AN INVESTIGATION OF LARGE-SCALE
HEAT TRANSFER PROCESSES AND SEA-SURFACE TEMPERATURE
FLUCTUATIONS IN THE NORTH PACIFIC OCEAN
by
NATHAN EDWARD CLARK
Sc.B., Brown University
(1962)
SUBMITTED IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF
PHILOSOPHY
at the
MASSACHUSETTS INSTITUTE OF
TECHNOLOGY
May, 1967
Signature of Author........ .-......................--....
Department of Meteorology, May 12, 1967
Certified by................ ........
y
.. .................
Thesis Supervisor
........
Accepted by........................................
Chairman, Departmental Committee
on Graduate Students
AN INVESTIGATION OF LARGE-SCALE
HEAT TRANSFER PROCESSES AND SEA-SURFACE TEMPERATURE
FLUCTUATIONS IN THE NORTH PACIFIC OCEAN
by
Nathan Edward Clark
Submitted to the Department of Meteorology on May 12, 1967,
in partial fulfillment of the requirement for the degree of
Doctor of Philosophy.
ABSTRACT
The atmospheric and oceanic parameters of sea-surface
temperature, air temperature, wet-bulb temperature, cloud
cover, and wind speed are used to compute monthly average
values of the incoming radiation, the effective back radiation, the latent and sensible heat transfers, and the total
heat transfer across the sea surface over the North Pacific
Ocean from 200 N to 550N for the period extending from 1951
through 1957.
The yearly average of the total heat flux across the
0
0
surface is integrated over the ocean from 15 N to 60 N, and
it is found that the ocean loses 7 x 1014 cal/sec in this
area on a yearly average basis.
The 12 monthly 7-year mean or normal values for each
heating term are Fourier analyzed, and it is found that each
term has a regular yearly cycle with maxima and minima separated by 6-month intervals. The yearly cycles of the seasurface temperature are compared to the yearly cycles of the
total heat transfer across the ocean surface, and it is found
that the sea-surface temperatures have their maximum values
near the end of the heating cycle (August or September) and
have their minimum values at the end of the cooling cycle
(February or March). These results agree with the theoretical
results of a one-dimensional model of the seasonal thermocline.
The normal values of the 12 monthly means of the heat
advection in the surface layer of the ocean are determined as
residuals from the surface layer heat balance equation. The
results obtained agree fairly well with what is known about
current patterns of the North Pacific and also with the results
of another independent investigation.
The departures of each monthliy value fro thLe normal values, i.e. the monthly anomalies, are computed for each of the
heating terms and also for the sea-surface and air temperatures
-ii-
for all 84 months. It is found that the sea-surface temperature anomaly patterns are geographically coherent, that they
can be fairly extensive in geographic size, and that on the
average they persist in time to about three months or slightly
longer. The air temperature anomalies are greater in magnitude
but are closely related in time to those of the sea-surface
temperature. The persistence of sea-surface temperature anomalies is related to the dynamics of forced and free convection
within the upper mixed layer of the ocean and to the transfer
of heat through the seasonal thermocline.
The anomaly patterns of the heating terms also show geographic coherence but have little month-to-month persistence.
Correlation studies between the anomaly series of the
month-to-month change in sea-surface temperature and those of
the total heat transfer across the sea surface and also those
of the horizontal temperature advection due to wind drift show
that there are statistically significant relations between
these quantities. The positive correlation is improved by
adding the temperature change due to surface heat transfer to
the temperature change due to horizontal advection and correlating the results with the observed sea surface temperature
changes.
A multiple linear correlation analysis between the three
terms, (1) the observed change in sea-surface temperature,
(2) the sum of temperature change due to surface heat transfer
and that due to horizontal advection, and (3) the vertical
velocity induced beneath the Ekman layer, shows that the percentage of variance of the first term accounted for by relation
to other variables is increased from 20 to 25 per cent in winter, 13 to 18 per cent in spring, 12 to 18 per cent in summer,
and 13 to 19 per cent in fall by including the effects of the
third term as well as the second.
A statistically significant positive correlation is found
between the anomaly series of the sea-surface temperature and
those of the specific humidity of the air, indicating that
the amount of moisture in the surface layer of the atmosphere
over the ocean is a function of the sea-surface temperature.
A statistically significant positive correlation is also
found between the anomaly series of the latent heat transfer
between ocean and atmosphere and those of the atmospheric water
vapor divergence field. It is felt that this type of analysis,
if applied to a smaller area with better data, would be helpful
in obtaining a better transfer formula for the latent heat flux.
Thesis Supervisor: Hurd C. Willett
Title0:
pro
fess
soo-
f
-iii-
TABLE OF CONTENTS
I.
II.
III.
IV.
V.
INTRODUCTION
A. Review of sea-surface temperature
fluctuation studies
1
B. Review of ocean surface layer heat
transfer studies
11
THEORY
A. Development of surface layer heat
balance equation
21
B. Development of heat transfer
formulas
21
C. Error analysis
27
DATA PREPARATION
A. Monthly averaged data
34
B. Daily averaged data
37
HEAT EXCHANGE CALCULATIONS
A. Results using monthly averaged data
38
B. Results using daily averaged data
62
DISCUSSION OF SEA-SURFACE TEMPERATURE AND
HEAT TRANSFER ANOMALIES
A.
Spatial and temporal scales of anomalies
1. Sea-surface and surface air
temperature
2. Heat transfer terms
B. Relationships between sea-surface
temperature fluctuations and heat
transfer anomalies
70
70
77
84
-iv-
VI.
VII.
ADDITIONAL RESULTS
A. Comparison of observed sea-surface
temperature and total surface heat
transfer cycles with theoretical
results
115
B. Relationships between anomalies of the
sea-surface temperature and anomalies
of the specific humidity of the air
119
C. Relationship between anomalies of the
latent heat flux and anomalies of the
atmospheric water vapor divergence
121
CONCLUSIONS
A. Summary of results
128
B. Suggestions for further study
140
ACKNOWLEDGEMENTS
143
BIBLIOGRAPHY
144
BIOGRAPHICAL NOTE
148
APPENDIX:
FIGURES Al THROUGH Al19
-V-
List of Figures
Figure 1.
Graphs showing yearly cycles of total
surface heat transfer and of seasurface temperature (yearly mean
removed) along 20 N.
Figure 2.
Graphs showing yearly cycles of total
surface heat transfer and of seasurface temperature (yearly mean
removed) along 35 N.
Figure 3.
Graphs showing yearly cycles of total
surface heat transfer and of seasurface temperature (yearlX mean
removed) along 35 N and 40 N.
Figure 4.
Graphs showing yearly cycles of total
surface heat transfer and of seasurface temperature (yearly mean
removed) along 500N.
Figure 5.
Graphs showing yearly cycles of surface heat transfer terms and of seasurface temperature (yearly mean
removed) for ship station NOVEMBER
(300N - 140 W) during 1962.
Figure 6.
Graphs showing yearly cycles of surface heat transfer terms and of seasurface temperature (yearly mean
removed) for ship station PAPA
(50 N - 145 W) during 1962.
Figure 7.
Typical anomaly pattern of seasurface temperature, December 1957.
Figure 8.
Typical anomaly pattern of air temperature, December 1957.
Figure 9.
Correlations between anomalies of
total surface heat transfer and
those ofA SST based on 84 months
of data, 1951-1957.
Figure 10.
Correlations between anomalies of
horizontal temperature advection
and those of A SST based on 84
months of data, 1951-1957.
-vi-
102
Figure 11.
Correlations between anomalies of
the sum of change in SST due to
surface heat transfer and to horizontal temperature advection and those
of A SST based on 84 months of data,
1951-1957.
Figure Al.
84-month average values and standard deviations of the incoming
radiation corrected for reflection
and cloud cover.
Figure A2.
84-month average values and standard deviations of the effective
back radiation from the sea surface.
Figure A3.
84-month average values and standard deviations of the latent heat
transfer between ocean and atmosphere.
Figure A4.
84-month average values and standard deviations of the sensible heat
transfer between ocean and atmosphere.
Figure A5.
84-month average values and standard deviations of the total transfer
of heat between ocean and atmosphere.
Figure A6.
84-month average values and standard deviations of the sea-surface
temperature.
Figure A7.
84-month average values and standard deviations of the surface air
temperature.
Figure A8.
84-month average values and standard deviations of the cloud cover.
A8
Figure A9.
84-month average values and standard
deviations of the observed wind
speed.
A9
Figure A10.
84-month average values and standard
deviations of the specific humidity
of the air.
A10
Figure All.
84-month average values and standard
deviations of the Bowen Ratio.
All
-vii-
Figures A12
to A23
7-year average values and standard
deviations of the incoming radiation. January through December
1951-1957.
A12-A23
Figures A24
to A35
7-year average values and standard
deviations of the effective back
radiation. January through December
1951-1957.
A24-A35
Figures A36
to A47
7-year average values and standard
deviations of the latent heat transfer. January through December
1951-1957.
A36-A47
Figures A48
to A59
7-year average values and standard
deviations of the sensible heat
transfer. January through December
1951-1957.
A48-A59
Figures A60
to A71
7-year average values and standard
deviations of the total heat transfer. January through December
1951-1957.
A60-A71
Figures A72
to A83
7-year average values and standard
deviations of the sea-surface
temperature. January through
December 1951-1957.
A72-A83
Figures A84
to A95
7-year average values and standard
deviations of the surface air
temperature. January through
December 1951-1957.
A84-A95
Figures A96
to A107
7-year average values and standard
deviations of the change in seasurface temperature between months.
January through December 1951-1957.
A96-A107
Figures A108 7-year average values of heat advecto A119 tion determined from the heat balance
equation. January through December
1951-1957.
A108-A119
-viii-
List of Tables
Table I.
Effect of observational errors on
heat transfer calculations.
Table II.
Monthly average heat transfer calculations using daily average variables and monthly averages of daily
variables for ship station NOVEMBER
(30oN - 1400W).
Table III.
Monthly average heat transfer calculations using daily average variables for ship station PAPA (50 N 1450W).
Table IV.
Integrated values of surface heat
advection determined from the heat
balance equation.
Table V.
Monthly average values and standard
deviations of the heating terms and
sea-surface temperature for ship
station NOVEMBER (300N - 140 0 W).
Table VI.
Monthly average values and standard
deviations of the heating terms and
sea-surface temperature for ship
station PAPA (500N - 145 0 W).
Table VII.
Latitudinal averages of correlation
coefficients between anomalies of
surface heat transfer and those of
monthly change in sea-surface
temperature.
Table VIII.
Latitudinal averages of correlation
coefficients between anomalies of
horizontal advection due to wind
drift and those of monthly change
in sea-surface temperature.
Table IX.
Latitudinal averages of correlation
coefficients between anomalies of
the sum of the temperature change
due to surface heat transfer and
that due to horizontal heat advection and those of the observed
change in sea-surface temperature.
101
-ix-
Table X.
Latitudinal averages of the percentage of variance of the observed
change in sea-surface temperature
accounted for by relation to (1) the
sum of the temperature change due to
surface heat transfer and that due
to horizontal heat advection and to
(2) the vertical velocities induced
beneath the Ekman layer.
107
Table XI.
Latitudinal averages of the correlation coefficients between anomalies of the sea-surface temperature
and those of the specific humidity
of the air.
120
Table XII.
Latitudinal averages of the correlation coefficients between anomalies of the latent heat transfer
and those of the divergence field
of atmospheric water vapor.
125
-1-
I. Introduction
A.
Review of Sea-Surface Temperature Fluctuation Studies
In 1920 Helland-Hansen and Nansen conducted their pioneer-
ing investigation into the causes of large-scale fluctuations
in sea-surface temperatures.
Since then numerous studies have
been made concerning this intriguing and elusive problem.
In
each case, an attempt was made to relate large-scale anomalies
of sea-surface temperature to fluctuations in space and time
of various meteorological and oceanographic parameters, wind
speed, surface drift currents, solar radiation, etc.
Helland-
Hansen and Nansen began by correlating sea-surface temperature
changes in the North Atlantic Ocean with anomalies of the northsouth component of the geostrophic wind.
Their results indicated
that the temperature anomalies were strongly related to the wind
anomalies and that they were not transported to any great degree
by the motion of sea-surface currents.
These results were
supported by the work done by Neumann, Fisher, Pandolfo and Pierson
(1958) and by J. Bjerknes (1962) in the same region.
Bjerknes demonstrated that it is probably the interplay
between the time changes in Qa, the net transfer of heat to
the atmosphere including both latent and sensible heat, and
those of
Q
and Qd, the advective heat supply and heat supply
due to up-welling, that regulates the trends of sea-surface
temperature.
He found that a long trend of cooling north of
-2-
500 N extending from the 1890's to the early 1920's is accompanied
by a strengthening of the Icelandic low, and the warming trend
of sea-surface temperatures until the early 1940's is accompanied
by a weakening of the Icelandic low.
Superposed on the long
cooling trend, rising and falling trends of 2 to 5 years durations
were found.
In these trends the ocean always cooled when the
prevailing west wind strengthened and warmed up when the wind
weakened.
During the long trend of northern cooling from the
1890's to the early 1920's, the surface water warmed up west
and north of the subtropical high pressure zone, presumably
because of increased wind and Gulf Stream advection.
The statistics presented by Bjerknes indicate only that
increased wind speed is accompanied by a cooling of the underlying surface waters and vice versa in some regions of the
North Atlantic.
The other mechanisms postulated for the
strengthening or reversal of these trends must depend upon
physical explanations which are based upon experience obtained
from further investigations.
At present, there is no way to
test the hypothesis that the trend of cooling of the sea surface due to increased west winds is reversed by an increase
in Gulf Stream advection, since heat transport cannot be
determined from the mean mass flow, which is the only quantity
that can be computed from wind-driven current theory.
The results obtained by Neumann et al indicate that the
-3-
same negative correlation exists between the zonal index of
atmospheric circulation over the North Atlantic and the surface
temperature anomaly on a monthly time scale.
Their work also
supports the idea that the temperature anomalies are not advected to any great extent by surface drift.
Correlations
between anomalies (departure from a normal value for a given
month) of sea-surface temperature for adjacent areas at time
lags in 2-month increments from 0 to 24 months were computed.
The coefficients were significantly highest when simultaneous
anomalies were correlated.
This indicates that processes
causing simultaneous changes over large areas, such as evaporation and surface divergence, are probably more important
than advection.
Namias (1959) has attempted to explain the warming of the
surface waters of the Eastern North Pacific Ocean during 1957
and 1958 by relating it to the abnormal velocity components of
the atmospheric circulation.
Although he obtained some of the
gross features of the observed temperature distribution, his
method is open to at least two objections.
The first is that
he considered a normal (time average) state of the atmosphere
for a particular region to correspond to a normal state of the
sea-surface temperature distribution for the same region.
From the anomalous wind components, he computed the anomalous
surface water displacements which, when placed over the normal
-4-
pattern of sea-surface temperature for the period in question,
enabled him to compute the anomalies of temperature due to
advection of the surface water.
The second objection is that of
using only the surface water advection term from the heat
budget equation for a local surface layer (see Chapter II).
Namias (1965) modified this procedure somewhat and computed the anomalous drift of surface water as before but
used an observed set of isotherms instead of the normal set
in order to compute the advection of temperature anomalies.
His correlation coefficients were improved somewhat over those
of the previous work.
However, the horizontal advection term
was still the only term evaluated, although he did mention the
possible effects of up-welling associated with the divergence
of the wind field and the transfer of latent and sensible heat
across the sea surface.
Arthur (1966) has extended Namias' work to include the
effects of the normal and anomalous wind drift currents on the
anomalous temperature gradients that occur within a specified
period of time.
His method thus takes account of all terms in
the horizontal temperature advection equation,
_
f f
SL~
a J~
-5-
where the capital letters denote the basic or normal values,
the primes denote the anomalies, and tl and t2 denote the
beginning and end of the period in question.
The first
and third terms on the right-hand side of the equation are
identified as the steps in Namias' method, while the second
term represents Arthur's extension.
Again, both the effects
of local changes in surface heat transfer and of vertical
velocities are neglected.
In fact, the assumption is made
that
dtt
Qr
(1.2)
d
in which the normal change in sea-surface temperature is
balanced by the effects of the normal surface heat transfer,
Q.
At present, evaluations of the extension are being carried
out at the Extended Forecast Division of the U. S. Weather
Bureau.
A much more rigorous approach was adopted by Berson (1962)
in which he attempted to determine from both theory and data
the two major effects on the heat balance in the surface layer
of the ocean, the effect on latent and sensible heat transfer
from the ocean, and the effect on the intensity of vertical
mixing and on horizontal transport within the upper mixed layer
from variable large-scale surface wind systems.
His results
-6-
(determined from data for a middle latitude zone of the North
Pacific Ocean during September 1959 through August 1960)
indicate that in the central longitudes and during autumn
and winter when anomalies of net radiation are small, a large
proportion of the anomalous evaporation and turbulent heat
conduction is generally tapped from the accumulation of heat
due to the effect of wind stress anomalies on advection.
He also developed a numerical method, including both advection
and surface heat transfer, for computing sea-surface temperature
anomaly changes when the wind stresses are given.
Berson concluded that feedback from ocean to atmosphere can
be treated numerically by subtracting the thermal energy equation
of the troposphere from the thermal energy equation of the active
layer of the ocean and obtaining a differential equation expressing the exchange of heat across their common boundary.
The main
problem in this method is in determining the changes in the depth
of the mixed layer or seasonal thermocline of the sea.
Since
very little data on this quantity exists, he is forced to use a
linear relationship between wind speed and layer depth determined
from various ocean stations in the North Pacific.
This relation-
ship holds only for the monthly changes during a particular year
and not for changes in the same month over many years.
During
the study presented below, data from ocean station PAPA (50*N145°W) was looked at and it was found that the linear relationship
-7-
varies from year to year and that no simple relationship holds
for changes from year to year for a particular month.
Roden (1962, 1963) has made a number of studies on the
power spectra of various meteorological and oceanographic
variables.
In his 1962 paper, he investigated records of sea-
surface temperature, cloudiness and wind speed for eight ocean
areas between Europe and South America.
He looked at the
frequency range between zero and six cycles per year and found
that most of the power of temperature anomalies is concentrated
at low frequencies and that there are no periodicities, that
a significant and inverse relation exists between temperature
and wind anomalies in the NE trade wind regions, and that along
the Brazilian coast there is a direct relation between temperature and north wind anomalies and an inverse relation between
temperature and east wind anomalies.
He found no relation
between temperature and cloud cover, implying that solar
radiation anomalies are not important in determining sea-surface temperature changes.
Probably the most ambitious program being conducted at
present is that of the Fleet Numerical Weather Facility (FNWF)
at Monterey, California.
Based on the theoretical work of
Laevastu (1960) and data collected on a world-wide basis from
naval vessels, analyses and forecasts are made of wave structure,
surface currents, thermal structure, surface heat transfer,
mixed layer depth, and sound channels.
The analyses are made
-8-
twice daily at 00Z and 12Z and 24-hour forecasts are made on
a daily basis.
The results are sent out via high-speed
computer data links to the Fleet Weather Centrals where they
are plotted on local area charts and transmitted in facsimile
or message form to fleet users.
The synoptic oceanographic analyses made at Monterey show
that pronounced changes in the surface layers of the ocean are
quite common.
The sea-surface temperature (SST) changes can
frequently be of the order of half the annual range and, in
some areas, can even exceed the total annual range.
The studies
indicate that the average period of the SST changes is shorter
(a few days) in higher latitudes, the areas of passing cyclones,
than in the lower latitudes where semi-permanent anticyclones
predominate.
The magnitudes of the changes are usually largest
near the stronger gradients of SST (current boundaries) and
The explanation
smallest in areas of small horizontal gradients.
of these synoptic changes can, in most cases, be found in the
atmospheric driving forces affecting both the dynamics of the
ocean surface layer and the surface heat transfer.
The advective changes by surface currents, computed according
to a method derived by Whitting in 1909 and described in Chapter
V., are quite apparent on the analyzed charts and account for a
major part of the SST anomalies in some areas.
Changes
caused by surface heat exchange and vertical velocities in the
mixed layer can be as large as 1.50 F in a 24-hour period, and
-9-
as large as 5*F at middle and high latitudes during the summer
season.
These driving forces change in the same way as the
weather conditions at the surface.
Thus, the changes in the
surface layers of the sea normally change at least as rapidly
as does the surface weather, and it is concluded that oceanographic analyses and forecasts should be carried out with the
same frequency as those of the surface weather.
In working with their data, the FNWF define three types
of SST anomalies:
(1) a deviation in a given region at a
specified time from the normal or long-period average for
the same region, (2) the difference between a temperature
actually observed and some reference temperature, and (3)
deviations of SST values at given places from smoothed latitudinal values for the same period of time.
Some conclusions
drawn regarding the first type of anomaly are that they are
relatively large in horizontal extent, that some of them
can be persistent over long periods while some appear and
disappear quite rapidly, and that some of the anomalies migrate
with the average surface current.
The computation of the third type of anomaly is done
numerically by scale and pattern separation.
The method
consists of repeated application of a smoothing operator which
reduces first-the amplitudes of the shortest wave lengths
and gradually affects longer and longer wave lengths.
Removal
-10-
of small-scale disturbances from the initial analysis leaves
the large-scale pattern; finally, removing the zonal portion
of the large-scale pattern gives the large-scale anomalies.
The large-scale anomalies change slowly from season to
season with fluctuations occurring as the surface weather
changes over the areas, while the small-scale anomalies can
be classified into two additional groups.
The first of these
corresponds to changes induced by local up-welling, heat exchange, and mixing.
The second group consists of those caused
by advection and meanders along current boundaries and eddies;
analysis of the advection anomalies at major current boundaries
shows that they are persistent to the extent that changes
occur according to the prevailing local winds.
In general,
warming occurs ahead of cold fronts (when southwesterly
winds are prevalent), and cooling occurs after frontal passages (when northerly winds prevail).
In addition to oceanographic analyses and forecasts, surface weather forecasts are also made on a daily basis which
take into account the exchange of heat between the ocean and
atmosphere.
Thus, we have an operational system that uses the
information obtained from detailed studies of air-sea interaction processes.
-11-
B.
Review of Ocean Surface Layer Heat Transfer Studies
An extensive review of this subject is given by Malkus
in Volume I of The Sea, Chapter 4.
The following will briefly
touch on the works mentioned in this review as a background
to the description of air-sea interaction studies that have
been conducted since the publication of The Sea in 1962.
The transfer of heat between ocean and atmosphere by turbulent conduction and the flux of water vapor may be determined
indirectly by two independent methods.
The first is that of
using energy and mass budget requirements, while the second
method depends upon the use of exchange formulas that have
been developed from the laws of small-scale molecular and turbulent transfer.
In the energy budget method of determining the heat flux,
the following equation is used to express the balance between
the amount of heat absorbed by the sea surface due to radiation,
the amount of this heat transported by ocean currents, and
the amount of heat supplied to the atmosphere:
R =O
+ Qe + S +
,
(1.3)
where R is the radiation surplus and equals the solar radiation
minus that which is reflected from the surface and that which is
reradiated back into space; Qs is the amount of sensible heat
-12-
that is transferred between ocean and atmosphere; Qe is the
amount of latent heat given up by the sea due to evaporation;
S is the amount of heat stored in the ocean; and Qv is the
amount of heat advected by ocean currents.
The total exchange,
Qs + Qe', between ocean and atmosphere may then be computed
from the relation,
Qs + Q
(1.4)
= R - S - Q
The main studies using this procedure have been conducted
by Sverdrup (1942), Jacobs (1951), Budyko (1956), London (1957)
and Houghton (1954), each of whom attempted to evaluate
Qs + Qe using various methods of computing R.
The critical
variable affecting R is the amount and type of cloud cover
in the area- under investigation.
In his work, Budyko showed
that the incoming radiation is approximately a linear relation
of the mean cloudiness, while the long-wave back radiation
decreases in a non-linear fashion with the mean cloudiness.
For annual averages, the storage S and the flux-divergence
v were considered negligible in comparison to the other
terms.
According to Malkus (1960),the most important demonstration
in marine meteorology of the past twenty years (Jacobs, 1951;
Montgomery, 1940; Bunker, 1960; Riehl et al, 1951) is that the
-13-
transfer of latent and sensible heat and momentum from sea
to air is governed largely by two parameters, the air-sea property difference of temperature or vapor pressure and the
prevailing wind speed averaged over some time period.
Thus,
while the average air-sea fluxes over long periods and large
regions are computable from planetary heat and mass budget
requirements, the fluctuations which build this picture and give
rise to the enormous departures from it are directly related
to transient atmosphere phenomena.
The basic premise underlying this development is that
turbulent exchange is the dominant mechanism affecting the
vertical distribution of a property from the air-sea interface
to a distance of several tens of meters above it, so that
the fluxes obey the equation,
F
= -K
p
where F
dp
p dz'
(1.5)
is the flux of the property p, dp/dz is the vertical
gradient of the property, and K
p
is the eddy transfer coeffic-
ient which is many orders of magnitude larger than the corresponding molecular transfer coefficient.
Thus, using the above
equation and making several assumptions about the nature of
the turbulence, equations can be derived relating the flux of
momentum, water vapor and sensible heat to a wind speed at
-14-
some height and the difference between the property at that
height and at the sea surface (see Chapter II).
The main works using this method that are reviewed by
Malkus are those of Budyko (1955, 1956), Drozdov (1953), Jacobs
(1951, 1951a), Sverdrup (1957), London (1957) and Houghton
(1954).
The annual mean distribution of Qe and Qs and the
other heat-balance components of the sea surface are presented.
The resulting annual heat budget of the ocean forms a foundation for discussing the global heat and water budgets, the
climatological picture of air-sea heat exchange, and the
average seasonal variation in several regions of this exchange.
She also reviews a study done by Colon (1960) in which he used
both the energy budget method and the transfer equation method
to investigate the monthly and seasonal distribution of the
various components of the surface layer heat balance equation
for the Caribbean Sea.
As a result of the observations of solar radiation made
during the International Geophysical Year, Budyko (1963) has
completed a new Atlas o
the heat balance of the earth.
The
Atlas contains 69 world charts including the annual means of
solar radiation reaching the earth's surface, the radiation
balance of the earth's surface, and the heat exchange between
ocean and atmosphere due to evaporation and turbulent heat
conduction.
It also contains charts of the mean annual con-
ditions of the redistribution of the heat in the oceans due
to ocean currents, of the radiation balance of the surface
atmosphere, of the heat of condensation, and of the redistribution of heat in the atmosphere due to atmospheric motion.
The charts are considered to be more accurate and are more
detailed than those presented in the Atlas prepared in 1956.
Some of the results obtained from this study are that the
meridional transfer of heat in the world oceans is about
60 per cent of the heat transfer of the atmosphere, that the
components of heat balance of the Atlantic, Pacific, and
Indian Oceans differ slightly from each other, and that for
each ocean the main part of the heat of the radiation balance is expended on evaporation.
It was also determined that
the earth absorbs 168 kcal/cm 2 annually.
Two-thirds of this
amount, 112 kcal/cm2/yr, is absorbed at the earth's surface,
while the rest is absorbed by the atmosphere.
The earth's
surface loses 40 kcal/cm2 per year by effective long-wave radiation, and, as a result, its average radiation balance is
equal to 72 kcal/cm2 /yr. Of this amount, 59
/yr
,cajcm
is expended on evaporation, and 13 kcal/cm2/yr is expended on
the turbulent loss of heat to the atmosphere.
Wyrtki (1965, 1966) has used climatic data obtained from
the U. S. Bureau of Commercial Fisheries to calclate the heat
exchange at the surface of the Pacific Ocean north of 200 S
-16-
for the period 1947 to 1960. In the 1965 study, the average
annual components of the surface heat transfer are given
along with the implications of their distribution
regard to ocean circulation.
with
It was found that the North
Pacific gains heat at a rate of 3 x 10
14
cal/sec.
The heat
gain occurs mainly along the eastern side of the ocean and south
of 250 N, while the main heat loss occurs in the region of the
Kuroshio Current.
From a simple theoretical model, Wyrtki found
that the average temperature distribution in the range of the
subtropical anticyclone is maintained in the presence of these
heat sources and sinks by a horizontal circulation of the order
of 10 million m /sec in a shallow surface layer.
The 1966 study is an atlas of the monthly variation of
heat exchange and surface temperature north of 200 S in the North
Pacific Ocean.
He used the same data as that of the 1965
study and averaged them over two-degree latitude-longitude
squares and by months for the 1947 to 1960 period.
For each
ten degree square, the seasonal variation of the heat transfer
components, the total heat exchange at the surface, and the
sea-surface temperature were presented.
The U. S. Bureau of Commercial Fisheries at La Jolla,
California, began to publish monthly charts of meteorological
variables and heat transfer at the air-sea interface in 1965
(Johnson, Flittner and Cline; 1965).
The data are collected
-17-
from synoptic marine radio weather reports from ships at sea
and analyzed by electronic computer.
The analyzed results
are prepared in chart form covering the entire North Pacific
Ocean and distributed to participating and interested agencies.
In 1965 Garstang reported the results of a project designed
to study the role of diurnal variations of sensible and latent
heat over the tropical ocean.
The transfer equations used
incorporate a drag coefficient that is linearly dependent upon
wind speed and atmospheric stability.
It was found that pro-
nounced diurnal variations in sens.ible heat transfer occur
along with semi-diurnal oscillations in cloudiness and precipitation.
Within synoptic scale disturbances the transfer of
either sensible or latent heat may increase by an order of
magnitude.
Integrated over the entire disturbance, the energy
flux is found to double the undisturbed values.
Since the role
of the energy input to the systems is fundamental to understanding the structure and behavior of both synoptic and mesoscale systems, he concludes that emphasis should be placed on
including these variable quantities in prediction schemes.
Due to the fact that the distribution of energy transfer
is strongly dependent upon the frequency of synoptic scale
systems, mean maps based on climatological data bear little
relation to the actual synoptic maps of energy flux.
The
transfers also differ in magnitude in the tropical regions
-18-
from the values given by previous investigations (e.g. Budyko,
1956) and indicate that a revision may need to be made
on the current estimates of the heat balance of the world
oceans.
Kraus and Morrison (1966) have conducted a statistical
analysis of wind, air, dew-point, and sea-surface temperature
records from all nine weather ships in the North Atlantic Ocean.
The study showed that local variations between years are highly
significant when compared to variations within months.
The
fluctuations show a consistent pattern of more that 500 miles
in the atmosphere and a persistence over several months.
The
horizontal extent of sea-surface temperature anomalies appears
to be small; however, they tend to be more persistent than
the air temperature anomalies.
It was also found that short-
period variations in the flux of latent and sensible heat are
due predominantly to atmospheric variations.
This effect is
greatest in the winter; during the summer the effect of seasurface temperature anomalies is somewhat greater.
The values
obtained for the heat fluxes were higher than those given by
Jacobs (1951) and may be due to the different record periods
used or the absence of a correction term in Jacobs' work due
to the covariance between the meteorological variables of wind
speed and air-sea temperature difference and between wind
speed and air-sea vapor pressure difference on a daily basis.
-19-
C.
Purpose of Thesis
The purpose of this work is to determine what relation-
ships, if any, exist between the atmospheric parameters of
wind speed, water vapor content, cloud cover and radiation and
the monthly and seasonal fluctuations of sea-surface temperature and of heat transfer between the ocean and atmosphere.
The data used in this study are much more extensive (7 years,
1951-1957, of data covering the North Pacific Ocean from 200 N
to 550 N) than any that have been analyzed previously, and it
is felt that much information has been obtained concerning
the interaction of the two media.
It is obvious from the review of previous studies that
it is necessary to include the effects of both surface
advection and local heat transfer in order to account for
fluctuations in time and space of the sea-surface temperature.
An attempt is made in this report to relate fluctuations in
the transfer of heat between ocean and atmosphere and fluctuations in the surface advection of heat due to wind-drift to
fluctuations in the sea-surface temperature.
Chapter II of the report outlines the theoretical considerations behind the transfer formulas used in the study.
Chapter III describes the type of data available and the procedures used to put the data into a workable form.
Chapter
IV discusses the results of the heat transfer computations
and also the relations between the yearly cycle of the total
-20-
heat transfer across the sea surface and the yearly cycles of
the sea-surface and air temperatures.
Chapter V describes
the spatial and temporal scales of anomaly patterns (an
anomaly is the monthly departure of a value from some calculated mean or normal value) of the heat transfer terms
and of the sea-surface and air temperatures; it also discusses
the results of correlation studies made between the time
series of the heat transfer term anomalies and those of the
sea-surface temperature anomalies.
Chapter VI describes the
results of further correlation studies made between other
atmospheric and oceanic paramteres.
Finally, Chapter VII
summarizes the results of the previous chapters and discusses
areas of possible future studies of this and other types.
-21-
II. Theory
A.
Development of Surface Layer Heat Balance Equation
For a column of water of unit cross-sectional area and of
depth -D (z positive upwards), the heat balance equation may
be written as
v
o
-D
o
0
t oo
00
o
o -b
where Ts is the temperature of a unit volume of water, Qr is
the radiation surplus, Qa is the sum of latent and sensible
heat transfer between ocean and atmosphere, Q
is the horizontal
advection of heat into the column, and QD is the vertical flux
of heat into the column at the depth -D.
This equation expresses
the relationship that exists over the time interval t between
the storage of heat in the column of water and the total exchange
of heat energy between the column and its environment.
B.
Development of Heat Transfer Formulas
1.
The radiation balance Qr may be broken up into three
terms, Qi, QR and Qb where
Qr
=
Qi - QR - Qb
and
(II.2)
Qi is the incoming solar radiation reaching the earth's surface,
and Qb is the effective back radiation equal to the difference
-22-
between the long wave radiation from the sea surface and the
long wave :radiation from the atmosphere.
The incoming radiation Qi (cal/cm2 / day)is determined
from the following equation proposed by Berliand (1960),
Q
=
Qio (1- ac
2
- bc)
,
(II.3)
where Qio is the incoming solar radiation with a clear sky,
c is the fractional cloud cover, b is a constant equal to .38,
and a is a function of latitude varying from .37 at 200 to .41
at 55° .
The values of Qio are taken from a table given by
Berliand (1960) as a function of latitude and month.
The amount of radiation reflected from the sea surface
QR is determined from
QR
=
i
r ,
(II.4)
where r is the percentage of radiation reflected and given in
a table by Budyko (1965) as a function of latitude and month.
Cox and Munk (1955) have calculated r taking into account the
effect of wind speed on the nature of the reflecting sea surface.
However, between 20'N and 550N their values do not
differ significantly from those of Budyko.
2.
The effective back radiation Qb (cal/cm2/day) is
determined from the semi-empirical equation proposed by Berliand
-23-
and Berliand (1952),
A(3T)(
=
k
0>
f-5Q~
(
Ga))
(11.5)
where S = .97 is the ratio of the radiation of the sea to that
of a
black body;
-=
1.75 x 10
Sea'
sea
-7
is the Stefan-Boltzmann constant;
eDair are the absolute temperature5(')
;
e is vapor pressure of the air in mb;
c is the fractional cloud cover; and
k is a function of the latitude.
The first term, which takes into account the effect of
sea-surface temperature, humidity, and cloudiness, varies
between 20 and 200 cal/cm2/day.
The constant k has been
evaluated considering the vertical extent of clouds as well
as the height of the cloud base from the earth's surface;
it varies from .51 at the equator to .75 at 55*N due to the
decrease in cloud base height in the polar regions.
The
second term represents the effect of stability on the back
radiation and varies between -20 and 20 cal/cm2/day.
3.
The exchange of latent and sensible heat between ocean
and atmosphere may be determined from the theory of turbulent
transfer and some assumptions about the nature of the wind
field just above the sea surface.
-24-
If turbulent exchange is the dominant mechanism affecting
the vertical distribution of a property near the surface,
then the vertical flux of the property p obeys the equation
dp
(11.6)
Fp = -Kp dz11.6)
where Kp is the eddy transfer coefficient.
For neat and water
vapor, Fp is constant in the vertical in a thin boundary layer
in which heat and water vapor are not accumulating.
In the
lowest few tens of meters over the sea surface, the air is
well-stirred, shear-turbulence dominated, neutrally stable
and barotropic.
Under these conditions, the following
exchange formulas may be used to determine the flux of momentum
, sensible heat Qs and latent heat Qe
(11.7)
Km du
dz
dT
Qs = -cpKs --dz
Q = LE = -LK
e
,
(11.8)
and
(11.9)
e dz
where Km, Ks and Ke are the eddy exchange coefficients of
momentum,
speed,
sensible heat and water vapor, u is
the wind
T is the temperature of the air, and q is the specific
-25-
humidity of the air.
When turbulent shear flow is the dominant process, the
assumption that Km = Ks = Ke = K is made, and
K
(II.7a)
/ du
dz
=
may be substituted into (11.8) and (11.9).
In addition, the
relation
(II.10)
To= f
may be obtained from either observations or from the work on
turbulence done by Rossby and Montgomery (1935).
In this
equation, the drag coefficient CD is a function of surface
roughness, anemometer height, and von Karman's constant.
If the vertical derivatives of (11.7) -- (11.9) are
expressed in terms of finite differences between the heights
z
a
and o,
u
= u
T
=
9 =
T
a
a
-
- T
-
= u
o
s
a
,
, and
qs'
then the transfer equations may be written as
(II.11)
-26-
It
=
CDua
(11.12)
,2
CD (Ts - Ta) ua , and
Q = p
LC
Qe = LE =
q )
(qs -
(II.13)
(11.14)
.
u
The drag coefficient CD depends upon the wind speed close
to the sea surface and is, therefore, not a constant.
Deacon
and Webb (1962) have derived a linear relationship between the
drag coefficient and wind speed measured at a height of 10
meters above the sea surface from the observational results
of various investigations.
This relationship,
) x 10 -
= (1.00 + 0.07
C
3
(II.15)
is applicable for near neutral stability conditions and shows
a comparatively slow increase of drag coefficient with wind
speed.
Since winds are measured at heights of around 10 meters,
the drag coefficient is dependent upon the effect of atmoFor a wL
spheric stability.
at
meter above thl
,urface,
the 10 meter speed will be greater under stable than under
unstable conditions.
Therefore, the drag coefficient should
be smaller for stable conditions thah. that determined from
I,.r1
r
VI.L
l.J
d
l~
a.LLLJ_
unst1
ir
f .
A
.--
e
conditinns.
-
---
In
his
study
is
on heat transfer between ocean and atmosphere over a tropical
ocean, Garstang (1965) used some theory of Monin and Obukov
-27-
(1954) to obtain a relationship between the drag coefficient
at 6 meters C6 , the wind speed u6 , and the bulk Richardson
number RBg
C6 = (1.46 + 0.07,&6 - 4.2 RB) x 10- 3
(11.16)
In the present study, however, (11.15) is used to determine
the drag coefficient since the measurements necessary to use
the Monin-Obukov procedure were not available.
C.
Error Analysis
The accuracy of the exchange formulas in determining the
transfer of heat depends upon the validity of the assumptions
underlying the theory, the accuracy of the observational data,
and the manner in which the formulas are used to calculate the
heat transfer.
If we accept the theory as valid, then the effect of
observational errors is shown in Table I.
(Part of the table
is taken from Roden (1959) as he used some of the same exchange
formulas that were used in this study.)
Since the exchange formulas are non-linear in nature, using
monthly mean data instead of daily or hourly values in determining the heat transfer may lead to errors.
In order to
determine the magnitude of these errors, daily values of
-28-
Table I
4 Qi/Qi
Cloud
Cover
C
Error
AC
0.0
+ 0.1
+4
+ 0
1.0
+ 0.1
+1
+ 10
0.2
+ 0.1
+6
+1
2.0
+ 0.1
+1
+ 5
0.4
+ 0.1
+9
+3
3.0
+ 0.1
+1
+3
0.6
+ 0.1
+ 13
+ 5
4.0
+0.1
+1
+ 3
0.8
+ 0.1
+ 22
+ 8
5.0
+ 0.1
+1
+ 2
1.0
+ 0.1
+ 48
+ 17
6.0
+ 0.1
+1
+ 2
Wind
Velocity Error
m/sec
A
Ts
OC
%
6 Qe
Ta
-
Error
(T -T
b/Q
S
A QsQs
,,
Humidity
Error
a
)
SQs/Qs
%
//
' Qe/Qe
4%
S
+1
+ 56
+ 56
+1
+1
+ 2
+1
+ 30
+ 30
+1
+1
+3
+1
+ 22
+ 22
+1
+1
+ 3
+1
+ 17
+ 17
+1
+1
+4
+1
+ 14
+ 14
+1
+1
+ 5
+1
+ 12
+ 12
+1
+1
+7
+1
+ 11
+ 11
+1
+1
+ 10
+1
+9
+9
+1
+1
+ 20
-29-
sea surface temperature, air temperature, dew-point temperature,
cloud cover and wind speed were obtained for weather ship
stations NOVEMBER (30N - 140W) and PAPA (50N - 145W) for the
period of January to December of 1962 (see Chapter III).
Monthly average values of Qi' Qb' Qs, Qe and the sum of these
four heat transfer terms QBAL were computed first from the daily
values and averaged over the number of days in the month; in
the second case, the variables were averaged over the number
of days in the month and the heat transfer terms computed from
these averaged values.
The results are shown in Tables II
and III. Q(v) denotes heat transfer averaged over the daily
values, and Q(v) denotes heat transfers computed from the
monthly averaged variables.
For Station NOVEMBER, the yearly average per cent differences between the two methods of computing the monthly average
heat transfers are 1.4 for Qi, 1.7 for Qb' 4.8 for Qe, 7.8
for Qs and 17.7 for QBAL.
For Station PAPA, the per cent
differences are 1.4 for Qi, 1.9 for Q6 9.5 for Qe, 45.1 for
Qs and 10.4 for QBAL'
The accuracy of the exchange formulas increases as the
value of the heat transfer increases since the errors resulting
from both observational and averaging effects are inversely
proportional to the heat transfer.
The teoretica
accuracy
of the formulas is also higher when large heat transfers occur.
Table II
Ship Station NOVEMBER (30N - 140W)
North
January
% dif
February
% dif
March
% dif
April
% dif
May
% dif
June
% dif
July
% dif
August
% dif
September
% dif
October
% dif
November
% dif
December
% dif
Qi(v)
Qi(V)
Qb(V)
217
221
128
1.4
307
1.8
355
1.3
432
313
359
440
1.7
466
1.1
456
1.3
465
1.5
499
1.7
437
1.6
346
1.2
287
.8
205
1.7
131
1.9
141
1.9
145
1.4
134
144
147
137
2.5
471
462
472
507
444
350
289
209
141
1.1
125
1.7
129
2.0
129
2.4
132
2.3
136
1.3
155
.7
136
1.4
143
128
131
132
135
137
156
138
Q (v)
Qe (V)
266
3.4
313
3.0
217
4.8
194
5.0
247
3.3
198
3.3
258
2.7
192
4.2
210
11.2
306
1.7
263
4.9
319
10.0
257
Qs(v )
31
14.8
303
207
184
40
1.6
30
12.4
17
10.6
238
192
251
184
187
311
250
350
38
7.9
16
16.4
28
.4
13
.6
10
.0
26
2.4
31
5.9
29
20.1
Qs(v )
BAL(v)
-208
6.8
-186
6.9
-38
43.9
87
18.0
40
36.4
117
10.9
52
22.4
165
8.1
84
32.9
-121
3.0
-162
9.8
-278
12.9
%AL(v)
-193
-173
-21
103
55
130
63
178
111
-125
-147
-314
Table
III
Ship Station PAPA (50N - 145W)
North
January
% diif
February
% dcif
March
% dif
April
% dcif
May
% di f
June
% dif
July
% di f
August
% dif
September
% di f
October
% dif
November
% di f
December
% dif
Qi(v)
59
1.8
121
2.1
201
1.7
334
1.4
357
1.2
355
.7
346
.7
286
1.1
236
1.4
153
1.4
79
1.3
50
1.8
Qi(V)
60
(V)
105
b(v)
108
2.1
124
204
338
362
357
348
289
240
155
80
51
121
3.4
118
2.3
136
2.2
112
1.4
100
.8
94
1.3
95
1.7
119
1.9
124
1.5
132
1.6
121
2.4
125
120
139
113
101
95
97
121
126
134
124
Qe (v)
40
55.1
56
18.4
104
2.7
114
2.7
77
3.0
57
3.9
28
1.9
52
5.5
159
1.9
191
7.6
355
7.1
116
4.6
Qe(v)
62
45
101
117
75
55
28
55
162
176
330
110
Q(v)
-29
44.8
10
8.6
30
4.8
44
17.3
8
11.5
5
.6
-7
41.4
-1
382.2
22
9.0
23
5.1
110
9.6
30
5.8
Q (v)
-16
10
29
36
9
5
-4
2
20
22
99
32
QBAL (v)
-57
63.5
-65
12.1
-52
9.5
40
15.6
160
2.5
193
1.9
231
.6
140
3.0
-63
.1
-185
8.6
-517
6.7
-217
.7
%AL (v)
-93
-57
-47
46
164
197
230
135
-63
-169
-482
-216
-32-
Since large transfers occur at times of strong winds, the wind
shear keeps the atmospheric stratification close to neutral
values, and these are the conditions for which the formulas
were derived.
It is also seen from Tables II and III that the errors
resu
-ing from averaging procedures are smaller for Qi and
This result is due to the fact that
Qb than for Qe and Qs.
there is a correlation on a daily basis between the wind speed
u
a
and the air-sea temperature difference Ts - Ta and between
the wind speed ua and the air-sea specific humidity difference
qs-
qa
by
If the monthly average heat transfers are given
Qe
(qss - qa) u a + CV (qs - qa) Ua
(11.17)
and
Qs
C (T
-
T
)
u
+
CV (T
- T
)
U
(11.18)
where the bar denotes an average over daily values and CV
denotes "covariance betwe en" the second terms in (11.17)
and (11.18) represent the contributions of the correlations
between the variables and the first terms the contributions
of the monthly averaged variables.
If only monthly averaged
variables are used, the second terms are neglected, and small
errors are produced.
Since monthly averaged variables were the only ones avail-
-33-
able over a long time period (see Chapter III), they were used
in this study to compute the heat transfer values.
However,
the investigation of the data from the two weather ships
indicates that the errors involved in this procedure are, in
the average, only 10% of the actual values.
The work of Malkus
(1962) and Kraus and Morrison (1966) also support this conclusion.
-34-
III. Data Preparation
A.
Monthly Averaged Data
Monthly averaged values of sea surface temperature, air
temperature, wet-bulb temperature, cloud cover and wind speed
were obtained for an 84 month period, January 1951 to December
1957, from Dr. O. E. Sette of the Bureau of Commercial Fisheries
at Stanford University.
The data are in the form of averages
for two-degree latitude and longitude squares and cover the
North Pacific Ocean from 200 S to 600 N.
However, in this study
only those values north of 200 N were used since the coverage
is excellent in this region; usually more than 500 observations
per two-degree square were available for the 84 month period,
and in some squares more than 5000 observations were given.
The data were first copied onto Marsden Square sheets
(10-degree latitude-longitude squares with subdivisions for
each 2-degree square within them) and those values obviously
in error were eliminated.
Overlapping averages of 9 values in
each quadrant of the 10-degree square were taken, and these
values were plotted in their respective positions on a map of
the North Pacific.
The 4 values in each of the 10-degree
Marsden Squares were then averaged to obtain a value at each
5-degree latitude-longitude intersection.
Finally, the 162
resulting values were punched onto cards for further use on
a 7094 IBM computer.
-35-
For the computation of geostrophic winds and heat transfer
terms, monthly average sea-level pressure values were taken from
charts supplied by the Extended Forecast Division of the U. S.
Weather Bureau and were also punched onto cards.
The 162-point network of points, extending from 20*N to
550 N and from coast to coast, gives a picture of the largescale features of each of the variables used in the computations
and each of the computed heat transfer terms.
It is felt that
this network is adequate for one to be able to learn something
about the large-scale interaction of ocean and atmosphere
on a monthly and seasonal time scale.
In his work using the same data, Wyrtki (1965, 1966) subtracted 1.2*F from all of the sea-surface temperatures.
This
correction was suggested by Saur (1963) to convert injection
temperatures to actual ocean temperatures.
This correction was
not used in this study, however, since the reported air and wetbulb temperatures are probably also biased on the high side due
to radiational effects of the ships on which the measurements
were made.
Since no information on this effect was available
and since the temperature values always appear as differences
in the exchange formulas, it was decided to use the sea surface
temperatures as reported.
The values of the specific humidity at the sea surface
were computed from the saturation vapor pressures over water
-36-
at the reported sea surface temperatures with the following
relationships:
_ e
qs =
p
(III.1)
s ,
and
e
s
(111.2)
= .98e ,
o
E = .622, es is the
where qs is the specific humidity,
saturation vapor pressure over the sea, e
is the saturation.
vapor pressure over fresh water taken from the Smithsonian
Meteorological Tables (1958), and p is the sea-level pressure
in millibars.
The values of the specific humidity of the air were
obtained from the following relationships:
(T
- Tw ) (Cp + wCp)
e
- w) L,
= (w'
a
= i
w+e
(111.3)
(111.4)
and
e
a
p
where Ta is the temperature of the air approaching the wetbulb, T
is the temperature of the saturated air leaving the
wet-bulb, CP is the specific heat at constant pressure of the
-37-
dry air, C
is the specific heat of water vapor, w is the mixing
ratio of the approaching unsaturated air, w' is the mixing
ratio of the saturated air, L is the latent heat of vaporization,
and qa is the specific humidity of the unsaturated air.
B.
Daily Averaged Data
Daily averaged values of sea surface temperature, air
temperature, dew-point temperature, cloud cover, and wind speed
were obtained for the 12 months of 1962 for weather ship stations
PAPA (500 N - 1450 W) and NOVEMBER (300 N - 140 0 W) from Dr. Glenn
Flittner of the Bureau of Commercial Fisheries at La Jolla,
California.
The data were first tabulated by month and then
punched onto cards for further analysis and use.
-38-
IV. Heat Exchange Calculations
A.
Results Using Monthly Averaged Data
Values of Qi', Qb' Qe
Qs and the total heat transfer across
the sea surface
Q = Qi
=
(IV.l)
Qb - Qe - Qs
were computed for each of the 84 months of the period under
investigation, using the formulas developed in Chapter II.
The results were averaged over the 7 years for each quantity
and each month and are shown along with their respective
standard deviations in Figures A12 through A71.
They were
also averaged over all of the 84 months in the period, and
these results are shown in Figures Al through All.
1.
The values of Qi (incoming radiation corrected for
cloud cover) averaged over the 84-month period are shown in
Figure Al.
0
They range from 124 cal/cm2/day at 55 N to
432 cal/cm2/day at 200 N and show a strong dependence upon latitude; variations along a latitude circle represent variations
in the average cloud cover.
Maximum incoming radiation occurs
between 200 N and 300 N from the Hawaii Islands to the Philippines due to the comparatively low cloud cover in this area
as shown in Figure A8.
The values of the standard deviation
-39-
at each of the 162 points reflect the fluctuations in Qi due
to the yearly cycle of incoming radiation and to the fluctuation
in the cloud cover from year to year during the 7-year period.
The values of Qb (effective back radiation) averaged over
the 84 months are shown in Figure A2.
They range from 87
cal/cm2/day at 50*N to 143 cal/cm2/day at 350 N and show
little seasonal change.
Effective back radiation is high in
the region of the Kuroshio Current due to the large sea-air
temperature difference and low in the high latitudes due to
the large amount of cloud cover in this region.
The standard
deviations of Qb are all comparatively low and indicate that
there is little seasonal or year-to-year change in the back
radiation.
The 84-month averages and standard deviations of Qe
(latent heat transfer) are presented in Figure A3.
The average
values range from 84 cal/cm2/day at 55*N to 384 cal/cm2/day
at 35*N.
The amount of heat lost from the sea surface due to
evaporation is lowest in the high latitudes due to the low
vapor pressure difference in this region and highest in the
region of the Kuroshio due to the large vapor pressure differences there.
The relatively large values of the standard
deviations reflect the large seasonal fluctuations of Qe'
especially in the area of the Kuroshio Current.
The average
values of Qe show the same distribution over the map as those
-40-
However, the magnitudes are larger
computed by Wyrtki (1965).
than his values due to the fact that Wyrtki subtracted 1.20 F
from the sea-surface temperature values, giving him lower
values of the sea-surface vapor pressure.
Part of the differ-
ences may also be attributed to the use of a drag coefficient
in this study that is a function of the wind speed and not a
constant value as was used by Wyrtki.
The values of the 84-month averages and standard deviations
for
Qs
(sensible heat transfer) are shown in Figure A4.
The
average values range from 6 cal/cm2/day at 40*N - 125 0 W to 106
cal/cm2/day at 350 N - 145 0 E.
The largest values occur in the
region of the Kuroshio, where the air-sea temperature difference is large, while the smallest values occur along 20'N
and the California coast, where the sea-air temperature
difference is small or reversed.
The large values of the
standard deviations reflect the seasonal variation of the
sensible heat transfer, especially off of the Asian coast,
where large masses of cold air move out from the continent
during the winter months.
The 84-month averages and standard deviations of the total
heat transfer across the sea surface SLHT or Q are shown in
Figure A5.
The average values range from -328 cal/cm2/day
at 350 N - 145 0 E to 117 cal/cm2/day at 30*N - 115 0 W.
The
preponderance of negative values indicates that during the
year the North Pacific Ocean loses more heat than it gains
-41-
at the surface above 200 N.
The largest amount of heat is lost
in the region of the Kuroshio, where back radiation, evaporation and sensible heat loss are high.
There is a small heat
gain along 200 N between 145 0 E and 165 0E due to high incoming
radiation and another heat gain along the California and
Mexican coasts which is caused by small values of evaporation
and sensible heat loss.
In comparing these values to those
given by Wyrtki (1965), it is seen that his map shows much less
heat loss above 200 N.
This result is due to the larger values
of sensible and latent loss used in this study, giving larger
values of total heat lost at the surface.
The extremely large
values of the standard deviations in Q indicate that there are
large seasonal and year-to-year variations in the total heat
transfer.
The values of the total heat transfer were integrated over
the North Pacific north of 200 N.
ocean loses
200 N.
7 x 10
14
The result indicates that the
cal/sec through the surface north of
Wyrtki (1965) calculated the total integral of heat
transfer over the entire North Pacific from the equator poleward and found that the ocean gains 3 x 1014 cal/sec, or
about 10 per cent of the incoming radiation.
The difference
between the two calculations shows that there is a heat gain
0
of 10 x 1014 cal/sec between the equator and 20 N and that it
is this region that accounts for the yearly heat gain of the
North Pacific.
The magnitude of this heat gain between the
-42-
equator and 200 N is uncertain, however, due to the different
values of latent and sensible heat used by Wyrtki and this
study.
Since the values in the latter were larger, resulting
in a larger heat loss above 200 N than Wyrtki's work would
show, the magnitude of the heat gain should be smaller than
calculated.
The 84-month average values and standard deviations of
sea-surface temperature, air temperature, cloud cover, observed
wind speed, specific humidity and Bowen Ratio are shown in
Figures A6 through All.
They are given here as reference
maps and will not be discussed.
2.
The values of Qi, Qb, Qe, Qs, and Q averaged over 7
years for each month are shown in Figures A12 through A71.
The values of the standard deviations over the 7-year period
are shown in order to give some idea of the variability of
each of the heat transfer terms on a year-to-year basis.
In order to facilitate the analysis of these maps, the
first 2 harmonics (first harmonic has a period equal to one
year or 12 months) of the Fourier Series for each transfer
term were computed at each of the 162 points on the map.
Along with the 2 Fourier coefficients, the percentage of the
series variance accounted for by each of the harmonics was
also computed.
This type of analysis was found to be extremely
useful in interpreting the.monthly and seasonal cycles of the
-43-
data, since the 2 harmonics together usually accounted for over
95 per cent of the variance at each point; in fact, over most
areas of the map the first harmonic or yearly cycle accounted
for over 90 per cent of the variance.
By determining the phase
of each of the harmonics and printing them in map form, it was
also possible to compare the relative times of maximum and
minimum values of each series at each of the points.
The values of Qi averaged over the 7-year period for each
month are presented in Figures A12 through A23.
The largest
values of Qi at each latitude occur in June, while the smallest
values occur in December due to the motion of the earth around
the sun and the tilt of the earth's axis of rotation with
respect to the plane of revolution.
The Fourier analysis shows
that there is a regular yearly cycle as expected.
The first
harmonic accounts for over 90 per cent of the variance at most
points, with the exception of the area from 200 N to 30*N and from
to 1400 W, where the percentage is reduced to 60 due to
110ow
the effect of cloud cover variation without a uniform yearly
cycle.
Figures A24 through A35 show the 7-year average values and
standard deviations of the effective back radiation Qb.
Due
to the low values of cloud cover and the large values of the
sea-air temperature differences, the largest values of
Qb
occur in January.
The Fourier analysis again shows a regular
yearly cycle with the maximum values occurring in December
and January and the minimum values in June and July.
Relativ-
ely low values of the standard deviations indicate that there
is little year-to-year variability of Qb*
The low values of
the amplitude of the first harmonic in the Fourier analysis
also show that there is only a small range between the maximum and minimum values of Qb throughout the year.
The 7-year average values and standard deviations of Qe
for each month are given in Figures A36 through A47.
The
results of the Fourier analysis show that there is a pronounced
yearly cycle in Qe and that the maximum values occur in
December, while the minimum values occur in June.
The first
harmonic accounts for more than 90 per cent of the series
variance in all areas except from 200 N to 250 N and 110 0 W
to 170 0 W, where the percentage falls to as low as 5.
This
result is due to the fact that both the specific humidity difference and the observed wind speed fail to have regular yearly
cycles in this area.
The amplitude of the first harmonic and
the 7-year standard deviations are largest in the area from
300 N to 450 N and 180 0 E to 125*E, reflecting the effect of the
large masses of cold, dry air that move off of the Asian
continent during the winter months.
Figures A48 through A59 show the 7-year average values
and standard deviations of Qs for each month of the year.
Again, the Fourier analysis shows that a regular cycle occurs
with the maximum values in November and December and the
-45-
minimum values in May and June.
This yearly cycle is due to
the fact that both the sea-air temperature difference and the
observed wind speed also have regular yearly cycles with maxima and minima in the same months.
Negative values of Qs
(heat gained by sea surface) occur in the high latitudes from
May through September, reflecting the fact that the air temperatures are warmer than the sea temperatures during these
months.
The relatively large values of the standard deviation,
especially in the summer months, indicate the large year-toyear variability that occurs in Qs!
Figures A60 through A71 show the 7-year averages and
standard deviations of the total heat transfer through the
surface Q.
It is seen that negative values of Q, implying
heat loss from the ocean surface, occur from September through
April, while positive values, implying heat gain by the surface layer, occur from April through September.
The results
of the Fourier analysis show that there is a regular yearly
cycle in Q with the maximum values occurring in June and the
minimum values occurring in December.
The first harmonic of
the series accounts for over 95 per cent of the variance in
all areas except those in which there are no regular cycles
of Qi'
Qb) Qe'
and Qs"
In looking at the maps for the different months, it is
seen that heating of the surface layer begins first in March
-46-
along the 200 N latitude from 130 0 E to 165 0 E and begins last
in the region of the Kuroshio Current in the latter part of
April.
The heating continues in all areas from May through
August except in the region between Hawaii and Baja California,
where some cooling occurs due to the high evaporation in all
months.
Cooling of the surface layer first begins in August
at 200 N and 125*E to 135 0 E, then in the region of the Kuroshio
during September, and finally spreads to all regions by October.
The cooling continues from October through March, with
the largest values occurring in December and January off of
the Asian coast, where over 700 cal/cm2/day are lost due to
the high latent and sensible heat transfer from ocean to atmosphere.
The large values of the standard deviations reflect
the large year-to-year variability in Q due to the corresponding
large variability in
3.
Qe and Qs"
The values of the sea-surface temperature SST, the
surface air temperature SAT, and the change in sea-surface
temperature from month to month4 SST averaged over the 7-year
period for each month are shown in Figures A72 through A107.
Figures A72 through A83 show the average values and the
7-year standard deviations for SST.
Fourier analysis of the
data at each of the 162 grid-points shows that there is a
regular yearly cycle in all areas, with the first harmonic
accounting for over 95 per cent of the variance at all points.
The maximum amplitudes of the yearly cycle occur in the area
of the Japan Sea and the Kuroshio Current, where the yearly
range of the sea-surface temperature is 18 to 240 F.
The
minimum values of the amplitudes occur in the area from 120 0 W
to 170 °W and 200 N to 300 N, where the yearly range is only
4 to 70 F.
The Fourier analysis also show that the maximum values of
sea-surface temperature occur in the latter part of August and
the early part of September and that the minimum values occur
in the latter part of February and the early part of March.
In comparing the yearly cycles of the total heat transfer
across the sea surface Q and the sea-surface temperature SST,
it is seen that the two are strongly related.
In all areas
of the maps except the southeast corner, the sea surface reaches
a maximum temperature at the end of the heating cycle (positive
Q) and a minimum temperature at the end of the cooling cycle
(negative Q).
Figures 1 through 4 show the cycles of Q and SST at various
points along the latitudes 200 N, 35N, 400 N and 500 N.
(For
a more extensive presentation of this type of analysis, see
Wyrtki, 1966.)
Large changes in Q throughout the year produce
corresponding large changes in SST, whereas small changes in
Q produce small changes in SST.
This relationship between
200 N - 1550 w
20N - 1450E
300'
2
20
N
- 125
W
300
/DAY
200
200
100
100
O
-100
-100
200
.300
D
F
D
F
A
3
A
0
D
3
A
O
D0
a
- I
I
D
F
A
S
L
A
II
I
F
Ii
A
I
Ir
J
Figure 1
I
* I
I
A
O
D
Im
A
I=
O
-300
D
LI
r
0
D
F
A
3
A
O
I
II
n
II
B
m
F
A
3
A
O
D
D
35oN - 145 0 E
200
/CM
2
35ON -155oW
35ON -125oW
/DAY
100
-100
200
300
-300
400
D
I
I
F
A
I
J
A
0
D
I
I
1
I
A
D
F
A
J
A
IOr- T-T
Figure 2
I
0
-
D
D
F
A
J
A
0
D
-50-
35oN - 170
0
E
40ON -150oE
-401
D
F
A
J
A
O
D
Figure 3
D
F
A
J
A
O
D
50ON - 165oW
50ON - 170oE
200
SON
- 1450 W
200
100
0
-100
-100
-200
-200
-300
-300
'
-
A
J
A
0
D
D
F
A
J
A
0
D
D
F
A
J
A
0
-400
D
I
A
I
J
I
A
0
D
D
F
I
A
J
I
A
I
0
D
0
F
A
J
A
0
D
I
F
IOr T-T
K
I
F
Figure 4
-52-
the average or normal values of the total heat transfer across
the sea surface and the sea-surface temperature suggests that
the year-to-year changes in the total heat transfer for a
particular month will be accompanied by year-to-year changes or
anomalies in the sea-surface temperature for the same month.
This is indeed the case, as will be shown in Chapter V.
Figures A84 through A95 show the 7-year average values and
standard deviations of the surface air temperature SAT.
The
results of the Fourier analysis show that a regular yearly
cycle is also present in the air temperature, with over 95
per cent of the series variance accounted for by the first
harmonic.
The maximum and minimum values of SAT occur at the
same times as those of the sea-surface temperature, August
through early September and February through early March
respectively.
In both the sea and air temperatures, the
maximum values tend to occur somewhat earlier (early August)
in the western part of the ocean than in the eastern part
(early September).
Lthat thLLe heaLLig cyLe
This result is also reflected in the fact
Lasts
approximately
1
month
LIonger
in
the eastern part of the ocean than in the western part; there
is a definite shift from August to September in the time when
the heating cycle ends as one moves eastward across the ocean.
The strength of the California Cuirrent and t-he accomp-
anying cold advection in the months of June, July and August
-53-
may play an important role in determining the end of the heating
cycle in the western part of the North Pacific.
During these
months, the strong advection of cold water keeps the sea-surface temperature somewhat lower than it would be if there were
no advection.
These lower temperatures, in turn, keep the tran-
sfer of latent and sensible heat down and let Q remain positive
(heat gain by sea surface) throughout August and early September.
As the advection weakens, the ocean temperature rises
until the incoming radiation is balanced by the heat lost from
the surface due to back radiation and latent and sensible
heat transfer.
At this point, the surface layer of the ocean
begins to lose more heat than it gains, and the sea-surface
temperature begins to fall.
The relationship between the yearly cycles of the
normal values of SST and SAT indicate that year-to-year
fluctuations or anomalies in SST will be accompanied by corresponding fluctuations in SAT.
From correlation studies between
the anomalies of the two data series, reported on in Chapter
V, it was found that this is exactly what occurs.
Figures A96 through A107 show the 7-year average values and
standard deviations of the month-to-month change in the seasurface temperature ASST.
These values were computed by taking
the difference between the average value of a particular month
and the following month and the average value of the same
month and the preceding one.
We thus have linear approximations
-54-
to the change in SST across each of the 84 months in the data
series.
The results of the Fourier analysis show that there
is a regular yearly cycle inL SST, especially in the lower
Above 350 N the second harmonic becomes important
latitudes.
and accounts for approximately 15 per cent of the variance in
these areas.
The maximum increase of the sea-surface temperature
over a month occurs from the middle of May to the middle of
June, while the maximum decrease occurs from the middle of
December to the middle of January.
During the times of little
or no change in the sea-surface temperature (early march and
early September) the heat content of the ocean surface layer
changes very little, and a balance must exist between the
transfer of heat across the sea surface and the advection within
the surface layer.
The monthly mean values of the observed daily wind speed
were averaged over the 7-year period and subjected to Fourier
analysis.
The results show that there is a regular yearly
cycle over most of the ocean areas with the exception of the
area 200 N to 3001Vand from 110*W to 170 0 E.
Over the rest of
the ocean, the first harmonic accounts for over 90 per cent
of the variance and, in the central and western parts, for over
95 per cent.
The largest amplitudes, 250 cm/sec, occur
between 350 N and 500 n and from the Asian coast eastward to
longitude 175 0 E.
Large amplitudes also appear in the Gulf
of Alaska, a region of frequent cyclogenesis.
As expected,
-55-
the largest values of the wind speed occur during January and
the minimum values in July.
The values of the Bowen ratio (Qs /Qe) were also averaged
over the 7-year period for each month and Fourier analyzed.
The results show that a regular yearly cycle occurs only in
the middle latitudes of the western part of the North Pacific.
In this area the ratio has-an annual range of .4 to 1.4 with
the maximum values occurring in January and the minimum values
in July.
In the other areas, the Bowen ratio has a value
of about .1 with little annual variation.
4.
Using the heat balance equation in the form
dkwhere H is the depth of the upper mixed layer of the ocean,
v is the velocity within the mixed layer, T is the temperature
of the mixed layer, and Q is the total heat transfer across
the sea surface, an estimate of the total heat advection within
the mixed layer, or the left-hand side of the equation (IV.2),
can be obtained using the observed values of the change in
heat content of the layer and the calculated total heat transfer.
Using the values of JT/4t, orA SST, and Q determined
previously and values of H taken from a study of the thermal
-56-
condition of the North Pacific Ocean by Pattullo and Cochrane
(1951), estimates of the heat advection within the mixed layer
were determined at each of the 162 grid-points for each of the
12 months.
The values are given in Figures A108 through A119.
Due to the uncertainties in the monthly distributions of
the mixed layer depths H, not much confidence can be placed
in these computed values of heat advection.
However, several
features of the monthly maps appear to be in agreement with
what is known about the current patterns in the surface layer
of the North Pacific, especially during the months from OctDuring these months, there is strong advec-
ober through May.
tion of warm water from the lower latitudes to the higher latitudes in the western part of the ocean; this is particularly
true in the region of the Kuroshio Current.
In the eastern
part of the ocean off of the California and Mexican coasts,
there is a weaker advection of cold water from the higher
latitudes to the lower ones.
through September.
This is also the case from June
However, during these 4 months the maps
show that there is cold advection of water within the surface
layer in the region of the Kuroshio in contrast to what is
expected due to the transport of the geostrophic and winddrift currents.
The reason for this discrepancy probably lies
in the fact that the monthly charts of the mixed layer depth
for the region show a marked decrease from June through September, going from 250 feet in May to 50 feet in June.
Looking
-57-
at the first term on the right-hand side of equation (IV.2),
it is seen that for positive values of Q and JT/Jt, small
values of H can give negative values or cold advection for
the left-hand side.
Another uncertainty in the method is the amount of heat
that is lost from the mixed layer across the seasonal thermocline.
Since this heat loss is believed to be compensated by
an up-welling of cold water through the region of the thermocline, the large negative values calculated for the advection
could reflect this heat loss or up-welling.
In the following
simple model of the surface layer, it will be shown that the
heat received or lost at the surface is balanced by a change
in the temperature of the surface layer, horizontal advection
of heat into or out of the layer, and loss of heat across the
region of the thermocline.
The equation of heat flow may be written as
where T is the temperature, u, v and w are velocity components
in the x,y and z directions, and K is the vertical heat exchange coefficient (only vertical mixing considered).
Using
the equation of continuity in the form
C)
(IV.4)
-58-
and placing the direction of flow along the y-axis (u = o),
we get
'
(IV.5)
Integrating over the mixed layer depth H in which T is
independent of z and letting V =
fvdz, then
-CP
since w = o and/k
T/Jz = Q/
cp at z = o.
In the discontinuity layer or the region of the seasonal
thermocline, we assume that the downward flux of heat due to
thermal conduction is balanced by water ascending with the
constant velocity w.
We can then write the relation
w (T - TD ) = A-
JT
(
in which (T - TD ) is the temperature gradient across the
discontinuity layer.
Substituting equation (IV.7) and the vertically integrated
continuity equation,
-H
-59-
or
JV
-
into equation (IV.6) we obtain
4V7
-
(IV.8)
This equation states that the total heat transfer across the
ocean surface (Q) is balanced by the change in temperature
of the mixed layer, horizontal advection within the layer,
and a loss of heat into the region of the thermocline.
The
last mechanism heats the water ascending from below, which
passes into the horizontal flow within the surface layer.
Using a value of 2 x 10- 5 cm/sec for the vertical velocity
through the discontinuity layer (Wyrtki, 1960) and 200 C for
T - TD, a value of 35 cal/cm2/day is obtained for the heat
loss from the surface layer due to vertical mixing at the
bottom of the layer.
Although this value is not large, it
would reduce the magnitude of the right-hand side of equation
(IV.2) and help to give more reasonable values for the horizontal heat advection.
The values of the heat advection determined from the heat
balance equation were integrated over each latitude circle
at 5-degree intervals from 200N to 550 N and also over the
-60-
entire North Pacific Ocean from 17.5 0 N to 57.5 0 n.
of the calculations are shown in Table IV.
The results
It is seen from
the table that warm advection occurs over the ocean from January through May, with the largest values occurring in March
where 14.6 x 1014 cal/sec is advected into the surface layer.
Cold advection occurs from June through December, with 6.6
x 10
14
cal/sec being advected out of the surface layer above
17.5 0 N in September.
As was mentioned above, the values of cold advection during
the summer months are uncertain due to the unreliable nature
of the mixed layer depth values, especially in the western
part of the ocean.
In a study on the meridional heat transport
by ocean currents, Bryan (1962) used a method combining
hydrographic station data and climatological values of surface
wind stress to compute the meridional transport of heat across
various latitudes in the Atlantic and Pacific Oceans.
Using
data obtained during August of 1955 at 320 N in the North Pacific
Ocean, he calculated that there was a southward transport of
heat by large-scale motions of approximately 2.8 x 1014
cal/sec.
From Table IV it is seen that a southward transport
of heat of 2.2 x 1014 cal/sec occurs at 320 N.
.
While
the magnitudes of the two calculations differ by a factor of
1.3, the direction of heat transport is the same for both studies.
Bryan concludes that more east-west sections are needed to
determine whether this southward heat transport is characteristic
Table IV
LAT
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
DEC
-. 3
-. 2
55 0 N
.2
.1
.6
.4
.3
-. 1
.0
-. 1
-. 3
-. 2
50 N
.2
.4
.8
.5
.2
-. 2
.0
-. 1
-. 7
-.9
-. 5
45 0oN
.3
•7
1.2
.5
-. 1
-. 4
-. 2
-. 3
-1.4
-. 5
-,6
40 ON
1.0
2.3
1.2
.4
-. 2
-. 2
35 0 N
1.0
2.9
1.6
.9
.0
-. 5
300 N
-. 3
1. 3
2.7
1.3
.4
-. 3
-. 6
25 0 N
-. 4
.7
2.4
1.1
.3
-. 5
-. 4
-. 4
20 0 N
.2
.4
1.8
.9
.7
.2
-. 3
.5
2.1
7.3
14.6
7.5
3.2
-1.5
-2.2
-3.3
Total
Area
Integral
Units:
1014 cal/sec
-1.1
-1.1
.7
-.7
-. 4
.1
-.6
-. 4
.2
-. 1
-1.0
-.1
-. 4
-2.
-. 6
*I
-2.I
-6.6
-.8
-5.6
-1.2
-5.8
-1I.4
-. 1
-3.8
-62-
of other years and other seasons.
The results of the present
work indicate that this transport is not characteristic of
all seasons, since a northward transport of heat occurs
during the winter and spring months at all latitudes above
200 N.
B.
Results Using Daily Averaged Data
In order to find out how much variability occurs during
a month in the heat transfer terms and in the sea-surface
temperature, daily values of certain meteorological parameters
were obtained for ocean stations PAPA (500 N - 145*W) and
NOVEMBER (300 N - 1400 W) in the North Pacific Ocean for 1962
(see Chapter III).
The data were used in the heat transfer
equations developed in Chapter II to compute
Qi
and Q on a daily basis.
Qb' Qe' Qs
These daily values of the heat
transfer terms were then averaged over each of the 12 months
to get monthly averaged values; in addition, the standard
deviations for each of the terms were calculated for each
month.
1.
The monthly averaged values of Qi, Qe + Qs , and Q
are shown in Figure 5 for station NOVEMBER and Figure 6 for
station PAPA; at the bottom of each of the Figures, the
value of the sea-surface temperature for each month minus the
-63-
500
2
oorAL/CM
/DAY
S Qi
\f
/d
400
300
*
•~~~'
.
;.
.''*.
* *
.*.....,,/ Q
+Q
•
.***
N
•-'
\
,
200
100
0
A
M
.J
A
J
S
-100
-200
STATION
NOVEMBER
(300 N - 1400 W )
-300
T3
2
I
I
-----J
I
F
M
---A
I
I
M
-1
-2
-3F-
Figure 5
I
I
J
J
A
A
I
S
0
0
N
N
I
D
D
-64-
400t CAL/CM 2/DAY
~--4
-.
Ni
300
200
100
/
"
A
STATION
M
J
J
A
J
A
N
D
N
I
PAPA
(500 N - 145*W)
S
Figure 6
S
O
-65-
yearly mean is also given.
At each of the stations the heating
cycle begins in March and ends in September.
The variations
in the total heat transfer Q during the summer months at
station NOVEMBER reflect the variations in the latent and
sensible heat flux Qe + Qs throughout these months.
At station
PAPA the large negative value of Q in November is a consequence
of the large value of Qe + Qs during this month.
The values
of the effective back radiation show little monthly variation
and have values of about 120 cal/cm2/day for station PAPA
and 135 cal/cm2/day for station NOVEMBER.
2.
Tables V and VI show the monthly averaged values and
standard deviations over the daily values of Qi,
Qb' Qe' Qs'
and SST for stations NOVEMBER and PAPA respectively.
Q
The
values of the standard deviations for Qi are comparatively
low and reflect the fluctuations in the cloud cover during
the individual months.
The standard deviations for Qb are
also low (approximately 20 per cent of the average values)
and again reflect the fluctuations in the daily cloud cover
values.
However, the standard deviation values for Qe and
Qs are extremely large, at times exceeding the magnitudes
of the average values themselves.
These large fluctuations
throughout each of the months are due to the combined effects
of fluctuations in the vapor pressure or temperature differences between sea and air and the daily values of the surface
Table
V
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
DEC
217
307
355
432
466
456
465
499
437
346
287
205
41
61
64
84
76
81
90
84
72
51
33
42
128
141
145
134
141
125
129
129
132
136
155
136
21
31
24
23
27
16
17
16
17
20
22
32
Qe
266
313
217
194
247
198
258
192
210
306
263
319
s
169
235
134
103
115
103
116
105
192
138
147
199
Qs
31
40
30
17
38
16
28
13
10
26
31
29
S
34
39
27
20
27
17
12
10
15
13
24
37
Qi
s
s
s
SST
s
-162
-278
-208
-186
-38
87
40
117
52
165
84
-121
221
296
170
157
152
147
167
157
233
178
184
233
18.7
18.5
18.1
18.8
19.8
20.2
21.5
22.6
23.3
23.0
22.0
20.2
.5
.0
.0
.3
.7
.3
.9
.4
.5
.7
1.0
.2
Monthly averaged values and standard deviations of Qi
0
0
and SST for station (NOVEMBER) 30 N - 140 W);
Units:
cal/cm 2/day for Q's and 00 for SST.
',Q0, Qs
Q
~.-_-- --L--LI
I
Table VI
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
DEC
59
121
201
334
357
355
346
286
236
153
79
50
15
30
47
68
77
61
59
60
50
32
17
12
105
121
118
136
112
100
94
95
119
124
132
121
s
Qb
s
36
29
29
25
28
20
15
20
27
32
33
37
Qe
40
56
104
114
77
57
28
52
159
191
355
116
s
98
84
94
71
75
69
37
52
153
177
-29
10
30
44
8
5
-7
-1
22
23
110
30
58
36
58
65
36
23
12
15
51
45
117
97
-57
-65
-52
40
160
193
140
-63
-185
-517
173
125
160
129
99
94
59
73
202
218
367
271
5.9
5.9
5.9
6.2
9.0
11.6
13.1
13.5
10.3
8.6
6.9
.4
.3
1.0
1.0
.6
.8
.6
.3
Qi
Qs
s
Q
SST
.2
.5
7.2
.5
231,
Monthly averaged values and standard deviations of Qi'
and SST for station PAPA (500 N - 1450W).
Units:
cal/cm2 /day
for Q's and oC for SST.
%#
240
.6
Qe,
Qs'
Q
158
-217
-68-
wind speed.
However, since the transfer equations are non-
linear in nature, it is not possible to determine which of
the terms is primarily responsible for the large daily
variations in Qe and Qs"
The large standard deviation values for Q are a result
of the combined effects of daily fluctuations
in Qi
Qe and Qs.
Qb'
However, since Qe and Qs have such large standard
deviation values with respect to their average values, it is
these 2 terms that are primarily responsible for the large
daily variation of the total heat transfer Q.
Compared to the heating terms, the standard deviations
of the sea-surface temperature are quite low.
This result
suggests that the large variability of the total heat transfer
on a daily basis is compensated by fluctuations in the depth
of the mixed layer and by fluctuations in the strength of the
horizontal advection within the mixed layer caused by the wind
stress on the ocean surface.
A study of the mixed layer
depth values for station PAPA during 1962 by the Pacific Oceanographic Group of the Fisheries Research Board of Canada shows
that the standard deviations of the daily layer depth values
over a month can be as large as 40 per cent of the monthly
averaged values; the yearly average of the standard deviation
values is 17 per cnt of the monthly mean layer
epths.
A
study of the daily values of the total heat transfer across the
-69-
surface and the daily change in sea-surface temperature at
stations NOVEMBER and PAPA showed that there was very little
correlation between the two terms.
-70-
V.
Discussion of Sea-Surface Temperature
and Heat Transfer Anomalies
A.
Spatial and Temporal Scales of Anomalies
1.
In addition to the 7-year average values of SST and
SAT for each month, the departures of each yearly value for
a given month from the average or normal value also were
computed.
These departures or anomalies were printed out in
map form for each month and each quantity and also punched
onto cards for further analysis.
Figures 7 and 8 show typical
maps for the sea-surface and air temperature anomaly patterns.
(Many of the results reported below first appeared in a
report written in 1966 by Dr. Hurd C. Willett and myself
for the National Science Foundation.
This report will be
quoted extensively, especially when discussing the magnitude
and duration of the SST and SAT anomalies.)
In looking at the anomaly maps of both SST and SAT for
the 7-year period, it is seen that values of the same sign
and magnitude occur in fairly extensive geographical areas.
0
The magnitudes of the SST anomalies have values of 2 F to
40 F except in the coastal areas and particularly off of the
Asian coast in winter, where larger values occur.
The SAT
anomalies are slightly higher than those of SST; however,
this difference is usually not greater than 50 per cent.
Ex-
ANCPALY OF SEA SLRFACE TEMP.
195
7
JANUARY
55h
J
-
-2
-2
130
4
-
3
1
1
-2
140
a
N
6
17
25
23
18
14
16
0
4
9
8
4
18
17
12
12
1
17
19
24
19
16
18
22
19
4 2
5
4
1
2
4
6
16
19
0
24
25
2
4
-6
-7
-S
-9
-7
-4
-2
6
13
19
21
19
14
3
-
1
7
13
13
1
2
7
6
8
16
12
3 -5
150
140
7
-6
0e
14
6
18
13 -19
5 -7
150
Figure 7.
-15
-8
160
-17
-18
-6
-16
-13
-4
170E
-13
9
-2
-10
-12
-8
-14
1 -
180
t9
-1
-1
3
170W
8
160
Anomaly pattern of sea-surface temperature.
December 1957.
2
-4
(OF x 10)
130
120
110
ANCMALY OF
1957
---
JANUARY-
8
55N
50N
-1
34
45N
40-
(
-5
14
35N
2
-7
4
-3
3
-1
-2
-9
1
-2
-6
130
-13
-10
140
4
-3
-1
9
6
2
-8
-3
-15
-17
-11
-
2
- 2
-2
3
-
12
3
-
-1
2
-3
-I3
20N
9
150
Figure 8.
-5
-
3
-16 -i6 -1
i
-
-15
160
-15
-10
17WE
17
-6
-
o
31
3
22
18
13
3
-1 -13 -31
1
6
33
26
18
I
Z
-4
T2
25
23
1
--
12 -11
1 - 6
180
9
4
3
31
F
-
-14
4
23
1
2315
19 -18
9
-15
-5
23
2623
3
Z
8
1
- 13
25N
36
-5
I 1<:Di14
-1
3
-4
-
-4
-
- 0
3
5
5- /
-3
-
-4
4
3
(3
-7
-18
170W
160
Anomaly pattern of surface air temperature.
150
December 1957.
1
1
-25 -12
-4
6
-25
-32 -14
4
14
-1
140
(oF x 10)
130
120
2
23
T3
110
-73-
cept for these coastal regions, there is no outstanding
difference between anomaly magnitudes for the different
months and seasons.
Although there is a large degree of month-to-month
persistence in the SST anomaly patterns, in many cases
geographically extensive areas of large anomalies are eliminated or even reversed in sign in a single month's time.
It
is apparent that the effect of even a strong anomaly pattern
in SST is not sufficient in some cases to prevent a change
in the large-scale wind pattern and other meteorological
parameters from completely destroying the anomalies.
In contrast to the month-to-month persistence, year-toyear persistence of the large-scale features of the anomaly
patterns is not at all apparent.
There appear to be no more
cases of notable year-to-year persistence of a strong anomaly
pattern than there are of complete elimination or even
reversal of such a pattern.
This conclusion applies to both
sea-surface and air temperatures.
For weak anomaly patterns,
the year-to-year comparison generally suggests that a random
relationship exists.
For a more objective analysis of the persistence of the
monthly mean SST and SAT anomalies, a selection was made of
15 points, the intersections of the 25th, 35th
and 45th
parallels with the five meridians 150 0 E, 170 0 E, 170 0 W,
-74-
150 0 W and 130 0 W.
This group of points covers most of the
central North Pacific and permits a reasonable north-south
and east-west comparison of conditions to be made.
For each of the 15 grid-points, a count was made of the
total number of changes of sign of the anomalies that occurred
during the 84-month period and also of the longest unbroken
period of months for which the sign of the anomaly remained
unchanged.
The average number of changes of sign of the SST anomalies
over the period is 25, indicating that the average duration of
an anomaly of the same sign is 3 and 1/3 months.
The corres-
ponding figures for the monthly mean anomalies of SAT are 28
and 3 months, respectively.
For the SST anomalies, the
average maximum period of consecutive months without change
of sign is 14, with the maximum number of months at a point
being 20.
The same numbers for the SAT values are 13 and 19
months, respectively, and indicate that there is no significant
difference between the stability of the sea and air temperature
anomalies on a monthly mean basis.
There is no latitudinal variation in the stability of
the anomaly values of SAT.
SST anomalies, however, show a
tendency toward maximum stability at 350 N and toward minimum
stability at 250 N.
In addition to the "count study", auto-correlations of
SST and SAT anomalies were made at 1, 2 and 3 months lag for
-75-
all 162 points on the map over the 84-month period.
For the
SST anomalies, the average value of the correlation coefficients
over the 162 points is .53 at 1 month lag, .35 at 2 months lag,
and .28 at 3 months lag; the maximum values of the individual
coefficients occurred at 350 N for all 3 lags.
For the SAT
anomalies, the corresponding values are .40, .28 and .25,
respectively; the latitudinal averages were approximately the
same at all latitudes.
These results tend to support the
conclusions obtained from the "count study".
Lag correlations at 1 and 2 years also were computed for
the SST and SAT anomalies on a seasonal basis, i.e. the 3
individual months of a season were correlated with the corresponding 3 months of the same season 1 and 2 years later.
There
is a consistent negative auto-correlation at all of the 15
selected points mentioned above at both the 1 and 2 years lag
for both the sea and air temperature anomalies.
This negative
correlation again supports the conclusion of the "count study"
that there is no year-to-year persisitence of these anomalies.
In order to determine how the sea and air temperatures are
related to one another, contemporary correlation coefficients
between the SST and SAT anomalies were computed for the entire
162-point network on a monthly seasonal, yearly, and 84-month
basis.
The following discussion of the results will concern
the 15 selected grid-points for the monthly, yearly and 84-month
period basis and the entire 162 grid-points for the seasonal
-76-
basis.
The average value of the correlation between SST and SAT
anomalies over the 12 months and the 15 points is .75, with
only 4 of the 180 coefficients slightly negative.
By month
the coefficients ranged from .53 for January to .88 for May
and September.
The only significant variation of the corre-
lation by latitude or longitude is that the average value of
the 36 coefficients on the 150 0 E meridian is .67, while the
average value of the other 4 meridians is .77.
This difference
reflects the effect of the Asian continent on the air masses
that move over the ocean during the winter months.
For the correlations computed on a seasonal basis, the
162-point average values are .51 for winter, (December, Janand February) .69 for spring, (March, April and May)
uary
.83 for summer (June, July and August) and .78 for fall
(September, October and November).
For all seasons except
winter, the largest values occurred along 350 N.
When the correlation between the anomalies is computed by
the 12 months of each of the 7 years separately, the average
value of the 15 points is largest in 1957 (.76) and smallest in
1953 (.55).
Again the largest values tend to occur along the
35th parallel.
The combination of all 84 months of the period raises the
overall average correlation to .77 and reduces the geographical
range of the coefficients.
The latitudinal averages go from
-77-
.73 on the 25th parallel to .80 on the 35th, and the meridional
values from .69 on the 150 0 W meridian to .79 on the others.
The principal facts learned from the above studies are
that there is very little long-term persistence of both the
sea and air temperature anomalies and that the anomaly patterns
of both aea and air temperature are strongly related to each
other.
As Dr. Willett points out, "If feedback from ocean to
atmosphere is to be accepted as a primary factor in the maintenance of long-term persistently anomalous patterns of the
general circulation, then the water temperature anomalies must
be markedly more stable (persistent) from month-to-month and
year-to-year than are those of the atmospheric circulation.
We know that the large-scale anomalies of the atmospheric circulation fluctuate in large amplitude in relatively short
periods of time.
To maintain a long-term persistent anomaly
of the general circulation, we require some factor of control
that is stable over correspondingly long periods, to the extent
that it can pull the general circulation back into line when
it goes off on a tangent."
From the investigation of the
persistence of the sea-surface temperature anomalies in the
North Pacific Ocean, it does not seem that these anomalies
provide the required control factor.
2.
Monthly departures from the 7-year average values
of the heat transfer terms Qi, Qb' Qe' Qs and Q also were
-78-
computed, printed out in map form, and punched onto cards.
The monthly departures or anomalies of the incoming
radiation Qi reflect the year-to-year variability in the cloud
cover over the ocean.
Their spatial scales are not as ex-
tensive as those of the sea-surface and air temperatures and
are not as geographically coherent, i.e. high and low values
as well as those of opposite sign occur close together at times.
The magnitudes of the anomalies range from 5 to 20 per cent
of the monthly average values and show very little month-tomonth persistence.
Auto-correlations of the Q. anomalies at 1, 2 and 3
months lag were computed over the 84-month period.
The
average value of the 162 correlation coefficients on the map
is .04 at 1 month lag with very little latitudinal variation;
a maximum value of .12 occurred at 200 N, 25*N, and 30*N and
a minimum value of -.08 occurred at 550 N.
The average values
at 2 and 3 months lag were both negative, indicating that
there is very little persistence in the
Qi anomalies due to
the lack of persistence in the cloud cover anomalies.
The anomaly patterns of the effective back radiation
Qb
also show very little geographical size or coherence, although
they are more extensive than those of Q..
As was shown by
the size of the standard deviation values of Qb discussed in
Chapter IV, the magnitude of the anomalies are comparatively
small, ranging from less than 1 per cent to 10 per cent of the
-79-
monthly average values.
The average value of the correlation
coefficients is .02 at 1 month lag and negative at both 2 and
3 months lag.
These results show that there is very little
persistence in the Qb anomalies, as is the case for those of
Q..
The anomaly patterns of the latent heat transfer Qe are
much larger in geographical size and magnitude than those of
Qi and Qb and also tend to have more geographical coherence.
At times, 25 per cent of the entire map is covered by an
anomaly pattern of the same sign and magnitude.
The magnitudes
of the anomalies vary widely over areas of the map and from
month to month, ranging from less than 1 per cent to over
75 per cent of the monthly average values.
There is surpris-
ingly little persistence in the latent heat anomaly patterns,
with the average value of the auto-correlation coefficient
at 1 month lag being -.15; the coefficients at 2 and 3 month
lags are also negative.
It would be interesting to know which factor in the transfer equation is primarily responsible for the month-to-month
and year-to-year fluctuations in the latent heat transfer.
However, due to the non-linear nature of the equation, it
is not possible to separate the effects of fluctuations in
the vapor pressure difference and in the surface wind speed.
A perturbation analysis of the transfer equation was tried in
which the equation was written as
-80-
QeoC
+
- -
U
q'
+
u
+
(V.1)
q'u'
where Qe is a monthly value of the latent heat transfer, and U are the 7-year averages of the specific humidity and
wind speed for a particular month, and
q' and u' are the
departures of particular monthly values from the average
values.
If the 4th term and either the 2nd or 3rd terms on
the right-hand side of equation (V.1) were small compared to
the other 2 terms, we could say that the remaining 2nd or
3rd term was responsible for the fluctuations in Qe since
q u is a constant.
However, it turns out that the last 3
terms are about the same order of magnitude and no information
is gained.
One can find out something about the relative importance
of the ocean and atmosphere in fluctuations of the vapor
pressure or specific humidity, since the variance of the
difference can be written as
V (qs -a
)
=
CV qs(q s - qa )
where V denotes "variance of",
+ CV qa' (q-a
q)
,
(V.2)
CV denotes "covariance
between", and qs and qa are the specific humidities of the
sea and air, respectively.
-81-
The two covariance terms were computed over the 7-year
period for the mid-season months of January and July.
For
the month of January, it was found that fluctuations in the
specific humidity of the air account for 2/3 or more of the
variance in (qs - qa ) at 135 of the 162 points on the map,
that fluctuations in the specific humidity of the sea surface
account for 2/3 or more of the variance of (qs - qa ) at 14
of the 162 points, and that the fluctuations of both quantities are of approximately equal importance at 13 of the 162
points.
For the month of July, fluctuations in qs account
for 2/3 or more of the variance in (qs - qa) at 69 points,
fluctuations in qa account for 2/3 or more of the variance
at 67 points, and the fluctuations of qs and qa are of equal
importance at 26 points.
From this it is concluded that
fluctuations in the specific humidity of the air are more
predominant than those of the sea surface during the winter
months, and that the two fluctuations are of equal importance during the summer months.
The anomaly patterns of the sensible heat transfer Qs
are similar to those of Qe in that they are geographically
coherent and can cover a large portion of the entire map
during some months.
Their magnitude also varies widely from
month to month, ranging from less than 1 per cent of the
monthly average values to over 100 per cent during the summer
months when the sensible heat transfer is small.
The
-82-
anomalies also have very little month-to-month persistence;
the average auto-correlation coefficient at 1 month lag is
-. 15 and the values at 2 and 3 months lag are also negative.
The same type of perturbation analysis of the transfer
equation for sensible heat was attempted as was done for the
latent heat equation.
Again, the 4 terms of the expansion
were of approximately equal magnitude, and nothing was learned
about the relative importance of fluctuations in the sea-air
temperature difference and the wind speed in determining
fluctuations in the sensible heat transfer.
However, a
covariance analysis of the sea-air temperature difference
was done with the following results:
for January, fluctuations
in the air temperature Ta account for 2/3 or more of the variance in the sea-air temperature difference (Ts - T a) at
132 points, fluctuations in T
s
variance in (T
s
T
s
and T
a
account for the same amount of
- T ) at 20 points, and the fluctuations of
a
are of equal importance at 10 points;
for the month
of July, the corresponding numbers are 99, 45 and 18,respectively; as for the specific humidity, air temperature fluctuations
are more predominant during the winter months while both air
and sea temperature fluctuations are of approximately equal
importance during the summer months.
The anomaly patterns of the total heat transfer Q vary
widely over the map for a particular month and from month-tomonth during the 84-month period.
Since the anomalies of
-83-
Qi and Qb are comparatively small, these fluctuations are
primarily due to the fluctuations in Q
e
and Q
.
s,
The magni-
tudes of the Q anomalies range from less than 1 per cent of
the monthly average values to over 200 per cent, with the
largest values occurring in the summer months.
Although
there is some geographical coherence in the anomalies, it is
not as great as that for Qe and Qs.
From the values of the
auto-correlation coefficients computed at 1, 2 and 3 months
lag over the 84-month period (-.14, -.14 and -.09),
there
appears to be very little persistence in the anomaly patterns.
This result is not surprising, since there was also very little
persistence found in the 4 quanitities that make up the total
heat transfer.
-84-
B.
Relationships Between Sea-Surface Temperature Fluctuations
and Heat Transfer Anomalies
In Section A we saw that sea-surface temperature anomalies
of fairly extensive geographical size occurred in all of the
84 months under investigation.
The magnitudes of the anomalies
varied widely from area to area and from month to month.
It
was also noted that the anomaly patterns exhibited some persistence in that the auto-correlations at the individual
points remained positive on the average to about 3 months lag.
Two questions arise from these observations:
(1) What
causes the anomaly patterns of sea-surface temperature to be
so different from year to year during the same month and in
the same area; and (2) What accounts for the month-to-month
persistence of the anomaly patterns.
The first question will
be dealt with below in sub-sections 1, 2, 3, 4 and 5, in which
the results of correlation studies between anomalies in the
heat transfer between ocean and atmosphere and anomalies in
the sea-surface temperature will be discussed.
The second
question is not easily answered, but will be discussed in
sub-section 6.
1.
In an attempt to account for the large-scale anomaly
patterns that occur in the sea-surface temperature, a series
of correlation studies was made in which the 7-year average
values of Qi, Qb' Qe' Qs' Q and6SST for each month were
subtracted from the individual monthly values and the results
-85-
correlated against each other over the 7-year period.
correlation coefficients were computed in 3 ways:
The
(1) using
the 7 pairs of anomaly values for each of the 12 months;
(2) using the 21 pairs of values for each season of the 7
years; and (3) using all 84 pairs of values for the 7-year
period.
The anomalies of the change in sea-surface temperature
over a monthASST were used instead of the SST anomalies since
it is the change in temperature over a time period that is
affected by heat transfer during a particular month; using the
SST anomalies themselves would not be correct since part of
the anomaly could have been caused by heat transfer during
a preceding month.
The results of correlating the anomalies of the individual
heat transfer terms Qi' Qb' Qe and Qs with the anomalies of
ASST show that no one term is primarily responsible for producing fluctuations in the sea-surface temperature.
However,
when the anomaly patterns of the total heat transfer across
the ocean surface Q were correlated with the anomaly patterns
of6 SST, significant results were obtained.
On a seasonal basis, i.e. using 21 pairs of values in
each correlation coefficient, the average values of the
162 coefficients on the map are .36 for winter (December,
January, and February), .21 for spring (March, April and May),
.27 for summer (June, July and August) and .27 for fall
(September, October and November).
The largest individual
-86-
values occur in winter, in which there are 67 points at which
the coefficients are larger than the 5 per cent confidence
limit.
The corresponding numbers for the other seasons are
49 for spring, 41 for summer, and 40 for fall.
The latitudi-
nal averages of the correlations for winter are shown in
Table VII along with those for the other three seasons.
Although the magnitudes of the coefficients are not very
large, the fact that they are positive over the entire map
in each of the seasons indicates that the sea-surface temperature anomalies are related to anomalies in the total heat
transfer across the sea surface.
In addition to the seasonal correlations, all 84 months
were combined and a correlation was computed at each of the
162 grid-points over this period.
Figure 9.
The results are shown in
Of the 162 coefficients on the map, 102 reach the
5 per cent confidence level of .22 and 86 reach the 1 per
cent level of .28.
In order to get some idea of the variability of the relation between anomalies of Q and those ofA SST, pattern
correlation coefficients between the two quantities for
each of the 84 months of the period were computed.
They range
from -.27 in March of 1953 to .73 in December of 1953.
Over
the 7-year period, the month of December has the largest average value of .43, while the month of March has the lowest
average value of .10.
-87-
Table VII
Winter
Spring
Summer
Fall
Year
55 0N
.20
-. 18
.29
.16
.12
50 0 N
.34
.18
.31
.31
.28
45 0 N
.38
.24
.. 26
.36
.31
40 0N
.40
.24
.40
.42
.36
350 N
.45
.35
.31
.34
.36
300 N
.45
.27
.21
.25
.30
25 0N
.27
.32
.20
.20
.25
20 0 N
.35
.26
.18
.09
.22
Average latitudinal correlation coefficient
between anomalies of total heat transfer Q
and anomalies of change in sea-surface
temperature 6 SST.
CCEF.
CCRR.
A/V/
4e5/7 VS,
ALL MONTHS
Awvo
n Q
1951-1957
55N
.17
o
e
.24 .26 .15-.01 .02 .24 .28 .28 .39 .43 .41 .40 .29 .32(
50N
45N
.06 .08
.15
.21
.33
.35
14 .28
.21
.16
.30
.36 .32
.27
._O
.23 .37
.40 .44 .49
.21
20N
.d
20 *
0,%*0
.17-.01-.
.25 .23
.18 .26 .35
.34 .34 .48
.41
.57 .53
.46 .46
.42 .51
.54 .54
.51
.42 .41 .44 .37 .15
.44 .47
.51 .47
.52 .47
.42 .44 .46
.25
.32
.29
.34 .30
.28
.29 .20
.19
.17
.28
.36 .39 .54
.54 .48
.28 .23
.35 .41 .41
.34 .26
.20 .10 .12 .09
.34
.21
.19
.33
.36 .24
.16
.15
.13
.28 .42 .51
.45 .35 .20 .25
.35 .49 .44
.11 .08
.03
.04 .10
.11
.36 .31
.31
.37
.29 .19
.20
.21
.21 .26
.10
.11 .20
.30 .38
140
150
130
120
110
C
130
Figure 9.
.42 .40
160
170E
.31 .26 .13 .18
180
170W
.11 .32
160
.36 .17
.08 .18
150
140
.15
Correlations between the anomalies of total heat transfer across the sea-surface Q and anomalies ofd SST based on 84
months of data from 1951 through 1957.
-89-
There is very little variability between years in the
12-month average value of the pattern correlation coefficients;
they range from .21 in 1951 to .28 in 1953, 1955 and 1957.
Whether averaged over the 7-year period for each month or over
the 12 months of each of the 7 years, the pattern correlation
coefficients are all significant at the 5 per cent confidence
level and many are significant at the 1 per cent level.
2.
Another method of changing the temperature of the sea
surface is by the horizontal advection of surface water across
sea-surface temperature iso-therms,
Adopting a method first
proposed by Namias (1959, 1965), in which he considers that
the advection of surface water is about 450 to the right of
the surface geostrophic flow, the change in sea-surface temperature was computed at each grid-point for each month using
the north-south and east-west gradients of sea-surface temperature and the north-south and east-west components of the surface
geostrophic flow.
The present method differs from that of
Namias in that contemporary values of the sea-surface temperature
gradient were used instead of previously determined "normal"
values.
Using the empirical expression for the transport of water
450
to the right of the surface wind stress proposed by
Ekman,
v/w = 0.0127/(sin# )1/2
(V.3)
-90-
where v is the velocity of the surface water, w is the velocity
is the latitude, the advection
of the geostrophic wind, and 0
of temperature can be determined from the relation
(V.4)
v .V T
where v is the water velocity and V T is the gradient of seasurface temperature at a grid-point.
In computing the values of the east-west and north-south
and vg, monthly
components of the geostrophic wind, u
The
averaged values of the sea-level pressure were used.
components were computed at each of the 162 grid-points using
finite difference approximations to the geostrophic wind
equations in spherical coordinates.
The horizontal gradients
and (Ts)
of the sea-surface temperature (Ts)
were
computed by applying finite difference approximations to the
gradient relationships in spherical coordinates,
JTST
1
1
(Ts)
(Ts1 = Rcos
__s
jI
R1
R cos(
s
AA
(V.5)
and
(TR
1
Ts
1R
Ts
where R is the radius of the earth and A and 0
longitude and latitude in radians.
denote
The grid spacing was 50
-91-
and A
and A
were 100 or T/18 radians.
Once the components of the geostrophic wind were known,
the surface water velocity components, u
and v , were then
determined from
Us =/
0.0127
(ug + vg)
(V.6)
and
v
0.0127
=
s W2(sin )l/
2
(vg
U).
g
The surface temperature advection was then computed from
-v .
T
-
-
u s (T
(T)
.
(V.7)
Anomalies of the surface temperature advection were computed for each of the 84 months in the same manner as for the
heat transfer and temperature terms, and the results correlated
with the anomalies ofA SST on a seasonal and an 84-month
basis.
On the seasonal basis, the average values of the 162
coefficients on the map are .22 for winter, .19 for spring,
.10 for summer, and .21 for fall.
As in the case of the surface
heat transfer correlations, the largest values occur in winter,
where there are 35 values reaching the 5 per cent confidence
-92-
level.
The corresponding numbers of the other seasons are
25 for spring, 17 for summer, and 27 for fall.
The latitudinal
averages of the correlations for each season are shown in
Table VIII.
The correlations are consistently highest. at
45*N and lowest at 200 N during all of the seasons.
The correlation coefficients based on the 84-month data
series are shown in Figure 10.
Of the 162 values on the map,
74 reach the 5 per cent confidence level and 44 reach the 1
per cent level.
By and large, the best results were obtained
between latitudes 300 N and 500 N and in the central part of
the ocean away from coastal boundaries. The poor results at
the latitudes 200 N and 250 N agree with those found by Namias
(1965) and may be due to the low variability of temperature
in this region, in which observational errors could be as large
as the anomalies themselves.
The area of low correlation in
the western part of the map is probably due to the effect of
the variability in the Kuroshio Current.
This current's
large northward transport of water and its meanders could
easily overshadow the effect of temperature advection due
to surface wind-drift in this area.
Pattern correlations between each of the two sets of
anomalies were also computed for each of the 84 months in the
period.
The values range from -.18 in January of 1955 and
July of 1953 to .55 in January of 1956.
If the correlation
values are averaged over the 7-year period for each month,
-93-
Table VIII
Winter
Spring
Summer
Fall
Year
55N
.11
.12
.20
.22
.16
50O0 N
.29
.32
.12
.34
.27
45ON
.33
.40
.18
.41
.33
40O0 N
.34
.23
.16
.31
.26
350N
.39
.25
.05
.20
.22
30 0 N
.32
.26
.06
.01
.16
250 N
.09
.04
.08
.08
.07
200 N
-.12
-.07
-.04
.12
-.02
Average latitudinal correlation coefficients
between anomalies of horizontal temperature
advection and anomalies of change in seasurface temperature
L SST.
CORR.
COEF.
A A/OM
45597-
VS,
A-vo /
--
,
T
ALL MONTHS 1951-1957
55N
.09-400
50N
.38
45N
'.13 .11 .12
.35 .14 .00 .12
.29 .30
.22
.12-.0
.22 j33 .37 .33 .2.2.41 .39 .21 .21
.11 .26 .25 .22 .30 .30 .32
,34 .47 .44 .40 .43 .44 .45 .39 .3
.01 .06 .20 .21 .19 .24 .25 .24 .31 .40 .38 .40 .39 133 .35 .46 .34 .32
.31 .03 .19
30N
.10
.29
.15-.03-.02-.01
.05
.15 .13
.00 .01 .02 .10
.18
;07-.09-.06
25N
-.06 .07
20N
-. 10-.01-.)2
130
Figure 10.
.26 .30 .31 .31 .35 .25 .32 .34 .33 .37 .33
.25 .26
.1.0 .11
.27
.06 .03 .15 .07
.10 t06 .02 .09-;07 .06-.01
.21
.16
.02-.04-.15-.02-.01-.10-.17 .03 .08-.04-.17-.23-104-.01-.01
140
150
T60
170E
180
.21 .21 .28 .26 &1
.22
.15 .23 .32 .39 .28
.07 .20 .16
429
42
170W
160
150
602-.12
140
i07
414
.16
.08 .21 .18-*07-.05
130
120
Correlations between the anomalies of horizontal temperature advection and the anomalies of6 SST based on 84 months
of data from 1951 through 1957.
.06
110
-95-
September has the largest value of .32, while June has the
lowest value of .04.
There is a great deal of variability
between years in the 12 month average values of the pattern
correlations, with the results ranging from .08 in 1954 and
1955 to .28 in 1951.
About half of the average values are
significant at the 5 per cent confidence level.
Several other methods of computing horizontal temperature
advection were also tried.
The empirical formula of Witting
(1909),
v = K
where v
s
(V.8)
,v
is the velocity of the surface water, v
g
is the
geostrophic wind averaged over some time period, and K is a
constant equal to 4.8 was used to compute the u and v components of the surface drift.
However, when the results of the
sea-surface temperature advection anomalies computed from these
components were correlated with the anomalies ofASST, no
significant correlation patterns were obtained for any
season.
This method is the same one that is used by the U.S.
Fleet Numerical Weather Facility (FNWF) at Monterey, California,
to compute temperature advection due to wind drift.
They
base their calculations on daily rather than monthly averaged
data, however, and this may account for the difference between
their success in using the method
and our failure.
-96-
An attempt was also made to compute the vertically integrated horizontal heat advection in the upper mixed layer of
the ocean due to Ekman transport and to correlate the results
with the observed changes in the temperature of the layer
eSST.
In order to determine this advection,
(v
-cp
p
)
(V.9)
'T) dz
e
-D
where ve is the horizontal velocity vector, T is the temperature at depth -z, and -D is the depth of the mixed layer,
the assumption was made that T is independent of z and equal
Then (V.9) can be reduced to
to the surface temperature T s.
vedz
-cp
.Ts
(V.10)
.
Using the relationships
f
dz = Ue =~/f
and
to
j
(V.11)
,-rdz = Ve = -T/f,
-D
where U
e
and V
e
are the vertically integrated east-west and
north-south components of the Ekman transport,
rA
and l
are the east-west and north-south components of the surface
wind stress, and f is the Coriolis parameter, the vertically
-97-
integrated horizontal heat advection was then computed from
Ue(Ts)
VT s = -c p+e
-c ppfTdz
e
where (Ts)
and (Ts)
+
e (T
are the east-west and north-south com-
ponents of the sea-surface temperature gradient.
ponents of the wind stress
=
C
(V.12)
?
The com-
andS were determined from
2 + V- 21/2
(V.13)
and
'
where,
2
= Y CDCt4
2 1/2
s )
s/2
is the density of air, us and vs are the components of
the surface wind velocity, and CD is the drag coefficient and
a function of the surface wind strength.
The results of correlating the anomalies of the horizontal heat advection due to Ekman transport and the anomalies
of4 SST also showed no significant correlation patterns in
any of the seasons or over the entire 84-month period.
From the above studies on the effect of surface current
advection on changes in the sea-surface temperature, it seems
as if surface winds are capable of driving a shallow surface
layer of the ocean at 450 to the right of the wind direction
quite freely back and forth over the sea surface, at least in
the central parts of the ocean.
Although the results of this
-98-
study are based on monthly averaged data, it seems reasonable
to assume that the same type of movement can occur on a daily
basis due to the passage of storm systems.
The work of the
FNWF at Monterey supports this idea, since their temperature
analyses and forecasts are made on a daily basis, and their
results have been verified to some degree by actual observations.
3.
In an attempt to improve the correlations between
the large-scale anomaly patterns of the sea-surface temperature
and the possible physical mechanisms responsible for these
anomalies, the changes in sea-surface temperature during a
month due to heat transfer across the sea surface were added
to the changes in sea-surface temperature due to horizontal
wind-drift temperature advection.
The change in temperature of the upper mixed layer
(assumed equal to the change in the sea-surface temperature)
can be computed from the relation
Q
C
(V.14)
'H
where Q is the total heat trnasfer across the sea surface,
is the density of sea water, C
is the specific heat of
sea water, and H is the depth of the mixed layer.
The change
in temperaturedT over a period of timent can then be
-99-
written as
tT
A t
where - v
Q
CHP
v
V T ,
(V.15)
VT is the contribution due to horizontal advection.
Since the year-to-year variations in the monthly values of
the mixed layer depth H are not known over large areas of the
North Pacific, the "normal" values of H given by Pattullo
and Cochrane (1951) were used in all of the computations.
This procedure could lead to considerable errors in determining the monthly values of Q/y CpH, since the results of
data taken at weathership stations in the North Pacific
(see Chapter IV) have shomwn that large variations in H can
occur between years for the same month.
As was also mentioned
in Chapter IV, there is some doubt as to the validity of the
mixed layer depth values for some months in the eastern part
of the ocean.
(This uncertainty may be cleared up in the near
future, however, since Mrs. Margaret Robinson at the Scripps
Institution of Oceanography is presently preparing maps of
the seasonal thermocline depth for all areas of the North.
Pacific based on all available bathythermograph observations.)
Again, monthly anomalies of the computed temperature
change were calculated and the results correlated on both a
seasonal and an 84-month basis.
On the seasonal basis, the
average values of the 162 correlation coefficients on the map
-100-
are .38 for winter, .22 for spring, .28 for summer and .28
for fall.
For the winter season, 78 values reach the 5 per
cent confidence level and 44 reach the 1 per cent confidence
level.
For the other 3 seasons, the corresponding numbers
are 51 and 24 for spring, 43 and 19 for summer, and 45 and
18 for fall.
The latitudinal averages of the correlations
are shown in Table IX.
The correlation coefficients based on the 84-month data
series are shown in Figure 11.
Of the 162 values on the map,
112 reach the 5 per cent confidence level and 91 reach the
1 per cent level.
Again, the best results were obtained bet-
ween the latitudes 300 N and 500 N and in the central part of
the ocean away from the coastal boundaries.
The results near
the coasts and along the 20th and 25th parallels are better,
however, than those obtained using only the horizontal temperature advection equation, indicating that anomalies in the total
heat transfer across the sea surface play an important role
in determining sea-surface temperature fluctuations in these
areas.
The pattern correlations computed between the anomalies
of the computed change in temperature andL SST range from
-.27 in March of 1953 to .75 in December of 1953.
When the
pattern correlations are averaged over the 7-year period for
each month, December has the largest value of .41 and March
has the lowest value of .07.
As in the previous two correla-
-101-
Table IX
Winter
Spring
Summer
Fall
Year
55N
.21
-. 09
.30
.20
.16
0
50O
N
.38
.14
.32
.36
.30
45 0 N
.43
.23
.27
.41
.34
40 0 N
.44
.28
.40
.42
.38
35 0 N
.49
.41
.33
.32
.39
30 0 N
.48
.27
.22
.26
.31
25 N
.28
.31
.21
.20
.25
20 0 N
.35
.25
.18
.09
.22
Average latitudinal correlation coefficients
between anomalies of A SST and anomalies of
the sum of sea-surface temperature change
due to heat transfer through the sea surface
and that due to horizontal wind-drift advection.
CCRR. CCEF.
A
O/VM L
SST
ALL
VS .
,voI
Tr-D
MONTHS 1951-1957
.15 .130.
55N
.27 .28 .11-.01 .1
.28 .26 .12 .03-.02 .29 .29 .33 .48 .52 .49 .47 .34
45N
43N
20N
.17 .24
.31
.22 .31
.31
.11
.37 .49 .49 .38 .41 .48
.18
.42
.29 .35 .39
.28
.19
.22
.39
.41 .36
.36
.52
.49 .47
.57 .58 .47 .39
.33
.54 .54
.50
.22
.35
.31
.05
.29
.30
.21
.15
.26 .34
.48 .42
.33 .18 .24 .33
.39
.40
.36
.13
.14
.22
.24
.16
Figure 11.
140
150
160
170E
.27
.331
.49 .49
.42 .32 .27
.16
.40 .53 .51
.46 .47 .47
.28 .20 .35
.21
.46 .38 .21
.57 .51
.39
.42
.49 .41
.52 .56
.44 .49 .46
.23 .33 .26
.15
.44 .25 .29
.26
.38
.36
.24
.22
130
.21
".13 .15
180
.12
170W
.09
.31
160
.36
.36
.37
.15
.07
.06
.47 .44 .10
.06-.04-.01 .14 .13
.17
.09 .17
.14
150
140
.08 .04 .15
130
120
.27 .33
110
Correlations between the anomalies of the sum of the change in SST due to heat transfer across the sea surface and to
horizontal temperature advection and the anomalies of6 SST based on 84 months of data from 1951 through 1957.
-103-
tion studies, there is little variability between years in
the 12-month average values of the pattern correlations; the
results range from .24 in 1951 to .29 in 1955.
The results obtained by combining the effects of heat
transfer across the sea surface and those of horizontal temperature advection are somewhat better than using either of the
terms by itself.
The main problem in this method is that the
temperature differences computed from the surface heat transfer term are much larger in magnitude than those determined
from the advection term.
When the two are added together,
the former term greatly overshadows the effect of the latter
and not much information is gained.
The temperature differ-
ences or anomalies computed from the surface heat transfer
term are, in fact, larger by an order of magnitude than any of
The
those observed in the 84-month period under investigation.
reasons for this
discrepancy probably lie in using unreliable
values for the depth of the mixed layer H and in the fact that
there is a loss of heat through the bottom of the mixed layer
due to vertical eddy diffusion, which is unaccounted for in
this study.
Before more reliable values of temperature
anomalies in the mixed layer can be obtained, much will have
to be learned about the causes of fluctuations in the mixed
layer depth and about the nature of heat loss across the thermocline region.
-104-
4.
An attempt was made to see if vertical velocities
induced at the bottom of the Ekman surfac layer had any effect
upon the sea-surface temperature anomalies.
These velocities
were computed from the relation
w
e
-
f
(V - V),
e
(V.17)
where we is the vertical velocity beneath the layer, f is the
Coriolis parameter, f
is the latitudinal variation of the
Coriolis paramater, V is the vertically integrated total meridional transport, and Ve is the vertically integrated Ekman
meridional transport.
(A derivation of this equation is given
on pages 43 through 45 in the report prepared by Willett
and Clark for the National Science Foundation in 1966.)
The monthly averaged values of V and Ve were taken from
the work done by Fofonoff (1960-1963) on transport computations
for the North Pacific Ocean for the Fisheries Research Board
of Canada.
The vertical velocities were computed at each of the 162
grid-points for all 84 months, and the monthly anomalies were
obtained by subtracting the 7-year average value for each
month from the 7 individual monthly values.
These results
were then correlated with the anomalies of4 SST on a seasonal
basis.
The results of the correlation study show that anomalies
-105-
in the vertical velocity beneath the Ekman layer, alone, have
little effect on the sea-surface temperature anomalies.
The
average values of the 162 grid-point correlation coefficients
are -.06 for winter, .00 for spring, -.11 for summer and -.15
for fall.
The slight negative correlation between the two
anomaly patterns for three of the seasons indicates that an
anomalous upward transport of water from beneath the Ekman
layer tends to lower the temperature of the surface layer.
The relationship is not very strong, however, since the average correlation coefficients are very small for each of the
four seasons.
In order to test the combined effects of heat transfer
across the surface layer, horizontal temperature advection,
and vertical transport of water beneath the Ekman layer, a
multiple linear correlation analysis was computed for the
three terms,,ASST, the sum of temperature change caused by
heat transfer across the ocean surface and of temperature
change caused by horizontal advection (TTD), and the vertical
velocity induced at the bottom of the Ekman layer (VV).
The
computations were done at each of the 162 grid-points and for
each of the seasons using the relationship
2
2
R 3.21
2
r3 1
- 2r31 r32 r12
2,
1 - r12
+ r3 2
(V.18)
-106-
where R
3.21
is the percentage of the variance ofA SST
accounted for by relation to TTD and VV, r3 1 is the linear
correlation coefficient between4 SST and TTD, r3 2 is the
linear correlation coefficient betweene SST and VV, and rl2
is the linear correlation coefficient between TTD and VV.
The results of the analysis are shown in Table X.
It
is seen from the table that the percentage of the variance of
A
SST accounted for by relation to the other variables is
increased from 20 to 25 per cent in winter, 13 to 18 per cent
in spring, 12 to 18 per cent in summer, and 13 to 14 per cent
in fall by including the effects of the vertical velocities
beneath the Ekman layer.
The magnitudes of the increase are
the largest in the higher latitudes, indicating that upwelling associated with Ekman layer divergence may be an
important factor in determining the temperature of the surface
layer in this region.
As for the statistical significance of the results of
the multiple correlation analysis, 58 of the 162 values on
the map reach the 5 per cent confidence level and 36 reach the
1 per cent level for winter, 45 of the 162 values reach the
5 per cent level and 15 reach the 1 per cent level for spring,
34 of the 162 values reach the 5 per cent level and 10 reach
the 1 per cent level for summer, and 36 of the 162 values
reach the 5 per cent level and 13 reach the 1 per cent level
for fall.
By chance alone, it is expected that 8 of the 162
-107-
Table X
Winter
TTD
TTD
and
VV
Summer
Spring
TTD
TTD
and
TTD
VV
TTD
and
Fall
TTD
TTD
VV
and
VV
'll
17
7
19
25
14
20
15
22
22
30
11
16
20
24
400 N
23
27
20
25
22
27
35 0 N
28
31
17
23
15
19
30 0 N
25
30
10
14
10
12
0
25O
N
16
21
7
12
10
15
20 0 N
14
18
8
14
7
12
Average
20
25
12
18
13
19
55 0N
12
19
50O0 N
19
450 N
13
18
Percentage of variance of
A SST accounted
for by relation to TTD and to TTD plus VV.
-108-
values on the map would reach the 5 per cent level and 2 of
the 162 values would reach the 1 per cent level.
5.
In order to see how large the errors are in computing the
temperature of the mixed layer, the observed values ofASST
(6 SSTO) were subtracted from the computed values of4 SST (A SSTC)
for all grid-points and all months.
These results were averaged
over the 7-year period for each of the 12 months and 162 points;
the departures from these mean or normal values were then computed
and punched onto cards for further analysis.
From the maps of the normal values of (6 SSTC-ASSTO), it
is seen that the errors are all an order of magnitude larger than
the observed values and have the largest values in all 12 months
in the western part of the ocean.
This last fact is probably due
to having neglected the transport of heat by the Kuroshio in calculating the
SSTC values.
In fall and winter, the (ASSTC-aSSTO)
values are negative for all points, indicating that the temperature
decreases due to heat loss were over computed.
During the summer
months, the (A SSTC-L SSTO) values were positive at all points,
indicating that the temperature increases due to heat gain were also
over computed.
During the spring months, the (A SSTC-A SSTO) values
were of smaller magnitude and mixed in sign; this indicates that the
computed changes in temperature are closer to the observed values
and that over computation occurred for both heat loss and heat gain.
If the computed heat transfers and the observed tempera-
-109-
ture changes are the correct order of magnitude, then there
must be heat advection occurring in both horizontal and vertical directions.
In his 1965 paper, Wyrtki, using the same
data as in this report, calculated that the heat gain of
3 x 1014 cal/sec of the surface layer of the North Pacific
Ocean must be compensated by deep water ascending with an
average velocity of 1.25 x 10- 5 cm/sec.
This upward transport
of cold water and an outflow of warm deep water of 8 x 106
m /sec, partly through the Indonesian waters and partly in
the eastern tropical Pacific from the counter current to the
south of the equational current, maintains the thermal structure of the ocean in the presence of a net heat gain.
In order to maintain the temperature distribution of
the ocean in the presence of horizontal inequalities of
heat exchange, there must be a horizontal advection of heat
in the surface layer of about 10 x 10
6
3
m /sec.
Wyrtki
comments that this is a weak circulation in comparison to the
Kuroshio transport of 65 x 10
6
3
m /sec and is, therefore,
relatively sensitive to transient changes in the strength
of the heat sources and sinks.
From Wyrtki's study and the
present one, it has been shown that pronounced changes in
these sources and sinks do occur on a seasonal and monthly
basis and could help to account for the anomalies of seasurface temperature by influencing the horizontal surface
currents.
-110-
In spite of the deficiencies of the methods described
above, it appears that heat transfer across the sea surface,
horizontal temperature advection due to wind drift, and vertical velocities induced beneath the Ekman layer profoundly
affect the temperature of the ocean's surface, at least on a
monthly and seasonal basis.
In order to improve our knowledge
of sea-surface temperature fluctuations, more and better observations of oceanographic and meteorological parameters are
needed, especially at shorter time scales; we need to develop
better equations to express the transfer of heat between ocean
and atmosphere and also between the surface layer of the ocean
and its lower regions; more information is needed concerning
the variations of the surface wind stress at all time scales
and the response of ocean surface currents to these variations;
and, finally, a theory is needed that treats the ocean and
atmosphere as two interacting media, from which the transfer of
various properties across their common boundary can be computed.
6.
In section A of this chapter, we found that there was
a fair degree of persistence in the sea-surface temperature
anomaly patterns at 1, 2 and 3 months lag.
We did not,
however, find any significant persistence in the anomalies
of the total heat transfer across the sea surface or in the
anomalies of the horizontal temperature advection.
This per-
sistence in the sea-surface temperature anomalies must then be
-111-
related in some way to the dynamics of forced and free convection that occur in the upper mixed layer of the ocean.
Following the reasoning of Kraus and Rooth (1961), we
consider that the sea is cooled at the surface by evaporation,
conduction and infra-red radiation.
However, heating by the
absorption of solar radiation extends to an appreciable depth,
and, therefore, there must be a layer in which heat is transported upwards.
This turbulent upward flux of heat is associat-
ed with free convection and tends to produce a layer of nearly
constant potential temperature.
In addition to this free
convection, the surface layer is stirred by the action of the
wind, which also tends to produce a layer of constant potential
temperature through forced convection.
Thus, changes in the
heat transfer across the ocean surface and the accompanying
changes in the surface wind stress should have a profound
influence on the temperature and depth of the ocean's upper
mixed layer.
Using data obtained from weathership station Echo (350 N480 W) for July, August and September of 1958 and 1959, Kraus
and Rooth computed 10-day means of the heat loss B across the
surface layer and the heat added to the surface layer through
solar radiation S.
The difference between S and B is the heat
balance of the quasi-isothermal of mixed surface layer.
Com-
paring the values of (S-B) with the values of the mixed layer
depth, they found that there is a direct relation between the
-112-
two quantities.
In other words, when (S-B) is small, indicat-
ing large cooling of the surface, the depth of the mixed layer
increases; when (S-B) is large, indicating small cooling or
heating of the surface, the depth of the mixed layer decreases.
In periods of large cooling, the increased effects of free and
forced convection combine to increase mixing in the surface
layer and to deepen the effects of this mixing.
In periods of
small cooling, when the winds are usually reduced in magnitude,
free and forced convection are weakened, and their effects are
confinced to a shallower surface layer.
This relation between the heat transfer balance at the
ocean surface and the depth of the mixed layer may offer a
clue to the persistence of sea-surface temperature anomalies.
An increase in the cooling rate of the ocean surface, due
primarily to increases in sensible and latent heat transfer,
causes an increase in the intensity of mixing in the surface
layer and also in the depth of the mixed layer.
Even though
the temperature of the mixed layer does decrease due to the
increased loss of heat, this decrease is not as large as it
would be if there were no change in the layer depth.
The
increased mixing and the associated increase in layer depth
allow the heat loss to be tapped from a greater volume of
water, thus reducing the change in temperature of the surface layer.
This process would tend to keep both positive
and negative temperature anomalies more stable than would be
-113-
expected if no fluctations in mixed layer depth occurred,
On the other hand, a decrease in the cooling rate or an
increase in the heating of the ocean surface causes a decrease
in the intensity of mixing and in the depth of the mixed layer.
The corresponding change in temperature of the mixed layer is
therefore larger than it would be if there were no change in
the layer depth, since a smaller volume of water is affected
by the heat transfer.
This process would tend to stabilize
negative temperature anomalies and unstabilize positive anomalies when (S-B) is negative,
When (S-B) is positive, it would
tend to stabilize positive temperature anomalies and unstabilize
negative anomalies.
From the remarks made in the above two paragraphs, it is
concluded that fluctuation in the mixed layer depth may play
an important part in the persistence of sea-surface temperature
anomalies on a month-to-month basis.
These fluctuations may
also help to explain the pronounced changes in anomaly patterns
that were observed at times during the 84-month period, since
some changes in the total heat transfer across the sea surface
can render the sea-surface temperature anomaly patterns more
unstable than would be expected if no fluctuations in the mixed
layer depth occurred.
The effects of horizontal heat advection due to wind drift
may also be an important factor in the persistence of se-surface temperature anomalies.
However, the auto-correlations of
-114-
the temperature advection anomalies for the 84 month period show
no persistence at 1, 2 and 3 months lag (the average values of
the 162 grid points on the map were negative for all 3 months).
-115-
VI.
A.
Additional Results
Comparison of observed sea-surface temperature and total
surface heat transfer cycles with theoretical results.
After this thesis had been completed, two papers were
found in the January, 1967, issue of Tellus by Kraus and
Turner in which a general theory of the seasonal thermocline
is presented.
The results of the theory give the depth of
the mixed layer or seasonal thermocline and the temperature
of the layer as a function of time and quite general external
energy inputs.
The backbone of the theory is two equations,
a thermal energy balance equation for the mixed layer and a
mechanical energy balance equation for the layer, which are
as follows:
dT
h-
dt
+
(T-
s
T )
hdt
where h is the depth of the layer,
S + B - S -B'
e
S + B
(VI.l)
Ts is the temperature of
the layer, Th is the temperature below the layer, S is the
penetrating component of the solar radiation,
8
is the
extinction coefficient, B is the heat exchange between ocean
and atmosphere due to back radiation and latent and sensible
heat transfer, andA is the Heaviside unit function, defined
as
-116-
S
dh)
1 )=for-
dhO7 0
and
dT
1 dTss h 2 +A
2dt
(T
Th) dh
dh
dt
-
G - D+
S
(VI.2)
where the first term on the left-hand side represents the
potential energy change associated with the change in temperature of the layer, the second term on the left-hand side
represents the potential energy change due to the entrainment
of water as the depth of the thermocline increases (dh/dt :
0),
and the terms on the right-hand side represent the energy
change due to mechanical stirring by the wind, dissipation of
energy within the layer, and convection due to internal
heating.
Equations (VI.1) and (VI.2) can be transformed into
dT
2
(S + B)h -
(G - D +
)
(VI.3)
n
and
(G
Sdh
S
LI
L
.
-
D +
- (S + B)h
(VI.4)
-117-
When the thermocline is rising,.A = 0 (dh/st
O) and
2 G - D + S/
S+B
h
(V5)
(VI.5)
Substituting (VI.5) into (VI.3), we get
dT
s
2 G-
D + S/
1 (S + B)
G.
h2
dt-
2 G -
2
D + S/
"
(VI.6)
Differentiating (VI.5) with the assumption that seasonal
variations of the solar radiation S are larger than the
corresponding changes of G and D, we get
dh
dh( ,
dt
1
(S + B)
22 - h) dS
p
.
(VI.7)
dt
Since h)2/p , we find that h and -S are in phase and
that the layer depth h has a minimum at the time of the summer
solstice when dS/dt = 0.
Equation (VI.6) shows that the tempera-
ture Ts is still increasing when dS/dt = 0 and that, in fact,
the warming is most rapid (dTS/dt has maximum values) when
h is close to its minimum.
The temperature Ts reaches its
maximum value later in the season.
When the thermocline is falling, the depth h is large,
and the product (S + B)h will always exceed the stirring terms
in equations (VI.3) and (VI.4).
Since (S + B) is negative
at this time, this product will account for most of the rates
-118-
of increase of layer depth and decrease in temperature.
Using a symmetrical saw tooth heating function for (S + B)
and letting D = 0 andP =oo
variation of T
S
, Kraus and Turner calculated the
and h with respect to time throughout one
complete heating and cooling cycle.
The results obtained look
just like the curves for the heating and sea-surface temperature cycles presented in Figures 1 through 6 of this thesis
(see Chapter IV).
In fact, the observed results, (1) that
the heating cycle reaches a maximum in June, (2) that the
maximum change in temperature occurs between May and June,
(3) that the sea-surface temperature reaches a maximum somewhat before the end of the heating cycle (August vs. September),
and (4) that the minimum of sea-surface temperature occurs at
the end of the cooling cycle, agree almost exactly with the
theoretical results.
We thus conclude that during the heating
part of the cycle, the depth of the mixed layer and its temperature are controlled by a combination of wind stirring and
penetrative radiation; during the cooling part of the cycle,
they are controlled primarily by surface cooling.
In other
words, during the heating cycle both terms inside the square
brackets of equations (VI.3) and (VI.4) are important; during
the cooling cycle, the first term in the square brackets of
equation (VI.3) and the second term of equation (VI.4) are
predominant.
-119-
B.
Relationship between anomalies of the sea-surface temperature
and anomalies of the specific humidity of the air.
Since the values of the specific humidity of the air qa
had been computed for each of the 84 months in order to obtain
the values of the latent heat transfer, the anomalies of q
were computed and the results correlated on a contemporary
basis with the anomalies of SST.
The results of the correlation
analysis on a seasonal basis are shown in Table XI.
It is seen from the table that a strong relationship exists
between the two anomaly series; the average values of the 162
coefficients on the map are .36 for winter, .46 for spring,
.61 for summer and .58 for fall.
Of the 162 coefficients on
each of the maps, 86 reach the 5 per cent confidence level and
60 reach the 1 per cent confidence level for winter, 112 reach
the 5 per cent level and 71 reach the 1 per cent level for
spring, 122 reach the 5 per cent level and 102 reach the 1
per cent level for summer, and 130 reach the 5 per cent level
and 107 reach the 1 per cent level for fall.
In all of the
seasons, these numbers are significantly larger that those
expected by chance alone (8 for the 5 per cent level and 2 for
the 1 per cent level).
Pattern correlations also were computed between each of
the anomaly series for all of the 84 months.
The monthly
values of the coefficients averaged over the 7 years range from
.37 in February and March to .60 in june.
If the pattern
-120-
Table XI
Winter
Spring
Summer
Fall
Year
55 0N
-.16
.24
.54
.52
.28
50 0N
.35
.41
.77
.58
.53
45 0 N
.59
.45
.79
.64
.62
40 0 N
.39
.51
.73
.68
.58
0
35O
N
.36
.52
.72
.63
.56
30 0 N
.38
.59
.59
.65
.55
25 0 N
.48
.54
.35
.54
.48
20 0 N
.47
.45
.40
.38
.42
Average latitudinal correlation coefficients
between the anomalies of SST and those of
the specific humidity of the air.
-121-
correlations are averaged over the 12 months of each year,
these average values range from .38 in 1951 to .62 in 1957.
The above results suggest that there is a direct relationship between the temperature of the sea surface and the amount
of moisture in the air directly above it, especially during
the summer and fall months.
This result is not surprising
since the vapor pressure of the sea surface also increases
as the temperature increases and this, in turn, causes an
increase in the amount of evaporation that occurs between
the ocean and atmosphere.
The fact that the correlations
are highest during the summer months probably reflects the
fact that, during the winter months, the water vapor is carried
away by relatively strong winds.
During the summer and fall
months, the winds are weaker (see Chapter IV, Section A.3)
and the moisture is allowed to build up over the area of the
sea surface from which it evaporated.
C.
Relationship between the anomalies of the latent heat flux
computed from the transfer equations and the anomalies of
atmospheric water vapor divergence.
Following the derivations given by Barnes (1965), for a
column of air one square centimeter in cross-sectional area
and extending from the surface of the earth to the top of the
atmosphere, we have
-122-
dw
+
- +
dt
where W = g
Q = -
' AQ = E - P ,
(VI.8)
q dp
a
= total water vapor in column,
q
= integrated vector transport of
V dp
water vapor,
qa
=
p
= pressure
specific humidity,
po = surface pressure,
E
= evaporation,
P
= precipitation, and
g
= acceleration due to gravity.
Exapnding equation (VI.8) in geographical coordinates
(A ,
, p,
JW
-W +
d t
t),
we get
o
1
a cosd
8j
++ 1 dQ
a d
_
Oi:tan$
a
(VI.9)
where A and / are longitude and latitude, respectively, and
a is the radius of the earth.
If d W/dt is small compared to the other terms (total
water content of the column remains nearly constant during a
period of timed t),
and since
-123-
S--
whereW
=
p/
qa
a
p
p
t,
d
= 0,
(VI.iY
p
then
=
a Q
cosA
JA
+
cos
)
E - P .
(VI.11)
Equation (VI.//) states that the divergence of the vertically
integrated vector transport of water vapor is equal to the
difference between the amount of evaporation and precipitation
occurring at the earth's surface.
Since the values of the latent heat flux or evaporation
(Qe) computed from the transfer equations developed in Chapter
II were available, it was decided to compute the values of
V • Q for the 84 months of the period from 1951 through 1957 and
compare the anomaly patterns of the two quantities.
However,
an extremely crude method had to be used in computing the
values ofV
Q , since the data were available on only a
monthly average basis and only for the surface layer of the
atmosphere.
Other investigators have found that daily and
hourly values have a large effect on the monthly average
values ofV' Q.
Using the u and v components of the surface geostrophic
wind found previously and the monthly averaged values of
qa, the values of V- Q were computed at each of the 162 gridpoints by the following method: we let
-124-
Q = qug
u
and Q=
qa v
,
then
1
.4
V
Q
where.A andA/
aa cos
cos
] QA
-
-
+
a (Q Cos )
r a
,
(VI.12)
are 100 or7T/18 radians.
The average vlaues of the 162 correlation coefficients on
the map between the anomalies of Qe and those of V7
Q are
.36 for winter, .25 for spring, .10 for summer and .22 for
fall.
For winter 78 of the 162 coefficients reach the 5
per cent confidence level and 58 reach the 1 per cent level;
the corresponding numbers for the other three seasons are
56 and 27 for spring, 22 and 6 for summer, and 37 and 20 for
fall,
The latitudinal averages of the correlations are given
in Table XII.
From the table it is seen that the largest correlations
occur during the winter months and between the latitudes
25"N and 509N.
The relatively high correlations during the
winter and fall months are probably due to the fact that the
transfer formula used to determine the latent heat flux most
accurately represents the actual heat flux when the magnitudes
Then the heat flux is small, as it
of this quantity ar large .
is during the spring and summer months, the conditions under
-125-
Table XII
Winter
Spring
Summer
Fall
Year
55 0 N
.15
-.02
-.03
.11
.05
50 0 N
.46
.23
.20
.26
.29
45 0 N
.57
.26
.22
.31
.34
40 0 N
.50
.38
.28
.39
.39
35 0N
.50
.40
.23
.31
.36
30 0N
.35
.37
-. 02
.19
.22
25 0 N
.24
.29
-. 02
.14
.16
20 0 N
.07
.07
-. 04
.02
.03
Latitudinal averages of the correlation
coefficients of the anomalies of Q
those of
V * Q.
e
and
-126-
which the transfer formula was developed are not fulfilled,
and the values computed become less reliable.
The poor results obtained at the latitudes 200 N and 550 N
in all of the seasons probably reflect the fact that there
were no values of Q
given at the latitudes 15oN and 600N,
and a linear approximation had to be introduced in order to
get these quantities.
(The values of6 Q/,/in the divergence
equation at 200 N and 550 N were computed by taking the difference
between the values of Q
at 25*N and 150 N and between the
at 600 N and 500 n and dividing the results by
values of Q ;
7/18 radians.)
-S,
Another source of error is the fact thatv "
Q is the
difference between the amount of evaporation and the amount
of precipitation that occur in a given area.
the anomalies of Qe with those of V*
In correlating
Q , this fact was not
taken into account, since the values of the precipitation
were not available.
In spite of the deficiencies of the method in determining
S
Q , the results of the correlation study between the
anomalies of 0
andV"
Q show that there is a statistically
significant relation between the two series, i.e. when there
is a stronger or weaker than average value of the latent
heat flux between ocean and atmosphere, there is also a
stronger or weaker than average value of the atmospheric
water vapor divergence field.
It is felt that this type of
-127-
analysis would be a valuable tool in helping to develop a
better transfer formula for the latent h#eat flux between
ocean and atmosphere, especially if it were applied to a
smaller area over which daily and hourly values of the
necessary meteorological and oceanographic paramters could
be obtained.
-128-
VII.
A.
Conclusions
Summary of Results
1.
Evaluation of the Transfer Formulas
a.
In Chapter II, the effect of observational errors
and the effect of using monthly averaged data rather
than daily or hourly data on the accuracy of the
exchange formulas for determining Qi, Qb' Qe and
Qs were investigated.
It was found that the accuracy
of these formulas increases as the values of the
heat transfers increase, since the errors resulting
from both observational and averaging effects decrease
with increased heat transfer.
The theoretical accuracy
of the formulas for determining Qe and Qs is also
higher when large transfers occur; since large transfers occur at times of strong winds, the wind shear
keeps the atmospheric stratification close to neutral
values, and these are the conditions for which the
formulas were derived,
b.
A comparison was made of monthly average heat
transfers determined by using the monthly averages of
daily heat transfers and by using monthly average
variables in the exchange formulas.
It was found
-129-
that the errors involved in using the latter method
are, on the average, about 10 per cent of the actual
values due to correlations on a daily basis between
the wind speed and air-sea property differences.
2.
Heat Exchange Calculations
The study concerning the 7-year means of-the monthly
averaged heat transfers produced the following results:
a.
The values of
Qi, Qb' Qe'
s and Q show large
seasonal and year-to-year variability, especially
those of
b.
Qe Qs and Q.
The 84-month averages of the total heat transfer
across the sea surface (Q) were integrated over the
North Pacific from 200 N to 55 0 n.
that the ocean loses
The result shows
surface north of 200 N.
7 x 1014 cal/sec through the
Wyrtki (1965) calculated
the total integral of heat transfer over the North
Pacific from the equator poleward and found that the
ocean gains 3 x 1014 cal/sec.
The difference between
the two calculations shows that there is a heat gain
of 10 x 1014 cal/sec between the equator and 200 N
and that it is this region that accounts for the
yearly heat gain of the North Pacific.
-130-
c.
Fourier analysis of the 12 mean monthly values of
each of the heat transfer terms showed that there are
regular yearly cycles in each of the terms over most
of the North Pacific Ocean from 20'N to 550 N.
The
maximum values of Qi occur in June, those of Qb in
January, those of Qe in December, and those of Qs in
November, and December.
There is also a regular
yearly cycle in Q, with cooling of the ocean surface
layer occurring from September through April and warming of the surface layer occurring from April through
September.
d.
Fourier analysis of the 12 mean monthly values of
the sea-surface temperature (SST) showed that there is
also a regular yearly cycle in this term, with the
maximum values occurring in August and September
and the minimum values occurring in February and March.
e.
By comparing the yearly cycles of SST and Q,
it was found that the maximum values of SST occur
near the end of the heating cycle (positive Q) and
that the minimum values occur at the end of the cooling
cycle (negative Q).
f.
There is a regular yearly cycle in the surface
-131-
air temperature (SAT), and the maximum and minimum
values occur at the same times as those of SST.
The
maximum values of SST and SAT both occur earlier in
the western part of the North Pacific than in the eastern part; this difference may be due to the effect of
the cold advection towards the equator by the California current.
g,
Fourier analysis of the change in SST from month
to month (A6SST), showed that there is a regular
yearly cycle in this quantity.
The maximum increase
of SST occurs from the middle of May to the middle of
June and the maximum decrease of SST occurs from the
middle of December to the middle of January.
During
March and September, when the ocean surface layer
temperature does not change very much, there must be
a balance between the heat transfer across the sea
surface and the advection that occurs within the
surface layer.
h.
Fourier analysis of the mean monthly values of
the observed wind speeds showed that a regular
yearly cycle is present, with the maximum values
occurring in January and the minimum values occurring
in June.
The same type of analysis showed that there
-132-
is also a regular yearly cycle in the Bowen ratio
(Qs/Qe) in only the middle latttudes and only then in
the western part of the ocean, where the annual range
of values is from .4 to 1.4;,in the other areas the
Bowen ratio has values of around .1 with very little
annual variation,
i.
Using the values of Q and A SST, the mean monthly
values of the total heat advection within the mixed
layer were computed.
The results obtained agree with
what is known about the current patterns of the North
Pacific during the winter and spring months.
During
the summer and fall months, the values show that there
is cold advection of water in the region of the Kuroshio current, in contrast to what is expected.
This
discrepancy may be due to the fact that the heat lost
from the surface layer across the thermocline region
is neglected.
j.
The integrated values of the heat advection from
17.5N to 57.5 0 N show that warm advection occurs over
the ocean from January through May and that cold
advection occurs in the remaining seven months.
Compared to a study done by Bryan (1960), the cold
0
advection of about 2.2 x 1014 cal/sec at 32 N.
-133r
agrees with the direction of heat transport calculated
by him using other methods.
k.
The monthly mean values of Qi, Qb' Qe' Qs and
Q were determined from daily data taken at stations
0
NOVEMBER (300 N - 1400 W) and PAPA(50°N - 145 W)
in 1962.
The yearly cycles of these quantities were
found to show the same type of variation as the results
using the monthly average data.
The large daily fluctua-
tions in Q are not accompanied by large fluctuations
in SST, which indicates that large fluctuations in
the depth of the mixed layer may occur.
Very little
correlation was found between the daily fluctuations
of Q and those of SST on a contemporary or lag basis.
3.
Discussion of Sea-Surface Temperature and Heat
Transfer Anomalies
The results of studying the anomaly patterns of the
heat transfer terms and of SST and SAT are as follows:
a.
The anomaly patterns of both SST and SAT can occur
in fairly extensive geographical areas.
The magnitudes
0
0
of the SST anomalies are about 2 F to 4 F, while
those of the SAT anomalies are slightly larger.
-134-
b.
Although some month-to-month persistence is found
in the SST anomalies, patterns can be eliminated or
reversed in a month's time.
The average of all auto-
correlations of the SST anomalies is down to .28 at
3 months lag; for the SAT anomalies, the autocorrelations
are down to .25 at 3 months lag.
c.
There is a high degree of correlation between
the anomalies of SST and SAT: the average maximum
correlation (.83) occurs in the summer months and
at the 35th parallel.
d.
It is concluded from the above results
that
there is very little long-term persistence of both air
and sea temperatures (negative autocorrelations
occur at 1 and 2 years lag) and that the anomaly patterns of both air and sea temperatures are strongly
related.
To maintain a long-term persistent anomaly
of the general atmospheric circulation, a control
factor is required that is stable over long time
periods so that it can pull the general circulation
back into line when it goes off on a tangent.
does not seem that sea-surface temperature
can provide this control factor.
It
anomalies
-135-
e.
There is very little persistence in
patterns of the heating terms Qi
Q.
Qb
,
the anomaly
Qe, Qs and
(Autocorrelations are negative for all terms at
1, 2 and 3 months lag.)
The anomaly patterns can be
large geographically, especially those of Qe and Qs"
f.
The magnitudes of the Q anomalies vary widely
over the map for a particular month and from month
to month during the 84-month period under investigation.
Their values range from less than 1 per cent
of the mean monthly average values to over 200 per
cent in the summer months.
4.
Relationships Between Sea-Surface Temperature Fluctua-
tions and Heat Transfer Anomalies
The results obtained from studying the relationships
between the anomaly patterns of SST and those of the
heat transfer terms are as follows:
a,
By correlating the anomalies of the total heat
transfer across the sea surface (Q) with the anomalies
of the change in SST from month to month (4 SST), it
was found that the two series have a significant
positive correlation for all seasons, the largest of
which occurs in the winter months.
This fact indicates
-136-
that sea-surface temperature anomalies are related to
anomalies of the total heat transfer across the sea
surface.
b.
The anomaly series of the horizontal temperature
advection due to wind drift currents are also significantly positively correlated to the anomaly series of
A SST in all seasons.
The largest average seasonal
correlation coefficient occurs in winter; the coefficients for all seasons are lower than the respective
values for the surface heat transfer anomaly series.
c.
From the study on the effect of surface current
advection on changes in the sea-surface temperature,
it seems as if surface winds are capable of driving
a shallow surface layer of the ocean quite freely
back and forth over the sea surface in a direction of
around 450 to the right of the wind stress.
d.
By adding the changes in SST due to heat transfer
across the sea surface to the changes in SST due to
horizontal wind-drift advection and correlating the
anomalies of the results with the anomalies ofb SST,
the average seasonal correlations were imporved somewhat over those obtained by using either of the two
I,
-137-
anomaly series alone.
The main difficulty in this
method is that the temperature changes computed from
the surface heat transfer term are much larger in
magnitude than those determined from the advection term.
In fact, the temperature changes computed from the surface heat transfer term are larger by an order of magnitude than any of those changes observed in the 84month period under investigation.
The reasons for
this discrepancy probably lie in using unreliable
values for the depth of the mixed layer and in the
fact that there is a loss of heat through the bottom of
the mixed layer due to vertical eddy diffusion, which
is unaccounted for in this study.
e.
An attempt was made to see if vertical velocities
induced at the bottom of the Ekman surface layer had
any effect on the SST anomalies.
The results of the
correlation study showed that these velocities, alone,
have little effect on the SST anomalies.
There is a
small negative correlation between the two anomaly
series for all seasons except winter, indicating that
an anomalous upward transport tends to lower the
temperature of the surface layer.
f.
In order to test the combined effects of heat
t
-138-
transfer across the surface layer (Q), horizontal
temperature advection (HTA), and vertical transport
of water beneath the Ekman layer (VV), a multiple
linear correlation analysis was computed for the three
terms,, SST, the sum of the temperature change caused
by Q and HTA, and VV.
The analysis showed that, by
including the effects of VV, the average percentage of
the variance inA SST accounted for by relation to
other variables is increased from 20 to 25 per cent
in winter, 13 to 18 per cent in spring, 12 to 18 per
cent in summer and 13 to 19 per cent in fall.
g.
The differences between the computed and observed
changes in SST are all an order of magnitude larger
than any of the observed values and have the largest
values in the western part of the ocean in all 12
months of the year.
The computed values of A SST are
closer to the observed values ofA SST during the spring
months and farthest from during the summer months.
h.
The small month-to-month persistence of the SST
anomalies must, in some manner, be related to the
dynamics of forced and free convection in the mixed
layer.
Fluctuations in the strength of mixing within
the layer and the associated fluctuations in the depth
-139-
of the layer can stabilize some anomaly patterns and
unstabilize others.
There is very little persistence
in the anomaly patterns of the surface heat transfer
and in those of the horizontal heat advection, and
it is concluded that these have little effect on the
persistence of the SST anomalies.
5.
Additional Results
a,
It was found that the observed results concerning
the cycles of the total surface heat transfer and of
the sea-surface temperature [(1) that the heating
cycle reaches a maximum in June, (2) that the maximum change in temperature occurs between May and June,
(3) that the sea-surface temperature reaches a maximum somewhat before the end of the heating cycle
(August vs. September), and (4) that the minimum of
sea-surface temperature occurs at the end of the cooling cycle] agree almost exactly with the theoretical
results of a one-dimensional model of the seasonal
thermocline.
b.
It was found that there is a strong relation
between the anomaly patterns of SST and those of the
specific humidity of the air over the ocean surface;
-140-
this relation is particularly strong during the
summer and fall months.
From this, it is concluded
that the amount of moisture in the air increases with
the temperature of the sea surface below it and
decreases as the ocean surface temperature decreases.
c.
A statistically significant correlation was found
between the anomalies of the latent heat flux from
ocean to atmosphere and those of the atmospheric
water vapor divergence in the surface layer of the
atmosphere.
When there is a stronger or weaker than
average value of the latent heat flux, there is also
a stronger or weaker than average value of the atmospheric water vapor divergence field.
It is felt that
this type of analysis would be a valuable tool in
helping to develop a better transfer formula for the
latent heat flux between ocean and atmosphere.
B.
Suggestions for Further Study
The suggestions for further study can be divided into the
following three categories:
1.
Use of available data and theory.
It is felt that
further information about the interaction of ocean and
atmosphere could be obtained by using the data and transfer formulas that are presently available.
Studies of
-141-
the type presented in this paper could give a better insight into the heat exchange processes and their effects
if applied to other areas of the world oceans, especially
in the tropics.
(At present, it is planned to continue
this study in the tropical areas of the North and South
Pacific Oceans during the next year.)
The studies could
concentrate their efforts on large or small-scale processes,
since the data to do this are available.
Probably the
hardest part of any endeavor of this type would be to
organize the data into a workable form, since their present form,as it stands, is almost useless for any statistical analysis.
2,
Develop new transfer formulas to be used with old or
new data.
In order to develop new transfer formulas,
experimental work will have to be done on the nature of
turbulent transfer between the ocean and atmosphere.
Since most experimental work is done on small-scale processes, it is felt that new data on oceanic and atmospheric
parameters should be obtained in order to test the new
theories.
A good example of this type of analysis is
the work done by Garstang (1965) concerning the effects
of small-scale sensible heat transfer on synoptic scale
weather systems.
He found that diurnal variations in
the sensible heat flux influence the energy flux of synop-
-142-
tic scale systems; integrated over the entire disturbance,
the energy flux is found to be double the undisturbed
values.
The inclusion of this synoptic scale energy
flux profoundly affects the large scale energy flux patterns,
and casts some doubt on the validity of the present
relative magnitudes of the various terms in the heat
budget of the ocean and atmosphere.
This type of result
suggests that new studies should be made with new formulas
and new data in order to eliminate these uncertainties.
3,
Develop a theory that treats the ocean and atmosphere
as two interacting media.
At present, there is a need to
develop a theory that treats the ocean and atmosphere
as two interacting media, from which the transfer of
various properties across their common boundary can be
computed.
A good example of this type of analysis is
the work of Kagan and Utina (1963) in which they develop
equations for describing the ocean and atmosphere and
solve them numerically by using the same boundary conditions for both systems at their interface.
In this
way, they are able to calculate the relative parameters of
ocean and atmosphere when different driving conditions
are applied,
Admittedly, this is a difficult problem,
but it is one which must be solved before long-range predictions can be made for both atmospheric and oceanic
properties.
Acknowledgements
The author would like to thank Professor Hurd C. Willett
for his encouragement and advice, without which this work
would not have been possible.
He would also like to thank Dr. 0. E. Sette and Dr.
Glenn A. Flittner of the Bureau of Commercial Fisheries for
making available the data that was used in this thesis.
The
Bureau of Commercial Fisheries also provided financial
support in the form of a fellowship grant for the 1965-66
and 1966-67 academic years.
The staff of Dr. Willett's research group, Helen Aronovitz, Wanda Burak, Madeleine Heyman, Marilyn Keegan, and
Dena Varelas did an excellent job of putting the original
data in a form useful for statistical analysis.
Miss Isabelle Kole did the drafting of the many figures
in this report and Miss Betty Goldman and Mrs. Jane McNabb
typed the final copy.
All computations were done on an IBM 7094 computer at
the M.I.T. Computation Center.
Finally, the author would like to thank his wife for
her constant support and encouragement throughout his entire
graduate career and also for typing and editing the original
manuscript.
-144-
BIBLIOGRAPHY
Arthur, R. S., 1966, Estimation of Mean Monthly Anomalies of
Sea-Surface Temperature, J. Geophys. Res., 71(10), 26892690.
Barnes, A. A., 1965, Atmospheric Water Vapor Divergence:
Measurements and Applications, AFCRL Special Reports,
No. 28.
Berliand, M. E. and Berliand, T. G., 1952, Determination of
effective radiation of the earth as influenced by cloud
cover, Izvestia of the Academy of Sciences USSR, Series
Geophizicheskaia, No. 1.
Berliand, T. G., 1960, Climatological method of total radiation,
Meteorologiya i Gidrologia, No. 6, 9-12.
Berson, F. A., 1962, On the influence of variable large-scale
wind systems on the heat balance in the active layer of
the ocean, Tech. Mem. 25, 75 pp., National Meteorological
Center, U. S. Weather Bureau.
Bjerknes, J., 1962, Synoptic survey of the interaction of sea
and atmosphere in the North Atlantic, Geofysike Publikasjoner, Geophysica Norvegica, 24(3), 115-145.
Bryan, K., 1962, Measurements of meridional heat transport
by ocean currents, J. Geophys. Res., 67(9), 3403-3414.
Budyko, M. I., 1956, The heat balance of the earth's surface,
U. S. Dept. of Commerce Translation by N. Stepanova
1958, 259 pp.
Budyko, M. I. and Kondratiev, K. Y., 1964, The heat balance of
the earth, Research in Geophysics, Vol. 2, M. I. T. Press,
529-554.
Deacon, E. L. and Webb, E. K., 1962, Small-scale interactions,
Chapter 3 of Vol. 1 of The Sea, Interscience Publishers,
New York and London.
Fofonoff, N. P., 1960-1963, Transport computations for the
North Pacific Ocean 1951-1960, Manuscript Report Series,
Fisheries Res. Board of Canada, Pacific Oceanographic
Group, Nanaimo, B. C.
-145-
Garstang, M., 1965, Distribution and mechanism of energy exchange between the tropical ocean and atmosphere, Report
to U. S. Army Electronics Research and Development Laboratory, Grant No. DA-AMC-28-043-64-G5.
Helland-Hansen, B. and Nansen, F., 1917, Videnskapsselskapeta
Skrifter. I. Mat.-Naturn. Klasse 1916, No. 9, Kristiania
1917.
Jacobs, W. C., 1951, The energy exchange between sea and
atmosphere and some of its consequences, Bulletin of the
Scripps Institution of Oceanography of the University of
California, 6(2), 27-122.
Johnson, J. H., Flittner, G. A., and Cline, M. W., 1965,
Automatic data processing program for marine synoptic
radio weather reports, U. S. Fish and Wildlife Service
Special Scientific Report No. 503.
Kagan, B. A. and Utina, Z. M., 1963, On the thermodynamic
interaction between sea and atmosphere, Okeanologiya,
3(2), 250-259.
Kraus, E, B. and Morrison, R. E., 1966, Local interactions
between the sea and the air at monthly and annual time
scales, Quart. J. R. Met. Soc., 92(391), 114-127.
Kraus, E. B. and Rooth, C., 1961, Temperature and steady
state vertical heat flux in the ocean surface layers,
Tellus, 13(2), 231-237.
Kraus, E. B. and Turner, J. S., 1967, A one-dimensional model
of the seasonal thermocline, Part II. The general theory
and its consequents, Tellus, 19(1), 98-105.
Laevastu, T., 1960, Factors affecting the temperature of the
surface layer of the sea, Soc. Sci. Fennica Comm. Phys.
Math.
Laevastu, T., 1966, Short-period changes and anomalies of
temperature in the ocean and their effects on sound
propagation, FNWF Tech. Mem. No. 18.
Namias, J.,
1959, Recent seasonal interactions between North
Pacific waters and the overlying atmospheric circulation,
J. Geophys. Res., 64, 631-646.
-146-
Namias, J., 1963, Large-scale air-sea interactions over the
North Pacific from summer 1962 through the subsequent
winter, J. Geophys. Res., 68(22), 6171-6186.
Namias, J., 1965, Macroscopic association between mean monthly
sea-surface temperature and the overlying winds, J.
Geophys. Res., 70(10), 2307-2318.
Neuman, G., Fisher, E. and Pandolfo, J., 1958, Studies on the
interaction between ocean and atmosphere with application
to long range weather forecasting, Final Report Under
U. S. Air Force Contract No. AF 19(604)-1284.
Pacific Oceanographic Group, Oceanographic Atlas of Ocean
Weather Station "PAPA" 1956-1963, 1965, Manuscript Report
Series No. 187, Fisheries Res. Board of Cancda.
Pattullo, J. G. and Cochrane, J. D., 1951, Monthly thermal
condition charts for the North Pacific Ocean, MS Report
No. 3, Bathythermograph Section, Scripps, Inst. Oceanog.,
30 pp.
Roden, G. I., 1959, On the heat and salt balance of the
California Current region, J. Mar. Res., 18(1), 36-61.
Roden, G. I., 1962, On sea-surface temperature, cloudiness
and wind variations in the tropical Atlantic, J. Atm.
Sci., 19(1), 66-80.
Roden, G. I., 1963, On sea level, temperature, and salinity
variations in the central tropical Pacific and on Pacific
Ocean islands, J. Geophys. Res., 68(2), 455-472.
Rossby, C. G. and Montgomery, R. B., 1935, The layer of
frictional influence in wind and ocean currents, Pap. in
Phys. Ocean. and Met., M. I. T. and W. H. O. I., Vol.
III, No. 3.
Willett, H. C. and Clark, N. E., 1966, Interaction of atmosphere
and ocean during climatic fluctuations, Report for NSF
Grant No. 24832.
Witting, R., 1909, Zur Kenntrio den vom Winde erzengten
Oberflachenstromes, Ann. Hydrogr. Marit. Met. 73, 193.
Wyrtki, K., 1961, The thermohaline circulation in relation to
the general circulation in the oceans, Deep-Sea Res.,
-147-
8(1), 39-64.
Wyrtki, K,, 1965, The average annual heat balance of the
North Pacific Ocean and its relation to ocean circulation,
J. Geophys. Res., 70(18), 4547-4560.
Wyrtki, K., 1966, Seasonal variation of heat exchange and
surface temperature in the North Pacific Ocean, Report
for NSF Grant No. GP-5534 at the Hawaii Institute of
Geophysics, Univ. of Hawaii.
-148-
Biographical Note
The author was born on February 26, 1940, in Milford,
Connecticut, and attended Milford public schools through high
school.
He entered Brown University in 1958 and was graduated
in 1962 with a Bachelor of Science degree in physics.
During the summers of 1962 and 1963, he was employed
as a research assistant by the Woods Hole Oceanographic Institute in Woods Hole, Massachusetts.
He entered the M.I.T. Graduate School in 1962 and has
been supported since the academic year of 1965-66 by a
research fellowship given by the U.S. Bureau of Commercial
Fisheries.
He is married to the former Joanna Bunker Rohrbaugh and
they, as yet, have no children.
e~c3
YEARLY AVERAGE AND STANDARD DEVIATION OF QI
1951-1957
124 133019 140 139 140 144 15,
55N
89
S79
136 135
66
67
50
221 199 178
81
96
109
45N
93
95
90
89
133 137 140 143 148 150 151 150 151
77
77 76
73 77 77
66 70
64
169 165 165 165
64
64
65
71
93 107
157 168 185
93 105
84
170 175 176 175 178 180 182 185 190 197 216 240
91 101 120
86
76 80
77
74
72
72
72
67
255 237 217 711 209 209 209 214 218 221 223 225 227 227 228 229 231 250 282(
96 115
89 88
88
87
90
88
83
81
81
78
75
73
73
76
83
92
101
282 275 269 268 268 269 271 275 279 282 285 284 282 280 276 270 265 275 304
98
85
89 83
96
99 104 105 106 109 106 101
95
91
89
92
97
104 101
314 323 333 340 344 344 343 345 348 349 348
98 103 108
97 102 108 105 103 100
97
100
348 348 351 350 345 342 333 319 305 292 290 305 339 38
8
80 80
75
77
85
80
95
108 108 109 110 109 106
342 360 379 395 403 402 404 409 411 408 405 403 402 402 400 395 388 371 349 326 306 297 303 344 406
79 73 79
71
68
68
71
92 80
96 96
97
99
99 101 103 101
99
102 iCl 107 107 108 104 101
20N
380 393 407 417 421 426 431 432 431 427 423 421 419 417 413 409 399 380 357 332 313 300 300 336 392 434
58 65 66 68 66
80
78
73 72 63 58
78
78
81
80
80
80
78
81
85
83
88
97
92
88 91
130
Figure Al.
2
140
150
160
170E
180
170W
160
150
140
130
120
84-month average values and standard deviations of the incoming radiation corrected for reflection and cloud cover.
(cal/cm /day)
110
YEARLY AVERAGE AND STANDARD
DEVIATION OF QB8
1951-1957
55K
50K
45N
111
37
104 i01
33
31
90
33
94( '95
29
34
94
29
94
26
96 100 10!
26 27 3;
90
32
89
31
87
29
87
27
87
25
88
25
90
26
90
24
89
22
88
21
88
21
92
21
98
22
98
32
95
32
94
31
94
28
94
27
92
26
91
25
91
23
90
20
91
19
92
19
94
18
97 105
18
16
23
135 126 118 115 111 108 106 105 105 10C3 102 101 101 101 102 103 105 110 118(
43 36 33 32 32 29 26 24 22 20 19 17 15
14 14
14 14
15 181
131 126 120 117 115 114 112 112 113 113 112 111 111 112 112 112 111 114 121
38 34 31 28 27 23 20 18 16 14 13 11
11 12 12
13 14
15 18
20N
126 129
42 43
129 126 122 120
40 35
28 25
111 113
29
27
114 115 116 116 116
26
23
21
19
19
103 103 105 107
19
17
16
15
130
Figure A2.
140
118 118 119 119 119 120 120 121 121 119 120 120 119
23
20
18
17
16
14
14
13
12
10
10
10
12
118
16
119 119 119
15
15
15
108 109 110 111 111 112
15
13
13
14
14
14
150
160
170E
121 121 122 122 122 123
15
15 13
13
11
10
111 112 113 113 114 116 117
16
16
14
13
13
12 11
18C
170%
160
122
10
117 115
14
15
115 119 126 131'
16
19
19
119 116 113 112 115 121
12
14
16
19
20
19
129
19
115 113 111 110 109 111 115 121 122
11
13
13 14
15 16 15
16 18
150
140
130
120
110
84-month average values and standard deviations of the effective back radiation from the sea surface. (cal/cm2/day)
YEARLY
AVf:AGE
ANC STANDORC
DEVI
T
ICN CF CE
1951-1957
1C4 1C3
83
76
108
76
ES
91
106 102 ICC
71
70
67
96
65
97
60
89
69
84
62
95
60
96 104 111
63
68
68
94 100
65
67
116 1081
68
61
V122 108 123 143 147 143 148 143 131 121 116 116 112 116 122 126 128 120
174
20N
(
137
101
181 169
142
'02 1C5
99 101
154 147 145
137 125
59
83
74
76
72
69
70
74
71
67
55
99
49
117 106
SC
90
81
76
75
69
62
59
46
41
92
193 215 313 353 384 365 339
183'rf61 190 24 224 214 197
?12 297 279 254 239 230 219 208 202
178 170 151 127 115 109 1C1
97
91
309 316 321
214 178 167
"87
364
192
327 332 317 299
170 173 161 149
333 3Y2 283 281
164 15f 128 121
129
197 194
82
76
190 180 169 153 127 110
68
57
52
44
38
32
282 273 264 259 253 253 254 251 244 241 230 213 194 17,5 160 134 124
121 11C 10C2
96
93
86
85 79
75 68
58
54
52
44 45
38
43
282 277 '71 275 280 290 297 292 296 301 299 295 291 282 261 235 200 187 169 154
113 106
92 91 89 9O 95 88 78 79 71 79 66 65 67 68 64 70 70 66
328 211 301 293 290 287 283 287 296 299 299 308 305 299 294 308 314 303 303 297 277 266 257 232 212 196
136 123 115
92
85
78
75
El
94
98 100
84
78
8C
89
84 84 83 89 91 86 82 83 85 72 54
12 0
Figure A3.
14C
150
160C
170E
18C
1701
160
150
140
13C
120
110
84-month average values and standard deviations of the latent heat transfer between ocean and atmosphere. (cal/cm 2 /day)
YEARLY AVERAGE ANC STANCARC CEVIATION OF CS
1951-1957
5CN
45K
9
9,-96
C6/91
95
2/d
(cal/cm
99
-45
83
39
72
35
53
42
54
50
64
58d
84
49
72
45
66
42
6C
36
53
31
48
3C
45
27
41
24
39
21
38
21
37
25
38
29
36
32
39
312
42
75
88
64
76
52
67
49
63
41
52
31
42
24
36
22
34
19
31
17
30
19
30
20
32
21
31
22
31
20
29
95 95 102
113 105 107
93
93
78
80
67
69
59
59
52
52
44
44
34
36
29
32
26
29
24
28
24
27
24
26
23
25
23
24
18
22
1
2:
106
99
106 100
85
86
73
74
65
68
56
55
46
44
43
40
40
37
35
31
31
29
29
26
27
24
26
23
25
22
24
20
22
19
18
17
11
16
9
12
76
86
72
73
65
64
60
59
54
53
49
49
43
42
37
34
36
31
34
29
32
27
32
26
31
25
29
23
28
24
27
22
27
20
27
19
25
17
23
16
22
15
21
14
19
15
14
11
8
12
62
65
48
48
38
38
31
30
28
26
28
24
25
21
23
18
23
17
23
17
25
19
27
22
26
22
25
2C
25
20
24
19
26
17
26
14
25
14
26
15
25
15
23
16
22
18
18
16
11
16
34
32
28
25
25
22
22
16
20
14
19
13
18
13
18
15
18
17
18
16
17
18
19
18
18
17
18
16
18
16
21
15
23
15
22
15
24
18
27
17
27
18
27
17
27
18
24
16
19
14
130
Figure A4.
JOO
82
76
103
6
3
60
93
76
140
150
160
170E
180
17CW
160
150
140
130
84-month average values and standard deviations of the sensible heat transfer between ocean and atmosphere.
ay)
15
9
120
110
2
(cal/cm
YEARLY
AVERAGE ANC STANCARC CEVIATION OF
SLHT
1931-1957
-82 -74 -92-105-1121
213 193 196 206 233
230 238
-107-101-104 -92 -80 -74 -64 -61 -54
224 211 203 194 191 187 181 176 173
-76
45N
-80-124-149-143-126-126-109
-81
-62
-53
-48
-55 -67 -75 -78 -591
176 185 188 189 191
-39 -44
-49 -50
-50
-29
5 289 288 271 259 242 235 217 200 186 183 177 172 176 183 182 179 178
-201-208-249-245-217-196-175-151-123 -92 -72 -59 -50 -52 -54 -51 -45 -10
387 359 359 333 314 292 267 252 235 209 201 191 184 181 176 169 164 158
3-328-307-272-230-205-178-141-119-104 -84 -66 -58 -52 -52 -51 -45 -37
-9
46
1 434 415 385 346 321 296 266 251 242 227 217 208 193 184 173 157 144 134 130
402 368 353
173-164-143-117 -97 -89 -78 -67 -62 -55 -52 -53 -51 -49 -55 -54 -48 -40 -21
351 347 325 299 267 252 243 234 226 220 21C 207 193 181 168 151 142 132 117
-194-133 -74 -34 -22 -23 -15
-3
-6 -14 -29 -41 -37 -40C -48 -50
361 319 298 270 257 240 220 205 204 204 205 209 199 180 171 158
-85 -49 -24
252 236 227
130
Figure A5.
-6
3 11
20
17
6
-1
194 179 167 161 171 183 181
140
150
160
170E
-5 -19 -16 -13 -13
184 168 157 149 148
180
17CW
7
66 1
118 116 1
-57 -69 -77 -76 -67 -47 -21
157 140 130 129 123 119 135
37 111
143 152
-36 142 134
-60 -83-103-101-101 -95 -35
40 100
130 139 139 136 125 133 151 143 123
160
150
140
130
120
84-month average values and standard deviations of the total transfer of heat between ocean and atmosphere.
(cal/cm2/day)
110
YEARLY AVERAGE AND STANDARD DEVIATION OF SST
1951-1957
55N
432 434 442 449 457 468
58
57
56
53
44 48
0
481 493
5
57
•
B
50N
438 437 442 446 449 453 458 464 469 476 485 497 509 518
61
53
48
46 45
45 48
52
55
57
57
56
55
50
45N
3 464 468 477 479 482 490 497 500 504 508 512 516 521 528 535 540 542
2 83
76
72
66
60
55
53
50
51
53
55
57
58
58
56
54 49
40N
624
09593 580 584 585 583 584 586 587 587 585 586 588 591 593 594 593 588 577 56d
888
83 79
74 69
65 61
58 57 56
57 59 60 60 58
55 51
48
42
32
35N
6 5 664 685 694 695 689 685 681 678 675 670 667 664 661 660 661 662 661 656 646 634 614 592 597
5r'8374 68
67
67
66 65
62
60
59
58 58
59 58
57
54
50
45
42
38
34 29
29
30N
716 735 745 747 746 744 741 738 736 733 730 726 723 719 717 715 712 737 697
70
62
59
57
57
57
56
56
55
52
51
49
48 47 45
42 38
34
31
682 667 651 636 636 654
29 28
27 27
31
4
783 786 788 791 792 792 791 788 786 782 778 774 770 765 761 755 748 739 727
46 44 45
43
42
42 41
42 41
39 37 35
34 32
30
27 24
22
21
714 701 688 678 677 690
20
21
23
27
33 44
815 816 819 822 822 821 819 815 811 807 802 798 794 788 782
30
29
29
28
26 26
26
26
26
25
24 23 22
22
21
776 769 759 751
20
19
19
18
742 732 725 724 732 750 781
19
22
23
26 29
34 36
160
140
25N
_
20N
130
140
Figure A6.
150
160
170E
180
170W
150
130
84-month average values and standard deviations of the sea-surface temperature. (OF x 10)
120
110
YEARLY AVERAGE AND STANDARD
DEVIATION OF
SAT
1951-1957
55N
415 417C427 436 443 453 459 467
57
61
64
66
67
69
70
74
50N
421 422 428 434 439 443 449 455 462 469 477
74 66
60
55
53
54
56
58
60
62
62
45N
459 441 441 449
114 13
97
88
574 5 1 548 540 543 949
121 116 111 105
96 88
456 464 473 482 489 495 500 505 512 515
80
72
65
60
57
56 58
60
61
61
486 495
62
62
5
03-7
61
521 527 531 532 533
61
59
57
55
51
553 558 563 567 569 571 573 576 579 582 583 581 577
81
74
69
65 63
62
63
63
62
59
57
53
50
567 557
46
40
59
619 636 646 650 648 649 650 650 650 649 647 644 644 644 645 647 645 643 631 620 603 585 590
119JI11
5 -98
94 91
85
79
75
70
67
65 64 63
62
60
57
53
49
45
41
38
33
32
677 694 737 713 718 718 718 718 716 714 711 707 704
96 90
83
77
72 70
66 63
60
58
56
55 53
25N
20N
a
753 759 755 77D 773
69 62
58
54
51
702 700 699 696 689 68)
51 49
46
42
38
35
666 651 636 622 625
33
32
31
31
32
773 773 771 769 766 762 757 753 748 745 740 732 722 711 697 683 670 660 663 680
49
46 44 44
42
42 41
39
37 34
31
27
24
24
24 24
26 30
33
44
l
796 800 8)3 806 807 806 805 802 798 794 790 786 782 776 771 763 755 746 737 726 715 707 704 714 734 765
39 36
34
32
30
28
28
28
29
28
29 28
27 26 25
22
20
23
21
21
23
25
29
32
37
38
130
140
Figure A7.
150
160
170E
180
170W
160
150
140
130
84-month average values and standard deviations of the surface air temperature. (OF x 10)
120
110
YEARLY
AVERAGE AN
STANCARC
CEVIATION OF CLOUC
CoVrp,
1951-1957
45h
84
S11
82<8 1
12
10
81
10
81
9
81
9
80
10
78
13
76
85
9
85
9
86
8
86
7
85
7
84
7
83
7
83
7
83
7
83
6
83
7
82
7
80
8
11
"N
72
-14
77
11
83
9
85
7
85
7
84
8
85
8
84
8
83
8
83
7
83
7
82
7
82
7
82
6
81
6
81
6
80
6
75
6
70
9
71
8
75
7
79
6
80
6
80
7
80
6
8C
6
79
6
79
6
78
5
78
6
78
6
77
6
77
5
77
5
77
5
76
6
72
6
65(
8
70
8
72
8
73
8
73
8
72
8
72
7
72
7
72
7
71
7
7C
7
70
7
70
7
71
7
71
7
71
7
72
6
73
7
71
7
'64
9
68
11
67
8
64
7
63
6
63
8
63
8
t3
7
62
6
61
6
61
6
62
7
62
6
62
7
61
7
61
8
63
7
63
7
65
6
67
7
70
7
72
8
72
8
69
10
61
12
52
11
67
IC
63
9
60
9
56
8
54
9
54
8
54
8
52
7
52
7
53
7
53
7
54
7
54
7
54
7
54
7
55
7
57
7
60
6
64
8
69
9
73
9
75
10
73
13
64
13
51
11
6C
9
58
9
55
10
53
9
51
9
50
9
49
7
49
7
49
8
50
7
51
8
51
7
52
8
52
7
53
8
54
8
56
9
60
9
64
10
69
10
72
9
74
9
75
10
68
9
57
9
130
140
Figure A8.
150
160
17CE
18C
17CW
160
150
140
130
84-month average values and standard deviations of the cloud cover. (% of sky covered)
120
48
8
110
YEARLY AVERAGE ANC
STANDARC DEVIATION OF OBSW
1951-1957
917 904a894 888 871
272 281 239 242 211
844 822 805Vy(
191 203 237y
0
910
243
45r
'733 783 8c2
245 246 246
G19 957
239 239
958 938 928 913 905 908 903 890 868 837 78,
224 222 199 190 181 191 201 206 197 181 17.
901 919 928 943 928 904 898 894 882 871 868 865 848 827 786 7
232 22P 214 223 214 210 180 175 180 192 195 208 199 187 151 1
799 833 8 3 900 897 900 894 886 870 854 844 824 808 807 796 781 767 758 7,
182 191 205 201 197 189 191 191 193 172 168 167 169 182 183 176 169 115
857 860 8,0 853 850
169 186 1'4 184 190
753 719 712 732 749 746 741
175 123 109 120 140 152 153
735 733
146 153
832 814 800 787 778
187 183 178 180 173
767 751 739 726 705 692 683 709 735 665
175 166 156 162 156 139 121
89 104
90
723 713 707 705 712
150 143 131 133 135
717 714 7C5 696 677 657 637 640 660 615 562
141 132 124 118 105
99 96
88
98
77 75
780 720 674 640 625 629 627 620 630 544 663 675 683 69t4 711 719 719 723 720 695 659 619 590 595 574
163 140 144 126 120 125 172 123 134 123 10t 110 112 104 112 113 112 105 104 101 104 101 120 106 102
20N
7
73C 698 6,58 642 624 623
172 158 144 118 117 10
130
140
Figure A9.
150
675 639 65
17 117 124
160
70 683
123 118
706 713 713
106 106 110
170E
180
17CW
712 732
117 112
160
748 744 753 747 715 683 653 616 574 497
115 124 126 122 110
99 102
97 85
64
150
140
130
84-month average values and standard deviations of the observed wind speed. (cm/sec)
120
110
YEARLY AVEFAGE ANC STANDARD
DEVIAfIIQN CF SPECIFC
HL/'D Jl /7-"/
1951-1957
50 50
120
3
5C1#
45%
54
13
56
14
57
15
59
15
51
50
51
52
53
16
14
13
13
12
54
13
55
14
56
14
58
14
60
15
61
16
63
16
65
16
66
15
-59
28
56
24
56
22
56
21
57
20
58
19
60
18
61
16
62
16
64
16
66
17
67
17
68
17
69
17
70
17
71
17
72
16
72
15
80
35
78
32
79
30
79
28
79
27
80
25
81
23
82
22
82
21
83
21
84
22
85
21
86
20
86
19
85
18
84
17
83
16
S0
14
7e
12
111 110 110 110
41
38
36
34
110
31
109 108 107 16 10 6 106 106 106 104 101
29
28
26
25
24
24
22 21
19
17
97
15
93
14
89
13
85
12
7
11i
125 129 133 135 135 136 135 134 133 131 129 128 126 124 123 122 120 117 112 106 101
46 43 43 41
38 35
3 31 29 28 26
25 24
22 20 18 17 15 14 13 12
97
12
94
12
97 105
13 1
I
156 157 158 159 159 158 158 156 154 152 150 147 145 142 1 0 137 134 129 123 117 112 17 105
1
10B .117
38 35 34 33 31 28 27 26 25 24 22 21 20 18 16 14 12 11 10 11 11 1' 16
17 21
176 117 178 177 176 175 174 172 170 167 165 163 161 158 155 151 146 142 137 132 127 122 121 127 198 153
23 22 21 21 20 18 17 18 18
18 18 17
15 14 14
12 11 11
11 11
12 14 17 19 22 23
130
Figure A10,
140
150
160
170E
18C
170%
160
150
140
130
120
84-month average values and standard deviations of the specific humidity of the air. (g/kg x 10)
110
LkM
<
Lin
44-
U
(Ni
0
Ica
(N
- (1
Ln
M
CO
'4
-
-44 M-
cv
'l
1).
4 (Vj
(-4.C
r-
NJ V'
'T4
cm fl-
(NJU
17
C? to.
4
4
C\) (M
rn'
m
(NI (4$
4..
(C
~ II
4444
C-4 Ln
C)'
L44(f
-All-
J
1
-4
-4 -4
414
-4-4-i:
-4 -4
4
-4
.44
-4
4
-.4
-44
-4
4
-
-4
--44
.-
~
N
-4
-4 -4
4 -4
4
-
4
-4 -4
..
4-
o,-4
.-t.--4
1
IN(4
4
.44 .-
--
fNJ (f)
-4
N
(.44-
~
(Nj 444
,
44JN
T
4
-1-
-4 -4
0
-44
.4,-4
-4
4
-4 -4
-4
--
-'
-44
14141
-4
-44-
--
V4
'-4
C)d
.)
'
-4
'
41
44
"4
:-4
4
td
CDC
41
-4v
2
$
W
7-YEAR AVERAGE AND STANDARD DEVIATION OF QI
JANUARY 1951-1957
55N
50N
45 N
90
?3
70
6
125 106
12
7
64
5
67
2
32
7
32
7
29
6
27
4
27
5
45
4
46
5
43
6
43
5
46
5
48
4
51
5
51
4
48
6
47
7
47
6
47
6
47
5
73
6
76
9
74
10
76
11
77
9
79
6
83
10
79
7
71
9
76
8
77
6
7.
7
73
7
98 105 112
4
5
10
143 140 143 152 160
8
5
4
8
11
77
q
157 156 157
10
10
11
153 150 147
i7
16
12
147 146 144 145 150 152
6
9
9
12
10
11
155 169
12
16
194
13
186 195 225 217
16
19
21
221 217 214 210 206 204 2-3
9
12
11
6
9 16
13
218 229 240 252 252 264 279 278
24
27
24 27
35
40
30
19
282 279 274 269 262 263 2b8 262 260 2o3 259 2"9 227 213
14
11
11
15
23
24
17
19
17
18
18
9
23 31
277 279 282 306 316 321 330 325 326 328 331 328 320 321 3e5 317
40
39
37
22
17
22
15
23
20
14
19
23
29
26
20
23
130
Figure A12.
140
150
160
90
20
111 109 111 Ill 113 115 112 109 108 In6 1)9 110 114 1 5
7
7
9
14
13
14
9
9
9
6
6
6
8
16
176 198 213 204 194 202 208 213
16
10
12
17 22
20
18
11
$7
20N
29
31
804 9
170E
18C
70OW
160
196 20O 2L,0 199 101
15
16
16
14
12
323 323 297 223
23
28
39
28
l3,
110
276 255
28
33
130
215 247 286
29
26
13
232 247 293 345
26
19
23
17
120
7-year average values and standard deviations of the incoming radiation. January 1951-1957. (cal/cm 2 /day)
11i
7-YEAR AVERAGE AND STANDARD
FEBRUARY
DEVIATION OF QI
1951-1957
55N
90
15
45N
"V139126 111
33
15
10
168 1 8 172 159 144
23
17
19
10
10
218 215 220 212 195 192
38 '33
27
15
21
14
89
9
85
11
86
7
H8
11
63
14
671 68
11
12
68
13
64
66
1.1
8
66
9
85
9
87
7
92
9
37
6
86
12
91
8
118 127 130 132 135 139 131 117 115 119 121
8
12
13 16
12
22
18
13
16
17
i4
152 160
7
10
163 168 171
12
13
12
176 171 162 155 157
19
14
12
16
15
190 198 204 206 204 206 207 207 208 199 197
12
11
15
18 20
11
8
10
8
8 10
85
9
99
12
921
14
118 1!7 118 128 135
10
10
8
9
15
160 156 151 154 168 191
9
10
10
9
14
23
199 195 191 195
8
1D
15
19
211 246 277
23
33
0
226 236 251 255 249 250 255 267 279 279 271 272 275 271 265 263 262 258 249 247 248 258 282 313 333
34
11
9
11
18
17
12
12
20
22
24
47
50
13
13
43 21
11
10
18
13
6
7
11
14
252 276 296 303 298
44
33
26
25
28
20N
301 315 333 348 348 336 347 355 346 332 332 329 315 299 237 282 289 315 346 375
26
20
12
23
26
22
16
17
17
22
28 20
17
24
19
17
28
55
54
27
0o
342 343 346 351 349 359 374 382 389 389 385 396 399 387 378 368 348 322 308 290 275 289 311 339 376 417
29
22
38
49
57
49
28 23 50 61 31 13 23
25 24
27
28
27
27
43
33
22
14
19 24
22
130
Figure A13.
140
150
160
170E
18C
170W
160
150
140
130
120
7-year average values and standard deviations of the incoming radiation. February 1951-1957. (cal/cm 2 /day)
110
7-YEAR AVERAGE AND STANDARD
DEVIATTON OF Qi
MARCH 1951-1957
55N
126 125~127 125 134 143
16
3
22
29
17
32
50N
145 147 147 155 161 157
36
33 28
21
19
15
45N
'226 212
0
39
!2
26
179 176
16
22
153 151 152
19
19
20
179 179 18t
195 192 184
20
22
25
35
22
22
154 157
16 21
145 155,
39
56
161 157
29
35
185C
33
179 181 187 194.193 197 213 227
16
17 1
4
15
15
20
20
29
278 267 249 226 204 206 2')8 208 208 205 2)9 215 216 210 214 223 219 217 222 249 29.
34
34
32
16
10
13
22
20
23
14
18
14
18
11
9
18 22
19
16
12
17
33
317 298 28
52
43
6& 34
264 250 245 252 251 250 250 248 248
15
7
13
11
9
11
16
17
11
308 309 323 332 330 332 329 337 334 320 314
43 22
22
21
26
19
12
19 20
26
32
254 257 249 249 248 243 245 250
1)
13
17 23
28
18
14
16
316 323 322 312 309
22
13
17
22
23
415 425 446 452 456 447 442 446 437 423 418 422 419 414 410 394 380 358 332
44 35
21
20
27
24
33
40
39
36
36
32
27
31
43
48
32
33
55
Figure A14.
140
150
160
170E
1.8
170W
160
322 361
25 '32
308 299 293 286 274 233 319 372 407
30
26
19
17
26
11
33
29
1
316 350 385 405 412 404 400 411 404 382 380 398 396 388 380 377 364 351 339 317 292
19 24
38
30
31
28
27
32
35
44 45
26
19 20
34
30
33
26
26
29
32
130
271
11
150
?87 312 358 412
33
55
37
23
328 337 286 305 345 436 482
43
47
52
68
47
27
33
140
130
120
7-year average values and standard deviations of the incoming radiation. March 1951-1957. (cal/cm2/day)
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF QI
APRIL 1951-1957
K)
55N
220 2384228 224 242 237 254 268
55
69
64
75
40
25
38
62
V
50N
188 193 205 213 217 237 249 240 239 238 241 260 240 3151
59
51
38
25
22
28
29 34
34
23
26
32
40
48
45 N
J348 320 283 247
\-80
57
50
37
229 225 230 251 258 253 249 242 245 257 273 289 294 306 322
28
23
23
17
28
27
31
36
29
18
22
24
32
24
33
373 340 302 285 271 262 261 275 23) 270 270 270 272 230 294 318 318 330 355
50
28
35
24
17
11
12
2C
24
16
23
29
29 24
27
27
11
14
261
348 330 319 316 310 306 3tO 31C 320 320 327 326 321 319 324 336 344 345
24
26
21
13
17
22
17
19
22
18
19
22
31
37
37
24
12
17
30N
25N
20N
363 367 371 381 390 389 389 380 375 380 384 38' 376 381 389 384 386 385 374
53
24
27
22
21
23
24
18
16 27
25 29
31
29
39
40
25
22
28
368
31
374
40
3,51 365 385 398 422
?6
24
44
57
51
S31
377 407 428 453 470 474 473 457 A42 453 463 446 431 429 433 437 440 430 407 386 372 367 364 373 426
35
48
38
53
43 32
26
37
36
35
34
26 21
-4 47 38 20 31 54 47 44 77 59 35
466 484 499 500 503 507 5C6 5C4 489 477 477 481 472 453 441 433 444 431 387
34 32
28
36
37
28
23
19
4,1 26
42
37 42
41
48
45
38
38
53
130
Figure A15.
140
150
160
170E
180
170W
160
1959
370 369 349 324
52
44
49
56
140
1
3
353 426 490
51
25
27
120
7-year average values and standard deviations of the incoming radiation. April 1951-1957. (cal/cm2 /day)
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF 01
MAY 1951-1957
55N
235
46
4
233 217 232 252 276
61
36
25
39
72
V
50N
232 225 222 233 242 250 262 265 255 241 248 265 291 312(
26
39
31
22 26
23 22
22
22
16
29
;3 39
43
45N
"2386 362 306 284 272 266 255 257 263 273 278 283 291 285 286
2 66
52
29
22
21
22
21
33
34
23
21
27 23
31
417
37
97
410
88
30
98 377 357
38
39
22
316 297 292 289 282 287 293 301 302 330 311
13
12
24
23
28
24 29
24 24
26 21
97 365 341 325 315 318 331 348 360 367 363 355 357
28
30
20
16
19
18
22
35
26 20
29
30
19
3') 314 344 385
26
14
25
37
318 326 334 339
19
17
10
11
373 42
18
21
364 372 374 372 373 393 434
9
23
26
11
22
34
43
56
44
373 390 413 434 433 427 430 435 441 452 455 458 454 445 433 440 452 439 421 4'7 396 392 402 432
49
38
19
27
32
16
15
17
15
27
34
23 20
21
31
18
20
?9
41
39
46
50
48
50
25N
20N
424 438 474 512 527 528 524 525 526 521 514 509 4S9 494 487 486 491 469 441 422 436 391 365 397 487
30
23
34
25
22
26 25
24
19
24 26
30
30
31
38
38
47
39
54
52
56
69
61
69
70
Po
493 5C9 526 531 528 535 533 523 511 512 499 497 484 478 483 483 474 456 437 392 363 345 325 361 440
31
28
29
34
37
33
35
37
42
32
11
35 65
55
49
51
46 48
68
69
59
80
78
64 55
130
Figure A16.
140
150
160
170E
180
170W
160
150
140
7-year average values and standard deviations of the incoming radiation. May 1951-1957.
130
120
2
(cal/cm /day)
514
32
110
7-YEAR AVERACE AND STANDARD DEVIATION OF QI
JUNE
1951-1957
55N
233 255 283 286
2 4
28
58
40
50N
186 190
13
19
45N
262 239
57
37
227 245
25
89
194 205 225 234 242 247 257 249 232 236 257 2721
15
24
23
14
18
40 46
16
25
30
23 26
'315 259 235 223 207 2'35 211 234 258 263 261 267 273 267 263 277 294 321 363
-61
47
32
23
13
13
12
15
23
31
34
31
27
22
23
29
34
22
31
tr
337 299 273
20
23
33
258 254 251 250 272 2E8 291 297 311 316 317
17
25
16
13
12
?4
23
26
35
23
22
349 344 331 321 325 322
27
31
35
38
33
27
327 345 358 367 382 395 395 386
22
29
38
24- 19
29
24 21
319 31P 317 352 416
28
32
39
33
55
376 359 342 367 429
32
26
32
41
67
368 384 416 434 457 467 460 450 451 454 464 469 465 475 483 471 448 424 397 332 371
25
29
15
27
31
22
21
24
23
30
32
37
23
1 ' 22
20
15
24
31
38
40
437 477 522 537
52
44
37
23
60
20N
548 553 548
18
14
34
455 503 549 556 550
27 17
15
23
22
130
Figure A17.
140
543 549 553 554 553 540 532 525 504 470 437 395
45
32
17 20
24
22
20
22
29
29
29
32
389 419 423 470
47
30
36
374 363 373 387 392 467
28
33
46
64
56
35
554 555
20
24
543 546 542 535 534 533 519 501 493 467 428 373 334 337 337 338 357 409 495
27
26
32 43
34
24
25
30
30
44
48 27
23
46
53
40
36
40
33
150
160
170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the incoming radiation. June 1951-1957. (cal/cm2 /day)
110
7-YEAR AVERAGE
AND STANDARD DEVIATION OF QI
JULY
1951-1957
55N
225 2?6 220 220 2331
50
37
34
43
41
0
50N
182 182
21
19
45N
o
o
0e
wc
176 177 178 184 190
17
17
18
14
19
199 232 2)5 238 212 226 265
36
11
13
15
14
22
16
'261 2?6 215 210 2v2 202 205 2'06 21) 216 219 222 224 230 240 247 264 316 389
\.52
40
20 26
19
16
12
13
19
20
26
21
2J
14
18
25
29
31
37
332 306 284 269 258 274 293 302 310 321 330 329 322 311 310 3)3 331 334 397
38
42
38
33
36
33
39
3
36
38
41
38
36
30
36
33
23
37
52
422 400 374 361 365 395 418 4?7 437 444 441 437 422 399 374 342 311 317 353
55
59
54
46
42
37
52
48
34
20
26
33
40
45
59
54
33
42
63
439 467 481 487 483 464 450 457 468 487 499 501 507 512 498 492 482 437 387 3430 316 312 328 383 453'
52 49
40 40
51
50
54
42
18
18 27
22
25
21
32
30 40
32
33
36
23
24
61
45
26
20N
530 504 512 517 518 502 498 509 513 511
58
54
57
31
31
47
65
46
31
31
508 515 528
41
39
32
460 469 482 494 508 5U5 498 494 492 491
64 55
43 32
16
38
39
40 40
50
493 500 512 511 497 497
66 43
33
32
30
21
(2'
130
Figure A18.
140
150
160
170E
180
532 516 514 503 445 383 335 318 306 317 4 2 489
23
30
28
29
'2
23
22
23
25
80
65
49
170W
160
86 439 400 358 324 308 320 399 471 475
13
35
36
41
55
16
40
34
48
31
150
140
130
120
7-year average values and standard deviations of the incoming radiation. July 1951-1957. (cal/cm2/day)
110
7-YEAR AVERAGE
AND STANDARD DEVIATION OF QI
AUGUST
1951-1957
55N
0
50N
164
16
45N
144
11
163 165 168 168
22
20 24
18
155166 182 201 216 217 226
13
19
33
47
49
47
62
162 168 173 130 135
9 10
13
17
13
268 219 198 198 207 216 218 219 216 203 203
2
39
19
20
26
33
36 25
20
7 13
195 209
17
21
217 244
32
34
218 226 229 237 246 251 285 335
20
29
19
20
22
21
31
69
404 3 5 351 327 3C6 2Q8 299 304 310 313 318 313 314 332 337 325 319
33
0
54
37
18
15
30
38
36
26
27
22
24
33
39
23
17
310 32 327 373
32 38
39
55
43
450 43519 418 420 419 411 405 399 397 41C 422 427 432 438 428 407 385 360 339 341 356
67'
2 25
43
50
35
27
24
22
20
15
22
43
37
29
32
31
27
26
34 42
27
36
62
0
443 455 448 455 467 459 457 467 470 470 468 463 4ud 475 480 480 467 437 400 372 345 310 331 351
45
21
13
16
28
29
31
24
21
23
26
28
10 27 23 26 26 22 27 40 44 37 ?8 24
25N
20N
(
444 432 440 464 466 446 456 491 501 500 499 491 472 477 483 488 477 438 388 347 318 272 274 350 461
33 36
40
31
26 28
25
26
31
37
46
47 38
?2
24
3)
24
15
37
35
31
39
42
35
31
386 378 391 412 413 419 445 469 474 476 471 458 451 462 470 481 458 424 399 3'1 303 291 339 377 455 473
33
42
54 47
40
20
27
37
43
44
5C
40
34 25
19
24 44
41
15
13
28
38
46
43
26
24
130
Figure A19.
140
150
160
170E
3
11 C
170W
160
150
140
139
123
7-year average values and standard deviations of the incoming radiation. August 1951-1957. (cal/cm2 /day)
110
C::3
7-YEAR AVERAGE
AND STANDARD DEVIATION OF QI
SEPTEMBER 1951-1957
55N
124 13
13 -21
ba1
.4 152 148 150 178 196
.6 32
30
35
42
53
"d
50N
181 169
29 33
45N
229 214 2U4 199 195
"-49 26
21
16
21
147 147 149 151 158 162 165 172
19
13
8 10
16
18
13
15
186 176 181 195
17
17
15
12
182 231 217
22
28
27
196 200 214 219 221 226 234 243 261 271
8
8
16
15
9
17
19
23
8
47
280 273 266 264 262 254 241 241 249 254 263 275 28
274 271 275 278 290
30
20
22
18
21
20
19
24 27
14
13
19
17
9
7
14
17
14
25 338 351 355 357 349 336 325 317 316 324 339 337 337 339 331 322
30
26
24
26
21
26
29
26
26
24
17
16
22
23
10
15
20
368 369 375 387 403 416 417 407 396 394 393 391 386 395 409 388 379 393
27 32
37
26
22
25
26
15
20
24 21
11
24
17
17
22
23
12
20N
391 417 424 421
43
34
34 31
130
Figure A20.
140
311 310 320
27
32
45
388 370 338
16
37
51
376 412 429 433 433 436 438 423 415 420 428 435 437 444 451 430 417 424 418
36 32
36
27
29
19
23
?3
30
27
13
7 20
22
24
26 21
15
27
3)91
44
388 343
48
55
42
'i
314 323 380 438
46 49
53
36
328 348 404 464
50
54
49
33
413 427 441 445 444 447 446 435 437 442 442 444 442 433 443 405 352 335 342 387 429 430
45
31
34
36
24
16 25
34
35
32
15
32
42
44
40
52
55
37
41
38
34
28
150
160
170E
18C
170W
160
150
140
130
120
2
7-year average values and standard deviations of the incoming radiation. September 1951-1957.
(cal/cm /day)
110
C--z,
7-YEAR AVERAGE AND STANDARD DEVIATION OF QI
OCTOEER 1951-1957
81
12
50N
45N
'196 103 169 148
-21 16
7
11
211 209 194
9
7
13
82 '91
13
13
91
14
85
11
87
12
83
12
9
15
120 116 110 107 104 107 110 1.16 120 119 115 115 128
14
21
20
11
8
6
5
6
6
6
5
7
17
1411
29
142 141
10
11
177
15
186 178 174
13
10
13
138 136 133
18
17
12
138 148 153 154 157 155 157 155
7
9
11
9
13
9
11
21
171 17C 168 175 185 187 189 194 200 233 209 224
17
14
8
5
6 12
13
5
6
11
16
12
231
230 234 237 242 240 235 236 231 230 232 232 231 236 248 255 253 251 261 290
19
20
14
14
15
17
20
17
11
.8 16
16
11
9
13
14 20
18
27
308 298 293 294 304 309 308 315 319 316 311 306 307 319 305 301 3J2 3)8 306 295
59
37
23
18
23
28
26
17
17
16
15
15
17
14
15
13
12
7
14
?1
324 343 347
27
24
22
356 361
25
25
130
Figure A21.
279 272 280 3)1 3'36'
32
26
24
21
2
357 372 372 364 371 381 375 362 356 358 367 372 366 358 356 350 334 309 290 293 329 378
15
30
26 32
23
25
19
17
22
23
22
17
18
24
19
17
24 43
39
46
40
46
360 364 374 388 392 395 399 395 382 371 373 386 403 396 3
371 367 362 347 329 326 345 380 411
24
17 24 20
18
15
21
26
27
29 32
22
14
28
44
33 35
38
41
37
52
51
50
32
140
150
160
170E
180
170W
1')
150
140
133
120
7-year average values and standard deviations of the incoming radiation. October 1951-1957. (cal/cm 2 /day)
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF QI
NOVEMBER 1951-1957
55N
kl
6
41
7
43
8
43
10
40
7
4V 7
ad'
co
A .0O
50N
45N
124 109
-22
16
167
12
10
63
8
66
10
66
6
66
5
66
7
66
7
66
6
67
6
64
7
63
9
60
6
60
62
-3. - 8
94
90
91
94
96
96
97
91
91
95
95
93
91
91
90
7
3
6
4
9
8
6
10
12
11
9
5
9
8
7
152 138 124
17
60
10
6
120 120 121
4
7
5
120 122 125
6
8
126 126 124 121
8
10
6
123 125
5
8
159 156 154 154 155 155 156 158 161 162 163 164 168 172
8
8
8 11
9
8
9
12
9
7
10
11
12
12
126 128
9
98 106
10
18
138 1541
-11
-11 16
172 174 183 205
9
13
12
11
217 220 218 216 220 220 219 223 227 223 218 220 223 232 235 234 232 231 230 221 211 217 234 269 300\
31 15
8
9 12 10 11 12
9
8 13
8
7 10 11 11
14 12 11 10 15
17 24 19 -t
238 242 246 266 289 300 296 306 312 302 289 291 300 307 308 304 298 288 280 260 242 245 243 -293 346
25 21 19 19 20 15
18 19 15
7
9 12 13 11 1.1 13 15 13 10 13 21 34 48 37 18
20N
276 291 307 329 352 353 346 352 353 350 343 337 342 342 326 324 314 297 287 282 280 261 251 287 337 366
42 27 17 17 13 14 17 25 23 16 22 26 23 30 42 39 25 27 30 23 292
28 36 36 20
-130 -...140
Figuire A22.
-
150 -
. 160
1708
180
170W .--
160
157
140
130
-
120
7-year average values and standard deviations of the incoming radiation. November 1951-1957. (cal/cm2/day)
110
7-YEAR AVERAGE
AND STANDARD DEVIATION OF QI
DECEMBER 1951-1957
S
22A3
1
9
4
19
4
22
3
20
3
19
3
20o
3
39
5
39
6
39
5
40
3
39
4
40
3
39
5
38
It
37
2
36
3
36
4
35
4
36
4
70
6
65
6
63
5
64
5
65
5
64
8
64
8
64
9
66
9
64
8
66
6
64
6
61
3
62
6
60
5
62
5
106 101
7
5
94
6
95
7
98
7
97
8
97
8
95
7
95
7
97
10
95
11
95
9
93
4
93
5
95
5
94
6
96 104 116
6
7
12
72
8
94 1
11
10
23
6
15 150 148 _453dF
I50 141 137
135
137 138
138 137
135 133
1
35
4 135
13 5 137
13
7136
3 17
137 136
1
28 18 1 71 11
6
5
6
8
14
12
9
6
10
12
14
11
178 188 197 201 196 191
29 18
16
14
14
10
189 192 192
3
9
15
190 192 193 192 191 188 185
13
12
9
12
16 19
11
195 217 231 244 252 250 254 257 253 252 255 250 245
24 26
32
23
21
19
12
12
17
17
17
13
13
20N
C
242 258 271 282 295 301
30
38
30
19 20
21
130
Figure A23.
140
150
309 311
17
16
1600
132
3 135
15
3135136
5 136
46 136
13
7
5
6
10
10
18
74
10
142 158 176
6
15
9
187 186 195 180 178 183
8
9
6
9
13
14
196 222 244
20
22
1
249 245 242 242 235 231 218 205 203 237 242 283
22
28
16
13
16
10
12
2.1 25
31
32
22
309 305 298 291 29) 285 278 279 28D 270 254 238
15
2'7 22
15
21
26 27
21
40
36
24 17
170E
67
5
170W
160
150
140
220 215 212 238 284 320
20
23
22
24
28
23
130
120
7-year average values and standard deviations of the incoming radiation. December 1951-1957. (cal/cm2 /day)
110
7-YEAR AVERAGE AND
STANDARD DEVIATION OF QB
JANUARY 1951-1957
55N
117 12001i5
43 32
37
115 113
14
17
45N
20N
104 101
17
11
120 111 113 119 132
37
20 30
29 28
a
107 111 114
7 14
21
113 103 101 106 112 117 120
16
15
15
14
16
16
15
149
14
136 131
18
12
131 128 118 117 117 114 117
13
20
20
17
14
10
18
112 104 106 111 114 112 117
13
13
9
9
17
21
20
176
10
161 158
12
12
156 146 136
17
14
12
132 126
12
13
117 114 113 113 116
13
13
11
9
12
118 119
13
18
17P 209 220 204 185 175
20J2026-19
10
10
165 161
13
13
158 149 141
14
13
11
136 129 124
10
12
119 116 114 114 117 120
11iC o
11
9
13
14
122 122
15
18
122 120
13
16
24
128 1k 0
20
16
182 192 193
10
9
12
179 162 159 156 150 148 146 143 140 134 130 128 123 122 124 125 127 125
12
12
8
8
6
7 10
10
6
6
9
8
6
6
9
11
14
15
125 132 140 145
13
12
9
160 156 154
12
9
11
154 148 148 150
12
14
13
13
125 133
17
12
40
145 142 142 142 142 136 132 132 130 130 131 130 129 124
10
5
7
8
4
5
8
"9
7
7
6
7
11
17
132 128 126 131 134 134 135 134 135 137 138 137
8
11
11
7
8
7
6
4
5
9
13
10
130
Figure A24.
140
150
160
170E
180
130 129 131 128
6
5
10
10
17CW
160
6 129 127 125 123
9
11
11
6
12
150
140
140 146
11
9
126 126 130 14b 145
12
130
11
7
6
120
7-year average values and standard deviations of the effective back radiation. January 1951-1957. (cal/cm2 /day)
7
110
7-YEAR AVERAGE ANC
STANDARC
FEBRUARY
55N
DEVIATION OF QB
1951-1957
115
4
S6
5CN
118 114 108 105 103
14
12
16
14
16
415N
35N
140 13C
27
12
16
384
142
11
169 156
9
11
139 131
11
12
156 167
17
16
25N
138 142 146 143 138 136 135
11
6
9
13
15
14
13
149 145
13
11
Figure A25.
97 101 1CO
98
11
9
14
96
99 102 103 106
14
10
12
15
98 100 108 114
94
93
98
99
98
13
17
14
9
9
13
19
13
137 130 126 iI8 1C9 104 105 106 105 103 106
1C
14
18
14
12
13
11
8
10
10
8
128 125
12
12
17
11
10
115 126
9
15
121 119 114 111 112 112 112 115 122 135
7
9
8
11
10
8
6
6
11
18
1 4
18
173 167 156 149 144 143 143 140 134 133 133 130 127 125 124 125 124 124 126 130 135 139 142
12
10
10
9
8
7
7
6
5
7
9
7
6
8 11 10
'8
8
8
13 22 19 1
137 139 138 135 139 140
7
7
9
1C
6
6
124 124 126 125 124 123 124 127 128 130 129 130 129
5
8
11
11
10
10
10
7
8
10
11
8
5
130
114 110
e
13C 123 119 109
18
15
20 18
166 152 150 147
11
9
13
14
30N
2 N.
135 136 133
14
17
18
4 110 105 113
140
150
160
170E
180
138. 135 134 131 128 125 126 125 128 132 130 139
6
127
2
10
10
10
11
13
9
5
8
17
6
11
130 128 122 115 114 113 111 117 122 124 132 138
5 10
14
16 13
9
10
10
15
16
14
10
170 W
160
150
140
130
120
7-year average values and standard deviations of the effective back radiation. February 1951-1957. (cal/cm2/day)
110
7-YEAR AVERAGE
AND STANDARD DEVIATION OF QB
MARCH 1951-1957
55N
5 N
105 104 101 101
27 20
15
11
45N
121
22
119 115 110
15
13
16
137 129 128
13
9
12
141 133
11
11
146 147 146 145 141
21
13
14
13
12
4
20N
131 134 133 135
7
8
14
12
108 105 101
18
20
16
124 118 112
17
17
17
Figure A26.
140
97
16
93
14
92
16
95
12
101 104 100
19
14
14
95
11
94
14
98 102 103 107 117
13
10
10
13
14
99
14
98 103 111
15
19
21
106 105 105 104 100 100 104 107 108 110
11
9
11
13
12
9
8
9
8
10
130 125 119 114 112 112 112 111 107
10
10
10
8
7
6
7
7
8
137 130 129 126 122 120 123 125 122
11
9
8
8
10
11
9
13
13
25
121
10
C106108 111 112 115 122 135 1 2
6
6
5
9
10
6
9
118 118 118 118 118 120 117 119 129---38 140
10
11
9
9
6
7
8
2
10
8
135 129 124 125 124 122 123 128 131 127 126 126 124 122 119 118 113 111 121 130 137
11
9
9
10
8
6
7
11
17 14
13
11
11
9
6
8
11
10
11
5
121 121 121 121 121 118 116 116 116 116 119 123
6
5
7
5
5
5
4
8
6
3
8
11
130
93 103 115
16
13
30
102
14
123
18
137
29
940;-95
24
27
150
160
170E
180
124 126 126
14
11
11
170W
121 119 116 111 112
11
9
4
11
10
160
150
140
110 107
9
10
.130
- - -
16 125132 140-14
9
7
10
120
110
- - - - - - - - - - - - - - - - --..
. .
2
7-year average values and standard deviations of the effective back radiation. March 1951-1957. (cal/cm /day)
7-YEAR AVERAGE AND STANDARD DEVIATION OF QB
APRIL 1951-1957
551
102
24
0
45N
111 104
19
19
129 124 116
17
13
17
9
80
22
86
19
87
13
87
9
97 103
12
14
97
16
95
16
96
9
88
14
86
12
90C 98
11
6
98
8
95
9
94
14
91
15
92
11
97 101 105 107 111 1
6
7
9
13
9
111 105 102 102 106 104
13
9
9
9
10
8
98
5
98
8
99 ICO 102 105 109 113 116 1
11
12
10
10
11
8
3
95
16
114 117 116
7
9
10
96 103
7
10
121
13
122 123 121 120 119 122 126 124
9
8 10
9
9
9
12
13
116. 118 117 118 119 118 12C 123 124 125
6
8
9
8
7
7
7
11
13
10
125 121 118 118
4
7
8
13
102 101 103 107 111 112 110 112 112 110 111 115 114 113 -115 116 120 119 111
9
8
6
5
5
5
3
5
9
6
8
8
8
6
7
9
7
5
4
130
Figure A27.
140
150
160
170E
111
15
107 110 111 110 112 113 112 111 112 116 119 118 123 1 7
10
9
6
5
8
9
11
13
14
11
8
7i 10-
115 116 116 115 113 1t 115 116 118 118 11i9 122
13
10
7
6
6
9
10
6
6
8
9
12
106 104 102 108
6
11
13
9
95 105 104 110 1161
30
19
13
18 '24
77
24
11
1 4 144 138 127 121 115 113 110 109
27.r13:14P'14
15
13
11
7
10
11
114 117 115
19
12
14
.99
26
180
170W
160
150
121 119
15
17
1:5
13
110 114
10
10
113 111 113-122 126
8
10
12
9
8
140
130
120
7-year average values and standard deviations of the effective back radiation. April 1951-1957. (cal/cm2/day)
110
7-YEAR AVERAGE AND STANDARC DEVIATION OF QB
MAY
1951-1957
55N
5CK
7C
70
71
72
12
16
13
10
73
75
8281
72
68
72
80
9_
17
13
14
19
12
8
15
23
77
81
9 "
-8
81
77
72
73
78
89
98
5
6
8
5
6
11
13
13
45N
105 104
20
19
90
17
86
9
82
7
80
9
78
8
78
7
79
11
82
Ic
84
8
85
6
86
7
82
5
80
6
85
6
91 100 107
5
7
7
40N
110 1 7 106 105
99 1D7
14
15
35N
9 106 10Z 107
210'1
3
9
3CN
90
13
25N
20N
S
96
92
90
89
87
87
88
90
91
91
92
92
93
97
7
9
6
7
4
6
7
11
Ii
11
9
6
6
5
2
105 100
6
9
95
10
92
8
93
7
95
5
11
97 101 103 103
6
4
6 IC
3
92
9
96 101
9
5
92
8
94
8
98 101 102 103 104
8
7
6
5
4
130
Figure A28.
140
104 105 107 107
5
7
8
5
150
99
6
160
117
6
104 105 108 110 110 110 110 113 121 1 5
13
9
8
10
7
2
4
8
10
96 100 104 103 103 104 105 109 113 115 117 118 119 118 120 124 123 120 118 117
13
8
6
4
4
7
5
5
4
5
5
5
8
11
6
4
7
8
6
7
9C
9
4
116 118
9
11
121 125
13
1
109 113 117 119 118 123 123
2
5
7
8
9 12 16
120 123 123 120
7
6
7
10
119 119 120
20
19
118 123
16 14
98 102 105 107 105 105 108
IC
8
5
9
14
12
12
112 115 119 117
11
7
6
12
11ii 109 112
14
10
18
113
19
117 126
14
11
128
8
160
140
120
110
170E
180
17CW
150
10
11
130
116
7-year average values and standard deviations of the effective back radiation. May 1951-1957. (cal/cm 2 /day)
7-YEAR AVERAGE AND STANDARC DEVIATION OF QB
JUNE 1951-1957
55N
67
ii
15CN
45N
20N
71
79
19
72
19
68
15
68
8
731
26
4h
91
50
6
50
7
51
6
54
8
60
10
62
8
63
8
64
6
67
8
65
7
62
4
67
3
74
11
78'
13
75
-16
67
15
64
11
-61
7
55
3
52
4
53
5
60
5
64
6
63
9
62
9
64
9
66
8
65
6
66
7
71
8
77
10
84
6
83
9
79
9
74
8
72
4
70
4
67
1
65
4
71
5
75
6
73
6
71
8
74
8
77
7
79
6
82
.7
85
6
86
9
92 1001
6
12
82
8
83
7
80
6
80
6
81
3
80
6
79
7
83
4
88
6
89
6
89
6
93
5
95
7
97
5
98
6
97
6
95 100 109
7
7
12
95
7
96
5
97
5
96
5
97
5
99
9
99
5
98 102 105 104 1C06 106 104 104 104 107 111 110
9
7
6
8
7
3
4
7
7
9
9
6
73
10
76
5
82
3
84
5
89
7
72
6
81
4
89
5
92
4
96 102 105 106 107 107 107 105 103 106 109 108 109 108 104 104
5
6
7
9
8
5
4
2
7
6
8
11
7
4
8
11
82
3
89
4
97
5
99 100 101 101 102 103 103
4
3
2
5
7
8
5
130
Figure A29.
140
150
160
170E
98 100 104 108 107 111
8
8
6
5
11
12
180
170W
160
140
1-6
104 103 103 104 114
10
8
14
8
P
114 110 104 102 105 103
13
9
16
20
16
14
150
90
9
130
7-year average values and standard deviations of the effective back radiation. June 1951-1957.
102
11
6-110- 114
9
120
(cal/cm2/day)
.
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF QB
JULY 1951-1957
55N
5N
45N
68
18
68
18
71
6
51
6
51
6
47
6
49
11
48
6
48
6
49
9
52
7
53
5
55
6
57
7
60
7
65
6
75
11
,
55
4
55
4
51
4
50
6
49
4
50
6
48
6
49
5
50
5
53
6
55
6
58
4
63
5
64
5
72
5
84
7
94
11
77
8
73
4
71
7
67
7
63
9
66
9
69
7
70
6
73
6
76
9
78
9
80
9
82
9
82
7
84
9
82
7
83
4
88
8
97
11
81
5
81
6
78
9
76
10
79
11
87
7
92
8
93
8
97 100 1CO 102 103 102
6
6
8
7
10
10
99
12
93
11
87
6
87
10
93
15
90
8
89
9
88
11
90
9
95 100 105 107 108 109 109
8
8
9
7
4
6
10
113 115 109
8
9
8
99
8
90
7
87
3
87
4
89
12
98 106
11
102 102 102 101 103 106 108 110 111 118 120 109
13
9
10
9
8
8
6
6
9
8
7
7
98
7
92
6
89
6
87
6
P8
15
98 10
12
8
98 101 101 10C2 108 112 105 100
4
7
6
5
6
5
3
5
98
9
94
13
91
7
91
11
99 103
7
7
82
6
87
6
83
"
9
90
8
95
8
95 100 102
6
9
11
88
9
94
7
98
5
98 101 102 102 100
4
5
7
7
8
Figure A30.
66
15
58
17
77
6
140
62
16
63
-21
7C
7
130
49
52C! 8
S0 14
14.
150
160
99 100
6
7
170E
95
7
180
170W
160
150
140
130
120
7-year average values and standard deviations of the effective back radiation. July 1951-1957. (cal/cm2/day)
97
5
110
7-YEAR AVERAGE ANC STANDARC DEVIATION OF QB
AUGUST 1951-1957
55KN
,3 590-65
6 - 7
10
73
19
83
21
86
20
83~
20
V
50N
4' N
89
7
56
10
55
10
54
7
56
8
56
7
55
5
57
5
60
5
61
7
62
6
68
8
73
7
74
8
811
8
"
84
68
20
60
8
61
9
6C
10
61
9
64
11
64
9
63
7
60
3
60
3
65
8
69
11
70
7
74
7
78
6
79
5
88
10
4
3
92
12
87
7
83
5
81
3
79
7
81
6
84
8
86
6
88
5
87
4
87
5
92
10
97
13
96
8
96
2
94
6
91
8
96 103
9
14
92
90
8 -'9
92
9
94
7
95
8
94
7
93
7
95
5
96 100
4
6
104 106 107 110 111 109 107 103
9
8
6
5
7
6
6
8
98
8
98 102 1 7
5
7
5
99 103 103 102 102 104 109 111 112 114 113 109 104 100
6
6
5
3
3
6
7
6
5
5
4
6
9
9
96
14
20N
68
11
86
10
85
5
85
2
89
4
94
5
95
6
95
6
79
5
81
6
85
2
92
4
97
4
96
4
96 103 106 104 103 100
5
5
5
5
5
4
99 104 110 112 117 114 108 102
6
7
8
8
7
6
9
7
79
9
80
9
85
9
91
9
93
9
94
5
97 100 100
5
4
4
94
7
130
Figure A31.
140
150
160
99
6
170E
96
8
94
5
180
98 104 115 118
7
8
10
12
170W
160
113 110 102
10
9
5
150
140
96
20
93
7
94 105 110
6
4
95
11
89
12
92 102 106
10
6
7
95
7
94
9
97 102 103 100
11
8
10
9
130
120
7-year average values and standard deviations of the effective back radiation. August 1951-1957. (cal/cm2/day)
110
7-YEAR AVERAGE
AND STANDARC DEVIATION OF QB
SEPTEMBER 1951-1957
55N
5CN
45N
85
28
109
10
20N
91
16
92
10
90
10
7C
79487
1iI20
22
88
21
84
22
87 105 116
21
22
26
87
19
82
20
74
10
75
9
75
8
75
8
79
12
81
12
82
8
83
9
87
12
98 107 11413
12
16
87
8
84
8
82
9
85
8
89
5
86
6
86
7
91
7
92
9
92
8
93
8
99 106 109 104
8
11
4
19
109 107 107 106 105 104
14
7
6
5
7
10
103 104 106 104
10
7
7
6
108 110 111 114 116 116
9
5
4
4
6
10
115 115
11
10
106 102 101 103 104 107 110 114 115 115
7
6
6
6
6
7
5
3
8
10
C105108 107 105 104 109 111 111 11
7
7
8
6
5
6
8
6
15
116 116 117 116 114 116 117 116 113 110 110
10
8
6
4
7
6
5
7
7
8
13
117 118 119 120 115 114 120 121 119 113 105 105
5
8
6
5
4
5
5
5
8
14
12
13
89
10
93
6
93
6
95
5
97
6
99 103 103 106 107 108 114 116 118 118 115 115 119 118 114. 109 105
9
9
6
6
6
5
5
7
5
3
5
4
4
6
8
14
13
91
9
90
7
89
5
92
6
93
7
97
7
130
Figure A32.
140
150
99
7
98
5
160
117
105 111 113
15
10
11
99 101 101 101 103 105 105 109 112 112 113 115 108 107 105 107
5
4
7
7
9
9
7
7
10
15
11
9
10
9
14
ii
170E
180
17CW
160
150
140
130
7-year average values and standard deviations of the effective back radiation. September 1951-1957.
123
11
106
10
120
2
(cal/cm /day)
99
9
110
C::z
7-YEAR AVERAGE AND STANDARD DEVIATION OF QB
OCTOBER 1951-195
55N
101 104011 109 100
20
20
15
17
13
5CN
108
18
45N
35 132
S13
121 112 110
12
12
9
104 10i
18
15
98
11
97 102
11
Ic
106 104 105
104 106
8
13
13
11
8
95
18
97 110
18
17
10C5 108 1C09 107 100 101 113
9
8
8
5
5
7
11
120
17
112 112 110 112
115 118
8
9
7
6
109
3
107 111
3
10
10
5
155 1 3 133 132 125 122 116 112 111 112 111 113 116 116 115 115 117 117 117 121 122
14
2
10
8
12
8
7
9
11
10
4
2
7
9
7
6
5
5
7
10
11
16 I
144
21
20
130 124 122 119 118 116 114 116 118
10
10
10
10
6
9
11
1C
11
141 137 126 119 117 113 111
27 20
12
10
6
9
11
25N
20N
(
119 120 120 121 119 119 122 122 121 122 125 125
7
6
9
8
4
7
9
7
9
10
10
10
114 116 118 12C 123 125 127 128 125 122 123 126 123 120 120 122 127
8
5
8
7
8
10
8
8
6
4
4
6
7
11
8
5
4
117 115 108 106 106 106 107 110 114 116 116 119 122 124 126 123 120 123 125 123 120 116 118 128 131
13
10
8
7 13 11 11 10
7
8
7
9 10
7
6
6
7
3
1
6 13
9 12 13
9
102
6
99
5
130
Figure A33.
96
5
94
6
140
95 101 104 103 104 104 104 106 108 109 112 114 113 113 116 118 120 116 114 116 116 113
10
8
9
9 10
9
9
8
8
5
7 11
14
8
8
8 10
9 13 10
7
5
150
160
170E
180
17CW
160
150
140
130 '
120
7-year average values and standard deviations of the effective back radiation. October 1951-1957. (cal/cm2/day)
110
7-YEAR AVERAGE ANG STANDARC
NCVEMBER
CEVIATION OF Q8
1951-1957
55N
118 1160117
22
5
17
50N
117 121 125 124
24
23
24
15
45N
"170 144 133
S17
13
196 192 178 158 149
16
0
9
10
13
16
185 190
22.1,20
-~
119 117 117 115 111 109 105
15
14
9
10
11
16
13
104 1081
10
13
12? 118 113
13
13
13
110 109 107 105
10
8
12
11
104 113 118
9
10
20
141 135 130 128 129 129 125 121 115 109 110 109 108
5
3
6
7
8
14
11
11
8
5
9
11
10
111 118 126
9
8
13
131 129
8
10
128 13C
9
13
-7159 148 141 132 126 124 121
6
8
9
14
8
4
7
12
129 125
11
12
122 124 123 121 119 116 118 119 118 120 125 134 1 0
12
15
14
11
7
8
12
13
9
8
8
9
154 159 154 143 136 131 127 125 124 124 123 124
15
16
12
8
9
7
8
6
6
6
1C
12
126 130
12
11
124 121 116 115 118 119 118 123 125 125 124 125
14
15
11
7
7
7
6
5
7
7
13
13
128 131 129 126 128 129 128 125 123 127 126
10
9
9
7
6
6
7
8
9
13
12
42
20N
120
11
115 118 117 109 1041
15
17
23
17
22
129 127 126 127 128 127 126 129 134 144 1 3
9
6
6
8
8
7
7
6
8
9
140 154
10
12
103 103 102 105 108 108 109 110 111 113 114 116 119 118 114 117 117 111 109 113 118 116 116 121 129
8
9
7
5
4
4
4
5
4
5 12
13
10
8 12
13
11
11
13
8
6
7
7 10
10
130
Figure A34.
140
150
160
170E
180
170W
160
150
140
130
120
2
7-year average values and standard deviations of the effective back radiation. November 1951-1957. (cal/cm /day)
130
9
110
7-YEAR AVERAGE AND
STANDARD OEVIATION OF QB
CECEMBER 1951-1957
55N
133 1290'14 113 121
23
30
23
28 22
5tN
124 122
11
12
45N
159 148 141 135 133
1
19
17
16
15
182 168 157
12
9
13
128
16
150 145 139
18
18
16
119 117
7
11
118 116
22
25
114 116 114 111 1C8 106 106
15
13
16
13
10
13
16
124 120 116
20
17
16
116 113
15
11
118"
26
104 105
18
19
109'
17
115 1C9 105 105' 100 101 111
9
7
6
15
15
14
13
134 128 125 122 118 116 113 112 112
15
14
13
12
9
6
13
7
14
109 111
12
11
121
19
118 128
10
14
Ul
9T 178 168 158
11
16
14
16
148 142 138 135 133 130
18
20
17
14
13
12
127 124 122 121 121
9
7
9
14
16
121
13
121 121 125 131
12
11
8
12
3CN
187 191 185
18
17
18
173 160 152 144 141 139
13 .11
12
11
9
10
25N
142
12
147 145
8
14
143 141 136 133 135 134 132 135 135 131 129
14
12 12
11
5
6
7
10
8
8 10
20N
119
17
118 117 118 119 119 122 127 127 126 125 122 118 117 114 117 121 119 115 114
13
8
6
9 11
11
9
8
12
12
7
5
8
10
7
7
5
8
11
130
Figure A35.
140
150
160
136 136
12
12
170E
136 133 132 130 128 130 130 130 128 126 128
10
8 10
11
7
10
9
7
9
9
10
180
170W
127 126 129 129 129 125 120 117
14
9
6
5
4
9
15
18
160
150
140
141
2
132 144 15
12
12
1
122 140
17
15
111 110 113
10
14
16
126
15
130
120
7-year average values and standard deviations of the effective back radiation. December 1951-1957.
154
13
i35 139
11
6
(cal/cm2/day)
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF QE
JANUARY
1951-1957
55t
87
*0B
50N
152 147
43
32
45N
*
94
92
107
69
91
44
118 132
70
51
6'
158 138 132 149 152 128 104 121 147 170 179 168S
30
36
56
59
53
39
34
35
35
63
96
96
'335 269 236 245 266 232 178 180 171 152 147 147 131 144
"255 115
66
37
80
65
45
34
50
33
35
50
47
39
483 419 426 435 450 402 325 289 260 215 186 191
213 73
71
35
98
94
60 46
49
34
42
48
697 656 605
60
81 1C6
115
49
170 188 186
48
67 68
171 165
60
62
187 196 205 216 215
51
38
47
50 46
180 15
68
67
563
532 481 401 346 301 259 235 226 229 232 225 219 219 183 149
116 144 120
96
82
63
48 66
81
94
72
60
65 69
72
62
522 5C5 537 563 556 535 535 670
140 14
98
93
82
87
89
87
412 405 387 367 344 334 338 306 276 275 256 234 223 201 190
1C6
96
84
86
66
68
83
97
94 90
79 73
64
39
51
154 13
30
565 512 460 423 405 417 412 161 341 367 377 398 365 366 357 323 310 296 269 257 232 238 248 225 191
115 118
76
9
100 101 112
85
76
62
74 106
71
57
64
51
73
71
65
83
56
49
77
55
32
20N
K2
467 460 418 383 381 370 351 '67 405 394
151 111
73
66
95
89 E88 85
91 1CO
130
Figure A36.
140
150
160
170E
353 344 322
112
87
52
18C
322 323 335 306 276 294 316 293 316
61
91 100
95
90 100 125 134 134
17C0
160
150
140
130
7-year average values and standard deviations of the latent heat transfer. January 1951-1957.
340 345 295 224
113 108
85
72
120
(cal/cm2/day)
110
7-YEAR AVERAGE
AND STANDARD DEVIA ICN OF QE
FERRUARY
1951-1957
55N
50
35
50N
134 133
38
42
"152
8
319
131
15
143 161
67
73
9 344 339 369
86
63
54
76
22 32 433 52
67
156 132 104
73
63
40
194 218 228 232
56
44
39
75
-69
50
84
29
54( '70
_46
47
69
17
87
11
86
43
92 113
38
58
97 106 129
24
29
41
18C 145 134 111 113 124 122 129
77
66
54 47
33
46
44 41
368 349 305 251 216
32
45
82
97
e8
596 546 493 465 453 397 347 301
70 102
84
61
59
53
29
54
193 163
68
62
149 142 131
60 63
52
275 248 229 207 185
80
85
95
85
81
182
70
135
36
130
50
141 129
44 44
1001
40
134 122 113
36
39
35
92
23
140 139 130 10
33
42
32
31
181 186 172
42
33
30
155 133 1 2
33
45
38
419 428 458 472 470 457 425 281 367 336 317 291 282 282 286 261 252 261 265 246 215 183 155 126 127
222 118
65
59
64
66
57
64
80
52
16
35
60
62
68
66 91
73
46 22
46 53
51
37
30
431 411 392 365 373 353 336 121 289 288 292 296 291 313 347 337 316 320 324 300 249 200 155 128
194 133
92
71
68 70
93
88
79
66
50
79
84
67
95 100 101
52
51
70
92
86
69 44
2
20N
331 351 384 360 342 305 290 309 305 315
93 1C7 108
70
70 62
74
89
86
78
130
Figure A37.
140
150
160
170E
320 301 282 303
79
72
71
61
18C
170W
137
50
co,
342 344 313 290 322 320 283 248 217 188 187 184
96
90
39
60 118 158 112
88
75
57
57
27
160
150
140
130
7-year average values and standard deviations of the latent heat transfer. February 1951-1957.
120
2
(cal/cm /day)
110
7-YEAR AVERAGE
AND STANDARD DEVIATION OF QE
MARCH 1951-1957
55
26
71
41
61
35
84
32
127 121 114 100
34
4C
40
26
87
24
83
27
99 105
29
31
1Co 101
71
86
150 139
73
51
45N
'95
\-70
113 145 173
68
34
34
135 133
14
19
177 178 186 17C 147 143 145 127 111 108
45
47 51
37
26
41
44 43
59
48
96 10
43
51
108 111,
31
39
121 131 138
36
35
38
127 128
23
28
207 246 291
55 46
56
289 267 255 245 220 190 174 164 152 141 143 158 165 169 157 135)
59
68
76
85
5e
52
54
62
68
66
59
49 41
46
33
341
402 402 390
107
98 109
348 307 273 243 239 224 201 187 174 169 176 190 193 195
95
77
72
74
77
62
47
54
60
66
57
50
39
35
188 154
37
35
268 3C4 288 305 338 319 298 280 266 252 240 26C 263 245 220 230 242 246 240 232 215 192 189
71
66
67 62
93 100 107
70
52 43
56
79
87 59
40
30
33
42
29
31
29
35
45
145 1"17
30
18
342 299 258 247 262 262 259 272 289 308 297 306 306 282 262 289 303 318 299 275 243 204 189 181 159
50
65
87
69
62
65
73
65
71 108 116 110
94
59
55
57
59
54
68
55
42
33
42
30
34
$
20h
9
279 274 264 270 279 276 280 293 306 337 334 351 376 364
49 60
73
59
46
61
74
73
94 151 161 113
98 116
130
Figure A38.
140
150
160
170E
18C
170W
ODc
336 313 333 342 300 289 284 249 254 245 202 186
108
83
86
77
78
75
71
44
56
62
47
47
160
150
140
130
120
7-year average values and standard deviations of the latent heat transfer. March 1951-1957. (cal/cm2 /day)
110
7-YEAR AVERACE AND STANDARD DEVIATION OF QE
APRIL
1951-1957
79
39
68
35
45NA
'4
2
36
22
63
26
89
32
98 103
43
40
107 123 154 176 178
33
34
37
40
51
264 248
79
71
126 171
53
49
199 174
52
55
Figure A39.
E8
25
92
32
93
76
78
47
67
26
60
24
6A
28
92
44
87
40
86
29
83
23
78
38
80
35
86
34
91(
46
122 130 127 113 101
31
39
48
35 42
92
42
92
35
96
30
95
37
97
48
96
39
90
25
185 190
62
71
186 178 147 133 125 125 133 134 129 121
65
60
30
16
16 22
24 25
28
30
216 209 197 185 173 159
71
74
77
72
56
34
153 161
30
31
168 166
41
37
108 105;
18
22
166 159 151 139 124
32
28
35
34
27
188 210 214 204 2CC 209 214 196 194 211 207 192 201 211 210 205 201 190 162 159 157 121
44
57
67
137 163
50
55
175
57
2)5 183 172
67 49
32
130
237 727
47
50
88 115 102
42 46
32
7
5cA80
.49 56
43
36
40
37
37
34
36
48
41
48
72
67
37
42
30
44
44
32
22
11i
27
187 213 ?39 278 255 243 270 279 271 280 275 259 249 243 227 195 190 189 163 149
74 87
73
59
39
45 38
48
57
61
58
57
52
60
52
67
71
61
37
27
208 233 241 239 ?66 286 279 274 299 289 284 290 306 305 306 301 253 224 255 273 249 207 171
27
32
42
49
53
46
46
54
59
70
66
68
76
90 107 108
74 66
59
42
37
36
36
140
150
!60
170E
18C
170W
160
150
140
130
120
7-year average values and standard deviations of the latent heat transfer. April 1951-1957. (cal/cm2 /day)
110
7-YEAR
AVERAGE AND STANDARD DEVIATION OF QE
MAY
1951-1957
K/o
47 46 "V47 42
16
10
22 ,22
42
10
41
16
44
22
5f
14
V
45N
v- 27
6
40N
66
36
35N
9
51
30N
79 128
69
58
6
34
E9 11150
54 44
36
151 147
33
53
25N
163 147 139 144
51
49
62
65
20N
158 147
26
27
13r
Figure A40.
90
23
52
29
65
26
63
17
63
10
67
11
62
10
60
8
55
13
42
10
39
8
45
8
52
9
60,
11
62
23
75
23
88
28
79
26
85
21
83
8
81
19
70
25
76
37
27
13
41
20
83
23
75
32
60
32
58
18
66
18
87
19
96 108 108 111 111 104 101 106 113 113
28
27
27
28
33
32
38
39
36
26
97
33
95
33
93
17
98 100 104 10
17
24
20
27
60
17
186 172 156 136 138
37
53
60
48
39
130 122 114
39
38
37
145 147 144 135 143 157
49
57
53
30
26
33
147 152 168
48
59
66
17!, 211 223 228 234
47
62
54
51
42
141
38
24
150
169
32
S9 104
29
31
126 136 144 143
57
38
32
32
140 143 137 138
21
22
22
31
157 146 157 187 210 205 198 193
26
18
23
31
33
27
28
37
161 160 194 233 241 235 246 264
41
20
26
39
42
52
74
78
182
41
83
39
128
29
16
9
167 151 145 129
48
57
26
19
290 268 266 268 236 209 194
31
45
59
66
59
60
84
102
24
185 162 131
80
28
40
'12 210 234 256 267 250 252 245 253 283 306 311 280 242 259 256 216 208 205
36
48
52
61
57
50
66
80
54
25
40
63
59
76 104
91
44
48
25
160
170E
18C
170W
160
150
140
7-year average values and standard deviations of the latent heat transfer. May 1951-1957.
130
120
(cal/cm2/day)
110
7-YEAR AVERAGE AND STANDARD DEVIATION CF CE
JUNE
1951-1957
$
20N
32
21
47
7
6?
31
28
7
27
11
23
7
22
13
24
11
19
11
22
15
29
19
29
17
35
16
50
14
62
13
67'
19 -
28
12
34
11
31
13
28
12
31
14
40
14
36
14
30
16
26
17
25
17
26
19
35
15
46
15
56
17
70
14
68
16
5
2
64
22
77
20
84
19
79
25
69
23
75
32
81
27
75
26
62
22
46
24
45
26
47
21
56
17
74
16
88
11
99
10
89
19
6,
2'
119 113 115
37
37
36
123 128 114 115 109 106
26
38
45
50 39
37
103
25
89
23
90
20
96 105 112 121 128 125 103
22 15
17 19 16
18 28
115 121
34 23
146 144 136 129 129 121 139 145 144 162 176 182 179 177 164 138 108 106
18
26
25
13
23
47 31
33 42 40 22 16 25 23 25 35 25 25
51
18
97 102
23
31
28
27
29
13
-1
S4
36
32
31
18
14
12
-11
79
63
'32
26
141 139 126 126 145 158
47 17
18
31
32 29
126
19
158 156 164 168 170
28
12
23
27
18
87
20
179 176 196 2 19 224 243 276 285 278 245 191 157 138
24
50
34
28 47 38 42 51 59 52 48 71 58
118
29
0,011
231 225 212 212 212 212 207 215 217 219 207 225 248 265 253 275 322 326 327 330 289 251 236
35
17
27
32
32
33
25
12
10
26
36
34
28
50
52
66
80 100 103 91
51
46
86
130
Figure A41.
140
150
160
170E
18C
170%
160
150
140
7-year average values and standard deviations of the latent heat transfer. June 1951-1957.
130
190 171 149
46
21
26
120
2
(cal/cm /day)
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF QE
JULY 1951-1957
h,
23
15
15c-31
18
19
44
18
55
21
66
42
49
43
4M
59
0
50N
45N
41
63
3'
79
5
5 5274 w85
4 8753
29
4
6
7
5
12
8
2C
15
25
21
17
12
15
13
21
13
25
11
31
10
36
17
38
20
51
28
64
23
79
17
20
18
15
11
17
9
17
11
11
10
10
13
17
16
16
9
18
13
17
19
29
17
39
18
44
5
47
12
54
15
67
15
55
29
SC
23
46
13
43
16
41
24
38
30
39
30
46
25
58
24
61
22
74
27
90
22
98
13
97
11
94
15
95 102
23
26
103 104
27 22
94
26
85
27
92 104
26
30
70
27
106 112 125 132 141 143 156 162 148 131 129 132 118
32
30
30
32
32
34
29 26
23
26
30
31
33
97 116 138 152 167 168 163 163 169 183 195 209 212 221 224 236 230 223 198
63
62
28
24
24
30 40
27
18
28 45
49
44 37
32
37
37 33
38
223 221 222 222 224 239 263
27
18
35
36
35
56 68
:2
20N
87
37
38
176 156 147 128
34 40
34
45
92
34
7
236 209 227 276 274 281 300 309 310 309 287 267 247 210 179 134
49
46
61
61
46
54
44 45
39 48
50
60
53 47
22
64
80
45
69
32
249 260 258 253 233 250 264 254 260 261 256 272 284 280 283 297 301 282 280 275 250 224 164 122 113 137
43 47
51
31
22 32
40C 50
51
49 42
34
42
40
59
66 54
28
47
56 41
22
33
37 38
36
130
Figure A42.
140
150
160
170E
18C
170W
160
150
140
7-year average values and standard deviations of the latent heat transfer. July 1951-1957.
130
120
(cal/cm2/day)
110
7-YEAR AVERAGE AND STANDARD DEVIA
ION OF QE
AUGUST 1951-1957
55N
50N
45N
V
60
44
53
69
5
20
25
25
18
25
33
28
30
27
27
23
25
16
28
11
28
8
27
12
30
20
44
18
48
18
43
13
51
12
59
14
66
17
62(
12
38
26
50
27
51
28
43
19
43
18
37
26
47
18
57
24
63
20
64
17
68
9
75
7
85
9
65
12
102 110 111 111 103
20
19 21
14
7
99
14
98
13
76
24
39
28
35
15
40
19
40
23
146 1 8 118
79
71
48
99
33
99
25
99
36
94 106 122 118 113 109 100
32
37
54
5C
33
18
17
173 162 164 175 172 176 178 170 161 151 150 147 146
33
23
17
35
42
4C
25
32
48
48 41
22
12
206 213 227 223 216 220 211
104
54
52
38
41
41
37
$
33 c59
14
62
52
118 123 163 182 190 186
107-J848
52~ 54
55 45
20N
27
142 130 119 108 113
18
23
26
28
24
215 215 209 206 205 204 217 213 204 209 212 200 186 178 173 162 138 113
30
22
22
31
29
24
51
68
54
35
33
26 44
39
29
36
32
3
262 3C4 311 263 250 270 256 241 231 227 253 267 261 287 306 312 323 301 282 264 249 220 204 169 132
72 1C5 125
78
53
71
66
45
25
27
50
74
65
5i
57
42
38
47 59
59
31
22
37
31
50
329 346 33Q 288 276 273 264 248 251 245 254 298 299 300 311 344 366 351- 333 328 288 253 237 205 181 184
95
94
82
61
48
52
55
37
33
36
75
5C
54
52
47
56 66
74
85
54 54
72
40 27
43
31
120
Figure A43.
140
150
160
170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the latent heat transfer. August 1951-1957. (cal/cm2 /day)
110
7-YEAR AVERAGE AND STANDARD
SEPTEMBER
DEVIATIOf
CF QE
1951-1957
)
'k
76
46
83cJ83
58
57
86
42
80
47
88
41
107 12j
29
34
91
34
95
35
95
24
95
37
91
40
102 129 106
28
36
37
121 119 116 111 112 111
31
40
36
39
37
26
120 126 111
13
18
27
V
50K
69
37
45h
79
9
60
36
86 103
34
34
76
27
88
16
89
16
90
26
110 116 129 147 140
37
29
41
33
33
96
31
184 185 210 212 205 221 227 235 224 208 204 184 174 174 164 156
37
58
62
53
57
65
69 48
30
35
42
29
36
35
23
13
314 289 278
104
99 84
30N
369 325 296
126 72
4C
25N
321
5
20N
93
305 309 277 280
55
88
52
50
65
30
144 120
20
18
78
16
280 296 306 293 285 283 270 246 239 224 200 192 167 158 128
65
65
68
76
68
61
54
52
48
37
39
36
26
33
14
96
22
299 313 300 281 262 260 246 236 243 230 214 203 180 154 136 121 121 131
66
65
45
52
43
51
41
33
53
57
35
32
38
30
38
37
21
22
285 250 249 256 273 285 285 288 275 257 248 254 258 257 264 258 242 246 213 185 175 154
57
43 51 40 46
7
87 65 38 34 12 22 25 60 71 52 30 33 50 52
58 45
301 267 25C 259 254 270 270 253 242 248 253 269 278 264 248 266 268 249 269 267 245 256 256
49
24
31
38
56
48 44
44
24
40
52
64
53
44 41
42
80 99
82
67 45 45
50
130
Figure A44.
140
150
160
170E
180
170W
160
150
140
130
141 139
22
38
195 203 203
61
38
41
120
7-year average values and standard deviations of the latent heat transfer. September 1951-1957. (cal/cm 2 /day)
110
7-YEAR
AVEFAGE AND STANDARD DEVIATION OF QE
OCTCeER
1951-1957
55N
195i172 169 159 144 132
139
78
86
69 73
67
50N
150 138
33
36
45N
'126 114
L-37 33
142 188 208
38
29
19
274 265 291
64
52
51
545 475 418 402
174
73
40
56
20h
$
136 15C 152
39
23
36
169 184 190 190 184 176
42
40
32
42 42
30
199 210 207 200 209 218 214 207 205 212 210 200 154 1CO
31
44
36
33
4C
59
40
49
35
32
40
54
43
30
?02 289 2E8 290 298 286 282 282 265 255 243 239 218 192 148 10
32
33
57
65
55
37
49
91
70
43
43
17
28
41
38
34
470 438 390 '39 3C8 308 303 300C 305 316 314 325 300 266 242
96 76
71
65
77
84
73
49
29
58 100
85
52
41
39
218 182 162 126
36
30
38
23
376 343 321 309
40
37
54
51
214 203 198 183 151 145\
23
28
23
22
19
301 304 313 315 312 319 331 312 280 256 236
62
70
76
6C
47
58
73
53
47 24
22
546 443 356 317 283 276 279
106
64
53
35
48
48
41
287 320
45
89
396 323 317 332 290 289 289
81
46
56
41
49
28
50
?GO 332 325 355 358 332 320 317 310 279 268
67
66
64
70 8C
70
71
90
80
64
62
130
Figure A45.
140
175 174
39
38
150
160
332 349 370 344 339 342 300 270 262 250 232 240 222 191 190 200
94
90
87
70
68
61
52
52
33
30
33
42
40
46
35
42
170E
18C
170W
160
150
281 275 276 254 246 240 234 208
88
58
37
35
50
46
57
56
140
130
120
7-year average values and standard deviations of the latent heat transfer. October 1951-1957. (cal/cm2 /day)
110
7-YEAR AVEPAGE
AND STANDARD DEVIATION OF QE
NOVEMBER
1951-1957
192
55N
151 141 124
30
22
28
99
227
85
50 N
45N
O '
219 214 216 213 186 165
74
61
42
92
93
76
168 159 158 168
21
24 28
41
'267 227 251 ?01 301 273 294 289 259 221 208 209 187 191
-65
35
43
57
71
67
76
95
80
65
63
45
38
51
422
73
130 1
47 1
166 169
50
50
205 191 184 178 120
73
61
65
57
40
397 450 453 404 368 357 361 343 301 281 260 235 236 233 215 203 172 11
43
58
36
50
41
44
81
85
68
54
33
22
38
44
48 56
33
16
613 605 564 501 454 410 357 351 361 348 321 299 277 276 259 233 207 181 143
83
36
63
87
99 105 109 108
94 77
50
42
62
50 40
22
26
54
56
457 494 514 536 531 487 439 403 388 373 355 329 322 334 325 309 295 287 282 262 246 208 184 161 164\
97
88 40
35
88 78
73
64
46
37
93
96
84
72
61
59
54 54 47
35 16
20
30
54
68
569 493 448 395 380 367 327
138 98
69
42
54
73
84
20K
2
317 319 334
52
38
51
357 332 322 321 309 290 290 314 316 306 309 272 227 225 216
92
83
82
63 53
63 62
38
33
50
40
51
51
93 102
-0%
438 391 361 319 335 344 340 ?40 330 348 359 351 341 303 283 312 315 290 293 317 330 317 298 288 263 236
64
92 62
36
56
66 70
59
57
68
85
54
60
41
59
58
35
23 42
66
87
73
78
93 69
32
130
Figure A46.
140
150
160
170E
180
170W
160
150
140
130
7-year average values and standard deviations of the latent heat transfer. November 1951-1957.
120
2
(cal/cm /day)
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF QE
DECEMBER 1951-1957
V
192 179
59
45
45SN
156 1446.135 129 140 196 216 199 'j
69
88
82
99 105
89
87 89
a'
168 182 183 167 151 156 169 180 206 202 192 184,"
49
4C
45
39
36
47
64 71
77
73
64
52'-.
255 228 270 266 238 239 248 228 204 183 181 185 184 206 208
8
64
89
75
60
77
94
62
49
39
37
44
50 72
85
41
1
93
67
80
86 103
112
98
1CC
87
65
60
68
83
94
84
197 200 189 159
83
72
38
39
72
64
32
12
41 494 549 596 653 625 567 509 479 440 389 353 330 315 290 273 263 279 274 247 218 182 142 132
95 71 90
90 153 154 139 154 190 165 117 11C 115 86
41
58
89 108
98
74
72
40
25
34
537 542 534 542
96
84
81 123
543 522 466 '35 451 431 384 368 358 346 341 341 338 336 303 277 238 191
145 132 146 143 121 101
90
84
74
59
54
45
84 90
70
69
61
38
604 569 527 479 470 428 367
197 93 111 124 135 105
96
20N
170 158 1
38
46
55
176 406 385 378 386 392 372 361 375 396 367 339 293 253 216 216 225 210
75
63
52
44 74
85
96 100
89 128 105
90
83 100
81
79 111
86
544 5CO 484 455 417 389 368 ?86 405 384 366 368 358 329 296 338 380 347 323 315 315 305 305 298 278 264
95
SO 109 103
81
79
62
74 138 134 109 106 117 140 146 133 144
98
90 103 122 109
91
80
67
50
130
Figure A47.
140
150
160
170E
18C
170W
160
150
140
130
120
7-year average values and standard deviations of the latent heat transfer. December 1951-1957. (cal/cm2/day)
110
7-YEAR
AVEPAGE AND STANDARD
JANUARY
DEVIATrIO
CF CS
1951-1957
7
$
90
45
74
23
130
Figure A48.
6;
20
61
31
92 112 1711
51
59 100
98
52
81
48
81
50
68
55
53
55
47
45
34
27
34
23
53
26
79
32
98 103
41
43
250 256 207 172 131
82 107
65
48
45
95
34
83
27
71
26
52
23
42
26
42
32
33
34
38
28
54
37
67
44
73
47
70
42
282 292 294 257 216 163 129 109
64
57
65
44
51
34
24
23
93
26
70
26
56
23
50
26
43
38
44
32
53
31
59
37
59
34
49
37
381
2B
98
27
83
25
72
24
63
28
51
30
44
34
43
33
45
33
47
29
47
28
35
30
24
25
79 301 284 244 20'
47
38
51
55
64
164 132 105
72
45
26
78
69
130 116
63
52
45N
231 198 179
73
40
47
85
6 c 93
81 .73
94
168 148 135 122
41
28
26 2C
I'
80 142 112
73
45
30
99
19
86
24
81
21
75
20
71
18
67
13
60
14
59
23
52
16
45
17
45
23
42
22
41
25
40
24
37
18
37
15
25
12
'19I
81
21
69
22
67
21
60
21
51
16
46
9
47
10
49
20
56
23
55
17
47
13
40
13
40
7
39
10
38
12
36
17
38
24
35
21
38
15
42
10
35
13
26
13
47
11
41
12
40
10
39
13
39
13
44
13
42
10
38
18
41
16
39
12
32
10
28
11
28
12
20
15
25
15
35
19
35
23
29
24
36
14
43
10
45
13
37
10
143
150
160
170E
18C
170W
160
150
140
130
120
7-year average values and standard deviations of the sensible heat transfer. January 1951-1957. (cal/cm 2 /day)
23
5
110
7-YEAR AVEiAGE AND
STANDARD DFVIATION OF QS
FEBRUARY
1951-1957
45N
48
58
60
36
73
30
82
45
51
40
48
34
36
25
29
26
26
35
28
28
47
28
52
2B
47
36
45[Z
33
'154 150 193 181 156 140 126
-74
45
58
32
52
57
73
79
56
51
40
35
28
19
18
18
21
21
25
23
23
27
18
31
18
33
26
36
22
3i
2
216
193 161 132
39
50 62
96
54
70
41
50
28
36
25
32
25
30
26
24
25
26
21
30
19
32
I7
32
13
2'
1i
151 122
39
30
93
19
8C
27
69
31
54
26
46
27
38
24
33
24
33
25
33
18
34
16
33
10
31
9
24
11
54
31 251 227 187 163
36
43
44
37
41
20h
t58
53
67
38
252 238 237
74
39
61
148 1-6 149 133
'68 43
28
21
37
89
43
105
43
120 113
19
20
94
49
21
9C
17
75
14
67
17
60
12
53
4
48
12
44
18
45
17
47
22
42
24
43
26
41
22
38
13
37
7
36
8
32
13
23
16
11
14
131
62
96
40
76
28
64
20
60
15
57
20
47
21
38
14
33
13
33
9
35
6
36
13
35
15
41
14
49
20
45
22
42
18
38
13
35
15
37
10
32
10
30
16
23
21
9
22
8
20
49
24
43
45
19
39
13
36
10
32
12
25
11
27
9
27
11
28
11
27
9
25
6
22
9
27
9
36
13
34
13
926
22
12
26
19
30
17
26
18
27
18
26
19
17
18
18
18
C
19
130
Figure A49.
140
150
160
170E
180
17CW
160
4
150
140
130
120
7-year average values and standard deviations of the sensible heat transfer. February 1951-1957. (cal/cm2 /day)
15
9
110
7-YEAR AVFRAGE AND STANPDRD
DEVIATICN CF CS
KARCH 1951-1957
%
57
a
S03
-34
41
34
44
50
41
70 104 13cA
55 112 152
96
57
84
38
71
10
52
13
34
22
35
24
33
23
25
20
23
25
27
22
26
17
40
16
59
42
99 116
35
32
98
36
74
37
65
45
46
31
36
12
31
12
28
12
22
15
18
21
21
20
21
16
26
17
32
21
36
18
115 131 147 138
38 46
36
28
110
35
89
41
72
44
53
28
42
14
32
17
27
20
25
22
22
23
28
21
34
20
35
17
36
19
31
14
22
16
67
55
0
5214
ee
57
2r113
137 147 141
57
38
51
52
126
4
99
0
3C
81
25
63
23
46
19
43
2C
42
20
35
19
31
20
29
20
27
18
31
18
36
18
35
13
32
13
31
12
21 \4
10
11I
87
33
8
23
73
21
69
23
70
29
65
27
55
27
42
12
36
9
35
15
34
20
37
22
34
21
31
17
31
14
33
13
33
12
35
10
33
5
33
6
30
7
26
9
28
16
18
11
76
14
56
12
39
11
31
12
32
12
30
8
25
12
19
10
18
9
29
19
34
26
36
25
32
23
28
17
30
15
33
14
34
7
36
7
33
13
31
13
24
9
22
9
28
21
22
15
13
10
36
9
29
7
25
7
22
8
19
6
18
8
18
10
15
10
14
12
21
15
30
23
34
24
32
23
33
17
28
11
27
9
28
10
31
14
31
20
29
18
26
13
24
13
32
20
31
15
22
9
130
Figure A50.
140
15,
160
*170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the sensible heat transfer. March
1951-1957. (cal/cm 2 /day)
17
9
110
7-YEAR AVERAGE
AN\
STA;NDARO
DEVIA
7
ICNF OF CS
APRIL 1951-1957
55N
'o
50 N
45N
'-17
15
16
19
8
14
42
12
27
13
10
15
10
221
21
17
20
21
19
27
26
17
19
10
17
17
21
21
24
15
23
8
24
11
22
10
20
9
20
12
20
-3
18
11
25
22
25
22
25
18
24
26
20
29
12
20
21
14
23
16
25
15
22
13
23
15
20
12
15
10
19
9
17
7
10
1
12
16
21
17
24
29
23
41
23
44
19
39
19
39
24
43
23
4C
18
28
19
19
17
19
11
18
13
18
16
19
16
16
12
14
15
15
14
12
10
5
8
24
51
29
45
24
4C
17
36
11
36
16
37
22
35
20
3C
15
24
10
19
6
19
7
21
8
25
11
23
11
20
10
19
13
20
14
18
12
13
9
II
5
27
7
24
12
22
13
20
12
17
8
16
8
20
13
19
14
19
12
24
7
25
8
2C
8
20
10
22
12
24
11
24
8
22
9
21
7
20
10
21
11
19
10
10
6
8
1833
14
-15
2330
18
14
'17
45
y
32
12
18
8
7
5
7
11
10
14
12
14
10
8
11
7
16
13
13
17
12
10
18
IC
22
14
21
14
22
14
21
14
22
10
22
6
20
7
21
11
23
11
22
9
19
13
7
14
2
13
14
8
5
5
3
5
7
6
10
8
9
8
9
7
10
12
11
22
10
17
9
11
14
E
14
'12
14
11
20
6
22 22
5
5
20
7
18
11
24
17
27
15
25
10
25
10
21
12
15
11
K2
120
Figure A51.
140
150
160
170E
18C
170W
160
150
140
130
120
7-year average values and standard deviations of the sensible heat transfer. April 1951-1957. (cal/cm2/day)
10
7
110
7-YEAR
AVERAGE AND STANDARD DEVIATICN OF
MAY
1951-1957
\i
V
55
50N
0
V -22
-3
10
-6
8
-7
9
-2
12
-1
19
-e -14
13
7
-8
7
-6
13
-8 -10
10
9
-9 -13 -11
8
11
8
-2
5
-7-11
9
-4
7
-9
26 14
-4
19
1
18
-1
15
-3
13
1
11
200
12
19
C
19
3
21
-0
12
-3
9
-9
7
6
5
-1
6
12
8
6
19
-3
19
8
9
14
16
16
18
15
14
12
9
11
10
12
12
11
18
8
19
11
20
8
15
7
9
4
8
-0
8
0
5
2
6
1
2
-6
7
4 2 - -3
14-i'
15
24
9
24
13
21
16
18
15
18
12
14
9
1C
IC
C10 8
10
8
10
8
13
9
13
9
15
12
15
10
12
5
10
5
9
6
8
8
6
8
6
8
3
t7
V
1
4
-7 -I Al
-12
2C
20
16
15
Q,
-15 -11
20
18
45N
20NK
CS
8
15
11
8
11
9
12
12
11
13
10
12
9
8
9
5
10
4
10
3
11
4
12
4
13
4
15
4
17
9
15
10
15
8
18
6
17
7
15
7
16
5
17
6
18
9
12
5
14
14
10
9
6
8
5
7
4
7
4
9
4
8
2
6
3
6
6
4
12
6
14
5
14
5
19
9
20
16
15
12
13
7
18
8
21
12
21
11
21
9
25
12
27
14
21
7
11
10
6
5
4
5
4
6
6
5
6
7
6
6
5
4
-2
7
-3
11
3
10
5
8
11
14
10
16
8
12
8
9
9
8
13
5
20
9
22
14
19
12
21
14
28
9
33
10
28
10
23
11
130
Figure A52.
140
150
160
170E
180
170
160
150
140
130
120
7-year average values and standard deviations of the sensible heat transfer. May 1951-1957. (cal/cm2/day)
2
17
6
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF
QS
JUNE 1951-1957
-5
-1
7
13
12
5
10
-9 -10 -12 -12 -14 -13
7
9
6
5
9
10
-9
10
-1
5
3
7
-5 -11 -14 -12 -13 -18 -19 -18 -18 -18 -14
9
9 10 12
8
8
8 10 12 13 11
-9
9
-5
9
-8
tj ,
17
-11 -13 -13 -11
15
13
10
7
-14
37
-8
7
-4
11
-3
11
-5
11
1
9
3
13
5
10
3
9
-2
7
-6
9
-5
11
-3
9
6
8
5
5
2
5
3
6
4
8
1
10
-3
12
-2
11
3
10
1
7
2
5
1
3
4
4
5
5
4
2
1
1
2
4
5
4
5
6
7
9
8
-1
9
0
8
-0
11
4
9
45KE
-3
19
1
11
-c
2
4
3
11
4
4
4
4
3
6
4
7
4
7
5
7
5
6
4
3
3
C
5
7
4
7
4
8
5
P
4
7
- 4
6
4
5
6
6
7
6
7
4
6
-2
12
-4
14
130
Figure A53.
3
6
140
150
1]6C
170E
18C
21
8
-8 -13 -12 -10
8
9
8
7
170W
61
5S
6
-4 -413
6
4
-8
6
-4
8
-1
8
1
6
-4
7
2
12
-3
7
-2
4
-1
6
2
5
5
6
6
6
8
4
5
6
-9
17
1
00
7
10
10
12
13
11
6
5
10
8
7
4
4
7
7
7
8
10
12
1
13
2
14
10
8
15
6
18
12
21
18
20
17
10
17
9
18
9
16
3
14
5
15
14
19
23
19
23
18
25
17
32
22
29
20
17
18
16
19
19
11
17
6
160
150
140
130
120
7-year average values and standard deviations of the sensible heat transfer. June 1951-1957. (cal/cm 2 /day)
\1
10
5
110
7-YEAR AVEPAGE AND STANDARD
JULY
DFVIATICN OF CS
1951-1957
55N
-8
12
1
9
5
7
5
11
-8 -10 -14 -16 -14 -13 -12
16
8
8
8
5
4
6
-8
6
-4
4
-0
6
-7
10"ll -14
11 .,11 11
14
1
50 N
-4
6
-6 -12
6
7
-
45h
6 14
6
7
8
8
7
11
7
-6
7
-7
9
-8
12
-'
12
1 -j0
7 11
-5
18
-6
18
-5
13
-3
10
0
8
7 -12
109 12
-4
12
-4
7
-2
9
-1
8
-0
6
-0
5
-2
10
10
6
8
4
4
3
5
6
-8
10
-7
8
-8
4
-5
5
-2
9
2
7
4
5
3
4
5
5
-2
5
C
1C
3
9
2
6
00
7
3
6
7
6
8
7
9
8
7
6
6
5
3
5
-3
10
2
10
8
10
42
7
-8
7
-4
7
1
5
4
7
5
4
4
7
4
9
5
5
6
5
7
7
8
9
9
e
7
8
4
9
5
8
11
5
13
7
11
6
8
5
7
5
8
6
9
4
8
10
4
5
6
5
5
11
5
15
7
13
lI
12
7
14
8
18
13
14
7
11
4
10
4
10
7
11
7
8
7
6
7
11
3
19
8
21
9
16
6
14
8
17
9
15
10
15
7
14
15
6
6
-2
6
13
3
17
5
18
7
15
6
13
6
14
6
16
12
13
13
10
8
10
5
5
14
9
5
4
9
1
10
3
R
8
7
15
11
12
12
11
20
20
13
21
15
20
9
17
5
10
7
5
7
42
1i0
Figure A54.
140
150
160
170E
18C
170W
160
150
140
130
120
7-year average values and standard deviations of the sensible heat transfer. July 1951-1957. (cal/cm2/day)
5
4
110
AVERAGE
7-YEAR
AND STANDARD
AUGUST
DEVIATICN
CF QS
1951-1957
55N
1
-3
8
12
3
10
7
6
7
14
"V
50
5
!3
-5
0
5
-6
8
-10
6
-8
-7
6
6
-8
-3
-5
5
-t
4
5
-3
8
-1
8
-3
9
-2
7
4
6
5
8
2
10
-4(
10
-7
6
6
5
-4
5
-2
6
-2
7
00
6
2
6
4
6
5
6
3
7
-1
6
-0
7
8
6
8
6
5
4
7
-1
11
-1
8
1
5
3
9
3
9
5
6
5
6
4
5
7
6
7
4
8
2
8
2
6
3
O
6
12
7
11
7
8
7
7
6
6
8
9
5
9
5
9
5
10
5
10
5
8
4
7
4
8
3
7
3
7
4
7
3
3
-15(
6
5
3
1
6
-1
8
8
8
5
5
9
8
11
6
12
6
13
5
11
4
12
4
12
2
11
3
9
4
8
5
6
5
8
6
8
6
5
5
7
3
10
3
10
3
8
3
7
7
9
10
16
13
17
7
4
5
12
9
18
17
16
16
18
14
22
14
18
12
15
10
13
7
11
8
11
10
8
10
6
12
8
12
13
12
11
7
17
7
21
6
22
3
23
8
20
14
21
13
24
13
16
11
2
8
21
24
15
21
11
21
11
21
10
18
8
16
5
15
7
14
6
11
5
6
3
7
12
7
9
11
9
23
17
31
13
28
14
29
15
34
13
33
20
32
21
27
18
17
15
9
11
42
20N
0' -1
19
8
-2
7
17
10
15
120
Figure A55.
140
150
160
170F
18C
170OW
160
150
140
130
120
7-year average values and standard deviations of the sensible heat transfer. August 1951-1957. (cal/cm2 /day)
10
7
110
7-YEAR AVERAGE
AND STANDRC DEVIATION OF QS
SEPTEMBER 1951-1957
1 ic:04
50N
45N
24
18
16
32
7
24
14
23
7
18
10
20
15
-3
20
-2
17
3
13
5
15
4
16
2
14
4
14
8
9
8
5
4
6
3
7
9
6
13
6
4
11
4
8
2 15
15
14
17
15
16
12
17
11
22
12
22
9
18
12
11
10
9
12
9
14
9
14
7
9
5
5
12
5
16
3
'
6
-9
10
32
27
15
26
15
31
17
32
14
29
13
32
14
34
15
36
11
34
6
28
8
27
12
23
10
19
11
16
8
15
4
17
3
17
3
9
4
-4
5
32 " 35 3P37
13'l l1111
35
15
30
12
27
11
28
9
30
10
33
12
33
14
33
11
35
10
32
8
27
11
25
8
21
8
17
3
17
4
15
4
13
3
8
3
3
4
5
9
4
17
30
9
25
6
25
8
23
9
17
7
17
9
22
11
28
10
26
3
24
7
24
6
26
7
22
6
18
5
16
7
15
5
14
4
14
6
12
5
11
4
8
3
7
5
8
5
5
6
24
11
15
11
5.
3
15
5
16
7
15
7
18
11
21
14
25
12
21
2
17
2
19
4
20
5
19
6
16
7
15
8
15
6
14
7
12
12
12
9
14
7
12
8
10
10
7
11
4
9
20
7
12
3
17
5
18
9
16
5
15
7
13
9
13
5
14
4
12
6
13
6
13
9
11
7
8
7
11
7
11
14
10
12
15
11
20
12
20
13
21
15
22
14
17
11
14
7
18
120
Figure A56.
11
4
35
16
140
150
160
170E
180
170W
160
15"y
140
130
120
7-year average values and standard deviations of the sensible heat transfer. September 1951-1957. (cal/cm2 /day)
13
5
110
7-YEAR AVERAGE AND STANDARD DFVIATION OF QS
OCTOBER 1951-1957
45m
"
49
40
39
29
50
32
5
46
46
24
45
20
38
19
40
20
47
26
34
22
38
17
40
15
48
16
56
23
52
23
22'
35
48
51
48
53
53
51
33
56
46
41
42
38
32
30
18
7
15
14
21
14
15
20
17
14
18
16
23
16
15
16
12
12
18
12
11
61
24
5
25
58
16
68
16
68
14
61
11
60
17
62
19
65
2C
63
18
58
14
55
20
49
19
46
13
42
14
36
9
29
8
22
9
14
6
3
8
P7 2 17-
75
22
77
24
69
16
60
16
53
14
45
7
43
11
42
12
43
12
47
7
48
8
46
15
49
13
45
8
36
7
29
7
22
5
18
3
14
6
8
4
7
3
78
28
72
15
56
7
47
11
41
12
34
9
32
1C
31
8
31
7
30
8
31
10
34
IC
36
6
37
9
38
12
33
9
29
9
27
5
23
5
18
4
18
2
19
4
19
7
18
7
16
59
21
42
11
28
9
23
7
20
7
22
9
24
8
22
5
25
10
26
11
28
12
34
13
31
12
29
13
28
11
24
8
21
8
22
5
22
8
21
9
22
8
21
7
20
10
23
8
22
5
32
11
24
8
22
9
18
10
16
9
19
4
2C
6
19
7
18
13
18
12
21
12
23
12
23
11
18
9
15
8
19
9
.22 20
17
13
20
10
21
10
24
9
21
5
19
9
21
6
19
3
160
150
84
8
$
29
14
66
76
21
75
24
20 k
32
8
6
62
I?0
Figure A57.
140
150
1]60
170E
IBC
170W
140
130
120
7-year average values and standard deviations of the sensible heat transfer. October 1951-1957. (cal/cm2/day)
16
5
110
7-YEAR
AVEPAGE AND STANDARD
DEVIATION CF
QS
NCVENBER 1951-1957
127 119, '97
111
95
61
B
ii
50N
'126
-73
20N
62
26
65
30
78\
56
*o
94
54
84
44
72
33
66
22
56
9
57
7
60
19
54
16
46
16
45
21
--
98 126 145 133 106 112 113
47
36
46
49
38
44 52
86
39
68
35
60
32
49
21
39
14
49
17
56
27
46
23
40
21
39
14
2
1
98
34
78
27
68
23
54
13
43
8
44
12
42
20
36
19
35
16
30
9
1
L52 162 155 139 '12
17
13
23
32
14
r
74
31
121 115 112 107
89
97
83
67
15q 145 168 157 129 109
41
34
45
26
17
22
106 1C6
36
25
84
29
106 112
26
34
93
9
85
16
75
23
75
28
76
33
69
31
61
27
53
18
45
12
41
12
37
18
32
14
28
9
23
4
15
5
10
6
102
10
95
4
Ha
19
76
15
66
12
56
11
54
6
54
4
53
14
46
18e
46
19
44
17
38
15
33
11
33
8
34
11
32
11
31
7
30
3
27
3
23
11
17
10
1
6
91
36
16
22
54
1i.
45
6
36
11
32
11
31
11
31
8
32
5
35
11
38
16
35
16
33
15
29
11
25
11
22
7
26
6
30
6
28
6
29
6
35
7
33
6
29
13
25
16
21
17
49
2U
35
10
27
9
24
6
2C
4
17
6
2C
10
20
10
19
7
17
10
16
16
2C
13
23
8
21
5
22
8
26
11
25
5
18
4
15
6
23
9
30
14
32
10
33
13
26
14
20
11
13?
Figure A58.
149
150
16C
170E
18C
1704
160
150
140
130
120
7-year average values and standard deviations of the sensible heat transfer. November 1951-1957. (cal/cm2/day)
18
9
110
7-YEAR AVERAGE AND
STANDARC DFVIA'TION OF QS
DECEMBER 1951-1957
148
91
131 10 '
66
64
127 113 108
53
41
55
97
46
92
32
e88 82
34
33
85
36
88
48
87
50
88
47
76
39
62
33
71
33
"180 157 188 171 138 129 121
-69 78 110
P3
60
63
64
98
42
78
29
67
27
65
25
61
25
61
24
62
30
55
36
44
34
41
28
43
17
38
18
237 212 215 179 150 146 126
63
42
48
47
53
54 41
i08
43
94
41
76
24
66
16
60
23
61
25
61
33
54
35
46
26
44
20
37
13
26
8
92
42
82
43
72
27
63
16
61
17
60
22
59
29
53
28
49
20
42
18
35
12
25
12
21
14
216 199 171
72
64 53
193 172 145 128 112 302
56
45
32
34
43
3q
145 132 122 104
59
69
62
46
87
37
77
35
76
31
74
29
69
23
67
24
63
23
61
20
60
22
59
18
57
19
55
20
48
16
46
12
39
9
31
13
25
14
22
13
2
I1
S34
112
33
91
72
60
52
3C
24
22
16
43
17
45
13
48
17
44
11
47
7
51
15
52
18
49
19
48
24
47
21
51
23
47
19
41
14
39
10
39
23
29
27
26
18
30
13
27
14
76
26
63
23
53
18
44
17
35
11
30
9
29
10
34
9
38
16
35
14
36
9
38
12
34
14
33
16
29
22
31
18
37
21
35
21
38
20
39
15
39
25
36
31
36
29
39
19
33
13
138
20N
169r'79 159 132 141
150 117
132
74
55
1i?
Figure A59.
140
150
160
170E
18C
170W
160
150
140
130
120
7-year average values and standard deviations of the sensible heat transfer. December 1951-1957. (cal/cm2/day)
30
6
110
7-YEAR AVERAGE AND
STANDARD OEVIATION OF SLHT
JANUARY
1951-1957
55N
-353-330-317-277-274-280-268-236-192-209
101
93
92
88 104 115 104
73
60 61
4TN
599-564-516-496-414-318-305-21-240-224154 172
97 122 114
86
63
78 58
149 131
123
66
-223-193-212-2
88
82
68
-745-706-600-481-419-369-294-246-247-234-244-265-283-282-235
77 149 133
88
69 73
60
66
74
89
68 83
95
87 115
-64 -695t85
825-971-92-842-755-696-615-49~-424-360-305-270-26-241-25-242-23-236-186-1
163/- 347
97
45
82 119 148 210 173 127 111
89
76
95 118 134 111 102 102 105 110
1
92
93
'58
-756-698-696-706-672-627-604-507 -426-415-391367-339-320-321-285-248-244-223-203-197-177-164 -94 -3
218 137 144 140 114 117 113 107 136 121 109 108
78
86 112 116 115 121 111 107
91
57
67
46 4
-671-571-480-405-370-369-344-280-247-277-293-326-294-282-261-231-220-202-176-175-165-189-207-153 -78
198 179 106 110 131 140 148 102
79
72 106 143 101
83 79
64
88
95
99 118 75
59
90 56
45
20N
$
-412-384-311-255-240-223-196-215-259-24-4
8-94-71-162-158-175-30-107-159-193-169-223-276-273-179 -47
225 161 113
67 109 109
99 110 117 117 146 123
81
87 120 128 119 110 123 159 169 153 130 118 101
95
130
Figure A60.
140
150
160
170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the total heat transfer. January 1951-1957. (cal/cm2 /day)
110
7-YEAR AVERAGE AND STANDARC
DEVIATION OF SLHT
FEBRUARY 1951-1957
\i
-169-178cr74-175-193-233-259-23
97 145
76
64
82
71
47
48
0
\o.
O
-267-252-268-218-170-146-115-126-131-144-190-209-191-15
73
88 108 94
72
63
37
33
61
60
73 75
82 74'-Q
{-23-308-374-392-383-371-355-248-176-147-107-109-124-123-136-146-137-129-105
45N
*
116 138
91
92
104 155 136 108
83
66
58
78
73
63
54
68
64
55
-570-6 7-604-585-64-583-54-489-406-305-236-191-146-131-119-101-110-122-123-109 -7
224 173 137
95 141 110
75 104 153 158 132 101
93
90 98
87
64
59
66
51
46
-2 7-4 025
7
91
-818- 45-646-579-545-453-367-303-263-216-187-159-132-128-132-141-125 -97 -46
84 136 121 108 102 80
49 85 116 121 129 115 113 106
70
57
49
46 51
-497-515-528-518-497-469-381-332-299-256-233-200-184-187-195-164-157-169-177-160-129 -87 -31
326 175 101
74 77
83
69 78 96
65
17
50
84 83
92
92 114 100
60
35
67 71
60
26-124-110-145-200-184-160-171-185-175-124
276 205 139 100
82
97 130
-448-374-318-270-273-245-204-162-112-11
-69
5 78
90
115 102
84
51
92 111
91 116
123 119
61
93
94
94
-163-175-209-173-152-101 -65 -81 -71 -84 -91 -60 -33 -70-129-137-1 -106-154-174-146-102 -54
145 148 143 92
85
87 101 119 129 104
94 88
97 92 126 113
33 91 164 194 149 140 123
11
77
38
87
130
Figure A61.
140
150
160
170E
180
170W
160
150
52
83 110
140
118
130
120
7-year average values and standard deviations of the total heat transfer.
February 1951-1957. (cal/cm 2 /day)
79
55
110
7-YEAR AVERAGE AND STANDARC DEVIATION OF SLHT
MARCH 1951-1957
55N
-73 -80-126
80
73
82
5rN
-206-180-159-131-101 -99 -91 -68 -50 -51 -66 -87-111-114K
134
92
27
33
54
64
60 47
49 52
47
29
61
75
45N
-89-171-225-209-180-178-138 -99 -86 -89 -66 -42 -40 -50 -68 -81 -66
132
77
68
80
85
98
66
30
42
42
56
83
67
54
51
55
40
4CN
-2253 219-288-362-348-294-253-221-173-128 -97 -80 -67 -49 -55 -80 -91 -93 -60
311 158
84
96
95
90 102 119 134 87
62
67
78
93
99 93
80 57
66 45
30N
25N
20N
-
- 9-1 9159
6
35N
-436-434-404-325-261-206-153-146-130 -94 -72 -62 -53 -67 -93 -94 -93 -71
43 161 158 167 129 104
98
97 103
89 69
80
93 103
94
81
56
49 55
38
12
52
-193-230-184-187-220-188-154-114 -95 -89 -79-104 -99 -76 -57 -72 -85 -99 -98 -99 -88 -53 -28
96
91
98
98 138 138 143
89
62
70
96 119 118
81
56
43
58
64 38
38
27
39
74
&
-233-139 -46
75
96 113
-21
85
1
91
130
Figure A62.
36
92
-8 -16 -18
91
90 90
39
83
140
37
64
-8
93
-5 -26 -78 -75 -82 -73 -50 -38 -71 -96-124-113-106 -89 -50 -25
86 101 166 182 166 137
91
68
60
60 70 97
77
35
53
96
36
82
28
21
110 103
150
160
89
70 1 4
57
3
25 103
68
54
2 -51 -65 -86-112-109 -80 -66-10-131-109-101-112 -94 -97 -57
130 201 222 169 138 149 113
77
90 99 115 115 105
89 110 105
170E
180
170W
160
150
140
7-year average values and standard deviations of the total heat transfer. March 1951-1957.
130
120
(cal/cm2/day)
49 139
73
77
110
7-YEAR
AVERAGE
AND STANDARC
APRIL
DEVIATION OF
SLHT
1951-1957
55N
64
66
2N
0
20N
49
58
41
42
41 110 182
71
81
77
145
95
32
72
Figure A63.
49
74
53
36
68
38
7
40
31
39
30
59
33
75
40
70
49
56
48
56
57
75
69
71
71
67
21
71
18
69
-6
52
13
70
31
62
39
66
44
67
47
73
48
61
66
63
77
78
82
60
98 121
41
58
-9 -46 -50 -64 -75 -56 -30
57
56
76 89
96
80
81
6
53
20
30
28
35
28
43
26
46
40
44
57
49
69
46
94 126
36
33
11
65
32
42
43
35
30
34
16
51
19
50
26
40
42
45
53
58
71 115 154
54
40
41
57
-7
55
57
43
41
40
27
57
49
68
54
67
27
56
26
65
50
58
45
75
31
96
29
86
32
54
30
54
37
49
61
57
63
56
83 142 182
44 33
2
174 170 159 134
99 125 113
hB
90
77
30
93
69
81
89
78
40
66
12
69
17
73
8
86
18
88
34
80
34
68
23
21
82 101
36
99
34
79
36
90
149 145 148 116
67
62
59
71
79
85
78
73
83
79
53
80
55
42
16
108 1.00 100
180
170W
194 222 178
72
52
55
130
37
96
47
72
4 -24
57
55
-94 -84 -73 -60 -53 -49 -39 -14
112 105
72
64
85 92
95
86
100
77
25032
22
13 123 133
27
53
'229 177 105
6 50
57
120
75
10
72~
140
150
160
170E
88 154
70
39
-2
-2 -14 -44 -17
4 -44 -85 -31
96 103 142 165 125 103 100
79
75
160
150
140
130
120
81 182
56
54
110
7-year average values and standard deviations of the total heat transfer. April 1951-1957. (cal/cm2 /day)
-14
-44 -17
4 -44 -85
-31
81 182
7-YEAR
AVERAGE AND STA:J)4RI' DEVIATION OF SLHT
MAY 1951-1957
55N
I.Z,
V
45N
2 3 2'12 190
11(
72(-61
3
130
35
139
38
114
47
86 106 119
45
21
17
114 124 131 132 136 149
15
25
26
14
17
35
152 152 148
37
29
32
'276 240 179 137 128
0
59
47
48
49
113
41
88
38
9P
34
99
40
108 110 115 133 150 158
38
39
33
26 29
37
159 156 169 20
36
31
54
7
73
55
84
55
93
60
184 157 1C9
68
33
46
359N
120 1401,1'47 131 112 126
39
45
51
56
35
34
7jT 51
67
61
45
68
52
76
97
62
81
57
79
56
77
50
69
68
69
57
92 119 136 156 147 112 102
65
77
66
50 50 67
41
87
47
89 115 127 140
34
35
30
16
138 138 160 21,
18 27
28
4
97 104 112 109
43
43
36
32
117 135 178 209
33
53
55 '55
r'
30N
197 154 150
110
59 42
)25N
157 189 233 261
85
69
95
92
2CN
235 265 254 213 198 199 190 215
34
22
54
76
67
68
63
51
130
Figure A64.
171 174 167 173 187 179
84
85
80
59
30
31
140
272 266 244 255
63
82
94
58
150
160
171 160 171 178
47
51
43
29
153 110
26 46
255 208 152 135 131 107
81
38
24
48
58
69
83 106
206
68
95 108 100
44
47 49
91
71
92
76
95 108 120 170 25
85
94
53 44
5
61
73
87
86
62
31
46
58
93 115 105 107
164 133 113 120
73 70
91 101
112 122 107
105 102
83
64
81
20 -14 -18
-8 -54 -76
86 124
96 111
74 60
170E
170W
180
160
150
140
52
99
130
36
79
96 218
50
77
0
73
120
7-year average values and standard deviations of the total heat transfer. May 1951-1957. (cal/cm 2 /day)
83
77
154
30
110
7-YEAR AVERAGE AND
STANDARC DEVIATION OF SLHT
JUNE 1951-1957
55N
144 160ofr80 183 167 140 105 104\LV
54
52
46
52
51
30
17
80
4
50N
117 126
22
22
45N
182
37
147 131
29
31
209 155 119
46
28
40
129 139 151 158 172 172 175 168 144 120 118 12Z
19
22
27
26
27 22
36
34 28
14
35
45
126 136 142 145 171 188 192 196 199 185 166
17
23
33 40
48
36
35
36
18
31
27
98 101 117
41
53
42
116 124 141 165
47
36 41
38
158 152 173
19
29
27
193 204 203 191 167 147 132 175
43 46
37
35
33
36
41
45
141 144 135 115 111 126 135 154 161 174 206 216 204 181 161 135 111 137 226
50
45
53
53
76
80 81
73
66
47
39 35
31
31
42
39
37
42
72
219 213 235 246 251 247 233 202 207 218 236 240 244 232 232 222 172 132 100
103 65
49
43
28
27
29
26
37
40
32
53
70 42
47 46 47
34 35
224 254 303 317 303 287 278 273 271 273 274 268 263 230 196 170 108
102 49
48
56
45 38
35
39
40
38
24
48
76
50
49 64
58
42
20N
134 181 232 237 231 235 242 220 220
53
26
30
50
53
49
38
27
27
130
Figure A65.
140
150
160
217 232 214
38
64 52
170E
180
180 143 137
44
74
75
170W
87
48
36 -11 -29
67 80
90
77 108 165 200
42
50
59
57
-6
63
69 119 140 232
87 134 120
71
94
8 -31 -82-130 -86 -33 -17
87 104 127 128 112
64 65 125
160
150
140
130
42 111 223
89 49 60
120
7-year average values and standard deviations of the total heat transfer. June 1951-1957. (cal/cm2/day)
110
7-YEAR AVERAGE
AND STANDARC DEVIATION OF SLHT
JULY 1951-1957
55N
108 13.
17 _2
50
O
130
18
45N
280 277 260 245
105 101
44
31
C
110
89
98
87
130
Figure A66.
108
90
130 130 116 115 132 143 140
23
18
37
38
22
12
15
85
34
97
28
137 132 124 118 110 129
17
19
18
24
33
30
'174 170 151 143 144
-53
38
33
37
30
154 159 151 165 169 171 155 143 136 137 134 126 157 25:
22
29 36
28
40
48
30
28
13
26
34 40 40
4!
190 181 163 158 158
53
66
56
37 32
174 194 202 199 194 198 180 152 128 125 123
47
71
70
58
50 39
40
35
33
37 40
118 147 230
40 44
6;
242 217 202 200 194 203 221 221 212 211 200 188 156 127 118 112
54 63
60
38
24
57
76
69 57
42
41
48 45
52
64 59
89
45
95 144
32
40
221 203 195 199 198 197 190 177
35 45
68
52
13
40 64 66
'189 183 180 187 182 147 116
70
59
75
67
46 84 128
20N
,0 133 114
4
31
24
180 178 160 132 123
74 64
37
44 47
94
35
82
41
66
53
64
59
68 102 190
49
83 50
5
68
25
81 218 316
42 123
83
63
157 192
97
51
173 119 124 130
67 89
77 69
115
63
84
40
68
54
52
64
33
46
4 -20
60 63
128 161 138 117
63
39
60 84
127 123
98
81
121 132 125 122 129
82 100 70
54 64
110
74
84
75
58
66
40
46
10 -35 -41 -28
69 86
82
33
140
160
170E
150
180
170W
160
150
140
130
47 169 251 236
38
63
72
51
120
7-year average values and standard deviations of the total heat transfer. July 1951-1957. (cal/cm 2 /day)
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF
SLHT
AUGUST 1951-1957
35
31
52
46
86
27
72
16
50
35
84
71
84
28
72
25
73
30
82
24
73
16
72
34
75 105
50 39
104 108 107 100
22
32
26
35
93
34
94
22
95
29
94 113 188
23
24
50
A
o31
S
j
45N
~118 113 103
6 45
26
156
78
3
2 8 2 6 r-7
15
63- ,4
79
26
97 113
24
36
86
17
95
24
91
27
88
19
127 112 111 117
41
32
35
30
'47 134 136 117 114 128 119 103 106 114
2 64
58
39
54
59
55
59 68 60
7
83
19
112 123 134 124 112 104 105 106 133
34 26
16
28 34
19
28
30 31
124 129 142
90
69
38
149 141
31
33
121 120 125
45
52
54
143 155 127 132 145 132 139
97
65
68
48
60
58
54
141 140
42
35
146 150 148 154 141
39
58
57
39 56
69
2 N
100
36
26
93 102
85 137 180 124
85
20N
-40 -69 -48
12
111 139 138 102
130
Figure A67.
140
23
81
59
106
P5 133 150
13
79
60
30
72
66 105
72
71
150
160
105
75
158 133 107
69 104 127
119 110
77 107
170E
60
75
180
130 141
57
48
209
45
156 171 159 143 124 109 106 123 148
60
64
51
36
19
27
47
45
46
148 159 136 101
80
73
45
42
81
35
72
66
60
68
35
61
29
59
106
109
77
75
55
64
53
56
1 -25 -41 -55 -57 -45
51
63
74 49
55
70
63 221
62
65
51
89
57
72
45
50
-1 -57 -69 -73-123-113 -88 -52
73
81 103 102
55
86 114
73
54 162
73
61
170W
160
21
50
150
140
7-year average values and standard deviations of the total heat transfer. August 1951-1957.
130
120
2
(cal/cm /day)
179
43
110
7-YEAR
AVERAGE AND STANDARC
DEVIATION
OF
SLHT
SEPTEMBER 1951-1957
55N
-39 -22 -39 -52 -60
46 42
35
39
59
8,
>1
50N
24
59
5
4 N
1
42
-8 -31
30
34
17
29
3
22
-5
20
32
31
11;
31
-41 -46 -,2 -87 -77-104-124-134-115 -87 -73
53
85
95
83
85
93
93 67 42
46 62
-39 -20 -21 -12
49 49
43
24
-7
19
5
22
50
21
12,
25
120 -79 -61 -65 -92-119-116-116-119 -94 -51
139 129 119 94
96 99 108
89
75
72 81
-42 -22
71
53
110 1 5
35
'68
107
-150 -92 -47 -45 -31
153
96 73
84 114
25N
20N
/
13 -18 -22 -20 -16 -19 -23 -20 -11
34
21
31 45
49 49
40 23
35
55
59
12 -11 -19 -31 -58 -73 -52 -22 -14
60
58
61
44
52
41
49
43 54
-1
55
7
47
10
43
6
41
4
46
23
41
26
40
65
31
17
72
14 -24 -59 -47 -27 -12 -17
72
87
85
55
65 55
67
8
50
35
47
14
69
20
65
44
40
50
51
59
67
60
57
64
52
90 133 179
38 40
3
42
67
48
92
35
83
35
84
79 145 208
56
54
43
-58
119
19
82
76
68
74
70
65
67
50
53
32
13
92 120
-4
98
18
55
45
39
55
11
47
34
48
41
61
76
36
87
29
59
50
33
-22
79
47
46
73
58
53
56
48
85
44
66
57
62
89
44
84
57
80
71
52
78
44
62
62
56
80
48
59 D51
42
86
62
92
130
Figure A68.
140
150
80
83
160
170E
180
170W
160
150
38
3 -20 -48 -41
107 123 105
82
65
140
130
68 106 115
70 27
37
120
7-year average values and standard deviations of the total heat transfer. September 1951-1957. (cal/cm2 /day)
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF
SLHT
CCTOBER 1951-1957
>0
55N
V
50N
-170-154-159-179-185-212-235-233-225-217-199-201-205-141
38
40
54
37
49
59 64 58
66
65
50
61
70
62
" -86 -75-130-201-226-212-230-229-222-231-238-218-205-203-204-193-175-110 -35
53
64 44
7
25
49
56
53
53
57 82
52
66 56
44 56
78
60 47
45N
25N
20N
-196-197-204
107
99 94
-366-2 99251-245-290-306-288-286-293-305-291-279-269-243-227-205-191-160-123 -59
150
5 99
76
73
42
32
69
87
80
61
65 114 90
56
63
27
38
55
44
9
41
-22-32042
-_7-441-395-332-268-229-229-225-230-241-252-249-263-228-174-137-110
-70 -38
187f--V5 -- ,4 132 131
98
87
79
84
98
91
62
32
67 122 103 60
48
47 43 31
40
20
18
\46
30
-456-386-308-274-230-181-156-138-130-136-153-166-167-174-191-169-129 -98 -79 -61 -62 -64 -44
222 103
54
71
66
53
72
63 78
87 99
80 54 72
96 63 63
27
24 24
19
27
39
6
39
S139
-398-258-145 -89 -36
81
58
40
56
-33 -46 -48 -79-100-131-167-139-125-123 -81 -54 -50 -47 -41 -73 -68 -37 -12
62
57
45 119 123 116 118
98 100
87
69 75
51
51
50 49
62 85
65
-174 -85 -75 -50 -28
106
53
65
55
53
-22 -22 -28 -55
32
55
73 91
-52 -98-115 -90 -62 -41 -47 -37 -30 -50 -52 -74 -62 -53 -31
85
91 110 101 94 102
76
60
81 105
71
63
48
82
84
150
170E
130
Figure A69.
140
160
180
170W
160
150
140
130
120
7-year average values and standard deviations of the total heat transfer. October 1951-1957. (cal/cm 2 /day)
10
87
74
86
110
7-YEAR
AVERAGE
ANC STANOARC DEVIATION
OF SLHT
NOVEMBER 1951-1957
55N
-396-366a36-310-291-260-265-30
210 6O 101
46
51
49
66 162
50N
-405-391-384-380-362-323-287-284-262-262-274-264-258-24
187 201 187 151 131 111
79
43
31
32
55
69
66
79
45N
-439-360-415-487-472-413-440-435-373-314-289-276-241-256-277-251-238-232-153
1
84
36 108 126 103 122 154 126 106
99 68
51
66 89
85
92
73
46
-521-6 8 607-562-643-632-548-486-471-480-445-377-343-305-266-267-259-232-221-182-105
232 2 3 122
82 111
62
70 68
74 121 127 98
80 45
30
52
66
72
77
43
17
-29 -415-570-681-767-749-688-591-5 18-464-399-393-404-379-341-307-272-267-243-210-182-147 -86 -44
113_' I 521'47
66
78 119 49
75 109 125 137 149 145 124
96
63
58
85
69 50
21
29
47
-501-539-553-529-535-474-413-362-340-328-310-280-271-277-258-236-221-217-212-198-192-148-107
-52 -33
128 116
47
40 114 102
95 86
55
43 110 123 114
97
80
70
65
71
63
44 15
22
45
70 89
-546-437-372-289-244-21R-190-165-164-192-230-201-183-174-154-134-146-186-192-200-225-187-139 -96 -45
201 137
87
63
80
97 11
75
50
65 117 113 114
82
63
71
75
52 36
60
52
45
75 123 133
2
20N
vo
-314-238-184-119-111-116-122-119-107-128-146-151-141-101 -93-131-143-123-130-172-199-204-196-148 -74 -18
147
85
79
49
58
75
93
88
79 89 104
79
83 57 79
80 41
15
43
87 114
91 102 125 103
57
130
Figure A70.
140
150
160
170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the total heat transfer. November 1951-1957. (cal/cm 2 /day)
110
7-YEAR AVERAGE
AND STANDARC DEVIATION
OF SLHT
DECEMBER 1951-1957
V
55N
134
195 213 212
188 150
158
158V
.06
-405-375-356-356-350-331-308-315-328-337-363-347-322-327
93
60
98
88
84
82
78
89 117 128 133 121 106
93-
50N
3-534-509-444-432-428-382-334-300-295-295-290-312-305-281-281-277-245
64
83
71
61
69
76 106 128 122 103
60
3 209 169 127 151 168 111
45N
6-713-639-565-562-518-460-415-357-328-309-306-319-309-282-270-236-182
P 132 142 168 176 146 148 134
92 73
94 114 133 127 103
87
48
21
584-548-491-440-405-3789-343-324-312-324-311-282-245203 202 170 158 168 118
56
80 119 148 134
98
92
-739-716-667-642-613-585-508-462-474-450-396-377-363-348-343-343-337-334-296-271-224-167-132-103
-94'
56
62 68
7
89
86
73 115 124
95
83
71
157 139 124 170 202 185 193 189 160 135 115 118 110
-699-611-532-450-420-366-2:9-299-336-309-304-322-330-302-290-305-334-309-279-239-206-160-156-153-109
90 125 114 100 138 113
98 115 125 131 119 165 138 110
73
57
89
152 148 162 166 176 135 1,8 102
'a
20N
-497-423-382-334-276-239-209-236-261-239-229-236-220-194-161-207-258-232-224-229-245-236-242-225-162-113
70
97
98
5 95 166 159 124 123 136 161 177 163 195 147 122 112 149 145 127 105
126 134 146 134 105
130
Figure A71.
140
150
160
170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the total heat transfer. December 1951-1957. (cal/cm 2 /day)
110
7-YEAR AVERAGE ANC STANDARC CEVIATION OF SST
JANUARY 1951-1957
55A
V
5Ch
394 394 3-8 4C2 407 417 427 439
6
8
7
11
12
10
9
10
%
391 397 404 407 411 414 416 418 420 424 433 446 459 471(Z\
16
13
16
14
12
9
9
7
6
7
6
4
8
9
399 393 403 421 435 444 454 460 464 466 466 466 465 470 478 486 493 496 500
10
13
17
18
17
18
19
19
18
16
15
13
11
10
10
9
6
6
7
518 515 526 534 539 543 546 547 545 541 538 537 538 542 547 549 549 540 533
20
12
12
12
15
18
19
20
19
16
14
14
14
14
13
10
7
8
6
637 634 633 631 629 626 621 616 612 609 607 609 612 615 615 612 605 589 570
11
10
12
10
10
10
9
10
11
12
13
14
15
14
11
8
7
9
6
3CN
652 670 684 689 692 691 689 688 686 685 682 679 676 676 676 676 676 676 670 659 647 629 614 610 6'26
21
15
11
6
4
3
5
7
8
8
7
7
8
8
8
9
9
8
5
3
4
6
6
6
1
25hj
730 738 741 743 748 751 750 747 744 743 742 740 735 734 735 732 727 721 710 698 686 668 656 656 668
11
7
8
5
5
5
7
10
1C
8
8
8
8
7
6
4
4
4
4
4
8
9
9
10
19
2CK
778 783 785 789 793 793 793 790 787 785 783 780 774 769 764 759 750 742 733 727 719 708 704 713 731 756
7
6
5
3
3
6
8
7
6
6
5
3
3
5
7
4
9
7
5
7
11
7
6
10
16
15
130
Figure A72.
140
150
160
170E
180
17CW
160
150
140
130
120
7-year average values and standard deviations of the sea-surface temperature. January 1951-1957. (OF x 10)
110
7-YEAR AVERAGE ANC STANDARC DEVIATION OF SST
FEBRUARY
1951-1957
S383
15
5CA
383 386 389 397 407 418 432
9
7
8
12 11
17 12
378 388 397 401 405 407 409 411 412 417 426 438 449 460
18
10
8
9
8
7
8
8
8
8
7
6
6
9
45h
380 374 388 404 414 425 439 448 453 457 458 459 461 464 470 478 483 488 494
3 11
12 12 15
13 11
13 14 11 13 13 13 12 11
9
6
8 11
53
11 13 495 492 504 512 516 522 528 530 529 527 526 527 530 532 536 539 537 531 525
13
16
14
6
9
9
14
13
9
10
12
IC
10
12
13
12
10
10
8
6
7
5 8 562 598 614 619 612 610 608 608 608 605 602 599 596 595 596 601 603 604 600 592 577 561 564
31-21 C15'- 7
6
3
5
7 10
10
9
8
7
9
7 10 10
9
7
6
6
6
5 .3
629 656 672 672 673 672 67C 668 667 667 665 664 664 662 662 662 665 664 660 650 638 623 609 603 612
26
9
7
5
4
4
5
8 10
9
9
9
9
5
4
6
6
5
3
3
6
5
6
7
9
721 725 725 728 732 734 734 732 731 732 730 729 727 724 723 721 718 712 704 694 680 668 658 645 647
15
7
4
3
3
4
7
9 10
12 12 10
8
5
4
4
4
4
3
6
6
5
9 15 17
2Ch
77C 771 775 780 784 785 783 781 777 774 771 769 767 761 757 753 747 739 734 726 712 704 702 703 717 741
6
4
3
4
3
4
6
6
7
8
8
7
7
6
7
5
5
7
7
6
7
9 10
16 18 16
130
Figure A73.
140
150
160
170E
180
17CW
160
150
140
130
120
7-year average values and standard deviations of the sea-surface temperature. February 1951-1957. (OF x 10)
110
7-YEAR AVERAGE ANC STANCARC CEVIATION OF SST
MARCH 1951-1957
55%
374 372 380 385 393 405 418 432
9
10
8
14
11
7
5
7
b,!
5N
377 385 392 398 402 405 408 410 411 415 422 433 444 459Z-2C
16
7
7
9
6
8
9
9
9
8
6
6
6
370 366 382 397
-10
11
12
13
407 420 436 447 454 459 462 463 463 464 466 471 476 482 490
15
14
11
11
12 IC
10
11
13
13
13
10
7
5
7
481 480 494 5C4
9
9
11
11
510 519 527 530 532 531 529 530 532 533 534 533 531 525 520
12
12
10C 10
10
9
8
9
11
11
11
8
7
6
7
608 604 603 602 603 604 601 600 598 596 595 597 6C1 603 601 595 586 571 556
7
8
8
9
8
6
6
8
10
11
10
10
10
8
6
6
8
8
6
631 653 663 670 673 670 668 667 666 666 663 663 662 662 662 664 666 663 657 645 630 616 603 600 610
28
14
8
5
5
7
8
9
9
8
7
8
9
11
11
8
6
4
3
5
7
8
8
5
7
729 728 727 732 735 734 735 733
11
8
8
8
8
10
13
13
731
13
732 73C 728 726 723 721 720 716 709 700 687 672 657 647 643 648
11
9
8
8
9
8
7
6
3
4
6
4
8
8
9
10
779 777 779 784 787 786 784 779 775 774 771 769 767 762 757 751 743 733 727 718 706 695 692 701 715 738
7
6
5
5
4
6
8
8
10
8
6
4
6
6
5
4
4
7
9
7
5
10
14
14
14
10
130
Figure A74.
140
150
160
170E
180
17CW
160
150
140
130
120
7-year average values and standard deviations of the sea-surface temperature. March 1951-1957. (OF x 10)
110
7-YEAR AVERAGE ANh STANDARC DEVIATION OF
SST
APRIL 1951-1957
386 384oe90 396 399 408 423 43
15
20
22 23
14
7
8
12
379 386
9
7
395 400 406 409 411 413 415 420 428 439 452 464
7
7
5
5
6
7
8
8
9
9
7
7
395 385 394 405 413 422 436 449 456 459 459 461 462 467 471 477 484 490 496
-14
11
9
9
11
11
507 499 508 514 516
521
13
12
10
11
13
13
9
8
7
8
528 533 534 532
11
7
6
6
10
9
10
10
10
10
8
6
10
53C 531 534 536 538 539 537 530 523
7
8
7
7
7
8
9
7
8
625 616 613 611 610 61C 610 608 605 603 6C2 605 607 608 606 600 591 575 559
10
10
10
11
11
10
9
7
7
8
8
8
7
6
5
7
9
8
8
3Ch
25NJ
2CI
642 675 691 690 687 682 678 677 677 677 677 678 675 671 671 672 673 670 661 650 637 622 606 600 6'1
17
13
12
9
8
9
8
6
5
6
7
7
8
9
9
8
6
5
5
7
7
5
8
8
7
749
S11
751 749 750 752 749 747 746 744 742 741 739 736 731 730 729 725 717 705 692 680 670 654 637 638
10
12
10
9
8
8
8
8
7
7
7
8
8
7
7
6
7
8
9
9
8
11
13
10
799 798 799 800 799 797 794 789 785 780 776 773 768 762 760 758 751 742 734 726 713 705 701 700 712 740
4
5
7
6
5
4
5
7
8
9
S
8
5
5
6
7
7
8
10
12
17
14
11
10
9
9
130
Figure A75.
140
150
160
170E
180
17C0
160
150
140
130
120
7-year average values and standard deviations of the sea-surface temperature. April 1951-1957. (OF x 10)
110
7-YEAR AVERAGE ANC STANDARC
CEVIATION OF SST
VAY 1951-1957
55I
408
6
SCh
391 398
11
11
4SN
425 428 435 451 471
8
8
8
8
9
408 414 420 426 429 433 438 443 451 465 483 498t
9
7
6
7
7
7
7
7
8
7
7
13
4439 417 416 425 429 437 449 459 466 469 472 474 477 482 489 499 509 517 521
24
13
12
10
11
13
14
11
11
11
12
9
8
8
7
7
8
10
12
552 537 532 534 534 539 545 549 550 547 544 543 546 551 554 556 556 550 54
17
11
9
10
13
14
13
12
10
9
T
7
9
8
7
7
8
8
9
661 648 640 636 636 637 637 633 630 625 622 621 623 624 622 616 605 590 575
12
11
13
13
15
14
13
13
12
10
11
13
13
11
7
8
9
7
7
695 714 726 726 720 717 714 712 712 711 708 704 699 695 694 694 691 686 677 663 650 633 619 616 622
1C
9
9
15
19
19
17
13
13
13
14
13
13
13
13
14
14 10
8
8
8
6
8
9
10
790 781 780 783 783 783 780 775 771 767 762 758 754 750 747 744 737 729 718 704 691 676 661 653 650
11
10
11
14
17
15
11
8
12
14
14
12
11
11
9
11
11
8
8
12
13
10
10
6
10
2
20C1
828 820 819 821 819 814 808 802 797 792 787 785 782 778 774 769 765 759 748 732 719 713 708 711 721 753
8
8
9
7
7
6
6
6
9
8
8
8 11
IC
11
14
16
16
12
16
19
14
12
7
8
6
130
Figure A76.
140
150
160
170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the sea-surface temperature. May 1951-1957. (OF x 10)
110
7-YEAR AVERAGE ANC STANCARC CEVIATION OF SST
JUNE 1951-1957
438 43944-47 454 462 478 495 510
12 15 14 19 18 15 13 12
5C1
423 424 431 438 443 449 456 463 468 474 485 502 518 5321
12
8 10
10 10 11 13 15 16 14
11
E
6
7
451N
488 463 459 461 458 460 470 478 483 488 495 502 5C8 513 521 531 540 548 552
-16 14
12
11
11
9
7
6
9 10
8
6
7
8
9
14 18
16
15
598 583 574 571 565
14
13
9
14
15
563 567 573 576 577 581 586 590 592 593 590 585 576 562
14
12
11
13
11
9
11
12
9
7
8
14
14
14
706 693 682 675 671 667 666 669 671 672 674 676 677 673 662 647 631 610 586
14
15
14
4
9
19
16
17
20 18
15
16
15
8
20
13
14
17 20
744 758 767 768 768 763 757 754 750 747 744 743 739 735 732 728 725 717 701 682 663 646 629 629 6t11
9
5
9
9 10 12
15 16 14 12
9 10 15 15 12 11
9
9 11 13 14 11 11
13 17
817 817 820 822 824 821 817 814 810 805 799 792 784 776 769 759 752 743 729 712 696 682 667 662 666
10 10
9
7
9
9
9
8
8 11
8
8 11 10 11 13 12 13 13 11
10
9 14 14 15
2CN
C
844 844 848 848 844 839 835 832 826 820 812 805 801 794 785 779 771 761 751 740 729 720 712 720 734 772
7
7
7
5
6
7
9
9
10
10
10C 10
9
9
10
9
6
8 10
9 11 10 12
8
13 11
130
Figure A77.
140
150
160
170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the sea-surface temperature. June 1951-1957. (OF x 10)
110
C::Z
7-YEAR AVERAGE
ANC STANCARC CEVIATION OF SST
JULY 1951-1957
476
12
00 510 522 536 544 544
15 15
16 18 16 16
4
0
-e
472 468 47C 478 482 491 502
9
11
10
11
10
9 11
45
"560
0
530 525
2C
15
521 511 513
14
18
18
522 531
17
12
512 518 526 535 547 557
12 13 12 12
13 13
531 542 551 559 564 568 573 577 582 584 576
10
9
8
10
16
16
16
17
16
12
5
700 6 7 665 648 64C 632 623 624 63C 634 639 643 645 646 646 642 638 631 622 605 58
3
48
7 2 742
1824
1
21
18
17
18
752 758 756 747 737 733
127
23
20
17
16
21
20
19
16
12
1C
12
13
16
15
15
15
14
11
10
731 729 727 729 733 731 727 723 717 705 693 675 656 631 605
16
15
15
12
10
10
11
9
10
11
12
13
12
11
11
14
13
798 806 811 809 806 800 795 793 793 790 785 781 777 772 767 760 748 734 716 696 679 665 652 655 675
13
9
6
8 12
13
10
8
8
5
3
5
8 10
10
9
9
10
11
13
13
11
14
15
17
845 847 847 843 841 840 838 835 831 825 817 809 802 796 790 780 766 752 738 722 709 701 697 695 705
8,
8
8
4
7
9
8
7
8
8
7
7
7
7
6
7
7
8
9
10
9
12
17
16 23
(7
20N
851 854 856 855 853 850 847 843 838 830 823 818 812 805 798 788 779 767 754 747 739 734 736 747 766 805
7
8
8
7
7
7
6
5
7
7
5
5
5
5
8
9
9
7 12
9 12
9
7
8 16 14
130
Figure A78.
140
150
160
170E
180
17CW
160
150
140
130
120
7-year average values and standard deviations of the sea-surface temperature. July 1951-1957.
(OF x 10)
-110
7-YEAR AVERAGE ANC STANDARE DEVIATION OF
SST
AUGUST 1951-1957
502 5C9 0 530 541 550 566 577 578!
15
9 15 19 16
7
11 11
0 %
538 527 527 529 531 536 544 553 562 569 578 588 596 594'
21 22 22 19 11
6
7
9 12 11 11 10 10 11
39 601 594 595 586 582 587 592 595 601 607 612 616 619 621 622 621 613 590
3 28 17 14 21 28 30 25 15
9 14
17 16 13 13 14 14 11 10
45h
771 7f
726 709 701 697 689 687 689 690 689 69C 692 693 692 686 679 668 654 630 596
8
1 17 14
11
9 17 22 23 17 10
8 14 16
14 11
14 15 16 12 12
78 797 807 804 799 793 786 780 775 771 769 765 761 756 752 749 743 731 719 701 680 651 621 633
2C1'
11s'- 6
9
8
8
5
5
9 11
8
4
6
9 10 10 11
13 14 13
10
9 10
822 828 833 831 826 823 819 816 813 808 803 798 792 784 776 768 760 747 731 713 696 677 659 673
iC
7
6
6
9
7
6
5
4
4
5
4
6
8 11 10
7
7
6
7
7
6
6
9
6
2CK
8
7
6
7
6
6
7
8
8
7
6
5
7
12
11
7
6
7
7
6
8
10
10
15
848 847 849 852 852 852 850 849 845 R40 833 826 819 813 806 799 791 781 771 763 753 749 753 768 793 829
7
7
6
6
8
6
4
4
5
6
6
4
3
4
5
6
3
6
5
7 10 12
14 10
9
8
130
Figure A79.
140
15C
160
170E
180
17CW
160
150
140
130
120
7-year average values and standard deviations of the sea-surface temperature. August 1951-1957.
(OF x 10)
110
7-YEAR AVERAGE ANC STANDARC CEVIATION OF SST
SEPTEMBER 1951-1957
49C
8
529 537 548 559 567
10
11
16
16
13
o aOC
5C1
538 527 522 519 519 523 532 544 557 566 574 582 590 588
9
9
8
6
6
6
9
12
13
13
13
11
10
13
%60 7 590 591 593 586 580 580 581 582 586 593 602 610 617 622 624 625 616 591
44
18
10
11
12
14
15
16
13
11
12
15
16
12
11
11
10
10
17
45N
746
19
0 713 700 699 696 686 679 676 673 672 673 677 682 685 685 681 673 662 642 61
16
10
7
8
10
14
17
17
17
15
12
12
11
8
8
12
13
12
11
17
7 8 776 787 791 793 792 788 781 771 763 756 749 745 744 744 744 743 735 726 711 692 667 637 639
22- 15
1
8
6
5
5
7
11
13
14
12
10
11
11
7
6
11
13
12
12
13
14
13
808 819 823 824 824 822 820 817 813 806 8CC 795 790 786 781 775 766 756 744 727 709 689 673 681 71
11
7
7
6
4
6
5
5
5
7
8
6
3
5
7
8
8
9
11
11
12
14
19 12
10
836 838 838 839 839 839 839 838 838 833 826 822 818 810 801 791 780 768 756 739 724 715 710 725 759
1C
2CN
10
9
7
7
8
7
6
8
7
5
4
5
6
9
9
8
10
12
10
11
17
25
16
14
845 844 844 847 849 850 847 846 845 839 832 829 825 817 807 759 790 781 775 765 757 755 758 776 803 830
12
9
10
10
9
9
8
7
5
5
5
5
4
7
9
11
15
20
16
6
9
14
19 13
10
10
130
Figure A80.
140
150
160
170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the sea-surface temperature. September 1951-1957. (OF x 10)
110
7-YEAR AVERAGE ANC STANDARC DEVIATION OF SST
CCTOBER
1951-1957
55N'
475
CW
489 500 511 523 534
13 15 18 17 18
503 491 490 489 491 496 5C2 509 517 528 539 550 560
14 12
8
8
9
7 10 13 16 18 20 19 16
45N
'566 545 547 555 554 545 543 545 546 547 551 558 566 576 587 594 599 592 57
-18
18 16 13
15 18 23 21 16 11
9 11 15 17 19 16 12 10 1
660 653 659 659 653 645 639 635 630 628 631 637 643 648 651 649 643 626 601
12 13 14
19 23 24 25 24 19 15 13 10
9 10 13 11
8
7 1
49 755 753 752 746 739 731 721 713 708 706 707 711 714 711 705 694 680 659 632
17 15 12 11
9
9
9
9
7
5 10 12 15 18 20 18 18
7
9 12
763 783 792 794 798 798 796 792 787 780 773 767 763 76C 759 757 752 745 735 720 706 689 675 674 6'99
16 11
7
6
8
8
7
8
9
8
9
9
8
6
7
6
6
4
4
6
9 12 11
9 16
8CS 816 822 825 827 828 828 825 821 817 813 808 803 799 794 786 776 767 756 743 731 721 717 721 743
7
6
7
7
7
5
5
6
6
7
6
6
6
7
7
6
6
6
6
7 10 11 12 13 21
20N
831 836 841 841 842 844 843 840 835 832 828 823 818 811 804 797 788 780 774 767 758 756 758 768 788 820
7
7
8
8
7
4
5
4
4
6
6
8
9
9
9
9
9
6
6
8 13 11
11 14 17
14
130
Figure A81.
140
150
160
170E
180
170W
160
150
140
130
7-year average values and standard deviations of the sea-surface temperature. October 1951-1957.
120
(OF x 10)
110
SON
-44-to-"3-3
__:
--
----
--
L
.
.
.
447 44 450 4.3 4. 46. 472 4A2 494 507 520 52
_
lOf7-~E~A~A~019 484 491 4§4is
6
11
41
1 6
4?16
5 4#5
514 526
12 49
17 498
17 449to 502Y3*4
is
4s
A4 538 407 556 555
S3
2 543
to
598
1 5 65
612 59
18 ~13
16 4 S4A
17 5 19 5819 520 4581
16
13 6016 6019 49961
16 11 5976 575
I
713 16
1113
4 14 1 5 1 11
1
'JOB 61 -3 76'9 7C9 705 699 643--964 673 665 6- -6ii
S-
30C
2
z
7
6 7
8
7
7
8
6
5
10 . 10
811 820 828 833 834 830 828 824
S. . ......
4
S1o0
Figure A82.
40
0
9
29Q1_el
4
9
-
9
a00
7
1060
. .........
16 730 727 726 725 725 723 714 701 688 614 658 655 6 6
9
6
5
8 9
790 799 80 & 6--8
o 808 666 804 60o 7 f789 184 iTf 16-T3 f67 761 7s5 t4
S6
2CN
. 144
726 745 758 762 764 783 762 759 755 750
11 13 16 16 13
9
1
11
66 67 1 6T4 673 665--655 638 613 613
8
sOS
7
"
8
8
8
?78@
. 1
11
"16~
4 166 .
7
7
9
727
9
10
a 11 798 t53
9
7
6
6
15
146
1 0
1
0
16
It7766 699 6498 716
14
4
12
12
2
10
15
739 739 74 767 803
14 12 10 11 1O
f
. 12 6
7-year average values and standard Beviations of the sek-Surface temperature. November 1951-1957. (oF x 10)
10
7-YEAR AVERAGE
ANC STANDARC
OEVIATION OF SST
CECEEER 1951-1957
416 417&-14 413 421 438 453 461
12
11
11
11
13
12
10C 13
V
5C0
412 412 421 424 427 430 434 436 440 446 458 471 483 4'
14 14
9
9
9
7
6
7
7
9 12 12 11
'442 421 439 450 456 461 47C 474 474 476 480 483 486 492 501 511 518 521
-19 15
12
12
12
10
8
9
10 .11
11
10
11
12
14
13
10
9
557 550 559 563 564 565 565 562 558 555 555 557 560 566 573 575 576 568
12
9
6
7 10
9
7
9 12 11
10 11 13 14 16 16 12
6
669 665 664 661 660 656 648 641 636 633 632 635 639 643 643 638 631 614
9
9
8
7
6
5
6
8
10
11
10
12
13
15
16
14
10
7
688 711 724 725 724 724 723 719 717 713 709 706 703 701 7CC 702 702 699 692 680 667 651 634 633 6'53
24 13
7
5
8
9
6
3
3
5
7
8
9
9
9
9
8
8
9
6
7 10 12 i5
751 762 768 773 776 778 777 772 769 767 764 762 759 756 752 750 747 737 726 713 699 683 673 679 696
11
6
4
3
4
4
4
6
8 10 12
9
7
6
6
7
6
5
6
5
9 11 11 10 21
2CN
792 800 806 811 812 812 810 807 804 799 793 790 787 783 778 775 770 760 750 742 731 720 719 734 757 783
8
5
5
3
4
4
4
6
5
6
7
6
5
4
5
5 10 11 10 10 12 12 13 12 16 14
130
Figure A83.
140
150
160
170E
180
17CW
160
150
140
130
7-year average values and standard deviations of the sea-surface temperature. December 1951-1957.
120
0
( F x 10)
110
7-EAR AVEFAGE
AND SIANARD
JANUARY
DEVIATION OF SAT
I551-1957
IV
559]9
to
41
15
9
376 374 36
19 21 39
310 302 316 151 319 359 418 425 437 445 449 449 452 455 458 46t 465 467 469
13
422
-
No-
28
346 359 373 3e81 382 34.8 394 399 405 410 412 415 419 429
21 13
8
9
10 13 14
50N
35N
37
2O
-
6 403 404 424
15 24 20 24
29
25
28
522
20
31
17
13
12
12
14
15
17
48 468 484 497 5C4 50,4 513 115 515 619 523 54 524 524 519 51
23 222322
22
1
2
19 18 14
9 10 11 1
1
443 461 497 517 526 533 547 59 566 513 577 577 579 518 581 587 593 594 692 588 582 571 557
38 '
f117r11 12 12
I1
T17
14 4 14 12
15 E
14 O
- 16
11
12
8
9 -12 13
12
5
33
6920
18 15
10
6
.
630 6
643
7
5
9
46
11
446
11
64
12
11
644
11
41 650851
11 11
9
7
652646 634
6
7 12
2 605
14 13
560
11
9
11
e
610
10
25K
654 670 681 691 702 708 711 712 712 713 712 7017 704 708 712 707 703 697 687 674 e62 640 624 624 646
29 18 11
4
7
8
9 11 12
12 15 14
5
12
4
5 11 1
18 11
8 1i
20t
736
13
46 751' 75A
12
10
Figure A84.
15
8
140
764 764 765 763 759 1571 755 75? 749 748 746 742 737 726 712 707 702 686 676 684 703 731
5
8 11
10 10
11 13
9
6
4
6
5
5 11 14 17
5
9
2
8 13 15
150
160
_170E
18C
170
160
150
140
130
_120
7-year average values and standard deviations of the surface air temperature. January 1951-1957. (oF x 10)
110
7-YEAR AVEPAGE AND STANDARC DEVIA
FEBRUARY
"r"
ION CF SAT
1951-1957
_352
55A
19
-
237
50t%
25
)3305
45N
12 29
40A
399 3
22
24
325 336
28
25
*et~
351 365 376 384 389 394 399 403 406 409 418 429 439.
24 20
19
18 14 13 14 13 12 12
11 14
17-2
56 376
27
~
3480' i58 368 373 376 387 399
24
28 23 10 10
13 18
26
398 419 434 443 451 452 453 455 459 465 468 470
23
21
18
16
15
13
12
11
11
11
13
15
474
19
391 396 415 434 449 466 481 495 502 507 512 513 517 520 523 524 520 513 509
21
17 20 24 25 23 21 21
20 16 13 10 10 10 10
10 11
11 13
i
U'
35t
4 9 474 494 512 520 522 535 547 .555 562 568 57C 571 573 575 579 585 587 586 581 573 558 546 554
60'45 F26 18 12 11
14
18
18
15
14
15
8
8
8
7
6
8
9 11 10
15 13
9
30A
558 574 592 603 614 620 625 631 634 637 638 639 640 639 639 642 643 643 638 628 615 601 593 595 608"
26 19 16 11
8
9
10
12
10
8
8
e
10
9 10 12 13
9
7
6
4.
11 13
5
25A
659 6172 679 687 695 698 703 7C7 708 708 706 705 704 700 697 697 696 692 684 671 658 645 641 641 643
29
9 13 13
15 14
9
11
9 11 13 iC 10
9
9
8 10
7
9 11 11
7 16 21 23
20
741
--
746 749 755 760 762 764 781
12
11
130
Figure A85.
9
9
140
8
10
150
9
7
160
758 755 752
8
10
1708
13
752 751
IC
lC
8
-92 -703 726
744 736 734 731 726 720 709 696 686 64
9 10 .14 13 16 21 23 19
7 1. 10
7
4
17.%
160
150
140
130
120
7-year average values and standard deviations of the surface air temperature. February 1951-1957. (OF x 10)
110
7-YEAR AVEFAGE ANC STANCARC
OEVIA'TION CF SAT
MARCH 1951-1957
358 3546,364 369 369 371 372 3701
25
26
18
21
22
24 37
50
%
343
20
353 366
17
12
375 3E8 392 395 399 402 404 410 415 418
11
10
6
8 12
14
13
11
12
21
348 340 345, 356 373 393 413 43C 440 445 449 453 455 455 456 459 461 463 472
-12
11
10
7
10
12
14
13
10
10
13
16
17
16
15
14
12
9
13
430 428 440 453 467 4E4 500 509 514 516 516 519 522 521 518 516 513 508 508~
14
13
10
10
12
13
13
IC
8
IC
13
16
18
17
18
14
12
10
11
544 548 555 562
16
17
15
12
580 6C1 617 628 635 637 642
23
14
16
15
14
14
14
682 6t0 698
10
10
12
645
11
568 576
11
11
580 58C 578 579 580 583 588 586 581
10
S
11
13
15
16
15
13
11
647 647 645 643 644 644 644 645 647 643 636 622 609 597 585 588 606\
9
10
12
13
15
16
16
14
10
6
4
6
8
10
13
9
1
708 712 714 718 721 720 715 711 708 707 7C5 701 700 697 690 681 667 654 640 627 627 638
13
11
11
11
id
11
10
13
15
17
16
14
11
8
5
3
4
5
10
19
13
12
754 757 762 769 773 773 773 770 767 763 755 75C 749 742 738 734 727 717
7
7
7
8
7
7
7
0
10
9
10
14
18
15
11
6
5
10
120
Figure A86.
575 567 553 543
11
12
11
9
140
150
160
170E
18C
170k
160
150
710 701 690 679 671 679 696 721
13 12
9 11 22 18
16 14
140
130
120
7-year average values and standard deviations of the surface air temperature. March 1951-1957.
(OF x 10)
110
7-YE'R AVEFAGE ANC
APRIL
STANDARO
OEVIATICN
CF SAT
1951-1957
55
371 369 384 393 393 401 413 425
16
20
19
18
10
8
10
16
372 378 386 394 4C2 4C2 402 407 412 415 424 435 446 4614
11
8
7
5 7 12 15 14 12 10 10 11 14 11
501
45N
-405 387 388 395 404 415 426 436 448 452 452 453 455 459 466 473 479 485 495
-12
15
15
14
14
15
14
11
13
16
18
1,5 11
11
12
12 12
9
9
f
497 484 489 '94 4SS 505 509 515 521 522 520 520 522 525 529 531 t28 522 520
10
11
11
14
16
15
14
9
10
10
9
10
10
9
9
12
12
11
10
599 593 593 593 592 592 592 591 591 591 590 590 591 594 593 587 578
12
10
IC
12
13
13
11
7
5
6
6
8
8
7
9
12 13
564 551
11
10
626 6£4 672 674 673 669 668 667 665 664 664 662 659 657 656 657 656 652 645 634 621 60, 593 591 60
12
12
14
14
15
14
10
8
10
13
11
9 11 12 10
8
6
6
8
10
11
10
9
7
727 737 743 744 744 741 740 '737 733 733
11
11
14
16
18
15
10
6
11
12
2Ch
789 7S4 797 794 791 790 7E8
5
6
9
9
9
5
6
130
140
150
733 72e 722
11
12
14
717 715 714 709 701 690 678 662 651 638 630 637
13
11
8
7
6
8
12
13
8
10
10
7
e73 178 774 770 765 760 754 748 744 738 730 723 712 695
9
160
11
8
17CE
7
6
18C
8
6
170k
6
6
160
7
10
150
11
16
140
20
e88 683 685 701 731
16
10
e
10
9
130
Figure A87. 7-year average values and standard deviations of the surface air temperature. April 1951-1957.
120
(oF x 10)
110
7-EAR
AVEFAGE
AhC
MAY
STANCARC DEVI~1TION CF
SAT
1951-1957
411 412C*23 432 432 440 453 46
12 11
10 11
9
8
7 13
S39 402 407
17
16
14
418 427 430 432 437 443 448 458 472 485 494
12
9
7
7
8
9
8
8
7
7
11
'452 422 419 425 429 438 448 45E 466 469 471 474 478 486 496 505 512 519 530
-14
10
18
553 532 525
13 12
16
F18 16
525
18
18
17
13
526 532 538 542
19
18
18
17
12
8
8
545 543
15
11
939
8
10
11
11
10
9
7
9
1
539 542 549 554 556 554 549 '4
9
9
10
11
9
9
9
10
647 635 628 E26 626 628 630 626 622 618 612 611 612 613 613 608 599 585 572
10
13
16
18
19
17
18
17
15
13
12
14
16
14
10
8
9
10
10
688 7C5 717 717 712 709 707 705 7C3 702 699 693 687 682 680 683 679 671 663 651 636 619
12
8
8
9 13 14
15
14
13 12 13
12
12 12 15 19 19 14 10
9
8
7
0,5 607
7
9
6 "20
a
78C 714 774 778 779 778 776 773 769 762 753 747 743 736 734 735 728 716 704 689 674 655 639 637 641
18 13 11 11 12 12
12 10
13 12 11
11
12 12 15 17 13
8
6 11 13
7
7
9 11
20 h
&
822 E17 816 816 815 810 805 8C4 798 790 784 779 776 773 769 763 756 747 735 719 704 691 682 687 701 737
10
9
7
5
5
6
7
7
9
9
7
E 12
14
12
14 17 17
16 17 20 15 10
7
8
7
130
Figure A88.
140
150
16C
170E
18C
17C
160
150
140
130
120
7-year average values and standard deviations of the surface air temperature. May 1951-1957. (OF x 10)
110
7-'EAR AVEFACE ANC STANCDRC
CEVIATICh CF SAT
JUNE 1951-1957
55h
438 44311453 460 466 478 490 506
15
19
21 22 20 14 12 17
0 %0f
5CN
430 433 440
18
15
7
P
0 -
445 450 457 465 472 478 483 491 502 515
f
8 10 11 12 13 13 15 15 14
502 471 462 463 462 468 478 486 493 5C2 509 515 521 525 530 536 543 552 562
18 16
18 18 16
13 10 11
11 11 11 10 10 10 12
14 16 15 16
45h
603 582 513 568 563 565 571 517 580 584 591 595 597 598 596 591 585 579 575
22 18
18 20
21
17 15 15 15 14
12 13 13
9
8 12 16 15 13
702 690 621 673 669 687 668 671 670 671 676 678 677 670 658 642 625 607 591
17 17
19 23 24 21
17 19 22 22 20 17 14
8
5 11 16
18
14
746 757 764 766 767 760 753 749 747 745
13
9
11
10
9
10
13
15
14
12
817 814 816 818 820 815 8C8 8C5
iC
9
7
4
6
8
1C
10
743 742 738 734
8
11
16
18
802 799 796 792 785 775 767 757 745 733 718 700 684 675 664 656 664
8
6'
6
6
6
7
8
11
13
13
16
17
14
17
20
8
17
838 838 840 840 837 833 830 125 820 816 813
6
5
7
6
7
9
10
7
6
7 10
130
Figure A89.
14')
150
160
132 727 719 709 693 673 654 638 626 625 641)
15
9
8
9
11
14
14
11
I1 '11 A
170E
808 800 791 782 771 758 748 737 722 711 707 700 703 719 760
E
5
5
6
19 16
6 13 10
9 11 11 14 18 20
18C
170k
160
150
140
130
120
7-year average values and standard deviations of the surface air temperature. June 1951-1957. (OF x 10)
110
7-YEAR AVEFAGE AC STANCARD
DEVIATION CF
SAT
JULY 1951-1957
482 494, 510
15
476 474 4 80 485 490 5C1
14
13
12
13
'573 542 531 526 !20 524 534 542
42 25
16
15
18
18
19
16
9
8
1,5 16
521 528 535 540 544
17 15 17 15 1:
514 52-3 529 535 542 550 557 568(
9 13 13 11
12 13 14 15
547 557 566 570 574 575 578 582 582 5'87 592
10
6
10
13
18
17
17
17
18
13
12
670 651 640 633 628 63C 637 642 646 650 652 650 647 641 635 629 618 607 595
28
22
19
18 _19
21
24
21
15 _8
10
13
18
18
17
17
16
11
9
758 748 737 733
22
18
16
14
804 8C9 810
12
!o
0
8
841 137 833
9
10
10
806 802 796
9
12
9
791i 7E9 786
7
832 829 825 822
8
4
8
11
8
Figure A90.
140
150
7
726 725 731 730 T27 720 711 699 685 670 651 629 606
18
15
9
6
8
9
11
14
16
15
14
11
10
783 778
7
9
774 772
e
8
768 762
8
9
751 738 725 710 691 673 658 646 652 6l8'
10
10
9
9
13
16
11
14
15
822 820 815 e09 801 796 791 782 768 753 742 729 711 699 690 685 690 707
10
11
6
6
e
8
8
7
9
10
9
8
11
15
12
20
17
27
840 E38 839 840 839 836 833 '31
6
6
7
7
6
9
11
12
10
731 727
13
15
160
629
13
e21 819 815 809 804 796 783 770 760 747 734 725 720 722 131 761 798
7 10
7
8
10
10
9
9 12 13
7 11 10
e 11 21 17
170E .
1C
170w
160
150
140
130
120
7-year average values and standard deviations of the surface air temperature. July 1951-1957. (oF x 10)
110
7-YEAR
AVERAGE
ANC STANCARO
DEVIA'TION
CF SAT
AUGUST 1951-1957
55%
504 509 525 541 552 560 568 577'
9
2 11 13 14 15 17 18
SON
539
24
531 533
25
600 594 594 592 590
20
15
16
24
28
22
534 535 539 546
17
10
4
9
554 564 571 576 585 594
11
13
15
15
15
16
597t
13
592 598 601 604 609 613 616 61'8 617 618 618 613 601
30
23
15
10
15
17
17
15
15
17
18
12
12
704 696 694 690 6E8 688 688
14
9
12
21
23
23
17
687 686 6e88 689 686 679 671 661 649 631 606
12
11
16
16 13
13
15
16
18
14
14
785 779 774 770 764 762 757 753 748 745 742 735 724 711 694 675 650 622
7
6
7
10
10
10
5
4
7
9
9
9
12
14
14
13
13
13
815
824
13
6
827 822 816 813 809 PC6 802 798 795 791 787 778 770 764 754 738 723 706 690 670 64e 660 701'
4
4
8
5
5
5
4
6
7
6
839 824 834 833 828 826 826 826 823 820 817 813
5
4
3
6
9
8
7
8
9
10
9
7
6
6
11
807 797 786
8
10
15
9
9
7
6
776 763 747 731
12
7
7
8
6
8
10
717 705 691
10
12
15
13
12
11
81 698 739
17
14
15
'0
20
837 835 835 835 834 833 834 833 831 827 825 822
4
5
3
4
6
5
5
4
6
6
4
4
130
Figure A91.
140
150
'60
170E
18C
814 808 799 785 773 764 754 743 733 728 733 753 783 817
6
6
6
8
8
3
6
8
13
16
18
13
9
10
17Ck
160
150
140
130
7-year average values and standard deviations of the surface air temperature. August 1951-1957.
120
(OF x 10)
110
7-YEAR
AVEFAGE AND
STANCARD DEVIAFION
CF SAT
SEPTEMeER 1951-1957
55N
490 492C506 522 533 540 547
14 19 18 16 16
7C0
5 53
19
54C 527 520 516 516 522 530 540 552 564 572 577 583 5861
14 13 13 12
10
9 12 13 13 13 14 11 10
17
45"
0615 585 5,C
'-14
13
11
582 577 571 568 57C
15
15
17
19
17
692 682 680 676 668 661 656 653
9
12
14
16
18
21
22
19
771 772 77C 762 751 742
11
10
9
11
14
17
573 58C 589 597 605 .613 618 616 614 610 6CC
16
14
14
16
13
9
12
10
10
i12 '18
652 656 661 667 673 674 670 660 649 634
16
16
14
13
10
9
15
14
12
12
11
1,
734 727 722 723 726 727 727 721 710 696 680 660 635
17
13
12
13 14
10
6
11
17 15
14
14
11
784 757 804 807 809 809 807 o00 792 786 781 774 769 766 765 761 753 743 731 715 697-681 667 674
11
8
7
7
7
7
6
5
4
6
8
5
4
6 8 8
7
7
11
12
12
15
1.
12
822 827 829 827 826 827 825 821 817 814 810 804 798 793 787 779 766 757'747 729 712 704
10
7
9
7
8
6
7
7
7
5
4
5
7
8 10
11
9
8
11
12 13
16
V
831 E34 835 834 835 835 834
5
8
10
7
7
6
8
130
Figure A92.
140
150
7(34
832 827 822 819 814
8
7
4
3
3
6
160
17CE
18C
702
22
\
718 754
15
11
D,
807 800 791 780 773 765 752 T42 79 -42
6
7
14 14 11 13 .13 12
11 15
72
17Ch
120
160
7\ 0A
150
140
130
79
8
7-year average values and standard deviations of the surface air temperature. September 1951-1957. (OF x 10)
a
81i
7
110
7-V'EAR AVEFAGE
ANC STANCARO
DEVIATION OF SAT
CCTCEER 1951-1957
450 451
23
1V
28
58 466 481 495 502 5071
31
32
23
20
25
30
......
o
488 478 475 473 473 475 480 488 499 511 525 535 541 547C
15
14 14
c1 10 11 16 20 23 22 23 23 23 -.17
.
550 530 527 530 529 523 521 521 521 522 527 538 548 557 569 579 585 582 572
1 20
17 13 15
17 18 16
14
8
9 10 14 17 20 17 14 11
9
645 6366 627 620 623 623 621 616 610 604 6CO 6CC 605 612 619 626 631 631 629 617 598
9
9
8
7
8
9 12
13 12 10
18 21 2C 17
12 11
15 18 19
12
1
6 9 676 694 7C7 715 717 72C 716 712 7C6 697 687 680 679 681 683 687 689 685 677 665 648 626 624
6
8 12 14
10
7
17 15
13 12 12 10 10
18 18
13 14
17 '20 20
12,114 14 12
725 741 756 765 773 776 776 772 766 759 752 744 737 735 733 733 731 725 716 704 690 673 i45 658 614
6
6
8 10 12
13 -11
9
9 11 11 11
12 13
13 11
10 10 11
9
2( 16 11
8
781 7 2 803 808 811 811 8C9 F07 803 798 793 785 780 777 774 768 759 749 738 725 712 702 696 698 723
18
5
8
9 10
IC 10
10 11 10
8
7
9 10 14 14 13 10
1C
9 10
6
4
5
5
20h
813 819
8
7
130
Figure A93.
825 828 830 828
6
5
140
5
5
150
826 t24 F23
5
5
160
9
0Q%
819 815 80E 8C2 799 794 785 773 766 760 751
9
170E
7
7
18C
8
8
17C
7
7
160
9
10
150
10
9
140
739 738 741 749 171 804
14
11 13 16
13 15
130
7-year average values and standard deviations of the surface air temperature. October 1951-1957.
120
(OF x 10)
110
7-YEAR AVEFAGE ANh STANCARD DEVIATIC
CF SAT
NCVEIreER 1951-1957
55F
403 407 413 421
19
415 412 412
24 23 18
45t
25
27
23
430 440 455 465S
25 25 26 35
420 425 433 441 451 461 471 486 500 504(
13
11
13
13
12
16
19
18
18
22
453 440 436 144 453 455 456 46C 467 473 480 489 498 508 519 530 539 537
'21
16
17
16
17
15
16
15
13
11
10
9 11
16
21
20
15
i0
537 535 538
9
14 2C
547 552 553
18
16
15
640 643 649 654 656
10
14
17
13
12
551 547
17
17
546 549 553 562 574 581 589 593 592 561
18
16
15
10
11
17
21
19
14
8
9
653 647 638 630 k31 634 640 648 652 652 647 637 621 60,2
12 15
19
16
11
9
12
18
16
13
9
10
15
16
666 682 697 709 720 724 727 727 726 721 715 708702 701 703 705 704 700 693 679 665 651 638 640 662'
23
19
12
11
12
11
9
7
8
9
12
15
15
12
10
5
7
10
11
10
10
10
14
11
1*
741 754 767 778 786 787 784 780 177 771 765 76C 756 754 754 751 742 732 721 707 694 161 674 678 696
2C
13
8
9
12
10
9
9
9
9
13
14
13
10
7
6
6
9
10
11
13
13
17
15
14
2C
788 EC1 811 816 820 819 815
12
7
6
5
5
5
9
130
Figure A94.
140
15C
00
ll1 806 802 799 794 789 785 780 772 762 755 748 739 730 719 716 729 750 783
8
8
8
14 13
13
5
9
13
10
9
9 10
7
6
9 15
15 14
e10
170E
18C
170W
160
150
140
130
120
7-year average values and standard deviations of the surface air temperature. November 1951-1957. (oF x 10)
110
7-'EAR AVEFAGE ANC STNCARD CEVIATIC
CF SAT
1951-1957
CECEMBER
367
20
3630362 365 377 390 406 424
24
23
18
10
11
19
23
36 371 382 391 396 4CC 404 406 411 419 430 444 458
16
8 11
12 12
24 22 22 2C 16
16 15 12
_'3 6 7 365 375 391 404
23
30
36
35
415 430 439 445 451 455 460 465 472 482 495 503 503 502
12
11
11
12
12
11
12
10
14
14
12
19
13
22
458 4 3 458 464 478 '94 505 512 519 521 522 524 528 532 535 543 552 557 558 550 541
6
6
8 10 12 14 16 16 14 11
12 11
15 13
16 15
29J 5 19 19 18
11 537 562 581 585 594 6C2 609 609 606 603 601 601 603 605 609 615 618 613 607 593 575
4 6
8
9
7
13
20
18
16
18
13
13
11
14
14
14
14
14
15
19
17
28P 21
55rJ'9
7
12
589 613 64C 656 667 673 678 66O 680 678 674 672 671 671 670 672 673 670 664 652 642 628 416 615 61
13
10
10
8
7
11
13
13
14
13
12
11
12
11
12
12
11
11
17
16
14
19
25
32
690 7C5 719 731 740 745 748. 743 740 740 735 73C 730 729 727 726 720 711 701 688 675 663
13
5 12
6
5
7
9
8
8
9 11
9 10 11 10
10
11 12 11
14 12
18
ic
20h
758 710 780 787 791 793 791 784 780
11
10
6
8
7
6
14
]0
7
130
Figure A95.
140
150
160
777 770
12
10
767 767 764 761 757 750 741 730 721 710 701
5
5
9 15 15
7 12
5
3
E
8
170E
18C
170%
160
150
140
130
653
13
655 671
14 18
97 706 729 755
16 12
18 17
120
7-year average values and standard deviations of the surface air temperature. December 1951-1957. (OF x 10)
110
7-YEAR AVERAGE ANC STANDARC DEVIATION OF
4SST
JANUARY 1951-195
55N
-14 -14a.-12 -10 -10 -13 -15 -14)
8
10
7
5
5
3
3
5
h,
-16 -11 -12 -11 -11 -11 -12 -12 -12 -13 -13 -14 -15 -1
5
3
1
2
3
2
3
3
3
3
3
3
4
6
-29 -23 -25 -22 -19 -18 -16 -13 -11 -11 -11 -11 -12 -12 -13 -15 -16 10
11
6
6
5
7
9
9
5
3
4
3
3
5
5
4
3
-40 - 0 -31 -31 -30 -28 -26 -23 -22 -20 -17 -15 -14 -14 -13 -15 -15 -16 -18 -19 -17
7
1
11
8
9
7
6
6
8
10
9
6
5
5
5
6
8
8
6
4
6
-47
2
40 -31
14
-26 -25 -26 -28 -28 -28 -27 -24 -22 -20 -19 -18 -18 -17 -18 -18 -18 -19 -19 -16
4
7
5
6
5
5
4
5
5
6
6
6
6
6
6
7
6
5
4
5
14
6
-33 -29 -25 -25 -25 -26 -27 -27 -26 -24 -23 -22 -21 -20 -19 -18 -17 -16 -15 -15 -16 -16 -14 -15 16
7
5
2
3
4
3
4
3
3
2
4
6
6
5
5
4
4
4
3
4
6
9
8
-15 -18 -21 -22 -21 -22 -Z2 -21 -20 -18 -17 -17 -16 -16 -14 -14 -14 -12 -11 -11 -11 -10 -10 -18 -24
6
2
2
3
3
2
2
3
2
3
4
4
3
3
3
3
3
3
3
5
8
10
12
12
10
20 N
-9 -14 -16 -14 -12 -14 -15 -14 -14 -14 -12 -11 -12 -12 -11 -11 -12 -11
4
2
3
6
5
2
3
2
3
3
3
2
4
5
5
4
7
8
130
Figure A96.
140
150
160
170E
180
170W
160
150
-8
7
-9 -10
8
9
140
-8
9
130
-8 -16 -21 -21
9
12
12
8
120
110
7-year average values and standard deviations of the change in sea-surface temperature between months. January 1951-
1957. (OF x 10)
7-YEAR AVERAGE ANC
STANDARC DEVIATION OF
A SS
FEBRUARY 1951-1957
-10 -1Icki10
S4
6
-9
10
-7
8
-6
5
-4
5
-
4
8
-7
5
-6
5
-5
6
-5
5
-4
4
-4
3
-4
3
-4
4
-4
4
-6
3
-7
3
-7
4
-6t
5
.5 -14 -11 -13 -14 -12
,8
6
9
5
3
6
-9
7
-7
7
-5
6
-4
4
-2
4
-2
3
-1
4
-3
4
-6
4
-8
5
-9
5
-7
3
-9
9
-7
9
-5
7
-5
5
-4
4
-3
5
-5
5
-7
4
-8
3
-9
2
-8
3
-
8
S4 -1
-12
-13 -15 -15 -15 -15--13 -11 -10
151310
5
4
3
4
5
5
4
5
-9
5
-7
5
-6
6
-7
6
-6
5
-6
5
-6
5
-7
4
-9 -10
3
2
-9
4
-7
4
-6
3
-11
10
-7
6
-14
23
15
-18-18-17 -15 -15-12 10
4
5
4
6
3
-9 -11 -10
8
7
3
-9 -11 -11 -11 -10 -10 -10
3
3
4
3
3
3
2
-9
2
-7
4
-7
4
-7
5
-6
5
-5
4
-7
3
-7
2
-7
2
-9
2
-7
3
-6
4
-5
3
-1
-5
-7
-6
-7
-8
-8
-7
-6
-6
-6
-5
-6
-7
-6
6
-6
-6
4
-5
-6
-7
5
4
-6
4
4
-5
4
-7-10
5
5
4
4
5
5
5
5
4
2
2
3
3
4
6
4
5
-2
3
-3
2
-4
3
3
-6
4
-6
2
-6
3
-6
3
4
-4
4
-4
3
-4
5
-4
5
5
-5
3
-7
5
-7
6
-6
8
-6
5
-8
5
1
4
20N
-3
-3
3
130
Figure A97.
3
140
150
-5
-5
4
160
-6
170E
180
5
-4
3
-4
170W
160
150
-3
140
130
120
-8
5
-9
5
110
7-year average values and standard deviations of the change in sea-surface temperature between months. February 1951-
1957. (OF x 10)
7-YEAR AVERAGE AND STANDARC DEVIATION OF
MARCH
LI SST
1951-1957
55N
4
1
1
2
4
14
17
14
12
8
6
7
6
2
'II
1C -2
Vj
50N
0
7
-1
3
-1
3
-0
4
0
3
1
2
1
3
1
3
1
2
2
1
1
2
0
3
2
3
21
6
7
7
6
5
3
4
0
5
-1
5
-1
3
-1
4
0
5
1
5
1
3
1
2
1
3
0
4
1
3
1
2
-0
3
-0
3
1
5
7
2
6
12
4
6
2
5
1
3
-0
-1
5
00
4
2
4
3
5
2
3
2
3
2
5
2
5
2
3
1
2
-0
1
-0
2
-0
3
-
5
,7
5
4
8
2
5
2
4
1
3
1
5
1
5
3
4
3
4
4
4
4
4
4
4
4
5
3
5
3
5
1
4
0
3
-1
2
-1
2
-1
4
1
-4
45N
1
7
5
7
18
10
9
10
6
9
3
7
3
5
2
4
3
5
3
5
5
5
5
6
4
7
4
6
5
5
4
5
3
5
4
4
4
3
4
1
4
-0
3
-1
4
-1
4
-2
3
-1
3
-1
14
9
13
6
12
6
11
4
10
4
8
3
7
4
7
4
7
6
5
6
5
5
6
4
5
4
3
4
4
3
4
4
3
3
2
3
0
4
-1
6
-0
5
1
6
-2
7
-4
4
-5
5
15
3
14
2
12
3
11
4
7
2
6
1
6
3
5
4
4
6
3
7
3
7
2
4
1
3
1
2
2
5
3
3
%2
2
1
4
-0
5
0
6
1
8
1
8
-1
7
-2
7
-2
9
130
Figure A98.
140
150
160
170E
180
170W
160
150
140
130
120 ,
7-year average values and standard deviations of the change in sea-surface temperature between months. March 1951-
1957. (OF x 10)
-1
7
110
7-YEAR
AVERAGE ANC STANDARD DEVIATION OF
A157-
APRIL 1951-1957
\i
V
45N
8
5
20
5
18
4
15
3
16
3
201
5
6,
N"#
'6'
7
6
7
4
8
2
8
3
9
4
10
3
11
3
12
4
13
4
14
3
14
4
16
4
20
5
20C
8
35
11
26
7
17
5
14
5
11
5
8
4
7
3
6
3
6
5
5
4
5
4
6
5
7
6
9
6
12
6
14
6
17
6
18
6
16
7
40
11
40
8
35
7
29
4
19
4
15
4
12
5
10
5
10
4
10
6
9
6
8
4
8
1
7
3
7
6
9
5
10
5
12
3
13
3
13
3
11
4
38
27,'13
34v
8
31
5
27
7
22
5
1
5
17
4
17
6
17
6
18
7
17
7
16
6
15
6
14
7
12
8
11
8
11
6
11
4
10
2
10
3
10
3
1Q
3
10
4
32
14
31
5
31
5
28
6
24
8
24
7
23
7
23
7
23
9
23
8
23
7
21
6
19
5
17
6
16
7
15
8
13
7
11
5
10
3
10
4
10
4
9
5
8
6
8
5
6
31
5
27
4
27
4
26
7
24
8
24
6
23
5
22
6
20
P
18
8
16
7
15
5
14
5
13
5
13
4
12
5
11
5
10
4
9
5
9
8
10
7
9
7
7
6
5
4
1
7
25
5
22
4
20
4
19
4
16
3
14
4
12
3
12
4
11
5
9
4
8
5
4
8
4
8
4
9
5
11
7
13
8
11
7
7
10
7
12
9
7
8
4
5
5
3
7
$7
do
130
Figure A99.
140
150
160
170E
180
170W
9
7
160
150
140
130
120
7-year average values and standard deviations of the change in sea-surface temperature between months. April 1951-
1957. (OF x 10)
8
5
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF 4 SS3T
MAY 1951-1957
6
9
55N
%a
50N
45Nh
51
12
47
10
46
11
28:3'28
12
13
29
16
32
12
35
7
37
6
371
7
22
6
19
5
18
3
19
4
19
3
20
3
23
4
25
3
26
4
27
5
29
6
32
8
33
6
34
6
39
8
33
4
28
6
23
6
19
4
17
3
15
4
14
5
15
4
18
3
21
4
23
6
23
5
25
4
27
7
28
7
30
7
42
8
33
4
29
7
25
7
21
6
20
5
20
6
21
6
23
5
26
5
28
7
28
7
28
6
28
4
26
4
24
5
23
5
20
4
0
20N
44
1
4
9
41
10
39
8
34
8
33
9
31
9
29
8
28
6
30
7
33
8
34
6
36
5
36
8
35
8
33
5
28
2
24
3
20
5
18
6
14
7
15
5
51
10
42
5
38
3
39
3
41
3
41
3
40
5
39
7
37
8
35
6
34
5
33
5
32
7
32
6
31
5
28
5
26
4
24
3
20
4
16
5
13
5
12
5
12
5
15
2
16
34
a
33
6
36
6
36
3
36
2
36
4
35
6
35
5
33
5
32
3
29
3
27
3
25
4
23
4
20
5
16
6
14
5
13
6
12
6
10
3
9
3
6
7
6
7
13
4
14
6
23
4
23
4
25
5
24
4
23
3
21
3
21
3
21
2
21
3
20
3
18
3
16
4
17
4
16
4
13
4
11
4
10
5
10
5
9
3
7
3
8
7
8
8
6
6
10
5
11
5
130
Figure A100.
1957.
2
8 53
19s12
140
150
160
170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the change in sea-surface temperature between months. May 1951-
(OF x 10)
16
5
110
7-YEAR AVERAGE AND STANDARD DEVIATION OF
;SS
JUNE 1951-1957
\iV
55N
50N
45N
39c;42
7
7
43
8
47
8
50
8
47
8
41
4
36
3
31
4
32
6
31
7
33
6
37
6
39
6
41
7
42
6
43
7
41
7
37
4
34(
4
1
9
57
12
55
7
48
5
41
5
38
5
37
6
36
7
33
7
36
8
40
7
43
6
44
9
43
9
42
9
39
9
37
7
34
3
57
57
56
54
49
45
43
43
43
45
49
51
52
50
46
42
38
33
28
2
5349,,j47
9
11 P0
48
11
49
9
49
7
49
5
48
3
46
3
45
5
48
6
52
6
53
4
53
5
51
5
47
4
41
3
35
4
30
6
25
5
21
4
15
3
16
5
52
6
46
5
43
3
41
6
43
9
42
9
41
7
41
6
41
7
39
7
39
6
39
6
39
5
39
5
37
5
33
5
29
4
24
3
20
3
17
4
15
4
16
5
17
6
20
5
26
8
28
7
33
7
34
7
30
7
29
9
29
9
29
7
30
7
30
8
29
7
28
7
26
5
24
5
23
5
22
3
18
4
15
3
12
2
10
4
9
5
9
6
13
6
18
9
21
7
28
13
12
3
17
4
19
6
17
5
17
6
18
6
20
4
21
4
21
5
19
3
18
3
16
3
15
5
14
4
12
3
10
6
%7
8
4
10
3
11
8
10
10
11
11
6
14
4
18
4
23
10
53
5
8
20NK
34
6
130
Figure Al01.
140
150
160
170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the change in sea-surface temperature between months. June 1951-
1957. (OF x 10)
26
6
110
7-YEAR AVERAGE AND STANDARC DEVIATION OF
4 SST
JULY 1951-1957
55N
33
6
35C-42
7
7
44
8
44
5
44
4
41
2
34
5
N
V
01
50N
45N
P
158
8
51
9
48
10
46
10
44
8
44
6
44
6
46
5
47
4
48
4
47
6
44
6
39
6
31
5
69
12
08
6
68
4
64
7
61
11
59
14
57
13
56
10
56
7
56
7
55
7
54
6
53
4
50
6
46
8
41
9
32
7
64
63
64
63
62
62
61
58
57
56
56
54
51
47
43
39
35
27
17
7
5
4
5
7
8
9
8
7
7
9
9
8
4
4
6
9
7
4
6
55 J51 ,48
17'L
0
6
47
7
50
7
52
7
53
8
52
9
52
8
52
7
48
9
45
11
42
11
39
9
37
8
33
7
30
6
29
7
27
8
25
8
21
6
18
6
20
6
39
8
35
5
33
6
32
7
29
7
30
7
31
8
32
8
32
6
31
5
30
5
28
7
27
10
25
9
22
9
21
8
18
4
15
4
15
4
16
5
17
7
16
5
15
5
23
3
34
5
13
6
12
6
12
5
11
5
10
4
11
4
12
4
13
4
13
3
13
3
13
2
14
1
14
4
13
4
13
8
12
9
11
6
9
5
9
6
10
4
12
4
13
5
16
9
25
8
38
4
2
4
1
4
0
4
2
3
4
5
7
5
8
4
8
4
9
5
10
4
11
4
11
4
9
5
9
5
11
5
10
6
10
5
11
3
12
5
12
6
15
8
21
9
24
7
29
7
63
65
11/
doq
20N'
130
Figure A102.
140
150
160
170E
180
170W
10
6
160
150
140
130
120
7-year average values and standard deviations of the change in sea-surface temperature
between months. July 1951-
1957. (OF x 10)
29
6
110
7-YEAR
AVERAGE ANC STANDARC DEVIATION OF
JS'
AUGUST 1951-1957
iiV
45N
8
7
10
6
8
5
6
6
8
7
9(
8
a'P
33
6
29
6
26
7
21
4
19
6
16
5
15
6
17
8
20
8
21
7
20
6
18
5
17
5
12t
6
30
15
33
9
36
9
38
10
34
12
29
12
25
8
26
8
22
8
21
7
22
10
24
12
25
9
25
7
24
4
22
3
16
4
2
0
24
12
26
11
30
10
32
10
32
1C
28
11
23
10
20
8
17
6
15
6
16
6
18
8
20
9
21
6
22
4
21
2
21
2
18
3
14
5
17
18
13
lr- 10-Q 9
16
9
19
12
23
11
26
10
24
3
20
8
17
7
15
5
10
4
6
4
7
4
9
4
11
5
13
3
15
2
17
3
18
3
19
4
18
5
17
4
13
4
6
6
7
4
8
4
10
6
11
7
13
7
12
5
11
4
9
3
7
4
7
4
7
4
7
5
7
6
8
5
9
3
11
2
14
3
16
4
15
4
13
4
11
5
13
4
20
8
&
-4
9
-5
8
-2
5
-1
5
-1
7
1
7
2
6
3
6
4
6
5
4
7
4
8
3
8
3
6
5
6
5
7
4
8
3
10
5
9
4
8
3
7
9
7
12
15
10
27
13
-3
-5
9
-6
9
-4
8
-2
7
-0
5
0
3
1
2
4
3
5
3
5
3
6
2
7
3
6
3
5
2
6
3
%
6
7
10
11
13
9
6
9
5
11
6
11
7
15
7
19
9
-4
8
8
130
Figure A103.
1957.
0
6C
6
23
29
5
10C
20N
7
5_
140
150
160
170E
180
170W
160
150
140
130
120
13
6
110
7-year average values and standard deviations of the change in sea-surface temperature between months. August 1951-
(OF x 10)
7-YEAR AVERAGE ANE STANDARC
DEVIATION OF
-SST
SEPTEMBER 1951-1957
-14 -150:-24 -26 -25 -28 -27 -22
10
12
10
7
8
9
9
12
55N
%
V
6
,o
50N
-18 -18 -19 -20 -20 -20 -21 -23 -23 -21 -20 -19 -18 -15(
8
9
9
7
5
4
5
6
6
6
8
8
8
7
45N
'-37 -29 -24 -20 -16 -19 -22 -24 -25 -27 -28 -27 -25 -22 -17 -14 -11 -10
"23
18
12
8
6
9
11
7
3
5
8
9
10
9
9
8
7
5
-34 -28 -21 -19 -18 -21 -25 -28 -29 -31 -31 -28 -25 -19 -14 -10
13
10
10
9
8
10
10
8
5
7
10
10
9
8
7
6
-46 -35 -31 -28 -22 -21 -17 -17 -18 -21 -24 -26 -27 -25 -23 -19 -15 -10
7
5
4
5
3
8
9
9
9
9
8
7
7
7
6
7
9
7\
-30 -23 -21 -19 -14 -13 -12 -12 -14 -14 -15 -16 -15 -12
4
4
5
5
6
4
5
7
6
6
6
5
4
9
-17 -12 -11 -10
5
4
4
4
20N
-8
5
-5
5
-4
6
130
Figure A104.
-5
7
140
-8
4
-7
4
-7
4
-8
6
-8
5
-7
6
-7
6
-6
5
-4
4
-2
4
-5
6
-4
5
-4
4
-5
4
-5
4
-4
5
-3
5
-1
5
-0
6
-1
150
160
170E
180
5
170W
-6
6
-2
4
-7
4
-4
4
0
4
4
3
6
3
\-2
4
-9
7
-6
6
-4
4
-1
3
2
2
4
3
5
4
6
5
8
5
0
4
-4
11
0
5
1
5
1
3
3
2
5
2
5
3
6
4
8
6
9
8
5
7
0
11
-1
4
-1
3
-2
5
-1
5
1
5
2
5
3
6
3
5
3
9
-0
7
-2
9
160
150
140
130
120
7-year average values and standard deviations of the change in sea-surface temperature between months. September
1951-1957. (oF x 10)
-4
7
110
7-YEAR AVERAGE ANC STANDARC DEVIATION OF
j
rS 7-
CCTOBER 1951-1957
-26 -27d-34 -37 -38 -39 -38 -3%
6
6
6
6
6
6
8
10
&
V
5CK
-43 -40 -37 -35 -34 -35 -37 -40 -43 -42 -40 -38 -36 -32(
9
7
5
5
5
5
6
8
8
7
7
8
8
7
45N
-55 -54 -49 -44 -43 -43 -42 -42 -43 -46 -48 -49 -46 -42 -38 -35 -31
15
11
12
11
10
9
9
11
12
13
13
12
10
8
7
6
5
-52 - 0 -50 -50 -47 -44 -41 -41 -41 -42 -44 -46 -48 -48 -46 -42 -37 -32 -27 -22 -17
1010
7
7
9
10
8
7
7
8
9
11
12
10
9
7
6
4
4
4
6
42
-6
-50 -45 -40 -41 -40
21-1125
5
6
6
7
4
-38 -36 -35 -36 -38 -40 -41 -41 -39 -36 -31 -27 -23 -19 -15 -12
6
4
4
2
3
5
6
7
6
5
4
4
3
3
5
5
4
5
-41 -37 -33 -31 -30 -30 -29 -29 -29 -28 -28 -30 -30 -30 -28 -25 -21 -17 -15 -14 -11
11
6
4
3
4
5
4
3
4
4
4
5
4
3
4
4
4
3
3
3
3
-8
5
-8 -13 -19
8
7
1
-28 -24 -20 -16 -15 -16 -16 -17 -19 -19 -19 -19 -20 -17 -14 -12
6
6
7
5
5
5
6
7
7
6,
5
4
4
3
5
5
-9
4
-8
3
-8
4
-6
3
-4
4
-4
7
-6 -13 -21
11
8
12
-18 -12
5
6
-6
6
-7
8
-9
8
-6
3
-4
6
-8 -10 -15 -18 -14
9
9
7
9
7
130
Figure A105.
-8
6
-7
5
140
-8 -10 -10 -11 -13 -13 -11 -11 -10
5
4
4
4
3
4
5
4
5
15b
160
170E
180
-8
5
170W
-6
5
-5
5
160
150
140
130
120
110
7-year average values and standard deviations of the change in sea-surface temperature between months. October 1951-
1957. (OF x 10)
DSS7-
7-YEAR AVERAGE AND STANDARC DEVIATION OF
NCVEMBER 1951-1957
55N
\ ) 0
V
50N
) -31C&35 -38 -40 -37 -35 -35'
9 11
8
5
6
8
6
9
0
00,,
-46 -39 -35 -33 -32 -33 -35 -36 -39 -41 -41 -40 -39 -361
11
10
6
3
3
4
6
6
7
8
9
7
6
5
45N
-59 -54 -52 -49 -42 -37
12
10
8
7
8 11
N
-36 -36 -36 -36 -38 -40 -42 -43 -42 -41 -36
10
8
6
7
7
8
8
10
.9
7
4
-53 -50 -52 -52 -50 -48 -45 -40 -37 -36 -37 -37 -38 -40 -42 -41 -39 -37 -34 -29 -2
11
6
8
7
8
11
12
13
12
9
9
9
8
8
6
8
8
7
3
4
N
-50 -48
13
N
-38
8
N
-29 -27 -27 -26 -26 -25 -25 -27 -26 -25 -25 -24 -22 -22 -21 -18 -15 -15 -15 -15 -16 -20 -22 -21 -24
4
3
4
4
5
4
3
3
3
5
5
4
4
4
3
5
5
3
3
2
5
8
9
7
7
20 N
-45 -42 -43 -44 -44 -43 -39 -38 -37 -36 -36 -37 -38 -38 -38 -35 -31 -28 -25 -23 -21
9
6
5
6
7
7
9
9
8
8
7
7
8
7
6
5
5
4
4
3
4
-36 -34 -35 -37 -37 -37 -37 -35 -33 -32 -31 -30 -30 -30 -28 -25 -24 -22 -20 -19 -19 -20 -21 -23
7
6
4
6
6
3
4
4
4
5
5
5
5
6
5
4
4
4
2
3
4
4
4
7
-20 -19 -17 -16
3
4
5
5
130
Figure A106.
19
5
140
-15 -16 -17
5
3
2
150
-17 -16 -17 -18 -17 -16 -14 -13 -12
3
2
3
4
4
4
3
3
3
160
170E
180
170W
160
-9 -10 -12 -13 -14 -18 -20 -17 -16 -19
6
5
4
3
5
7
9
6
4
4
150
140
130
120
110
7-year average values and standard deviations of the change in sea-surface temperature between months. November 1951-
1957. (OF x 10)
7-YEAR AVERAGE ANC STANDARC DEVIATION OF
A-SSE
CECEVBER 1951-1957
\i~I
5CN
-22 -256-26 -27 -27 -26 -27 -29
9
6
10
10
9
9
9
11
-31 -25 -21 -20 -19 -19 -21 -23 -25 -27 -28 -29 -29 -26t
9
6
6
4
3
3
5
5
6
6
8
8
7
5
45N
-43 -39 -36 -34 -28 -23 -22 -20 -18 -19 -21 -24 -27 -29 -29 -31 -28
7
8
7
5
6
6
4
5
4
3
4
6
8 10
9
7
5
12
-
48
2C17
7
11
5
6
4
4
5
6
6
7
8
7
5
6
9
10
8
6
6
4a
4 -41 -40 -39 -40 -38 -36 -35 -33 -30 -28 -28 -28 -28 -29 -29 -29 -26 -25 -24 -21 -18
9
7
8
6
6
4
5
6
7
7
7
8
8
6
6
7
7
5
4
5
5
-39 -39 -37 -36 -35 -35 -36 -36 -35 -33 -32 -30 -28 -27 -26 -25 -25 -24 -23 -21 -20 -21 -20 -20 8
3
8
5
5
5
5
5
4
4
4
4
5
4
5
5
5
4
3
3
5
6
5
6
-27 -27
2
20n
2
9
-29 -30 -29 -28 -28 -29 -29 -27 -24 -23 -22 -20 -19 -18 -17 -16 -15 -15 -16 -18 -19 -18 -21
2
4
5
4
4
4
3
2
3
4
4
4
6
5
4
4
4
4
6
7
5
6
9
-16 -19 -21 -22 -21 -18 -17 -18 -17 -15 -14 -14 -16 -15 -16 -15 -13 -12 -11 -12 -14 -15 -17 -15 -16 -22
1
2
1
2
3
2
2
2
1
4
5
4
4
5
7
7
6
5
3
6
9
7
4
5
7
8
130
Figure A107.
140
150
160
170E
180
170W
160
150
140
130
120
7-year average values and standard deviations of the change in sea-surface temperature between months. December
1951-1957. (OF x 10)
110
HEAT ACVECTICN FRCM BALANCE
JANUARY
EQ ,. (CALCM-21,AY-1)
1951-1957
551\
61
9
107 -29
289 248
218
29
20K
$
-10
-9
-8
73 105
-28
-43
-26
-8
-6
28
-86
-95
-76
-80
-83 -88
-39
-51
-64
-68
-78-102-124-112
459 235
41
-99-122-147-171-171
285 131
13 -14
130
Figure A108.
140
150
160
96
-12
-96-132
-58
25
-12
107
6 -50 -96
-59 -48
3 6 -15-109 -75 -20
183 16,2
-14
87 -10
10
86 20'
19
346 206 187 169
-61
25
54
7
505 443 257 163 10 3
32 -!4
78
33
-33
-10
35 -17 -42 -64 -92
-87
-90
-84 -53
-87
30
151 177
113 128 154
18
20
48
-97 -82 -41
-94
-81
-10 -13
-21
4 -22
-33 -15
-20
-18
-49 -55
-32 -11
-60 -48
60
86
39 105 172
160
150
140
130
-29 -26 -11
170E
-7 -32
180
170W
-34
-54
8
22
c
-76-108'
52 -81-166
92 -23-131
120
110
7-year average values of heat advection determined from the heat balance equation. January 1951-1957. (cal/cm2/day)
HEAT
ACVECTICN FRCV BALANCE
EQ%.
(CALCM-2AY-1)
FEBRUARY 1951-1957
-9-10C5
115
V 64
63
19 131
319 275 297
538 453
20N
(
379 370
323 317 301 217
437 285
174 141
123
123 152 134
93
174
130
Figure A109.
140
69
129
'8
77
187 17(
85
37
43
14
31
41
60
71
84
82
90
86 124
70
50
51
61
66 104
68
33
19
15
58
65
163
86
70
77
39
52
64
18
-7
-12
5
12
192 130
92
103
87
43
44
15
9
-11
-4 -37
-59 -48 -53
27
7
7
38
49
61
52
60
23
19
4 -22 -17
-1
4
14
33
71
72
45
49
83
60
-3
-2
-2
60
60
69
69
50
50
49 117
116
55
-19
160
67
43 113
114
86
269 209 204
3-2 240 240
22 -35
150
82
9 -36
-49 -35
170E
-24 -52
180
-39
170W
160
150
56
2 -64 -51 -72-101-125
140
-46
-91-173-189
7 -24 -80-117-155
130
7-year average values of heat advection determined from the heat balance equation. February 1951-1957.
120
110
2
(cal/cm /day)
HEAT ACVECTICN FRCM BALANCE EQN.
(CALCM-20AY-1)
MARCH 1951-1957
7
38
125 117
93
74
94
85
87 143
89
42
59
68
68
81
77
143 125 108
87
90
97
91
93
60
111 125 113
94
74
55 -23
154 165
162 158
119
99
65
33
-4 -81-151
187 181 206 172 108
111 143
148 158
113
89
89
70
-9 -76-145
111 123 124
114
142 109
101 125
108
150
140
130
169 14 c1
206 154 132
45N
-150 189 227 225
230 225 217 220 180 156 140
371 323 253
229 231 201
168
182 213
179 105
Figure A110.
113
508 474 447 348 285
246 230
140
125 111
228 221 223 201
350 362 295
130
154 151 138
101
311 356 4C0 369 294
248 337 325
181
185
131
88
150
97
214 203 216 258 226 178
165 184
84
160
80
170E
180
130 127
17CW
112
160
171 103
126
146 214
-6
84
33 -69-147
120
7-year average values of heat advection determined from the heat balance equation. March 1951-1957. (cal/cm2/day)
110
HEAT ADCVECTICN FRC
BEALANCE EQJ.
(CALCM-2CAY-1)
APRIL 1951-1957
176 131
55N
5fN
99 214 239
2C4
95
149
24 -19
130
Figure A111.
85
95
90
13
42
65 103
70
55
66
73
75
76
86
77
89 121
85
' 69
65
63
97
88
59
74
64
45
17
6
10
16
33
42
61
81
60
187
215
192
190
165
157
165
146
111
64
50
31
31
50
48
58
56
20
353 295
236 209
197 145 1 71
148
107
78
60
62
68
71
67
46
43
19 -27 -97
237 187 154
112 102
103
81
76
99
108
75
77
92
81
61
55
51
35
21
90 123
68
29
79 103
93
124 111
84
76
66
68
60
50
20
14
-2
37
6"
101 125
136 157 143
78
61
150
140
76
33
38
15
27
13 -12
140
150
'~
160
3t
170E
180
44
170W
160
-11
27 -47-148
125 155
130
7-year average values of heat advection determined from the heat balance equation. April 1951-1957.
-88-153
72 -60-135
120
110
2
(cal/cm /day)
HEAT ACVECTICN FRCP
MAY
BALANCE EQN.
(CALCM-2jAY-1)
1951-1957
66 114 141
V
-
50N
-75
459
24
0c
-31
-44
-35
-27
39
44
62
80 155 156
-57
-54
-35
-51
-55
-57
-39
16
23
32
61
-7 -22
-30
-28
-39
-46 -39 -14
61
75
15 -16
-34
-53
-41
41
112 140
176
-50
1591
100 105
94
118 113 104
79
35
133
94
53
-3
-95-
-27
-45
-49 -22
-12
-3
13
-3 -15
19
91
111
98
97
67
35
-3
184 -40 -98-109 -73
-46 -57
-13
27
61
71
53
68
100
88
59
74
74
36
13
25
51
5C
63
72
78
43
30
60
99
106
85
76 117 118
140
130
25 -70
-99 -37
130
140
Figure A112.
0 -ku
-20
5 -70 -63
118
3
54
!
-44
27
235 332
20N
9 -16
. 0
-52
15
8 111
227
-48
a ,
205 21
15
150
25
47
22
160
170E
180
17CW
160
150
-8
3 -19-143
68 -15
120
2
7-year average values of heat advection determined from the heat balance equation. May 1951-1957. (cal/cm /day)
-64
110
cz::
HEAT ACVECTICN FRC
BeALANCE
EQN.
(CALCM-20AY-1)
JUNE 1951-1957
\j
50N
-69 -38 -36
45N
-57
-21
-13
-13
-26
2
19
-3
-90 -62 -25 -15
24
18
15
-13
-34 -41
-40 -55
-45
-50 -58
-64 -87 -96
3
27
13
-8
-7 -15 -27
-33 -19
3
37
79
69
56
47 -53-119- 130 -81
-48
-5
1
-68
-66
-21-149-181- 139 -90 -31 -53
2CN
-32
-32-103
-98 -73 -62 -39
130
140
Figure A113.
0
-96 -44
25N
150
5
160
1 -1
-37
55N
11
49
-9 -27 -31
45
170E
180
-40 -63
-21
-57
-5
4
23
46
54
15-128
36
34
15
8
16
27
26
23
17 -99- 11
-35 -12
-2
11
66
71
47
9
25
32 -11
-35
9
3
57
83 104 105
82
45
34
35
19
36
81
77 117 218
171
80
53
-60 -68 -77 -66
-2 -29 -33
-55 -57 -40
2
17CW
160
150
140
130
-59-182
32-105
9
-7-135
120
110
2
7-year average values of heat advection determined from the heat balance equation. June 1951-1957. (cal/cm /day)
HEAT ACVECTICN FRCP BALANCE
EQN. (CALCM-20AY-1)
JULY 1951-1957
55N
50N
-15
4' N
-53 -36
-9 -17
-21
18
38
18
30 -1
3
9
31
62
56
22 -33
-6
29
61
61
48 -12-168
35
19 -22 -38 -54
-52 -56
-43
-22
18
45
53
70
13-14
25
26
17
-1 -18 -34
-45 -46
-42
-7
0
13
10
45
47 -37-
-90 -50 -29 -27
-26
-31
-20 -19 -27
-30 -17
10
9
33
45
24
51
63
-72
-35 -69-104
-95 -90-108-113-129 -81 -51
Figure A114.
-6
64
-15
-123-108-126-131-123
140
-3 -20
23
-42 -44
35-109 -19
130
-8
4
-33 -42 -54 -38
-27
-17C-119-167-127
-0
1
150
-61
160
-85 -27 -25 -27
-49 -36 -33
i70E
-16
180
1
25
-5
34
54
66
53
70 102
131 122
41 -70-166
-30 -27
27
43
66
79 124 174
160 117
12-101-136-146
17CW
160
150
140
130
120
7-year average values of heat advection determined from the heat balance equation. July 1951-1957. (cal/cm2/day)
110
HEAT ACVECTICN FRCM
BALANCE EQN.
(CALCM-2CAY-1)
AUGUST 1951-1957
55N
-21 -61
-14
-45
-73 -63
92
-54
-28
-60
14
31
16 -35
130
140
Figure All5.
6
1
6
-1
-14
-5
6
5
7
-47 -51
-29 -19
-24 -15
-36 -40 -32
-38 -52
-48 -55 -53 -45 -53
-56
-83-103-107-110-112 -89
-38 -32 -47
-37 -30 -66
-37 -30 -6
150
-5
-27 -41 -38
-87
-49
-56
-90-100-112-106-110 -94 -97
-49-102- 107 -64 -?0-123-134-132 -98
-124
-38 -40 -33 -25
-16
-129-136-103-1C0- 103 -82 -!0
$
-20 -29
-98 -76 -77
-98 -76 -77
160
170E
-62
-62
-56
1
1
180
-45
-9
6
28
28
16
16
17
17
17CW
-53 -25
-75 -57 -28
-93 -60 -17
8
52
-48 -46
65
10
27
104 115
-5(
5 -41-1
7 -26-15
1
-6
46
60
125 116
76 132 154 207 227 202
76
32 154 207 227 202
160
150
140
159
159
130
-57-101
42 -43-152
92
22-114
89
89
-3 -87-128
-3 -87-128
120
'110
7-year average values of heat advection determined from the heat balance equation. August 1951-1957. (cal/cm2 /day)
HEAT ACVECTICN FRCM BALANCE EQN.
SEPTEMBER
(CALCM-2DAY-1)
1951-1957
55N
-64 -570o
-109
45N
214-170-114 -74
23 -44
20N
-72
'103 -81
-7
14 -22
-16
1
-94 -79 -85
-53 -55 -48
-4
-17
6
-57 -44 -87 -78 -40
-16
-3
-3
6 -20
-36
-73-123-146-143-121
-60
-24
-18 -24 -58-11
-33
-72
-74
-51
-46
-A6-121-131-121-110 -63
Figure A116.
150
160
-90-118
-98
-80
-75-106 -93-103-109 -65 -56 -53
-46 -95-104
140
-69 -96-104 -81 -67 -61
170E
-54 -77-14
-24
-77 -65 -5C
130
-99-108-123-126-123-120-125-121 -76-106
-56 -96-138-184-174-162-141-113 -78
-62-107-151-139-115 -94
-90 -85 -73
97-115-126-127-101 -6
180
-44 -74
170W
-26 -18
-33 -30
-26
-47
-8
-23 -20 -36-133-197'
-20
3
0
26
42 -13-117-208
-92 -72 -76 -74
-26
20
51
73
160
150
140
130
58 -68-117-136
120
110
7-year average values of heat advection determined from the heat balance equation. September 1951-1957. (cal/cm 2 /day)
HEAT ACVECTICN FRCM BALANCE
OCTOBER
EQN. (CALCM-2CAY-1)
1951-1957
-134-162-165-128-122-104-120-151-188-163-163-132 -59
)182-205-129 -62
-46
178
$
20N
72
-18
145 -20 -75
-29
-54
130
Figure A117.
2
-51
-6 -17
140
-60
-92-110-139-146-158-2C4-257-266-200-164-108 -62 -83
-38
38
57
158 117
72
10 -25 -18
10 -39
22
-49 -75 -91
8
-8 -27
-39
-60
-82-124-179-232-202-151-102
-75
-78 -61
-80 -79
-91 -82
-61 -41
6 -53
-40 -60 -63 -78 -77
-95 -33
-19
i70E
180
-83 -94 -76
150
160
-5
-31
-75
-60
-65
-60
-57
-84-122-123-118-106 -99-111 -91 -47 -74
-89 -86
-85
-91
-85 -86 -93 -44
-6 -10 -61
-48 -27 -51
-48 -45 -48 -28
27
27 -12
-35 -34
-15
-38
42
3 -15
17CW
160
-51
150
-42
1
140
130
-92\
-61-108
-71-122-153
120
110
7-year average values of heat advection determined from the heat balance equation. October 1951-1957. (cal/cm 2 /day)
HEAT ACVECTICN FRCM BALANCE
EQN.
(CALCM-20AY-1)
NCVEMBER 1951-1957
55N
5fN
-128-105-101
45K
-16 -73 -27 -12
-95-108-162-217-235-278-271-236-199-139 -69
-82-109 -83 -94-186-286-331-368-426-349-270-235-156 -63
3
132
1 4
181 180
247 225 103
45
32
2 -78-156-237-305-327-277-226-176-163
-64 -1
-45 5
12
206
342 389
340 239
165
99
33
7
-80 -62
20C
159
1C4
106
-16
-8
193
284 147
77
47
-4
36 -42 -94-107-138-186-167-119
-15
-3 -91-131-187-222-188-186-186-143
-44 -51 -54-115-128-118-122-161-141-103
-81-104
-97-106-125-120.-74
-94
8
-91-102
-90
-51
-45
-54
-4
-6
25
17 -20
-19 -15
-62-105
140
130
120
110
-28
-96-110-106-118
2
20h
122
23 -32-102-109-128-147-169-160-155-154-132-126-124-105
130
Figure A118.
140
150
160
170E
180
170W
-52
160
13
150
7-year average values of heat advection determined from the heat balance equation. November 1951-1957.
2
(cal/cm /day)
HEAT
ACVECTICN FRCM 8ALANCE EQN.
CECEMBER
(CALCM-2AY-1)
1951-1957
55N
-40 -95-112-153-102
o
24
55
3
co
-85 -91
45N
-31
-11
q2
208 188 159
279 208
20
-65 -51
-53 -61-119-146-166-136 -80
-71 -47
7T
11-123-122 -4C -66 -78
-71 -86-120-144-138-121-104 -78
48
-5 -88 -31
-98-101-122-144-123 -92-119 -82 -31
1 -33
-24
-62 -46
-87-115-143-136-133-152-179-9-159-180-168-156-129-108
342
165
384
153 -33-152-178-196-265-250-205-194-137-100 -68
298 101
130
Figure A119.
-2 -29 -54 -97-173-209-179-165-2C1-182-159-148-127 -95 -73
-4 -8q-127-113-113-100 -51
140
150
160
-36 -24 -17
170E
180
-60
-86-107 -91-121-153-122-100 -A61\
-54 -48 -10
51
33
33 -15
-65-115 -39
74
66
87
17CW
160
-99
150
7-year average values of heat advection determined from the heat balance equation.
93
140
-61-125-107 -55 -75
63
41
130
31
60
120
-55
110
December 1951-1957. (cal/cm 2 /day)