Skip to main content

Full text of "DTIC ADA1045954: An Open-Ocean Marine Fog Development and Forecast Model for Ocean Weather Station Papa."

See other formats


'^AlO* 59S 


NAVAL POSTMAOUATE SCHOOL MONTEREY CA F/0 4/2 

AN OPEN-OCEAN MARINE FOO DEVELOPMENT AND FORECAST MODEL FOR OCE—ETC(U) 
JUN SI R L CLARK 






























ADA104595 


‘(:>'NPS68-81-005 









NAVAL POSTGRADUATE SCHOOL 

Monterey, California 




THESIS 


AN OPEN-OCEAN MARINE FOG DEVELOPMENT AND 
FORECAST MODEL FOR OCEAN WEATHER 
STATION PAPA 

by 

Robert Louis Clark 
June 1981 


Approved for public release; distribution unlimited. 


Prepared For; 

Naval Air Systems Command 
Washington, D.C. 

81 9 2B 03^ 








READ IWSTRUCTtONS 

■ErORE COMRUETTWO FORM 
nflCMT*! catalog MUMACa 


UNCLASSIFIED 


/ ■ AR Pn j ty . I*' RO^T ACCEWOW MO. ^ 

( ADl-Ai-Q4^^ r 


1/ 


IKCUAITV CL»tl>IC»TIOi« or Twit ■•Ot fWttmt Data Sataaaa) 


REPORT DOCUMENTATION PACE 


rwa luiinrar 


An Open-Ocean Marine Fog Development and 
:Forecast Model for Ocean Weather Station 
'Papa . -- 



t. ACRroMiiMa OAO. ACAeAr mumaia 


7 , *uTMOAfa> 


COATAACT OA OAAAT NUMACAfaj 


Robert Louis /ciarJc 





ir 


• CAroAMIAO OAaAMIlATIOM MAMC AnO AOOAlll 

Naval Postgraduate School 
Monterey, California 93940 


AAeOAAM CLEMENT. AAOJCCT. TASK 
AACA • AOAK unit NUMACAS 


N6685681WR81003 


It CONTAOLLINa OKAICC NAME ANO AOOACM 

Naval Postgraduate School 
Monterey, California 93940 




W. MEAQATflum- 


J\an 


81 


•'7 


14. MONITOAimC ASCmCV name a AOaactS/lf MtHrwmt hmm CamttmlUmg 0«Nas> 

Naval Environmental Prediction Research 
Facility 

Monterey, California 93940 


129 


IE. ICCUAITV Class, (mi Utlm rSAart) 

Unclassified 


tta. OCClassiaication/ooamgaaoimc 
SCNEOULS 


IS. OtSTAiAuTlON statement (ml lHlm Kmmmni 


Approved for public release; distribution unlimited. 


IT. OISTAIAUTION STATEMENT (ml lAa M«,rMC mmimrm^ln Blmm» M, II Afftaran* Immm Mmmmn) 


IS. suaalcmsntaay notes 


IS. 


KEY VOAOS (Cmmilmmm mm rmrmrmm rnimm il mmmmmmmmr mmt lASAflfY *T MnA wiSfJ 


Advection Fog 
Fog Forecasting 
Marine Fog 


North Pacific Ocean Fog 
Open-ocean Fog 
Visibility Forecasting 


Fog Modeling 


10. aASTAACT (Cmmummm mm tmmmmmm rnimrn II mmmmmmmtf mmt lAaMIIY Sr SIraS mmmtrnr} 

■>"'Marine fog forecasts during the simmer period in the North Pacif; 
are not made presently with auiy acceptable degree of accuracy. Ob¬ 
jective fog development models exist and are used with some success 
for localized coastal regions of the western U.S.; scarcity of 
accurate data has hindered creation of a reliable open-ocean model. 
The Eulerian single-station approach, utilizing a segment of the 
complete accurate data of Ocean Weather Station Papa (50N,145W). 


00 1473 

(PaRe 1) 


EDITION OA I NOV SS IS ORtOLRTt 
S/N OIOI'OIA* ASOI 

I 


UNCLASSIFIED 


security CLASStAICATION OA THIS RARE 


.'J 

AE fWhmm Oats I 











#20 - ABSTRACT - (CONTINUED) 

*is applied in this study to an objective marine fog forecasting 
model. 

The time-series study of significant atmospheric variables 
at OWS Papa, when coupled with a chronological synoptic over¬ 
view, delineates accur.ately fog/no fog sequences in the summer 
months of 1973 and 1977. Actual observed fog situations are 
evaluated by the general model and presented in relation to oper 
ocean fog indices, NOAA 5 satellite coverage and synoptic 
history. 

The open-ocean forecast model is tested on an independent 
data set for the month of July 1975 at OWS Papa, with favor- 
cible results. 

The research delineates four required indices that must 
all be positive to forecast fog. These indices, when plotted 
daily in the region of OWS Papa allow a single station to 
predict, with some confidence out to twenty-four hours, the 
occurrence of advection fog.^ 


Access ion For 
NT 13 

F'ilC; TA'; 
IjiDiiM'.lCl!') '5 

J . X C . • 


By_ 










Approved for public release; distribution unlimited. 

An Open-Ocean Marine Fog Development and 
Forecast Model for Ocean Weather 
Station Papa 

by 

Robert Louis Clark 
Lieutenant, United States Navy 
B.S., United States Naval Academy, 1975 


Submitted in partial fulfillment of the 
requirements for the degree of 


MASTER OF SCIENCE IN METEOROLOGY AND OCEANOGRAPHY 

from the 

NAVAL POSTGRADUATE SCHOOL 
June 1981 


Author 
Approved by: 














ABSTRACT 

Marine fog forecasts during the summer period in the North 
Pacific are not made presently with any acceptable degree of 
accuracy. Objective fog development models exist and are 
used with some success for localized coastal regions of the 
western U.S.; scarcity of accvirate data has hindered creation 
of a reliable open-ocean model. The Eulerian single-station 
approach, utilizing a segment of the complete accurate data 
of Oceaui Weather Station Papa (50N,145W) is applied in this 
study to an objective marine fog forecasting model. 

The time-series study of significant atmospheric variables 
at OWS Papa, when coupled with a chronological synoptic over¬ 
view, delineates accurately fog/no fog sequences in the sum¬ 
mer months of 1973 and 1977. Actual observed fog situations 
ace evaluated by the general model and presented in relation 
to open-ocean fog indices, NOAA 5 satellite coverage and 
synoptic history. 

The open-ocean forecast model is tested on an independent 
data set for the month of July 1975 at OWS Papa, with favor¬ 
able results. 

The research delineates four required indices that must all 
be positive to forecast fog. These indices, when plotted daily 
in the region of OWS Papa allow a single station to predict, 
with some confidence out to twenty-four hours, the occurrence 
of advection fog. 


4 








I. INTRODUCTION AND BACKGROUND - 

II. APPROACH AND OBJECTIVES - 

III. DATA - 

IV. FOG DEVELOPMENT MODEL AND OPEN OCEAN INDICES 

V. INDEPENDENT TEST - 

VI. CONCLUSIONS AND RECOMMENDATIONS - 

APPENDIX A: INTERNATIONAL VISIBILITY CODES - 

APPENDIX B; DATA SOURCES - 

APPENDIX C; LEIPPER COASTAL FORECAST PARAMETERS 

AND INDICES - 

APPENDIX D: OPEN OCEAN FORECAST PARAMETERS AND 

INDICES - 

APPENDIX E; DAILY RAOB PLOTS - 

TABLES - 

FIGURES - 

LIST OF REFERENCES - 


INITIAL DISTRIBUTION LIST 




















LIST OF FIGURES 


1. Ocean Station Papa (SON, 145W) and General Sea 

Surface Circulation - 92 

2. Mean Sea Level Pressure in July (mb) - 93 

3. Surface Synoptic Display, 10 August 1977, OOZ - 94 

4. Surface Synoptic Display, 11 August 1977, OOZ - 95 

5. Surface Synoptic Display, 12 August 1977, OOZ -- 96 

6. Surface Synoptic Display, 13 August 1977, OOZ - 97 

7. Surface Synoptic Display, 14 August 1977, OOZ - 98 

8. July 1973 Inversion Height and Horizontal 

Visibility- 99 

9. July 1973 Air Temperature and Dew Point 

Temperature-100 

10. July 1973 Sea Level Press., and Sea Surface Temp. - 101 

11. July 1973 Wind Direction and Wind Speed-102 

12. July 1973 Temp. Index and Moisture Index-103 

13. July 1977 Inversion Height and Horizontal 

Visibility-104 

14. July 1977 Air Temperature and Dew Point 

Temperature- 105 

15. July 1977 Sea Level Pressure and Sea Surface 

Temperature- 106 

16. July 1977 Wind Direction and Wind Speed-107 

17. July 1977 Temperature Index and Moisture Index — 108 

18. Aug- 1973 Inversion Height and Horizontal 

Visibility- 109 

19. Aug. 1973 Air Temperature and Dew Point 

Temperature- 110 

20. Aug*1973 Sea Level Pressure and Sea Surface 

Temperature- 111 


6 















Aug. 197 3 Wind Direction and Wind Speed- 

Aug.1973 Temperature Index and Moisture Index 

Aug. 1977 Inversion Height and Horizontal 
Visibility - 

Aug.1977 Air Temperature and Dew Point 
Temperature - 

Aug.1977 Sea Level Pressure and Sea Surface 
Temperature - 

Aug.1977 Wind Direction and Wind Speed - 

Aug.1977 Temperature Index and Moisture Index 

July 1975 Inversion Height and Horizontal 
Visibility - 

July 1975 Air Temperature and Dew Point 
Temperature - 

July 1975 Sea Level Pressure and Sea Surface 
Temperature —- 

July 1975 Wind Direction and Wind Speed - 

July 1975 Temperature Index and Moisture Index 


























TABLE OF SYMBOLS AND ABBREVIATIONS 


C Degrees Celcius 

ft. foot 

Fig. Figure 

G/KG Grams/Kilogram 

GMT Greenwich Mean Time 

HT Height 

I.H. Inversion Height 

Kt Knots 

M Meters 

mb Millibars 

NAS Naval Air Station 

NOAA National Oceanic and Atmospheric Administration 

NFS Naval Postgraduate School 

NM Nautical Miles 

OWS Ocean Weather Station 

PST Pacific Standard Time 

RAOB Radiosonde Observation 

RH Relative Humidity 

SAT Surface Air Temperature 

SST Sea Surface Temperature 

T Temperature 

Z Zulu Time 


9 








ACKNOWLEDGMENTS 


The author expresses his sincere appreciation to his 
thesis advisor, Dr. Glenn H. Jung of the Naval Postgraduate 
School, for his long hours of professional and patient guid¬ 
ance. His advice and assistance both as a Professor of 
Ocecinography and as a friend were immeasurable. 

Deep gratitude is also expressed to Professor Emeritus 
Dale F. Leipper who provided hours of detailed background, 
countless articles and constructive and concise comments on 
the preparation of this paper. 

In particular, the author wishes to thank his wife Mac 
for her loving understsuiding and encouragement. 


10 


















1. INTRODUCTION AND BACKGROUND 

Marine fog forecasting is one of meteorology's difficult 
remaining problems. Although the individual major variables 
affecting fog formation are well known, the great variability 
of factors involved such as cloud cover, wind, radiation, 
temperature, surface currents and synoptic weather features 
serve to make the forecast problem quite complex. Compound¬ 
ing difficulties in research on fog formation, marine meteoro¬ 
logical sampling stations are scarce? also reports from 
ships of opportunity are seldom made by trained meteorolo- 
logical observers and are lacking in the accuracy necessary 
for a forecast model (Schrock and Jung, 1976). 

For the purposes of this study, marine fog was defined as 
listed in the Federal Meteorological Handbook, No. 2 (1969): 

fog - a visible aggr^'gate of minute water particles 
based at the earth's surface which reduces 
horizontal visibility below 1,000 meters 
(5/8 mile). 

This reduction in visibility by water vapor corresponds to 
code figures 90 to 94 as described in this reference (Appendix 
A) . 

When fog occurs and extends downward to the ocean surface, 
its impact is substcintial on both commercial and military 
activities. Effects and restrictions on naval operations have 
been accurately delineated (Wheeler and Leipper, 1974)? more 
recently, specific aircraft carrier operational situations 


11 









in relation to restricted visibility have been described 
(Selsor, 1980). 

Historically, the approach to the problem of accurate fog 
forecasting has been varied. Thorough knowledge of the pre¬ 
sent synoptic condition is a prerequisite for a future pre¬ 
diction. Misciasci (1974) listed three main categories of 
analytical fog research: (1) climatological, (2) synoptic, 
and (3) statistical-numerical. To describe more fully all 
branches of fog research, Schrock and Jung (1976) expanded the 
number of research categories to six: (1) microphysics and 
physics of fog formation and dissipation, (2) statistical- 
numerical modeling and forecasting, (3) depiction of fog areas 
over the open ocean from satellite information, (4) marine fog 
climatology, (5) synoptic modeling and forecasting, and (6) 
classificational descriptive definitions. The research in 
this first approach utilized the synoptic modeling and fore¬ 
casting method to establish offshore conditions for one specific 
coastal fog study. 

A second approach that has shown some success in marine 
forecasting is the statistical-numerical modeling and fore¬ 
casting category. Utilizing a statistical regression analysis 
with a limited set of numerical model output parameters (MOP's), 
work was initiated at the Naval Postgraduate School (NPS), 
Monterey, California starting with Schramm (1966) and further 
refined by Nelson (1972), Aldinger (1979), Yavorsky and 
Renard (1980), and Selsor (1980). The final model output 


12 





statistics (MOS) three-category scheme iSelsor, 1980) is an 
operationally-oriented method, yet the data input for the 
necessary coverage and accuracy remains lacking. 

A third approach, which is the focus of the present re¬ 
search, utilizes a very localized time series of significant 
atmospheric varicJales as a fog development model in conjunc¬ 
tion with a large scale synoptic overview to provide the fore¬ 
cast basis. This approach and objectives are described in 
the next chapter. 


13 




II. APPROACH AND OBJECTIVES 


Global fog forecasting and, more specifically, the 
ability to forecast timing and extent of marine fog is still 
in its early stages. Only the United States Navy's Fleet 
Numerical Oceanography Center has an operational objective 
fog forecast program. It is based on a computerized statis¬ 
tical analysis of five weighted fog-related model output 
parameters which are products of numerical analysis or primi¬ 
tive equation models. 

According to Renard (1976) there still exist two major 
problems that are symptomatic of inaccurate marine fog 
forecasts: 

1. Initial-state depiction of marine fog is incomplete 
and inaccurate. 

2. Currently available climatological marine-fog statis¬ 
tics are contradictory and/or inaccurate. 

A major objective of the present research is to reduce the 
first of these stated problems by providing, for one location, 
an accurate and complete history of fog related variables. 

This is accomplished by utilization of a single station of 
professionally-recorded data. Although this is a little-used 
approach to the forecasting of marine fog, the time-series 
method itself is not historically new. 

The use of a synoptic approach to predict fog formation, 
coupled with a day-to-day sequence focusing on a localized 
area, was first described in Leipper's (1948) classic paper. 


14 



"Fog Development at San Diego." This systematic approach 
was successful once the effects of local influences were 
understood as to their impact on a set of initial atmospheric 
conditions. This material was reviewed and the area of appli¬ 
cation extended to the Santa Monica area in a subsequent 
article by Leipper (1968). 

Since that time there have been several recent coastal 
studies, such as McConnell (1975), Peterson and Leipper (1975), 
Schrock and Jung (1976), Beardsley euid Leipper (1976), zmd 
Evermeunn eind Leipper (1976), that have applied Leipper's model to 
specific localized coastal regions of the eastern Pacific. 

The success of these studies at limited coastal locations 
and the proposed synoptic marine fog model patterned by 
Evermann suggested the possibility of adapting a working 
coastal fog development model to the open ocean situation. 

The overall objective of this study is to attempt to 
adapt a successful coastal fog development model to a com¬ 
pletely marine format. In this case the location chosen at 
which to develop the model is in the Northeast Pacific, 
specifically the Gulf of Alaska region. The research applies 
eui Eulerian single-station approach utilizing the complete and 
accurate data record of Ocean Weather Station Papa (SON, 145W) 
in the central Gulf of Alaska (Figure 1). The reasoning be¬ 
hind using the Eulerian single-station approach with a time- 
series study of atmospheric variables is the representative 
location of Ocean Weather Station Papa with respect to an open 


15 








ocean location for fog application and the completeness and 
accuracy of the data record. 

Canadiaui operation of Ocean Weather Station Papa began in 
December 1950. The station has since been used primarily to 
make meteorological observations of surface and upper air 
variables and to provide an open ocean air-sea rescue service. 
The station itself, located in time zone plus ten west, is 
manned by two vessels operated by the Marine Services Branch 
of the Ministry of Trauisport (CCGS Vancouver, CCGS Quadra) . 

Each vessel spends six weeks on station in rotation before 
returning to the mainland. Weather observations are made and 
recorded by trained weather observers. 

A prerequisite to accurate fog forecasting is the detailed 
knowledge of the present mesoscale features of the lower atmos¬ 
phere as well as a knowledge of the general synoptic state 
and recent history. The Leipper coastal forecasting approach 
centers around the atmospheric vertical temperature structure 
of the lower three thousand meters (below 900 mb). Leipper's 
model is a four-stage series that was extended northward to the 
central California coast where Peterson and Leipper (1975) 
adapted it to a three-stage model. 

In general, the first stage is initiated when the North 
Pacific subtropical high moves eastward to an inland position 
over northern California. Subsidence causes the continental 
air to be warm and dry and this factor is further enhanced by 
a downslope and offshore flow with adiabatic warming. The 
flow of warm air over the cooler coastal water initiates a 


16 





lit' 


low-level surface inversion. This inversion restricts verti¬ 
cal movement and confines any increase in the moisture con¬ 
tent to the layer adjacent to the sea surface. With the air 
mass above the inversion relatively dry, there is increased 
radiational cooling. Stage two is indicated by a decrease 
in the offshore flow so that the offshore air mass is relatively 
stationary over the colder ocean surface. A surface inversion 
is formed and the surface layer becomes nearly saturated at 
a temperature very close to that of the underlying sea surface. 
Stage three is the return of the normal northwesterly flow 
and advection of the now nearly saturated air mass with a 
further trajectory over the cold upwelled water. Due to the 
effects of radiational cooling from the top of the layer, 
cooling from below by the sea surface, and mixing within the 
layer, saturation occurs and fog forms. Once formed, the fog 
layer cools by radiation from the top of the cloud layer to a 
temperature several degrees below that of the sea surface. 

There are two effects that serve to intensify and maintain the 
fog. There is a flux of heat and moisture upward at the sea 
surface. The lapse rate is super-adiabatic because of radia¬ 
tional cooling from above combined with the heating from below. 
Secondly, since the layer remains cooler than the sea surface 
temperature, the evaporative process can continue as the 
evaporation is governed by the sea surface vapor pressure 
gradient. Stage four occurs as the layer deepens to four- 
hundred feet or more. The onshore winds and convective in¬ 
stability due to radiational cooling serve to increase this 


17 






I 




layer depth. In the layer-deepening process, a point is 
reached where the depth of the fog layer below the inversion 
is such (1300 feet at San Diego) that radiational cooling 
can no longer increase the cloud depth. It then becomes a 
stratus regime with the saturated layer lifting off the sur¬ 
face. The entire process usually evolves through a period of 
approximately five days, according to Leipper (1948). 

After the general development scheme was explained, Leipper 
analyzed several fog-related variables and defined three non- 
diurnal indices based on these parameters that provide a 
necessary condition for the formation of fog at San Diego. 
Appendix C is a worksheet taken from the Naval Weather Service 
Facility Local Area Forecasters Handbook for San Diego, Cali¬ 
fornia, that utilizes the Leipper variables as well as indices 
for making a fog forecast. This objective fog forecasting 
method is presently in use at NAS Imperial Beach, NAS North 
Island, cind NAS Miraunar, all located in the San Diego area. 

The variables presently used in the Leipper method are as 
follows: 

1. The base of the inversion of the 1200Z Montgomery 
Field (San Diego, Ca.) upper air sounding. 

2. The highest air temperature adsove the base of the 
inversion, if an inversion exists with a base below 
3000 feet. 

3. The sea surface temperature at 0800 Pacific Staindard 
Time (PST) taken at Scripps Pier (La Jolla, Ca.). 

4. The dew point temperature taken at Lindbergh Field 
(San Diego, Ca.) at 1700 PST on the preceding day. 

These variables are combined to form three indices that all 

must be favorable on a given day to forecast fog assuming 


18 







L- 


little moisture in the upper air. The period of observation 
or "fog day" was defined as the period from 1630PST on one 
calendar day through 1629PST the following day. These indices 
are listed below; 

1. Height of the Inversion Base; to be favorable, the 
height of the inversion base must be less than 1300 
feet [Variable (1) above]. 

2. Temperature Index; if an inversion exists with a 
base below 3000 feet, the temperature index is the 
highest temperature above the inversion base minus 
the sea surface temperature [Variable (2) minus (3) 
above]. Surface air temperature is substituted for 
the highest temperature above the inversion base if 
no inversion exists with a base below 3000 feet. 

To be favorable, the quantity must be greater than or 
equal to zero degrees C. 

3. Moisture Index; this index is defined as the differ¬ 
ence between the dew point temperature and the sea 
surface temperature [Variable (4) minus (3) above]. 

A favorable value is any positive value or any nega¬ 
tive value between zero and minus five degrees C. 

The earliest time-series approach to fog forecasting in 
an open ocean environment was Leipper's (1945) Forecasting 
Summer Fog at Shemya which also suggested application to 
other northern Pacific areas. Ogata and Tamura (1955) con¬ 
ducted a sea (advection) fog study at Ocean Weather Station 
Extra (39N, 153E) that was a continuous time-series examina¬ 
tion of synoptic conditions. Grisham (1973) utilized synop¬ 
tic ship reports in compiling a statistical fog comparison 
that indicated a positive relationship between the occurrence 
of marine advection fog and the temperature and moisture index 
approach. A study by Misciasci (1974) was similarly conducted 
at Ocean Weather Stations Sierra (48N, 162E) and Quebec (43N, 


19 


F 


da 



167W). Misciasci's results indicated that there was indeed 
a possibility of extending a coastal type fog development 
model to the open ocean. For example, the analysis showed 
that although the temperature index did not appear to give a 
good indication of fog and non-fog situations, the moisture 
index did have a positive correlation. Also noted was a rela¬ 
tion between layer thickness and fog duration as well as 
little diurnal dependence shown by the variables. 

An approach to fog forecasting on the West Coast using 
some concepts and some questions was described in a working 
paper by Leipper (1976). A similar approach to offshore fore¬ 
casting was presented by the Calspan Corporation (Rogers, 1981). 
Since 1972, the Calspan Corporation, in conjunction with the 
Naval Postgraduate School (NFS), has been investigating marine 
fog formation offshore along the western coast of the United 
States. Although the research was initially directed at the 
microphysics of the fog problem, their recent applications 
have been in relating micro- and meso-scale processes to 
synoptic-scale events to propose an experimental "decision 
tree" for use in conceptual forecasting of coastal (marine) fog 
(Rogers, 1981). Their research distinguishes between at least 
four different types of offshore west coast (marine) fog 
(Pilie, 1979): 

1. Fog triggered by instability and mixing over warm 
water patches. 

2. Fog developed as a result of lower (thickening) 
stratus clouds. 


20 




3. Fog associated with low-level mesoscale convergence. 

4. Coastal radiation fog advected to sea via nocturnal 
land breezes. 

Of special note are the results concerning temperature. Once 
fog had formed, their results support the observation that 
air temperature within the fog is not dominated by the ocean 
temperature and the air temperature exiting downwind from 
the fog patch was some 0.5 degrees (C) cooler than the air 
entering the upwind fog edge, inferring an on-going process 
of long-wave radiational cooling within the patch. 

In view of the many previously-cited positive indications 
of a consistent fog process existing between the marine coastal 
regime and the open ocean, the present research set an objec¬ 
tive of modifying a working coastal model based on varicibles 
and indices presently in use to the open ocean regime of 
Ocean Weather Station Papa. 

The research approach, in shifting the analysis to an open 
ocean format, expects several factors specifically to change. 
Upwelling and diurnal effects will not be as strong as ob¬ 
served near coastlines and the surface radiational heating, 
seen as a dissipating mechanism, will no longer be a critical 
factor. Advection strength, duration and trajectory should 
become factors that will replace at sea the adiabatic down- 
slope air flow stage of the Leipper coastal model. 

The Eulerian single-station approach utilizing time-series 
studies at OWS Papa will examine data acquired in the months 
of July and August for the years of 1973 and 1977. These 


21 



months are most representative of the summer fog regime based 
on climatology for the North Pacific Ocean. The developmental 
fog model, based on adapted and modified indices, will be 
then compared to cin independent data set for a specific 
period during an additional year at the same station. The 
data base is examined in the next chapter. 


22 


III. DATA 


All data for the present research were originally recorded 
at Ocean Station Papa, and they were obtained through the 
Naval Weather Service Detachment at the National Climatologi¬ 
cal Center, Asheville, North Carolina (Appendix B). Addi¬ 
tionally, the satellite coverage for these two years (1973 
and 1977), although first recorded at Elmendorf A.F.B. Alaska, 
was received from the satellite archives at the University 
of Wisconsin, Madison. 

Figures 8 through 27 are the record of pertinent data 
(variables) for this research. For each of the four months 
examined, the following time series figures are presented 
every twelve hours at OOZ and 12Z: (1) Wind direction in 

degrees, (2) Wind speed in knots, (3) Air temperature, dew 
point temperature and sea surface temperature, all in degrees 
C, and (4) Sea level pressure in millibars. The inversion 
height in millibars is taken from the plot of the daily OOZ 
RAOB. If an inversion does not exist it is specifically noted 
on the time series plot. A key figure is the recorded hori¬ 
zontal visibility which is plotted using a remge code. 

Appendix A is a listing of this code in nautical miles along 
with a corresponding World Meteorological Organization equiva¬ 
lent in kilometers. For this study, fog is represented by 
the codes 94 through 90. The time series figures show the 
fog blocks labeled 94 and less than or equal to 93. As 


23 





described in Chapter I, any visibility less than 5/8 mile 
(1000 meters, or the 94 category or less) is considered open 
ocean fog. Additional variables analyzed in time series 
graphical form but not reproduced in this report are: 

1. maximum temperature of the sounding; 

2. air temperature minus dew point temperature; 

3. dew point temperature minus sea surface temperature; 

4. air temperature minus sea surface temperature; 

5. station reports of fog (all visibility codes less 
than or equal to 98). 

These figures are omitted since the specific form of the first 
four figures did not contribute significantly to the open 
ocean fog development indices, presented in detail in Chapter 
IV, as being critical to a successful forecast model. The 
fifth figure was omitted because the reported range of fog 
codes was much broader th 2 Ln the focus of this study. 

In the 124-day period of the analysis, there were eighteen 
specific cases of open ocean fog lasting twelve hours or more 
that fit the study category of restricted visibility; i.e., 
visibility less than or equal to 5/8 statute miles. Ten of 
the cases were associated with synoptic frontal activity while 
eight were non-frental in origin and maintenance. The months 
of July 1973 and July 1977 each had two occurrences of fog, 
one being frontal and the other non-frontal. August 1973 
and August 1977 had the majority of the fog occurrences; 

August of 1973 had seven (four frontal in nature and three 
non-frontal), and August of 1977 had seven (four frontal and 


24 





three non-frontal). Of the entire four-month period, there 
were nine days of data missing from the record so that a 
115-day record is used for the percentage calculations. 

For excunple, there were 468 hours of "fog case" periods, as 
compared to the 2760 total hours (115 days), resulting in a 
16.9 percent rate of fog for the period. 

The station reports of Ocean Weather Station Papa indi¬ 
cate a much higher incidence rate of fog. For the same period, 
station reports listed some twenty-four cases of fog versus 
the eighteen of this study. Fog was reported present for 
some 660 hours (versus the 468 hours of this study), resulting 
in an occurrence rate of 23.9 percent. This greater amount 
of station-reported fog is due to the reporting procedure 
that includes the "light fog" cases of restricted visibility 
codes of 98 through 95 in addition to the codes (94-90) used 
in this study as the defined cutoff for the presence of fog. 

These frequency rates, 16.9 percent for codes 90 through 
94 and 23.9 percent for codes 90 through 98, compare most 
favorably with computer-assisted climatological marine fog 
frequencies for the eastern North Pacific Ocean. Ocean 
Weather Station Papa, for the years 1963-1974, has a clima¬ 
tological fog frequency of nearly 20 percent for the codes 
90 through 96 (Renard, 1975). 

Although the variables and indices developed for the non- 
frontal open ocean fog appear to be related to frontal fog 
as well, the forecasting of frontal fog is not addressed by 


25 








this study for two major reasons. First, frontal fog as 
compared to the non-frontal case is much shorter in duration. 
In the four-month period examined in this study, nine of the 
ten cases of frontal fog were recorded as persisting for only 
the shortest duration interval (12 hours), whereas in the 
eight non-frontal cases the average duration was for thirty- 
nine hours. Although ideal for a further study, the overall 
economic cuid strategic impact of the frontal fog is reduced 
because of its shorter duration. Secondly, since the frontal 
fog initiation time is tied directly to the large-scale synop¬ 
tic frontal feature,- it is much more easily forecast than the 
non-frontal case. Frontal movement is now routinely and 
accurately observed at sea through satellite coverage, with 
the direction of movement and speed well defined by loop 
analysis. 

Appendix E presents the chronological series of vertical 
temperature soundings plotted from the surface to the 500- 
millibar level on skew T-log p coordinates. Independent 
soundings taken at OOZ daily exist for 108 days of the 124- 
day study periods at OWS Papa. 

In the following chapter, these parauneters are combined 
with adjustments to the Leipper coastal model indices to pro¬ 
vide an objective open-ocean fog model. The synoptic frame¬ 
work is integrated with the fog development model then to 
provide the forecast criteria. 


26 












IV. FOG DEVELOPMENT MODEL AND OPEN OCEAN INDICES 


There are eight pericxis of non-frontal fog in the months 
covered by this study. Although lasting an average of 
thirty-nine hours, the fog sequences cover durations of 
twelve to one-hundred eight hours. 

As previously described, the Leipper coastal model is 
based on a four-stage development. In subsequent studies, 
Peterson (1975) and Evermann (1976) present coastal models 
of three and five stages respectively. A key relationship 
between these three development models of coastal fog is 
the similarity of chronological physical development. The 
general development model is an idealized chronology of evo¬ 
lution focusing on key physical steps which, combined, com¬ 
prise a typical physical sequence. In nature, the sequence 
may be retarded, accelerated, interrupted, etc,, by fluctua¬ 
tions of the larger scale circulation. 

The eight widely-varying periods of non-frontal fog in 
this study are each seen as at least a portion of a single 
idealized development model. Stage I of the open ocean model 
occurs with the positioning of the mid-Pacific subtropical 
high in a location generally to the southeast of the forecasting 
station, OWS Papa (Fig. 2), so that the station falls under 
the synoptic influence of the high. The surface sea level 
pressure at OWS Papa begins to rise as the high pressure system 
intensifies and increases in magnitude under the influence of 


27 




the long i-'ave ridge. In seven of the eight fog sequences 
(Table III) tnis rise in pressure is the initiating feature. 

In one case (Case 2) a major low pressure system is located 
just to the south of the recording station, and it will be 
treated as an anomalous case. Its development, though having 
very similar variables and indices, is not the subsidence/ 
advection sequence that seems to be typical for this area. 

Table I shows that in each of the eight cases, the period 
of occurrence of fog is preceded by a steady increase in the 
surface pressure. The period of intensification was found to 
last from twelve hours to as long as five days with an average 
period of intensification lasting some two and a half days 
(1.12 day standard deviation). The pressure increase during 
this intensification period averaged 4.6 millibars (3.13 mb 
standard deviation) per day and it is the first indication of 
a subsidence-dominated regime that contributes to the fog for¬ 
mation at OWS Papa. Close analysis of the daily RAOB history 
(Appendix E) further confirms subsidence prevailing over the 
station in this initiating stage. The subsidence effectively 
creates a near-surface layer with a warm, relatively dry upper 
air mass overlying a shallow cooler and more moist marine 
layer. 

A second major definitive factor in Stage I in each of 
the eight cases, is a steady increase in the dew point tem¬ 
perature prior to the formation of the open ocean fog. The 
average period of rising dew point temperature was 2.2 days 


28 






(0.82 day standard deviation), very nearly the period of the 
sea level rise in pressure (2.5 days). The magnitude of 
increase in dew point temperature ranged from 1.5°C to 5.5®C 
for specific periods (Table II) and the average rate was a 
1.5°C rise per day with a standard deviation of 0.67°C. There 
was a steady increase in the water content per unit volume 
of the marine layer, indicated by this rise in dew point 
temperature. This is a result of advection taking place 
simultaneously with the subsidence noted earlier. 

In all seven cases examined in this model (Case 2 omitted) 
there was a relatively long uninterrupted flow of low level 
air from the south due to the circulation on the western side 
of a high pressure system that dominates the recording station 
Recorded wind directions ranged from 120°T to 290°T for the 
seven cases depending on the exact orientation of the high 
pressure system in relation to OWS Papa. Additionally, the 
wind speeds remained relatively steady (3 to 15 knots), prior 
to the formation of fog. 

Due to the station variables' dependence on orientation 
relative to the position of the subtropical high and the 
magnitude of the winds, this open ocean non-frental fog is 
termed an advection fog. In the advection process, the sur¬ 
face air mass is advected northward. In the overlying air 
mass, measurements indicate a temperature higher than that of 
the air initially over OWS Papa at the outset of Stage I. 

Thus, the air temperature is seen to rise prior to the onset 


29 







of fog for each of the cases. Additionally, in each of the 
eight cases the low layer of marine air is found to be warmer 
than the sea surface in Stage I as well as during all or part 
of the period that fog was actually present. Thus in the 
advection process, there is cooling of the air mass from 
below by the sea surface as well as turbulent mixing due to 
the low level winds. These factors, in conjunction with the 
subsidence-induced layer capping the low level air mass, 
serve to modify the trapped marine layer and to lead towards 
saturation. 

Stage II is initiated at saturation with the advection 
fog forming as a very shallow layer with the inversion very 
near the surface. Over the average duration (39 hours) of 
fog present at the recording station, the surface layer thick¬ 
ened slightly as the measured inversion height increased al¬ 
though not one of the eight cases examined in the study had 
inversion heights that were recorded over 1000 feet during 
the period that fog was present. The mechanism for maintenance 
and thickening of the fog in this stage appears to be radiational 
cooling from the top of the layer as well as cooling from 
below by the sea surface. Throughout this stage, winds remain 
steady in magnitude, averaging nearly 10 knots, auid direction. 
The rise in dew point temperature and surface pressure, measured 
prior to the saturation stage, ceases during Stage II. 

Stage III commences with the dissipation of the existing 
fog. In each of the cases, dissipation was due to frontal 


30 



effects. Within six to twenty-four hours after the recorded 
end of the fog periods, there was a passage of either a 
cold or occluded front over the OWS Papa area. The average 
period of time between fog termination and frontal passage 
was 13.5 hours. Tables I and II show the rates of decrease 
for sea surface pressure and dew point temperature that 
accompany the dissipation stage for each fog case. The sea 
surface pressure decreased for an average of 1.6 days (0.49 
day standard deviation) in the post-fog stage at an average 
rate of 5.2 millibars per day (2.0 mb standard deviation). 
This decrease is very consistent with the frontal passage 
initiating the fog dissipation. The dew point temperature 
decreased over an average period of 1.7 days (standard devia¬ 
tion of 0.99) at an average rate of 1.9“C per day (1.67“C 
standard deviation). This is very consistent with the 1.6 
day average period of pressure decrease. The dew point 
temperature decrease is a direct result of the post-frontal 
clearing and decrease in humidity. 

The post-frontal period also was indicated by a 
decrease in the air temperature and either an elevated (above 
3000 feet) or non-existent inversion. An analysis of the 
daily synoptic charts for this stage showed a decrease in 
the intensity eind magnitude of the sub-tropical high. In 
decreasing in magnitude, the high pressure system's influence 
retreated in a southeastward direction away from OWS Papa as 
the short wave trough and associated frontal activity moved 


31 










over the recording station area. The entire three-stage 
process lasted an average of some six days. 

With the open ocean fog development model described, the 
Leipper coastal fog model variables and indices were examined 
in an attempt to modify them based on the development model 
so that an accurate set of forecast indices would be estab¬ 
lished for the open ocean case. The variables which are 
recommended for use in the open ocean forecast model are as 
follows (all data recorded at OWS Papa): 

1. The base of the inversion (in mb) as recorded from 
the OOZ RAOB. 

2. The highest air temperature above the base of the 
inversion if an inversion exists with a base below 
3000 feet, measured at 002 in ®C. 

3. The sea surface temperature measured at OOZ and 12Z 
in “C. 

4. The dew point temperature measured at OOZ and 12Z in 
»C. 

5. The surface wind direction in degrees true and wind 
speed in knots at OOZ and 12Z. 

These five parameters, when combined as described below, 
form four indices that all must be favorable on a given day 
to forecast fog. The period of observation is defined as 
from 1400 local (OOZ) the preceding day through 1400 local 
the recording day with the forecast valid for the next twenty- 
four hours. The open ocean fog forecast indices (Appendix D) 
are; 

1. Height of the inversion base; to be favorable the 

height of the inversion base at OOZ must be less than 
1000 feet. 


32 







2. Temperature index: the temperature index is the 
highest temperature above the inversion base at OOZ 
minus the sea surface temperature (002) recorded 
the previous day. To be favoradjle the quantity must 
be greater than or equal to 0®C. 

3. Moisture index: the moisture index is the dew point 
temperature (OOZ) minus the sea surface temperature 
(OOZ) recorded the previous day. To be favorable 
the quantity must be greater than or equal to a 
negative 0-5®C. 

4. Advection index: the advection index is a combination 
of the wind direction auid wind speed at OOZ. To be 
favorable the winds must be between 120'’T and 290‘’T 
amd greater them three knots but not greater than 

15 knots. 

Index accuracy for the advection fog periods of this study 
is excellent. The inversion height index and advection index 
were favorable for each of the 276 hours of fog present (Table 
III) in cases developing past Stage I of the model. The 
temperature index was favorable for all but twelve of the fog 
hours or was correct more than 95 percent of the time. The 
moisture index was correct (favorable) all but 48 hours of 
the advection fog periods or more than 82 percent of the time. 
When combined into the forecast model, the indices correctly 
indicated the presence of every fog case and were completely 
favorable for over 78 percent of the total durations of the 
fog periods. 

The indices not only successfully delineated the presence 
of fog, but were successful in indicating the initiation and 
cessation of fog periods or the limits between Stages I and 
II and Stages II and III of the development model as well. 

In every case, at least one index was unfavorable in the 


33 






r 


24-hour period prior to the formation of fog, and, in every 
case but one, at least one index was unfavorable in the first 
twenty-four hour post-fog stage (Table IV). Since only one 
index is needed to indicate a no-fog case, the open ocean 
indices had an overall success rate of 93 percent of delineat¬ 
ing the no-fog periods of Stages I and III as well as the 
success in indicating the Stage II fog presence. 

The major modifications made to the Leipper coastal model 
are summarized: 

a. Favorable inversion height is 1000 feet in this open 
ocean model, down from the 1300 feet of the coastal 
model. 

b. The timing of recording observations for the open 
oceam model is uniformly OOZ and all readings are 
made at OWS Papa. 

c. Winds are a factor in the open ocean case as the fog 
is of the advection type. 

d. The temperature depression between dew point tempera¬ 
ture cuid sea surface temperature is only -0.5®C 
instead of the full -5.0°C of the coastal model. 

This is likely due to the fact that, in the open 
ocean case, the surface layer is always in contact 
with the sea surface and is relatively moist at all 
times. The large temperature depression of the 
coastal and inland regions is not observed. 

A single case of open ocean fog is now presented in rela¬ 
tion to the theoretical development model. Figures 3 through 
7 support this analysis as they are the chronological synoptic 
history covering the three stages of development for fog 
Case 6 (Figures 23-27, and Tables I-IV). At OOZ on 10 August 
1977, the synoptic overview shows that the siibtropical high 
is relatively weak (1020 mb) and is pressed up against the 
North American continent. Dominating the OWS Papa region, in 


34 








fact dominating the entire eastern North Pacific, is a large, 
intense low pressure system (996 mb). Although the period 
of fog initiation was 48 hours away, there is already an 
upward trend in the dew point temperature that began at OOZ 
09 August. The development model stage I begins at OOZ 10 
August as the rising dew point temperature couples with a 
rising sea surface pressure when the subtropical high begins 
an intensification process and the low pressure system to 
the west of the station begins to dissipate. Over the next 
48 hours prior to fog formation, the sea surface pressure 
over the recording station will intensify some 22 millibars. 
By OOZ 11 August, the synoptic overview shows the low has 
decreased in intensity some 6 millibars and has split into 
two closed circulations. The high has intensified auid grown 
greatly in magnitude. The recording station is now experi¬ 
encing a long, low-level air flow northward. By 12Z 11 
August a steady sea level air temperature rise is noted at 
the station as the advection process is initiated, moving a 
warmer air mass northward. At OOZ 12 August, the high center 
has moved well northward and has intensified just to the east 
of the recording station. The rapid strong intensification 
of the high pressure system with the associated strong sub¬ 
sidence has caused the inversion height to move down to less 
them 400 feet above the ocean's surface by the beginning of 
Stage II at OOZ 12 August. At this point of saturation, the 
fog forms emd lasts for a 48-hour duration. Sea surface 
pressure, dew point temperature and air temperature all have 


35 





reached their maximum values by OOZ 13 August and have begun 
to decrease. By 12Z 13 August the fog has ceased since a 
cold front associated with the low center located in the 
Aleutian Islands is in close proximity to the west. The low- 
level inversion of Stage II became an unfavorable index as 
it rose to an elevation of over 2500 feet due to the frontal 
effect. The decrease in dew point temperature, air tempera¬ 
ture, and sea surface pressure continued through Stage III; 
the decrease in dew point temperature lasting 48 hours after 
Stage II was over. The OOZ 14 August synoptic overview shows 
the subtropical high dissipating in magnitude to the east of 
the recording station as the cold front has passed and the 
low pressure to the west has intensified. By this time, the 
moisture index had become unfavorable. By 12Z 14 August, the 
wind index and temperature index had become unfavorable and 
the OOZ sounding on 15 August showed the inversion height 
unfavorable. 

The entire sequence of the three stages of development for 
this case 6 lasted some five days. The open ocean model accur¬ 
ately fits this example case amd the fog sequence has been 
accurately delineated by the open ocean indices. Additionally, 
pre-fog and post-fog variable tendencies correctly indicated 
formation and dissipation. 

Analysis of DMSP satellite coverage for this specific case 
6 effectively complemented but was not essential in the open 
oceam forecast method developed. NOAA-5 visible and IR 


36 






coverage on 9, 11, 12, and 15 August 1977 was examined for 
the Gulf of Alaska area. The IR coverage on 9 August is 
dominated by the intense low pressure system over the North¬ 
west Pacific. The IR coverage on 11 August definitely showed 
the intensification of the subtropical high in the eastern 
Gulf by a widespread area of clearing. The low pressure to 
the west of the OWS Papa is much reduced in size and patches 
of low level stratus (possibly fog) are seen in areas near the 
recording station. Satellite coverage similar to that on 
this specific day may have excellent future use in delineating 
fog patch extent as the low level stratus is bordered by large 
clear areas in the anticyclonic regime. The IR satellite 
coverage on 15 August nicely showed the cold front activity 
that had earlier caused the dissipation of the advection fog 
at OWS Papa. Although there are problems in interpreting 
differences between low-level stratus and surface fog, satellite 
support of the open ocean fog development model does effec¬ 
tively delineate fog patch size associated with the high 
pressure system. Additionally, frontal activity is accurately 
located via satellite data, thus cessation times of the fog 
periods may be more accurately predicted since all cases of 
advection fog in this study were dissipated by frontal activity. 

Satellite coverage for Case 7, a 108-hour period of fog 
during the month of August 1977, was examined. Infrared 
coverage on 15 August, some twenty four hours prior to the 
fog initiation, indicated that OWS Papa was in a clear area 


37 


s 








with high level clouds to the west and widespread low level 
stratus to the south. Towards the east and the center of 
the high pressure system, there was general clearing, while 
to the north of the station was another large stratus patch 
extending nearly to the southern Alaska coast. A 20 August 
visible satellite photo during the latter part of the fog 
period indicated that OWS Papa was obscured by cloud cover. 
Because it was a visible satellite photo, cloud height and 
type were less easily discernible. Clear patches were clearly 
identifiaible on the visible photos, thus maximum possible fog 
patch sise was accurately delineated. A large clear area was 
present north of OWS Papa towards the center of the high pres¬ 
sure system. There appeared to be an extensive coastal fog 
bank present in the Gulf, just south of the eastern portion 
of Alaska. Again, the patch size is accurately shown on the 
satellite display by the presence of no-cloud areas surround¬ 
ing the cloud mass. The high pressure circulation itself is 
well defined and the curvature of flow is easily seen in the 
cloud pattern. To the west of the recording station, a major 
frontal band extends in a north-south direction as the low 
pressure system to the west intensifies. The leading edge of 
the front is shown by the change from an open cellular regime 
to that of the high cirrus bands of the front. Frontal width 
is additionally quite visible. On 21 August the infrared 
coverage is seen some twelve hours after the dissipation of 
fog at OWS Papa. The coverage clearly shows the front in 











close proximity directly to the west of the station. Frontal 
activity and frontal width are defined by the high clouds seen 
in the IR image. Low level clouds, possibly a stratus or 
fog deck, are seen to the east of the recording station, and 
they remain widespread, extending eastward nearly to the con¬ 
tinental border. The satellite coverage investigated served 
to enhance the synoptic display as well as the open-ocean fog 
development model in this case as well. While the satellite 
data were not critical for the forecast indices, they did serve 
to show the possible extent of known fog patches as well as 
specifically delineating non-fog areas. 

An independent data set for the month of July 1975 will 
be examined in the next chapter. The open oceain variables 
and indices as well as noted trends in the 1973 and 1977 
years will be analyzed and the accuracy of the method itself 
determined. 


39 



V. INDEPENDENT TEST 

The open-ocean development model was tested on an inde¬ 
pendent data set. The Ocean Weather Station Papa meteorologi¬ 
cal records for the month of July 1975 were chosen. The 
significant atmospheric variables were plotted along with the 
fog indices (Figures 28-32). Prior to analysis of the ob¬ 
served horizontal visibilities# the indices were examined 
according to the forecast method (Appendix D) to determine 
periods of a favorable fog forecast. The forecast periods 
determined from the indices were then compared to the periods 
of observed advection fog. 

On three occasions, all of the indices were favorable in 
the forecast model, and, on two occasions, all of the indices 
were favorable except for the wind direction of the advection 
index. For the July 1975 period, there were five periods of 
observed advection fog that correlated in timing and duration 
with the favorable periods predicted by the open-ocean fore¬ 
cast model. 

For the month of July 1975, there were a total of nine fog 
cases, four frontal and five of the advection type. The 
seven-hundred forty-four-hour observation period had restricted 
visibilities for one-hundred thirty two hours or 17.7% of 
the total recording period. This percentage is consistent 
with the fog development data for 1973 and 1977 (16.9%) as 
well as climatology (20%) . 


40 




stage II or the fog formation times of the five advection 
cases are listed: 

1. Case I - 12Z July 1 

2. Case II - 12Z July 2 

3. Case III - OOZ July 8 

4. Case IV - 12Z July 9 

5. Case V - OOZ July 15. 

As noted in the fog development model cinalysis, the trends 
for an increase in the dew point temperature prior to advec¬ 
tion fog formation and decrease after the fog period were 
consistent for four of the five independent advection fog 
cases. Similarly, the trend for a decrease in sea level 
pressure in the post-fog Stage III was consistent for all 
five of the advection fog cases. Least consistent with the 
fog development model was the trend for an increase in the 
sea level pressure during Stage I of the fog development model. 
In three of the five cases there was an increase in the sea 
level pressure immediately prior to fog formation, while in 
the other two cases, OWS Papa was already in a high-pressure- 
dominated regime at the outset of Stage I. 

The test on the independent data set verified not only 
the open-ocean forecast method but the three-stage fog develop¬ 
ment model as well. Detailed analysis of the synoptic reports 
at six-hour intervals for this time period indicated a similar 
development scheme and dissipation process by frontal activity. 

If the wind direction part of the advection index were 
expanded sixty five degrees (from 290 tc 355 “T), the fore¬ 
cast model would have accurately predicted each of the five 


41 





advection fog cases. Without the index corrected, the pre¬ 
diction accuracy was 66.7% or correct for 48 of the 72 hours 
of advection fog. Further studies may expand the limits for 
this part of the advection index to better account for flow 
in the northeast quadrant of the subtropical anticyclone 
that is consistent with the first two stages of the open-ocean 
development model. 

As well as accurately delineating the actual advection 
fog periods of the independent test case, the pre-fog period 
(Stage I) and the post-fog period (Stage III) were correctly 
delineated in all cases by at least one unfavorable index 
within the twenty-four hour period immediately prior to fog 
formation and within the first twenty-four hour period immed¬ 
iately after fog dissipation. This result implies the actual 
physical mechanics of the fog stages were well represented 
in the chosen indices. 

Although the size of the independent data set was rela¬ 
tively small (25% of the development model data set), consistent 
fog frequencies with earlier work serve to substantiate and 
support the positive nature of the results for a single sta¬ 
tion forecast model. Extensive further testing of the fore¬ 
cast model and related indices at Ocean Weather Station Papa 
will serve to increase the accuracy of the existing indices, 
such as modification in the advection index as more fog cases 
are subjected to the model. Application of the development 
model and forecast method to data of other professionally 


42 







VI. CONCLUSIONS AND RECOMMENDATIONS 


The accurate forecasting of open-ocean fog is of vital 
concern to any ocean-going operation. This study has made 
a strong step in the direction of delineating accurate fog 
occurrence for the summer months at OWS Papa based on atmos¬ 
pheric variables and significant fog-identifying indices. 

The success of the Euleriaui single station approach using this 
objective forecast model/ when coupled with a synoptic over¬ 
view and satellite coverages, presents a most accurate analy¬ 
sis of the fog situation at the recording station. The 
successful coastal fog model was indeed adaptable to the 
open-ocean case although several different physical mechanisms 
were found to be at work, most notably the dissipation caused 
by passage of cold fronts. 

There are specific recommendations applying to this approach. 
By further extending this successful model to other stations 
of professionally-measured marine stations, an evaluation can 
be made as to the applicability of this model and to the 
consistency of processes at work in the open ocean. These 
further studies would serve to correlate as well as to refine 
the presently-developed variables and indices. 

At Ocean Weather Station Papa, this model should be further 
tested and refined with data from additional years. Specif¬ 
ically the indices should be continually examined and smoothed. 
The moisture index, with further research and adaption of its 


44 






measured temperature depression, should become more accurate 
than the present 82%. 

Although this model was developed for the high fog fre¬ 
quency summer months, further research could be directed 
toward the development of an all-season model. 

Frontal fog was not included in this study as it was of 
short duration (90% of the cases 12 hours or less). However 
it did occur at a higher frequency, thus there is a need for 
research in this area. A first look at data in these frontal 
fog periods shows mcuiy strong correlations with the non-frontal 
cases. The moisture index was consistent and also quite 
accurate. The advection index, as well as the influence of 
subsidence on inversion height, did not play a major role in 
the frontal fog cases. 

Examination of specific segments of the fog development 
model, such as the subsidence mechanism and magnitude within 
intensifying subtropical anticyclones, would certainly con¬ 
tribute to the accuracy and understanding of the processes in 
the first stage of the open-ocean model. 

Research into the microphysics of the problem is a pre¬ 
requisite to the complete physical understanding of the pro¬ 
cesses interacting in the fog formation and dissipation steps. 
Early work by Businger (1973) and Wyngaard (1973) on turbulent 
transfer and turbulence over land areas in the atmospheric 
surface layer have been extended to the "over ocean" applica¬ 
tion by Davidson with the Environmental Physics Group of the 










Naval Postgraduate School, Monterey, California (1978). 
Additionally their research has provided a moist forced- 
entrainment model for mixed layer depth changes in the 
atmosphere that appears to have excellent possibilities for 
application to future fog prediction capability (Davidson, 
1980) . 


More accurate understanding of subtle changes in the 

microphysics of the oceanic layer such as in the temperature 

2 

structure ftinction parameter, , and the humidity structure 

2 

function parameter, , may only enheuice the accuracy of 
open-ocean fog forecasting. 

In the long term, successful objective forecast models 
will be integrated with the microphysics of the atmosphere 
as data stations and recording accuracy allow. In turn, this 
product will be blended with climatology, enhanced by the 
model output statistics approach as computer capability is 
increased, to present a consistent, accurate, easily-transmitted 
prediction product. 

Of immediate impact is the fact that in the region of 
Ocean Weather Station Papa for the summer period, there now 
appears to be an accurate method for a single station to fore¬ 
cast fog formation and dissipation periods. All four of the 
open-ocean indices are within the measurement capability of 
ciny U.S. Naval Vessel. When recorded and plotted daily and 
when combined with a shore-based synoptic report and satellite 
links, they together will allow any single unit or task group 


46 





T 


to predict with some confidence one of the most dangerous 
atmospheric phenomena to any open-ocean operation, the long 
duration advection fog with restricted visibilities of less 
than one kilometer. 


47 









APPENDIX A 


ABRIDGED VERSION OF INTERNATIONALLY USED PRESENT WEATHER AND 
VISIBILITY CODES (UNITED STATES DEPARTMENTS OF COMMERCE, 
DEFENSE, AND TRANSPORTATION, 1969) 


Present Weather Present Weather 


Code 


Code 


Value 

Definition 

Value 

Definition 

00-03 

Characteristic change 

30-39 

Duststorm, sandstorm. 


of the state of the 


drifting or blowing 


sky (cloud) during 


snow. 


the past hour. 

40 

Fog at distance, but 

04-09 

Haze, dust, sand, or 


not at station during 


smoke. 


the past hour (visi¬ 

10 

Deep light fog. 


bility less than 1 km). 

11-12 

Shallow heavy fog. 

41-49 

Deep heavy fog at the 

13-17 

Lightening, thunder. 


time of observation 


or precipitation 


(visibility less than 


within sight, not 


1 km) . 


reaching the ground. 

50-59 

Drizzle, or drizzle 

18-19 

Squall(s), funnel 


and rain. 


cloud(s) during the 

60-63 

Slight to moderate 


past hour. 


rain. 

20 

Drizzle during the 

64-65 

Heavy rain. 


past hour. 

66 

Slight freezing rain. 

21-23 

Rain, snow or rain 

67 

Moderate or heavy 


and snow during 


freezing rain. 


the past hour. 

68 

Slight rain or drizzle 

24 

Freezing drizzle 


and snow. 


during the past 

69 

Moderate or heavy rain 


hour. 


or drizzle and snow. 

25-27 

Shower(s) during the 

70-79 

Solid precipitation 


preceding hour. 


not in showers. 

28 

Fog during the past 

80-89 

Showery precipitation 


hour. 


or precipitation with 

29 

Thunderstorm during 


current or recent 


the past hour. 


thunderstorms. 


Visibility Ship 

Station 

W 

90 

Less than 50 m Code 

Figure 

Plot 

91 

50-199 m 90 

<50 yds 

92 

200-499 m 91 


50 yds 

93 

500 m - 0.99 km 92 


200 yds 

94 

1 - 1.99 km 93 


1/4 NM 

95 

2 - 3.99 km 94 


1/2 

96 

4 - 9.99 km 95 


1 

97 

10 - 19.99 km 96 


2 

98 

20 - 49.99 km 97 


5 

99 

50 km or more 98 


10 


99 


25 


48 



APPENDIX B 


Time Period 


July/August 

1973 


July/August 

1977 


Computer- 

** 

National 


DATA SOURCES 


Cfcservaticn Site 

Data Avcdlable 

Sources 

Ocean Weather 
Station Papa 

Surface Observa¬ 
tions (microfilnv' 
computer-tape) * 

NWSD Asheville, 
N.C. of (NCC)** 


RACES (microfilm) 

NWSD Asheville, 
N.C. of (NOC) 


Sea Surface Tatp- 
eratinre (micro- 
filin/ccirputer- 
t^) 

NWSD Asheville, 
N.C. of (NCC) 


Satellite Coverage 

Space Science 
and Engineering 
Center 

University of Wis¬ 
consin, Madison 

Ocean Weather 
Station Pc^)a 

Surface Observa¬ 
tions (microfilnv' 
conputer-t^e) 

NWSD Asheville, 
N.C. of (NCC) 


RACBS (microfilm) 

NWSD Asheville, 

N.C. of (NCC) 


Sea Surface Taip- 
erature (micro- 
filnv/ccitputer- 
tape) 

NWSD Asheville, 

N.C. of (NCC) 


Sea Surface Tatp- 
erature 

Oceanogrcphic ob¬ 
servations at 

Ocean Station P 
Institute of Ocean 
Sciences, Patricia 
Bay, B.C. 


Satellite Coverage 

Space Science and 


Engineering Caiter 
university of Wis- 
ccxisin, Madison 


tape - Common marine tape format deck 128 
Climatological Center 


49 








APPENDIX C 


EXAMPLE OF LEIPPER FOG FORECAST PARAMETERS AND INDICES 
A Worksheet from the Naval Weather Service Facility, San Diego. 


PARAMETERS 





a) 

Base of inversion on 12GMT 
(SAN) 

sounding 

— 

FT 

b) 

Highest air temperature 
of inversion (SAN) 

above base 

Ta®_ 

C 

c) 

Sea temp at 0800LST on preceding 
day (Scripps) 

Tw=_ 

C 

d) 

Dew point temp at 1630 LST 
preceding day (NZY) 

on 

TD 1630= 

P 

C 

e) 

Mixing ratio at 10,000* 
12GMT SAN sounding 

on 

the 

MR= g/kg 

INDICES 


FAVORABLE 

UNFAVORABLE 

a) 

Base of inversion® FT 


Below 

1300' 

Above 

1300' 


b) 

Ta minus Tw = C 


Above 0 

C 

Below 0 

C 


c) 

TD 1630 minus Tw= C 

p - 


Above -5 

C 

P low -5 


d) 

MR at 10,000' = g/kg 


Less than 
3.5 g/kg 

More than 
3.5 g/kg 



NOTE: All must be favorable for forecast to become favorable 

to indicate fog formation. 

3) Forecast amd Verification 

a) Forecast valid for period 1800-0600 LST following time 
of observation used in computations. 

b) Verification (circle one) 

Visibility Above 2 miles Below 2 miles 

Obstruction to vision Fog Ground Fog Haze None 

COMPUTED BY: _ 


DATE ; 


50 




APPENDIX D 

OPEN OCEAN FOG FORECAST VARIABLES AND INDICES 


Parameters 

1. Inversion Height (mb) (OOZ) 

2. Wind Direction (“T) and Wind Speed (knots) (OOZ, 12Z) 

3. Sea Surface Temperature (“C) (OOZ, 12Z) 

4. Dew Point Temperature (°C) (OOZ, 12Z) 

5. Maximum Temperature (air) above the Inversion (®C) (OOZ) 


Indices 


Favorable 


Unf avorcOole 


1. Inversion Height (OOZ) 

2. Temperature Index- Maximum 
temperature above inversion 
(OOZ) minus sea surface temp¬ 
erature at (OOZ) the preceding 
day. 

3. Moisture Index - Dew point 
(OOZ) minus the sea surface 
temperature at OOZ the 
preceding day. 

4. Advection Index - Wind 
direction (’T) auid Speed 
(knots) 


Below 1000 ' 

Values 
greater 
than O^C 


Values 
greater 
than -0.5°C 


Winds 

greater than 
or equal to 
3, less than 
or equal to 
15;120-290°T 


Above 1000' 

Values less 
than 0®C 


Values less 
than -0.5°C 


Winds 
greater 
than 15 knots 
291-119°! 


Note: 1. All must be favorable to forecast fog at the 

recording station. 

2. Forecast is valid for 24 hours (1400 local of the 
recording day through 1400 local of the following 
day) . 


51 





DAILY RAOB PLOTS 



July 1973 _ _ _ 4 July 1973 










July 1973 








DIAGRAM 



10 July 1973 - 12 July 1973 






diagram 



July 1973 











DIAGRAM 



18 July 1973 - - 20 July 1973 







(SELECT PORTION) 



57 


23 July 1973 _ _ _ 25 July 1973 




(SELECT PORTION) 






SKEW T. log p DIAGRAM 



July 1977 -9 July 1977 




(DOO) 

SKEW T, log p DIAGRAM 



10 July 1977 - 12 July 1977 












61 


14 July 1977 - 16 July 1977 




USAF 




SKEW T, log p DIAGRAM 
(SELECT PORTION) 



23 July 1977 - 25 July 1977 






(SELECT PORTION) 









SKEW T. log p DIAGRAM 
(SELECT PORTION) 



65 











JSAF skew T. log p DIAGRAM 
(SELECT PORTION) 





DIAGRAM 



68 


15 August 1973 _ _ _ 17 August 1973 





DIAGRAM 









diagram 



70 


1973 _ - - 26 August 1973 










DIAGRAM 













SKEW T. log p DIAGRAM 



1977 





DIAGRAM 



73 


1977 






SKEW T. log p DIAGRAM 
(SELECT PORTION) 












1977 






SKEW T. log p DIAGRAM 
(SELECT PORTION) 



20 August 1977 








SKEW T. log p DIAGRAM 



1977 











25 August 1977 - 27 August 1977 

26 August 1977 - 28 August 1977 





79 


1977 - 31 August 1977 










DIAGRAM 












81 


July 1975 - - 8 July 1975 



82 


10 July 1975 - - _ 12 July 1975 









14 July 1975 







(SELECT PORTION) 



84 







DIAGRAM 



25 July 1975 





(SELECT PORTION) 







87 


30 July 1975 - 31 July 1975 









SURFACE PRESSURE CHANGE 



AVERAGE 2.5 11.5 4,6 mb 1.6 8.4 - 5.2 mb 










































POINT TEMPERATURE CHANGE 



AVERAGE 2.2 3.2 1.5 C 1.7 3.2 





































































TABLE III 


SUMMARY OF THE NON - FRONTAL FOG CASES. 


CASE 

YEAR 

MONTH 


DURATION 
IN HOURS 

1 

73 

JUL 

25/002 

12 

2 ^ 

77 

JUL 

13/002 

24 

3 

73 

AUG 

5/002 

48 

4 

73 

AUG 

10/12 2 

48 

5 

73 

AUG 

20/12 2 

12 

6 

77 

AUG 

12/002 

48 

7 

77 

AUG 

16/122 

108 

s” 

77 

AUG 

26/12 2 

12 


a) Does not fit open ocean model 

Does not develop p-'.st Stage I of model 

1) Total Non-frontal Fog Duration - 312 Hours 

2) Total Advection Fog Duration - 276 Hours 

3) Average Non-Frontal Fog Duration - 39 Hours 

4) Average Advection Fog Duration - 46 Hours 


90 







































LIMITING INDICES FOR EACH FOG CASE. 







00 

INV HT 

TEMP 

AND 

MOIST 

INDEX 

INV HT 

TEMP 

AND 

MOIST 

INDEX 

h- 

INV HT 

WIND 

DIRECT 

INV HT 

WIND 

DIRECT 

<0 

INVHT 

TEMP 

INDEX 


in 

INV HT 

TEMP 

AND 

MOIST 

INDEX 

INV HT 

TEMP 

AND 

MOIST 

INDEX 


INV HT 

TEMP 

AND 

MOIST 

INDEX 

INV HT 

TEMP 

AND 

MOIST 

INDEX 

CO 

MOIST 

INDEX 

TEMP 

AND 

MOIST 

INDEX 

CM 

DIFFER 

MOi 

DEVELC 

ENT 

)EL 

PMENT 

- 

IH ANI 

INV HT 

MOIST 

INDEX 

— 


UNFAVORABLE 

INDICES FOR 

THE 24 HOUR 
PRE- FOG 

PERIOD 

UNFAVORABLE 

INDICES FOR 
THE 24 HOUR 
POST - FOG 

PERIOD 


91 



















Figure 1. Ocean Station Papa (50N,14 5VJ) and General 
Sea Surface Circulation 















145W 130W 






naval POSTMAOUATE school MONTEREY CA F/0 «/2 

AN OPEN-OCEAN MARINE FOR DEVELOPMENT AND FORECAST MODEL FOR OCE—ETC(U) 


. . i JUN SI R L CLARK 

MPgAH-Wl-OOB _ 














































Figure 6. Surface Synoptic Display, 13 August 
















HORIZONTAL VISIBILITY ^SEE APPENDIX A) vs. 














DEW POINT TEMPERATURE CO vs. DAY 

























Figure 11. July 1973 Wind Direction and Wind Speed 












Figure 13. July 1977 Inversion Height and Horizontal Visibil 







































Figure 17. July 1977 Temperature Index and Moisture Index 















HORIZONTAL VISIBILITY ^SEE APPENDIX A) vs. DAY 















Figure 21. Aug. 1973 Wind Direction and Wind Speed 









>• 



■< 



Q 


> 

(3 


< 

Z 


o 




o 


«A 

UJ 


> 

o 


o 

• 

UJ 

q: 



a 

1 

z 

Ul 


UJ 

z 

X 

u 

o 

H* 

CQ 

CO 

N 

< 

QC. 

UJ 

> 

O 

o 

K 

§ 

z 

H 

< 

< 

UL 

UJ 

UJ 

CO 

> 

cc 

< 

o 

o 

UJ 

03 

h- 

oc 

< 

< 

< 

LU 

cr 

tr 


UJ 

0. 

Q 

D 

s 

UJ 

f- 

UJ 

Q 

< 

h- 

< 

Q£ 

UJ 

O 

r 

u; 

a 

(0 

S 

< 

u. 


UJ 

q: 


f- 

D 


S 

CO 


D 

< 


s 

UJ 



CO 


X 

1 

1 

i 


< 


s 



POINT IN *C vs. DAY 

ST AT 00 2 THE PRECEDING DAY MINUS 0.5 
























Figure 26. Aug, 1977 Wind Direction and VJind Speed 


















HORIZONTAL VISIBILITY ^SEE APPENDIX A) vs. DAY 







DEW POINT TEMPERATURE ('C ) vs. DAY 











LIST OF REFERENCES 


Aldinger, W.T., 1979: Experiments on Estimating Open Ocean 
Visibilities Using Modf I Output Statistics . M.S. Thesis, 
Dept, of Meteorology, Naval Postgraduate School, Monterey, 
Ca., 81 pp. 

Beardsley, J.W., and Leipper, D.I\, Fog on the Central Cali¬ 
fornia Coast for 1973: Analysis of Trends , M.S. Thesis, 
Naval Postgraduate School, Monterey, California, March 
1976. 

Businger, J.A.: 1973, Turbulent Transfer in the Atmospheric 
Surface Layer , Workshop on Micrometeorology, D. Haugen, 
Editor, American Meteorological Society, 67-98. 

Calspan Report No. 6673-M-l, An Investigation of Marine Fog 
Forecast Concepts , by C.W. Rogers, E.J. Mack, R.J. Pilie 
and B.J. Wattle, January 1981. 

Davidson, K.L., Houlihan, T.M., Fairall, C.W., and Schacher, 
G.E.: 1978, Observation of the Temperature Structure 
Function Parameter C^ Over the Ocean , Boundary-Layer 
Meteorol., l5, 506-523. 

Davidson, K.L., Schacher, G.E., Fairall, C.W., and Houlihan, 

T.M.: 1980, Observations of Atmospheric Mixed-Layer Changes 
Off the California Coast (CEWCOM-76) , Second Conference on 
Coastal Meteorology. January 30-February 1, 1980. 

Evermann, G.S., and Leipper, D.F., Marine Fog Development 
Along the West Coast During 1973 Using Transient Ship 
And Coastal Station Observation , M.S. Thesis, Naval 
PostgraduateSchool, Monterey, California, September, 

1976. 

Grisham, P.O., An Investigation of Marine Fog Formation and 
Occurrence over the North Pacific Ocean Area , Research 
Paper, Naval Postgraduate School, Monterey, ^Calif., 
September 1973. 

Haurwitz, B. and Austin, J.M., Climatology, First Edition , 
McGraw Hill Book Company, Inc., 1944. 

Leipper, D.F., Forecasting Summer Fog At Shemya , Advance Copy, 
Weather Central, Alaska 11th Weather Region, AAF, 1 June 
1945, WCA 303F.0. 

Leipper, D.F., "Fog Development at San Diego, California," 

Sears Foundation: Journal of Marine Research, v. VII, 





Leipper, D.F., "The Sharp Smog Bank and California Fog Develop¬ 
ment ," Bulletin of the American Meteorological Society , 

V. 49, No. 4, p. 354-358, April 1968. 

Leipper, D.F., Fog Forecasting on the West Coast; Some Con¬ 
cepts and Some Questions , Naval Postgraduate School, 
Monterey, California, §~March 1976. 

McConnell, M.C., Forecasting Marine Fo^ on the West Coast of 
the United States Using a Linear Discriminant Analysis 
Approach , M.S. Thesis, Naval Postgraduate School, Monterey, 
California, September 1975. (Advisor; R.J. Renard) 

Misciasci, F.J., Fog Occurrence and Forecasting at Two North 
Pacific Ocean Stations, May and June, 1953 , M.S. Thesis, 
Naval Postgraduate School, Monterey, Calif., September 

1974. 

Naval Postgraduate School Report NPS-5lRd75041, Climatological 
Marine-Fog Frequencies Derived from a Synthesis of the 
visibility-weather Group Elements of the Transient-Ship 
Synoptic Reports , by R.J. Renard, R.E. Englebretson, and 
J.S. Daughenbaugh, April 1975. 

Nelson, T.S., 1972; Numerical-Statistical Prediction of 

Visibility at Sea~ M.S. Thesis, Department of Meteorology, 
Naval Postgraduate School, Monterey, Ca., 36 pp. 

Ogata, T. and Y. Tamura, "The Sea Fog Over the Open Sea," 

Kishocho Kenkyu Jiho, Journal of Meteorological Research, 
Part 1, V. 7, pp. 633-642, 1955. 

Pacific Marine Science Report 77-25, Oceanographic Observations 
At Ocean Station P (50°N, 145°W) , Volume 83 and 84, 17 
June - 3 August 1977. 

Peterson, C.A., and Leipper, D.F., Fog Sequences on the 
Central Caiifornia Coast with Examples , M.S. Thesis, 

Naval Postgraduate School, Monterey, California, September 

1975. 

Pilie, R.J., Mack, E.J., Rogers, C.W., Katz, U., and Kocmond, 

W. C., "The Formation of Marine Fog and the Development 
of Fog-Stratus Systems Along the California Coast," 

Journal of Applied Meteorology , v. 18, p. 1275-1285, 
October, 1^79. 

Schramm, W.G., 1966; Analysis and Prediction of Visibility 
at Sea . M.S. Thesis, Department of Meteorology, Naval 
Postgraduate School, Monterey, Ca., 54 pp. 


125 














Schrock, J.A., and Jung, G.H., Transient Ship Synoptic Re¬ 
ports, an Evaluation of their Contributions"to a Fog 
S^dy of 19 August - 5 September 19 74, and 1-5 December 
T§15 , NPS Report 58JG7606i, Naval Postgraduate School, 
Monterey, California, June 1976. 

Selsor, H.D., Further Experiments Using A Model Output 

Statistics Method In Estimating Open Ocean Visibility , 
M.S. Thesis, Naval Postgraduate School, Monterey, 
California, December 1980. (Advisor: R.J. Renard) 

U.S. Departments of Commerce, Defense and Transportation, 
Federal Meteorological Handbook No. 2, Synoptic Code , 
(also NAVAIR 50-1D-2), Superintendent of Documents,U.S. 
Government Printing Office, Washington, D.C., 1969b. 

Wheeler, S.E., and Leipper, D.F., 1974: Marine Fog Impact 
on Naval Operations . NPS Report 58Wh74091, Dept, of 
Oceanography, Naval Postgraduate School, Monterey, Ca., 

118 pp. 

WMO TECHNICAL CONFERENCE ON THE APPLICATIONS OF MARINE METEOR 
OLOGY TO THE HIGH SEAS AND COASTAL ZONE DEVELOPMENT, The 
Observation, Analysis, Forecasting and Climatology of 
Marine Fog by R.J. Renard, Geneva, 22-26 November 1976. 

Yavorsky, P.G., and Renard, R.J., 1980: Experiments Con¬ 
cerning Categorical Forecasts of Open-Ocean Visibility 
Using Model Output Statistics . NPS Report NPS63-80-002, 
Dept, of Meteorology, Naval Postgraduate School, Monterey 
Ca., 87 pp. 


126 

















INITIAL DISTRIBUTION LIST 


No. 


1. Defense Technical Information Center 
Caaneron Station 

Alexandria, VA 22314 

2. Library, Code 0142 
Naval Postgraduate School 
Monterey, CA 93940 

3. Chairman (Code 68Mr) 

Department of Oceamography 
Naval Postgraduate School 
Monterey, CA 93940 

4. Chairman (Code 63Rd) 

Department of Meteorology 
Naval Postgraduate School 
Monterey, CA 93940 

5. Dr. Glenn H. Jung, Code 68Jg 
Department of Oceanography 
Naval Postgraduate School 
Monterey, California 93940 

6. Lt. Robert L. Clark 
Division of Math and Science 
Oceauiography Department 
U.S. Naval Academy 
Annapolis, Md 21402 

7. Director 

Naval Oceanography Division 
Naval Observatory 
34th and Massachusetts Avenue NW 
Washington, D.C. 20390 

8. Commander 

Naval Oceanography Command 

NSTL Station 

Bay St. Louis, MS 39522 

9. Commanding Officer 

Naval Oceanographic Office 

NSTL Station 

Bay St. Louis, MS 39522 

10. Commanding Officer 

Fleet Numerical Oceanography Center 
Monterey, CA 93940 


Copies 

2 

2 

1 

1 

2 

1 


1 


1 

1 

1 


127 



11. Conunanding Officer 

Naval Ocean Research and Development 
Activity 
NSTL Station 

Bay St. Louis, MS 39522 

12. CommeUiding Officer 

Naval Environmental Prediction 
Research Facility 
Monterey, CA 93940 

13. Chairman, Oceanography Department 
U.S. Naval Academy 

Annapolis, MD 21402 

14. Chief of Naval Research 
800 N. Quincy Street 
Arlington, VA 22217 

15. Office of Naval Research (Code 480) 
Naval Ocean Research and Development 

Activity 
NSTL Station 
Bay St. Louis, MS 39522 

16. Scientific Liaison Office 
Office of Naval Research 

Scripps Institution of Oceanography 
La Jolla, CA 92037 

17. Library 

Scripps Institution of Oceanography 
P.O. Box 2367 
La Jolla, CA 92037 

18. Library 

Department of Oceanography 
University of Washington 
Seattle, WA 98105 

19. Library 
CICESE 

P.O. Box 4803 

San Ysidro, CA 92073 

20. Library 

School of Oceanography 
Oregon State University 
Corvallis, OR 97331 

21. Commander 

Oceanographic Systems Pacific 
Box 1390 

Pearl Harbor, HI 96860 


128 






1 


22. Dr. Dale F. Leipper, Code 68Lr 
Department of Oceanography 
Naval Postgraduate School 
Monterey, California 93940 

23. Mr. Charles H. Clark 1 

417 W. Ccunpbell Avenue 

Sherrill, New York 13461 

24. CALSPAN Corporation 1 

Attn: E. J. Mack 

P.O. Box 400 
Buffalo, N.Y. 14225 

25. Commanding Officer 1 

Naval Western Oceanography Center 

Box 113 

Pearl Harbor, HI 96860 

26. Comittauider 1 

Naval Air Systems Command 

Attn: AIR-370, CDR K. VanSickle 
Washington, D.C. 20360 


129