VARIATION
IN TROPICAL WIND PATTERNS (1968 - 1979)
by
Crystal Lynn Barker
SUBMITTED TO THE DEPARTMENT OF METEOROLOGY AND PHYSICAL
OCEANOGRAPHY IN PARTIAL FULFILLMENT OF THE
REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
in
METEOROLOGY
At
The
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
May 21, 1982
Mras sac iusetts Institute of Technology,
1982
Signature of Author _-_ (
Department ofMeteorology and Physical Oceanography,
May 21, 1982
Certified by
-
-Re cgi
------------------------------------d E. Newell1
hesis Super
Accepted by
isor
------------------------------
Ronald G. Prinn, Chairman,
on Graduate Studies
WITHDRAWN
i ? i f".-;'?
-----
Departmental
---------
Commi ttee
VARIATION IN TROPICAL WIND PATTERNS
(1968 - 1979)
by
CRYSTAL L. BARKER
Submitted to the Department of Meteorology and Physical
Oceanography on 21 May, 1982, in partial fulfillment of
the requirements for the degree of Master of Science in
Meteorology.
ABSTRACT
The monthly mean zonal and meridional wind components from the
NMC operational tropical objective analysis were used to compute
mass flux, zonal mass flux, and relative angular momentum (RAM).
The mass flux results suggest that the v data are unreliable,
although the u data seem excellent.
The remaining results show interesting variations and relationships between some of the phenomena of the tropics. The Walker
Circulation anomalies have a negative correlation with those of the
Hadley Circulation and may sometimes precede them. Changes in the
SOI and southern Pacific SST (also negatively correlated) seem to
occur before, or simultaneously with, changes in either circulation.
The scenario for an El Nino includes warm water and low pressure
off the coast of South America, high pressure and high SSTs in the
Indian Ocean, a weak Walker Circulation and a very strong Hadley
Circulation.
Thesis Supervisor:
Title:
Reginald E. Newell
Professor of Meteorology
-2-
TABLE OF CONTENTS
Page
I. INTRODUCTION
II. DATA AND ANALYSIS
III.
COMPARISON AND DISCUSSION OF MONTHLY AND MEAN
MASS FLUX
IV. COMPARISON AND DISCUSSION OF MONTHLY AND MEAN
ZONAL MASS FLUX
V. COMPARISON AND DISCUSSION OF MONTHLY AND MEAN
RELATIVE ANGULAR MOMENTUM
VI.
INTERRELATIONSHIP OF RELATIVE ANGULAR MOMENTUM AND
ZONAL MASS FLUX WITH PRESSURE, SST, SOI,
AND STRENGTH OF THE MONSOON
VII.
CONCLUSIONS
ACKNOWLEDGEMENTS
BIBLIOGRAPHY
-3-
LIST OF FIGURES
Figure
I.
II.
III.
Page
Diagram of the Walker Circulation, Data from Work
by Boer and Kyle; Diagram from Newell (1979)
35
Mean January and July u; units: kts.
36
Seasonal Mass Flux; Newell et al (1972);
units:
IV.
V.
1012 g sec - i
Mean Zonal Mass Flux; 1976 - 1979,
January,
April
12
-1
July, and October; units: 10'2 g sec-
g sec39
Relative Angular Momentum Anomalies for Northern
Hemisphere, Southern Hemisphere, and the Globe;
units: 10"3
VII.
g cm2 sec - 1
40
Zonal Mass Flux Anomalies of the Northern Hemisphere
in the Western Pacific and over India;
units: 10 2 g sec -
VIII.
IX.
X.
41
Zonal Mass Flux Anomalies of the Southern Hemisphere
in the Western Pacific and in the Indian Ocean;
units: 1012 g sec - 1
42
Southern Oscillation Index; the Pressure Difference
between Tahiti and Darwin; (compliments of Rennie Selkirk);
units: mb
43
Mean Sea Surface Temperature; Newell (1979);
units:
XI.
38
Zonal Mass Flux 1976; January, April, July and
October; units: 102
VI.
37
oC
44
Actual and Filtered Monsoon (June-September); Rainfall of
India from 1901 - 1977; (Parthasarathy and Mooley);
units:
cm.
45
-4-
Fiurre
XII.
Page
Mean January Sea Level Pressure Distribution;
NASA ATLAS: units: mb
XIII.
46
Mean July Sea Level Pressure Distribution;
NASA ATLAS: units: mb.
XIV.
Diagram of Tropical Interrelationships
XV.
Diagram of the Tropical Interrelationships during
and El Nino
47
48
49
-5-
LIST OF TABLES
Page
50
Table I.
Symbols and Abbreviations
Table II.
Mean Relative Angular Momentum: 1969-1971,
1971 - 1979; Levels: 700-200 mb.
Units: 103 °g cm2 sec - 1
Table III.
Mean Relative Angular Momentum: 1976 - 1979;
Levels: 1000-200mb. Units: 1030g cm2 sec - 1
Table IV.
Relative Angular Momentum: 1977.
Units: 1030 g cm2 sec-1
Table V.
Number of Mean Zonal Mass Flux Cells per Month
19.6°N - 19.6 0 S Latitude
Table V'I.
Correlation Coefficients for: SST, SOI, RAM for the
N.H., S.H., and Globe, Zonal Mass Flux of N.H.
Indian and Pacific Oceans, and S.H. Indian and
Pacific Oceans
Table VII.
Correlation Coefficients.
Global RAM versus
Other Parameters at a - 1 Month Lag
-6-
I.
INTRODUCTION
The climate of the tropics has always fascinated meteorologists.
The mechanisms
that govern tropical cyclones, rainfall, and monsoons are
still not fully understood.
Perhaps most interesting is the question of
what effect the tropical atmosphere and oceans have on the entire global
climate, if any at all, and whether the tropics are affected by midlatitude changes.
One mechanism that affects both the tropics and higher latitudes is
the Hadley Circulation, first described by Hadley (1735).
He postulated
that air rose near the equator, then headed towards the poles where
descended before heading back toward the equator at low levels.
it
The Hadley
cells are an extremely important element of glcbal circulation and of the
heat, momentum, and total energy budgets.
Another phenomenon that seems to be linked to global weather is the
Southern Oscillation.
The Southern Oscillation was investigated by Sir
Gilbert Walker(1923-37) while he was Director General of Observatories in
India.
Walker was searching for a way of predicting monsoonal rainfall,
and discovered that he needed to consider changes all around the world.
He found correlations between different pressure anomalies,
fall, and later, sea and atmospheric temperatures.
seasonal rain-
he found that the
negative correlation of pressure anomalies in the Pacific and Indian Oceans
seemed to affect global weather.
(Other such correlations exist in the
North Pacific and North Atlantic, but on a much smaller scale.)
Southern Oscillation is characterized by the Southern
The
Oscillation Index (SOI)
which is positive when seasonal pressure is high in the southeast Pacific,
-7-
and low in the Indian Ocean.
(Originally, Djakarta and Santiago were
used -- nowadays a variety of stations have appeared in the literature,
with Easter Island and Darwin being the most common.)
Many authors have studied the Southern Oscillation and its relations
with winds, cloudiness, mid-latitude pressure, rainfall, air temperature
and sea surface temperature.
However, the most interesting fact is the
existence of a toroidal circulation in the Equatorial Pacific.
Troup (1965)
first mentioned it in connection with upper level wind anomalies and the
Southern Oscillation.
But it was Bjerknes (1969) who called this Pacific
circulation "The Walker Circulation".
He described it as having a westward
horizontal pressure gradient along the surface, and an eastward one at
upper levels.
Both Troup and Bjerknes recognized that the Walker Circula-
tion and its fluctuations were part of the variations in the Southern
Oscillation.
This idea was later supported by many authors, including
Webster (1973).
Later work showed that such east-west circulations appear around the
globe (Kidson, 1975; Newellet al, 1974)
(See Figure I). This is supported
by satellite cloud data that show cloudiness and rainfall near the upward
motion of the circulation, with cloud-free areas near the downward motion
(Julian and Chervin, 1978).
Krishnamurti (1971) asserts that the Walker
Circulation is just the "southern extension of a much more vigorous eastwest circulation that extends up to 400 N during the northern summer."
The mechanisms that maintain this Walker Circulation (as we shall call the
entire series of equatorial global circulations) have been open to many
theories.
Webster (1972) and Cornejo-Garrido and Stone (1977) attribute
the driving force of the circulation to latent heat.
-8-
Earlier work suggested that atmospheric changes were influenced by
the ocean currents (Schell, 1956).
Wyrtki (1975), however, has come up
with a plausible theory in regard to the east-west circulation and the
phenomenon known asEl Ni'o.
El Nino is the southern summer appearance of
upwelling
warm water off the coast of Peru which suppresses the usual cold
of this area.
The upwelling brings up rich nutrients and this area is one
of the major fishing grounds, mainly of anchovies, in the world.
Peru's
economy depends on it. It is only when this warming persists that fishing is severely disrupted.
El Nino has come to mean only those persistent
warmings, not the annual ones.
Wyrtki (1975) proposed that strong per-
sistentsoutheast trade winds cause an accumulation of water in the
Western Pacific; when the trades relax, the ocean sloshes back, bringing
warm water to the coast of Peru.
His results show that air changes lead,
and therefore initiate, sea changes -- not vice versa.
One of the most spectacular El Ninos occurred in 1972-73.
show a great deal of variability.
Ramage and Hori (1981)
El Ninos
comment on the
complications in prediction caused by partial and aborted El Ninos.
Bjerknes (1969) and many later authors have shown the correlation between
the Southern Oscillation, the Walker Circulation, and SST changes including El Ninos.
Bjerknes (1969) defined a strong Walker Circulation as that which
occurs when the surface pressure is high over the colder eastern Pacific
and low over the warmer Western Pacific, with a decreased precipitation and
low activity of the Hadley Circulation.
A weak Walker would cause an
enhanced Hadley Circulation, a large poleward movement of angular momentum
and strong mid-latitude westerlies.
Trenberth (1979) shows a possible
-9-
relationship between the SOI and the Indonesian area Hadley Cell.
Also, it must not be forgotten that Walker was searching for a way
to predict Indian monsoons when he discovered the Southern Oscillation.
Indeed, there is a strong relationship between the intensity of the
monsoon and aspects of the Walker Circulation (Weare,
1979).
The purpose of this present investigation into the atmosphere of
the tropics is to attempt to inquire into some of these phenomena; in
particular, Bjerknes'
theory that a weak Walker Circulation would cause
an enhanced Hadley Circulation and a large poleward transfer of angular
momentum.
To do this, a large data set (1968-1979) of monthly mean zonal
wind components (u) and (v) will be used.
The mass flux (or intensity of
the Hadley Cells) and the zonal mass flux (intensity of the Walker Circulation) will be calculated and their relation investigated.
The
relative angular momentum of the area 48.1°N to 48.1 0 S will also be
calculated and its trends noted.
Finally, anomalies in the above para-
meters will be investigated and their relationship to patterns of
surface pressure, SST, the monsoon, and the SOI will be examined.
-10-
DATA AND ANALYSIS
II.
The data set for this investigation is the operational tropical
objective analysis of the National Meteorolgical Center (NMC).
Daily
values of i and v from the Meteorological Satellite Laboratory of the
National Earth Satellite Service were used to find monthly means from
1968 - 1979 at six levels: 1000, 850, 700, 500, 300 and 200 mb.
Data
were missing for January 1968, October 1972 and November 1972 at 200
and 700 mb; forJanuaryand February 1968 and October and November 1976
at 300 and 500 mb; also for January 1968 - November 1974 at 850 mb,
and for January 1968 - November 1975 at 1000 mb.
Roughly 8% of the
days of the remaining months was also missing (Arkin, 1981).
The monthly
means were established on a Mercator grid at every 5o longitude and
stretching from 4801 North Latitude to 4801
South; the grid is 23
x72
points.
NMC used rawinsondes, aircraft reports, and satellite cloud-tracked
winds in its analysis.
Three different analysis techniques were used
during this time frame.
August 1974 (Bedient
The Cressman Analysis was used from 1968 to
et al, 1967).
September 1974 - August, 1978.
The Hough Analysis was used from
This, unfortunately, forces the
wind
field to be non-divergent, thus prohibiting observation of mean meridianal overturnings by
causing (v)(the long-term zonal average) to be zero
(Rosen and Salstein, 1980).
Then, from September, 1978, through 1979,
the Optimum Interpolation Technique was used (Gandin, 1963; Bergman, 1979)
and the [v] returned.
Rosen and Salstein (1980) and Chui and Lo (1979) investigated the
-11-
quality of the NMC data and found it generally good.
It compares
well with Sadler's (1975) streamlines produced using aircraft and
radiosonde data. The author found difficulties with the v, however.
There is a great deal of variation in the wind pattern from month
to month, and from year to year.
Roughly, the 200 mb flow in the summer
contains a wide, thick swath of easterlies that extends around the world
and can have extensions or streamers near Australia and Peru.
Sometimes,
this swath is interrupted with westerlies in the mid-Pacific.
In the
winter, the easterly stream narrows and can break into three streams -over South America, Southern Africa, and Indonesia.
appear as far north and south as 30'
These easterlies can
Latitude.
Long-term means were generated for the data set.
For the u field,
the years 1969-71 and 1973-79 were used to create long-term means at 700,
500, 300 and 200 mb.
Another set of long-term u means was created for all
six levels using the years 1976 -79.
The long-term
'
was established using
the years 1969, 1970, 1971, 1973 and 1979 at 700-200 mb.
In order to study the intensity of the Mass Flux and the Zonal Mass
Flux, the equations and techniques discussed by Newell et al, vols. I and
II (1972 and 1974) were used.
The Mass Flux is found from a two-dimensional
stream function:
Jp
[v] dp
and yields a picture of the circulations on a pressure-versus-latitude grid.
The units are 1012 gm sec - 1. As Newell points out in Appendix III, vol. I
Iv] in the middle latitude probably contains large errors, but is fairly
accurate in the tropics.
The stream function evaluated using each grid
-12-
point of each month of each year at every level (zero denotes the
missing months).
The pressure interval used was 75 mb for 1000 mb,
150 for 850, 175 for 700, 200 for 500., 150 for 300 and 50 for 200 mb,
and was from the boundaries to the mid-point of each level.
The Mass
Flux of the long-term means was created for the levels 700-200 mb.
The Zonal Mass Flux was determined by
ZMF
where u
=
aA
=
u - [ul
P
dp
The pressure interval was the same for the Mass Flux.
The latitude
span for each latitude was the distance in degrees from the mid-point
between the latitude to the south to the midpoint of the one to the north.
to 608
This varied from 100
ZMF was found for the nine latitudes between 19.60 north and south,
and each of these latitudes is represented by the contours on a pressure
The units are 1012 gm sec - 1.
versus longitude grid.
The Zonal Mass Flux
was then generated for both sets of long-term u means.
In order to observe the behavior of the relative angular momentum
(RAM)
of the tropics, tables were created for the latitude belts, using
the Newell et al (1972) equation:
2Ta
3
RAM
2=
g
P
2
I ] cos 2 dgdp
JoO
The latitude span and the pressure interval were the same as used
previously.
The 700 - 200 mb levels were used.
Another table was gen-
erated for all the years, and for 1976-79 using all six levels.
Tables
of latitude versus months of each year show the RAM of the latitude belt
-13-
on 103o
gm cm2 sec - '. At the bottom of each table, the total RAM for
the Northern Tropical Hemisphere (O - 48.1oN), the Southern Tropical
Hemisphere, and the total for the area
48.10 N to 48.1oS
are listed for
each month.
The RAMs of the two long-term u means were also calculated.
-14-
II.
COMPARISON AND DISCUSSION OF MONTHLY AND MEAN MASS FLUX
In order to see how the intensity of the Hadley Circulation varies
during this time, the streamlines of the two-dimensional stream function
were inspected.
Newell et al (1972) found that, in the summer and winter
months, a large Hadley Cell sat over the tropics with a weaker cell off to
the side in the summer hemisphere.
In the spring and fall, the two cells
were equal in intensity, and the upward motion between them was located
near the equator.
The maximum values for the large Hadley Cell were
1.8 x 1014 g sec - ' to 2 x 1014 g sec -1 .
In the mid-latitudes, the Ferrel
Cells appeared.(see Figure III).
When the mass flux of the mean winds was
generated, patterns very
similar to the findings of Newell et al (1972) appeared (see Figure III).
However, the intensities were much lower.
Of course, the mean v was
created using only the years 1969, 1970, 1971, 1973 and 1979, and only
1979 extended down to 1000 mb.
But since the maximum values of the Hadley
Cells were about .6 x 101 4 g sec -' , or one-third the values found by Newell
et al (1972) and others, these intensities would seem to be unreliable.
The maximums of the Hadley Cell seem to occur at 400 mb, a level which was
not available, so the high peaks just may have been missed.
factor may be the NMC analysis of v.
A more important
Lau and Oort (1981) found that the
NMC analysis created a weaker v field in the Northern Hemisphere (N.H.)
than the analysis of the Geophysical Fluid Dynamics Laboratory.
Though
the patterns may be useful (since the patterns of the mean were comparable
to other work) the author feels that the mass flux for this data set is
unreliable.
Therefore, we will have to rely
-15-
on the relative angular
momentum to tell us how the Hadley Circulation is functioning, with
only quick glances at the individual mass flux patterns.
IV. COMPARISON AND DISCUSSION OF MONTHLY AND MEAN ZONAL
MASS FLUX
Now to consider how the Walker Circulation, which is superimposed over
the Hadley Circulation, changes throughout this time frame, the zonal mass
flux (Newell et al, 1974) was used as a crude representation of the circulation.
Newell et al found that there was an equatorial cellular circulation
with rising motion over the western Pacific, and sinking air west of South
America.
There was another cell with rising motion over South America
and sinking motion west of Africa, another over India, and a fourth weak
cell over Africa.
The 200mb u values (see Figure II)
do not usually show
distinct patches of easterly and westerly winds connected with each cell,
but appear as a long swath of easterlies that undulates north and south of
the equator.
The zonal mass flux for this particular data set was calculated
as far north and south as 19.6
latitude.
Both sets of long-term mean values (1969-1971 and 1973-1979 at levels
700 - 200mb, and 1976-1979 at levels 1000 - 200 mb; see Figure IV), exhibit
similar behavior above 700mb.
The equatorial regions are dominated by
either four or six major cellular circulations.
The cells are defined as
alternating regions of positive and negative zonal mass flux.
These regions
often show cyclonic (negative values) or anticyclonic movement (positive
values).
When there are six of these cellular regions, these are situated over:
1) the western Pacific, (2) the Indian Ocean, (3) Africa, (4) the Atlantic,
(5) South America, and (6) the eastern Pacific.
There is rising motion
over the date line, over the western Indian Ocean and off the east cost of
South America; there is sinking motion over the eastern Indian Ocean, western
-17-
Africa, and off the west coast of South America.
When there are four
cells, they are over South America, the Indian Ocean, the Pacific and
the eastern Pacific.
There is rising motion over the western Pacific,
and off the east coast of South America, and sinking motion off the west
coasts of Africa and South America.
The six-cell structure dominates the
equator northward from November to May (see Table V).
The four-cell
structure exists south of the equator, but expands northward to 9.90N in
June through September. The four-cell structure again dominates in the
latitudes
further north
than 9.8°N, while the six-cell circulations
are prevalent south of 9.9°S latitude.
In August only two cells appear
at 14.7N and at 19.6°S (one in both the Western and Eastern Hemispheres).
From November to April, as many as eight cells can appear (an extra pair
forms over Indonesia).
Some of the anomalies of the yearly data are listed below (refer
again to Figure IV for the mean behavior):
In October, 1969, a broken structure (many scattered tenuous cells
was observed in the Eastern Hemisphere (E.H.)
The January, 1979, data displayed a strong equatorial African surface
cell and a strong southern mid-Pacific cell.
In April 1970, a large
positive-valued upper level circulation appeared in the S.H., which
lasted through July.
In October, 1970, the structure returned to
-1
normal, but the intensities were high (12 to 15x 1012 g sec ) in
the southern Pacific.
The southern Atlantic cell was weak in January and April, 1971.
The
Peruvian cell was strong in January, and the Indian one was strong in
-18-
April.
October, 1971, displayed very high intensities (28x1012g sec-1 in
the equatorial Indian cell).
A negatively valued cell dominated the Atlantic and the Indian cell
was strong throughout 1973.
In 1974, the Indian cell deteriorated into a very weak cellular
structure, but it still contained very high intensities (23x1012 g
sec -'
in January).
The structure in the early part of 1975 was characterized by scattered tenuous cells in the Pacific and Indian Oceans.
This scattered structure and variability continued in 1976 (see
Figure V).
The October structure of the Pacific and Indian regions of both 1978
and 1979 were characterized by high intensity cells (12 to 18x10
sec -' over India ;
13x10 1 2 g sec -1
over the 1978 northern Pacific,
and 17x 1012 g sec-' over the 1979 southern Pacific).
-19-
12 g
V. COMPARISON AND DISCUSSION OF MONTHLY AND MEAN RELATIVE
ANGULAR MOMENTUM
The absolute angular momentum of the atmosphere is made up of the
angular momentum due to the air moving with the angular velocity of the
earth, and the relative angular momentum due to the zonal wind speed with
respect to the earth.
The total angular momentum of the entire earth and
atmosphere isconserved and changed only if external pressures or torques
affect it (such as tides).
The angular momentum of the atmosphere alone,
however, is affected by forces from the earth, and vice versa.
The predominant easterlies that exist in the tropics provide a negative torque (through friction) on the earth.
The earth, in return, provides
a positive torque on the atmosphere and creates new atmospheric angular
momentum.
In the mid-latitudes, the predominant westerlies create the
opposite effect and provide a sink for atmospheric angular momentum. Now
the atmosphere conserves its momentum by ferrying relative angular momentum
out of the tropics, mainly by means of the Hadley Circulation, and carrying it poleward through the mid-latitudes with large-scale eddies.
This flux of angular momentum towards the poles shows seasonal variations with the maximum in the winter.
There is also an interhemispheric flux,
which generally moves from the winter to summer hemisphere (Newell et al,
1972).
Bjerknes (1966)
assumes that warm equatorial waters will speed up
the Hadley Cells, and that their poleward movement of angular momentum will
create stronger mid-latitude westerlies.
This assumes that the most evap-
oration, and thereforelatent heat liberation,
-20-
will occur over
this warm
water (a debatable point, see Cornejo-Garrido and Stone, 1977).
Also, the monsoonal easterlies affect the size of the flux into
the Northern hemisphere (N.H.).
(The easterlies can make the N.H. total
relative angular momentum negative.)
This flux moves directly down the
gradient into the N.H.
When the author looked at the relative angular momentum (RAM) between
48.1 0 N and 48.1 0 S for the levels 700-200mb, for the years 1969-1971 and
1973 - 1979, she found a general agreement with the normal cited by
Newell et al (1972); see Table II. However, though the negative RAM and
the maxima
and minima
occurred in the usual places, the numbers were
found to be a little high when compared with Newell et al (1972).
although the
And,
southern hemisphere (S.H.) to N.H. ratios were comparable,
the totals (N.H., S.H., and Globe) were also high.
The author suspects
that the difference is due mainly to the thinner latitude belts used
(catching more of the peaks) and the possibility that the NMC data (u)
may be slightly high (Lau and Oort, 1981).
Even though the data showed
basically the same pattern, there were some differences between the
individual years and the mean.
An anomaly is defined as having a value that differs from the mean
by more than 20 x 10O3 g cm2 sec-1.
The mean maximum value of the N.H. occurs in February at 28.7°S
with a value of 259 x 1030 g cm2 sec -'. Some high anomalous
peaks occurred in 1970, 1977 (see Table II), and in 1978.
maxima
N.H.
The N.H.
of both 1970 and 1977 occurred earlier than usual (January).
The mean maximum value of the S.H. occurs in July, at 28.7 0 S,
-21-
with a value of 221.3 g sec -1 . Low S.H. maxima
1973; high S.H. maxima
appeared in 1971 and
occurred in1978 and 1979.
The mean minimum occurs in August at the equator, with a value of
-85.4 x 1030 g cm2 sec -'.
in 1973 and early minima
The minima
A high minimum occurred in 1977, a low minimum
in 1974 (July), 1978 (June) and 1979 (July).
in 1973 and 1979 were also further north than usual (5°N).
When the RAM for the levels 1000-200mb for the years 1976 - 1979
were inspected, it was found that all the individual years and the mean
(see Table III) had a negative summer N.H. total (in either July or
August, or both).
The only other difference between the two means (except
for the increase that the addition of the lower level caused) was that
the Global maximum
results.
was in December, not January as in the 700-200mb
Some of the differences and similarities that the individual
data showed, when compared with the mean and its 700-200mb counterparts,
are listed below:
In 1976, the S.H. maximum was in September, not July.
The 1977 results were the same as the 700-200mb version, with
the following additions.
The S.H. maximum was early (June),
and the minimum was further south than usual (5SS, see Table IV).
In 1978 added anomalies included the observations that the S.H.
maximum was high and the minimum came in August (like the mean)
but was 50 further south.
In 1979 the minimum occurred in both June and July, at the
equator.
-22-
The results of the N.H., S.H., and Global totals will be discussed in the next chapter.
-23-
VI.
INTERRELATIONSHIP OF RELATIVE ANGULAR MOMENTUM AND ZONAL
MASS FLUX WITH PRESSURE, SST, SOI AND THE STRENGTH OF THE
MO SOON
Now we must
intercompare the results of this study and see how they
relate to other authors' findings and theories.
In order to do this,
anomalies of the N.H., S.H., and Global values of relative angular momentum (RAM) were calculated and plotted as time series (see Figure VI).
A
crude measure of the intensity of the zonal mass flux was devised by
listing the maximum value of the Indian Ocean positive cell and western
Pacific positive cell in January, April, July and October, for both the
N.H. and S.H.
A mean was calculated and the anomalies were plotted as
a time series (see Figures VII and VIII).
The SOI is represented as the
difference in pressure between Tahiti and Darwin (see Figure IX).
SST time series is taken from Newell (1979) (see Figure X).
The
The monsoonal
rainfall graph (Figure XI) is from Parthasarthy and Mooley (1978) and
the surface pressure maps were taken from Godbole and Shukla (1981);
see Figures XII and XIII.
These pressure maps indicate that in January, 1969, the N.H. highs
were higher, while the S.H. highers were lower than the mean (this supports
the findings of van Loon and Madden, 1981).
Atlantic low was extremely low ( -16mb).
In January 1970, the North
In 1972, the North Atlantic low
was again at a minimum (-16mb) as were the South American (4mb) and the
Indonesian lows (4mb).
In 1973, the North Atlantic low was still low (-12mb)
and reached a severe minimum in 1974 (-24mb).
the South American low was very low (-24mb).
-24-
In January of 1975 and 1976,
The July pressure pattern in 1970 showed rather high S.H. highs.
In 1974, the S.H. highs
w2re again a trifle high (24mb), while in 1976
the North Atlantic high was low (22mb).
The SOI (see Figure IX) generally shows agreement with the pressure
maps.
Negative SOI values mean high Darwin and low Tahiti pressures.
Severe minima
occurred in 1969/70, 1972, and in 1976/77.
There were
very high peaks in 1970/71, 1973/74 and 1975; lesser peaks occurred in
1968 and 1978.
The SST displayed high maxima in 1965, 1969, 1972 and 1976 (El Nino
years; see Figure X).
This rash of warm spells followed an era of more
stable sea surface temperatures.
The 1972 event was the most spectacular.
The monsoon time series, though a little sparse of data in this
time period, lists very dry years
occurred in 1961 and 1975.
as 1965 and 1972, while wet years
The El Nino years of the southeastern Pacific
are the dry monsoon years in India.
The
anomalies of N.H.
RAM, S.H. RAM and Global RAM all seemed to
occur at nearly the same time (see Figure VI).
In order to find out just
how the various times series were related, a cross correlation was performed between the SST, the SOI, the N.H. RAM, the S.H. RAM, the Global
RAM, and the zonal mass flux of the N.H. and S.H. Indian Ocean and the N.H.
and S.H. pacific Ocean (see Table VI).
related (-.575).
The SOI and SST are negatively cor-
The Global RAM and the SOI are also negatively correlated,
but the best correlation occurs when the SOI is lagged by one month (-457).
(see Table VII).
This is a negative lag, so an eventappears in the SOI one
month before the RAM displays it. Since the SOI and SST are correlated
with no time lag, distinct SOI and SST anomalies seem to occur first,
-25-
before the RAM changes.
The correlations between the N.H., S.H. and Global RAM are listed
in Table VI.
(Note the +.833 correlation between the S.H. and the Global
RAM.)
The zonal mass flux anomalies of the northern Pacific, southern
Pacific and southern Indian Oceans are in fairly close agreement with correlations from +.475
to +.62.
However, the northern Indian Ocean data do
not correlate as closely with the other three sets of zonal mass flux (correlations of only +.288 to +.393, see Table VI).
The various zonal mass fluxes are negatively correlated with Global
RAM.
Therefore, the Hadley Cell is strong when the Walker Circulation is
weak, and vice versa.
The Indian Ocean correlations with the Global RAM
(N.H. = -.345 and S.H. = -.368 at a one-month lag) are not quite as strong
as the Pacific zonal mass flux data, but do display the same inverse variation.
In order to investigate this negative correlation that the zonal
mass flux has with the RAM, the zonal mass flux patterns from Chapter IV
were compared with the mass flux patterns of Chapter III.
Although we
found the intensities of the mass flux to be unrealistic, the patterns
seemed to agree with other authors' work, so some value could be assigned
to them.
The variation between the Walker and the Hadley Circulations does
appear in the patterns; one is weak and tenuous when the other is strong.
There is, however, a modification of this behavior in the northern Indian
Ocean.
Here, as intimated by the lower cross-correlation, the inverse
variation between the two circulations is not as strong as it is in other
-26-
parts of the tropics.
does not weaken,
the surface.
When the Walker Cell is strong, the Hadley Cell
but exists at higher levels and does not extend down to
When the Walker Circulation is weak the Hadley Cell is
moderate, but does reach down to the surface.
As to which circulation
changes first, the correlations provide support for the Walker Cell
anomalies starting simultaneously with both the SST anomalies and the
Global RAM (the strongest link, however, is with the Global RAM).
From this web of interlocking data, a fairly coherent scenario
emerges.
When the pressure off the coast of South America is high,-the
SST in that area is low (see Fig. XIV).
Pacific and cold in the Indian Ocean.
The SST is warm in the western
The pressure is low at Darwin,
the Walker Cells are very strong, and the Hadley Circulation is weak.
The easterly winds in the Pacific would be strong, also.
The extrema
in the SST occur first (with the SOI extrema) followed by the minimum
in Global RAM.
The maxima
in the various segments of the Walker Cell
(ZMF) occurs at the same time as the SST changes or the RAM changes.
(The correlation coefficients support both times, but favor the RAM
peaks.)
The opposite scenario occurs when the SSTs off the coast of South
America are high (an El Nio), and the pressure is low; see Figure XV.
The tropical Walker Cells are weak, and the Hadley Cells are strong.
In the western Pacific, the SSTs are low and the pressure over Darwin
is high.
The scenario of the above model agrees with parts of Bjerknes'
hypothesis (warm South American seas, a strong Hadley Cell and a low
SOI).
The peak of the Hadley Circulation, or at least of the relative
angular momentum, occurs one month after the peak SST.
-27-
The Hadley
Circulation weakens after this time, and the Walker Cells gain
strength (accompanied by increased Pacific easterlies).
Bjerknes had
thought that the warm water in the central Pacific would increase the
evaporation, and therefore the winds and the Hadley Cell.
Cornejo-
Garrido and Stone (1977), however, have demonstrated that evaporation
is at a minimum in this region of warm water.
They feel that the max-
imum in condensation heating is associated with areas of convergence,
not merely warm seas.
The author tends to agree with Cornejo-Garrido and Stone, who
proposed that latent heat was a driving force for the Walker Circulation.
Newell, Ebisuzaki, Selkirk (personal communication) have evidence for
an area of convergence in the mid-Pacific that is enhanced by the El
(\j
Nino water as it stretches westward.
This area may provide the Hadley
Cells with the extra energy necessary to reach a maximum after the El
Nino has begun.
If this is true, then it is possible that the strong
Walker Cells of Figure XIV may receive more condensation heating from
the area of warm water and strong convergence in the western Pacific.
What triggers the strengthening of the Hadley Cells and the weakening of the Walker Cells is unclear.
Weare (1979) showed that high pres-
sures over India are followed a month or two later by high SSTs in both
the Indian and the Pacific Oceans.
High pressure occurred in mid-1972
and at the end of 1976, in accordance with the SOI.
Our results,
however, show a direct, not lagged correlation, of SOI and SST.
Weare
further suggested that latent heat may affect the Indian and Pacific
pressures, and that the change in the equatorial Pacific pressure
-28-
gradient would affect the winds, thus causing an El Nino.
However, our
correlations indicate that the SSTs rise before changes occur in the
circulations and winds.
Therefore, although this mechanism may explain
the strengthening and weakening of the cells, it seems unlikely to be a
trigger for the El Nino.
-29-
VII.
CONCLUSIONS
In conclusion, the author was able to construct a scenario of events
in the tropical Pacific and Indian Oceans.
There is a positive correla-
tion between the Global RAM and the SST (with the SST leading the RAM by
a month).
A negative correlation exists between the RAM and the SOI
There is also a strong inverse
(again with the SOI leading the RAM).
variation between the Hadley Cells and the Walker Cells.
This variation
usually occurs at the same time, although there is evidence that some
of the Walker Cells occasionally strengthen earlier.
Namias (1976) found
increased westerlies during an El Nino, and assumed they were due to an
increased poleward atmospheric RAM.
The results of this study would tend
to support Namias and disagree with Chiu and Lo (1979) who did not find
any such transport during an El Nino.
The problem of the triggering mechanism of the El Nino is still
unresolved.
Our results suggest that it is not either the Walker or the
Hadley Circulations that triggers the El Nino, but pressure changes.
is supported by the aborted El Nino of 1975.
This
The RAM and the zonal mass
flux showed extreme variations (the Walker Cells were high while the Hadley
Cells were weak), but the El Nino was very weak.
The SOI, however, showed
that pressures over the Indian Ocean and Indonesia were not as high as in
other El Nino years.
This, perhaps, is the reason for the aborted event.
Clearly, more such investigations, with larger and more accurate data
sets,
are necessary.
-30-
ACKNOWLEDGEMENTS
The author initially became interested in this topic while working
on an Undergraduate Research Opportunity Project (UROP) for Prof. R.E. Newell.
The author is extremely grateful to Prof. Newell for his guidance, enthusiasm, and encouragement.
He purchased the original data set with help
from the Department of Energy, and financed the computer time through
the National Science Foundation.
Thanks are due to Rennie Selkirk, John Anderson, and Diana Spiegal
for dealing with the problems associated with the original data tapes.
Rennie Selkirk also donated some of the time series in this thesis.
Additional thanks are due to Robert T. Willis, III and Richard W. Schaaf
for help with the computer work; to Kenneth G. Powell for his mathematical
and theoretical help, to my family, and to Evelyn Holmes for typing the
manuscript. Also, thanks to Sharon Gould-Stewart for daytime, and
David Albany for late-night sympathy and Alfredo Navato for last-minute help.
Again, the author is extremely grateful to Richard Schaaf and to her
family for their love and encouragement.
-31-
BIBLIOGRAPHY
Arkin, Phillip A., "The Relationship between Interannual Variability
in the 200mb Tropical Wind Field and the Southern Oscillation",
Graduate Thesis, University of Maryland, 1981 (1-25).
Bedient, H.A., W. G. Collins and G. Dent, "An Operational Tropical
Analysis System", Mon. Wea. Rev., 95, 1967, pp. 942-949.
Bergman, K. H., "Multivariate Analysis of Temperature and Winds Using
Optimum Interpolation", Mon. Wea. Rev. 107, 1979, pp. 1423-1444.
Bjerknes, J., "A Possible Response of the Atmospheric Hadley Circulation to Equatorial Anomalies of Ocean Temperature",Tellus, 18,
1966, pp. 820-829.
"Atmospheric Teleconnections from the Equatorial Pacific",
Mon. Wea. Rev. 97, 1969, pp. 163-172.
Chiu, W.-C.,
and A. Lo, "A Preliminary Study of the Possible Statistical
Relationship Between theTropical Pacific Sea Surface Temperature
and the Atmospheric
Circulation", Mon. Wea.
Rev., 107, 1979,
pp. 18-25.
Cornejo-Garrido, A.G., and P. H. Stone, "On the Heat Balance of the
Walker Circulation", J. Atmos. Sci., 34, 1977, pp. 1155-1162.
"Objective Analysis of Meteorological Fields, Gidrometeor,
Isdat., Leningrad", (Israel Program for Scientific Translations,
Gandin, L. S.,
Jerusalem),
1965, pp. 1-242.
Godbole, R.V., and J. Shukla, "Global Analysis of January and July Sea
Level Pressure", NASA Tech. Mem. 82097., (1981).
Hadley, G., "Concerning the Cause of the General Trade Winds", Phil. Trans.
26,
1735, pp. 153-168.
Julian, P.R., and R. M. Chervin, "A Study of the Southern Oscillation and
Walker Circulation Phenomenon", Mon. Wea. Rev. 106, 1978, pp. 14331457.
-32-
Kidson, J. W., "Tropical Eigenvector Analysis and the Southern Oscillation", Mon. Wea. Rev., 103, 1975, pp. 187-196.
Lau, N.-C., and A. H. Oort, "A Comparative Study of Observed Northern
Hemisphere Circulation Statistics Based on GFDL and NMC Analyses,
Part I: The Time-Mean Fields", Mon. Wea. Rev., 109, 1981, pp. 13801403.
Namias, J., "Some Statistical and Synoptic Characteristics Associated
with El Nino", J. Phys. Oceanogr., 6, 1976, pp. 130-138.
Newell, R. E., "Climate and the Ocean", Amer. Scien., 67, 1979, pp. 405416.
SJ.W. Kidson,E. G. Vincent and G. J. Boer,
The General
Circulation of the Tropical Atmosphere and Interactions with Extratropical Latitudes,
1 and 2, 1972 and 1974, The MIT Press,
pp. 258-
371.
Parthasarathy, B., and D. A. Mooley, "Some Features of a Long Homogeneous
Series of Indian Summ-er Monsoon Rainfall",
pp. 771-780.
Mon. Wea. Rev. 106, 1978,
Ramage, C. S., and A. M. Hori, "Meteorological Aspects of El Nino", Mon.
Wea. Rev., 109,
1980, pp. 1827-1845.
Rosen, R. D., and Salstein, D. A., "A Comparison Between Circulation
Statistics
Mon. Wea.
Computed from Conventional Data and NMC Hough Analysis",
Rev., 108, 1980,
pp. 1226-1247.
Schell, I. I., "On the Nature and Origin of the Southern Oscillation",
J. Met., 13, 1956, pp. 592-598.
Trenberth, K. E., "Interannual Variability of the 500mb Zonal Mean Flow
in the Southern Hemisphere", Mon. Wea. Rev., 107, 1979, pp. 1515-1524.
Troup, A. J., "The Southern Oscillation", Quart. J. Roy. Met. Soc., 61,
1965, pp. 490-506.
-33-
van Loon, Harry, and Roland A. Madden, "The Southern Oscillation,
Part I: Global Associations with Pressure and Temperature in
Northern Winter", Mon. Wea. Rev., 109, 1981, pp. 1150-1162.
Walker, G. T., India Meteor. Dept. Mem.,
, India Meteor.
24, 1923,
pp. 75-131.
Dept. Mem., 24, 1924, pp. 275-332.
, Mem. Roy. Meteor. Soc.,
2, 1928, pp. 97-134.
, Mem. Roy. Meteor. Soc., 3, 1930, pp. 81-95.
, Mem. Roy. Meteor. Soc., 4, 1932, pp. 53-84.
, Mem. Roy. Meteor. Soc., 6, 1937, pp. 119-139.
Weare, Bryan C., "A Statistical Study of the Relationships between Ocean
Surface Temperatures and the Indian Monsoon", J. Atmos. Sci., 36,
1979, pp. 2279-1403.
Webster, P. K., "Remote Forcing of the Time-Independent Tropical Atmosphere", Mon. Wea. Rev., 101, 1973, pp. 58-68.
"The Dynamic Responses of the Equatorial Pacific Ocean to
Atmospheric Forcing", _J.Phys. Ocean, 5, 1975, pp. 572-584.
Wyrtki, K.,
-34-
equator
30
Isa
6aster
.180'
1504W
120'W
Pacific Ocean
F
GURE
I.
(D-TAr
WA%-KER
Or THE
FROm
FROMRNEW
WORK
ELL
Africa
Atlantic Ocean
South America
D AGP0M
30*E
06
30*W
60'W
900W
sY
(1979))
-35-
Indian Ocean
CIRCULWATION
BOER
AN
90*E
60*E
K-uLE;
IN
120'E
Australia
T-,NU 1Rj
DiRcRtw\
U
50
S+
1 OE
----1
50051
CD)
I
3Ou0
o
-
!80w9o
R,
I
ou
OE
e n, (n 1
,..so w
,o
h-
(s) ~u
E
i so E
0
R
j0
r
:O0
mb
---- r- ---
-
-
DECEMBER MASS
FLUX
-- r
-r
JUNE - AUGUST
FEBRUARY
"
1012 gm sec 1
MASS
I
-.
LUX
74
a o
IA
!,4
-,
/
,5
S
90N
0
70
60
LATITUDE
LATITUDEOC
FiGv
t JE.
-e
-20
- 8
"c
50
f
40
30
I
1
20
t
10
0
t0
LATIITU0
20
30
LATITUDC
SEASON LN
MAss
FLuy.
(DAG-AM
OM
EVOC ,'FR t a
(19 7 )')
40
102
qm
s-
"
f
C) T
AP~~T
I
OVV
,sowviWO
100
20
too
20
100
c.q s
1000IN-2
) g0w
160e
160w-
0
/sow
a
60E
50,
EG~r-E
M.EI
M~pN ar4jM~ss
AR -
LL'< AN~O
FL ux
OCr-OSER
c G-19 )
(V7
UNITs:
F-0OP
W--
AIIIU~~
-sec-'
rcZn
180w
J6~C
O CrosYa
TU M
APRIL
T AN
10001
200
I(00 01--
6Ow
180 w
FIGURE
.
ZoNaL
MAss
FLux
IN 1\1(
UNITS: 1012
se
50
..
18ow
C
(A)>
<
0,
O
0
J/\I
fvl
r
M
100
200
300
1 68
,
6
I
70 1 712
79 I 75 I 7d f 77 1 78 I 79 1
72 1 73
'3
V)
(c)
. A-)" AA
.
-An
A
lu
V>
"3
V
V)
S6 8 t 6~9
70 1 71
755 1 76
72. r 73 i 7
' 77 I 76
1 79
"3
(C)
0J
V
i 8
I 69
R5TIVH
(3)
70
71
AN1-VLtrN
rHE Nlo~newtc
72
73
MOPCELM
79
75 f '4
ATIU
I
aM)5rner
HAw
,
AAI
C
CC)
77
'78
FOX
TOALVSt
(A ) T
THE
OtrrHBEUAl
77
UHE
LOGE
NEMSPHERE
(A)
A
'3
A
U.'
I A T 0
68
Ir
69
t
I
t
I
70
71
72
73
70
7
72
73
7t
I
7'5
I
1
1
7
77
I
I
1
77
78
78
79
0
(8)
0
O
8A
To I
68
I
MNAs. F-OC A
TE IWOLAX/ 0CTAA1/
ZOMA L
(A)
I
I-16E
AAO
(13
0aOF-
I
I
71
75
tHE
NoHrE v
7
THE PaOFic OcE,A/
t
[]EMISP4EXKE
79
1i%
10
(A)
A __t AA
_ I,
__ _
/V
/
_.
v
A
__
___
v
V
1_L_
__
__1
I ~
10
,.rTA
I
0I
68
69
70
,
71
72
73
74
75
76
77
78
75
I -
13)
10
J0-AY0
' 68
FiufE
IiE
'
9 ' 70 ' 7/
Zomj, NAss F ux
(A)
THE
AlA
OCE.AA
' 72 ' 73 ' 7
Or TH
4oMAjESsq
A W3
' 75 ' 75 ' 77 '78
(P)
So
5
WTHE
PkCIFt
fHE-RA
OCEAM
'79
4EMI SF,4ERE
,
3
I
zI
1
3
68
1 69
SoUrHne&
lAmnrrxr
70 t 71 1 7.2 1 73
OScuaa-rIoM
Aivo
DARww
i IDEX
r 7
7
PDESuE
7
77 1178
IFFIEEL2v
7
ar
WEE
El Niho occurrences
2
mean sea-surface temperature
0.86
0.43
AK
-
1 -
EE
-2
:-:r
osph-~ric f:ee
air temperature
-3
a e
7v
n
major volcanic eruptions
FiaURE
-
0,43
-0.86
SUvAcE
NON EASoNAL
OVER
TrE
FOR
THE
RECGtON
SEAN
t40
LATIowE.
0
W
>EL7T
-44-
T
coAsTr
ro
0
o -
5S
eRGcE
us
vPEA TURE
SoOu-r AkC2cA
or
o
Ito-F
I~.42
I~7S
I
too
I
I
1~
-
'I
Ii
Aft/~-Li
I
A
(uG009&"CMS)
r
I
I
ILI
40
YEARS
Fc ,u Re
-\TUMP
lRPkiNVPALL
VN-UE.5
P~NP
OIP
APFE
-
Ft LTv-9ic
APPt
RorA
-45-
tAO
M
ooN
t90-197
')
kTN-
'Et PTBE)
1)
n
r
c~_
----
(1961-76)
JAN. MEAN
---- --------
---. -- ,T
-~
--' r--l
'1
I ----1
T -----~-- -~
---
I-- ...
8':-:TC~~
,--24
12
RC-C-
'.7-,.,
d-~
-I
r ::::I:::
C--C
1
12
--- _
-- --
i
20
20
~ "'
/---
A
-
I
L
,..
T;ILC7-~
I.,
12-
-I---rCI
-t------trr~----t--)L1Cf~~
~L
-16
7
a....
. .
- 7-r
---
_
180W
160
FlGURE
-_
100
80
ANR
MZL.
EAN
N
GOO
AMD
60
12
I
A..
'
120
.'
-F '
&
~-
-
-
S:
*--,
9
-o ----
_ 40..
140
7a
- .
r'
16
4
1-
2,
24
2HN,
T
11
-1
':ie
44
, 8
r-------
~-~-rr
,--,,
4
24
-----------
r
40
SEA
IAUYLP(
20
0
L-PjESS&PRs'
1980
40
20
60
Dt5-vRlUIoN
U
ITS
rb
80
:i
1:
100
(CN"t
1
:, A
120
140
FROM
160
0eor
(1961-76)
JULY MEAN
.............
-
i
I
'-- 7-
t
r,
I
I
lmmmL
I-- .. ,71
,,
80
N
HY'
1 -
161
----
---
12
160
FtcGOURE
120
140
.
12
- - - - - -4
o/21
16
l1
0
-------
12~i9
180w
04
--
e
100
EAM
80
Tj
ODOLE
60
40
SEA
LEvRE
AND
SHUHLA
20
0
P-lUSLP%
0931))
tso
------ -----77
~t
40
20
14
:
:i:::.i:7t7777lj~iiii:
~
60
80
100
.
!
120
140
UDT
IsmonT
UNkTs:
mb
......
160
1I0E
S=. 538 (-1)
+
+L
7'-
WA~'LKER CE~LL STR
EQ
57T COLD (-)
,RAMIZ MIF -7.47 '
ZMIF/.501
zv/5o1: .3
L
15
Ft Gu9,e:
C-f)
/37Z
qol1/o
RAM
ZM F
£
w
D IA.GRNM
OF
=E-T~~
soNNi-
.39'?C1i
.536-)
TRO'P%L-
NTL
I
kGuL-PR~ mAomewTU
M N'SS'
F LU Y,
qoo
1350
M~
LPS
A/-B
=-C0.S\P-lN
(--)"B
coE5:r-cA\q-YqT
LERDS A% BNnAONTW5X
'I
7L~
-7'CF1LL
A/LKER CELL WAI&A
WAK~~ C-)
COOL
-I76'V
loQ.
55T WARMtC+)
.3 S3 c- Z)
Rtc~urAE
+
q 00
1q5
=I.
DIAG9,r.
Rk\M=
zo
ttAF
1.350
OF
FE feeW
TPgopicpoIN
RF-j-Tk\1E
OPtL-
Pt4GULAiZ
mP-,
FLC
R~PTOSrIP$
mA1A-ruA
qo
13 5"
ULRMG
70
Ei N in~
AmN
COVxR-:oTN
B
Lr.IC)S
C0EFI7OA'TA
ENn
NT"
u
=
mean eastward (zonal) wind component
SST
=
sea surface temperature
v
=
mean northward (meridional) wind
component
SOI
=
Southern Oscillation Index
RAM
=
relative angular momentum
ZMF
=
zonal mass flux
N.H.
=
Northern Hemisphere
zonal average of u
[u]
S= L
2T
u(X,,p,t)dX
0
[V]
=
zona) average of v
S.H.
=
Southern Hemisphere
u
=
deviation from zonal average
E.H.
=
Eastern Hemisphere
W.H.
=
Western Hemisphere
u*
= u- [ ]
=
mass flux
=
latitude, positive northward
=
longitude, positive eastward
=
radius of the earth, 6.371 x lOm
=
magnitude of the acceleration of
gravity, 9.81m sec 2
pressure
TABLE
I
-
M
5I AbOLS
ND
AB
REJ IAT\ONS
MEAN 200-700
48. 1
44. 7
41.0
37. 2
33. 1
28. 7
24. 2
19. 6
14. 7
9. 9
5.0
0.0
-5. 0
-9. 9
-14. 7
-19. 6
-24. 2
-28. 7
-33. 1
-37. 2
-41. 0
-44. 7
-48. 1
TOTAL
TOTAL
GLODE
NH
SH
TAB LEL
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
DEC
67. 8
92. 4
133.9
178. 9
223. 8
255. 7
235.0
169. 4
68. 6
0. 6
-31. 0
-26. 5
-15. 6
-10. 1
-10. 5
5. 8
41.0
83. 6
115.7
132. 9
138.6
122.2
111.7
69. 6
97. 3
140. 2
184. 8
227. 6
259. 7
238. 2
172. 2
68. 1
-3. 8
-41. 1
-41.0
-29. 3
-15. 1
-11.9
2. 9
35. 2
73.0
99. 2
114. 5
124. 8
122. 5
116.2
65. 8
90. 1
129. 2
169. 6
205. 4
230. 6
215. 7
161. 3
69. 8
-3. 7
-40. 1
-38. 9
-27. 1
-17. 8
-6. 9
23. 0
62. 8
96. 2
113.0
116.2
120. 9
113.9
109. 4
62. 0
84. 6
118. 1
146. 2
163. 8
172. 0
158. 5
126. 0
59. 4
-4. 3
-42. 7
-44. 5
-33. 1
-8. 0
31.7
89. 0
124. 1
137. 5
129. 6
120. 6
120. 2
103. 8
101. 0
61. 5
80. 6
104. 4
119.0
123. 6
124. 2
106. 2
72. 3
25. 2
-19. 5
-48. 0
-52. 1
-37. 3
3. 7
67. 1
140. 1
174. 8
181. 6
164. 1
138. 1
124. 2
90. 0
87. 1
53. 4
68. 8
90. 3
104. 0
98. 8
71.6
27. 4
-9. 5
-31.4
-51. 2
-66. 8
-70. 8
-57. 0
-9. 3
66. 0
153. 7
199. 4
211.3
183. 9
150.6
128. 2
93. 8
03. 6
60. 8
71.6
80. 2
72. 6
45. 3
7. 2
-23. 6
-39. 2
-51. 6
-67. 2
-78. 5
-79. 2
-64. 6
-17. 7
57. 5
152. 2
205. 8
221. 3
192. 2
154. 2
132. 0
88. 8
81.0
67. 2
76. 9
81.0
65. 6
33. 9
-1. 1
-24. 7
-37. 2
-52. 6
-68. 0
-81. 3
-85. 4
-68. 4
-18. 7
57. 7
149. 9
204. 2
214. 1
184. 7
148.7
128. 7
92. 0
85. 0
75. 4
86. 0
95. 8
90. 6
69. 2
37. 1
1.9
-25. 5
-45. 3
-59. 9
-71. 5
-72. 9
-59. 9
-16. 5
55. 9
146. 4
194. 0
209. 2
181. 7
147. 3
129. 3
96. 2
91. 4
84. 3
97. 3
112.2
118. 1
114.9
97. 8
58. 3
13. 7
-22. 6
-44. 3
-54. 2
-47. 0
-34. 1
0.3
56. 8
128. 6
167. 1
182. 3
166. 7
141. 4
128. 0
94. 9
91. 1
85. 3
99. 8
121. 3
139. 9
154. 2
156. 1
120. 8
64. 3
4. 5
-30. 9
-41.2
-31.9
-24. 2
-2. 8
38. 5
98. 0
135. 7
158. 7
156. 3
140. 9
130. 1
102. 4
96. 1
78. 7
102. 8
139. 2
173. 3
201. 8
213. 4
177. 6
112.8
32. 0
-17. 5
-34. 3
-24. 8
-18. 5
-7. 5
16. 3
55. 1
89. 5
120. 1
139. 5
142. 1
141. 5
110. 4
102. 5
1301.0
702.0
2083.9
1392.3
611.5
2003.7
1274.3
663.9
1958.2
1021.3
894.1
1915.3
723.6
1107.4
1831.0
320.0
1168.6
1488.6
38.1
1163.0
1201.1
17.1
1135.0
1152.1
217.5
1130.5
1356.0
552.1
1099.5
1651.5
858.0
1013.5
1871.6
1167.4
878.5
2045.9
-
MEAN
..EVEL.S
Fr.ATVE
ANG~ULAR
700-200
mb
M
MENTU
( \99
- 1911,
UNiTs
1979)
\9IT: 130
gc
m
FO R
sec
.I
MEAN 200-1000, 1976-79
JAN
1
(n
FEB
MAR
APR
MAY
.
JUN
JUL
AUG
SEP
OCT
NOV
DEC
48. 1
44. 7
41.0
37. 2
33. 1
28 7
24. 2
19. 6
14. 7
9. 9
5.0
0.0
-5. 0
-9. 9
-14. 7
-19. 6
-24 2
-20. 7
-33. 1
-37. 2
-41. 0
-44. 7
-48. 1
74.0
100. 9
146. 3
199. 4
250. 5
279. 8
241.7
156. 2
40. 3
-40. 4
-64. 5
-42. 3
-9. 1
-1. 6
-14. 6
-11.9
21 1
72. 5
122. 6
159. 4
172. 7
160. 7
143. 7
79. 3
112.9
162.4
213. 3
254.0
272. 5
232. 0
151. 3
40. 3
-44. 4
-76. 3
-62. 4
-29.2
-16. 1
-26. 7
-21.6
16.3
67. 9
109. 0
137.8
153.6
152.0
146.2
80. 4
103. 1
138. 3
172. 5
201.7
224. 6
206. 3
146. 2
48. 0
-31. 8
-66. 2
-53. 0
-30. 0
-23. 5
-20. 1
0.0
39. 6
86. 5
118.2
135. 5
144. 5
143. 0
139. 4
65. 6
91.4
128. 9
158. 5
170. 4
172. 6
156. 3
119.2
46. 5
-22. 4
-60. 8
-64. 6
-51.2
-23. 4
21.3
80. 7
122. 6
143.6
138. 4
133. 2
133. 1
128. 4
123. 2
67. 6
86. 3
109. 5
125. 3
131.8
128. 1
102. C
61.7
9. 7
-32.0
-56. 3
-64. 2
-56. 6
-15. 2
57. 1
135. 7
180. 9
201. 3
106. 8
164. 0
139. 2
114. 1
98. 5
58. 1
72. 4
94. 5
107. 8
99. 8
67. 4
10.7
-24. 7
-52. 3
-64.0
-78. 2
-98. 7
-96. 5
-43. 6
52. 2
154. 6
214. 4
237. 0
211.4
176. 3
144. 3
115.3
97. 4
69. 9
81.2
88. 8
79. 0
47. 2
0. 6
-37 5
-60. 1
-69. 5
-74. 8
-85. 7
-100. 5
-95. 8
-43. 2
51.6
157. 8
222. 7
246. 1
216. 4
177. 9
143. 1
111.7
92. 9
76. 0
86. 2
91. 5
75. 9
38. 8
-7. 6
-40. 7
-56. 8
-65. 6
-72. 7
-86.9
-106. 2
-102. 2
-48. 6
49. 5
156. 7
217. 7
235. 0
203. 4
167. 3
138. 8
113.6
98. 1
83. 7
93. 5
101. 9
95. 4
71. 9
34. 9
-6.0
-40. 2
-56. 2
-57. 9
-70. 5
-91.6
-90. 5
-39. 6
58. 7
165. 6
218. 9
232. 4
199. 9
165. 5
142. 9
124. 2
113. 1
101. 5
113. 7
125. 8
129. 2
122. 0
99. 1
55. 8
5. 4
-38. 1
-60. 0
-65. 8
-66. 3
-58. 4
-19. 0
56. 5
137. 5
180. 7
190. 9
103. 1
164 O0
148. 0
130. 8
120. 4
98. 1
114. 1
136. 7
155. 0
163. 7
155. 6
108. 6
43. 0
-23. 5
-63. 5
-70 5
-64. 9
-54. 4
-25. 5
27. 9
94. 3
142. 8
174. 0
176. 9
169 3
156. 5
137. 5
122. 2
93. 6
122. 3
161.3
196. 9
224. 1
227. 6
175. 9
96. 8
9.5
-48. 0
-57. 9
-34 2
-14. 3
-6. 6
14. 7
56. 7
96. 6
140. 0
169. 3
179. 8
171 3
146. 7
123. 4
TOTAL NH
TOTAL SH
GLOBE
1363.1
794.3
2157.4
1366.0
658.2
2024.2
1196.6
706.7
1903.3
993.8
917.4
1911.3
701.6
1173.7
1875.3
250.2
1213.3
1463.5
-11.1
1231.1
1219.9
-15.1
1176.1
1161. 1
204.9
1245.3
1450.2
555.5
1209.4
1764.9
784.9
1089.0
1873.9
1185.2
1060.4
2245.6
TABLE-
TI9
- MEAN
R
rNVE
SGUJLAR
MOMENrTOMV
(97o-9c )
,L) NI %' 5 :
1000- 200 mb
L
For V,~v.s
O1so0
c
"
197/
JAN
4''1
40
41
41
37
24
269 5
I:)
14
107
68
-25
-60
-52
-- 17.
-4.
I
9
6
6
9
2
6
-2
O
-4-9
4,
1
11
-1
--37
-41
-44
-4
1
IOTAL 4NH
IOTAL S11
cLO.E
0E
1977
48 1
44 7
41. 0
33
1
24
2
19 6.
14 7
9 9
(3)
5 0
0.0
--5 0
-9 9
-14
7
-- 19 6
-24. 2
-21,
7
-33 1
--37 2
-41. 0
-41 7
- 4 . I
TOTAL NI
IOTAL S14
GLO01
TmBLc
4
1
0
6
5
1458. 1
918.2
2276.3
1292 4
623 3
1915.6
JAN
FED
55 0
93 3
87. 1
129. 1
192. 3
253. 1
292 5
201 0 6
211.7
104. 1
-11.6
54. 0
225 2
293. 8
3:0. 1
200. 1
178. 2
43. 5
-57 5
-97 -1
-71. 9
-27. 5
-15 7
-19. 3
0. 2
43. 7
100. 2
149 1
180. 3
103. 3
160 6
137. 0
D2O 7
-
56. U
74. 6
96. C;
112 4
122. 5'
131. 5
121. 4
95 6
45. 9
-8. 4
-52 7
-68. 1
--51. 9
-0. 7
85. 5
167. 3
197. 0
204. 4
176. 4
141. 5
113. 2
93. 4
54. 3
44. 0
66.
95
114.
110
70
26.
-17.
-42
-59.
-76
-87
-74.
-22.
70.
168
216
233.
209.
176.
142
109.
117.
61
7.2
77
66.
2O
-16 3
0 1
28. 5
60 2
96. 4
119. 0
130. 2
134. 8
130 9
125. 6
65 0
83. 2
111. 0
130. 2
130. 5
1,149 5
151 I1
139. 5
7. 9
20 8
-35. 6
-6.
4
--45. 3
-12. 8
41. 5
95. 7
119. 6
131. 3
128. 8
129. 2
130. 7
123. 3
112. 1
1121. 1
767. 1
1898.2
1012.8
925.9
1938.7
762.4
1177.2
1939.6
275.9
1272.3
1568.1
77. 7
102. 4
141 0
172. 4
107 1
185 8
155 4
111.6
48. 6
-6
I
-- 36. 8
-36. 1
-24
3
MAR
APR
88. 0
MAY
-99. 8
-65.2
-51. 1
-40. 0
-15. 3
20. 4
68. 4
116.6
153. 1
160. 9
162. 5
152. 2
114. 5
155. 9
189. 5
199. 8
189. 7
147. 3
90. 3
17. 9
--38. 4
-62. 1
-51.4
-33. 9
-28. 4
-18 2
7. 5
44. 8
91. 3
126. 5
148. 2
159. 1
156. 3
150. 3
1284. 1
620.5
.190, 6
1065. 7
777.6
1843.3
957.4
910.6
1868.0
EtLT1IE
rOR
APR
71.0
91. 3
121. 9
142. 1
147. 4
152. 1
145. 4
123. 9
64. 2
-6. 0
-59. 6
-74. 4
-62. 3
-34, 7
17. 9
76. 6
108. 7
130. 5
138. 7
146. 8
151. 5
143. 5
130. 5
-121.5
1462. 3
856.0
2318 3
ILV
JUL
265 7
27O 5
210 2
121 5
20. 5
-54. 7
-89 7
-77. 0
-50 2
-37 6
-21. 9
52
35 8
74. 9
111. 1
135. 9
144. 6
137 0
127. 1
99 7
135
156.
155
134.
114
JUII1
226. 1
18 7
54 6
2
MAY
76 5
114. 0
170 41
04. 2
139 7
EO.. 9
9
5
flAil
FEO
TH
/ NGULNR
E-
1977
63. 1
7
7
9
7
1)
5
04. 6
28. 7
-26. 2
-67. 6
-03. 7
-71
5
-21. 6
02.
106.
121.
120.
132.
117.
67. 1
156 2
198 4
215. 6
193. 9
161. 1
132. 1
109 9
99. 3
730. 8
1178.7
1929. 6
OMENTUAi
2
18
8
4
0
4
3
9
4
5
3
8
4
0
1
8
3
2
6
4
2
5
JUN
47.
71.
102.
123.
120.
82.
19.
-33
7
1
5
8
3
3
6
8
0
-25
AU
5
0
1
9
0
4. 1
1105. 3
1109. 4
8
1
5
0
3
3
9
9
3
5
0
0
5
1
72 2
79. 8
83. 6
70. 7
37 I
-7. 0
-38. 8
-54. 5
-62. 9
-75 4
-92 3
-1 12. 6
-116. 0
-73. 7
15 9
123. 0
200. 1
246. 9
239. 6
208. 5
167. 8
126 3
100. 2
264.0
1306.9
1570.9
-43. 7
1182 2
1138. 5
-bo. 4
-71
-Q6.
-102.
-98.
-48.
47.
154.
217.
248.
233.
204.
167.
128.
103.
FoR
70 7
91 2
75. 8
C8. 0
87
-36.
-508
-71.
-80.
-90
-104.
-101.
-54.
36
141.
206.
233.
207.
171.
135.
101.
-38. 7
-5. 4
-69 1
-850 2
-95 7
-20 1
-41, 9
45. 5
142. 9
203 4
234. 9
217. 7
102. 5
143. 7
107. 5
05. 2
JUL
73. 1
U5 3
4
9
1
Ul'/
I1
(9
7
1
4
4
1
3
8
6
9
7
8
8
0
9
3
1
93 1
103 C
114
6
8
114. 2
105 0
84 0
65. 0
37 6
222
133
149.
14
115
60 5
20. 6
-22. 3
5
-- 55
1
-73
-73
-50.
-21.
9
3
1
0
6'
'75.
181.
90 7
-25. 6
-46. 8
-53. 7
-61. 1
-71
I o/
I
171
-63.
3
-57.
-39Y
150
197
122.
36.
21.
-40
--20.
5
711.U
-71. 3
-19. 8
79. 4
17Y. 0
218 6
219. 8
181. 9
I15 0
110. 2
97. 8
07. 3
145,
197.
196.
173.
152.
135
116.
10o.
-29. 4
1106. 5
1077. 1
201. 4
1190. 7
1403. 1
514.8
1149. 1
1663.9
830. 8
1063. 1
1693.9
OCT
NOV
DEC
99 6
118 8
145. 5
162 1
163. 0
148 -1
106. 0
46 2
-24 2
-68 7
-79 ?
-73. 1
-60. 6
-35. 8
12. 7
76 1
127. 2
169. 6
105. 5
186 4
173 2
1O0. 9
105. 3
115. 6
126. 0
122. 8
10B 8
87. 8
40. 7
-2. I1
-47. 1
-71. 5
-84. 3
-89 1
-83. 8
-- 50. 7
30. 5
129. 5
103. 8
204. 6
190. 2
172. 4
154 7
123 7
119.7
77. 1
105 1
150. 5
199. 2
241. 4
255. 6
200. 1
106 8
6. 2
-53. 7
-64. 7
-33. 5
-3. 4
3. 5
21 4
598 5
09. 1
124. 6
155. 6
176. 1
176.2
154. 3
132. 3
173. 8
1197. 2
1372.9
465. 7
1140.1
1605.8
780. 3
1072.5
1852. 8
AUC
80. 7
93. 1
98. 2
80. 6
38. 9
-12. 6
-50. 0
-71. 0
-78. 5
-02. 6
-94. 5
-120. 2
-128. 9
-85. 1
9. 9
125. 8
207. 4
247. 5
229. 2
194. 2
156. 3
118. 6
93. 6
-57. 13
1108. 4
1050.6
55. 6
SEP
77 3
84. 1
72. 7
87. 8
65 7
32 0
-6 4
-41. 3
-57. 0
-55. 3
-62. 4
-86. 8
-95. 3
-47. 4
57. 4
168. 7
220. 7
230. 9
197.
4
6
1 2. 3
134. 8
112. 2
WtEl
A
-O
700- 200 mb- AND
UN IT
- \0-
- C
35.
1
1
2
9
9
2
0
9
136
167.
-4
171
163.
146
123.
105
1477
127. 0
()
-
15
75
e-
100o0-
123
142.
154
149.
129
110
1230 4
1033 2
2263. 7
1207. 0
1072.2
227p9
OO2 mb
OCT
NOV
DEC
6
6
6
4
2
4
4
4
4
6
4
4
6
6
6
6+
6+
4
4
6
6
6
4+
6
4
4.
6
4+
6
4
4+
4
6
6
6
4+
4+
6
6
4+
6+
4+
6
6
8
6
14.7 0 S
6
6+
6
6
6+
6
4+
6
8
6
19.6 0S
8
8
6
4
4
4
4
4
6+
6+
JULY
AUG
4
2
4
4
FEB.
MAR.
19.6 0 N
4
4
14.7N
4
9.9°N
5.0"N
JAN.
EQ
5.00S
9.9°oS
9,90S
TA B-E 7
NUMLER
APR.
MAY
6
JUNE
6
OF MEAN ZotIAL MAss F'vx CELL
PER
SEPT
MONWTH
P11NOE 1-"1
1979
SO I TD
STI
C I TD
R:H
::H
ZIH
PF':;H
CORRF:ELATIN
r.NH
MATR I
RSH
RALL
ZINH
ZPNH
i. 88
1.00
.534
-.
m'_!.":
8,620-
1 ISH
1.0
20.
.
rLL
ZI NH
1,
-0. Vi.:576
-.
173
:-:
1.
-g3 :34r~
-I.455
-8. H26
.- I*30
0.3•3-0
-8.214
CozR=LATIrON
Zow
14
-0. 405
CEFFICETS
MAss Ft,x AtvMALC-
5H, lyoiv
Ocirt
ANvo
FR
0. 392
T50I, N/H. RAM, S.H
SST,
oF THE
5.H.
- 1~ ::
i,
PAciFIC
H. LIDrA
OCeAA
OCEA, Al
N,
1 -74
0.4,.._
&.oG,4L RAM,
, PAcIFIc OcEA,
RAI"I,
PRrCE 1'="
::
1979
12
C:ORRELATION MAI'RTR I"
,.lt
II F F
RRCG2
Ff:G 1
RALL
1.000
1. 000
-0. 255
RRLL
r. :-,7l5
-0:.34'
-0. 167
-0.215
I.7
TABLE
-0.457
. -44
0 529
J. 26
r0.33
:::
CorICR-LATIoM
I.00-
-0, 122
A
-1
H
Io
NTH
RA 11A
LAG
-03422
5'
C)
,
vERzsv.
AfS
LM
A
FI.3!
....
4-1"'='-
AL.
.- KAI
zoA
1. 000
0.5:3i
8 6;=2
0.274
-L. 38
--.
COEFICIEA/TS OF
/H. RAMhl q.H.
,
H
N.H. PAc-,c OLEM
00
-0. 405
-0. 024
-0. 077
so0,
AT
i.
F-U x
H
(tHE
ofr
PACFIC
Al
Oc
0
1.000
PARAMETF RS
(5T,
IvIRivA Oce
),
WHic
WHIC)
ARE