottom 



23% 




Fore Reef 
(24%) 



Reef Crest 
(13%) 




Reef Flat 
(63%) 



49% 



~95%o of total area is coral reef and hardbottom 
-0.37 km 2 ► -0.35 km 2 

Figure 5.14. (a) Proportion of benthic structure types in the Amaua and Auto CFMP re- 
serve, (b) Proportion of coral reef and hardbottom in each reef zone. Structure types or 
zones representing <1% of the total area are not shown. 



Amaua & Auto 

CFMP 

Reserve 



Coral Cover 
2(2) 

3 



Coral Richness Fish Biomass 

2 (2) 2 (2) 



Fish Richness 
2(2) 






Only two surveys were located within 
the Amaua and Auto CFMP reserve, and 
these were in the aggregate reef close 
to the seaward boundary of the reserve. 
Coral data at these sites includes one 
low and one high value for cover and 
two low values for richness. Fish data 
includes one low and one medium value 
for biomass and two low values for rich- 
ness (Figure 5.15). The small sample 
size greatly limits the scope of these find- 
ings and does not allow comparisons 
with American Samoa overall. Additional, 
more widely spread surveys are needed 
to more fully characterize the reef fish 
and coral communities within this MPA. 



Biogeographic Characteristics 

The Amaua and Auto CFMP reserve lies in a biogeographic region (Bioregion 6, Chapter 4) that is a hotspot 
for coral cover and fish biomass. The region's fish and coral communities are similar to those around north- 
central Tutuila, where the Tutuila unit of the National Park and the Vatia CFMP reserve are located. 



Figure 5.15. Fish and coral data collected in the Amaua and Auto CFMP re- 
serve. Pie charts depict the proportions of high (red), medium (pink), and low 
(white) values for coral cover, coral richness, fish biomass, and fish richness. 
Number labels represent the number of studies and sites (in parentheses) 
comprising each pie chart. 



Additional References 

Andrews 2004, Musburger 2004, Houk 2010 



Aoa CFMP Reserve 

Overview 

The village of Aoa is in NE Tutuila. The Aoa CFMP reserve (Figure 5.16) was established in 2005 to improve 
the coral reef habitat and restore fish and invertebrate stocks within the reserve. The -0.3 km 2 reserve in- 
cludes the entire Aoa Bay between Motusaga Point and Palau Point and extends offshore approximately 50 
yards from the reef edge. The reserve fronts a -2.2 km 2 watershed in intermediately impacted condition with 
moderate human population density and a medium density of pigs. In some areas of the watershed there is 
moderate to high potential for periodic natural erosion due to the soil type and steep slopes, but sedimenta- 
tion is moderated by the Aoa wetland. Nearshore waters are also impacted to a lesser extent by urban runoffs 
and insufficiently treated wastewater. The reserve is closed to all commercial and recreational fishing apart 
from when it is opened for subsistence fishing at certain times of the year. 




0.25 

Legend 

Benthic Structure Types 



Aggregate Reef 



Pavement 



Aggregated Patch Reefs g| "e^ 



Individual Patch Reef 
Spur and Groove 



s 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Habitat Composition, Reef Fish, 
and Coral Communities 
The Aoa CFMP reserve is domi- 
nated by coral reef and hardbot- 
tom structures, which together 
comprise -80% of the area within 
the reserve (Figure 5.17a). Coral 
reef structures in the form of ag- 
gregate reef cover -8% of the 
area, while pavement and reef 
rubble together comprise -70% 
of the area. In comparison, ag- 
gregate reef covers -5% of 
American Samoa overall. About 
73% and -10% of the coral reef 
and hardbottom in the reserve 
are in the reef flat and fore reef 
zones, respectively, compared 
to -9% and -22% around Ameri- 
can Samoa (Figure 5.17b). 



(a) Total Area by 

Benthic Structure Type 



(b) Zonation of 

Coral Reef and Hardbottom 




35% 



Bank/Shelf 
(5%) 

Fore Reef 
(10%) 



-0.34 km 2 



Reef Crest 
(11 



-80% of total area is coral reef and hardbottom 




Reef Flat 
(73%) 



-> -0.27 km 2 



Figure 5.17. (a) Proportion of benthic structure types in the Aoa CFMP reserve, (b) Pro- 
portion of coral reef and hardbottom in each reef zone. Structure types or zones repre- 
senting <1% of the total area are not shown. 

There were no fish and coral 

surveys from the island-wide comparison located within the Aoa CFMP reserve. 

Biogeographic Characteristics 

The Aoa CFMP reserve is located in a distinct biogeographic region that was a hotspot for coral richness 
(Bioregion 1 1 , Chapter 4). The region's coral communities are similar to those in NW Tutuila, where the Faga- 
malo CFMP reserve and No-Take MPAare located. 

Additional References 

Randall and Devaney 1974, Birkeland et al. 1987, Orcutt 1993, Birkeland et al. 1994, Birkeland et al. 2003, 

Houk2010 



Figure 5.16. Benthic habitat (by structure type) within the Aoa CFMP reserve. No reef fish or coral surveys from the island-wide 
comparison were located within the Aoa CFMP reserve. 




Aua CFMP Reserve 

Overview 

The village of Aua is located on the eastern side of Pago Pago Harbor. The Aua CFMP reserve (Figure 5.18) 
was established in 2002 to "manage, protect, and preserve the fish, shellfish, and the coastal area of the vil- 
lage of Aua" (ASDMWR 2003b). The -0.2 km 2 reserve extends from Ava Point to Muliti Point with a seaward 
boundary ranging from 200 yards to the edge of the reef area (ASDMWR 2003b). It fronts the northeast por- 
tion of a -10.4 km 2 watershed in extensively impacted condition. In addition to natural sedimentation caused 
by highly erosive soils on the steep slopes of the watershed and increased surface runoffs due to extensive 
urbanization, nearshore water quality has also been severely degraded by nutrient and chemical discharges 
by the tuna canneries and other historical industrial, commercial, and military activities adjacent to Pago 
Pago Harbor. There is a medium density of pigs in the watershed compared to all of American Samoa. The 







"■ "il*J < Ava Point 




Legend 

Benthic Structure Types 



Aggregate Reef 



Pavement 



Rock/Boulder 



Aggregated Patch Reefs D ■ P ^"I 6 "* W [ th 3 Emergent Vegetation 

\ -kt -d Paten Keets ^♦^ 



Individual Patch Reef 



Spur and Groove 



Pavement with 
Sand Channels 

Reef Rubble 



Algal Plain 
Mud 



□ 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



reserve is closed to all commercial and recreational fishing apart from when it is opened for subsistence 
fishing at certain times of the year. Destructive fishing methods, including the use of bleach, poisons, and 
explosives are banned. The use of scuba gear and nets for fishing and the breaking up of corals for fishing 
are also banned, as is fishing by outsiders (ASDMWR 2003b). 



Habitat Composition, Reef Fish, 
and Coral Communities 
Coral reef and hardbottom struc- 
tures together comprise -82% of 
the area within the Aua CFMP re- 
serve (Figure 5.19a). Coral reef 
structures in the form of patch reefs 
comprise -7% of the area, while 
reef rubble is predominant and cov- 
ers -60% of the area. In compari- 
son, individual patch reef and reef 
rubble cover less than 1 % and -2%, 
respectively, of American Samoa 
overall. About 75% and -16% of 
the coral reef and hardbottom in the 
reserve are in the reef flat and fore 
reef zones, respectively, compared 
to -9% and -22% around American 
Samoa (Figure 5. 1 9b). Also of note, 
an additional -8% of the coral reef 
and hardbottom is in the lagoon. 



(a) Total Area by 

Benthic Structure Type 



10% 



13% 




(b) Zonation of 

Coral Reef and Hardbottom 



Lagoon 
(8%) 



Fore Reef 
(16%) 




Reef Flat 
(75%) 



61% 



-0.23 km 2 



-82% of total area is coral reef and hardbottom 



> -0.19 km 2 



Figure 5.19. (a) Proportion of benthic structure types in the Aua CFMP reserve, (b) 
Proportion of coral reef and hardbottom in each reef zone. Structure types or zones 
representing <1% of the total area are not shown. 



There were no fish and coral surveys from the island-wide comparison located within the Aua CFMP reserve. 

Biogeographic Characteristics 

The Aua CFMP reserve is in a biogeographic region that includes Pago Pago Harbor and is a hotspot for fish 
biomass and has a unique coral community (Bioregion 5, Chapter 4). Note that high fish biomass may be due 
to the ban on sale offish from the harbor and while the coral community is "unique" relative to elsewhere in 
American Samoa it is not necessarily "healthy". 

Additional References 

Mayor 1924, Dahl and Lamberts 1977, Mc- 
Connaughey 1993, Orcutt 1993, Green et al. 
1997a, Fisk and Birkeland 2002, Coles et al. 
2003, Andrews 2004, Birkeland et al. 2004, 
Cornish and DiDonato 2004, Green et al. 2005 




Figure 5.18. Benthic habitat (by structure type) within the Aua CFMP reserve. No reef fish or coral surveys from the island-wide 
comparison were located within the Aua CFMP reserve. 



Image 20. Pago Pago Harbor and Rainmaker Mountain near Aua. 
Photo credit: Matt Kendall, NOAA Biogeography. 




Fagamalo CFMP Reserve 

Overview 

The village of Fagamalo is located in NW Tutuila. The Fagamalo CFMP reserve (Figure 5.20) was estab- 
lished in 2003 to "preserve the coral reef area of the village of Fagamalo" and amended in 2010 (ASDMWR 
2003c). The -0.4 km 2 reserve extends from Niutulua Point in the west to Tafaga Point in the east and offshore 
approximately 200 yards. It abuts the Fagamalo No-Take MPAand fronts a -2.1 km 2 watershed in pristine 
condition with very low human population density and a low density of pigs. While human impacts are mini- 
mal in the watershed, there is moderate to high potential for runoff and erosion because of the soil types and 
steep slopes with sediment transport into Fagamalo Bay primarily via Matavai Stream. The reserve is closed 
to all commercial and recreational fishing apart from when it is opened for subsistence fishing at certain 




0.25 

Legend 

Benthic Structure Types 

Aggregate Reef 



Pavement 



| Aggregated Patch Reefs £ j ^^ 



Individual Patch Reef 



Spur and Groove 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



□ 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 

Site Values for each Variable; 

Coral Cover 
Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.20. Benthic habitat (by structure type) and fish and coral survey data within the Fagamalo CFMP reserve. Coral cover, coral 
richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shading), or 
low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from CRSR, REA, 
and TCRMP. 



times of the year. Destructive fishing methods, including the use of bleach, electrical shocking devices, and 
explosives, are banned. In addition, fishing within Fagamalo streams is also prohibited (ASDMWR 2003c). 



(a) Total Area by 

Benthic Structure Type 



12% 




51% 



(b) Zonation of 

Coral Reef and Hardbottom 



Channel 
(1%) 

Bank/Shelf 
(19%) 



Shoreline 
Intertidal 
(14%) 




Reef Flat 
(4%) 

Reef Crest 
(3%) 



Fore Reef 
(59%) 



-0.38 km 2 



~85%> of total area is coral reef and hardbottom 



-> -0.32 km 2 



Figure 5.21. (a) Proportion of benthic structure types in the Fagamalo CFMP reserve, (b) 
Proportion of coral reef and hardbottom in each reef zone. Structure types or zones repre- 
senting <1% of the total area are not shown. 



Habitat Composition, Reef 
Fish, and Coral Communities 
The Fagamalo CFMP reserve 
is dominated by coral reef and 
hardbottom structures, which 
together comprise -85% of 
the area within the reserve 
(Figure 5.21a). Coral reef 
structures, primarily aggre- 
gate reef, cover just over half 
of the area. In comparison, 
aggregate reef covers only 
-5% of the mapped benthic 
habitat around American Sa- 
moa. About 4% and -59% of 
the coral reef and hardbot- 
tom in the reserve are in the 
reef flat and fore reef zones, 
compared to -9% and -22% 
around American Samoa 
(Figure 5.21b). Also of note, 
-14% of the coral reef and 
hardbottom is in the shoreline 
intertidal zone. 



Only 3 surveys were located within the 
Fagamalo CFMP reserve. Coral data at 
these sites includes one low and two me- 
dium values for cover and two medium 
values for richness. Fish data includes 
one high and two low values for biomass 
and one low and two high values for rich- 
ness (Figure 5.22). The small sample 
size greatly limits the scope of these find- 
ings and does not allow comparisons 
with American Samoa overall. Additional, 
more widely spread surveys are needed 
to more fully characterize the reef fish 
and coral communities within this MPA. 



Biogeographic Characteristics 

The Fagamalo CFMP reserve lies in a distinct biogeographic region (Bioregion 14, Chapter 4) that is a 
hotspot for coral cover and fish biomass and richness. The region's coral communities are similar to those in 
NE Tutuila, where the Masausi, Sailele, and Aoa CFMP reserves are located. 

Additional References 

Orcutt 1993, Fiskand Birkeland 2002, Musburger 2004 



Coral Cover Coral Richness Fish Biomass Fish Richness 
3(3) 2(2) 3(3) 3(3) 







Fagamalo 

CFMP 

Reserve 



Figure 5.22. Fish and coral data collected in the Fagamalo CFMP reserve. 
Pie charts depict the proportions of high (red), medium (pink), and low (white) 
values for coral cover, coral richness, fish biomass, and fish richness. Number 
labels represent the number of studies and sites (in parentheses) comprising 
each pie chart. 




Fagamalo No-Take MPA 

Overview 

The village of Fagamalo is located in NW Tutuila. The village signed a cooperative agreement with DMWR 
in May 2010 to join the No-Take MPA Program. The boundaries were finalized in December 2010 and the 
agreement was made to activate the no-take regulations (ASAC § 24.1008 (c)(i)) for an initial period of 10 
years. The completion of the revised management plan is still underway and expected completion is May 
2011. The -2.9 km 2 no-take boundary extends from Tafaga Point (in the west) to Oali'i (in the east) and ~2 
km offshore (Figure 5.23). It fronts a -2.1 km 2 watershed in pristine condition with very low human popula- 
tion density and a low density of pigs. While human impacts are minimal in the watershed, there is moderate 
to high potential for runoff and erosion because of the soil types and steep slopes. All types of fishing and 
extractive use are prohibited within the no-take MPA. 




Fagamalo No Take MPA Boundary 



0.5 

Legend 

Benthic Structure Types 



□ 



Aggregate Reef 



Pavement 



Aggregated Patch Reefs @ "e^ 



Individual Patch Reef 



Spur and Groove 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 

Site Values for each Variable: 

Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.23. Benthic habitat (by structure type) and fish and coral survey data within the Fagamalo No-Take MPA. Coral cover, coral 
richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shading), or 
low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from REA. 



(a) Total Area by 

Benthic Structure Type 



52% 




(b) Zonation of 

Coral Reef and Hardbottom 

Shoreline 
Intertidal 
(2%) \ Fore Reef 
" x (11%) 



39% 



Bank/Shelf 
(86%) 



Habitat Composition, Reef 
Fish, and Coral Communities 
Coral reef and hardbottom 
structures together comprise 
-46% of the area within the 
Fagamalo No-Take MPA (Fig- 
ure 5.24a). Coral reef struc- 
tures comprise -44% of the 
area and include aggregated 
patch reefs (-39%) and ag- 
gregate reef (-5%). However, 
over half of the mapped ben- 
thic habitat within the MPA is 
covered by algal plain. In com- 
parison, aggregate reef and ag- 
gregated patch reefs cover less 
than 15% of American Samoa 
overall. Only -1% and -11% of 
the coral reef and hardbottom 
in the MPA are in the reef flat 
and fore reef zones, respective- 
ly. Almost 90% of the coral reef 

and hardbottom is in the bank/shelf (Figure 5.24b). In comparison, only -50% of the coral reef and hardbot- 
tom around American Samoa is in the bank/shelf, -9% is in the reef flat and -22% is in the fore reef. 




i%i% 



-2.9 km 2 



-46% of total area is coral reef and hardbottom 



-> -1.3 km 2 



Figure 5.24. (a) Proportion of benthic structure types in the Fagamalo No-Take MPA. (b) 
Proportion of coral reef and hardbottom in each reef zone. Structure types or zones rep- 
resenting <1% of the total area are not shown. 



Only 4 surveys were located within or just 
outside the Fagamalo No-Take MPA and, 
of these, one was in the reef and hardbot- 
tom formations nearest to the shoreline 
and three were carried out on the offshore 
bank made up of reef and hardbottom 
formations. Coral cover and fish richness 
are relatively higher at these sites com- 
pared to all of American Samoa, whereas 
fish biomass values are relatively lower 
(Figure 5.25). No coral richness data was 
collected with these surveys. Additional, 
more widely spread surveys are needed 
to adequately characterize the reef fish 
and coral communities within this MPA. 



Coral Cover Coral Richness Fish Biomass Fish Richness 



Fagamalo 

No-Take 

MPA 



American 

Samoa 

overall 



1(4) 



0(0) 



1(4) 



1(4) 



e 



6 (339) 



5(137) 



(£(£ 




6 (347) 



Figure 5.25. Comparison offish and coral data collected in the Fagamalo No- 
Take MPA to data from all of American Samoa. Pie charts depict the propor- 
tions of high (red), medium (pink), and low (white) values for coral cover, coral 
richness, fish biomass, and fish richness. Number labels represent the number 
of studies and sites (in parentheses) comprising each pie chart. 



Biogeographic Characteristics 

The Fagamalo No-Take MPA lies in a distinct biogeographic region that is a hotspot for coral cover as well 
as fish biomass and richness (Bioregion 14, Chapter 4). The region's coral communities are similar to those 
in NE Tutuila, where the Masausi, Sailele, and Aoa CFMP reserves are located. Also of note, this is the only 
MPA that encompasses bank reef formations, making it a valuable and unique component of the MPA net- 
work. 



Additional References 

Orcutt 1993, Musburger 2004, Oram 2008 



Leone Pala Special Management Area 

Overview 

The Leone Pala Special Management Area (SMA) (Figure 5.26) is located in SWTutuila and was designated 
a special management area by the American Samoa Coastal Management Act of 1 990 because of its "unique 
and valuable characteristics" and the "imminent threat from development pressures" (ASCA§ 24.0503). It in- 
cludes both a -0.02 km 2 marine component, delineated by a straight line from the mouth of Leafu stream, and 
the adjacent wetland areas (ASAC § 26.0221). The primary reason for this and other designated SMAs is to 
regulate on-shore activities that could be harmful to unique marine ecosystems (Gombos et al. 2007). The 
Leone Pala SMA fronts a -14.7 km 2 watershed in extensively impacted condition with high human population 
density as well as a high density of pigs. Encroachment into the wetland area and nutrient, sediment, and silt 
discharges into the streams that flow into the lagoon have significantly impacted the ability of the wetland to 




Legend 

Benthic Structure Types 



■ 



Aggregate Reef 



Pavement 



Abated Patch Reefs || "e^ 



Individual Patch Reef 



Spur and Groove 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



filter sediment and nutrients. Management of Leone Pala SMA is primarily by the American Samoa Coastal 
Management Program (ASCMP) of the Department of Commerce, but no fishing regulations exist beyond 
territorial regulations and there is not a written management plan (Gombos et al. 2007). 



Habitat Composition, Reef Fish, and Coral Communities 
Coral reef and hardbottom structures together comprise -0.25% of the 
very small Leone lagoon that is the marine component of Leone Pala SMA. 
Instead, its benthic environment consists mainly of mud (Figure 5.27) with 
a mangrove shoreline that was too small to be included in island-scale 
mapping (NOAANCCOS 2005). Because coral reef and hardbottom struc- 
tures comprise only -0.25% of the lagoon, we do not include the zonation 
of coral reef and hardbottom in Figure 5.27. 

There were no fish and coral surveys from the island-wide comparison 
located within the Leone Pala SMA. 

Biogeographic Characteristics 

Leone Pala SMA lies adjacent to a biogeographic region that is a hotspot 
for coral cover, fish biomass, and fish richness (Bioregion 1, Chapter 4). 
However, this MPA lacks well developed reefs and is intended for protec- 
tion of wetland and nearshore habitats. 

Additional References 
Gilmanetal. 2007 



Total Area by 
Benthic Structure Type 



100% 




-0.02 km 2 

Figure 5.27. Proportion of benthic 
structure types in the Leone Pala 
SMA. Structure types or zones repre- 
senting <1% of the total area are not 
shown. 




Figure 5.26. Benthic habitat (by structure type) within the Leone Pala SMA. No reef fish or coral surveys from the island-wide com- 
parison were located within the Leone Pala SMA. 



Image 21. Mangroves at Leone Pala SMA. 
Photo credit: Matt Kendall, NOAA Biogeography. 




Masausi CFMP Reserve 

Overview 

The village of Masausi lies in NE Tutuila. The Masausi CFMP reserve (Figure 5.28) was established in 2002 to 
"conserve the marine resources in the ocean or in the village reef" (ASDMWR 2003d). The -0.2 km 2 reserve 
extends from Puputagi Point in the west to Folau Point in the east and offshore approximately 200 yards. 
It fronts a -1.6 km 2 watershed in minimally impacted condition with low population density concentrated in 
Masausi Village and a medium density of pigs. Because of the erosive soil types and steep slopes, there 
is moderate to high potential for periodic natural erosion with sediments carried into the nearshore waters 
fronting the watershed. Nearshore waters are also impacted to a lesser extent by urban runoffs. The reserve 
is closed to all commercial and recreational fishing apart from when it is opened for subsistence fishing at 
certain times of the year. Destructive fishing methods, including the use of bleach, poisons, and explosives, 



r 



J Masausi CFMP Reserve Boundary 




170 o 50'W 170 o 40'W 



Legend 

Benthic Structure Types 



■ 



Aggregate Reef 



Pavement 



| Aggregated Patch Reefs | J »£££% 



Individual Patch Reef 



Spur and Groove 



^ 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 
Site Values for each Variable; 
Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.28. Benthic habitat (by structure type) and fish and coral survey data within the Masausi CFMP reserve. Coral cover, coral 
richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shading), or 
low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from REA. 



are banned. The use of scuba gear for fishing, flashlights for night fishing, and the breaking up of corals for 
fishing are also banned, as is fishing by outsiders (ASDMWR 2003d). 



Habitat Composition, Reef Fish, 
and Coral Communities 
The Masausi CFMP reserve 
is dominated by coral reef and 
hardbottom structures, which 
together comprise -59% of its 
area (Figure 5.29a). Coral reef 
structures, mostly aggregate 
reef, comprise -18% of the 
area, while -40% of the area is 
covered by algal plain. In com- 
parison, aggregate reef covers 
-5% of American Samoa over- 
all. About 23% and -54% of the 
coral reef and hardbottom in the 
reserve are in the reef flat and 
fore reef zones, respectively, 
compared to -9% and -22% 
around American Samoa (Fig- 
ure 5.29b). Also of note, -10% 
of the coral reef and hardbot- 
tom is in the shoreline intertidal 
zone. 



(a) Total Area by 

Benthic Structure Type 



40% 




(b) Zonation of 

Coral Reef and Hardbottom 



Shoreline 

Intertidal 

(10%) 



Reef Flat 
%) 



28% 



Bank/Shelf 
(3%) 




Reef Crest 
(11%) 



Fore Reef 
(54%) 



~59%> of total area is coral reef and hardbottom 



-0.20 km 2 



-►-0.12 km 2 



Figure 5.29. (a) Proportion of benthic structure types in the Masausi CFMP reserve, 
(b) Proportion of coral reef and hardbottom in each reef zone. Structure types or zones 
representing <1% of the total area are not shown. 



Coral Cover Coral Richness Fish Biomass Fish Richness 



1(1) 



1(1) 



1(1) 



1(1) 



Masausi 

CFMP 

Reserve 




• 





Only two surveys were located within 
the Masausi CFMP reserve, and these 
were located in the area covered by algal 
plain rather than the reef and pavement 
areas. Coral data at these sites includes 
one medium value for cover and one high 
value for richness. Fish data includes 
one low value each for biomass and rich- 
ness (Figure 5.30). The small sample 
size greatly limits the scope of these find- 
ings and does not allow comparisons 
with American Samoa overall. Additional, 
more widely spread surveys are needed 
to more fully characterize the reef fish 
and coral communities within this MPA. 



Biogeographic Characteristics 

The Masausi CFMP reserve is located in a biogeographic region that is a hotspot for coral richness (Biore- 
gion 11, Chapter 4). The region's coral communities are similar to those in NW Tutuila, where the Fagamalo 
CFMP reserve and No-Take MPA are located. 



Figure 5.30. Fish and coral data collected in the Masausi CFMP reserve. Pie 
charts depict the proportions of high (red), medium (pink), and low (white) 
values for coral cover, coral richness, fish biomass, and fish richness. Number 
labels represent the number of studies and sites (in parentheses) comprising 
each pie chart. 



Additional References 
Orcutt 1993, Musburger 2004 




Matu'u and Faganeanea CFMP Reserve 

Overview 

The villages of Matu'u and Faganeanea are found on the south central coast of Tutuila. The Matu'u and Fa- 
ganeanea CFMP reserve (Figure 5.31 ) was established in 2005 with the primary goal of "protecting the coral 
reefs of Matu'u and Faganeanea to provide more fish for the future generation" (ASDMWR 2005). The -0.3 
km 2 reserve extends from the western tip of Utulaina Point to Matautuloa Point and offshore approximately 
1 00 yards (ASDMWR 2005). It fronts a -2.6 km 2 watershed in intermediately impacted condition, has moder- 
ate human population density and a low density of pigs. Because of the erosive soil types and steep slopes, 
there is moderate to high potential for periodic natural erosion with sediments carried into the nearshore 
waters fronting the watershed. Nearshore waters are also impacted to a lesser extent by insufficiently treated 



Matu'u & Faganeanea CFMP Reserve Boundary 





Legend 

Benthic Structure Types 



□ 



Aggregate Reef 



Pavement 



Rock/Boulder 



| Aggregated Patch Reefs | J ^« 



Individual Patch Reef 



Spur and Groove 



Pavement with 
Sand Channels 

Reef Rubble 



Emergent Vegetation 
Algal Plain 
I Mud 



□ 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 

Site Values for each Variable 

Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.31. Benthic habitat (by structure type) and fish and coral survey data within the Matu'u and Faganeanea CFMP reserve. 
Coral cover, coral richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium 
(pink shading), or low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are 
from ASEPA and REA. 



wastewater. The reserve is closed to all commercial and recreational fishing apart from when it is opened for 
subsistence fishing at certain times of the year. Loitering in the reserve and in village streams is also prohib- 
ited (ASDMWR 2005). 



Habitat Composition, Reef Fish, and 
Coral Communities 
The Matu'u and Faganeanea CFMP 
reserve is dominated by coral reef 
and hardbottom structures, which 
together comprise -94% of the area 
within the reserve (Figure 5.32a). 
Coral reef structures in the form of 
aggregate reef comprise -29% of 
the area, while reef rubble covers 
more than 50% of the area. In com- 
parison, aggregate reef and reef 
rubble cover -5% and -2%, respec- 
tively, of the mapped benthic habitat 
around American Samoa. About 2% 
and -70% of the coral reef and hard- 
bottom in the reserve are in the reef 
flat and fore reef zones, respectively, 
compared to -9% and -22% around 
American Samoa (Figure 5.32b). 
Also of note, -16% of the coral reef 
and hardbottom is in the reef crest 
zone. 



(a) Total Area by 

Benthic Structure Type 



(b) Zonation of 

Coral Reef and Hardbottom 



53% 






Reef Flat 




Bank/Shelf 


(2%) 


Reef Crest 


(11%)^ 






^16%) 



-0.32 km 2 



-94% of total area is coral reef and hardbottom 



Fore Reef 
(70%) 

-► -0.30 km 2 



Figure 5.32. (a) Proportion of benthic structure types in the Matu'u and Faganea- 
nea CFMP reserve, (b) Proportion of coral reef and hardbottom in each reef zone. 
Structure types or zones representing <1 % of the total area are not shown. 



Coral Cover 
2(2) 



Coral Richness Fish Biomass 

1(1) 2(2) 



Matu'u & 
Faganeanea 
CFMP 
Reserve 






Fish Richness 

2(2) 



Figure 5.33. Fish and coral data collected in the Matu'u and Faganeanea 
CFMP reserve. Pie charts depict the proportions of high (red), medium (pink), 
and low (white) values for coral cover, coral richness, fish biomass, and fish 
richness. Number labels represent the number of studies and sites (in paren- 
theses) comprising each pie chart. 



Only two surveys were located within the 
Matu'u and Faganeanea CFMP reserve, 
and these were both on the eastern end 
of the reserve. Coral data at these sites 
includes one low and one medium value 
for cover and one medium value for rich- 
ness. Fish data includes two medium 
values for biomass and one low and one 
high value for richness (Figure 5.33). The 
small sample size greatly limits the scope 
of these findings and does not allow com- 
parisons with American Samoa overall. 
Additional, more widely spread surveys 
are needed to more fully characterize the 
reef fish and coral communities within this 
MPA. 



Biogeographic Characteristics 

The Matu'u and Faganeanea CFMP reserve lies in a biogeographic region that is a hotspot for fish richness 

(Bioregion 4, Chapter 4). 

Additional References 

Randall and Devaney 1974, McConnaughey 1993, Orcutt 1993, Peshut et al. 2007 



Nu'uuli Pala Special Management Area 

Overview 

The Nu'uuli Pala SMA (Figure 5.34) is located in south-central Tutuila near the airport and, similar to Leone 
Pala SMA, was designated a special management area by the American Samoa Coastal Management Act of 
1990 because of its "unique and valuable characteristics" and the "imminent threat from development pres- 
sures" (ASCA § 24.0503). It includes both a -2.0 km 2 marine component, delineated by a straight line from 
Avatele Point to Mulinu'u Point, and the adjacent wetland areas (ASAC § 26.0221). The primary reason for 
this and other designated SMAs is to regulate on-shore activities in the wetland areas that could be harmful 
to unique marine ecosystems (Gombos et al. 2007). The Nu'uuli Pala SMA fronts a -17.6 km 2 watershed in 
extensively impacted condition with high population density and continued pressure from residential expan- 
sion. Increased turbidity and sedimentation within the lagoon result from the steep slopes and highly erosive 




Nu'uuli Pala Special Management Area Boundary 



0.25 

Legend 

Benthic Structure Types 

Aggregate Reef 



Pavement 



Aggregated Patch Reefs 1^%--— 



Individual Patch Reef 



Spur and Groove 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



soils in adjacent watersheds as well as from impervious surface runoffs in the urbanized areas. In addition, 
nutrient loading from insufficiently treated wastewater may impact nearshore waters. There is a medium den- 
sity of pigs in the watershed. Management of Nu'uuli Pala SMA is primarily by the American Samoa Coastal 
Management Program (ASCMP) of the Department of Commerce, but no fishing regulations exist beyond 
territorial regulations and there is no written management plan (Gombos et al. 2007). 



(a) Total Area by 

Benthic Structure Type 




(b) Zonation of 

Coral Reef and Hardbottom 



Reef Flat 
(78%) 



P 



73% 



* also some dredged and fore reef, but 
these areas are <1% of total area 



-2.0 km 2 



-5% of total area is coral reef and hardbottom 



-> -0.1 km 2 



Habitat Composition, Reef 
Fish, and Coral Communities 
Coral reef and hardbottom 
structures together comprise 
only -5% of the marine com- 
ponent of the Nu'uuli Pala 
SMA. In fact, no coral reef 
structures are found within 
this MPA. It is instead domi- 
nated by mud and mangrove 
habitats, which cover -73% 
and -13%, respectively, of its 
area (Figure 5.35a). In con- 
trast, these structure types 
together comprise only -1% 
of American Samoa overall. 
About 78% and -8% of the 
coral reef and hardbottom in 
the SMA are in the reef flat 
and fore reef zones, respec- 
tively, compared to -9% and 
-22% around American Sa- 
moa (Figure 5.35b). Also of 
note, -15% of the coral reef 
and hardbottom is in dredged 
areas. 

There were no fish and coral surveys from the island-wide comparison located within the Nu'uuli Pala SMA. 

Biogeographic Characteristics 

This SMA lies adjacent to a biogeographic region that is a hotspot for fish richness (Bioregion 4, Chapter 4). 
While Nu'uuli Pala is clearly a different and separate subregion, it has by far the largest area of mangrove 
habitat in American Samoa and may contribute to the adjacent region's fish richness by providing habitat for 
juvenile fish. 

Additional References 

Helfrich 1 975, Yamasaki et al. 1 985, Kluge 1 992, Ponwith 1 992, lose and McConnaughey 1 993, Orcutt 1 993, 

Peshutetal. 2007 



Figure 5.35. (a) Proportion of benthic structure types in the Nu'uuli Pala SMA. (b) Proportion 
of coral reef and hardbottom in each reef zone. Structure types or zones representing <1% of 
the total area are not shown. 



Figure 5.34. Benthic habitat (by structure type) and fish and coral survey data within the Nu'uuli Pala SMA. No reef fish or coral 
surveys from the island-wide comparison were located within the Nu'uuli Pala SMA. 



Ofu Vaoto Territorial Marine Park (also known as Ofu Vaoto Marine Reserve) 

Overview 

The Ofu Vaoto Territorial Marine Park was established in 1994 "to protect its unique coral reef wildlife habitat 
while enabling the public to enjoy the natural beauty of the site" (ASCA §18.021 4). It lies at the southwest tip 
of Ofu Island (Figure 5.36) and extends from the mean high water line seaward to approximately the ten fath- 
om depth contour from the western end of the Ofu Airport runway to Fatauana Point, where it abuts the Ofu 
unit of the National Park (ASCA §18.021 4). It fronts the southern tip of a -4.4 km 2 watershed but is minimally 
impacted by land-based human activity. The nearshore waters may be impacted by natural sediment runoffs 
because of the steep slopes and highly erosive soils. There is a medium density of pigs in the watershed 
compared to all of American Samoa, but waste discharge from piggeries is less likely to impact the nearshore 
waters of the Park since the portion of the watershed fronting the Park is largely uninhabited. While the De- 




Fatauana Point 
(14°11'2"S, 169 o 39'50"W) 



cc 


CR 


FB 


FR 



Ofu Vaoto Marine Park Boundary 



0.25 

Legend 

Benthic Structure Types 



0.5 



Aggregate Reef 



Pavement 



Aggregated Patch Reefs || "e^ 



Individual Patch Reef 



Spur and Groove 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 



1 

Fish/Coral Data 

O Survey Sites 

Site Values for each Variable 

Coral Cover 



Artificial 
Unknown 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Deep Water 



Figure 5.36. Benthic habitat (by structure type) and fish and coral survey data within the Ofu Vaoto Marine Park. Coral cover, coral 
richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shading), or 
low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from REA. 



partment of Parks and Recreation (DPR) has management authority for the Park, the Department of Marine 
and Wildlife Resources (DMWR) exercises primary authority over fishing regulations (Gombos et al. 2007). 
Fishing and shellfish harvesting are prohibited, with the exception of subsistence fishing and harvesting by 
Ofu Island residents according to territorial regulations (ASCA §18.0214). 



(a) Total Area by 

Benthic Structure Type 



11% 




(b) Zonation of 

Coral Reef and Hardbottom 

Shoreline 

Intertidal 

(2%) 



Bank/Shelf 
(10%) 



25% 



45% 




Reef Flat 
(42%) 



Fore Reef 
(46% 



-0.48 km 2 



-85% of total area is coral reef and hardbottom 



-►-0.41 km 2 



Figure 5.37. (a) Proportion of benthic structure types in the Ofu Vaoto Marine Park, (b) 
Proportion of coral reef and hardbottom in each reef zone. Structure types or zones 
representing <1% of the total area are not shown. 



Habitat Composition, Reef Fish, 
and Coral Communities 
The benthic habitat in the Ofu 
Vaoto Marine Park is dominated 
by coral reef and hardbottom 
structures, which together com- 
prise -85% of its area (Figure 
5.37a). Coral reef structures 
comprise -34% of the area and 
include spur and groove (-24%) 
and aggregate reef (-10%). In 
addition, pavement covers -45% 
of the area. In comparison, these 
three structure types comprise 
only -15% of the mapped ben- 
thic habitat around American 
Samoa. Of note, -11% of the 
mapped benthic habitat is of un- 
known structure type due to wave 
swash. About 42% and -46% of 
the coral reef and hardbottom 
are in the reef flat and fore reef 
zones, respectively, compared 
to only -9% and -22% around 
American Samoa (Figure 5.37b). 

Only two surveys were located within the 
Ofu Vaoto Marine Park, and neither of 
these surveys was in the reef and hard- 
bottom formations nearest to the shore- 
line. Coral data includes one low and one 
medium value for cover. No coral rich- 
ness data was collected with these sur- 
veys. Fish data includes one low and one 
medium value for biomass and one low 
and one high value for richness (Figure 
5.38). The small sample size greatly lim- 
its the scope of these findings and does 

not allow comparisons with American Samoa overall. Additional, more widely spread surveys are needed to 
adequately characterize the reef fish and coral communities within the Park. 

Biogeographic Characteristics 

The Ofu Vaoto Marine Park lies in a biogeographic region that includes all of Ofu and Olosega islands (Bio- 
region 18, Chapter 4). This area is a regional hotspot for coral richness, fish biomass, and fish richness. 

Additional References 
Maragos etal. 1995 



Coral Cover 
1(2) 



Coral Richness Fish Biomass 

(0) 1 (2) 



Ofu Vaoto 

Marine 

Park 





Fish Richness 

1(2) 

O 



Figure 5.38. Fish and coral data collected in the Ofu Vaoto Marine Park. Pie 
charts depict the proportions of high (red), medium (pink), and low (white) 
values for coral cover, coral richness, fish biomass, and fish richness. Number 
labels represent the number of studies and sites (in parentheses) comprising 
each pie chart. 



Pago Pago Harbor Special Management Area 

Overview 

The Pago Pago Harbor SMA (Figure 5.39) is located in central Tutuila and was designated a special man- 
agement area by the American Samoa Coastal Management Act of 1990 because of its "unique and valu- 
able characteristics" and the "imminent threat from development pressures" (ASCA § 24.0503). Its marine 
boundaries are defined by a straight line from Goat Island Point to the jetty at Leloaloa (ASCA § 26.0221) 
and include -1.2 km 2 of marine habitat. The primary reason for this and other designated SMAs is to regu- 
late on-shore activities in the wetland areas that could be harmful to unique marine ecosystems (Gombos 
et al. 2007). The Pago Pago Harbor SMA includes the inner harbor area and fronts the western portion of 
a -10.4 km 2 watershed in extensively impacted condition. In addition to natural sedimentation caused by 
highly erosive soils on steep slopes and increased surface runoffs due to extensive urbanization, nearshore 




0.25 

Legend 

Benthic Structure Types 

I Aggregate Reef 



Pavement 



Aggregated Patch Reefs | J ^« 



Individual Patch Reef 
Spur and Groove 



m 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Figure 5.39. Benthic habitat (by structure type) within the Pago Pago Harbor SMA. No reef fish or coral surveys from the island-wide 
comparison were located within the Pago Pago Harbor SMA. 



water quality has also been severely degraded by nutrient and chemical discharges by the tuna canneries 
and other historical industrial, commercial, and military activities adjacent to the harbor. There is a medium 
density of pigs in the watershed. Management of the SMA is primarily by the American Samoa Coastal Man- 
agement Program (ASCMP) of the Department of Commerce, but no fishing regulations exist beyond territo- 
rial regulations. There is no written management plan (Gombos et al. 2007). Sale offish or shellfish from the 
inner Harbor is prohibited due to contamination by heavy metals and other pollutants (ASEPA 1991). 



Habitat Composition, Reef Fish, 
and Coral Communities 
Coral reef and hardbottom struc- 
tures together comprise -44% of 
the area within the Pago Pago 
Harbor SMA (Figure 5.40a). 
Coral reef structures comprise 
-33% of the area and include 
aggregate reef (-18%) and ag- 
gregated patch reefs (-15%). In 
comparison, these two structure 
types comprise only -12% of the 
mapped benthic habitat around 
American Samoa. In addition, a 
large portion (-37%) of the ben- 
thic habitat in this SMA is cov- 
ered by mud. Also of note, -15% 
of the mapped benthic habitat in 
Pago Pago Harbor SMA is of un- 
known structure type due to high 
turbidity. About 23% and -40% 
of the coral reef and hardbot- 
tom in the SMA are in the reef 
flat and fore reef zones, respec- 
tively, compared to only -9% and 
-22% around American Samoa 
(Figure 5.40b). 



(a) Total Area by 

Benthic Structure Type 



15% 



18% 




(b) Zonation of 

Coral Reef and Hardbottom 



Shoreline 
Intertidal 
(3%) Reef Flat 
(23%) 



15% 




Fore Reef 
(40%) 



Bank/Shelf 
(34%) 



-1.2 km 2 



-44% of total area is coral reef and hardbottom 



-> -0.5 km 2 



Figure 5.40. (a) Proportion of benthic structure types in the Pago Pago Harbor SMA. 
(b) Proportion of coral reef and hardbottom in each reef zone. Structure types or zones 
representing <1% of the total area are not shown. 



There were no fish and coral surveys from the island-wide comparison located within the Pago Pago Harbor 
SMA. 

Biogeographic Characteristics 

Pago Pago Harbor SMA lies at the margin of a biogeographic region that is a hotspot for fish biomass and 
has a unique coral community (Bioregion 5, Chapter 4). However, note that high fish biomass may be due to 
the ban on sale offish from the harbor due to contaminant concerns and while the coral community may be 
"unique" relative to elsewhere in American Samoa it is not necessarily "healthy". 



Additional References 

McConnaughey 1993, Orcutt 1993, Green et al. 1997a, Fisk and Birkeland 2002, Coles et al. 2003, Craig et 

al. 2005, Green et al. 2005, Peshut et al. 2007 



Poloa CFMP Reserve 

Overview 

The village of Poloa is located on the NW tip of Tutuila. The Poloa CFMP reserve (Figure 5.41) was estab- 
lished in 2001 to "conserve, protect, and manage the resources in the village reef" (ASDMWR 2001). The 
-0.4 km 2 reserve extends from Legaotaema Point in the west to the boundary of Poloa with Fagali'i village 
and offshore to 100 yards beyond the seaward edge of the reef flat (ASDMWR 2001). It fronts a -1.1 km 2 
watershed in minimally impacted condition with low population density, while the northeastern tip of the re- 
serve fronts a -2.1 km 2 watershed of similar condition. While human impacts are minimal in the watershed, 
there is high potential for runoff and erosion because of the erosive soil types and steep slopes. There is 
a low density of pigs in the adjacent watersheds. The reserve is closed to all commercial and recreational 
fishing apart from when it is opened for subsistence fishing at certain times of the year. Destructive fishing 



J Poloa CFMP Reserve Boundary 




ITO^O'W 170 o 40'W 170°30'W 



0.25 

Legend 

Benthic Structure Types 



■ 



Aggregate Reef 



Pavement 



| Aggregated Patch Reefs | J »£££% 



Individual Patch Reef 



Spur and Groove 



^ 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 
Site Values for each Variable; 
Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.41. Benthic habitat (by structure type) and fish and coral survey data within the Poloa CFMP reserve. Coral cover, coral 
richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shading), or 
low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from KRS. 



methods including the use of bleach, poisons, and explosives are banned. The use of scuba gear for fishing, 
flashlights or lanterns for night fishing, and the breaking up of corals for fishing are also banned, as is fishing 
by outsiders (ASDMWR 2001). 



29% 



Habitat Composition, Reef Fish, 
and Coral Communities 
The Poloa CFMP reserve is domi- 
nated by coral reef and hardbot- 
tom structures, which together 
comprise -98% of the area within 
the reserve (Figure 5.42a). Coral 
reef structures comprise -25% 
of the area and include spur and 
groove (-18%) and aggregate 
reef (~7%). In addition, pave- 
ment covers -38% of the area. 
In comparison, these three struc- 
ture types comprise only -15% of 
American Samoa overall. About 
33% and -21% of the coral reef 
and hardbottom in the reserve 
are in the reef flat and fore reef 
zones, respectively, compared to 
-9% and -22% around American 
Samoa (Figure 5.42b). Also of 
note, over 30% of the coral reef 
and hardbottom is in the bank/ 
shelf, compared to -50% around 
American Samoa. 



Only one survey was located within the 
Poloa CFMP reserve. Coral data at this 
includes one medium value for cover. No 
coral richness data was collected within 
the reserve. Fish data at the site includes 
one low value for each of biomass and 
richness (Figure 5.43). The small sample 
size greatly limits the scope of these find- 
ings and does not allow comparisons 
with American Samoa overall. Additional, 
more widely spread surveys are needed 
to characterize the reef fish and coral 
communities within this MPA. 



(a) Total Area by 

Benthic Structure Type 

1% 




(b) Zonation of 

Coral Reef and Hardbottom 



Channel Shoreline 
(2%) Intertidal 
" (3%) 



Bank/Shelf 
(34%) 




Reef Flat 
(33%) 



Fore Reef 
(21%) 



Reef Crest 
(7%) 



-0.36 km 2 



-98% of total area is coral reef and hardbottom 



-> -0.35 km 2 



Figure 5.42. (a) Proportion of benthic structure types in the Poloa CFMP reserve, (b) 
Proportion of coral reef and hardbottom in each reef zone. Structure types or zones 
representing <1% of the total area are not shown. 



Coral Cover 

1(1) 



Coral Richness Fish Biomass 

0(0) 1(1) 



Fish Richness 

1(1) 






Poloa 
CFMP 
Reserve 



Figure 5.43. Fish and coral data collected in the Poloa CFMP reserve. Pie 
charts depict the proportions of high (red), medium (pink), and low (white) 
values for coral cover, coral richness, fish biomass, and fish richness. Number 
labels represent the number of studies and sites (in parentheses) comprising 
each pie chart. 



Biogeographic Characteristics 

The Poloa CFMP reserve is located in a biogeographic region that is not a hotspot for any of the coral and 

reef fish variables analyzed (Bioregion 15, Chapter 4). 

Additional References 

Orcutt 1993, Musburger 2004, Peshut et al. 2007 




Sailele CFMP Reserve 

Overview 

The village of Sailele is located in NE Tutuila. The Sailele CFMP reserve (Figure 5.44) was established in 
2005 to protect the reef area so it can allow sustainable use of the marine resources in the reserve. The -0.1 
km 2 reserve extends from Malo Point to Leanmanu Point and offshore to approximately 75 yards from the 
reef crest. It fronts a -0.7 km 2 watershed in minimally impacted condition, with low human population density 
concentrated in Sailele Village. There are no pigs reported in the watershed, so any impacts by waste dis- 
charge from piggeries most likely come from elsewhere. Because of the erosive soil types and steep slopes 
there is moderate to high potential for periodic natural erosion with sediments carried into the nearshore 
waters fronting the watershed. Nearshore waters are also impacted to a lesser extent by urban runoffs. The 
reserve is closed to all commercial and recreational fishing apart from when it is opened for subsistence fish- 
ing at certain times of the year. 



-— — 

J Sailele CFMP Reserve Boundary 




Legend 

Benthic Structure Types 

I Aggregate Reef 

Aggregated Patch Reefs 112 Pavement with 
yy y KK Patch Reefs 



Pavement 



Individual Patch Reef 
Spur and Groove 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Habitat Composition, Reef Fish, 
and Coral Communities 
The Sailele CFMP reserve is 
dominated by coral reef and 
hardbottom structures, which 
together comprise the entire 
area within the reserve (Figure 
5.45a). Coral reef structures in 
the form of aggregate reef com- 
prise -24% of the area, while 
pavement covers -45% of the 
area. In comparison, these two 
structure types cover less than 
15% of American Samoa over- 
all. About 48% and -35% of the 
coral reef and hardbottom in the 
reserve are in the reef flat and 
fore reef zones, respectively, 
compared to only -9% and 
-22% around American Samoa 
(Figure 5.45b). Also of note, 
-15% of the coral reef and hard- 
bottom is in the reef crest. 



(a) Total Area by 

Benthic Structure Type 



(b) Zonation of 

Coral Reef and Hardbottom 



Bank/Shelf 
(2%) 




Fore Reef 
(35%) 




Reef Flat 
(48%) 



45% 



Reef Crest 
(15%) 



100% of total area is coral reef and hardbottom 
-0.08 km 2 ► -0.08 km 2 

Figure 5.45. (a) Proportion of benthic structure types in the Sailele CFMP reserve, (b) 
Proportion of coral reef and hardbottom in each reef zone. Structure types or zones rep- 
resenting <1% of the total area are not shown. 



There were no fish and coral surveys from the island-wide comparison located within the Sailele CFMP re- 
serve. 

Biogeographic Characteristics 

The Sailele CFMP reserve is located in a biogeographic region that is a hotspot for coral richness (Bioregion 
1 1 , Chapter 4). The region's coral communities are similar to those in NW Tutuila, where the Fagamalo CFMP 
reserve and No-Take MPAare located. 



Figure 5.44. Benthic habitat (by structure type) within the Sailele CFMP reserve. No reef fish or coral surveys from the island-wide 
comparison were located within the Sailele CFMP reserve. 




Vatia CFMP Reserve 

Overview 

The village of Vatia is located on the north-central coast of Tutuila and partially overlaps the Tutuila unit of the 
National Park. The Vatia CFMP reserve (Figure 5.46) was established in 2001 to "manage, protect, and pre- 
serve the fish, shellfish, and the coastal area of the village of Vatia" (ASDMWR 2002b). The -0.6 km 2 reserve 
extends from Falelofia Point at the northern end of Polatai Islet to Craggy Point at the boundary with Afono 
village and offshore approximately 100 yards (ASDMWR 2002b). It primarily fronts a -4.9 km 2 watershed in 
minimally impacted condition with the population concentrated around Vatia Bay. Because of the soil types 
and steep slopes, there is moderate to high potential for periodic natural erosion and nearshore sediment 
impacts. Nearshore waters are also somewhat impacted by wastewater discharge concentrated near the vil- 




cc 


CR 


FB 


FR 



Craggy Poin 




Kilometers 



0.5 

Legend 

Benthic Structure Types 

Aggregate Reef 



Pavement 



Aggregated Patch Reefs g§ "e^ 



Individual Patch Reef 



Spur and Groove 



Pavement with 
Sand Channels 

Reef Rubble 



^1 Rock/Boulder 
§ Emergent Vegetation 

Algal Plain 
I Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 

Site Values for each Variable: 

Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.46. Benthic habitat (by structure type) and fish and coral survey data within the Vatia CFMP reserve. Coral cover, coral rich- 
ness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shading), or low 
(white shading). Grey shading indicates variables with no data at a given site. 
Fish and coral survey data are from KRS, MPABR, and REA. 



lage. There is a low density of pigs in adjacent watersheds compared to all of American Samoa. The reserve 
is closed to all commercial and recreational fishing apart from when it is opened for subsistence fishing at 
certain times of the year. Destructive fishing methods, including the use of bleach, poisons, and explosives, 
are banned. The use of scuba gear for fishing, flashlights for night fishing, and the breaking up of corals for 
fishing are also banned, as is fishing by outsiders (ASDMWR 2002b). 



14% 



Habitat Composition, Reef Fish, 
and Coral Communities 
The Vatia CFMP reserve is 
dominated by coral reef and 
hardbottom structures which 
together comprise -96% of the 
area within the reserve (Figure 
5.47a). Coral reef structures in 
the form of aggregate reef com- 
prise -30% of the area. In addi- 
tion, pavement covers -37% of 
the area. In comparison, these 
two structure types cover -12% 
of American Samoa overall. 
About 22% and -30% of the 
coral reef and hardbottom in 
the reserve are in the reef flat 
and fore reef zones, respective- 
ly, compared to only -9% and 
-22% around American Samoa 
(Figure 5.47b). Also of note, al- 
most 40% of the coral reef and 
hardbottom is in the bank/shelf, 
compared to -50% around 
American Samoa. 



Only three surveys were located within 
the Vatia CFMP reserve, and these were 
at the extreme western and eastern ends 
of the reserve. Coral data at these sites 
includes one low, one medium, and one 
high value for cover and one low value 
for richness. Fish data includes one low 
and one medium value for biomass and 
one low and two medium values for rich- 
ness (Figure 5.48). The small sample 
size greatly limits the scope of these find- 
ings and does not allow comparisons with 
American Samoa overall Additional, more 
reef fish and coral communities within this 



(a) Total Area by 

Benthic Structure Type 



1% 3% 



(b) Zonation of 

Coral Reef and Hardbottom 




30% 



Reef Flat 



Bank/Shelf 
(39%) 




Reef Crest 
(7%) 



37% 



Fore Reef 
(30%) 



-0.62 km 2 



-96% of total area is coral reef and hardbottom 



-> -0.60 km 2 



Figure 5.47. (a) Proportion of benthic structure types in the Vatia CFMP reserve, (b) 
Proportion of coral reef and hardbottom in each reef zone. Structure types or zones rep- 
resenting <1% of the total area are not shown. 



Coral Cover Coral Richness Fish Biomass Fish Richness 



3(3) 



1(1) 



2(2) 



3(3) 



Vatia 

CFMP 

Reserve 



a 






Figure 5.48. Fish and coral data collected in the Vatia CFMP reserve. Pie 
charts depict the proportions of high (red), medium (pink), and low (white) 
values for coral cover, coral richness, fish biomass, and fish richness. Number 
labels represent the number of studies and sites (in parentheses) comprising 
each pie chart. 

widely spread surveys are needed to more fully characterize the 
MPA. 



Biogeographic Characteristics 

The Vatia CFMP reserve lies in a biogeographic region that is a hotspot for coral cover and also includes 
the Tutuila unit of the National Park (Bioregion 12, Chapter 4). The region's fish and coral communities are 
similar to those around Fagaitua Bay on the SE side of Tutuila. 

Additional References 

Randall and Devaney 1974, Orcutt 1993, Coles et al. 2003, Andrews 2004, Musburger 2004, Houk 2010 




Federal or Federal/Territorial Co-Managed MPAs 
Fagatele Bay National Marine Sanctuary 



Overview 

Located on the SW side of Tutuila, Fagatele Bay National Marine Sanctuary (FBNMS) (Figure 5.49) encompasses 
-0.7 km 2 of fringing coral reef ecosystem within a collapsed volcanic crater and extends from the southern tip of 
Fagatele Point to the southern tip of Step's Point. The Sanctuary was designated in 1986 to "protect and preserve 
an example of a pristine tropical marine habitat and coral reef terrace ecosystem" (1 5 C.F.R. 922.1 00-1 04, FBNMS 
Regulations 1986). FBNMS fronts a -3.2 km 2 watershed in pristine condition. Human impacts within the watershed 
are minimal and although nearshore waters may be affected by runoff of highly erosive soils on the steep slopes of 
the watershed, these runoffs are not associated with significant nutrient loads. There is a low density of pigs in the 




Larsen 



0.25 

Legend 

Benthic Structure Types 

Aggregate Reef 



Pavement 



| Aggregated Patch Reefs £ j P ^™^ 



Individual Patch Reef 



Spur and Groove 



^ 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



□ 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 

Site Values for each Variable; 

Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.49. Benthic habitat (by structure type) and fish and coral survey data within Fagatele Bay NMS. Coral cover, coral richness, 
fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shading), or low (white 
shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from ASEPA, CRSR, MPABR, 
REA, and TCRMP. 



watershed compared to all of American Samoa, and no humans (or pigs) inhabit the portion of the watershed adja- 
cent to Fagatele Bay. However, the only landfill on Tutuila is located north of the ridge above FBNMS, and research 
is recommended to assess the potential for groundwater seepage into the Bay. Fishing is presently regulated differ- 
ently within two zones in FBNMS but may soon be modified to complete no-take with a management plan revision 
that is currently underway. Presently, fishing with designated gear types is allowed in the outer bay (seaward of a 
line between Matautuloa Benchmark and Fagatele Point) whereas hook and line and commercial fishing in the inner 
bay are prohibited. 



Habitat Composition, Reef Fish, 
and Coral Communities 
FBNMS is dominated by coral 
reef and hardbottom structures, 
which together comprise -91% 
of its area (Figure 5.50a). Coral 
reef structures comprise -64% of 
the area and include aggregate 
reef (-36%) and spur and groove 
(-28%). In comparison, these 
two structure types comprise 
only -11% of the mapped benthic 
habitat around American Samoa. 
About 7% and -36% of the coral 
reef and hardbottom in FBNMS 
are in the reef flat and fore reef 
zones, respectively, compared to 
-9% and -22% around American 
Samoa (Figure 5.50b). In addition, 



(a) Total Area by 

Benthic Structure Type 



23% 




(b) Zonation of 

Coral Reef and Hardbottom 



Reef Flat 

(7%) Reef Crest 

(4%) 



36% 



Bank/Shelf 

Escarpment 

(31%) 




Fore Reef 
(36%) 



28% 



-0.70 km 2 



Bank/Shelf 
(22%) 



-91% of total area is coral reef and hardbottom 



-> -0.64 km 2 



Figure 5.50. (a) Proportion of benthic structure types in Fagatele Bay NMS. (b) Propor- 
tion of coral reef and hardbottom in each reef zone. Structure types or zones representing 
<1 % of the total area are not shown. 



-22% and -31% of the coral reef and hardbottom are in the bank/shelf and bank/shelf escarpment, respectively. 
In comparison, -50% of the coral reef and hardbottom around American Samoa is in the bank shelf, whereas only 
-10% is in the bank/shelf escarpment. Of note, the relative proportions and inshore/offshore zonation of these reef 
and hardbottom features are replicated by those found in adjacent Larsen Bay. 



Eight surveys were located within Fagatele 
Bay and those were concentrated in the 
northeast aggregate reef and pavement 
areas. Coral data at those sites suggests 
relatively higher cover and richness values 
compared to all of American Samoa. Fish 
richness values are much higher than those 
elsewhere in American Samoa whereas 
biomass was comparable (Figure 5.51). 
Additional, more widely spread surveys are 
needed to more fully characterize the outer 
portions of the Bay. 



Coral Cover Coral Richness Fish Biomass Fish Richness 



3(7) 



3(6) 



4(7) 



4(7) 



Fagatele 

Bay 

NMS 



American 

Samoa 

overall 



a a 



6 (339) 



5(137) 



(3(3 




3 

6 (347) 

3 



Figure 5.51. Comparison offish and coral data collected in Fagatele Bay NMS 
to data from all of American Samoa. Pie charts depict the proportions of high 
(red), medium (pink), and low (white) values for coral cover, coral richness, fish 
biomass, and fish richness. Number labels represent the number of studies 
and sites (in parentheses) comprising each pie chart. 



Biogeographic Characteristics 

FBNMS lies in a biogeographic region that 

includes Larsen Bay and is a hotspot for coral cover as well as coral and fish richness (Bioregion 2, Chapter 4). The 

region's fish and coral communities are representative of southwestern Tutuila and have some similarities to coral 

communities around Aunu'u. 



Additional References 

Birkeland et al. 1987, Orcutt 1993, Birkeland etal. 1994, Green et al. 1999, Fiskand Birkeland 2002, Birkeland etal. 

2003, Coles et al. 2003, Andrews 2004, Birkeland et al. 2004, Green et al. 2005 



National Park of American Samoa - Ofu Unit 

Overview 

The Ofu unit of the National Park (Figure 5.52) was authorized by Public Law 100-571 in 1988 and formally 
established in 1993 following a lease agreement with the villages (16 U.S.C. 410qq-410qq-1, NPS 1997). 
Its boundary follows the southeast shoreline road of Ofu Island from Fatauana Point to Asega Strait and ex- 
tends 0.25 miles offshore. It also extends inland to include the southern slopes of Sunu'itao Peak (16 U.S.C. 
410qq-410qq-1, NPS 1997). The 1986-7 NPS feasibility study noted the "exceptionally diverse reef fish and 
coral communities" in this area as well as the lack of reef damage from crown-of-thorns starfish outbreaks 
(NPS 1988). The -1.5 km 2 marine portion of the Ofu park unit fronts the largely uninhabited southeast side 
of a -4.4 km 2 watershed. The nearshore waters are not significantly impacted by human activity but may 
be impacted by natural sediment runoffs because of the steep slopes and highly erosive soils. There is a 




Fatauana Point (UNIT'S, 169 o 39'50"W) 



National Park of American Samoa - Ofu Unit Boundary 



0.5 

Legend 

Benthic Structure Types 



■ 



Aggregate Reef 



Pavement 



| Aggregated Patch Reefs | J »£££% 



Individual Patch Reef 



Spur and Groove 



^ 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 



Fish/Coral Data 

O Survey Sites 
Site Values for each Variable; 
Coral Cover 



Artificial 



Unknown 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Deep Water 



Figure 5.52. Benthic habitat (by structure type) and fish and coral survey data within the Ofu Unit of the National Park. Coral cover, 
coral richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shad- 
ing), or low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from REA. 



medium density of pigs in the watershed overall, however the portion of the watershed actually fronting the 
Park is largely uninhabited. Fishing or gathering is prohibited in the park, except subsistence fishing by native 
American Samoans using traditional tools and methods in accordance with rules established by the National 
Park Service and village leaders (16 U.S.C. 410qq-410qq-1, NPS 1997). 



(a) Total Area by 

Benthic Structure Type 



(b) Zonation of 

Coral Reef and Hardbottom 



Reef Flat 
(48%) 



24% 



Habitat Composition, Reef Fish, 

and Coral Communities 

The benthic habitat in this park 

unit is dominated by coral reef 

and hardbottom structures, which 

together comprise -71% of the 

area (Figure 5.53a). Coral reef 

structures comprise -32% of the 

area and include mostly spur and 

groove (-24%) and aggregated 

patch reefs (-6%). In comparison, 

these two structure types comprise 

only -13% of the mapped benthic 

habitat around American Samoa. 

Of note, - 8% of the offshore area 

in the Ofu park unit was too deep 

for satellite based mapping. Also 

of note, -9% of the benthic habitat 

within the park unit is of unknown 

bottom type due to wave swash. 

About 48% and -33% of the coral 

reef and hardbottom in this park 

unit are in the reef flat and fore 

reef zones, respectively, compared to only -9% and -22% around American Samoa (Figure 5.53b). Also of 

note, almost 20% of the coral reef and hardbottom is in the bank/shelf, compared to -50% around American 

Samoa. 




29% 



-1.5 km 2 



Bank/Shelf 
(18%) 



-71% of total area is coral reef and hardbottom 




Fore Reef 
(33%) 



-1.1 km 2 



Figure 5.53. (a) Proportion of benthic structure types in the Ofu Unit of the National 
Park, (b) Proportion of coral reef and hardbottom in each reef zone. Structure types or 
zones representing <1% of the total area are not shown. 



Only 5 surveys were located within this 
park unit and none included each of the 
key variables. Coral richness values at 
these sites are relatively higher compared 
to all of American Samoa, while coral 
cover and fish biomass and richness are 
relatively lower (Figure 5.54) although the 
small sample size greatly limits the scope 
of these findings. 

Biogeographic Characteristics 
The Ofu unit of the National Park is locat- 
ed in a biogeographic region that includes 
all of Ofu and Olosega and is a hotspot 
for coral richness, fish biomass, and fish 
richness (Bioregion 18, Chapter 4). 



Coral Cover Coral Richness Fish Biomass Fish Richness 

1(3) 1(2) 1(3) 1(3) 



NPSA- 
Ofu Unit 



American 

Samoa 

overall 




6(339) 5(137) 

0£> 





Figure 5.54. Comparison of fish and coral data collected in the Ofu Unit of 
the National Park to data from all of American Samoa. Pie charts depict the 
proportions of high (red), medium (pink), and low (white) values for coral cover, 
coral richness, fish biomass, and fish richness. Number labels represent the 
number of studies and sites (in parentheses) comprising each pie chart. 



Additional References 

Itano and Buckley 1988, Friedlander 1993, Hunter et al. 1993, Maragos et al. 1995, Craig and Basch 2001, 

Craig et al. 2001, Smith and Birkeland 2003, Andrews 2004, Pendleton et al. 2005, Garrison et al. 2007 




National Park of American Samoa - Ta'u Unit 

Overview 

The Ta'u unit of the National Park (Figure 5.55) was authorized by Public Law 100-571 in 1988 and formally estab- 
lished in 1 993 following a lease agreement with the villages (1 6 U.S.C. 41 Oqq-41 0qq-1 , NPS 1997). It occupies -20 
km 2 of land and ~ 4.7 km 2 of marine habitats in the south and southeast of Ta'u Island. The 1986-7 NPS feasibility 
study noted that this area includes "the largest extent of both undisturbed lowland and montane rainforest and cloud 
forest left in American Samoa" and would "provide important habitat for seabirds, shorebirds, flying foxes, and forest 
birds" (NPS 1 988). Also noted is the presence of the prehistoric village of Saua, along the east coast, and Taisama- 
sama, or the Yellow Waters of Tui Manu'a cultural site, located centrally on the south shore. The marine component 
of the park has a seaward boundary that extends 0.25 miles offshore from Si'ufa'alele Point eastward to the Saua 
site (16 U.S.C. 41 Oqq-41 0qq-1, NPS 1997). Most of the marine component of the park fronts an uninhabited -8.5 




cc 


CR 


FB 


FR 



CC 


CR 


FB 


FR 



J National Park of American Samoa - Ta'u Unit Boundary 



o 1 

Legend 

Benthic Structure Types 



■ 



Aggregate Reef 



Pavement 



| Aggregated Patch Reefs | J »£££% 



Individual Patch Reef 



Spur and Groove 



^ 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 



Fish/Coral Data 

O Survey Sites 
Site Values for each Variable; 
Coral Cover 



Artificial 



Unknown 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Deep Water 



Figure 5.55. Benthic habitat (by structure type) and fish and coral survey data within the Ta'u Unit of the National Park. Coral cover, 
coral richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shad- 
ing), or low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from REA. 



km 2 watershed in pristine condition that covers the southern portion of the island. The eastern marine component 
fronts an uninhabited portion of a -37 km 2 watershed that occupies most of the island. Although there are no real 
human impacts to the nearshore waters, there is a moderate to severe potential for sediment runoffs because of the 
steep slopes and highly erosive soils. There is a low density of pigs in the watershed and the portion actually front- 
ing the park unit is uninhabited, so waste discharge from piggeries is less likely to impact nearshore waters. Fishing 
or gathering is prohibited in the park, except subsistence fishing by native American Samoans using traditional tools 
and methods in accordance with rules established by NPS and village leaders (16 U.S.C. 41 Oqq-41 0qq-1, NPS 
1997). 



(a) Total Area by 

Benthic Structure Type 



(b) Zonation of 

Coral Reef and Hardbottom 



17% 



Reef Flat 
(12%) 



Fore Reef 
(34%) 



50% 



20% 



Habitat Composition, Reef Fish, 
and Coral Communities 
The nearshore areas of the 
Ta'u unit of the National Park 
are dominated by coral reef and 
hardbottom structures, which 
together comprise -43% of the 
area within the park unit (Figure 
5.56a). Coral reef structures in 
the form of spur and groove com- 
prise -17% of the area. In com- 
parison, spur and groove covers 
only -6% of the mapped benthic 
habitat around American Samoa. 
Because of the steep drop off at 
the shelf edge, about half of the 
offshore area in the Ta'u unit of 
the National Park was too deep 
for satellite based mapping. In 
addition, -7% of the area is of un- 
known bottom type due to wave 
swash. About 12% and -34% of 
the coral reef and hardbottom in 

this park unit are in the reef flat and fore reef zones, respectively, while over half of the coral reef and hardbottom 
is in the bank/shelf (Figure 5.56b). This is similar to the proportions of reef zones around American Samoa overall. 





Bank/Shelf 
(54%) 



-4.7 km 2 



-43% of total area is coral reef and hardbottom 



-> -2.1 km 2 



Figure 5.56. (a) Proportion of benthic structure types in the Ta'u Unit of the National Park, 
(b) Proportion of coral reef and hardbottom in each reef zone. Structure types or zones 
representing <1% of the total area are not shown. 



Twelve surveys were located within the park 
unit. Coral data suggests comparable cover 
and relatively higher richness values com- 
pared to all of American Samoa, although 
only two surveys included coral richness 
data. Fish biomass and richness values are 
comparable or slightly lower relative to all of 
American Samoa (Figure 5.57). 

Biogeographic Characteristics 
The Ta'u unit of the National Park straddles 
two biogeographic regions that are hotspots 
for coral cover and coral and fish richness, 
and that share a unique coral community 
representative of Ta'u Island (Bioregions 19- 
20, Chapter 4). 



NPSA- 
Ta'u Unit 



American 

Samoa 

overall 



Coral Cover Coral Richness Fish Biomass Fish Richness 

1 (10) 1 (2) 1 (10) 1 (10) 



£• 



6 (339) 



5(137) 



(3<3 




6 (347) 



Figure 5.57. Comparison of fish and coral data collected in the Ta'u Unit of 
the National Park to data from all of American Samoa. Pie charts depict the 
proportions of high (red), medium (pink), and low (white) values for coral cover, 
coral richness, fish biomass, and fish richness. Number labels represent the 
number of studies and sites (in parentheses) comprising each pie chart. 



Additional References 

Green and Hughes 1999, Craig and Basch 2001 , Pendleton et al. 2005 



National Park of American Samoa - Tutuila unit 

Overview 

The Tutuila unit of the National Park (Figure 5.58) was authorized by Public Law 1 00-571 in 1 988 and formal- 
ly established in 1 993 following a lease agreement with the villages (16 U.S.C. 41 Oqq-41 0qq-1 , NPS 1 997). It 
lies between the villages of Fagasa and Afono on the north-central coast of Tutuila with a seaward boundary 
that extends 0.25 miles offshore (16 U.S.C. 41 Oqq-41 0qq-1 , NPS 1 997). The park partially overlaps the Vatia 
CFMP reserve. The 1986-7 NPS feasibility study noted that this area includes "the longest stretch of unde- 
veloped coastline and undisturbed forest on Tutuila" (NPS 1988). The -6.5 km 2 marine portion of the Tutuila 
park unit primarily fronts two watersheds. The western of these is a -5.1 km 2 watershed that is uninhabited 
and in relatively pristine condition. Nearshore waters may however, be impacted by sediment runoff resulting 




National Park of American Samoa - Tutuila Unit Boundary 



o 1 

Legend 

Benthic Structure Types 

Aggregate Reef 



Pavement 



| Aggregated Patch Reefs £ j ^^ 



Individual Patch Reef 



Spur and Groove 



^ 



Pavement with 
Sand Channels 

Reef Rubble 



Rock/Boulder 
Emergent Vegetation 
Algal Plain 
Mud 



□ 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 

Site Values for each Variable; 

Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Figure 5.58. Benthic habitat (by structure type) and fish and coral survey data within the Tutuila Unit of the National Park. Coral 
cover, coral richness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink 
shading), or low (white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from 
ASEPA, KRS, MPABR, and REA. 



from the erosive soil types on steep inland slopes. To the east is a 4.9 km 2 watershed in minimally impacted 
condition with the population concentrated around Vatia Bay. In this area there is moderate to high erosion 
and runoff potential and slight impacts from groundwater and surface water contamination. The southwest 
and eastern boundaries of the park extend slightly into adjacent watersheds, with the watershed to the south- 
west being in intermediately impacted condition. There is a low density of pigs in these adjacent watersheds 
and most of the area is uninhabited. Fishing or gathering is prohibited in the park, except subsistence fishing 
by native American Samoans using traditional methods in accordance with rules established by NPS and 
village leaders (16 U.S.C. 410qq-410qq-1, NPS 1997). 



(a) Total Area by 

Benthic Structure Type 




(b) Zonation of 

Coral Reef and Hardbottom 



Reef Flat 
(4%) 



Habitat Composition, Reef Fish, 
and Coral Communities 
Coral reef and hardbottom struc- 
tures together comprise -43% of 
the offshore areas of the Tutuila 
unit of the National Park (Fig- 
ure 5.59a). Coral reef structures 
comprise -28% of the area and 
are dominated by aggregate 
reefs, a structure type often with 
a high percentage of reef build- 
ing corals. Also, algal plain cov- 
ers -43% of the area within this 
park unit. The relative propor- 
tions of benthic structure types 
within the park unit are repre- 
sentative of American Samoa in 
general. About 4% and -59% of 
the coral reef and hardbottom in 
this park unit are in the reef flat 
and fore reef zones, respective- 
ly, compared to -9% and -22% 
around American Samoa (Figure 
5.59b). Also of note, 36% of the 
coral reef and hardbottom is in the bank/shelf compared to -50% around American Samoa. 




Fore Reef 
(59%) 



Bank/Shelf 
(36%) 



-6.5 km 2 



-43% of total area is coral reef and hardbottom 



-> -2.8 km 2 



Figure 5.59. (a) Proportion of benthic structure types in the Tutuila Unit of the National 
Park, (b) Proportion of coral reef and hardbottom in each reef zone. Structure types or 
zones representing <1% of the total area are not shown. 



Fifteen surveys were located within the 
park. Coral data suggests relatively high- 
er cover and similar richness values com- 
pared to all of American Samoa. Fish bio- 
mass and richness values were relatively 
lower compared to all of American Samoa 
(Figure 5.60). 

Biogeographic Characteristics 
Most of the Tutuila unit of the National 
Park overlaps with a biogeographic re- 
gion along the north shore of Tutuila that 
is a hotspot for coral cover (Bioregion 12, 
Chapter 4). The region's fish and coral 
communities are similar to those around 
Fagaitua Bay on the SE coast of Tutuila. 



NPSA- 

Tutuila 

Unit 



American 

Samoa 

overall 



Coral Cover Coral Richness Fish Biomass Fish Richness 



1 (10) 



1(2) 



1 (10) 



1 (10) 



a® 



6 (339) 



5(137) 










6 (347) 





Figure 5.60. Comparison offish and coral data collected in the Tutuila Unit of 
the National Park to data from all of American Samoa. Pie charts depict the 
proportions of high (red), medium (pink), and low (white) values for coral cover, 
coral richness, fish biomass, and fish richness. Number labels represent the 
number of studies and sites (in parentheses) comprising each pie chart. 



Additional References 

Green and Hunter 1998, Craig and Basch 2001, Coles et al. 2003, Pendleton et al. 2005 



Rose Atoll Marine National Monument and National Wildlife Refuge 

Overview 

Rose Atoll Marine National Monument (MNM) lies at the eastern end of the Samoan archipelago and was 
established in 2009 by Presidential Proclamation 8337 to protect the "lands, submerged lands, waters, and 
marine environment around Rose Atoll" and its "dynamic reef ecosystem that is home to a very diverse as- 
semblage of terrestrial and marine species, many of which are threatened or endangered" (Proclamation No. 
8337). The Monument has a rectangular seaward boundary approximately 50 nautical miles from the mean 
low water line of Rose Atoll (Figure 5.61). The volcanic hotspot responsible for the Samoan Island chain, 
the Vailulu'u seamount, lies just west of the monument (Appendix A). The Monument encompasses almost 
35,000 km 2 , including Rose Atoll National Wildlife Refuge (NWR), which was established by cooperative 
agreement between the Government of American Samoa and the USFWS in 1973 and includes the -6.8 km 2 




0.5 

Legend 

Benthic Structure Types 

Aggregate Reef 



Pavement 



^ Rock/Boulder 



Aggregated Patch Reefs || "e^ 



Individual Patch Reef 
Spur and Groove 



s 



Pavement with 
Sand Channels 

Reef Rubble 



5 Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 



Sand 

Artificial 

Unknown 



Fish/Coral Data 

O Survey Sites 

Site Values for each Variable: 

Coral Cover 



Fish Biomass 



cc 


CR 


FB 


FR 



Coral Richness 
Fish Richness 



Deep Water 



Figure 5.61. Mapped benthic habitat (by structure type) and fish and coral survey data within Rose Atoll MNM. Coral cover, coral rich- 
ness, fish biomass, and fish richness values at each survey site are classified as high (red shading), medium (pink shading), or low 
(white shading). Grey shading indicates variables with no data at a given site. Fish and coral survey data are from CRSR and REA. 



of land, submerged land, and waters within the mean low water line of Rose Atoll (RANWR 1974). Rose is 
one of the smallest atolls in the world and is uninhabited by humans. It provides important nesting grounds 
for the threatened green sea turtle and habitat for 17 species of federally protected migratory seabirds and 
shorebirds. It also supports the largest population of giant clams in American Samoa as well as many rare 
species of reef fish (Gombos et al. 2007). Because Rose Atoll is uninhabited, nearshore waters are not im- 
pacted by human use such as from urban runoffs and waste discharge from piggeries. Rose Atoll NWR has 
been closed to the public since its establishment to protect the fish and wildlife in the refuge and is managed 
exclusively by USFWS. Proclamation 8337 prohibits commercial fishing within the Monument and gives the 
Secretary of Commerce (through NOAA) primary management authority over fishery-related activities in 
the marine areas outside of the NWR (16 U.S.C. 1801 et seq., Proclamation No. 8337). The Secretary of 
Commerce has initiated the process to add the marine areas of the Monument to Fagatele Bay NMS in ac- 
cordance with the National Marine Sanctuaries Act (16 U.S.C. 1431 et seq.). 



(a) Total Mapped Area by 
Benthic Structure Type 



_2% 3 %. 



(b) Zonation of 

Coral Reef and Hardbottom 



Lagoon 
(2%) 




Back Reef 
(65%) 



14% 



40% 



Habitat Composition, Reef Fish, 
and Coral Communities 
Most of the -35000 km 2 within 
Rose Atoll MNM is open ocean 
and too deep to map with sat- 
ellite imagery. The -7.9 km 2 of 
mapped benthic habitat within the 
Monument is dominated by coral 
reef and hardbottom structures, 
which together comprise -63% 
of the area within the Monument 
(Figure 5.62a). Coral reef struc- 
tures comprise only -10% of 
the mapped benthic habitat and 
include aggregate reef (~2%), 
aggregated patch reefs (~3%), 
and spur and groove (-5%). In 
comparison, these three struc- 
ture types cover -18% of the 
mapped benthic habitat around 

American Samoa. Almost all of the spur and groove lies in the -1.2 km 2 of benthic habitat outside the mean 
low water line, whereas the lagoon area contains all of the aggregate reef and aggregated patch reefs. In ad- 
dition, pavement covers -40% of the mapped area and -25% of the mapped area within the lagoon is classi- 
fied as unknown because of cloud cover or because it is part of the deeper interior of the lagoon. None of the 
coral reef and hardbottom in the Monument is in the reef flat, while -8% is in the fore reef compared to -22% 
around American Samoa (Figure 5.62b). Almost two-thirds of the coral reef and hardbottom in the Monument 
is in the back reef zone, compared to only 
-3% around American Samoa. Coral Cover Coral Richness Fish Biomass Fish Richness 




Bank/Shelf 

Escarpment 

(5%) 

Bank/Shelf ._ » , Reef Crest 

(11%) [s%) (8%) 

-63% of mapped area is coral reef and hardbottom 
-7.9 km 2 ► -5.0 km 2 

Figure 5.62. (a) Proportion of mapped benthic structure types in Rose Atoll MNM. (b) 
Proportion of coral reef and hardbottom in each reef zone. Structure types or zones rep- 
resenting <1% of the total mapped area are not shown. 



There are 51 surveys distributed in the 
spur and groove and pavement areas sur- 
rounding Rose Atoll and in the inner la- 
goon. Coral data at these sites suggests 
relatively lower cover and richness values 
compared to all of American Samoa. Fish 
biomass is slightly higher relative to all of 
American Samoa, whereas fish richness 
is slightly lower (Figure 5.63). Of note, the 
fish and coral values within the lagoon are 
comprised of different fish and coral com- 
munities (unpublished MDS analyses and 



2(39) 



1 (11) 



2(40) 



2(40) 



Rose 

Atoll 

MNM 



American 

Samoa 

overall 





e>® 



6 (339) 



5(137) 



6 (344) 



0O 




6 (347) 



Figure 5.63. Comparison of fish and coral data collected in the Rose Atoll 
MNM to data from all of American Samoa. Pie charts depict the proportions 
of high (red), medium (pink), and low (white) values for coral cover, coral rich- 
ness, fish biomass, and fish richness. Number labels represent the number of 
studies and sites (in parentheses) comprising each pie chart. 



in Kenyon et al. 2010) and are considerably lower relative to the values located just outside the mean low 
water line in the fore reef and bank/shelf (Figure 5.61 ). 

Biogeographic Characteristics 

Rose Atoll MNM comprises a distinct biogeographic region (Bioregion 17, Chapter 4) that is a hotspot for 
fish biomass and has a unique coral community. Rose lies upstream in the South Equatorial Current relative 
to the rest of the Samoan Archipelago. Analysis of larval connectivity in the region suggests that Rose Atoll 
may be isolated from larval sources and less resilient to disturbance (Chapter 3). Also of note, Rose Atoll is 
dominated by crustose coralline algae and possesses a unique algal community (Tribollet et al. 2010). 

Additional References 

Wass 1981, Wass 1982, Green etal. 1997a, Wegmann and Holzwarth 2006, Schroeder et al. 2008, Tribollet 

etal. 2010, Vroom 2011 

RESULTS: M PA NETWORK ANALYSES 

How much of American Samoa is protected in the MPA network? 

Approximately 427 km 2 of the nearshore area around American Samoa is shallower than 150 m and can be 
considered potential reef ecosystem as defined previously. As of January 2011, only -8% (-32 km 2 ) of the 
potential reef ecosystem area around American Samoa is currently protected by the existing MPA network 
(Figure 5.64a). Considering the type of protection, only -3% of the total potential reef ecosystem (-12 km 2 ) 
has complete no-take restrictions whereas -5% (-20 km 2 ) has other regulations such as gear limits, devel- 
opment restrictions, and bans on commercial fishing (Figure 5.64b). Considering only the -69 km 2 of coral 



(a) 



(b) 



No-Take 



Restrictions Restrictions 
/ 

3% 5% 



53% 




Other 



small percentages of emergent 
vegetation, artificial structures, 
deep water, and unknown 
structure type not shown 



Total Potential Reef Ecosystem - -427 km 2 



32 km 2 

| | Coral Reef 
Hardbottom 

Unconsolidated Sediments 
Algal Plain 
Deep Water 



Figure 5.64. (a) Proportion of the total potential reef ecosystem area around American Samoa by benthic structure type for the entire 
suite of existing MPAs and for the rest of American Samoa. For simplicity, some structure types were aggregated into larger catego- 
ries. Aggregate reef, patch reef, aggregated patch reefs, and spur and groove were aggregated into coral reef. Pavement, pavement 
with patch reefs, pavement with sand channels, reef rubble, and rock/boulder were aggregated into other hardbottom. Mud, sand 
with scattered coral/rock, and sand were aggregated into unconsolidated sediments. Structure types or categories representing <1% 
of the total area are not shown, (b) Proportion of the total potential reef ecosystem with no-take restrictions (dark pink shading) and 
with other fishing restrictions (light pink). 



reef habitats around American Samoa, the existing MPA network protects -10% of the area of those features 
(Figure 5.65a) with only -3% (~2 km 2 ) having complete no-take restrictions and -7% (~5 km 2 ) having other 
regulations (Figure 5.65b). 



(a) 



£V/ S # 



y^s* 



(b) 



2% 




No-Take 
Restrictions 



-10% (~7 km 2 ) in existing MPAs 



24% 




Other 
Restrictions 
/ 



* small percentage of 
patch reef not shown 



39% 
Total Coral Reef Habitat - -69 km 2 



~7km 2 

Aggregate Reef 
Aggregated Patch Reefs 
Individual Patch Reef 
Spur and Groove 



Figure 5.65. (a) Proportion of coral reef habitat by benthic structure type for the entire suite of existing MPAs and for the 
rest of American Samoa. Structure types representing <1% of the total area are not shown, (b) Proportion of coral reef 
habitat with no-take restrictions (dark pink shading) and with other fishing restrictions (light pink). 

Which biogeographic regions and ecological hotspots are represented in the MPA network? 

Fourteen of the twenty ecologically distinct Bioregions identified in Chapter 4 include at least one MPA, leaving 
only six with no representation in the present 
MPA network (Table 5.2). Bioregions not cur- 
rently represented in the existing MPA network 
that have been identified as having unique reef 
fish and/or coral communities in Chapter 4 in- 
clude only Swains Island (Bioregion 16) and 
Aunu'u (Bioregion 8). Overlaying the 36 eco- 
logical hotspots defined among the Bioregions 
with MPA boundaries revealed which hotspots 
are at least partly protected by the existing net- 
work. This simple accounting indicated that 25 
out of 36 hotspots are at least partly protected 
by existing MPAs. Results were broadly con- 
sistent among all four variables (Table 5.2). 

Bioregions defined as hotspots for multiple 

variables may have greater ecological and 

conservation importance relative to regions 

that are hotspots for fewer variables. Seven of 

the Bioregions were defined as hotspots for 3 

, r ,. A . ,, / ui±r Image 22. Shoreline of Bioregion 2 inside Larsen Bay. 

out of the 4 variables (none were hotspots for pho f credit . Matt Kendal| N £ ^ Biogeography . 




Table 5.2. Biogeographic regions, ecological hotspots, and overlap with existing MPAs. Biogeographic 
regions and hotspots are defined in Chapter 4 of this assessment. The number of existing MPAs within 
each Bioregion is summarized in the last column. The bottom two rows summarize the number of 
hotspots for each variable with at least one MPA and the proportion of hotspots for each variable rep- 
resented by the existing MPA network. 



Bioregion 




Cover 


Richness 


Biomass 


Richness 


T . . Existing 
u , ' MPAs within 


1 


X 




X 


X 


3 


2 


2 


X 


X 




X 


3 


1 


3 




X 






1 





4 








X 


1 


3 


5 






X 




1 


2 


6 


X 




X 




2 


2 


7 


X 




X 




2 





8 


X 




X 


X 


3 





9 








X 


1 





10 


X 




X 


X 


3 





11 




X 






1 


3 


12 


X 








1 


2 


13 













1 


14 


X 




X 


X 


3 


2 


15 













1 


16 


X 




X 


X 


3 





17 






X 




1 


2 


18 




X 


X 


X 


3 


2 


19 




X 




X 


2 


1 


20 


X 


X 






2 


1 




10 


6 


10 


10 


36 


25 


Hotspots w/ an 
MPA Present 

Proportion of 

Hotspots 
Represented 




6 


5 


6 


6 








6/10 


5/6 


6/10 


6/10 







all 4 variables) and can be considered relatively high-value sites. These include the SW coast of Tutuila 
between Cape Taputapu and Sail Rock Point (Bioregions 1 and 2), Aunu'u (Bioregion 8), the eastern tip of 
Tutuila (Bioregion 10), Fagamalo area (Bioregion 14), Swains Island (Bioregion 16), and Ofu/Olosega (Bio- 
region 18). Existing MPAs protect portions of four of these high-value Bioregions (1, 2, 14, and 18). The high- 
value Bioregions not currently protected by the existing MPA network are Aunu'u (Bioregion 8), the eastern 
tip of Tutuila (Bioregion 10), and Swains Island (Bioregion 16) (Figure 5.66). 



The simple hotspot and bioregional summaries presented above do not take into consideration two key fac- 
tors that vary considerably among MPAs: size of the protected area and type of protection. The mere pres- 
ence of an MPA within a Bioregion does not guarantee sufficient protection. For example, the SW coast of 
Tutuila (Bioregion 1 , a hotspot for 3 of the 4 fish and coral variables) contains two of the existing MPAs and 
therefore it may appear that this Bioregion is being adequately protected with replication. However, the two 
MPAs in Bioregion 1 together comprise less than 0.4 km 2 of potential reef ecosystem, leaving the vast major- 
ity of the Bioregion unprotected. 



Rose Atoll 



Manu'a Islands (Bioregions 18-20) 




Legend 

■■■■ Bioregion boundaries 
— MPA boundaries 

| Bioregions that are hotspots for 3 fish/coral variables 
coral banks 

• other key features 



Figure 5.66. Distribution of existing MPAs relative to the locations of significant ecological features, including Bioregions that are 
hotspots for three fish/coral variables (Chapter 4) and the mesophotic coral banks surrounding Tutuila (Appendix B). 



Size and regulatory comparisons among MPAs 

MPAs around American Samoa have a very wide range of sizes and protect very different amounts of poten- 
tial reef ecosystem, from <0.02 km 2 to -9.1 km 2 (Figure 5.67, Table 5.3). It is important to consider both the 
proportions of habitats as well as their size when evaluating relative protection of coral reef and hardbottom 
features. For instance, many of the smallest MPAs possess a high proportion of coral reef and hardbottom 
structures, often greater than 80% of their area, and so encompass some key habitats very efficiently. How- 
ever, they may not be large enough to encompass the home range of the fish species they are intended to 
protect. The largest MPAs generally encompass a wider variety of bottom types and therefore have lower 
proportions of coral reef and hardbottom, often less than 50% of their area. These low proportions of coral 
reef and hardbottom can be misleading in judging the relative contributions of an MPA and must be consid- 
ered in the context of MPA size. For example, fifteen of the twenty-two MPAs were smaller than 1 km 2 indi- 
vidually. Collectively these fifteen MPAs encompass only -25% of the protected coral reef and hardbottom 
(-4.4 km 2 of 16.9 km 2 ) around American Samoa. In contrast, at Rose Atoll MNM, the largest MPA, only -50% 
of the potential reef ecosystem is coral reef or hardbottom, but this encompasses 30% of the total protected 
coral reef and hardbottom around all of American Samoa (-5 km 2 out of 16.9 km 2 ). This single MPA protects 
more coral reef and hardbottom than all 15 of the smallest MPAs combined. Larger MPAs are also more likely 
to be effective in protecting fish species with large home ranges. 



Existing MPAs (-32 km 2 ) 





Nu'uuli PalaSMA 
(2.0 km 2 ) 



Rose Atoll MNM* 
(9.1 km 2 ) 




Masausi CFMP Reserve 
(0.2 km 2 ) 



Alega Private Marine Reserve 
(0.1 km 2 ) 




NPS-Ofu unit 
(1.5 km 2 ) 




Sailele CFMP Reserve 
(0.08 km 2 ) 

• 

Leone Pala SMA 

(0.02 km 2 ) 



Pago Pago Harbor SMA 
(1.2 km 2 ) 



NPS-Tutuila unit 
(6.5 km 2 ) 





Fagatele Bay NMS 
(0.7 km 2 ) 




NPS-Ta'u unit 
(4.3 km 2 ) 



Vatia CFMP Reserve 
(0.6 km 2 ) 







Ofu Vaoto Marine Park 
(0.5 km 2 ) 



Fagamalo No-Take MPA 
(2.9 km 2 ) 



Legend 

Benthic Structure Types 

Aggregate Reef 



Fagamalo CFMP Reserve 
(0.4 km 2 ) 



Amaua & Auto 

CFMP Reserve 

(0.4 km 2 ) 



Poloa CFMP Reserve 
(0.4 km 2 ) 



Aoa CFMP Reserve 
(0.3 km 2 ) 



Amanave CFMP Reserve 
(0.3 km 2 ) 



Alofau CFMP Reserve 
(0.3 km 2 ) 



Matu'u & Faganeanea 

CFMP Reserve 

(0.3 km 2 ) 



Aua CFMP Reserve 
(0.2 km 2 ) 



* Rose Atoll MNM includes Rose Atoll NWR 



Pavement 



^£, Rock/Boulder 



**-«**§rr 



Individual Patch Reef 
Spur and Groove 



Pavement with 
Sand Channels 



I Reef Rubble 



J Emergent Vegetation 
Algal Plain 
Mud 



Sand with 
Scattered Coral/Rock 

Sand 

Artificial 

Unknown I Deep Water 



Figure 5.67. Proportion of potential reef ecosystem area by benthic structure type for each existing MPA. 
Pie sizes are scaled relative to potential reef ecosystem area. 

In addition to size differences, the MPAs around American Samoa vary in the type of protection they provide. 
Only two existing MPAs have at least some zone designated with the strongest level of protection, complete 
no-take. These include the entire Fagamalo No-Take MPA and the portion of the Rose Atoll MNM landward 
of the 50 fathom curve (including the NWR). These no-take areas comprise only -3% of the area identified 
as coral reef habitat around American Samoa. In 2000, Governor Tauese Sunia set a goal to protect 20% of 
American Samoa's coral reefs in no-take areas by 2010 (Sunia 2000). Existing regulations can be modified 
or zones created within present MPAs in consultation with DMWR's No-Take MPA Program to partly meet this 



Table 5.3. Potential reef ecosystem area (km 2 ) by benthic structure type for existing MPAs. Areas 
program, for the entire suite of existing MPAs, and for all of American Samoa. 


are given for each individual MPA 


Benthic Classifications 






CO (/) 

cc >, 


Detailed 
structure types 












I 






Private 
Reserve 


SMAs 


Territorial 
Marine 


T - . American 
Tota ~ 

Samoa 


M— 


CD 

s_ 

"CO 

s_ 

o 
O 


Aggregate reef 


0.7 


0.1 


0.3 


1.7 


0.1 


0.1 


0.2 


0.0 


3.1 


19.6 


Aggregated 
patch reefs 


0.1 


0.2 


0.0 


0.2 


1.1 


0.0 


0.2 


0.0 


1.8 


28.3 


Individual patch 
reef 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


1.6 


Spur and groove 


0.2 


0.4 


0.2 


1.2 


0.0 


0.0 


0.0 


0.1 


2.0 


19.1 


Total coral reef 


0.9 


0.8 


0.4 


3.1 


1.3 


0.1 


0.4 


0.2 


7.0 


68.6 


E 
o 

-i— < 
o 

-Q 
"D 

s_ 

CD 

X 


Pavement 


1.1 


3.1 


0.2 


1.9 


0.0 


0.1 


0.2 


0.2 


6.6 


28.8 


Pavement with 
patch reefs 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


2.2 


Pavement with 
sand channels 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.8 


Reef rubble 


1.0 


1.1 


0.0 


0.5 


0.0 


0.0 


0.0 


0.0 


2.6 


8.8 


Rock/Boulder 


0.2 


0.0 


0.0 


0.5 


0.0 


0.0 


0.1 


0.0 


0.8 


2.7 


Total hard bottom 


2.3 


4.2 


0.2 


2.9 


0.1 


0.1 


0.2 


0.2 


9.9 


43.4 


"D 


"D C 
° £ 

co .E 

s ~° 

O 


Mud 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


1.9 


0.0 


1.9 


4.1 


Sand with 
scattered coral/ 
rock 


0.0 


1.1 


0.1 


0.1 


0.0 


0.0 


0.0 


0.0 


1.3 


1.7 


Sand 


0.2 


0.1 


0.0 


0.9 


0.1 


0.0 


0.1 


0.0 


1.3 


36.5 


CD 

6 


Emergent 
vegetation 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.3 


0.0 


0.3 


0.3 


Algal plain 


0.2 


0.0 


0.0 


2.8 


1.5 


0.0 


0.0 


0.0 


4.5 


231.2 


Artificial 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.0 


0.1 


Unknown 


0.0 


1.7 


0.0 


0.6 


0.0 


0.0 


0.2 


0.1 


2.5 


4.3 


Deep Water 


0.0 


1.2 


0.0 


2.0 


0.0 


0.0 


0.0 


0.0 


3.3 


36.6 




Total Area 


3.6 


9.1 


0.7 


12.3 


2.9 


0.1 


3.2 


0.5 


32.4 


426.7 



goal. However, it should be noted that because only 10% of the total coral reef habitat in American Samoa is 
within existing MPAs, even if all were hypothetically re-zoned as no-take, the 20% goal would only be halfway 
achieved. New MPAs encompassing the same total area as all existing MPAs combined would need to be 
implemented. This hypothetical re-zoning example demonstrates that additional large, cross sectional MPAs 
with no-take restrictions such as recently implemented at Fagamalo by DMWR are necessary to accomplish 
this goal. In addition to establishing the no-take site at Fagamalo, DMWR's No-Take Program has identified 
additional priority sites (e.g. Aunu'u, Chapter 4), has conducted standardized biological surveys at those sites 
(Oram 2008), and plans continued engagement with other MPA programs and local communities to build 
strong commitments to achieving the 20% goal. As the MPA network expands under various authorities to 
meet this goal, MPA practitioners in American Samoa can use information in this assessment on larval con- 
nectivity (Chapter 3), fish and coral communities (Chapter 4), benthic features (Appendix B), and additional 
information to identify areas of high ecological value that could be added to the no-take components of the 
MPA network. 




CONCLUSIONS 

Our goal in summarizing the biogeographic 
features within existing MPAs was to provide 
an accounting of the MPA landscape using a 
consistent set of broadly important ecosystem 
variables. However, it should be noted that 
spreading protection among Bioregions or in- 
cluding representation of the particular ecologi- 
cal hotspots defined in Chapter 4 is not an ex- 
plicitly stated goal of the local MPA community 
in American Samoa. It is up to this community 
to identify the specific goals to be achieved by 
the network and a process to achieve them. 
These goals may include biogeographic repre- 
sentation, replication, quantitative targets (e.g. 
20% no-take), ensuring connectivity among 
sites, and protection of specific ecological or 
cultural sites. 




Image 23. Original CFMP sign posted at Fagamalo. New signage was 

under development in 2011 . 

Photo credit: Matt Kendall, NOAA Biogeography. 



While our analysis has focused on broad reef 
fish and coral variables, some existing MPAs 

have little to do with these general ecological variables since they may be designed to protect cultural re- 
sources or specific biota (e.g. Saua and Taisamasama cultural sites on Ta'u). There is a diversity of additional 
features of special importance that are worthy of protection that were not addressed using the general vari- 
ables focused on in this study. For example, the mesophotic banks around Tutuila have some vibrant coral 
reef communities that may be less vulnerable to climate change and nearshore stressors than shallower 
reefs (Riegl and Filler 2003, Bare et al. 2010) but are poorly represented in the existing MPA network. Only 
the Fagamalo No-Take MPA encompasses such features around Tutuila presently. Another feature, Vailulu'u, 
the only volcanically active seamount of the 65 in the EEZs of American Samoa and Samoa and the origin 
of the Samoan archipelago, lies between Rose Atoll and the Manu'a Islands (Appendix A). Vailulu'u lies just 
outside the Rose Atoll MNM and a small boundary modification would encompass its unique hydrothermal 
vent communities and likely eventual emergence as a new island (Staudigel et al. 2006). Another example 
of a special and unique area is off the southwest coast of Ta'u and includes several coral heads of the spe- 
cies Pontes lutea (Fisk and Birkeland 2002, Brown et al. 2009) that are remarkable for their enormous size. 
These features presently lie outside the National Park boundary on Ta'u and are not currently protected by 
the existing MPA network. 



In addition to protecting such special or unique features at single sites, replication of non-unique regions 
or habitats at multiple sites that are similar is an important principle of MPA network design. This spreads 
the risk of degradation or loss of a particular ecosystem or resource over multiple MPAs and enhances the 
resiliency of the protected ecosystem or resource. Given the susceptibility of reef ecosystems to anthropo- 
genic and natural disturbance (e.g. crown-of-thorns starfish, tsunamis, pollution), this should be an important 
consideration for MPA authorities in American Samoa. For example, protecting Larsen Bay would almost 
perfectly replicate the very similar reef ecosystem in the adjacent and already protected Fagatele Bay NMS. 
Also, the discontinuous coral banks around Tutuila (e.g. Taema, Nafanua, and many others) offer abundant 
choices to replicate protection of bank features such as those in the Fagamalo No-Take MPA. Additional re- 
gions lacking representation or replication in the existing network are identified in Table 5.2 and Figure 5.66. 
The connections among MPAs due to factors such as swimming within a home range for adult fish, onto- 
genic habitat shifts, and dispersal offish and coral larvae are also important to consider in network design. 
Telemetry or tagging studies are needed to quantify the scales and frequencies of fish movements among 
and across MPA boundaries. Hydrodynamic models are needed to quantify connections among MPAs due to 
larval dispersal. Chapter 3 of this assessment used a broad-scale hydrodynamic model to evaluate connec- 
tions among islands of the entire archipelago. It was found that Samoa's much larger islands and coral reef 
area are the largest source of larvae in the archipelago, some of which are circulated to American Samoa 



via the South Equatorial Counter Current (Chapter 3). Most of American Samoa (except Swains Island) in 
turn, lies upstream of Samoa in the South Equatorial Current and many larvae spawned in MPAs there may 
seed the reefs of Samoa. Distances between islands and MPAs and their positions in these currents partly 
dictate which species of larvae may be spawned in one MPA but then be transported to sustain the resident 
population in another. For example, the MPA at Rose Atoll MNM not only encompasses a large proportion 
of the protected reef habitats in American Samoa, but connectivity models indicate that some of its larval 
production gets exported downstream to the other islands in the archipelago. All existing and proposed MPAs 
should be evaluated in a similar context. For maximum benefit, the broad scale hydrodynamic models in 
this assessment must be coupled with finer-scale models to understand smaller current patterns and eddies 
around particular islands. DMWR and ASEPA are presently developing such a model around Tutuila to better 
understand localized larval transport and identify potential MPA sites that may provide resilient sources of 
larvae to the broader ecosystem and MPA network. 

A comprehensive and cooperative MPA strategy with protection goals that involves the many MPA manage- 
ment programs in American Samoa has been developed over the past 5 years (Oram 2006, 2008, Damitz 
2007). The strategy consists of 5 action plans covering governance and administration, MPA designation, 
education and outreach, research and monitoring, and enforcement, with each containing time-bound goals 
and objectives. One of the key overarching goals is the need for effective dialogue and collaboration among 
MPA programs to most effectively align resource protection needs with appropriate management authorities. 
Effective communication and collaboration among MPA programs is vital to not only minimize stakeholder 
confusion at the village level but also to prevent competitive obstruction among many well intending agencies 
working in a relatively small region. To date, the MPA network strategy has facilitated the establishment of an 
MPA Coordinator within the Territorial government's Coral Reef Advisory Group and the formation of an MPA 
Network Working Group. Key next steps in the evolution of the MPA network strategy are to, 1 ) define over- 
all quantitative resource protection goals, some of which have been stated by individual programs already 
such as establishment of 20% of coral reefs as no-take refugia (Oram 2008), 2) quantify what resources are 
protected within the existing MPA network (i.e. this analysis and others like it, e.g. Fisk and Birkeland 2002, 
NOAA NOS 2009), 3) identify ways to accomplish resource protection goals through additional MPAs or mod- 
ification of the size, regulations, or spatial arrangement of existing MPAs, and 4) identify which management 
authorities, or more likely which mix of programs, can contribute the appropriate combination of financial, 
material, and stakeholder support roles necessary to successfully manage each site and the overall network. 

The MPA network in American Samoa is an ever-changing landscape. As of January 2011, a number of po- 
tential MPAs were at various stages in the "proposal" process. These MPAs would add potential reef ecosys- 
tem and additional coral reef and hardbottom areas to the MPA network and may protect Bioregions, ecologi- 
cal hotspots, or special features noted in this report that are not currently protected by the existing network. 
It is also important to note that MPAs in American Samoa vary considerably in terms of their permanence. 
Some provide fixed protection in perpetuity barring new legislation and others provide more changeable 
protection for shorter durations at which point they must be renewed. Many CFMP reserves, for instance, 
are established for an initial period of 2-3 years, after which modifications to protection level, boundaries and 
regulations can be made by village leaders in response to changing needs and resource conditions. As the 
MPA landscape around American Samoa evolves, the components of this Biogeographic Assessment can 
be used to evaluate the ecological contributions of additions to the network on the basis of protected habitats 
(Appendix B), reef fish and coral communities (Chapter 4 and Appendix C), and larval connectivity (Chapter 
3). 



Once comprehensive benthic maps and MPA boundary files become available for the Samoan Islands of 
Upolu and Savai'i, a full accounting of the MPA Network for both American Samoa and Samoa will be pos- 
sible. This is a critical next step in regional MPA planning given the close proximity and high potential for 
interdependency of MPAs across both these jurisdictions. A comprehensive, archipelago-wide MPA strategy 
is recommended that maximizes the benefits of this potential connectivity and promotes resiliency of not only 
the wider MPA network, but also coral reef ecosystems more generally throughout the archipelago. 



ACKNOWLEDGEMENTS 



REFERENCES 



We gratefully acknowledge the assistance of several people who made very helpful contributions to this 
chapter. Phil Wiles at American Samoa EPA contributed valuable information from the ASEPA Piggery Com- 
pliance Program. Christin Reynolds at American Samoa DOC provided GIS boundaries for the Alega Private 
Marine Reserve, Ofu Vaoto Territorial Marine Park, and the Special Management Areas plus contextual data 
used to characterize the watersheds adjacent to each MPA. Frank Pendleton of USFWS provided a simple 
and understandable description of the complicated management regime at Rose Atoll Marine National Monu- 
ment. Sarah Bone of NPS helped us draft corrected GIS shapefiles of the NPSA boundaries. Ken Buja of 
the NOAA Biogeography Branch worked with Selaina Vaitautolu, the CFMP program manager, to produce 
GIS shapefiles of the CFMP reserve boundaries. Tisa Fa'amuli provided a description of the rationale and 
implementation for the Alega Private Marine Reserve. 



15 C.F.R. 922.100-104. Title 15: Commerce and Foreign Trade, Part 922: National Marine Sanctuary Program Regula- 
tions, Subpart J: Fagatele Bay National Marine Sanctuary. 

16 U.S.C. 1. Title 16: Conservation. Chapter 1: National Parks, Military Parks, Monuments, and Seashores, Subchapter 
I: National Park Service, Section 1: Service created; director; other employees. 

16 U.S.C. 410qq-410qq-1. Title 16: Conservation, Chapter 1: National Parks, Military Parks, Monuments, and Sea- 
shores, Subchapter LIX-O: National Park of American Samoa, Section 410qq: Findings and purpose, Section 410qq-1: 
Establishment. 

16 U.S.C. 431-433. Title 16: Conservation, Chapter 1: National Parks, Military Parks, Monuments, and Seashores, 
Subchapter LXI: National and International Monuments and Memorials, Section 431: National Monuments; reservation 
of lands; relinquishment of private claims. 

16 U.S.C. 668dd-668ee. Title 16: Conservation, Chapter 5A: Protection and conservation of wildlife, Subchapter III: 
Endangered species offish and wildlife, Section 668dd: National Wildlife Refuge System, Section 668ee: Definitions. 

16 U.S.C. 1431. Title 16: Conservation, Chapter 32: Marine Sanctuaries, Section 1431: Findings, purposes, and poli- 
cies; establishment of system. 

16 U.S.C. 1801. Title 16: Conservation, Chapter 38: Fishery conservation and management. 

American Samoa Administrative Code (ASAC) § 24.10. Title 24: Ecosystem Protection and Development, Chapter 10: 
Community-Based Fisheries Management Program. 

American Samoa Administrative Code (ASAC) § 24.1001. Title 24: Ecosystem Protection and Development, Chapter 
10: Community-Based Fisheries Management Program, Section 1: Authority 

American Samoa Administrative Code (ASAC) § 24.1008 (c)(i). Title 24: Ecosystem Protection and Development, 
Chapter 10: Community-Based Fisheries Management Program, Section 8: Fishing or taking fish in a village marine 
protected area. 

American Samoa Administrative Code (ASAC) § 26.0221 . Title 26: Environmental Safety and Land Management, Chap- 
ter 2: Coastal Management, Section 21: Special Management Areas. 

American Samoa Code Annotated (ASCA) § 18.0214. Title 18: Parks and Recreation, Chapter 2: Department of Parks 
and Recreation, Section 14: Establishment of Ofu-Vaoto Marine Park. 

American Samoa Code Annotated (ASCA) § 24.0503. Title 24: Natural Resources & Environmental Ecosystem Protec- 
tion & Development, Chapter 5: Coastal Management Program; Section 3: Designation of coastal zone and special 
management areas. 

American Samoa Department of Marine and Wildlife Resources (ASDMWR). 2001. Fisheries Management Plan for 
Poloa. 7 pp. + Appendices. 

American Samoa Department of Marine and Wildlife Resources (ASDMWR). 2002a. Fisheries Management Plan for 
Alofau. 9 pp. + Appendices. 

American Samoa Department of Marine and Wildlife Resources (ASDMWR). 2002b. Fisheries Management Plan for 
Vatia. 11 pp. + Appendices. 

American Samoa Department of Marine and Wildlife Resources (ASDMWR). 2003a. Fisheries Management Plan for 
Amaua and Auto. 9 pp. + Appendices. 

American Samoa Department of Marine and Wildlife Resources (ASDMWR). 2003b. Fisheries Management Plan for 
Aua. 11 pp. + Appendices. 

American Samoa Department of Marine and Wildlife Resources (ASDMWR). 2003c. Fisheries Management Plan for 
Fagamalo. 10 pp. + Appendices. 




American Samoa Department of Marine and Wildlife Resources (ASDMWR). 2003d. Fisheries Management Plan for 
Masausi. 8 pp. + Appendices. 

American Samoa Department of Marine and Wildlife Resources (ASDMWR). 2005. Fisheries Management Plan for 
Matu'u and Faganeanea. 6 pp. 

American Samoa Environmental Protection Agency (ASEPA). December 6, 1991. Public Health Directive: Contami- 
nated Fish in Harbor. 

American Samoa Environmental Protection Agency (ASEPA) Piggery Compliance Program. February 2011. 

Andrews, Z. 2004. Effects of the Community-Based Fishery Management Programme on the coral reef communities in 
American Samoa. MSc Thesis, University of Wales Bangor. 128 pp. + Appendices. 

Bare, A.Y., K.L. Grimshaw, J.J. Rooney, M.G. Sabater, D. Fenner, and B. Carrol. 2010. Mesophotic communities of the 
insular shelf at Tutuila, American Samoa. Coral Reefs 29: 369-377. 

Birkeland, C.E., R.H. Randall, R.C. Wass, B. Smith, and S. Wilkins. 1987. Biological resource assessment of the Fa- 
gatele Bay National Marine Sanctuary. NOAA Technical Memorandum NOS MEMD 3. Marine and Estuarine Manage- 
ment Division. Washington, DC. 232 pp. 

Birkeland, C.E., R.H. Randall, and S.S. Amesbury. 1994. Coral and reef-fish assessment of the Fagatele Bay National 
Marine Sanctuary. Report to the National Oceanic and Atmospheric Administration, U.S. Department of Commerce. 126 
pp. 

Birkeland, C.E., R.H. Randall, A.L. Green, B.D. Smith, and S. Wilkins. 2003. Changes in the coral reef communities of 
Fagatele Bay NMS and Tutuila Island (American Samoa), 1982-1995. Fagatele Bay National Marine Sanctuary Science 
Series. Pago Pago, AS. 226 pp. 

Birkeland, C, A. Green, C. Mundy, and K. Miller. 2004. Long term monitoring of Fagatele Bay National Marine Sanctuary 
and Tutuila Island (American Samoa) 1985 to 2001: summary of surveys conducted in 1998 and 2001. Report to the 
National Oceanic and Atmospheric Administration, U.S. Department of Commerce. 158 pp. 

Brown, D.P., L. Basch, D. Barshis, Z. Forsman, D. Fenner, and J. Goldberg. 2009. American Samoa's island of giants: 
massive Porites colonies at Ta'u island. Coral Reefs 28: 735. 

Coles, S.L., PR. Reath, PA. Skelton, V. Bonito, R.C. DeFelice, and L. Basch. 2003. Introduced marine species in Pago 
Pago Harbor, Fagatele Bay, and the national park coast, American Samoa. Final report to the U.S. Fish and Wildlife 
Service, Fagatele Bay National Marine Sanctuary, National Park of American Samoa, and American Samoa Department 
of Marine and Natural Resources. Bishop Museum Technical Report No. 26. Honolulu, HI. 182 pp. 

Cornish, A.S. and E.M. DiDonato. 2004. Resurvey of a reef flat in American Samoa after 85 years reveals devastation 
to a soft coral (Alcyonacea) community. Marine Pollution Bulletin 48: 768-777. 

Craig, P. and L. Basch. 2001. Developing a coral reef monitoring program for the National Park of American Samoa. 
National Park Service. 17 pp. 

Craig, P., C. Birkeland, and S. Belliveau. 2001. High temperatures tolerated by a diverse assemblage of shallow-water 
corals in American Samoa. Coral Reefs 20: 185-189. 

Craig, P., G. DiDonato, D. Fenner, and C. Hawkins. 2005. The state of coral reef ecosystems of American Samoa, pp. 
312-337. In Waddell J. (ed.), 2005. The status of coral reef ecosystems of the US and Pacific freely associated states. 
NOAA Tech Report NOS NCCOS 11. 522 pp. 

Dahl, A.L. andA.E. Lamberts. 1977. Environment impact on a Samoan coral reef: a resurvey of Mayor's 1917 transect. 
Pacific Science 31: 309-319. 



Damitz, B. 2007. American Samoa marine protected area network strategy. Technical report prepared for the American 
Samoa Coral Reef Advisory Group by Chromis LLC. August 2007. 53 pp. 

Fagatele Bay National Marine Sanctuary (FBNMS) Regulations. 51 Fed. Reg. 15878-15883 (29 April 1986). 

Fisk, D. and C. Birkeland. 2002. Status of coral communities on the volcanic islands of American Samoa. A re-survey of 
long-term monitoring sites. Report to the Department of Marine and Wildlife Resources. Pago Pago, AS. 134 pp. 



Friedlander, A. 1993. Preliminary assessment and recommendations for long-term monitoring of reef fish populations in 
the proposed National Park on Ofu Island, American Samoa. Draft report to the National Park Service. 39 pp. 

Garrison, V.H., K. Kroeger, D. Fenner, and P. Craig. 2007. Eutrophication comparison of coral reefs in Ofu and Olosega. 
Report to the American Samoa Department of Commerce. 19 pp. 

Gilman, E., J. Ellison, and R. Coleman. 2007. Assessment of mangrove response to projected relative sea-level rise and 
recent historical reconstruction of shoreline position. Environmental Monitoring and Assessment 124: 105-130. 

Gombos, M.J., R. Oram, and S. Vaitautolu. 2007. American Samoa Coral Reef MPA Summary, pp. 11-25. In Wusinich- 
Mendez, D. and C. Trappe (ed.), 2007. Report on the Status of Marine Protected Areas in Coral Reef Ecosystems of 
the United States Volume 1: Marine Protected Areas Managed by U.S. States, Territories, and Commonwealths: 2007. 
NOAA Technical Memorandum CRCP 2. NOAA Coral Reef Conservation Program. Silver Spring, MD. 129 pp. + Ap- 
pendices. 

Green, A.L. , C.E. Birkeland, R.H. Randall, B.D. Smith, and S. Wilkins. 1997a. 78 years of coral reef degradation in Pago 
Pago Harbor: a quantitative record. Proceedings of the 8th International Coral Reef Symposium 2: 1883-1888. 

Green, A., J. Burgett, M. Molina, D. Palawski, and P. Gabrielson. 1997b. The impact of a ship grounding and associ- 
ated fuel spill at Rose Atoll National Wildlife Refuge, American Samoa. Report to U.S. Fish and Wildlife Service, Pacific 
Islands Ecoregion. Honolulu, HI. 65 pp. 

Green, A. and C. Hunter. 1998. A preliminary survey of the coral reef resources in the Tutuila unit of the National Park 
of American Samoa. Report to the National Park of American Samoa. 42 pp. 

Green, A.L. , C.E. Birkeland, and R.H. Randall. 1999. Twenty years of disturbance and change in Fagatele Bay National 
Marine Sanctuary, American Samoa. Pacific Science 53: 376-400. 

Green, A. and T Hughes. 1999. Rapid ecological assessment of the coral reef resources in the Ta'u unit of the National 
Park of American Samoa. Draft report to the National Park of American Samoa. 14 pp. 

Green, A., K. Miller, and C. Mundy. 2005. Long term monitoring of Fagatele Bay National Marine Sanctuary, Tutuila Is- 
land, American Samoa: results of surveys conducted in 2004, including a re-survey of the historic Aua Transect. Report 
to the National Oceanic and Atmospheric Administration. U.S. Department of Commerce. 93 pp. 

Helfrich, P. 1975. An assessment of the expected impact of a dredging project proposed for Pala Lagoon, American 
Samoa. Sea Grant Technical Report UNIHI-SEAGRANT-TR-76-02. 84 pp. 

Houk, P. 2010. American Samoa Community Based Monitoring Program Data Assessment and Program Development 
Workshop Summary and Recommendations. 16 pp. 

Hunter, C.L., W.H. Magruder, A.M. Friedlander, and K.Z. Meier. 1993. Ofu reef survey baseline assessment and recom- 
mendations for long-term monitoring of the proposed National Park, Ofu, American Samoa. Final report to the National 
Park Service. 90 pp. 

lose, PK. and J. McConnaughey. 1993. Fishery resources in Pala Lagoon. American Samoa Government, Department 
of Marine and Water Resources Biological Report Series, No. 37. Pago Pago, AS. 42 pp. 

Itano, D. and T Buckley. 1988. The coral reefs of the Manu'a Islands, American Samoa. American Samoa Government, 
Department of Marine and Water Resources Biological Report Series, No. 12. Pago Pago, AS. 26 pp. 

Kenyon, J.C., J.E. Maragos, and S. Cooper. 2010. Characterization of coral communities at Rose Atoll, American Sa- 
moa. Atoll Research Bulletin No. 586. National Museum of Natural History, Washington, DC. 30 pp. 

King, M. and U. Faasili. 1998. A network of small, community-owned village fish reserves in Samoa. Parks 8:11-16. 

Kluge, K. 1992. Seasonal abundances of zooplankton in Pala Lagoon. American Samoa Government, Department of 
Marine and Water Resources Biological Report Series, No. 36. Pago Pago, AS. 33 pp. 

Maragos, J.E., K.Z. Meier, and C.L. Hunter. 1995. Reef mapping and beach monitoring project at the Ofu unit of the 
National Park of American Samoa and adjacent territorial reef park at Ofu Island. Report for the National Park Service 
prepared by Corial. Honolulu, HI. 61 pp. 

Mayor, A.G. 1924. Structure and ecology of Samoan reefs. Carnegie Institute of Washington Publications 340: 1-25. 



McConnaughey, J. 1993. The shoreline fishery of American Samoa in FY92. American Samoa Government, Depart- 
ment of Marine and Water Resources Biological Report Series, No. 41. Pago Pago, AS. 68 pp. 

Mesophotic Coral Ecosystems. 2010. A research cooperative between the NOAA Center for Sponsored Coastal Ocean 
Research, Perry Institute of Marine Science, and the Centre for Marine Studies at the University of Queensland. Web- 
site accessed March 2010, http://www.mesophotic.org. 

Musburger, C. 2004. Monitoring recommendations for the Community-Based Fisheries Management Program of Tu- 
tuila, American Samoa. Report prepared for the Department of Marine and Wildlife Resources, Pago Pago, AS. 48 pp. 

National Park Service (NPS). 1988. National Park Feasibility Study, American Samoa. Pacific Area Office, Honolulu, 
HI. 138 pp. 

National Park Service (NPS). 1997. General management plan/environmental impact statement, National Park of Amer- 
ican Samoa. Pacific Area Office, Honolulu, HI. 81 pp. 

NOAA NCCOS (National Centers for Coastal Ocean Sciences). 2005. Shallow water benthic habitats of American 
Samoa, Guam, and the Commonwealth of the Northern Marianas (CD-ROM). NOAA Technical Memorandum NOS 
NCCOS 8, Biogeography Team. Silver Spring, MD. 

NOAA National Ocean Service (NOS). 2009. Coral Reef Habitat Assessment for U.S. Marine Protected Areas: U.S. 
Territory of American Samoa. NOAA Coral Reef Conservation Program and NOS Special Projects Office. Silver Spring, 
MD. 26 pp. 

Oram, R.G. 2006. American Samoa coral reef marine protected area strategy. American Samoa Government, Depart- 
ment of Marine and Wildlife Resources. January 2007. Pago Pago, AS. 25 pp. 

Oram, R.G. 2008. Marine protected area master plan: A manual to guide the establishment and management of no-take 
marine protected areas. American Samoa Government, Department of Marine and Wildlife Resources Biological Report 
Series 2008-01 . Pago Pago, AS. 64 pp. 

Orcutt, A.M. 1993. Fishes observed during the Tutuila coastal resources inventory. Report prepared for American Sa- 
moa Department of Marine and Wildlife Resources, Pago Pago, AS. 67 pp. 

Pedersen Planning Consultants. 2000a. American Samoa Watershed Protection Plan, Volume 1: Watersheds 1-23. 
Prepared for American Samoa Environmental Protection Agency and American Samoa Coastal Management Program. 
390 pp. 

Pedersen Planning Consultants. 2000b. American Samoa Watershed Protection Plan, Volume 2: Watersheds 24-35. 
Prepared for American Samoa Environmental Protection Agency and American Samoa Coastal Management Program. 
313 pp. 

Pedersen Planning Consultants. 2000c. American Samoa Watershed Protection Plan, Volume 3: Watersheds 36-41. 
Prepared for American Samoa Environmental Protection Agency and American Samoa Coastal Management Program. 
124 pp. 

Pendleton, E.A., E.R. Thieler, and S.J. Williams. 2005. Coastal vulnerability assessment of National Park of American 
Samoa (NPSA) to sea-level rise. U.S. Geological Survey Open-File Report 2005-1055. 

Peshut, P.J., R.J. Morrison, and B.A. Brooks. 2007. Arsenic speciation in marine fish and shellfish from American Sa- 
moa. Chemosphere 71: 484-492. 

Ponwith, B.J. 1992. The Pala Lagoon subsistence fishery. American Samoa Government, Department of Marine and 
Wildlife Resources Biological Report Series, No. 27. 28 pp. 

Proclamation No. 8337, 74 Fed. Reg. 1577 (January 12, 2009). 

Randall, J.E. and D.M. Devaney. 1974. Marine biological surveys and resource inventory of selected coastal sites at 
American Samoa. U.S. Army Corps of Engineers. Pacific Ocean Division. 108 pp. 

Riegl, B. and W.E. Piller. 2003. Possible refugia for reefs in times of environmental stress. International Journal of Earth 
Science 92: 520-531. 



Rose Atoll National Wildlife Refuge (RANWR), American Samoa. 39 Fed. Reg. 71-76 (January 2, 1974). 



Samuelu, J. 2003. Village fish reserve report for Vaisala (re-survey). August 2003. Assessment, Research and Manage- 
ment Section Fisheries Division. Ministry of Agriculture, Forestry, Fisheries, and Meteorology. 6 pp. 

Schroeder, R.E., A.L. Green, E.E. DeMartini, and J.C. Kenyon. 2008. Long term effects of a ship-grounding on coral reef 
fish assemblages at Rose Atoll, American Samoa. Bulletin of Marine Science 82: 345-364. 

Smith, L. and C. Birkeland. 2003. Managing NPSAs coral reefs in the face of global warming: research project report 
for year 1. University of Hawaii at Manoa 10-03 Technical Report. Honolulu, HI. 32 pp. 

Staudigel, H., S.R. Hart, A. Pile, B.E. Bailey, E.T Baker, S. Brooke, D.P Connelly, L. Haucke, C.R. German, I. Hudson, 
and others. 2006. Vailulu'u seamount, Samoa: Life and death on an active submarine volcano. Proceedings of the Na- 
tional Academy of Sciences of the United States of America 103: 6448-6453. 

Sunia, T 2000. Letter from Governor Tauese Sunia to Lelei Peau (Chairperson of the American Samoa Governor's 
Coral Reef Advisory Group) regarding coral reef protection. August 2, 2000. American Samoa Government. 

Tribollet, A.D., T Schils, and PS. Vroom. 2010. Spatio-temporal variability in macroalgal assemblages of American 
Samoa. Phycologia 49: 574-591. 

Yamasaki, G., D. Itano, and R. Davis. 1985. A study and recommendations for the management of the mangrove and 
lagoon areas of Nu'uuli and Tafuna, American Samoa. Economic Development Planning Office. Pago Pago, AS. 99 pp. 

Vroom, PS. 2011. "Coral dominance": A dangerous ecosystem misnomer? Journal of Marine Biology. 2011 Article ID 
164127, 8 pp. doi:10.1155/2011/164127. 

Wass, R.C. 1981. The fishes of Rose Atoll. Unpublished Report, Office of Marine and Wildlife Resources, Government 
of American Samoa, Pago Pago, AS. 10 pp. 

Wass, R.C. 1982. The fishes of Rose Atoll - Supplement I. Unpublished Report, Office of Marine and Wildlife Resourc- 
es, Government of American Samoa, Pago Pago, AS. 11 pp. 

Wegmann, A. and S. Holzwarth. 2006. Rose Atoll National Wildlife Refuge research compendium. Report prepared for 
U.S. Fish and Wildlife Service by Sula Restoration Ecology. Honolulu, HI. 93 pp. 



Appendix A: Seamounts within the Exclusive Economic Zones 

of Samoa and American Samoa 

Laurie B. Bauer 1 and Matthew S. Kendall 2 

INTRODUCTION 

Seamounts are underwater mountains of volcanic origin. They are often formed near mid-ocean ridges or 
subduction zones at the edges of tectonic plates but also occur over upwelling plumes ("hotspots") within 
plate boundaries (Wessel 2001). Like all geologic formations, seamounts change in shape and height over 
millions of years as a result of the gradual processes of volcanic growth upward out of the seafloor, growth 
of coral reefs if emergent or shallow enough, and eventually the processes of erosion and subsidence or 
sinking of the reshaped structure back into the seafloor. The Samoan Archipelago is part of a hotspot chain 



i 




lA-fr- 



iFfl 



I2£- 



- ; : 



-ira 



-in 



-in 



Figure A.1 . Seamounts of the Samoan Exclusive Economic Zones. 



1 NOAA/NOS/NCCOS/CCMA/Biogeography Branch and Consolidated Safety Services, Inc., Fairfax, VA, under NOAA 
Contract No. DG133C07NC0616 

2 NOAA/NOS/NCCOS/CCMA Biogeography Branch 




that extends from the volcanically active Vailulu'u seamount in the east to west of the island of Savai'i (Hart 
et al. 2006) and includes examples of many stages in the seamount life cycle. The region also includes many 
seamounts not associated with the Samoan hotspot (Figure A. 1). 

Seamounts are not only interesting features geologically as described above, but also biologically in that 
they represent oases of biodiversity relative to the comparatively barren ocean and seafloor surrounding 
them. Seamounts offer an array of habitat opportunities, current fields, and depth zones for plankton, fish 
and corals to occupy, they play a role as "stepping stone" features connecting populations of reef fish and 
corals between islands, are known gathering sites for many pelagic fish species, and consequently are 
popular destinations for fishing and scientific study (Rogers 1994). 

The scientific definition of a seamount has evolved overtime (Staudigel et al. 2010) and has variously been 
based on some minimum height above the seafloor, gravity anomalies of even fully subsided or buried sea- 
mounts, and has included emergent islands by some definitions. In this assessment seamounts are defined 
as totally submerged but extending a minimum of 150 m above the seafloor. The objective of this appendix 
is to provide a characterization of seamounts within the Exclusive Economic Zones (EEZs) of Samoa and 
American Samoa. Of particular importance are those shallow enough to be colonized by reef fish and corals. 



METHODS 

Seamounts are typically mapped 
using sonar and satellite altime- 
try. While sonar based mapping is 
the most direct method and pro- 
vides detailed resolution, it is ex- 
pensive and generally limited in 
spatial coverage. Satellite altime- 
try in contrast, which can be used 
to detect seamounts indirectly 
due to variations in the Earth's 
gravity field (Wessel 2001, Wes- 
sel et al. 2010), is available at a 
global scale but is comparatively 
coarse resolution. 

Two datasets were used to char- 
acterize seamounts in the Sa- 
moan Archipelago. The Global 
Seamount Census (Wessel 2001 , 
Wessel et al. 2010) is a global 
database of -12,000 seamount 
features from satellite-derived 
bathymetry (Smith and Sandwell 
1997). The Seamount Biogeo- 
sciences Network provides a 
characterization for -1,800 sea- 
mounts worldwide, including 
available multibeam data, rel- 
evant literature, and morphologi- 
cal characteristics (Koppers et 
al. 2010a). Seamounts from both 
sources were plotted and ex- 
amined in concert with available 
bathymetry datasets. Three ba- 
thymetry data sources were used 



12 i 
10 - 

>, 8 " 
o 

c 

=> 6 - 

O" 

CD 

s_ 

^ 4- 
2 - 
- 



11 



10 



II 



500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 



Depth of top (m) 



o 

c 

CD 

D" 
CD 



18 

16 - 
14 - 
12 
10 - 
8 - 



4 - 
2 - 




b) 



500 



1000 



1500 



2000 



2500 



3000 



3500 



4000 



Height (m) 



Figure A.2. Frequency distribution of seamounts within the Samoan and American 
Samoan EEZ based on a) depth of seamount top and b) height. 



to obtain complete coverage of the study region. 
High resolution (5-40 m) bathymetry based on 
multibeam sonar was obtained from the Pa- 
cific Islands Benthic Habitat Mapping Center 
(http://www.soest.hawaii.edu/pibhmc.htm) for 
three seamounts: Vailulu'u, Muli (locally known 
as Northeast Bank) and Tulaga (Two-Percent 
Bank). Moderate resolution (180 m) bathymetry 
from merged multibeam and satellite data were 
downloaded from Koppers et al. (2010a). Last, 
one-minute bathymetry estimated from satellite 
altimetry, ship depth soundings, and other sourc- 
es (http://topex.ucsd.edu, Smith and Sandwell 
1997), was used where the two finer resolution 
datasets lacked coverage. 

In many cases, seamount locations disagreed 
slightly between the two data sources and were 
therefore moved slightly in this characterization 
to more precisely identify the approximate peak 
of each seamount based on bathymetry. A few 
features from the satellite derived dataset (Wes- 
sel et al. 2010) lacked significant bathymetric re- 
lief and were removed. These features may have 
been gravity anomalies representing "buried" 
seamounts (Wessel 2001, Wessel et al. 2010). 
Additional features not included in either sea- 
mount dataset were identified within the EEZs 
of Samoa and American Samoa based on the 
bathymetry. Summary information for each sea- 
mount feature was compiled including peak co- 
ordinates, depth of peak, and total height above 
the seafloor. The depth of the peak and base of 
the seamount was estimated in ArcGIS and the 
height was calculated as the difference between 
these estimates. Seamounts are summarized 
in tabular form, as histograms based on peak 
height and depth, and in map format for both the 
region and as individual maps for features with 
potential reef communities. 

RESULTS AND DISCUSSION 

A total of 65 seamount features were identified; 
48 within the EEZ of American Samoa, 16 with- 
in the EEZ of Samoa, and one, Tisa Seamount 
situated on the EEZ boundary (Figure A. 1; Table 
A.1). Approximately 20 of the seamounts in the 
study area are derived from the Samoan hotspot 
and lie along the axis of the archipelago. This 
group includes many of the largest and shallow- 
est seamounts in the region. In addition, there 
are two groups of smaller seamounts in the 
southeastern and northern regions of the Ameri- 
can Samoa EEZ. 



V Jilulu'u Soamouini 
VC*0Q4 BLtrtyrnrirlc (flip 



Mnri-tttr 

GiUMji-ISIi 




ii xwuvfii-viy -4sr -^q -3*w -xh -nH-jnp-^fi-vr- -jw f 



Figure A.3. Vailulu'u Seamount, the active hotspot for the Samoan 
Island Chain. 



PipjCua Guyat 

IIP** MM-iraH - ! ■■■■ ip*i- lit. ' 







1, 



rrtftj 




'VY. 





Figure A.4. Papatua Guyot (South Bank). 




Tuttpi Sflftmeunl 



iMI^tfl B J ny II HUK Mlp 



9c*Jt ■ era 1#H Me*fl«l Btmyni«r|c M»p 






pmmi 



. J _,.^., r . 



ira-?^* 




■rflB 



i.Tl 



1*« 






■■Y-V A« 



:-dMMDC--HQiDW-9Ul 




Figure A.5. Tulaga (East Bank). 



iiraax 




~l 

I 

I 
I 
I 



Figure A.6. Muli Guyot (NE Bank). 



**tf s«4 4+lhymrtrR^ Hip 



^3ji 







Figure A.7. Pasco Seamount. 



Figure A.8. Toafilemu Seamount. 



Table A.1. Locations and morphological characteristics of seamounts within the EEZ of American Samoa and Samoa. 1= Koppers et al. 
2010a; 2= Wessel et al. 2010; 3= estimates based on bathymetry from Koppers et al. (2010a); 4= estimates based on bathymetry from 
Smith and Sandwell (1997), 5=estimates based on bathymetry from NOAAPIBMC, 6= seamounts identified from visual inspection of the 
bathymetry. Names are based on seamount databases with local names provided in parenthesis where known. 



Name 




EEZ Source Longitude Latitude Depth of top (m) 134 Height (m) 134 


Malu Malu Seamount 


American Samoa 


1,2 


-169.7856 


-14.6006 


286 (3) 


1,884(3) 


Malulu Seamount 


American Samoa 


1,2 


-168.6422 


-14.4728 


2380 (3) 


1,396(3) 


Muli Guyot (NE Bank) 


American Samoa 


1,2 


-170.0822 


-14.0479 


49(5) 


2,926 (3) 


Papatua Guyot (South Bank) 


American Samoa 


1,2 


-170.6433 


-14.8880 


23(3) 


3,629 (3) 


Soso Seamount 


American Samoa 


1,2 


-170.2254 


-13.7608 


1,820(3) 


1,714(3) 


Tama'l Seamount 


American Samoa 


1 


-170.5393 


-13.7544 


2,666 (3) 


1,146(3) 


Tulaga Seamount (East Bank) 


American Samoa 


1,2 


-170.0267 


-14.5125 


78(5) 


1,313(3) 


Vailulu'u Seamount 


American Samoa 


1,2 


-169.0577 


-14.2160 


583 (5) 


2,700 (3) 


Unnamed Seamount 1 


American Samoa 


1 


-167.5555 


-14.7115 


3,722 (3) 


1,304(3) 


Unnamed Seamount 2 


American Samoa 


2 


-171.8754 


-10.6409 


3,342 (3) 


566 (3) 


Unnamed Seamount 3 


American Samoa 


2 


-171.5049 


-10.5500 


4,028 (3) 


749 (3) 


Unnamed Seamount 4 


American Samoa 


2 


-171.2254 


-10.4921 


3,861 (3) 


798 (3) 


Unnamed Seamount 5 


American Samoa 


2 


-170.4845 


-11.1755 


3,674 (3) 


1,180(3) 


Unnamed Seamount 6 


American Samoa 


2 


-170.3508 


-11.9043 


2,957 (3) 


2,012(3) 


Unnamed Seamount 7 


American Samoa 


2 


-170.3072 


-11.5180 


4,068 (3) 


940 (3) 


Unnamed Seamount 8 


American Samoa 


2 


-169.8827 


-11.1421 


3,037 (3) 


1,990(3) 


Unnamed Seamount 9 


American Samoa 


2 


-169.8170 


-11.5138 


3,552 (3) 


1,384(3) 


Unnamed Seamount 10 


American Samoa 


2 


-169.5490 


-10.3254 


3,237 (3) 


1,679(3) 


Unnamed Seamount 11 


American Samoa 


2 


-168.8920 


-11.2256 


3,968 (3) 


943 (3) 


Unnamed Seamount 12 


American Samoa 


2 


-168.4578 


-16.6080 


2,985 (4) 


2,086 (4) 


Unnamed Seamount 13 


American Samoa 


2 


-168.4156 


-16.3762 


4,420 (4) 


582 (4) 


Unnamed Seamount 14 


American Samoa 


2 


-168.3594 


-11.9369 


4,997 (3) 


238 (3) 


Unnamed Seamount 15 


American Samoa 


2 


-168.3415 


-12.2581 


5,091 (3) 


167(3) 


Unnamed Seamount 16 


American Samoa 


2 


-168.3411 


-16.9409 


2,690 (4) 


2,192(4) 


Unnamed Seamount 17 


American Samoa 


2 


-168.1575 


-16.8079 


3,146(4) 


1,523(4) 


Unnamed Seamount 18 


American Samoa 


2 


-168.1088 


-14.1747 


4,494 (3) 


499 (3) 


Unnamed Seamount 19 


American Samoa 


2 


-168.0070 


-16.2408 


2,438 (4) 


2,125(4) 


Unnamed Seamount 20 


American Samoa 


2 


-167.8859 


-15.9792 


3,480 (3) 


1,254(3) 


Unnamed Seamount 21 


American Samoa 


2 


-167.2823 


-12.6179 


3,162(3) 


1,756(3) 


Unnamed Seamount 23 


American Samoa 


2 


-167.2741 


-15.7566 


3,916(3) 


1,587(3) 


Unnamed Seamount 24 


American Samoa 


2 


-166.8266 


-15.6801 


4,444 (4) 


768 (4) 


Unnamed Seamount 25 


American Samoa 


2 


-166.1574 


-16.2741 


2,064 (4) 


3,025 (4) 


Unnamed Seamount 26 


American Samoa 


2 


-165.4911 


-15.1572 


4,666 (4) 


705 (4) 


Unnamed Seamount 29 


American Samoa 


6 


-167.7583 


-15.8242 


4,243 (3) 


758 (3) 


Unnamed Seamount 33 


American Samoa 


6 


-172.2244 


-11.8078 


4,092 (3) 


619(3) 


Unnamed Seamount 34 


American Samoa 


6 


-169.7597 


-10.3760 


3,572 (3) 


1,308(3) 


Unnamed Seamount 35 


American Samoa 


6 


-167.2844 


-13.0375 


4,198(3) 


647 (3) 


Unnamed Seamount 36 


American Samoa 


6 


-167.2790 


-13.3913 


4,363 (3) 


660 (3) 


Unnamed Seamount 37 


American Samoa 


6 


-170.1634 


-14.3718 


353 (3) 


1,206(3) 


Unnamed Seamount 44 


American Samoa 


6 


-170.0859 


-11.7090 


4,250 (3) 


650 (3) 


Unnamed Seamount 45 


American Samoa 


6 


-166.9862 


-14.9812 


3,711 (3) 


1,223(3) 


Unnamed Seamount 46 


American Samoa 


6 


-167.0048 


-14.8347 


4,704 (3) 


282 (3) 


Unnamed Seamount 47 


American Samoa 


6 


-167.3141 


-15.1643 


4,799 (3) 


234 (3) 


Unnamed Seamount 48 


American Samoa 


6 


-167.7249 


-15.1334 


4,858 (3) 


290 (3) 


Unnamed Seamount 49 


American Samoa 


6 


-167.8125 


-15.4669 


4,915(3) 


353 (3) 




Table A.1 cont. Locations and morphological characteristics of seamounts within the EEZ of American Samoa and Samoa. 1= Koppers 
et al. 2010a; 2= Wessel et al. 2010; 3= estimates based on bathymetry from Koppers et al. (2010a); 4= estimates based on bathymetry 
from Smith and Sandwell (1997); 5=estimates based on bathymetry from NOAA PIBMC; 6= seamounts identified from visual inspection 
of the bathymetry. Names are based on seamount databases with local names provided in parenthesis where known. 



Name 






Longitude Latitude 


Depth of top (m) 1 3 4 Height (m) 1 34 


Unnamed Seamount 50 


American Samoa 


6 


-168.1543 


-15.6757 


4,887 (3) 


269 (3) 


Unnamed Seamount 51 


American Samoa 


6 


-168.4162 


-14.5275 


4,104(3) 


730 (3) 


Unnamed Seamount 53 


American Samoa 


6 


-168.9760 


-11.1330 


4,608 (3) 


378 (3) 


Tisa Seamount 


American Samoa 
/Samoa 


1,2 


-171.2235 


-14.4073 


861 (3) 


2,140(3) 


Agavale Seamount 


Samoa 


1,2 


-172.4833 


-13.2333 


995 (3) 


1,986(3) 


Pasco Seamount 


Samoa 


1,2 


-174.4157 


-13.0865 


94(4) 


3,051 (3) 


Si'usi'u Seamount 


Samoa 


1,2 


-173.6039 


-13.2414 


1,269(3) 


2,359 (3) 


Taumatau Seamount 


Samoa 


1,2 


-172.2503 


-13.2894 


842 (3) 


1,890(3) 


Toafe'ai Seamount 


Samoa 


1,2 


-173.9573 


-12.2925 


488 (3) 


2,976 (3) 


Toafilemu Seamount 


Samoa 


1,2 


-174.3238 


-12.5384 


30(4) 


2,737 (3) 


Tuapi'o Seamount 


Samoa 


1,2 


-173.1259 


-13.2477 


425 (3) 


2,966 (3) 


Uo Mamae Seamount 


Samoa 


1 


-172.2427 


-14.9441 


645 (3) 


3,169(3) 


Unnamed Seamount 27 


Samoa 


1 


-170.7861 


-13.0671 


3,796 (3) 


1,004(3) 


Unnamed Seamount 30 


Samoa 


6 


-173.7718 


-12.8724 


3,032 (3) 


601 (3) 


Unnamed Seamount 38 


Samoa 


6 


-173.8263 


-12.9707 


2,349 (3) 


1,379(3) 


Unnamed Seamount 39 


Samoa 


6 


-174.2913 


-12.2940 


1,711 (3) 


1,250(3) 


Unnamed Seamount 40 


Samoa 


6 


-172.4459 


-13.0042 


3,098 (3) 


698 (3) 


Unnamed Seamount 41 


Samoa 


6 


-171.4520 


-13.2108 


3,451 (3) 


1,326(3) 


Unnamed Seamount 42 


Samoa 


6 


-170.8492 


-13.4912 


3,711 (3) 


941 (3) 


Unnamed Seamount 43 


Samoa 


6 


-171.1356 


-13.9646 


2,430 (3) 


1,297(3) 



Peak depth ranged from 23 m to deeper than 5,000 m. The frequency of peak height by depth exhibited a 
bi-modal pattern, with the majority of seamount peaks located in water over >2000 m in depth (Figure A.2a). 
The majority of seamounts (60%) were less than 1,500 m in height (Figure A.2b). 

Only Vailulu'u seamount (Figure A.3) is characterized as hydrothermally active, whereas the remaining sea- 
mounts are extinct volcanoes (Koppers et al. 2010b). The biological community varies among different loca- 
tions on Vailulu'u and includes polychaetes, crinoids, octocorals, sponges, and a population of cutthroat eels 
(Staudigel et al. 2006). 



In general, there is a lack of information on biological communities for other seamounts within the EEZ of 
Samoa and American Samoa. Seamounts Online (Stocks 2009) is a global database of user-contributed 
data on species distributions on seamounts. However, there was no data available for seamounts on the 
Samoan Archipelago at this time. The estimated depths of the top of the seamount features were used to 
determine whether shallow or mesophotic reefs were potentially present. Generally, mesophotic reefs range 
from -30 to 150 m depth, although deeper records of zoxanthellate corals and coralline algae have been 
documented (Hinderstein et al. 2010). The estimated peaks of the seamount features were only shallower 
than 150 m depth for 5 out of 65 features, suggesting that mesophotic reefs are potentially present. Three of 
these seamounts are located within the American Samoa EEZ (Figures A.4-A.6) while the latter two features 
are located on the western edge of the Samoa EEZ (Figures A.7-A.8). Paputua Guyot, locally known as 
South Bank, has been identified as a drowned atoll in recent bathymetric surveys although development of 
mesophotic reef communities is lacking (R. Brainard, personal communication, NOAA Coral Reef Ecosystem 
Division, Honolulu, HI). All the remaining seamount features are estimated to be greater than 300 m deep, 
although actual depths should be interpreted with caution due to the scale and estimation methods of the 
bathymetry data. For example, there was a large difference in the depth of Tulaga Seamount, locally known 
as East Bank, when measured by multibeam sonar (78 m, PIBHMC) versus satellite altimetry (> 700 m). 



CONCLUSIONS 

A wide range of seamount morphologies exist within the Samoan EEZs. Approximately one third of the sea- 
mounts in the study area are derived from the Samoan hotspot which is presently located at Vailulu'u with 
the rest scattered in two main groups in the American Samoa EEZ. The five seamount features with potential 
mesophotic reef communities were evaluated further and used as inputs for understanding reef connectivity 
in Chapter 3 of this assessment. 

ACKNOWLEDGEMENTS 

The seamount databases available online and cited in this document are an excellent resource. We gratefully 
acknowledge Dr. Anthony Koppers for allowing us permission to use the original seamount maps available 
at EarthRef.org in our report. 

REFERENCES 

Hart, S.R., M. Coetzee, R.K. Workman, J. Blusztajn, K.T.M. Johnson, J.M. Sinton, B. Steinberger, and J.W. 
Hawkins. 2004. Genesis of the Western Samoa seamount province: age, geochemical fingerprint and tecton- 
ics. Earth and Planetary Science Letters 227: 37-56. 

Hinderstein, L.M., J.C.A. Marr, F.A. Martinez, M.J. Dowgiallo, K.A. Puglise, R.L. Pyle, D.G. Zawada, and R. 
Appeldoorn. 2010. Theme section on "Mesophotic Coral Systems: Characterization, Ecology, and Manage- 
ment." Coral Reefs 29:247-251 . 

Koppers, A.A.P, H. Staudigel and R. Minnett. 2010a. Seamount catalog: Seamount morphology, maps, and 
data files. Oceanography 23(1): 37. 

Koppers, A.A.P, H. Staudigel, S.R. Hart, C. Young, and J.G. Konter. 2010b. Vailulu'u seamount. Oceanog- 
raphy 23(1): 164-165. 

Rogers, AD. 1994. The Biology of Seamounts. Advances in Marine Biology 30:305-350. 

Smith, W.H.F. and D.T. Sandwell, 1997. Global seafloor topography from satellite altimetry and ship depth 
soundings, Science 277: 1957-1962. Online: http://topex.ucsd.edu/index.html. 

Staudigel, H., S.R. Hart, A. Pile, E.T Baker, S. Brooke, D.P Connelly, L. Haucke, C.R. German, I. Hudson, D. 
Jones, A.A.P. Koppers, J. Konter, R.Lee, T.W. Pietche, B.M. Tebo, A.S. Templeton, R. Zierenberg and CM. 
Young. 2006. Vailulu'u Seamount, Samoa: Life and death of an active submarine volcano. Proceedings of 
the National Academy of Sciences 103(17): 6448-6453. 

Staudigel, H., A.A.P. Koppers, J.W. Lavelle, T.P. Pitcher, and T.M. Shank. 2010. Defining the word "sea- 
mount." Oceanography 12(1): 20-21. 

Stocks, K. 2009. SeamountsOnline: an online information system for seamount biology. Version 2009-1. 
World Wide Web electronic publication, http://seamounts.sdsc.edu. 

Wessel, P. 2001. Global distribution of seamounts inferred from gridded Geosat/ERS-1 altimetry. Journal of 
geophysical research 106(B9): 19431-19441. 



Wessel, P., D.T. Sandwell and S. Kim. 2010. The global seamount census. Oceanography 23(1): 24-33. 



Appendix B: Shoreline to Shelf Edge Benthic Maps of Tutuila, 

American Samoa 

Matthew S. Kendall 1 

INTRODUCTION 

Accurate maps of coral reef ecosystems are a critical component of reef science and management. Me- 
sophotic reefs around Tutuila, American Samoa makeup the majority of the area of coral reef ecosystems 
around the island but have not been comprehensively mapped and classified like their shallow water coun- 
terparts (see NOAA NCCOS 2005). To meet this need we created benthic maps of the mesophotic reef eco- 
systems and edge matched them to the existing shallow water maps to produce a comprehensive shoreline 
to shelf edge map of benthic features. 

METHODS 

Benthic features were visually interpreted from sonar imagery collected by the Pacific Islands Benthic Habi- 
tat Mapping Center (PIBHMC) (downloaded from http://www.soest.hawaii.edu/pibhmc/). Primary datasets 
used during map production were bathymetry and backscatter. Additional image datasets derived from these 
primary sources were also used during interpretation and included slope, rugosity, contours, and hillshade. 

Areas of consistent tone and texture in the sonar imagery were identified visually by toggling among the vari- 
ous sonar datalayers. Boundaries were drawn around these contiguous sonar signatures using the Habitat 
Digitizer Extension to ArcGIS 9.0 (http://ccma.nos.noaa.gov/products/biogeography/digitizer/welcome.html). 
All features were delineated at a scale of 1:10,000. The minimum mapping unit (MMU- size of the smallest 
feature to be delineated) was restricted to 4,000 m 2 to be consistent with benthic maps recently produced for 
shallow water areas (NOAA NCCOS 2005). 

Benthic features with consistent sonar signatures were attributed based on a classification scheme modi- 
fied from the recently completed "Benthic Habitats of American Samoa, Guam, and CNMI" (NOAA NCCOS 
2005). The scheme was originally designed for use with color satellite imagery. The spatial properties of 
the satellite and sonar imagery (i.e. 5 m bathymetry and 1 m back scatter grid resolution for sonar imagery 
versus 4 m color and 1 m black and white grid resolution for IKONOS satellite imagery) and scales of map- 
ping were similar (i.e. on-screen digitizing scale was 1:1 OK for sonar imagery and 1:6K for satellite imagery, 
MMU was 4,000 m 2 for both data sources). Sonar signatures were primarily ground truthed using video 
transect data from a towed camera system which was supplemented with drop camera video on specific 
features. Video transect data was collected between 2002 and 2006 and obtained from PIBHMC (Bare et al. 
2010). Supplemental drop camera data was collected in May 2010 and consisted of 119 sites on features in 
between the video transects. Each site was characterized by ~2 minutes of video for a total of ~4 hours of 
seafloor imagery. 

A key goal was to create comprehensive maps of the coral reef ecosystem of Tutuila from the shoreline to 
the shelf edge. The extent of the sonar data was approximately from the insular shelf edge to the base of 
the fringing reefs around Tutuila, however, in places it did not cover all the way to the shelf edge, include the 
base of fringing reefs (lower fore reef), or necessarily include 100% coverage throughout the shelf. Gaps 
between sonar swaths on the shelf were simply ignored during digitizing since they were typically narrow and 
occurred in regions of relatively homogenous seafloor. To fill in gaps in coverage at the shelf edge, the 100 
fathom isobath from NOAA Navigational Chart #83484 was overlayed in the GIS and used to assist in digi- 
tizing placement of the shelf edge polygon. The gap in coverage between the sonar data and the shoreline 
was filled using maps from the 2005 Benthic Habitats of American Samoa, Guam and CNMI data CD (NOAA 
NCCOS 2005). The satellite base maps (NOAA NCCOS 2005) were edge matched to sonar based maps 
principally along the seaward edge of the fore reef, a feature often visible in both sonar and satellite imagery. 
The shallow water map was generally clipped out in regions of overlap due to the higher interpretability of the 
sonar imagery of this zone relative to the satellite imagery. 

1 NOAA/NOS/NCCOS/CCMA Biogeography Branch 



Classification Scheme 

The classification scheme defined benthic habitats based on four attributes: 1) shelf zone, 2) general geo- 
morphological structure, 3) detailed structure, and 4) percent hardbottom. Every feature in the benthic habitat 
map was assigned a designation from each level of the scheme. We customized the classification scheme 
to be compatible with, 1) the available sonar data for American Samoa, and 2) the existing benthic maps of 
shallow reefs for American Samoa (NOAA NCCOS 2005). The primary differences between this scheme and 
the one used to map shallow reefs of Tutuila were that biological cover was not mapped whereas percent 
hardbottom was. These changes were necessary to the scheme due to the differences in sonar and satellite 
imagery. 

Zones 

Thirteen mutually exclusive zones were identified from land to deep ocean corresponding to typical insu- 
lar shelf and coral reef geomorphology. These zones included: Land, Shoreline Intertidal, Reef Flat, La- 
goon, Back Reef, Reef Crest, Fore Reef, Bank/Shelf, Bank/Shelf Basin, Bank/Shelf Escarpment, Channel, 
Dredged, and Unknown. Figure B.1 illustrates zone types across a typical cross-section of the island shelf. 
Zone refers only to each benthic community's location and does not address substrate or structure types that 
are found within. For example, the Lagoon zone may include patch reefs, sand, and reef rubble; however, 
these are considered structural elements that may or may not occur within the lagoon zone and therefore, 
are not used to define it. Note that some zone categories exist only in the nearshore map (NOAA NCCOS 
2005; e.g. shoreline/intertidal and reef crest) and others only exist in the sonar based portion of the map (e.g. 
bank/shelf escarpment and bank/shelf basin). See NOAA NCCOS (2005) and Bare et al. (201 0) for additional 
cross sectional figures and example photographs of each zone type. 

*. ............ ...... ...... ..*„ ...... .-. ..*.*.-»»*.«. MEWjJlizJiJitkJLiQP 



^.. t ..2 



Shoreline/ 
lute it id nl 



Reef 
FJat 




^rbig Iw'TttfeLmi 




Bank/Shelf 
Basin 



Figure B.1. Cross section of Zones. 



Land 



Terrestrial features at or near the spring high tide line. The shoreline is based on the official digital shoreline 
available at the time nearshore mapping was conducted (NOAA NCCOS 2005). As a result many of the land 
polygons may be smaller than the MMU used to delineate marine features. 



Shoreline Intertidal 

Area between the spring high tide line (or landward edge of emergent vegetation when present) and lowest 
spring tide level. Emergent segments of reefs are excluded from this zone and instead are defined below as 
Reef Crest. Typically, this zone is narrow due to the small tidal range in Tutuila. While present island-wide, the 
feature is often too narrow to be mapped on steep shorelines due to the scale of the imagery and the MMU. 



Lagoon 

Shallow area (relative to the deeper water of the bank/shelf) between the Shoreline Intertidal zone and the 
Back Reef of a reef. This zone is typically protected from the high-energy waves commonly experienced on 
the Bank/Shelf and Reef Crest zones. 

Reef Flat 

Shallow, semi-exposed area of little relief between the Shoreline Intertidal zone and the Reef Crest of a fring- 
ing reef. This broad, flat area often exists just landward of a Reef Crest and may extend to the shoreline or 
drop into a Lagoon. This zone is often somewhat protected from the high-energy waves commonly experi- 
enced on the Bank/Shelf and Reef Crest zones. 

Back Reef 

Area just landward of a Reef Crest that slopes downward towards the seaward edge of a Lagoon floor or 

Bank/Shelf. This zone is present only when a Reef Crest exists. 

Reef Crest 

The uppermost, and often flattened, emergent (especially during low tides) or nearly emergent segment of 
a reef. This high wave energy zone lies between the Fore Reef and Back Reef or Reef Flat zones. Breaking 
waves are often visible in aerial or satellite imagery at the seaward edge of this zone. 

Fore Reef 

Area along the seaward edge of the Reef Crest that slopes into deeper water to the landward edge of the 
Bank/Shelf platform. This feature is often referred to locally as the reef slope. Features not associated with 
an emergent Reef Crest but still having a seaward-facing slope that is significantly greater than the slope of 
the Bank/Shelf are also designated as Fore Reef. 

Bank/Shelf 

Deeper water area (relative to the shallow water in a lagoon or reef flat) extending offshore from the seaward 
edge of the Fore Reef or shoreline to the beginning of the escarpment where the insular shelf drops off into 
deep, oceanic water. If no Reef Crest is present, the Bank/Shelf is the flattened platform between the Fore 
Reef and deep open ocean waters or between the Shoreline Intertidal zone and open ocean. 

Bank/Shelf Basin 

Broad depressions of deeper water occurring in the Bank/Shelf. These features are surrounded by well de- 
fined inflections in bathymetry up to the relatively less deep waters of the Bank/Shelf. 

Bank/Shelf Escarpment 

This zone begins on the oceanic edge of the Bank/Shelf, where depth increases rapidly into deep, oceanic 
water and exceeds the depth limit of sonar imagery around Tutuila. This zone is intended to capture the in- 
flection point of the shelf to deep waters of the open ocean. 

Channel 

Naturally occurring channels that often cut across several other zones. 

Dredged 

Area in which natural geomorphology is disrupted or altered by excavation or dredging. 



Pinnacle 

High relief features occurring in or adjacent to Bank/Shelf Basin that are capped by coral reef or hard bottom. 

Unknown 

Zone indistinguishable due to gaps between swaths in sonar imagery or due to turbidity, cloud cover, water 

depth, or other interference in satellite imagery. 



Geomorpholoqical Structures 

Fifteen distinct and non-overlapping geomorphologic structure could be mapped by visual interpretation of 
sonar imagery. Structure refers only to predominate physical composition of the feature and does not ad- 
dress location (see Zone for shore to shelf edge location). The structure types are defined in a collapsible 
hierarchy ranging from four major classes (Coral Reef and Hardbottom, Unconsolidated Substrate, Other 
Delineations, and Unknown), to fifteen detailed classes listed and defined below. See NOAANCCOS (2005) 
for example photographs and satellite images of each classification. 

Coral Reef and Hardbottom 

Solid substrates including bedrock, boulders, and reef building organisms. A thin veneer of sediment may 
be present. Detailed classes within this category include Aggregate Reef, Individual Patch Reef, Aggregated 
Patch Reefs, Spur and Groove, Pavement, Pavement with Sand Channels, Pavement with Patch Reefs, 
Reef Rubble, and Rock/Boulder. 

Aggregate Reef 

Continuous, high-relief coral formation of variable shapes lacking sand channels of Spur and Groove. 
Includes linear coral formations that are oriented parallel to shore or the shelf edge. This class is used 
for such commonly referred to terms as linear reef, fore reef or fringing reef. 

Individual Patch Reef 

Patch reefs are coral formations that are isolated from other coral reef formations by bare sand, sea- 
grass, or other habitats and that have no organized structural axis relative to the contours of the shore 
or shelf edge. They are characterized by an often circular or oblong shape with a vertical relief of one 
meter or more in relation to the surrounding seafloor. Individual Patch Reefs are larger than or equal to 
the MMU. 

Aggregated Patch Reefs 

These features have the same defining characteristics as an Individual Patch Reef. However, this class 
refers to clustered patch reefs that individually are too small (less than the MMU) or are too close to- 
gether to map separately. Where aggregated patch reefs share sand halos, the halo is included in the 
polygon. 

Spur and Groove 

This structure has alternating sand and coral formations that are oriented perpendicular to the shore 
or reef crest. The coral formations (spurs) of this feature typically have a high vertical relief relative to 
pavement with sand channels and are separated from each other by 1-5 meters of sand or hardbottom 
(grooves), although the height and width of these elements may vary considerably. This habitat type 
typically occurs in the Fore Reef or Bank/Shelf Escarpment zone. 



Reef Rubble 

Dead, unstable coral rubble often colonized with turf, filamentous, calcareous, or encrusting macroal- 
gae. This habitat often occurs landward of well developed reef formations in the Reef Crest, Back Reef 
or Reef Flat zones due to storm waves piling up dead coral. Reef Rubble can also occur in offshore 
areas due to diseased or physically impacted reef communities. 

Rock/Boulder 

Large, irregularly shaped carbonate blocks or volcanic rock often extending offshore from the island 
bedrock or headlands. Can also occur as aggregations of loose rock fragments that have been de- 
tached and transported from their native beds. Individual boulders often range in diameter from 
0.25 -3 m. 

Unconsolidated Substrate 

Areas of the seafloor consisting of small, unattached or uncemented particles with less than 10% cover of 
large stable substrate. Detailed structure classes of softbottom include Sand, Mud, Sand with Scattered 
Coral and Rock, and Algal Plain. 

Sand 

Coarse sediment typically found in areas exposed to currents or wave energy. 

Mud 

Fine sediment often associated with stream discharge and build-up of organic material in areas shel- 
tered from high-energy waves and currents. 

Sand with Scattered Coral and Rock 

Primarily sand bottom with scattered rocks or small, isolated coral heads that are too small to be delin- 
eated individually (i.e. smaller than individual patch reefs). If the density of small coral heads is greater 
than 10% of the entire polygon, this structure type is described as Aggregated Patch Reefs. 

Algal Pain 

Low relief (<~0.25 m) substrate composed of a mixture of sand, live halimeda, halimeda sand, fleshy 
macroalgae, and rhodoliths. Relative abundance of these compositional elements is highly variable 
over scales of a few meters. 

Other Delineations 

Any other type of structure not classified as Coral Reef and Hardbottom or Unconsolidated Sediment. Usu- 
ally related to the terrestrial environment and/or anthropogenic activity. Detailed structure classes include 
Land, Artificial, and Emergent Vegetation. 



Pavement 

Flat, low-relief, solid rock in broad areas often with partial coverage of sand, algae, hard coral, gorgoni- 

ans, zooanthids or other sessile invertebrates. 

Pavement with Sand Channels 

Areas of pavement with alternating sand/surge channel formations that are oriented perpendicular to 
the shore or Bank/Shelf Escarpment. The sand/surge channels of this feature have low vertical relief 
relative to Spur and Groove formations. This habitat type occurs in areas exposed to moderate wave 
surge such as the Bank/Shelf zone. 

Pavement with Patch Reefs 

Habitats of pavement with occasional patch reef formations that make up less than 10% of the area of 

the polygon. This habitat type occurs nested in pavement areas on the Bank/Shelf zone. 



Land 

Terrestrial features at or near the spring high tide line. 

Artificial 

Man-made habitats such as submerged wrecks, large piers, submerged portions of rip-rap jetties, and 

the shoreline of islands created from dredge spoil. 

Emergent Vegetation 

This category includes all species of mangroves regardless of canopy density. This class was not used 
the sonar derived map for obvious reasons but was retained in the classification scheme due to its oc- 
currence in nearshore maps. 




Unknown 

Major structure indistinguishable due to data gaps from turbidity, cloud cover, water depth, ship orientation, 

line spacing, or other interference with an interpretable signature of the seafloor. 

Unknown 

Detailed structure indistinguishable as above. 

Percent Hardbottom 

Seven classes were used to denote the approximate proportion of each polygon occupied by hard bottom 
substrate. A polygon encompassing several patch reefs that were too small to be delineated individually is 
actually comprised of some area of patch reefs and some background structure such as sand. This category 
includes both "living" hard bottom such as patch reefs as well as "abiotic" features such as pavement and 
rock/boulder. This attribute can be used to estimate the combined amount of coral reef and hard bottom 
around the island based on the area of each polygon. 

<10% 

Used for all sand, mud, and sand with scattered coral and rock polygons. 



ACKNOWLEDGEMENTS 

Sonar data and towed camera imagery are essential tools for mapping coral reef ecosystems at this depth 
and are the basis of this mapping project in deeper waters. These data were collected and made available for 
this project by staff at the Pacific Islands Benthic Habitat Mapping Center and NOAA's NMFS/PIFSC/Coral 
Reef Ecosystem Division. We are particularly grateful to John Rooney, Alica Bare, and Kerry Grimshaw for 
obtaining and navigating those datasets. Laurie Bauer, Kim Roberson, Kevin Grant, Chris Caldow, and Ken 
Buja were intrumental in collecting additional ground validation data. Ken Buja provided help with the habitat 
digitizer and with sewing the shallow and deep water maps together. 



10-30% 

Used for some aggregated patch reef polygons and other discontinuous features. 

30-50% 

Used for some aggregated patch reef polygons and other discontinuous features. 

50-70% 

Used for some aggregated patch reef polygons and other discontinuous features. 

70-90% 

Used for some aggregated patch reef polygons, pavement, and other discontinuous features. 



REFERENCES 

Bare AY, Grimshaw KL, Rooney J, Sabater MG, Fenner D, Carroll B. 2010. Mesophotic communities of the 
insular shelf at Tutuila, American Samoa. Coral Reefs 29:369-377. 

NOAA NCCOS (National Centers for Coastal Ocean Science). 2005. Shallow-water benthic habitats of Amer- 
ican Samoa, Guam, and the Commonwealth of the Northern Mariana Islands (CD-ROM). NOAA Technical 
Memorandum NOS-NCCOS 8, Biogeography Team. Silver Spring, MD. 



90-100% 

Used for most individual patch reef, pavement, aggregate reef polygons, and other continuous features. 

10-90% 

This broad category was used for algal plain polygons due to the high variability in Rhodolith coverage 
of this bottom type which was not interpretable in the sonar imagery. Rhodoliths are hard algal nodules 
with -5-10 cm diameter that are not cemented to the seafloor. They can form highly variable coverage 
from sparse (a few per square meter) to a nearly continuous cover over sand substrates. 

RESULTS 

The map product from this work is available for download from the NOAA/NOS/NCCOS/CCMA/ Biogeogra- 
phy Branch website at http://ccma.nos.noaa.gov/about/biogeography/. 




Appendix C: Fish and Coral Data Plots 

Figures C.1-31. Bar graphs depicting the distribution of percent coral cover (C.1-8), coral richness (C.9-15), fish biomass 
(C. 16-23), and fish richness (C. 24-31) for each of the studies. Dashed red lines indicate the identified natural breaks in the 
data that were used to classify survey site values as high, medium, and low for each variable. 



Figured -ASEPA 




o 





Q) 

TO 

05 
03 



=3 

a 

03 

h- 



03 
O 
< 



03 
O) 
03 




03 
> 



CD 
O 



03 
> 



03 
03 



03 

=3 

'CO 
O) 
03 



3 


=3 


=3 


=3 


I 03 


=3 


03 


03 








1 o 


03 


CD 


03 


| < 


^ 


CO 

03 


03 
LL 



03 
O) 


< 



CD 

=3 

M— 

03 
D) 
03 
LL 



Figure C.2 - CRSR 




> 
o 
o 

"i5 

i- 

o 
o 



100 
90 
80 
70 
60 
50 
40 
30 
20 
10 






high 




medium 


low 






























i 
















i 






























IIiiihh. 










1 




Swains_SW1 
Swains_SW2 


■ cd + 

§ i 

i 1 
u 


Nu'uli 

Vatia 

Olosega Village 

Masefau 

Ta'u Village 

Amanave 

Utulei 

Ofu Village 


Faga'alu 

Fagasa 

Fagafue 

Fagaitua 

Fagamalo Cove 

Aunu'u 

Aoa 

Fagatele 

Fagamalo 

Afuli 

Rose_NW2 

Onesosopo 

Leioaloa 

Rose_SW3 

Amouli 

Rose_SW2 

Aua 

Rose_NW1 

Rose_SW1 




Figure C.3 - GCRMN 



Figure C.5 - MPABR 



100 
90 
80 
70 

r 6 ° 

0) 

> 

O 50 

o 



05 
O 

o 



40 
30 
20 
10 



high 


medium 


low 


j — 






m 






**** 




Fagamalo 

Papa Puleia 

Vaisala 


Samatau 

>iufaga Faga 

Safaatoa 

Saleapaga 


Palolo Deep 



medium 



100 



0) 

> 
o 
o 

75 

o 
o 




Figure C.4 - KRS 



Figure C.6 - REA 




100 



90 



80 





70 


^_^ 




ss 






60 


3_ 




0) 




> 




o 


50 


o 




05 




3_ 

o 


40 


o 






30 




> 
o 
O 



"o. 

CD 



>^ 


>^ 


^ 


>^ 


>^ 


CD 


CD 


CD 


CD 


CD 


CQ 


CQ 


CQ 


CQ 


CQ 





3 


CD 


CD 


O 


> 


CD 


O 


D) 


c 


CD 

c 

CD 


CD 


O 


_CD 


£ 


E 


Q_ 





< 


E 


< 





CO 




< 






1 
CD 





CD 
CD 
CO 



CD 
CQ 

CD 
D) 




CD 
CQ 

CD 
O 
O 
< 



_CD 
O 



Q_ 


O 





1_ 





CD 


=3 




CD 


CD 


CD 

3 


3 
< 



CD 
CQ 

CD 
-^ 
'CD 
Q) 
CD 



CD 

o 
O 



CD 
CQ 

=3 

CD 
CO 
CD 



20 



10 



high 



medium 



low 




COOTI^OU^r^^U^^^CMCMCpCOCO 

csi c\i csi 

< < =) 



■NMflO)^^S^-lO<DO)' 



roc\io)ooo)^-inT-o)CDoos^-ocosNooco 

._ ._..._ _ . ._ . . " -■ t- O 00 o o 

CM CO CM CM CM 



T-T-CNN^-tOWtDCNpOC\irOO)OCMOO^CN^COC\ICMroSttOCDCNOOT-CMCOCN|T-T-CNCN 

cncncncncnco^cococnRJcncncncncncn^co 

ww'-i-ww | -oi-^i-i-oooQ:i-i-^i-i-i-OQ:i-^oi-i-i-i-i-^i-i-i-i-a:i-i-i-i- 
* due to the large number of sites, not all site labels are visible on this axis 



CM < 



■ LO ' 



D h" h~ < CO 

< =) =) ^ O 
I- I" I" % 01 



O O CO 

CM CM CM CM CM ^ 

Z> => _i Z> o < 

I- i- o i- tz \- 



CN N N CO t- ^> 
O CO CM t- O CD 
CM ^ CM CO CM CM 

WDWhDQh 
0<0303 

qi \- a: i- co i- 



CD <vf 
CM CM 

CO CO 





Figure C.7 - SFR 



100 



90 



Q_ 






Q_ 




8U 


< 


70 




^~^ 






£ 






j- 


60 











> 






O 


50 




o 








40 




o 






o 


30 



20 



10 

















high 


































medium 














low 








































































































































































































































i 
































ii.. 








































































nun 






1 ■ ■ ■ ' 


Satoalepa 

Lefagaoali 

Savaia 

Safa 

Manase 

Vaova 

Fagasa 

Tafitoala 

Vailoa 

Sasina 

Vavau 

Luua 

Sapin 

Sataua 

Matautu Falealil 

Fatuvalu 

Malae 

Salimu 

Papa Palaul 

Saasaa 

Tauaoc 

Auala 

Saleilua 

Aufaga 

Matareva 

Sapunaoa 

Faleu 

Tafagamanu 

Fasitoota 

Foailalo 

Satalc 

Vaiala 

Tafatafa 

Salua Ta 

Neiafu Ta 

Lalov 

Poutas 

Fogatul 

Siufaga Falelata 

Apolima uta 

Faleapuna 

Lepuiai Ta 

Matatufu 

Afega 

Lepa 

Saoluafata 

Matautu Falelata 

Salesatele 

Apa 

Fagali 

Fuailoloc 

Salan 

Sapapali 



Figure C.9-ASEPA 

high 
16 



o 

o 

c 



o 
a. 

co 

s_ 

CD 
C 
0) 
O) 

</> 

o 

c 
-C 
o 

£ 

2 
o 



14 



12 



10 




CO 


3 


:= 


CO 


3 


& 


^ 


D) 


CO 


CO 

I- 


CO 

_l 


CD 
< 


CO 









_CD 

3 

03 
O) 
CO 



CD 


o 


3 


3 


3 


CO 


CO 


£= 


3 


CO 


CO 


o 


O) 


o 

CD 

_l 


CO 


CD 


o 


< 


CO 


CO 

> 


^ 


CO 
CO 


< 




CD 
CO 








^ 






CO 

3 



CD ^ 

=3 CO 

.ff ,7 s 



CO CO 

CO -^3 

CO CO 

a* > 

CO 



Figure C.8-TCRMP 



Figure C.10-CRSR 



0) 

> 
o 
o 

1 

o 




low 





o 




0) 




0) 




c 




(0 


(/) 


s_ 


</) 


+j 


<D 


0) 


o 


(0 


t* 


O 




O) 


rc 


O 


l. 




o 


O 


o 






s_ 




O 



1 1 1 ■ ■ 



=tfc 



CO 

> 



CO 
CO 



1 ■ ■ 


3 


CO 


>, 


CO 
CO 


CO 


CO 
00 


O) 


E 


CD 


CO 
LL 


< 


CJ 
CO 
CO 
CO 
CO 







high 












medium 


















low 














































































































































































































































































CO 

o 

< 


Amanave 

Fagaitua 

Fagatele 

Leioaloa 

Faga'alu 

Fagamalo 

Vatia 

Aunu'u 

Leone 

Masefau 


Lepula 

Faga 

Fagafue 

Hurricane House 

Fagasa 

Afuli 

Fagamalo Cove 

Nu'uuli 

Aua 


Ofu Village 

Asaga 

Fatumafuti 

Utulei 

Onesosopo 

Sili 

Olosega Village 





Figure C.11 - GCRMN 



Figured 3 - REA 



Q_ 


^^ 




+-» 


Q_ 


O 

(1) 


^^fl 


0) 


*** 


c 




(0 




(/) s- 




(0 ** 




CD s- 




O (0 




S .2 




— O) 




(0 o 




t_ -— 




o o 




U£ 




J- 




o 




E 



* 




o 

0) 

c 
(0 



30 



25 



20 



high 



0) 

a 

2 
o 

c 

CD 

=tfc 



0) 

(0 

CD 

c 

o 

5 

75 

o 
o 



10 



03 


ZJ 


o 


03 


03 


03 


03 


Q. 


03 
CO 
"03 


-2 

03 

E 


03 

E 

03 


O 

"co 


CO 

03 

a. 

03 


CO 
03 
LL 




13 

Q_ 


CD 
CD 

a 


> 


03 


CO 


03 


CD 


03 


03 
Q_ 


o 




CO 


03 
LL 


CO 


03 


CO 
03 


o 








CO 


4— 
_ZJ 


03 

Q_ 


03 
Q_ 



CO 



medium 



^- co oo ^- oo "<t 
cp cp o cp cp cp 

zjzjzjzjoi^zjozjzjzjzjozjkzjozjz):!)!- 1 !- 1 

u_ < < u i z> u i < < u_ u i < p u i < < < z) z) 

oi-i-ooi-ooi-i-oooi-i-ooi-i-i-i-i- 



low 



t-NincgcONCNOlOO) 
OOOOt-t-OOt-O 



CO I— I— 
O Z> Z> 

tx \- \- 



h W h 

Z) O 3 - ■ v -ww-----w VV -www- > ^ -> =» -> 3 => > 



Z> CO I- CO CO 
U-OZ>OOPPPPP 



CO CO 

o o 



O) O CO CD N 
O CM O O O 



C0l-C0C0C0<<<<<<<< 



Figured 2 - MPABR 



Figure C.14-SFR 



70 



60 



0) 

o 
o 

£> 

0) 
0) 
0) 

c 
o 
5 

I 20 

o 

10 



50 



40 



30 




10 





o 


8 




0) 






(/> 






c 


7 




(0 




(/) 


i_ 




0) 


+J 




0) 




6 


o 


</) 




a: 





5 




5) 




(0 


o 




j_ 




/\ 


O 


o 




o 








j_ 


S 




o 






E 






=jt 


2 



-i— < 

CO 


-i— < 
CO 


CO 


CO 


03 


.cz 


-1— > 
CO 


ZJ 
03 


CO 


CO 


CO 


CO 


to 


to 


ZJ 
03 


-i— • 
CO 


to 


CO 


^ 


to 


to 


-1— ' 
CO 


CO 


co 


03 


03 





03 


03 


f— : 


CD 


03 


03 


03 


03 


CD 














03 








CO 


CO 


CO 


f— : 


LU 


§ 


LU 


LU 


CZ 
CD 

o 

03 

E 

CD 
03 

H 


o 

z 


5 


O 


LU 


CD 


LU 


LU 


§ 


§ 


03 
E 
< 


5 


§ 


5 


DO 


5 


§ 


LU 


LU 


LU 


£Z 


o 

03 

ZJ 
£Z 

CO 


CO 

cz 

CD 
CO 

03 

_l 


ZJ 

~b 
c 

=5 
< 


03 
DQ 
CD 


"co 


CD 

> 
03 
£Z 
03 

E 

< 


ZJ 

~zj 
c 

ZJ 

< 


CD 

cz 
o 

CD 

_l 


<L 


03 

ZJ 

c 

03 

Z 


r 
o 

z 

03 
O 
O 


03 

ZJ 

A-> 

'03 
CO 
03 

LL 


CD 

£Z 
O 
CD 

_l 


03 

E 

CD 
03 

h- 




> 
03 
C 
03 

E 
< 


03 

ZJ 
£Z 

03 

z 


03 
DQ 


03 


03 

ZJ 

-1—' 

'co 

CO 
CO 
LL 




o 

CO 
CO 
CO 
CO 


CO 

ZJ 
03 

E 

< 

1 

o 


CO 

£Z 



CO 
CO 

_l 


CO 

'*— • 
03 

> 


03 

E 


CO 

h- 


03 

ZJ 
03 

E 

< 
6 






CO 
03 
LL 














Q_ 












CO 
03 
LL 




ZJ 
CO 
CO 


ZJ 

< 








ZJ 

< 


z 



_b 



03 

> 



high 


medium \ low 


























j_ 


i 














-1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1 




1 



IS 'co 8 "co c ^ =!=5 = .2 IS « i= "co 15 




m S 


'en 


O'CO 


'co 


'cc 


cc 


IS •"" 


CO 


h- 


c co 

as 

co ^ 


CD 
CC 


CC 3 


o 
n 


CC 

'3 


UL 


<D 




Q. 


CO 


CC 


z 




CD 



IS'-^ "ro IS 

^iCC I— 3 
CC ■ -3 CC CC 

3| -S 3 E 
cc" - co o 

CO" °- 

! < 

I 



3 CC cc cc ■ 

■S < cc cp . 



! Q. CC ^ CC TO 

1 ^ co = ro ro 

LL CC fO Q. 
CC 



LL CO 3 



CO 



CO 





Figured 5 -TCRMP 

high 



Figure C. 17 - CRSR 



medium 



low 




CO 

C/) 
0) 
CO 

E 
o 

(0 



i^nn 








high 






















med 


iurr 




















low 


I ouu 
mnn 






I uuu 










































ouu 
n - 


























































llllllin 


U n 


Rose_NW2 

Fagaitua 

Swains_SW2 

Fagamalo Cove 

Ofu Village 

Asaga 

Rose_NW1 

Olosega Village 

Rose_SW3 

Leone 

Afuli 

Fatumafuti 

Aua 

Rose_SW2 

Aunu'u 

Sili 

Fagaloa 

Faga 

Rose_SW1 

Amouli 

Nu'uli 

Amanave 

Fagatele 

Fagafue 

Swains_SW1 

Lepula 

Onesosopo 

Rose_SE1 

Vatia 

Utulei 

Fagamalo 

Faga'alu 

Aoa 

Leioaloa 

Masefau 

Ta'u Village 

Fagasa 



Figure C.16-ASEPA 

high 
25000 




< 



^ 


o 

'co 

> 


CD 


CD 


:= 


^ 


CD 


CD 

M— 

CD 
O) 
CD 


CD 


3 


CD 


CD 


O 


O) 


3 


3 


^ 


A-> 


£ 


D) 


CD 

co 

CD 


< 


CD 
< 


CD 

_l 


-i— > 
CD 


'CD 
CD 


CD 

> 


CD 

1- 


03 

"CD 
CO 
CD 

3 


^ 












LL 


LL 









B 

"CD 
O) 
CD 



CD 
CO 
CD 
D) 
CD 



CD 
£Z 
O 
CD 



CD 
CD 
O) 
CD 



Figure C.18-GCRMN 

high 



70 



60 



° 50 
c/> DU 

c 

CO 



O) 40 

C/) 
CO 

I 30 
o 



CO 



CO 



20 



10 



medium 



low 







^i~r 



CD 
Q) 
CD 
LL 

CD 
D) 
CD 



CO 



i ^ 


^ 


O 


CD 






CD 


CD 


i £ 


E 


CD 


CD 


CO 


CO 



CD 


CD 


O 





CD 


CD 


D 


CO 


E 


Q_ 


CD 


CD 


CD 
Q. 
CD 


> 


CD 
LL 


Q_ 







Figured 9 - KRS 

high 




Figure C.21 - REA 

high 



160 



140 



120 



^ 100 



E 



</> 
05 

E 
o 



(/> 



■4— • 


0. 


£ 


0_ 


^ 


CL 


>, 


-i— • 

0_ 


^ 


^ 


^ 





>^ 


^ 


^ 


^ 


^ 


-•— > 


Q_ 


CO 


CO 


CO 


CO 


CO 


> 


CO 


CO 


CO 


CO 


CO 




o 





CD 


zj 


CD 


ZJ 


DO 


CO 


oq 


CD 


CD 


O 


dq 


DQ 


DQ 


DQ 


DQ 


o 


o 





_C 


^ 


CO 


CO 


CO 


CO 


zj 


CO 


CD 


O 


o 


zj 


zj 


CO 


CO 


Q_ 


3 
03 


o 
O 




O) 


o 


ZJ 


CO 


c 


> 





c 


CO 


CO 


o 


D) 


CO 


1 

o 

O) 

o 

_l 





CD 

1 

CO 

ZJ 


LL 


'co 

CO 
LL 


CO 


CD 
t/) 
CO 


CD 
CO 

CO 

_l 


CO 

cz 

CO 

E 
< 


CO 
< 


1 


CO 

E 
< 


o 
o 

< 


o 
Q_ 



< 


CO 
LL 



80 



60 



40 



20 



o 

_Q 

CO 

_£Z 

I 

CO 

ZJ 

< 



medium 



low 




T— 


^1- 


1^ 


CD 


T— 


o 


O) 


CM 


CM 


r- 


^1" 


00 


CD 








CO 


^f 






^f 


CO 








O 


CO 


co 


CM 


CO 


C\| 


CN 


CM 


CN 


CO 


Tf 


CM 


CM 


^1" 


h- 


\- 


< 


1- 


C/) 


H 


h- 


~) 


h- 


") 


~) 


n 


~) 


_) 


-J 


C/) 


_) 


n 


_) 


_> 


<r 


_) 


LL 


<r 


_i 


LL 


h- 


\- 


h- 


a: 


h- 


h- 


h- 


h- 


O 


h- 


o 


O 



N-cocMOLOT-cDT-Locsja^LOLOLOco-ncoco 

^-CMt-^CNOO)CNCNCO(DCO^CDt-CDCOCM 
COCMCMCOCMCMCM^l-COCMCOCMCMCO^l-CMCMCM 



h CO CO h 

d o o 2 

I- 0£ 0£ I- 



CO D H H H 
O < => => => 
Ct H H h- I- 



< I- I- I- I- I- 
> Z) Z) Z) Z) Z) 



bq => co 

i- O a: 



rtNO)CONCNO(DOCN^T-OOONCOCO 
(OCMCNCOOlfi(NOCOO(NO)T-T-iOCOO 
COCMCMCMCMCOCOCMCMCOCO^CMCOCOCMCM 

h<D(ODhhD<hhh<hhhCO 
3>U_0<=>=>LJ->333>3330 

i-^OarHHi-o^HHHgHHHa: 



due to the large number of sites, not all site labels are visible on this axis 



Figure C.20 - MPABR 



Figure C.22 - SFR 



45 



40 



35 



0) 



o 


30 


o 




(/) 




s "- •* 




</) 


?5 


</> 




05 




E 
o 


20 


CD 





(0 



15 



10 





high 






medium 


























low 


































i i 






















I 






















































"I 

^C 
C 

c 

£ 
C 
i 

< 


-> ■+ 

o c 
D c 

D : 
> 

S J 
C 


n i 1 
D . c 

; : B 

3 1 C 
3 j j 

Ml 




3 -: 

1 i 

< 

L 


r, CO 

i ! 

3 O 

3 CO 

Z =5 
3 £Z 

CO 

z 


1 
( 

u 
< 
( 

£ 
( 

1 

< 


n co 

CO 
U LU 
D CO 

I 1 

I i= 

c 


1 

( 
D 

I 
( 


/) j to 
D ! CO 
> . LU 

>, i § 

Q ! co 
D ' g" 

r CO 
^ I LL 

o ■ 
o ■ 


1 

( 
1 

< 
( 
h 


n 

D 

> 

O 

D 
O 


CO 

L_ 
A-> 

£Z 


o 

CO 

E 


CO 

\- 




tn 

CO 
LU 

CO 

=5 
£Z 




CO 
CO 
LU 
CO 

£Z 

CO 

CO 

_l 




CO 
CO 
LU 

> 

CO 
DQ 




"co 
cr 

CO 

n 




to 


CO 

£Z 

CO 

CO 

_l 




to 


g 

CO 

=5 
CO 

E 
< 
o 

<r 




=5 

o 

< 




CO 
CO 
LU 

CO 

=5 
CO 

E 
< 
o 

< 




CO 
CO 
LU 



£Z 
O 


_l 




CO 

CO 



-£Z 

XL 
O 

z 

CO 

o 
o 

Q_ 


1 

< 

£ 
( 
< 


o to 

D 

; § 

D CO 

Z Z5 
D £Z 
D CO 
J CO 

z 





70 



60 



o 
o 

(0 50 

c 

CO 



_g> 40 



0) 
0) 
05 

E 
o 

0) 



30 



20 



10 



high medium 


low 


































T I 




ii 


1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 


1,1, III, III, ■ ■ ■■■■■■ ■,■■-,-,--, , 



§ § = w 

ills 

co — ' CO CO 

co com 



= CO CO 

CO ■- co 

CO > 

-^ CO CO 

■T 3 C7) C7) 



ZJ CO ~ 

co co o 

e 3 a 

CO CD 



"F O CO CO Z3 ■« 

.b -= +-! m r» its . lu 



Saim S H 



CO O CO 



co CO CO 



CO 



CO CO CO 



CO 



iSC/)C/>_Z5^>>_ 

^ o co 

2 CO CO 



= J5 5 'co o co 'co 2 S, i5 =5 'co 'co 

LL H « "CO g" O & 

^ Q_ 



'co 

CO 



_C0 Q. 
CO 

1 % 

^o OT 

."3 



C7) 





CI) 


CO 


> 

o 


CO 

£Z 


'co 
o 


to 


CO 
LL 


CO 

Z5 


CO 
CO 



CO 


CJ) 
CO 
LL 


CO 

_l 


Z5 
Q. 
CO 

CO 


< 



CO 
C7) 


U) 






LL 




ZJ 





ZJ 
CO 





Figure C.23-TCRMP 



Figure C.25 - CRSR 




o 
m 

0) 

a. 

</> 
o 

o 

CD 
Q. 
</> 

</> 
O 

c 
o 



</> 



200 

180 

160 

140 

120 

100 

80 

60 

40 

20 













high 
























medium 


























low 












































































































































































































































































i 






















































































































































































































































































)losega Village 

Aunu'u 

Asaga 

Fagatele 

Sili 

Aua 

agamalo Cove 

Rose_SW1 

Leone 

Rose_NW2 

Rose_SW3 

Utulei 

Ofu Village 

Afuli 

Swains_SW2 

Lepula 

Rose_SW2 

Fatumafuti 

Amouli 

Vatia 

Rose_NW1 

Swains_SW1 

Masefau 

Fagamalo 

Nu'uli 

Aoa 

Amanave 

Onesosopo 

Faga 

Leioaloa 

Fagaitua 

Faga'alu 

Rose_SE1 

Fagasa 

Ta'u Village 

Fagafue 



Figure C.24-ASEPA 

high 



Figure C.26 - GCRMN 






Figure C.27 - KRS 



Figure C.29 - REA 



o 


CO 

c 

CO 





Q. 

CO 


o 

(1) 
a. 

CO 

0) 
0) 

o 

c 
-C 
o 

5 



0) 



60 



50 



40 



30 



20 



10 







nigh 


















mec 


ium 


























low 




































































































































c 

( 

( 

( 
J 
( 


L C 
D : 

? - 

J 

1 

D 
J) 

D 
J 


L CL 
D CO 

> .£ 
o 
O 




CO 
DO 

CO 
O) 
CO 

A-> 

CD 
CD 

CO 

i 

CO 

Z3 




CD 

> 

o 
O 

CD 

'q. 

CO 
< 




CL 
CO 
CO 
CO 




CO 
DQ 
t/) 

"c 

CD 
C/) 

s_ 
CO 

_l 




CL 



£ 

.5 
"co 




CL 

zs 

CO 

o 

LL 




CO 

do 

zs 

CO 

o 
o 

< 




CO 
DQ 

CD 
> 
CO 

cz 

CO 

E 
< 




CO 
DQ 

O 

cz 

< 




CO 
DQ 

Z3 
CO 
CO 

E 
< 




CO 
DQ 
CO 

Z3 
'co 

O) 
CO 
LL 




CO 
DQ 

Z3 

a 


(J) 

CO 




CO 
DQ 
CO 

o 
o 

CL 




cz 
'o 
CL 

CO 
Q) 
CO 
LL 




O 
-Q 

CO 

CO 

Z3 
< 




CO 
DQ 

CO 
U) 
CD 

< 





50 



^ 45 

O 



(0 40 

c 

2 

r 35 


a 

2 30 

O 


Q. 
CO 

S. 

CO 
(O 


c 
o 

(0 



high 



25 



20 



15 



10 



medium 



low 



oo)(DtDOT-tD\r^t\r 

(DCOOt-t-NIOCMIOO 
COCOCNCNCNCOCOCNJCNCO 

h-h-D^h-h-h-^h-h- 
Z)Z)<>Z)Z)Z)>Z>Z> 
HHl-gl-1-l-gi- 



N|C\INCMlOLnO)CflT-oOT-r\|CNCDCDN^CMT-NNOOT- 
^^COCNCNCNCNCVJCNCNCNCNCNCNCNCOCNCO^C^COCOCsl 

HCOH<Z)Z)HHHHHHHHHHH<HHZ) : 
Z)OP^LL<Z)Z)Z)Z)Z)Z)Z)Z)Z>PP^PP 



Z)-CO 
LL 'O 



hhO'l^H^h 



CO 



CO 



CNJ CO N- CO CO t- 
CN CNJ CO CO CNJ CNJ 



I- < \~ \~ 
=> 5> Z> Z> 



CO 

a: co 



< h- 



^oooLOinoocvjO) 

t-COLOCNCOCOOOCOt- 
CNJCNJCNJCNJCNJCNJCNJCNJCN 



DWhDDWHWW 
ll03<<0300 
O Ql \- \- \- CL - ! - 01 01 



due to the large number of sites, not all site labels are visible on this axis 



Figure C.28 - MPABR 



Figure C.30 - SFR 



90 



80 



70 



high 







o 
o 


60 


CO 




s "- •* 




co 

CO 


50 







c 




-C 


40 


o 




tt 





CO 



30 



20 



10 



C/) 


C/) 


C/) 


r 


c/> 


a) 


CO 


CD 


CO 


§ 


LU 


§ 


o 

z 


LU 


CO 
DQ 


CO 
DQ 


CD 
> 
CO 


Z3 
~Z3 


CD 
> 
CO 


CD 


CD 


cz 

CO 


C 
3 


cz 

CO 


CD 


CD 


E 
< 


< 


b 


CO 
CO 


CO 

CO 
1 1 




< 



Z3 
~Z3 

< 



medium 



C/) 

CD 



C/) 

CD 



5 § 



CO 

'co 

CO 



CO 

E 

CD 
CO 



Z3 
_co 

CO 

E 

< 



C/) 
CO 
LU 

CO 

Z3 

CO 



Z3 

< 



t/) 




CO 

cz 

CO 



£? 



t/) 

cz 

c/) 



C/) 


CO 


co 


CO 


CO 


CO 


CO 


CO 


CO 


CO 


CO 





CO 


CO 


CO 




CO 


CO 





CO 


LU 


^ 


LU 


LU 




-£Z 

r 
o 

z 

CO 

o 


cz 

O 

CO 

E 


CO 


LU 


LU 


^ 


LU 


t/) 

cz 

CO 

CO 

_l 




cz 
o 


_l 


CO 

Z3 

'co 

O) 
CO 
LL 


CO 

'*— • 
CO 

> 




cz 
o 


_l 


CO 

E 


CO 

h- 


CO 

Z3 
CO 

E 

< 


CO 

Z3 

cz 

CO 

M— 

CO 

z 










o 


1- 






o 












CL 








Z3 

< 





low 



o 


CO 

c 

CO 





CO 


O 



a. 
co 

5. 

co 
co 


c 

o 



CO 

il 



12 



10 



_b 

5 £ 

co Q 

CO 

Z3 

cz 
^CO 
CO 

z 



CO 

> 











high 
























medium 
































ow 










































































































































































































































































































i 




Lepuiai Tai 

Savaia 

Vailoa 

Salua Tai 

Vaovai 

Poutasi 

Matautu Falealili 

Sasina 

Tafatafa 

Papa Palauli 

Aufaga 

Sapini 

Satalo 

Matatufu 

Vaiala 

Salesatele 

Saasaai 

Tafagamanu 

Lefagaoalii 

Matareva 

Malae 

Sapunaoa 

Tauaoo 

Matautu Falelatai 

Sataua 

Manase 

Afega 

Siufaga Falelatai 

Faleapuna 

Luua 

Saoluafata 

Vavau 

Auala 

Fasitootai 

Faleu 

Lalovi 

Salimu 

Fatuvalu 

Satoalepai 

Apolima uta 

Sapapalii 

Apai 

Saleilua 

Fagalii 





Figure C.31 -TCRMP 

high 



medium 



low 




Figure C.32. MDS plots based on coral community data for sites in each of the American Samoa studies. Sites are coded by 
Bioregion. The R value of the Global ANOSIM test among all Bioregions is provided where results were significant. For the 
MPABR data, the Nafanua Central site was excluded as an extreme outlier. 



ASEPA 



2D Stress: 0.13 



▼ T 

T 



3 

=3 

=3 
< 



CD 


3 


o 


CD 


£ 


CO 


CO 


CO 


E 


D) 


I— 


CO 


CO 
LL 




CO 



CO 

o 

< 



CO 


^ 




CO 


CO 


do 


> 


CD 




1_ 




o 




CO 




<fi 



=3 
=3 



CO 

=3 
CO 

E 
< 



_2 

CO 
CO 
O) 
CO 



CD 


CO 


C 


t/) 


o 


CO 


CD 


O) 


_l 


CO 



MPABR 












2D Stress: 0.11 


T 
T 


▲ 


▼ 


A 


T 




A + 


+ 


T 








Global ANOSIM 




+ 




R = 0.327 








p = 0.002 









TCRMP 



2D Stress: 0.07 



CRSR 



2D Stress: 0.07 



T * 



* * 



• T 
T 



REA 









2D Stress: 0.22 




* 








* * 




T 




+ 








* T 




T 




* 






# 


* • ▲y T 

** 


* 
* 






A * 




T 


Global ANOSIM 








R = 0.566 








p = 0.001 


* * * 







Region 


▼ 1 


T2 


T3 


T4 


?5 


6 


+ 7 


• 8 


+ 9 


^10 


11 


▲ 12 


▲ 13 


▲ 14 


▲ 15 


#16 


#17 


#18 


#19 


#20 




Figure C.33. MDS plots based on fish community data for sites in each of the American Samoa studies. Sites are coded by 
Bioregion. The R value of the Global ANOSIM test among all Bioregions is provided where results were significant. 



ASEPA 



2D Stress: 0.17 



T 
▼ ▲ 



KRS 














+ 




2D Stress: 0.15 

• 


T 


T 


T 






Global ANOSIM 


T 


T A 


T 




R = 0.375 










p = 0.003 








T 



TCRMP 



2D Stress: 0.1 



CRSR 











2D Stress: 0.15 


TA 
T * 

* 


▲ 


* 




T 


T * 
* T 


T 




* 




▲ 


T 


* 






T 


* 









REA 



2D Stress: 0.25 






* * 



Global ANOSl^ 
R = 0.372 
p = 0.001 



Region 


T1 


T2 


T3 


T4 


T5 


6 


+ 7 


• 8 


+ 9 


♦ 10 


11 


▲ 12 


▲ 13 


▲ 14 


▲ 15 


#16 


#17 


#18 


#19 


#20 



Figure C.34. MDS plots based on coral and fish community data for sites in each of the Samoa studies. Sites are coded by 
Bioregion. The R value of the Global ANOSIM test among all Bioregions is provided where results were significant. For the 
SFR fish community data, the Fagali'i site was excluded as an extreme outlier. 

GCRMN Coral 



SFR Coral 






















2D Stress: 0.2 




▲ 










▲ 






▲ 






T ▲* 


T 
* ▼ 


T 

▲ 
T 

▲ 


L 






▼ 




? 


T 




♦ 


T 


T 




♦ 


▲▼ 



2D Stress: 0.04 



SFR Fish 



GCRMN Fish 













2D Stress: 0.18 






▲ 


A T 






♦ 






▲ 






♦ 


▲ 


T 
T 




T 


♦ 


T 








▲ 


A A 


Global ANOSIM T 




v 


▲ 


▲ 
T 

▲ 


▲ 
▲ 

▲ 


R = 0.197 










p = 0.003 



















2D Stress: 0.05 






▲ 










▲ 










T 
T 


T 






▼ 






Region 










A 21 










▼ 23 








▲ 


▲ 24 
♦ 25 

26 
▼ 27 

28 


▲ 








A 29 
▲ 30 



Appendix D: Key attributes and activities of MPAs in the existing network of MPAs in 

American Samoa as of January 2011 

Data provided by Alice Lawrence 
Table D.1. General description of MPA implementation and management. 



MPA 




.evel of Manageme 






gnation ^ati 2§f ™. n * P[f" Laws & Regulations 


















Alofau CFMP Reserve 


Territorial 


DMWR a , village 


2001 


Cooperative 
Agreement 


ASAC§ 24.10 


Amanave CFMP 
Reserve 


Territorial 


DMWR a , village 


2009 


Cooperative 
Agreement 


ASAC§ 24.10 


Amaua& Auto CFMP 
Reserve 


Territorial 


DMWR a , village 


2003 


Cooperative 
Agreement 


ASAC§ 24.10 


Aoa CFMP 
Reserve 


Territorial 


DMWR a , village 


2005 


Cooperative 
Agreement 


ASAC§ 24.10 


Aua CFMP 
Reserve 


Territorial 


DMWR a , village 


2002 


Cooperative 
Agreement 


ASAC§ 24.10 


Fagamalo CFMP 
Reserve 


Territorial 


DMWR a , village 


2003 


Cooperative 
Agreement 


ASAC§ 24.10 


Fagamalo No-Take MPA 


Territorial 


DMWR b , village 


2010 


Cooperative Agree- 
ment; Management 
Plan being Finalized 


ASAC§ 24.1001, 
ASAC§ 24.1008 (c)(1) 


Leone Pala SMA 


Territorial 


ASCMP C , village 


1994 


None 


ASCA§ 24.0503, 
ASAC § 26.0221 


Masausi CFMP Reserve 


Territorial 


DMWR a , village 


2002 


Cooperative 
Agreement 


ASAC§ 24.10 


Matu'u & Faganeanea 
CFMP Reserve 


Territorial 


DMWR a , village 


2005 


Cooperative 
Agreement 


ASAC §24.10 


Nu'uuli Pala SMA 


Territorial 


ASCMP C , village 


1995 


None 


ASCA§ 24.0503, 
ASAC § 26.0221 


Ofu Vaoto Marine Park 


Territorial 


DPR, DMWR 


1994 


None 


ASCA§ 18.0214 


Pago Pago Harbor SMA 


Territorial 


ASCMP C , village 


1997 


None 


ASCA§ 24.0503, 
ASAC § 26.0221 


Poloa CFMP Reserve 


Territorial 


DMWR a , village 


2001 


Cooperative 
Agreement 


ASAC §24.10 


Sailele CFMP Reserve 


Territorial 


DMWR a , village 


2005 


Cooperative 
Agreement 


ASAC §24.10 


Alega Private Marine 
Reserve 


Private 


Alega village d 


1985 


Unwritten 
Agreement 


Unwritten Rules 


Vatia CFMP Reserve 


Territorial 


DMWR a , village 


2001 


Cooperative 
Agreement 


ASAC §24.10 


Fagatele Bay NMS 


Federal/ 

Territorial 

Co-Managed 


NOAA e , AS DOC 


1986 


Management Plan 
(1986), Management 
Plan Review Process 
underway (2009) 


16 USC 1431 etseq., 
Public Law 106-513, 
15 CFR 922.100-104 


NPSA-Ofu unit 


Federal 


NPS f 


1988 (lease 
finalized in 
1993) 


Management Plan 
(1997) 


16USC410qq- 
410qq-1 


NPSA-Ta'u unit 


Federal 


NPS f 


1988 (lease 
finalized in 
1993) 


Management Plan 
(1997) 


16USC410qq- 
410qq-1 


NPSA-Tutuila unit 


Federal 


NPS f 


1988 (lease 
finalized in 
1993) 


Management Plan 
(1997) 


16USC410qq- 
410qq-1 


Rose Atoll 
MNM/NWR 


Federal 


USFWS9, NOAA 


2009 (MNM), 
1973 (NWR) 


Management Plan 
Development Process 
underway 


Presidential Proclama- 
tion 8337, 16U.S.C. 
668dd-ee, Public Law 
105-57 



a - Selaina Vaitautolu (Community-Based Fisheries Management Program Manager), Taahinemanua@yahoo.com 

b - Lucy Jacob (No-Take MPA Program Manager), americansamoa.mpa@gmail.com 

c- Nathan llaoa (American Samoa Coastal Management Program Manager), nate@doc.as 

d - Tisa Fa'amuli, tisa@tisbarefootbar.com 

e- Gene Brighouse (Superintendent), Gene.Brighouse@noaa.gov; Kevin Grant (Deputy Superintendent), Kevin.Grant@noaa.gov 

f- Mike Reynolds (Superintendent), Mike_Reynolds@nps.gov; Tim Clark (Marine Ecologist), Tim_Clark@nps.gov 

g - Frank Pendleton (Rose Atoll Marine National Monument and National Wildlife Refuge Manager), Frank_Pendleton@fws.gov 



Table D.2. Conservation focus and management practices. 



Table D.3. Biological and socio-economic monitoring/assessment and community involvement. 







MPA 


_ y. Protection Level of _ - Fishing Vessel Anchoring 
Conservation _ _ . Permanence Constancy _ . . .. n n * - *■ 

Focus Protection J Restriction Restrictions Restrictions 


















Alofau CFMP 
Reserve 


Cultural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Conditional 


Year Round 


Commercial, 
Recreational 
Prohibited 3 


Unrestricted 


Unrestricted 


Amanave CFMP 
Reserve 


Cultural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Conditional 


Year Round 


Commercial, 
Recreational 
Prohibited 3 


Unrestricted 


Unrestricted 


Amaua& Auto CFMP 
Reserve 


Cultural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Conditional 


Year Round 


Commercial, 
Recreational 
Prohibited 3 


Unrestricted 


Unrestricted 


Aoa CFMP Reserve 


Cultural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Conditional 


Year Round 


Commercial, 
Recreational 
Prohibited 3 


Unrestricted 


Unrestricted 


Aua CFMP Reserve 


Cultural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Conditional 


Year Round 


Commercial, 
Recreational 
Prohibited 3 


Unrestricted 


Unrestricted 


Fagamalo CFMP 
Reserve 


Cultural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Conditional 


Year Round 


Commercial, 
Recreational 
Prohibited 3 


Unrestricted 


Unrestricted 


Fagamalo 
No-Take MPA 


Natural Heritage 


Ecosystem 


No Take 


Conditional 


Year Round 


All Fishing 
Prohibited 


Unrestricted 


TBD 


Leone Pala SMA 


Natural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Permanent 


Year Round 


Commercial, 

Recreational 

Restricted 


Unrestricted 


Unrestricted 


Masausi CFMP 
Reserve 


Cultural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Conditional 


Year Round 


Commercial, 
Recreational 
Prohibited 3 


Unrestricted 


Unrestricted 


Matu'u & 

Faganeanea CFMP 
Reserve 


Cultural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Conditional 


Year Round 


Commercial, 
Recreational 
Prohibited 3 


Unrestricted 


Unrestricted 


Nu'uuli Pala SMA 


Natural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Permanent 


Year Round 


Commercial, 

Recreational 

Restricted 


Unrestricted 


Unrestricted 


Ofu Vaoto Marine 
Park 


Natural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Permanent 


Year Round 


Commercial, 

Recreational 

Restricted 


Presently 
Unknown 


Presently 
Unknown 


Pago Pago 
Harbor SMA 


Natural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Permanent 


Year Round 


Commercial, 

Recreational 

Restricted 


Unrestricted 


Unrestricted 


Poloa CFMP Reserve 


Cultural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Conditional 


Year Round 


Commercial, 
Recreational 
Prohibited 3 


Unrestricted 


Unrestricted 


Sailele CFMP 
Reserve 


Cultural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Conditional 


Year Round 


Commercial, 
Recreational 
Prohibited 3 


Unrestricted 


Unrestricted 


Alega Private Marine 
Reserve 


Sustainable 
Production 


Ecosystem 


Uniform 
Multi-Use 


Permanent 


Year Round 


All Fishing 
Restricted 13 


Restricted 


Restricted 


Vatia CFMP Reserve 


Cultural Heritage 


Ecosystem 


Uniform 
Multi-Use 


Conditional 


Year Round 


Commercial, 
Recreational 
Prohibited 3 


Unrestricted 


Unrestricted 


Fagatele Bay NMS 


Natural, Cultural 
Heritage 


Ecosystem 


Zoned 
Multi-Use 


Permanent 


Year Round 


Commercial, 

Recreational 

Restricted 


Unrestricted 


Restricted 


NPSA- Ofu unit 


Natural, Cultural 
Heritage 


Ecosystem 


Uniform 
Multi-Use 


Permanent* 


Year Round 


Commercial, 
Recreational 
Restricted 


Unrestricted 


Unrestricted 


NPSA-Ta'u unit 


Natural, Cultural 
Heritage 


Ecosystem 


Uniform 
Multi-Use 


Permanent* 


Year Round 


Commercial, 
Recreational 
Restricted 


Unrestricted 


Unrestricted 


NPSA -Tutuila unit 


Natural, Cultural 
Heritage 


Ecosystem 


Uniform 
Multi-Use 


Permanent* 


Year Round 


Commercial, 
Recreational 
Restricted 


Unrestricted 


Unrestricted 


Rose Atoll MNM/NWR 


Natural Heritage 


Ecosystem 


Zoned w/ No 
Take Areas 


Permanent 


Year Round 


Commercial, 

Recreational 

Restricted 


Restricted 


Restricted 



a - The reserve is closed to all fishing apart from when it is opened for subsistence fishing at certain times of the 
b - Only subsistence fishing with traditional methods by villagers/relatives is allowed, 
c - Only subsistence fishing with traditional methods is allowed. 
* 50 year renewable lease and permanent federal funding. 



year. 







MPA 


Biophysical Assessments/ 


Socio-economic 

Assessments/ 

Monitoring 


Community Power to 
Community/Stakeholder Engagement Take/Enforce Man- 


Alofau CFMP 
Reserve 


Baseline Assessment; 
Biannual Monitoring Program 


Future Plans to Design 
and Implement Socio- 
economic Monitoring 


Involved in Management Planning, Involved 
in Site Management, Existence of Collab- 
orative Management Mechanisms 


Yes 


Amanave CFMP 
Reserve 


Baseline Assessment; 
Biannual Monitoring Program 


Future Plans to Design 
and Implement Socio- 
economic Monitoring 


Involved in Management Planning, Involved 
in Site Management, Existence of Collab- 
orative Management Mechanisms 


Yes 


Amaua & Auto 
CFMP Reserve 


Baseline Assessment; 
Biannual Monitoring Program 


Future Plans to Design 
and Implement Socio- 
economic Monitoring 


Involved in Management Planning, Involved 
in Site Management, Existence of Collab- 
orative Management Mechanisms 


Yes 


Aoa CFMP 
Reserve 


Baseline Assessment; 
Biannual Monitoring Program 


Future Plans to Design 
and Implement Socio- 
economic Monitoring 


Involved in Management Planning, Involved 
in Site Management, Existence of Collab- 
orative Management Mechanisms 


Yes 


Aua CFMP 
Reserve 


Baseline Assessment; 
Biannual Monitoring Program 


Future Plans to Design 
and Implement Socio- 
economic Monitoring 


Involved in Management Planning, Involved 
in Site Management, Existence of Collab- 
orative Management Mechanisms 


Yes 


Fagamalo CFMP 
Reserve 


Baseline Assessment; 
Biannual Monitoring Program 


Future Plans to Design 
and Implement Socio- 
economic Monitoring 


Involved in Management Planning, Involved 
in Site Management, Existence of Collab- 
orative Management Mechanisms 


Yes 


Fagamalo 
No-Take MPA 


Baseline Assessment 


Baseline Assessment; 

Monitoring Program in 

Development 


Involved in Management Planning, Involved 
in Site Management 


Yes 


Leone Pala SMA 


Presently Unknown 


Presently Unknown 


Involved in Management Planning 


No 


Masausi CFMP 
Reserve 


Baseline Assessment; 
Biannual Monitoring Program 


Future Plans to Design 
and Implement Socio- 
economic Monitoring 


Involved in Management Planning, Involved 
in Site Management, Existence of Collab- 
orative Management Mechanisms 


Yes 


Matu'u & Faga- 
neanea CFMP 
Reserve 


Baseline Assessment; 
Biannual Monitoring Program 


Future Plans to Design 
and Implement Socio- 
economic Monitoring 


Involved in Management Planning, Involved 
in Site Management, Existence of Collab- 
orative Management Mechanisms 


Yes 


Nu'uuli Pala SMA 


Presently Unknown 


Presently Unknown 


Involved in Management Planning 


No 


Ofu Vaoto Marine 
Park 


Presently Unknown 


Presently Unknown 


Presently Unknown 


Presently Unknown 


Pago Pago 
Harbor SMA 


Presently Unknown 


Presently Unknown 


Involved in Management Planning 


No 


Poloa CFMP 
Reserve 


Baseline Assessment; 
Biannual Monitoring Program 


Future Plans to Design 
and Implement Socio- 
economic Monitoring 


Involved in Management Planning, Involved 
in Site Management, Existence of Collab- 
orative Management Mechanisms 


Yes 


Sailele CFMP 
Reserve 


Baseline Assessment; 
Biannual Monitoring Program 


Future Plans to Design 
and Implement Socio- 
economic Monitoring 


Involved in Management Planning, Involved 
in Site Management, Existence of Collab- 
orative Management Mechanisms 


Yes 


Alega Private 
Marine Reserve 


Fish Catch Monitoring by 
Tisa; DMWR Key Reef Spe- 
cies Program Monitoring Site 


None 


Village Community Designated and 
Manages the Reserve 


Yes 


Vatia CFMP 
Reserve 


Baseline Assessment; 
Biannual Monitoring Program 


Future Plans to Design 
and Implement Socio- 
economic Monitoring 


Involved in Management Planning, Involved 
in Site Management, Existence of Collab- 
orative Management Mechanisms 


Yes 


Fagatele Bay 
NMS 


NOAA Biogeographic Assess- 
ment in 2008, 2010; Long-term 
Monitoring Program in Place 


Baseline Assessment; 

Monitoring Program in 

Development 


Involved in Management Planning, 

Existence of Collaborative Management 

Mechanisms 


Some 


NPSA- Ofu unit 


Baseline Assessment; 
Annual Monitoring Program 


Occasional 
Assessments 


Involved in Management Planning, 

Existence of Collaborative Management 

Mechanisms 


Some 


NPSA-Ta'u unit 


Baseline Assessment; 
Annual Monitoring Program 


Occasional 
Assessments 


Involved in Management Planning, 

Existence of Collaborative Management 

Mechanisms 


Some 


NPSA -Tutuila 
unit 


Baseline Assessment; 
Annual Monitoring Program 


Occasional 
Assessments 


Involved in Management Planning, 

Existence of Collaborative Management 

Mechanisms 


Some 


Rose Atoll MNM/ 
NWR 


Baseline Assessment; 
Regular Monitoring Program 


Occasional 
Assessments 


Some - Involved in Management Planning 

Process; Meetings held in Manu'a and 

Tutuila, November 2009 


No 




Table D.4. Current and future projects. 



MPA 




ects/Activitiei 




RJjffiWJfPWIP^WJfPH 




rojects/Activit 




Alofau CFMP 
Reserve 


Education & Outreach; Monitoring; 
Village Beach & Underwater Clean 
Ups; Community Reef Monitoring 


w/ No-Take MPA Program on Education & Out- 
reach Programs, Socio-economic Assessments, 
MPA Designation Work, VM PA Awareness, and 
Enforcement Partnerships 


Regular Biological Monitoring; 
Socio-economic Surveys; Edu- 
cation & Outreach; Training and 
Capacity Building; Enforcement 
Activities (e.g. Boat Patrols) 


Amanave CFMP 
Reserve 


Education & Outreach; Monitoring; 
Village Beach & Underwater Clean 
Ups; Community Reef Monitoring 


w/ No-Take MPA Program on Education & Out- 
reach Programs, Socio-economic Assessments, 
MPA Designation Work, VM PA Awareness, and 
Enforcement Partnerships 


Regular Biological Monitoring; 
Socio-economic Surveys; Edu- 
cation & Outreach; Training and 
Capacity Building; Enforcement 
Activities (e.g. Boat Patrols) 


Amaua & Auto 
CFMP Reserve 


Education & Outreach; Monitoring; 
Village Beach & Underwater Clean 
Ups; Community Reef Monitoring 


w/ No-Take MPA Program on Education & Out- 
reach Programs, Socio-economic Assessments, 
MPA Designation Work, VM PA Awareness, and 
Enforcement Partnerships 


Regular Biological Monitoring; 
Socio-economic Surveys; Edu- 
cation & Outreach; Training and 
Capacity Building; Enforcement 
Activities (e.g. Boat Patrols) 


Aoa CFMP 
Reserve 


Education & Outreach; Monitoring; 
Village Beach & Underwater Clean 
Ups; Community Reef Monitoring 


w/ No-Take MPA Program on Education & Out- 
reach Programs, Socio-economic Assessments, 
MPA Designation Work, VM PA Awareness, and 
Enforcement Partnerships 


Regular Biological Monitoring; 
Socio-economic Surveys; Edu- 
cation & Outreach; Training and 
Capacity Building; Enforcement 
Activities (e.g. Boat Patrols) 


Aua CFMP 
Reserve 


Education & Outreach; Monitoring; 
Village Beach & Underwater Clean 
Ups; Community Reef Monitoring 


w/ No-Take MPA Program on Education & Out- 
reach Programs, Socio-economic Assessments, 
MPA Designation Work, VM PA Awareness, and 
Enforcement Partnerships 


Regular Biological Monitoring; 
Socio-economic Surveys; Edu- 
cation & Outreach; Training and 
Capacity Building; Enforcement 
Activities (e.g. Boat Patrols) 


Fagamalo 
CFMP Reserve 


Education & Outreach; Monitoring; 
Village Beach & Underwater Clean 
Ups; Community Reef Monitoring 


w/ No-Take MPA Program on Education & Out- 
reach Programs, Socio-economic Assessments, 
MPA Designation Work, VM PA Awareness, and 
Enforcement Partnerships 


Regular Biological Monitoring; 
Socio-economic Surveys; Edu- 
cation & Outreach; Training and 
Capacity Building; Enforcement 
Activities (e.g. Boat Patrols) 


Fagamalo No- 
Take MPA 


Education & Outreach; Biological As- 
sessments; Learning Exchange; Cur- 
rent Surveys; Drop Cam Surveys; 
Larval Dispersal Modeling 


w/ CFMP on Education & Outreach Programs, 
Socio-economic Assessments and MPA Designa- 
tion Work; w/ Fishery Staff for Drop Cam Work; w/ 
EPA on Current Surveys and Modeling 


MPA Designation Process; Edu- 
cation & Outreach; Engagement 
Workshops; Monitoring Training; 
Socio-economic Surveys; Model- 
ing; Mapping; Network Develop- 
ment; Improve Enforcement 


Leone Pala 
SMA 


Presently Unknown 


Presently Unknown 


Presently Unknown 


Masausi CFMP 
Reserve 


Education & Outreach; Monitoring; 
Village Beach & Underwater Clean 
Ups; Community Reef Monitoring 


w/ No-Take MPA Program on Education & Out- 
reach Programs, Socio-economic Assessments, 
MPA Designation Work, VM PA Awareness, and 
Enforcement Partnerships 


Regular Biological Monitoring; 
Socio-economic Surveys; Edu- 
cation & Outreach; Training and 
Capacity Building; Enforcement 
Activities (e.g. Boat Patrols) 


Matu'u & Faga- 
neanea CFMP 
Reserve 


Education & Outreach; Monitoring; 
Village Beach & Underwater Clean 
Ups; Community Reef Monitoring 


w/ No-Take MPA Program on Education & Out- 
reach Programs, Socio-economic Assessments, 
MPA Designation Work, VM PA Awareness, and 
Enforcement Partnerships 


Regular Biological Monitoring; 
Socio-economic Surveys; Edu- 
cation & Outreach; Training and 
Capacity Building; Enforcement 
Activities (e.g. Boat Patrols) 


Nu'uuli Pala 
SMA 


Presently Unknown 


Presently Unknown 


Presently Unknown 


Ofu Vaoto Ma- 
rine Park 


Presently Unknown 


Presently Unknown 


Presently Unknown 


Pago Pago 
Harbor SMA 


Presently Unknown 


Presently Unknown 


Presently Unknown 


Poloa CFMP 
Reserve 


Education & Outreach; Monitoring; 
Village Beach & Underwater Clean 
Ups; Community Reef Monitoring 


w/ No-Take MPA Program on Education & Out- 
reach Programs, Socio-economic Assessments, 
MPA Designation Work, VM PA Awareness, and 
Enforcement Partnerships 


Regular Biological Monitoring; 
Socio-economic Surveys; Edu- 
cation & Outreach; Training and 
Capacity Building; Enforcement 
Activities (e.g. Boat Patrols) 



Table D.4. cont. 


Current and future projects. 






MPA 


Existing Projects/Activit 




k*vjI 7j*v/jn 




Sailele CFMP 
Reserve 


Education & Outreach; Monitoring; 
Village Beach & Underwater Clean 
Ups; Community Reef Monitoring 


w/ No-Take MPA Program on Education & Out- 
reach Programs, Socio-economic Assessments, 
MPA Designation Work, VM PA Awareness, and 
Enforcement Partnerships 


Regular Biological Monitoring; 
Socio-economic Surveys; Edu- 
cation & Outreach; Training and 
Capacity Building; Enforcement 
Activities (e.g. Boat Patrols) 


Alega Private 
Marine Reserve 


Education & Outreach; Fish Catch 
Monitoring; Turtle Nest Monitoring; 
Annual Palolo Worm Festival & Fish- 
ing Events; Practicing of Traditional 
Fishing Methods 


w/ DMWR's Marine Monitoring Program, Marine 
Mammal Monitoring Program 


REEF Fish ID program to be 
Implemented using Observations 
by Visitors to the Reserve; Tradi- 
tional Knowledge Documentation 
Project 


Vatia CFMP 
Reserve 


Education & Outreach; Monitoring; 
Village Beach & Underwater Clean 
Ups; Community Reef Monitoring 


w/ No-Take MPA Program on Education & Out- 
reach Programs, Socio-economic Assessments, 
MPA Designation Work, VM PA Awareness, and 
Enforcement Partnerships 


Regular Biological Monitoring; 
Socio-economic Surveys; Edu- 
cation & Outreach; Training and 
Capacity Building; Enforcement 
Activities (e.g. Boat Patrols) 


Fagatele Bay 
NMS 


Management Plan Review and 
Possible Additional Sanctuary Unit 
Designation Process; Biogeographic 
Assessment; Community Engage- 
ment; Scientific Research; Maritime 
Heritage Study 


w/ASCC Internship Program; w/ LBJ Hospital to 
open a Hyperbaric Chamber on Island; w/ Sea 
Education Association (S.E.A.) on Oceanographic 
Research, Education, and Cultural Exchange; 
w/ Hawaiian Islands Humpback Whale NMS and 
DMWR on Humpback Whale and Cetacean Re- 
search; w/ NCCOS on Biogeographic Assessment 
of the Samoan Archipelago; w/ Le Tausagi: Enviro- 
Discoveries Camps; w/ASG - Preserve America 
Initiatives; w/ University of Hawaii - Research and 
Monitoring 


Renovate Sanctuary Offices and 
Visitor's Center; Proposed Plans 
to Expand to Include Additional 
Units at Tutuila: Larsen Bay and 
Aunu'u, Manu'a: Ta'u, Rose Atoll, 
and Swains island; International 
Exchange Program; Education 
Workshop; Climate Change 
Strategy 


NPSA-Ofu unit 


Siapo Educational DVD; Natural 
History of AS Publication; Research 
Papers; New NP Ranger on Ofu 


w/ DMWR, NOAA, FWS, DOI-insular office, Office 
of Samoan Affairs, PICED, Coast Guard, Territorial 
EMS/LBJ Hospital, Territorial Historic Preservation 
Office, ASCC, Land Grant, Sea Grant, and the 
Villages of Vatia, Fagasa, Pago Pago, Sili, Ofu, 
Olosega, Ta'u, Fiti'uta, and Faleasou on Various 
Projects 


Research; Annual Monitoring; 
Education & Outreach; Encour- 
age Climate Change Research 
on Impacts to Coral Reefs at Ofu 
Field Station 


NPSA-Ta'u unit 


Siapo Educational DVD; Natural 
History of AS Publication; Research 
Papers 


w/ DMWR, NOAA, FWS, DOI-insular office, Office 
of Samoan Affairs, PICED, Coast Guard, Territorial 
EMS/LBJ Hospital, Territorial Historic Preservation 
Office, ASCC, Land Grant, Sea Grant, and the 
Villages of Vatia, Fagasa, Pago Pago, Sili, Ofu, 
Olosega, Ta'u, Fiti'uta, and Faleasou on Various 
Projects 


Research; Annual Monitoring; 
Education & Outreach 


NPSA-Tutuila 
unit 


Siapo Educational DVD; Natural 
History of AS Publication; Research 
Papers 


w/ DMWR, NOAA, FWS, DOI-insular office, Office 
of Samoan Affairs, PICED, Coast Guard, Territorial 
EMS/LBJ Hospital, Territorial Historic Preservation 
Office, ASCC, Land Grant, Sea Grant, and the 
Villages of Vatia, Fagasa, Pago Pago, Sili, Ofu, 
Olosega, Ta'u, Fiti'uta, and Faleasou on Various 
Projects 


Research; Annual Monitoring; 
Education & Outreach 


Rose Atoll 
MNM/NWR 


Planning Review Updates; Public 
Consultations 


w/ DMWR on Research and Monitoring Work; Cur- 
rently w/ FBNMS and NMFS to Develop Manage- 
ment Plans for the MNM 


Collaborate w/ NMFS and 
DMWR on Monitoring Efforts 



> 


00 


o 


o 


m 


O 


z G> 


o 5 


> > 


H — 


© O 


£> 


3 CO 


o £ 


o) m 


— CO 


2 £ 


© 2 


3 m 


o z 


3 ^ 


= O 


Q- -n 


§3 


z m 


O co 


CO > 


z 2 


o o 


o > 


O z 


CO > 


rt *> 




■D 


m 


i- 


> 


O 




O 





United States Department of Commerce 

Gary Locke 
Secretary 



National Oceanic and Atmospheric Administration 

Jane Lubchenco 
Administrator 



National Ocean Service 

David Kennedy 
Assistant Administrator