NASA-CR-201430
Tellus ( 1995 ). 47 A. 575 596 Copyright r Munksgaard, 1995
Printed in Belgium all rights reserved TELLUS
ISSN 02X0 6495
Planetary and synoptic-scale interactions during the life
cycle of a mid-latitude blocking anticyclone over the
North Atlantic
By ANTHONY R LUPO* and PHILLIP J. SMITH, Department of Earth and Atmospheric Sciences ,
Purdue University, West Lafayette , IN 47907 , USA
(Manuscript received 29 August 1994; in final form 6 March 1995)
ABSTRACT
The formation of a blocking anticyclone over the North Atlantic has been examined over its
entire life-cycle using the Zwack Okossi (Z O) equation as the diagnostic tool. This blocking
anticyclone occurred in late October and early November of 1985. The data used were provided
by the NASA Goddard Laboratory for Atmospheres on a global 2.0° latitude by 2.5° longitude
grid. The horizontal distribution of the atmospheric forcing mechanisms that were important to
500 mb block formation, maintenance and decay were examined. A scale-partitioned form of the
Z O equation was then used to examine the relative importance of forcing on the planetary and
synoptic scales, and their interactions. As seen in previous studies, the results presented here
show that upper tropospheric anticyclonic vorticity advection was the most important con-
tributor to block formation and maintenance. However, adiabatic warming, and vorticity tilting
were also important at various times during the block lifetime. In association with precursor
surface cyclogenesis, the 300 mb jet streak in the downstream (upstream) from a long-wave
trough (ridge) amplified significantly. This strengthening of the jet streak enhanced the anti-
cyclonic vorticity advection field that aided the amplification of a 500 mb short-wave ridge. The
partitioned height tendency results demonstrate that the interactions between the planetary
and synoptic-scale through vorticity advection was the most important contributor to block
formation. Planetary-scale, synoptic-scale, and their interactions contributed weakly to the
maintenance of the blocking anticyclone, with the advection of synoptic-scale vorticity by the
planetary-scale flow playing a more important role. Planetary-scale decay of the long-wave ridge
contributed to the demise of this blocking event.
1. Introduction
Blocking anticyclones are large-scale phenomena
which can have profound impact on mid-latitude
weather and climatic conditions not only over
the regions in which they occur, but also over
upstream and downstream areas as well { Rex,
1950, 1951; Illari, 1984; Agayan and Mokhov,
1989; Mokhov, 1993). The formation of blocking
anticyclones has proven to be a difficult problem
for both operational forecasting and for various
forecast models (Simmons, 1986; Tibaldi and
Molteni, 1990; Tracton, 1990; Tibaldi et al., 1993,
* Corresponding author.
1994 ), even though their climatological behavior is
w r ell known (Rex, 1950; Triedl et al., 1981; Lejenas
and Okland, 1983; Lupo and Smith, 1995). Many
studies have showm the importance of large-scale
forcing, such as long-w r ave baroclinic processes or
topography, in block formation and maintenance
(Charney and Devore, 1979; Sperenza, 1986; Dole,
1986). Other studies have described the configu-
rations of and or the interactions between long
waves in blocked flow's (Austin, 1980; Lejenas and
Madden, 1992), In particular, a series of papers
by Tung and Lindzen ( 1979) attempts to explain
blocking as caused by the resonant amplification
of large-scale planetary waves forced by those
mechanisms named above. McWilliams (1980)
Tellus 47 A (1995), 5, 1
576
A. R. LUPO AND P. J. SMITH
demonstrates that blocking anticyclones have
characteristics consistent with that of a solitary
wave (“modon”), which in their simplest form
consist of a high-low vortex pair. Modons are
uniformly translating, shape preserving, non-linear
analytic solutions of, in the case of the McWilliams
study, the inviscid equivalent barotropic potential
vorticity tendency equation. Frederiksen (1982)
considers the problem of blocking as instabilities
which grow in three-dimensional Hows. He studied
the problem using a variety of static stabilities
corresponding to differing flow- stabilities. All of
these studies show plausible mechanisms for block
formation and maintenance. However, they fail to
address such issues as why blocks tend to occur in
preferred geographic locations (Frederiksen, 1982),
why blocks are observed to form and decay on
time scales more consistent with synoptic-scale
phenomena (Lupo and Smith, 1995), or why
blocks are observed to ‘‘fluctuate” in intensity (or
regenerate) during their life-cycle (McWilliams,
1980). More recently, another set of studies has
demonstrated the importance of mid-latitude
transients on block formation (Alberta et al., 1991;
Colucei, 1985, 1987; Mullen, 1987; Shutts, 1983,
1986; Konrad and Colucci, 1988; Tsou and Smith,
1990; Tracton, 1990).
Diagnostic studies have utilized a variety of
methods to examine the atmospheric forcing
mechanisms that are important in block formation
and maintenance. Hansen and Chen ( 1982) used
atmospheric energetics equations in spectral space
to examine both a Pacific and an Atlantic blocking
case. They concluded that two processes that
lead to block formation are the non-linear forcing
of ultra long waves by intense cyclogenesis (mode
1 blocking), and the baroclinic amplification of
planetary-scale weaves (mode 2 blocking). Both
scenarios were preceded by intense upstream sur-
face cyclogenesis. Dole (1986) performed a com-
posite analysis of persistent Pacific region positive
(blocking anticyclones) and negative (“cutoff”
lows) anomaly events to describe their vertical and
thermal structures and their relationship to the
zonal How in the jet region. He found that during
development both classes of anomalies exhibited
a highly baroclinic structure and that formation
and decay often occurred very rapidly. He also
found that no regional atmospheric precursor was
evident until just prior to development. Mullen
(1987) used the quasi-geostrophic potential
vorticity equation to examine the importance of
synoptic-scale eddy forcing on composite (time
mean) Atlantic and Pacific blocking flows. He
found that eddy vorticity transports, occurring
one-quarter wave length upstream from the
blocking ridge, and eddy heat transports were
important during development, but that the
former was more important than the latter in
maintaining the composite blocks. Also, he found
that baroclinic instability was a very important
mechanism in the formation of synoptic-scale
eddies associated with blocking. Alberta et al.
(1991), Tsou and Smith (1990), and Tracton
(1990) used various forms of the height tendency
equation, the vorticity equation, or the omega
equation as their diagnostic tools. They all deter-
mined that anticyclonic vorticity advection was
important in block formation. The first two studies
also noted the importance of temperature advec-
tion in block formation, while Tracton (1990)
found this mechanism played a more indirect role.
Alberta et al. ( 1991 ) noted that once the block
was established, barotropic forcing processes
dominated the maintenance period. Alberta el al.
(1991) and Tsou and Smith (1990) also found
block formation to be preceded by a rapidly-
developing surface cyclone. Additionally, Tsou
and Smith (1990) noted the importance of an
associated jet streak just prior to block onset and
just after the period of rapid surface cyclone
development.
Since the studies mentioned above show- blocking
to involve both synoptic-scale and planetary-scale
processes, some diagnostic studies have attempted
to isolate the contributions of each and examine
their interactions. Tsou and Smith (1990) use a
partitioned form of the extended height tendency
equation to examine the role each scale and their
interactions play in block formation. They found
that the interaction component, dominated by the
advection of synoptic-scale vorticity by the
planetary-scale winds, W'as the largest contributor
to block formation. They also show that block
formation is the result of the superposition of
a mobile and amplifying synoptic-scale ridge
and a large-scale stationary planetary-scale ridge.
Following Tsou and Smith ( 1990 ), Tracton ( 1990)
uses a spectral decomposition of 500 mb height
fields to demonstrate that blocking reflects the
superposition of synoptic-scale (and smaller scale)
and planetary-scale wave modes. He used the
Tellus 47 A (1995), 5, 1
PLANETARY AND SYNOPTIC-SCALE INTERACTIONS
577
quasi-geostrophic vorticity and omega equations
as the basic framework of his investigation and
found results similar to the scale partitioned results
of Tsou and Smith (1990).
Many of the studies mentioned above focus on
the development of blocking anticyclones, and a
few examine their maintenance. Only one of these
studies (Dole, 1986) examines the decay period of
blocking anticyclones. Considering the difficulty
that forecast models encounter during block for-
mation and decay (Tibaldi and Molteni, 1990;
Tracton, 1990; Tibaldi et al., 1993, 1994 ), there is
a clear need for more studies involving the entire
life-cycle of blocking anticyclones (Tracton, 1990).
In this paper, a diagnosis of the entire life-cycle of
an individual 500 mb blocking anticyclone event is
performed. Using the Zwack-Okossi equation
(Zwack and Okossi, 1986; Lupo et al., 1992) as the
primary diagnostic tool, the horizontal distribu-
tion of various atmospheric forcing mechanisms
are studied. In addition, by partitioning the basic
data into their planetary-scale and synoptic-scale
components, the atmospheric forcing on each
scale, as well as the interaction between the scales
is also examined over the block life-cycle. Another
unique aspect of this work is that a blocking
anticyclone preceded by non-explosive cyclo-
genesis is examined. Previous studies similar to
this one, notably Tsou and Smith (1990) and
Tracton (1990), examine blocking anticyclones
that were preceded by explosive cyclogenesis.
Finally, the primary objective of this diagnosis is
to elaborate on the Tsou and Smith ( 1990) block
formation mechanism, with particular attention
given to the role that the intervening jet streak
plays in block formation.
2. Data
The data used in this investigation were anal-
yses obtained from NASA/Goddard Laboratory
for Atmospheres (GLA) (Schubert et al., 1993).
This assimilated data set covers a 5-year period
from 1 March 1985 through 28 February 1990.
Analyzed fields were provided on a 2.0° latitude by
2.5 longitude grid at 14 mandatory pressure levels
from 1000 to 20 mb at 6-h intervals for the entire
globe. These fields include upper air parameters
w and v (horizontal wind vector components in
m s), r (geopotential height (m l), T (absolute tem-
perature), rh (relative humidity), and q (mixing
ratio (g/kg) ), all of which were then interpolated
linearly in In (p) to 50 mb isobaric levels. Also
included were a variety of surface parameters; a
complete list of these can be found in Schubert
et al. (1993). The analysis scheme incorporated
data from a variety of sources. The basic com-
ponents of the assimilation system, model physics,
and parameterizations used are described in more
detail by Baker et al. (1987) and Schubert et al.
{ 1993). Finally, subsets of these data are available
upon further request from the Goddard Distri-
buted Active Archive Center ( DAAC) The subset
of data utilized for this study covers the period
from 26 October to 30 November 1985.
3. Computational procedures
The diagnosis was accomplished using the
Zwack-Okossi (Z-O) equation (Zwack and
Okossi, 1986) in its complete form (Lupo et al.,
1992). The Z-O equation is a geostrophic vorticity
tendency (d£ g /dt) equation derived by coupling
the vorticity and thermodynamic tendency equa-
tions through the hydrostatic thickness equation
(Lupo et al., 1992). The result is a generalization
of the Petterssen-SutclifTe equation (Petterssen,
1956, p.324) that allows for the diagnosis of
geostrophic vorticity tendency at a near-surface
pressure level as forced by dynamic and thermo-
dynamic forcing mechanisms vertically integrated
through the depth of the entire atmosphere.
To use the Z-O methodology for diagnoses
at pressure levels alolt (pi), it is necessary to
formulate a geostrophic vorticity tendency equa-
tion for level pi which includes the near-surface
geostrophic vorticity tendency. This is easily
accomplished by taking the horizontal laplacian
and Eulerian time derivative of the hydrostatic
thickness tendency equation, multiplying by the
gravitational acceleration, and dividing by the
Coriolis parameter (f) to yield the Z-O upper-air
tendency equation:
R .
+ - V :
J “pi
pvr+^+Sw]
/
d P
P '
( 1 )
Teilus 47 A (1995), 5, 1
578
A. R. LUPO AND P. J. SMITH
where, in this study, pi = 500 mb; p/ represents a
near-surface level (the first 50 mb pressure level
above the earth’s surface at any grid point ); V is
the horizontal wind vector; Q the diabatic heating
rate; S the static stability parameter (-T/0)
{cQjdp)\ oj the vertical motion (dp/d/); V the del
operator on an isobaric surface; and /?, < p , T,
and (K the gas constant for dry air, the specific heat
at a constant pressure, absolute temperature, and
potential temperature, respectively. The near-
surface geostrophie vorticity tendency (5f g /c/) p/ is
then given by the complete Z-O near-surface
tendency equation:
) = PD
■t V
dC A dw
- V • VC.. — (O + l - j —
dp 4 op
(a) (b) (c)
vadv vvte dive
./ 8V '
) + A • ( V x F) —
CL \
— k Va> x —
V f>/
/
ot J
(dl
(e)
If)
till
fric
ageo
(PD)/?p j‘ r/
./ “ Pt *7’
-f V7>
Q
dp.
( 2 )
(g) (hi (i)
tadv diab adia
In (2), F is the frictional force, £ a the absolute
vorticity, and £ ag the ageostrophic vorticity. Also,
PD is ( 1 /( pZ — />,)), where p t is the pressure at
some sufficiently high pressure level (30 mb in this
study ) chosen to encompass most of the atmo-
spheric mass. Forcing mechanisms (a) through (f )
on the right-hand-side of (2) are the dynamic
forcing mechanisms coming from the vorticity
equation, while the remaining terms (g), (h), and
(i) are the thermodynamic forcing mechanisms.
Finally, ( 1 ) was relaxed to produce height ten-
dencies at pi. These height tendencies were then
filtered to remove all information below 5A.v
( < 1000 km), thus removing subsynoptic-scale
noise present due to data, model, and computa-
tional errors.
In (1) and (2), o was calculated using a
complete form of the omega equation similar to
that of Krishnamurti (1968):
V 2 m%> 4- jl.
crv)
dp 2
, . (IV „
x V VC a -A • V x F 4- — x Vo)
x ' 1 op
d~' r R f
4- /w-rA + -v 2 v str-
op- P \
(3)
In (1), (2), and (3) the frictional force was
restricted to the boundary layer and was calcul-
ated using the Krishnamurti ( 1 968 ^ algorithm.
Diabatic heating included convective and stable
latent heat release (Fosdick and Smith, 1991;
Lupo et al., 1 992 ). boundary-layer sensible heating
(Lupo et al., 1992), and a longwave radiation
parameterization (Sasamori, 1968) that assumes
randomly overlapped clouds ( Harshvardhan et al.,
1987). The boundary-layer sensible heating cal-
culation included contributions from all isobaric
levels between the near-surface level to the first
isobaric level above the planetary boundary layer
( PBL ) as calculated by the Goddard Earth
Observing System Atmospheric General Circula-
tion Model (GEOS-1 AGCM ). The long-wave
radiation parameterization included contributions
due to water vapor, carbon dioxide, and ozone.
The ageostrophic vorticity tendency was calculated
from the ageostrophic wand, which is turn was
calculated as the vector difference between the
observed and geostrophie wind. All horizontal
(vertical) derivatives were calculated using 4th
(2nd)-order finite differencing. Vertical integrals
were calculated using the trapezoidal rule, and all
relaxation was accomplished using sequential
over-relaxation (SOR) (Haltiner and Williams,
1980, pp. 157-164).
All the computations described in this chapter
were carried out over the entire Northern
Hemisphere in order to reduce the influence of
the boundaries in the central region of the grid.
However, all comparisons between Z O and
observed height tendencies were made within a
particular region. This region was chosen so
as to encompass the blocking anticyclone and it’s
immediate vicinity in the upstream and down-
stream directions. For part of the life-cycle of the
Tellus 47A (1995), 5, 1
Response <•
PLANETARY AND SYNOPTIC-SCALE INTERACTIONS
579
block (prior to 1200 GMT 2 November), this
region was bounded by 70° N, 30 C N, 70° W, and
10 W. This area covered most of the North
Atlantic, Eastern North America, Greenland, and
Western Europe. The remaining part of the block
life-cycle was examined in a region bounded by
70° N, 30° N, 90° W, and 30°W. This region
covered the eastern third of North America and
the Western North Atlantic. Comparisons were
made using correlation coefficients between the
Z-O equation height tendencies and observed
± 6-h finite differenced height tendencies and mean
absolute values of Z-O and observed tendencies,
both over all grid points contained in the domains
defined above.
A scale partitioning procedure following that of
Tsou and Smith ( 1990) was employed to partition
the basic data fields into their planetary-scale and
synoptic scale components. The scale partitioning
was accomplished using a second-order two-
dimensional Shapiro (1970) filter applied 1250
times to the basic data fields. Applying the filter
in this manner yields a response function (Fig. 1,
curve a), which retains approximately 2%, 44%,
and 80% of the signal for waves having a hori-
zontal wavelength of 3000, 4500, and 6000 km
at 45" N, respectively. This particular response
function was chosen because blocking anticyclones
typically have wavelengths of 6000 km. Also, since
at 45°N 6000 km represents a hemispheric wave
number between 4 and 5, the filtered data can be
regarded as the planetary-scale component, and
that part which was filtered out (the difference
between the total and filtered component) becomes
the synoptic-scale component. It should be noted
here that the filtering procedure differs from that of
Tsou and Smith ( 1990), who used a Barnes filtering
procedure similar to Maddox (1980). While the
filtering procedure was different in order to save
computing time, the filtering parameters were
chosen such that our response function (Fig. 1,
curve a) was similar to that of Tsou and Smith
(1990) (their Fig. 1 ).
As noted previously, height tendencies were
filtered to remove small scale (<1000 km) noise
without significantly degrading the large-scale
information. This was accomplished using a
4th-order, 2-dimensional Shapiro filter applied
1250 times to all height tendencies. The response
function for this filter is curve b in Fig. 1.
Finally, the filtered data fields were used in a
partitioned form of the Z-O equation derived by
substituting for each variable X:
X=X+X\ (4)
where X(X') represents the planetary-scale (syn-
optic-scale) component. Using this technique in
( 1 ) yields an equation of the form:
at
P + S + /,
(5)
where I\ S , and / represent the planetary-scale,
synoptic-scale, and scale-interaction components,
respectively, of the forcing processes on the right-
hand-side. A similar equation for , occurs
when the terms in (2) are partitioned. An example
of the result of this process as applied to term (a)
on the right-hand-side of (2), or the contribution
to the total tendency by vorticitv advection, in
both ( I ) and (2 ), is:
fi*
pd ) ( - v vc, - r ■ v;'
‘/'i
r s
Fig. 1. Response curves for the (a) 2nd-order and (b)
4th-order 2-dimensional Shapiro ( 1970) filler.
- r vc:,- ■"•VCjd/;, (6)
/, I,
Tellus 47A (1995), 5. 1
580
A. R. LUPO AND P. J. SMITH
where /, is the advection of synoptic-scale vorticity
by the planetary-scale wind, / 2 is the advection
of planetary-scale vorticity by the synoptic-scale
wind, and I { + 1 2 = /.
4. Synoptic discussion
Fig. 2 is a graph of 500 mb central or maximum
height values spanning the pre-block 500 mb ridge
and blocking anticyclone period. Significant dates
and periods within the blocking anticyclone's life
cycle are labeled on the .v-axis or defined inside the
figure. The procedures for choosing the central or
maximum height values and time of block onset
and termination are specified in Lupo and Smith
(1995). In addition, the synoptic discussion that
follows is based on the dates in Fig. 2. The pre-
development 500 mb Atlantic region environment,
represented by 1200 GMT 28 October 1985,
shows a long-wave trough located over the western
North Atlantic, eastern Canada, and the north-
eastern United States (Fig. 3a). Embedded within
this trough was a closed cyclone located over
the Gulf of St. Lawrence (48°N, 65° W) and a
short-wave ridge to the east of the closed center
and south of Greenland. A stationary long-wave
ridge was located over the eastern Atlantic and
much of Europe. A surface cyclone located over
Newfoundland had reached the end of its develop-
ment phase with a central pressure of 990 mb.
By 1800 GMT 29 October ( Fig. 3c), the 500 mb
low was located over Newfoundland and had
reached its minimum height 6-h earlier. The short-
5600 I ■ i —
12/29 06/30 00/31
Tune (GMT V Dale (October November 19B5)
Fig. 2. Plot of 500 mb maximum or central height values
(gpm) for the 500 mb short-wave ridge or blocking
anticyclone versus time. Important periods of time within
the block’s lifecycle are separated by the vertical dashed
lines.
wave ridge, now located over southern Greenland
and extending approximately 500 km into the
North Atlantic, had propagated northeastward
with a northwest-to-southeast oriented axis. This
ridge became stationary at this time. Also, at this
time a cyclonically curved jet streak at 300 mb w r as
located along 50" W on the eastern ( western ) flank
of the long-wave trough (ridge) (Fig. 3d). The
surface cyclone noted above, with nearly the
same central pressure, had moved northwestward
slightly such that by this time there was little tilt
with height. After this time the cyclone decayed
rapidly and disappeared 12-h later. A second
surface cyclone was just forming along a surface
trough extending southward from the first low.
This second cyclone first appeared as a closed
system at this time with a central pressure of
996 mb at 42 C N, 5(F W, or 700 km southeast of
Newfoundland (Fig. 3b), and is of special interest
because it is the precursor cyclone to the block
development. References to a surface cyclone and
cyclogenesis in the next two paragraphs apply to
this cyclone.
By 0600 GMT 30 October ( Fig. 3e ), the 500 mb
short-wave ridge had intensified and a closed anti-
cyclone center appeared southeast of Greenland
(56 N, 32,5 r W), indicative of block onset. Also a
closed high pressure area at the surface (not
shown) appeared in approximately the same locale
with a central pressure of 1026 mb. This surface
anticyclone could be identified with the 500 mb
anticyclone throughout the block lifetime. The
300 mb jet streak (Fig. 3f) had also intensified
w r ith the maximum wind speeds increasing approx-
imately 25% in the first 12-h, 80% of which
occurred in the first 6-h, following the start of
surface cyclogenesis (1800 GMT 29 October).
The short-wave amplification and intensifying jet
streak upstream of the closed anticyclone, the
latter associated with the developing surface
cyclone, is a block development scenario similar
to the conceptual model proposed by Tsou and
Smith (1990).
The blocking anticyclone remained nearly
stationary for 42-h after onset. During this period
it intensified most rapidly, reaching maturity at
0000 GMT 31 October (Fig. 3g). At block onset,
the surface cyclone was midway through its
period of most rapid deepening and had achieved
a central pressure of 991 mb. Slower deepening
occurred over the next 12-h and the cyclone
Tellus 47A (1995), 5, 1
PLANETARY AND SYNOPTIC-SCALE INTERACTIONS
581
5<H) mb Heights (gpm) 1200 GMT 28 Oct 85 500 mb Wind Speed (m/s) 1200 GMT 28 Oct 85
500 mb Heights (gpm) 1800 GMT 29 Oct 85 500 mb Wind Speed (m/s) 1800 GMT 29 Oct 85
500 mb Heights (gpm) 0600 GMT 50 Oct 85 500 mb Wind Speed (m/s) 0600 GMT 50 Oct 85
Fig. 3- Regional 500 mb height (gpm) and 300 mb windspced (m s' ’) maps for (a) and (b) 1200 GMT 28 October,
(c) and (d) 1800 GMT 29 October, (e) and (f) 0600 GMT 30 October, (g) and (h) 0000 GMT 31 October. <i| and
(j) 1200 GMT 01 November, and (k) and (1) 1200 GMT 04 November 1985. The contour intervals are 60 gpm for
the height fields and 10 (m s' ') for the wind speeds. The surface cyclones are denoted by an “L” on the height maps.
The shaded regions are wind speeds exceeding 40 (m s 1 ).
Tellus 47 A (1995), 5, 1
PLANETARY AND SYNOPTIC-SCALE INTERACTIONS
583
attained its lowest central pressure of 988 mb at
1800 GMT 30 October (not shown). Throughout
the remainder of the block lifetime, the occluded
surface cyclone changed little in intensity and
became associated with a 500 mb low that was
part of the blocking dipole that was forming
at 0000 GMT 31 October and is quite evident at
1200 GMT 01 November (Fig. 3i).
From 1200 GMT 1 November (Fig. 3i) through
1200 GMT 2 November (not shown), the blocking
anticyclone propagated westward across the
North Atlantic and into extreme eastern Canada
(58 N, 62.5 W). During the 60-h period ending
with 1200 GMT 2 November, the central height
values of the blocking anticyclone remained
nearly constant (Fig. 2). A strong anticyclonically
curved jet nearly encircled the 500 mb anticyclone,
such that a strong easterly jet streak was located
between the 500 mb anticyclone/cyclone dipole
(Fig. 3j). After 1200 GMT 2 November, the block
again became nearly stationary and remained
so for the balance of its lifetime. The surface
anticyclone associated with the block had slowly
intensified reaching a maximum of 1037 mb by this
time.
The decay began after 1200 GMT 3 November
(represented by Fig. 3k, 1200 GMT 4 November)
and was characterized by the center of the anti-
cyclone moving southward and the block losing
its identity until it failed to meet the blocking
criteria defined by Lupo and Smith (1995) on
0000 GMT 5 November. Note from Fig. 2 that the
block decay is not marked by a decrease in central
height values. During the 36-h period preceding
1200 GMT 3 November, the block underwent
modest intensification (Fig. 2). After this time,
the 500 mb cyclone that was part of the dipole
became an open wave again as it moved east of
the blocking ridge (see Fig. 3k ). In addition,
another surface cyclone, located over the southeast
Atlantic coast of the United States, developed in
association with an advancing upper air trough.
During this development, the surface cyclone w ? as
located under a large region of 300 mb diffluent
flow with jet maxima aligned in a configuration
similar to that shown by Rodgers and Bosart
(1991) (Fig. 31). Therefore, unlike the configura-
tion at block onset, and unlike the conceptual
model of Tsou and Smith (1990), no jet streak:
was located on the eastern ( western ) flank of the
500 mb trough( ridge).
5. Results
5,1. Z O diagnostic results
To conserve space, the ensuing discussion will
focus on the results for map times representative of
the block development (pre-block ), intensification,
maintenance, and decay periods, even though all
6-h time periods during the block lifetime w^ere
examined. Presented are total calculated 500 mb
height tendencies and the contributions to the
total height tendency by vorticity advechon,
temperature advection, and adiabatic temperature
change (map depictions). These mechanisms were
chosen because they were consistently the largest
contributors to the total tendency in the region
studied. Also shown are the contribution of the
individual forcing terms at the center point (the
maximum 500 mb height value) of the 500 mb
anticyclone (bar graphs), which is denoted by an
“X” in each map. The bar graphs were used to
represent the development component of the
height tendency, since at the center point the
propagation component is zero. Comparison of
the calculated 500 mb height tendencies w ith the
centered (±6-h) finite difference height changes
for each map time resulted in average correlation
coefficients of 0.794 and mean absolute values
for the tw r o fields of 0.874 x10 3 (gpm/s) and
1.029 x 10 " 3 (gpm/s), respectively. Thus, the cal-
culated fields are reasonable representations of the
observed height changes.
Using 0000 GMT 30 October to represent the
block development period. Fig. 4a shows that the
region of ridge amplification east and southeast of
Greenland was experiencing height increases,
while height falls were prominent over the region
of the precursor cyclone. From Figs. 4b, e it is seen
that anticyclonic vorticity advection (AVA) was
the primary contributor to ridge amplification at
500 mb and had a spatial pattern similar to that
of the total height tendency. The adiabatic tem-
perature change also made a contribution to ridge
amplification (Figs. 4d,e). The height rise maxi-
mum due to this term was located just slig' tlv
northeast of the ridge axis ( see Figs. 3c, e ). This
maximum w r as due to a broad area of dow nward
motion, whose profile (not shown) exhibited a
maximum between 850 and 700 mb. The tem-
perature advection term, like the remaining terms
at the center point, was small (Figs. 4c, e). The
other dynamic terms, with the exception of the
Tellus 47A (1995), 5, 1
Fig. 4. (a) Regional 500 mb calculated height tendency fields, and (b), { c }, and (d) are the contribution to the total
tendency by vorticity advection, temperature advection, and adiabatic temperature change, respectively, for 0000
GMT 30 October 1985. (e) Bar graphs showing the contribution of each individual forcing term to the total tendency
at the block center marked with an “X” in (a), (b), (c) and (d). Contour interval: 0.5 units. Units: 10 1 gpm s '.
Tellus 47A (1995), 5, 1
PLANETARY AND SYNOPTIC-SCALE INTERACTIONS
585
vorticity tilting term, contributed to height falls
along the ridge axis.
An examination of the 300 mb wind field (Figs.
3b, c, d) reveals that the ridge axis was located on
the anticyclonic shear side of the 300 mb jet maxi-
mum. Also, the jet maximum itself strengthened
considerably in association with the low-level
cyclogenesis. This jet maximum also changed in
shape such that the curvature changed from
cyclonic to anticyclonic during block develop-
ment in response to changes in the shape of the
height field. Therefore, it is likely that the strong
anticyclonic vorticity advection found along the
ridge axis throughout the development period
was due to both the anticyclonic curvature and
shear components of the vorticity field. This region
of anticyclonic vorticity advection increased in
strength throughout the development period coin-
ciding with the strengthening of the jet maximum
discussed in Section 4. Thus, it is likely that
baroclinic processes, such as temperature advec-
tion and latent heating played an indirect role in
block formation the block by contributing to a
strengthening of the jet maximum. This result and
conclusion was also found by Tracton (1990).
The blocking anticyclone intensification period
is represented by 1200 GMT 30 October. At this
time, 500 mb height rise maxima w r ere located
northeast and west of the block center (Fig. 5a).
Height rises over the block center had begun to
diminish as had the forcing by each contributing
term on the right-hand-side of (1) (Fig. 5e). By
examining Figs. 5b, d, however, it is apparent that
the height rise maxima induced by vorticity advec-
tion and adiabatic temperature change were no
longer in close proximity to the block center as had
been the case 12-h earlier. Fig. 5e shows that AVA,
temperature advection, adiabatic temperature
change, and vorticity tilting contributed to block
intensification, while all the other terms inhibited
intensification over the block center.
1200 GMT 1 November is representative of
the 500 mb height tendencies during the block
maintenance period. Fig. 6a shows that the height
tendencies at the block center were generally
weaker than at previous times, and the anti-
cyclone center was near the zero tendency isopleth.
Figs. 6b, c, and d also show r that the forcing due to
each mechanism was not necessarily weaker in the
region. Rather, like the total tendency, the block
center was located close to the zero isopleth in
each field. Therefore, during this period, the total
forcing and the forcing contributions from each
individual term at the block center were very
weak, with no one term consistently dominating
the others (Fig. 6e). The total tendency at this
particular time exhibits height falls, but this
reflects the minor oscillations (as seen in Fig. 2)
of the zero-tendency line relative to the block
center that prevailed throughout the maintenance
period. Additionally, it was during the main-
tenance period and after 1200 GMT 1 November
that the block had moved westward, presumably
encouraged by the height rise region west of the
block.
The decay of the blocking anticyclone, repre-
sented by 1200 GMT 4 November, was marked by
slow southward propagation of the center and the
encroachment of height falls on the western flank
and center of the block (Fig. 7a). The height fall
maximum southwest of the block was associated
with the advancing upper air trough mentioned
previously. This height fall region was dominated
by cyclonic vorticity advection, warm air advec-
tion (Figs. 7b, c), and latent heat release (not
shown ). From Figs. 7b, d, e, it can be seen that
AVA, adiabatic warming, and vorticity tilting were
still contributing to height rises at the block center
at 500 mb, and prominent height rise maxima were
located south-southwest of the block. However, a
deep layer of upper tropospheric warm air advec-
tion associated with the advancing trough con-
tributed significantly to height falls at this time.
Combined with the negative contributions from all
the other dynamic forcing terms, this was sufficient
to produce 500 mb height falls over the block
center.
5.2. Scale- partitioned results
An overview^ of the scale-partitioned results will,
again for brevity, apply to the four representative
map times used in Subsection 5.1. In addition, a
detailed discussion of the results is restricted to
total partitioned 500 mb height tendency fields
and anticyclone center statistics in order to IT ms
on the influence of the partitioned forcing terms on
block development. An overall comparison of the
P, S, and I term contributions can be obtained
from the absolute values of the partitioned height
tendencies averaged over the region presented in
Table 1. These reveal that the planetary, /synoptic-
scale interaction height tendencies {I) and the syn-
Tellus 47A (1995), 5, 1
PLANETARY AND SYNOPTIC-SCALE INTERACTIONS
589
Table 1 . Mean absolute values of the regional parti-
tioned height tendency {10~ 3 gpm/s) for (a) the
first three periods and (b) the decay period of the
block
^\^T*ime period:
Scale
(a)
(b)
planetary (P)
0.1573
0.4475
synoptic (S)
0.4924
0.5449
interactions (I)
0.4899
0.6110
optic-scale height tendencies ( S ) were, on average,
significantly larger than the planetary scale (P)
height tendencies over the first three periods in the
life cycle of the blocking event. While the / and S
height tendencies were approximately equal in
magnitude over these periods, the P height tenden-
cies were only about one-third of the other two
components, a result similar to that of Tsou and
Smith ( 1990). Tracton (1990) found that the inter-
action component (/) was the dominant compo-
nent in his case study. During the decay period, the
Fig. 8. Height tendency contributions to the Z-O total height tendency by the (a) planetary-scale, (b) synoptic-scale,
and (c) planetary/synoptic-scale interaction components, (d) Bar graphs showing the contibution to the total
tendency by each term at the block center point (X) on 0000 GMT 30 October 1985. Contour intervals: (a) 0.1 units
and (b) and (c) 0.5 units. Units: 10 3 gpm s' 1 .
Tellus 47 A (1995), 5. 1
590
A. R. LUPO AND P. J. SMITH
Table 2. Abbreviations of the forcing processes represented in the interaction height tendencies figures
Vadl -V-VC
Vad2 -K'-Vf
Tadl — V • VT'
Tad2 -K'-Vf
Adi 1 Sc</
Adi2
Divl
Div2
Till
S' to
d
dp
do)
dp
dV „ ,
— — x Vco
dp
Til2
Vvrl
Vvr2
„
— — — X VCD
dp
dp
. dC
> fy
advection of synoptic-scale vorticity by the planctary-scale wind
advection of planetary-scale vorticity by the synoptic-scale wind
advection of synoptic-scale temperature by the planetary-scale wind
advection of planetary-scale temperature by the synoptic-scale wind
synoptic-scale vertical motion acting on the planetary-scale static stability field
planetary-scale vertical motion acting on the synoptic-scale static stability field
synoptic-scale divergence field interacting with the planetary-scale vorticity field
planetary-scale divergence field interacting with the synoptic-scale vorticity field
the cross product of the vertical shear of the planetarv-scale wind and the gradient of the
synoptic-scale vertical motion
the cross product of the vertical shear of the synoptic-scale wind and the gradient of the
planetary-scale vertical motion
the advection of planet ary -scale vorticity by the synoptic-scale vertical motion field
the advection of synoptic-scale vorticity by the planetary-scale vertical motion field
magnitude of the I and S terms remained nearly
the same. However, the magnitude of the P com-
ponent increased to a value that was nearly equal
in magnitude to the other two components.
The development period was dominated by
500 mb height rises over the amplifying small scale
ridge (see Fig. 4), forced primarily by the interac-
tion height rises (Fig. 8c). An area of planetary-
scale height rises (Fig. 8a) is located nearly coin-
cident with the interaction height rises, while a
synoptic-scale height rise maximum (Fig. 8b) is
located west of the block center. An examination of
the individual contributors (Fig. 8d) to height rises
shows that the vorticity advection term, which
dominated development, was forced by the S and
both terms in the / vorticity advections. It should
be noted here that Table 2 describes the abbrevia-
tions found in Figs. 8d, 9d. lOd, and 1 Id. A com-
parison with earlier map times (not shown ) reveals
that it was the increase in amplitude of the inter-
action component that accounted for the increase
in vorticity advection associated with the streng-
thening of the jet as discussed in Sections 4 and 5. 1
The height rises due to adiabatic warming noted in
Subsection 5.1 were contributed mainly by P and /.
The / contribution was mostly due to vertical
motion on the synoptic-scale interacting with the
planetary-scale static stability field.
The intensification period, in general, was
characterized by smaller 500 mb height rises over
the center of the block (see Fig. 5). The 1 height
rise component, due mainly to the advection of
synoptic-scale vorticity by the planetary-scale
w'ind, was again the main contributor to the height
rises (Fn:. 9di This period was also characterized
by a lack of significant contributions from the
thermodynamic forcing mechanisms (Fig. 9d).
The remaining dynamic forcing mechanisms were
largely due to synoptic-scale processes. As in
Fig. 5, it can be shown by examining Figs. 8a -c
that the height rise maxima were not weaker than
they were during development. These maxima,
in particular the interaction and planetary-scale
component maxima, were just located farther from
the block center than they were 12-h earlier.
As discussed in Subsection 5.1, the maintenance
period was marked by small 500 mb total tenden-
cies and small contributions by each forcing
mechanism. As was shown in Fig. 6, the block
center w r as close to the zero tendency line for
both the total height tendency and individual
forcing tendencies. For most terms, the total parti-
Tcllus 47 A (1995), 5, 1
PLANETARY AND SYNOPTIC-SCALE INTERACTIONS
591
Fig. 9. As in Fig. 8 except for 1200 GMT 30 October 1985.
tioned height tendencies also reflected this trend contributing to height rises over the block center;
(Figs. lOa-c). However, the important role of but, on the planetary-scale, cyclonic vorticity
the synoptic-scale vorticity advection by the advection (CVA) was the largest contributor to
planetary-scale wind in maintaining the block is height falls. All of the other P forcing processes,
still evident ( Fig. lOd ). with the exception of adiabatic temperature
In the decay period, the largest net height falls change, were also producing height falls. In addi-
near and over the block center were contributed by tion, most of the S and / terms were also forcing
the P terms (Fig. 11). At this time, AVA was still height falls. The warm air advection contributing
Tellus 47A (1995), 5, 1
592
A. R LUPO AND P. J. SMITH
Fig . 10. As in Fig. 8 except for 1200 GMT 01 November 1985.
to height falls in Fig. 7e was comprised mainly of the total height tendency. This qualitative observa-
the interaction advections (Fig. lid), while the tion agrees with the findings of Tracton ( 1990) who
dynamic mechanisms were mainly due to synoptic- found a high correlation between the advection
scale processes. of synoptic-scale vorticity by the planetary-scale
Throughout the block life-cycle, the interac- wind and the total vorticity tendency. Addi-
tion component, which is clearly dominated by tionally, the 300 mb wind field can be partitioned
the advection of synoptic-scale vorticity by the into a planetary-scale and synoptic-scale com-
planetary-scale wind (not shown), most resembles ponents (Fig. 12). Correlating each component of
Tellus 47 A (1995), 5, 1
PLANETARY AND SYNOPTIC-SCALE INTERACTIONS
593
the wind field with each of the total partitioned
tendencies reveals that the highest correlations
were consistently between the synoptic-scale com-
ponent of the total wind and the total interaction
height tendencies (see Fig. 12). The correlations
averaged 0.31 and ranged from 0.25-0.40. The
other comparisons yielded correlations between
— 0.1 and 0.1. Note in the example of Fig. 12 that
the interaction height rise regions are located close
to the synoptic-scale jet maxima.
6. Conclusions
The formation of a blocking anticyclone over
the North Atlantic Ocean during the fall of 1985
has been examined over its entire life cycle using
the Z-O equation as the primary diagnostic tool
and the Tsou and Smith (1990) block formation
model as a guide. Similar diagnoses (Tsou and
Smith, 1990; Tracton, 1990) have examined block-
ing anticyclones that were preceded by explosive
Tellus 47A (1995), 5, 1
SF 7
594
A. R. LUPO AND P. J. SMITH
300 mb Wind Speed (m/s)
300 mb Wind Speed (m/s)
(KMX) GMT 31 Oct 85
Fig. 12. The (a) planetary-scale and (b) synoptic-scale component of the 300 mb wind speeds (ms ') and
(c) planetary /synoptic-scale interaction component of the total height tendencies ( x 10 1 gpm s 1 ) for 0000 GMT
31 October 1985. Contour intervals are: (a) 2, (b) 10. and (c) 0.5 units, respectively.
cyclones. This paper examines a blocking anti-
cyclone preceded by a non-explosively developing
cyclone and, unlike other studies, examines the
entire block life-cycle. This paper, like many others,
demonstrates the importance of mid-latitude
transients, especially extratropica! cyclones, in
block formation and maintenance. Furthermore,
as Dole (1986) found with Pacific positive and
negative anomalies, this blocking anticyclone
decayed just as rapidly as it developed, or on a
time scale more consistent with synoptic-scale
phenomena. Interestingly, in the case studied
here, decay of the blocking anticyclone was also
associated with an upstream developing cyclone.
However, this cyclone was not accompanied by a
jet streak favorably positioned as was the case for
block development. More case studies are being
examined to determine if this result can be found in
other blocking anticyclones.
The relationship between blocks and jet streaks
following the results of Tsou and Smith ( 1990), in
w'hich it is suggested that intervening jet streaks
may play a role in the link between precursor
cyclones and blocking anticyclones. In particular,
the jet streak involved in this blocking anticyclone
strengthened significantly in association with
Tellus 47 A (1995), 5, 1
PLANETARY AND SYNOPTIC-SCALE INTERACTIONS
595
surface cyclogenesis. The increased anticyclonic
shear and curvature in turn strengthened the anti-
cyclonic vorticity advection field that dominated
the amplification of the downstream short-wave
ridge. The location of jet streaks relative to the
block center within the large-scale flow appeared
to be important throughout the block life-cycle. As
long as the jet maxima were located favorably, as
in Tsou and Smith (1990), the block developed
and was maintained. However, when they were
located in an unfavorable configuration, or one
favorable to cyclogenesis as in Rodgers and Bosart
( 1991 ), the block decayed.
In this diagnosis, it was found that anticyclonic
vorticity advection was the largest contributor to
block formation and maintenance at 500 mb.
During much of the block life-cycle adiabatic
warming resulting from downward motion, maxi-
mizing between 850 mb and 600 mb, and vorticity
tilting also contributed to 500 mb height rises over
the anticyclone center. The other thermodynamic
mechanisms w^ere very small until the decay
period, when temperature advection contributed
to height falls and the other dynamic forcing
mechanisms also contributed to height falls over
the block center. The block decayed when forcing
processes contributing to height falls overwhelmed
those contributing to height rises.
The partitioned height tendencies over the anti-
cyclone center demonstrated that the interac-
tion component, largely due to the advection of
synoptic-scale vorticity by the planetary-scale
wind, dominated the total height tendency from
development through maintenance. The adiabatic
warming that contributed to block formation
and maintenance described in Subsection 5.1, was
dominated by the synoptic-scale vertical motion
acting on the planetary-scale static stability field.
During maintenance, the S and / components were
jointly responsible for the height rise region west of
the block, and this region probably encouraged
the westward propagation ending at 1 200 GMT 02
November 1985. During decay the planetary-scale
forcing assumed a greater role in contributing to
the total height tendency field. The regional MAVs
shewed that the synoptic-scale and interaction
components were nearly of equal magnitude
throughout the block life-cycle. This result was
similar to that of Tsou and Smith (1990), but dif-
ferent from that of Tracton (1990) who showed
that the interaction component w'as clearly domi-
nant. Following and expanding on the suggestion
of Tracton (1990), more case studies are being
examined to determine if the relative importance
the planetary-scale, synoptic-scale, and interaction
components are dependent on season, flow' regime
character, or geographical location. Finally, a high
correlation between the synoptic-scale 500 mb
wind field and the total interaction height tenden-
cies was found.
7. Acknowledgments
The authors would like to thank co-workers
Marty Rausch and Don Rolfson for their com-
puter programming assistance and for their contri-
butions during discussions. Also, we would like to
thank Mike Seablom at the Goddard Laboratory
for Atmospheres in Greenbelt, Maryland for his
assistance in acquiring the data used for this study.
Finally, the authors acknowledge the support of
the National Aeronautics and Space Administra-
tion through NASA grant # NAG8-915.
REFERENCES
Agayan, G. M. and Mokhov, I. I. 1989. Quasistationary
autumn regimes of the Northern Hemisphere atmo-
sphere in FGGE. Atms. Oc. Pys. 25, 1 150-1156.
Alberta. T. L„ Colucci, S. J. and Davenport, J. C. 1991.
Rapid 500 mb cyclogenesis and anticyclogenesis. Mon.
Wea. Rev. 119 , 1 186-1204.
Austin, J. F. 1980. The blocking of middle latitude
westerly winds by planetary waves. Q. J. Roy. Meteor.
Soc. 106 , 327- 350.
Baker, W. E., Bloom, S. C., Wollen, J. S., Nestler, M. S.,
Brin, E., Schlatter. T. W. and Branstator, G. W. 1987.
Experiments with a three-dimensional statistical
objective analysis scheme using FGGE data. Mon
Wea. Rev. 1 15 , 272-296.
Chamey, J. G. and DeVore, J. G. 1979. Multiple flow
equilibria in the atmosphere and blocking. J. Atmos
Sci. 36 , 1 205 1216.
Colucci, S. J. 1985. Explosive cyclogenesis and large-scale
circulation changes: Implications for atmospheric
blocking. J. Atmos. Sci. 42 , 2701-2717.
Colucci, S. J. 1987. Comparative diagnosis of blocking
versus non-blocking planetary circulation changes
during synoptic scale cyclogenesis. J. Atmos. Sci 44
124-139.
Tellus 47 A (1995), 5, 1
596
A. R. LUPO AND P. J. SMITH
Dole, R M. 1986. The life cycles of persistent anomalies
and blocking over the North Pacific. Adv. Geophys. 29,
31-70.
Fosdick, E. K. and Smith, P. J. 1991. Latent heat release
in an extratropical cyclone that developed explosively
over the southeastern United States. Mon. Wea. Rev.
119, 193-207.
Frederiksen, J. S. 1982. A unified 3-D instability theory of
the onset of blocking and cyclogenesis. Mon. Wea. Rev.
110, 969-982.
Haltiner, G. J. and Williams, R. T. 1980. Numerical
prediction and dynamic meteorology. New York: John
Wiley & Sons, 478 pp.
Hansen, A. R. and Chen, T.-C. 1982. A spectral energetics
analysis of atmospheric blocking. Mon. Wea. Rev. 110,
1146-1165.
Harshvardhan, Davies, R., Randall, D. A. and Corsetti,
T. G. 1987. A fast radiation parameterization for
atmospheric circulation models. J. Geophys. Res. 92,
1009-1016.
Illari, L. 1984. A diagnostic study of the potential vor-
ticity in a warm blocking anticyclone. J. Atmos. Sci. 41,
3518-3525.
Konrad, C. E. and Colucci, S. J. 1988. Synoptic climatol-
ogy of 500 mb circulation changes during explosive
cyclogenesis. Mon. Wea. Rev. 116, 1431 1443.
Krishnamurti. T. N. 1968. A diagnostic balance model
for studies of weather systems of low and high
latitudes, Rossby numbers less than one. Mon. Wea.
Rev. 96, 518-530.
Lejenas, H. and Okland, H. 1983. Characteristics of
Northern Hemisphere blocking as determined from a
long time series of observational data. Tellus 35 A,
350-362.
Lejenas, H. and Madden, R. A. 1992. Traveling
planetary-scale waves and blocking. Mon. Wea. Rev.
120, 2821-2830.
Lupo, A. R., Smith, P. J. and Zwack. P. 1992. A diagnosis
of the explosive development of two extratropical
cyclones. Mon. Wea. Rev. 120, 1490-1523.
Lupo, A. R. and Smith, P. J. 1995. Climatological
features of blocking anticyclones in the Northern
Hemisphere. Tellus 47 A. in press.
Maddox, R. A. 1980. An objective technique for separat-
ing macroscale and mesoscale features in meteorologi-
cal data. Mon. Wea. Rev. 108, 1108-1121.
McWilliams, J. C. 1980. An application of equivalent
modons to atmospheric blocking. Dyn. Atmos. Oc. 5,
43-66.
Mokhov, LI. 1993. Diagnostics of the climate systems
structure. Gidrometeoizdat (in Russian). St. Peters-
burg, 272 pp.
Mullen, S. L. 1987. Transient eddy forcing and blocking
flows. J. Atmos. Sci. 44, 3 22.
Petterssen. S. 1956. Weather analysis and forecasting,
vol. I, 2nd edition, McGraw-Hill, 428 pp.
Rex, D. F. 1950. Blocking action in the middle
troposphere and its effect on regional climate (II). The
climatology of blocking action Tellus 2. 275 301.
Rex, D. F. 1951. Blocking action in the middle
troposphere and its effect upon regional climate ( I ).
Tellus 3. 196-211.
Rodgers, E. and Bosart, L. F. 1991. A daignostic study of
two intense oceanic cyclones. Mon. Wea. Rev. 119.
965-996.
Sasamori, T. 1968. The radiative cooling calculation for
application to general circulation experiments. J. Appl.
Meteor. 7, 721-729.
Schubert, S. D., Rood, R. D. and Pfaendtner, J. 1993. An
assimilated dataset for Earth science applications. Bull.
Amer . Met. Soc. 74, 2331-2342.
Shapiro, R. 1970. Smoothing, filtering, and boundary
effects. Rev. Geophys. 8, 359-387.
Shutts, G. J. 1983. The propagation of eddies in diffluent
jetstreams: Eddy vorticity forcings of blocking flow
fields. Q. J. Roy. Meteor. Soc. 109, 737-761.
Shutts, G. J. 1986. A case study of eddy forcing dunng an
Atlantic blocking episode. Adv. Geophys. 29, 135-161.
Simmons, A. J. 1986. Numerical prediction: Some results
from operational forecasting at ECMWF. Adv.
Geophys. 29, 305-338.
Sperenza. A. 1986. Deterministic and Statistical proper-
ties of Northern Hemisphere middle latitude circula-
tion. Adv. Geophys. 29, 199-226.
Tibaldi, S. and Molteni, F. 1990. On the operational
predictablity of blocking. Tellus 42A, 343-365.
Tibaldi, S., Ruti, P., Tosi, E. and Maruca, M. 1993.
Operational predictability of winter blocking: An
ECMWF update. Proc. ECMWF Seminars on Valida-
tion of forecasts and large-scale simulations over
Europe. ECMWF, Reading, Berkshire, UK.
Tibaldi, S., Tosi, E., Navarra, A. and Pedulli, L. 1994.
Northern and Southern Hemisphere seasonal
variability of blocking frequency and predictability.
Mon. Wea. Rev 122, 1971-2003.
Tracton, M. S. 1990. Predictability and its relationship to
scale interaction processes in blocking. Mon. Wea. Rev.
118, 1666 1695.
Triedl. R. A.. Birch, E. C. and Sajecki, P. 1981 Blocking
action in the Northern Hemisphere: A climatological
Study. Atmos.-Oc. 19, 1 23.
Tsou, C.-H. and Smith, P. J. 1990. The role of synoptic/
planetary-scale interactions during the development of
a blocking anticyclone. Tellus 42A, 174-193.
Tung, K. K. and Lindzen, R. S. 1979. A theory of station-
ary long waves, part 1: A simple theory of blocking.
Mon. Wea. Rev. 107, 714 734.
Tung. K. K. and Lindzen, R. S. 1979. A theory of station-
ary long waves, part II: Resonant Rossby waves in the
presence of realistic vertical shear. Mon. Wea. Rev. 107.
735 -750.
Tung, K. K. 1979. A theory of stationary long waves,
part III: Quasi-normal modes in a singular wave
guide. Mon. Wea. Rev. 107. 751 774.
Zwack. P. and Okossi, B. 1986. A new method for solving
the quasi-geostrophic omega equation by incorporat-
ing surface pressure tendency data. Mon. Wea Rev.
114, 655 666.
Tellus 47 A (1995), 5, 1
