1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Intraseasonal latent heat flux based on satellite observations
Semyon A. Grodsky1, Abderrahim Bentamy2, James A. Carton1, and Rachel T. Pinker1
Revised for Journal of Climate
March 23, 2009
1
Department of Atmospheric and Oceanic Science, University of Maryland, College
Park, MD 20742
2
Institut Francais pour la Recherche et l’Exploitation de la Mer, Plouzane, France
Corresponding author:
senya@atmos.umd.edu
40
Abstract
41
Weekly average satellite based estimates of latent heat flux (LHTFL) are used to
42
characterize spatial patterns and temporal variability in the intraseasonal band (periods
43
shorter than 3 months). As expected, the major portion of intraseasonal variability of
44
LHTFL is due to winds, but spatial variability of humidity and SST are also important.
45
The strongest intraseasonal variability of LHTFL is observed at mid-latitudes. It weakens
46
towards the equator reflecting weak variance of intraseasonal winds at low latitudes. It
47
also decreases at high latitudes reflecting the effect of decreased SST and the related
48
decrease of time mean humidity difference between heights z=10m and z=0m. Within the
49
midlatitude belts the intraseasonal variability of LHTFL is locally stronger (up to 50 Wm-
50
2
51
passing storms and is locally amplified by unstable air over warm SSTs. Although
52
weaker in amplitude, but still significant, intraseasonal variability of LHTFL is observed
53
in the tropical Indian and Pacific Oceans due to wind and humidity perturbations
54
produced by the Madden-Julian Oscillations. In this tropical region intraseasonal LHTFL
55
and incoming solar radiation vary out-of-phase so that evaporation increases just below
56
the convective clusters.
57
) in regions of major SST fronts (like the Gulf Stream and Agulhas). Here it is forced by
Over much of the interior ocean where the surface heat flux dominates the ocean
58
mixed layer heat budget, intraseasonal SST cools in response to anomalously strong
59
upward intraseasonal LHTFL. This response varies geographically due in part to
60
geographic variations of mixed layer depth and the resulting variations in thermal inertia.
61
In contrast, in the eastern tropical Pacific and Atlantic cold tongue regions intraseasonal
62
SST and LHTFL are positively correlated. This surprising result occurs because in these
1
63
equatorial upwelling areas SST is controlled by advection rather than surface fluxes. Here
64
LHTFL responds to rather than drives SST.
65
2
66
1. Introduction
67
Latent heat flux (LHTFL) links air-sea heat exchange with the hydrological cycle. This
68
evaporative heat loss term balances a significant portion of the surface heat gain due to
69
solar radiation (da Silva et al., 1994). Satellite sensors can measure sea surface
70
temperature (SST), near-surface winds, and humidity, and thus provide data for
71
estimating evaporation. Currently, several satellite-based global ocean latent heat flux
72
products are available (e.g. Chou et al., 2003 and references therein). In this study we
73
exploit the availability of a new global 16-year (1992 – 2007) record of weekly satellite-
74
based turbulent fluxes of Bentamy et al. (2008) to examine the observed geographic
75
distribution of intraseasonal LHTFL and its role in air-sea interactions.
76
77
Most observational examinations of LHTFL focus on its behavior on monthly and longer
78
timescales (e.g., da Silva et al., 1994; Yu et al., 2006). Recent studies of mid-latitudes
79
(Qiu et al. 2004) and tropics (Zhang and McPhaden, 2000) have shown that intraseasonal
80
variations of LHTFL associated with synoptic disturbances can alter SST by up to 1oC.
81
Modeling studies (Maloney and Sobel, 2004; Han et al., 2007) suggest that these SST
82
variations may in turn organize intraseasonal atmospheric convection and thus provide an
83
air-sea interaction mechanism for phenomena such as the 30-60 day Madden-Julian
84
Oscillations (MJOs) (Madden and Julian, 1994). Since LHTFL is also proportional to
85
evaporation its intraseasonal variations also contribute to variations of surface salinity,
86
thus increasing the impact of LHTFL on surface density.
87
3
88
In the tropics, high temperatures and thus saturated humidity combined with significant
89
synoptic variability such as MJO lead to significant intraseasonal LHTFL. MJO may be
90
driven in part by the evaporation-wind feedback (Neelin et al., 1987). MJO is a feature of
91
all tropical sectors, although it is most pronounced over the eastern Indian Ocean and
92
western Pacific Ocean in boreal winter and is strongly modulated by ENSO. MJO is
93
characterized by strong 2-4m/s fluctuations of surface winds and precipitation
94
(Araligidad and Maloney, 2008). As a result, it produces correlated fluctuations of both
95
LHTFL and shortwave radiation (SWR) with amplitudes of 30-50 Wm-2 and is observed
96
to cause 0.5oC fluctuations of SST (Krishnamurti et al. 1988; Shinoda and Hendon,
97
1998; Zhang and McPhaden, 2000). Moreover, recent research suggests that these
98
intraseasonal fluctuations may actively interact with lower frequency climate variations
99
in the tropics, just as in the Pacific, where the westerly wind bursts may trigger the
100
evolution of the El Niño/Southern Oscillation (ENSO) cycle (McPhaden, 2004).
101
102
Some of intraseasonal variability observed at low latitudes and subtropics is linked to
103
intraseasonal variability of midlatitude pressure systems. In particular, Foltz and
104
McPhaden (2004) have examined the intraseasonal (30-70 day) oscillations in the
105
tropical and subtropical Atlantic and found their link to fluctuations in the strength of the
106
Azores high. The subtropics and mid-latitudes are subject to synoptic meteorological
107
forcing originating in the mid-latitude storm systems. This additional variability has a
108
strong seasonal component, amplifies in the cold season, and varies from year-to-year. In
109
the Kuroshio extension region Bond and Cronin (2008) have found that in late fall
110
through early spring cold air outbreaks associated with synoptic events lead to intense
4
111
episodes of LHTFL and sensible heat loss. Similar origins of intraseasonal variability are
112
observed by Zolina and Gulev (2003) in the Gulf Stream region.
113
114
In summer and fall (when the ocean mixed layer shoals) cloud shading effects
115
accompanying synoptic disturbances become important sources of intraseasonal flux
116
variations. Based on experiments with a mixed layer model Qiu et al. (2004) suggest that
117
these summertime intraseasonal flux variations can induce SST variations with
118
climatologically significant ±1oC amplitudes. This and other observational evidence
119
suggest significant contributions by LHTFL variability in the intraseasonal band to the
120
state of the climate system. In this study we focus on geographical patterns of LHTFL,
121
consistency with SST, interplay with incoming solar radiation, as well as modulation by
122
longer period processes.
123
124
This study is possible due to several improvements to the climate observing system.
125
Beginning in the early 1990s a succession of three satellite scatterometers provides high
126
resolution surface winds. Brightness temperature estimates from the Special Sensor
127
Microwave Imager provide an estimate of relative humidity. When combined with
128
estimates of surface temperature it is possible to estimate LHTFL at weekly resolution
129
(Bentamy et al., 2008).Clouds and aerosols, the main factors affecting SWR, are available
130
from a variety of sensors flying in both geostationary and polar orbits (Rossow and
131
Schiffer, 1999; Pinker and Laszlo, 1992). Finally, an array of more than 90 moorings
132
distributed across all three tropical oceans (McPhaden et al., 1998) provides ground truth
5
133
at high temporal resolution which can be used to explore the accuracy of the remotely
134
sensed estimates.
135
136
2. Data and method
137
This research is based on the recent update of weekly satellite-based turbulent fluxes of
138
Bentamy et al. (2003, 2008). The three turbulent fluxes, wind stress ( τ ), LHTFL ( Q E ),
139
and sensible heat flux ( QH ) are estimated using the following bulk aerodynamic
140
parameterizations (Liu et al., 1979):
141
142
τ
 C D u a  u s (u a  u s )

143
QE
 CE u a  u s ( qa  qs )
L
144
QH
 C H u a  u s (Ta  Ts ) ,
C p
(1)
145
146
where  is the air density, L =2.45*106 J/kg is the latent heat of evaporation, C p =1005
147
J/kg is the specific heat of air at constant pressure. The turbulent fluxes in (1) are
148
parameterized using wind speed ( w  ua  u s ) relative to the ocean surface current
149
(relative wind speed is close to actual wind speed outside regions of strong currents), the
150
difference of specific air humidity and specific humidity at the air-sea interface ( qa  qs ),
151
and the difference of air temperature and SST ( Ta  Ts ). The lower indices (a) and (s)
152
indicate atmosphere at the reference level (normally 10m) and at the sea surface,
6
153
respectively. The bulk transfer coefficients for wind stress ( CD , drag coefficient), latent
154
heat flux ( C E , Dalton number), and sensible heat flux ( CH , Stanton number) are
155
estimated from wind speed, air temperature, and SST using the Fairall et al. (2003)
156
algorithm (COARE3 version). LHTFL is positive if the ocean loses heat, while QH is
157
positive if the ocean gains heat.
158
159
The variables needed for the evaluation of (1) are obtained from satellite measurements.
160
Wind speed relative to the ocean surface current ua  u s is measured by scatterometers
161
onboard the European Research Satellites ERS-1 (1992-1996), ERS-2 (1996-2001), and
162
QuikSCAT (1999-2007) (e.g. Liu, 2002). The humidity ( q a ) is derived from the Special
163
Sensor Microwave Imager multi channel brightness temperatures using the Bentamy et al.
164
(2003) method, while the specific surface humidity ( qs ) is estimated from daily averaged
165
SST. This version of LHTFL uses the new Reynolds et al. (2007) daily bulk SST while
166
the previous version of LHTFL (Bentamy et al., 2003) is based on the Reynolds and
167
Smith OIv2 weekly bulk SST. In the present version no correction is made for cool skin
168
and diurnal warming. The air temperature is determined from remotely sensed data ( q a
169
and SST) based on the Bowen ratio method suggested by Konda et al. (1996).
170
171
The turbulent fluxes are calculated using the COARE3.0 algorithm from daily averaged
172
values binned onto a 1° global grid over satellite swaths. Due to differences in sampling
173
by different satellite radars and radiometers, the final flux estimate is further averaged
174
weekly and spatially interpolated on a regular 1° grid between 80° S and 80° N using the
7
175
kriging method as described by Bentamy et al. (1996). The accuracy of the resulting
176
weekly fluxes is assessed by comparisons with in-situ measurements from moored buoys
177
in the tropical Atlantic and Pacific (PIRATA and TAO/TRITON), the northeastern
178
Atlantic and northwestern Mediterranean (UK Met Office and Météo-France), and the
179
National Data Buoy Center (NDBC) network off the U.S. coast in the Atlantic and
180
Pacific Oceans1. Quite high correlations (ranging from 0.8 to 0.92) are found between
181
satellite and in-situ LHTFL, while biases and standard deviations are generally low.
182
Standard deviations of satellite and in-situ LHTFL vary from 18 Wm-2 and 25 Wm-2. The
183
highest bias is found in comparisons with the NDBC buoys in the Gulf Stream region
184
where the time mean satellite LHTFL is 10 Wm-2 below in-situ values (or 7% of the
185
NDBC regional LHTFL mean). In the tropics satellite LHTFL overestimates in-situ
186
LHTFL by 8 Wm-2. These comparisons indicate significant improvements of the new
187
LHTFL product over the previous release described in Bentamy et al. (2003).
188
189
The same buoy network is used to evaluate accuracy of Ta retrieval. The time mean
190
satellite derived Ta is slightly colder than in-situ air temperature. The bias is weaker at
191
midlatitudes but magnifies up to -0.7 oC in the tropics (based on comparisons with the
192
TAO/TRTON and PIRATA buoy data). Standard deviation of daily average satellite and
193
in-situ Ta is around 0.6 oC and but is stronger (0.9 oC) in high gradient SST areas like the
194
Gulf Stream. As expected, weekly averaging decreases slightly (by around 0.1°C) the
195
standard deviation but doesn’t affect the bias much.
196
See ‘New Release of Satellite Turbulent Fluxes 1992 – 2007’ at
ftp://ftp.ifremer.fr/ifremer/cersat/products/gridded/flux-merged/documentation/flux.pdf
1
8
197
Intraseasonal signal is evaluated in few steps. First, the annual cycle is calculated from
198
the weekly data as a sum of the first three harmonics (Mestas-Nuñez et al., 2006). Next,
199
the anomaly is calculated by subtracting the annual cycle from the original signal.
200
Finally, the intraseasonal signal is calculated as the difference between the anomaly and
201
its 13 week running mean. This procedure retains periods shorter than 3 months that are
202
referred to as intraseasonal in this study. The variability of intraseasonal fluxes is
203
characterized by the running standard deviation that mimics the upper envelope of the
204
intraseasonal signal. Running standard deviation of the intraseasonal signal is calculated
205
using the same 13 week running window. Comparisons of the satellite intraseasonal
206
LHTFL with in-situ data from the TAO/TRITON moorings in the tropical Pacific, the
207
PIRATA moorings in the tropical Atlantic, and the RAMA moorings in the tropical
208
Indian Ocean are presented in the Appendix.
209
210
The LHTFL from this study is compared with LHTFL provided by the National Center
211
for Climate Prediction/ National Center for Atmospheric Research (NCEP/NCAR)
212
reanalysis (Kalnay et al., 1996), the Woods Hole Oceanographic Institution objectively
213
analyzed air-sea fluxes (OAFlux) of Yu et al. (2004) that combines satellite data with
214
model simulations, and with in-situ ship borne estimates collected by the International
215
Comprehensive Ocean-Atmosphere Data Set (ICOADS) of Worley et al. (2005). Mean
216
sea level pressure for this study is provided by the NCEP/NCAR reanalysis. In-situ
217
measurements from the TAO/TRITON moorings in the tropical Pacific Ocean
218
(McPhaden et al., 1998), the PIRATA moorings in the tropical Atlantic (Bourles et al.,
9
219
2008), and the RAMA moorings in the tropical Indian Ocean (McPhaden et al., 2008) are
220
also used for comparisons.
221
222
For several years now, uniform, long-term data from observations made from numerous
223
satellites relevant for inferring surface shortwave radiation (SWR) have been prepared as
224
homogeneous time series. The satellites that are being used for SWR retrieval usually
225
have between two to five channels in spectral intervals that are relevant both for inferring
226
SWR (visible) and for detecting clouds. Cloud data are provided by the International
227
Satellite Cloud Climatology Project (version D1) at a nominal resolution of 2.5° at 3hr
228
time intervals (Rossow and Schiffer, 1999). The original version of the SWR retrieval
229
scheme is described in Pinker and Laszlo (1992) and has been used at NASA/Langley for
230
generating the GEWEX/SRB product1. Since, several modifications have been introduced
231
to the inference scheme as related to aerosols (e.g. Liu and Pinker, 2008), data merging
232
(Zhang et al., 2007), and elevation correction (Ma and Pinker, 2009) (the SWR data used
233
in this study are derived with Version 3.3.1).
234
235
3. Results
236
Mean LHTFL and seasonal variations
237
First, presented are global patterns of the LHTFL and its annual and semiannual
238
harmonics. These components form the annual cycle that is used as a reference for
239
evaluating anomalies and intraseasonal signal. Spatial patterns of magnitude and phase of
240
these harmonics are similar to the Mestas-Nuñez et al. (2006) analysis that is based on the
241
three-year record (1996-1998) from the previous release of the LHTFL archive of
1
http://gewex-srb.larc.nasa.gov
10
242
Bentamy et al. (2003). Comparison of the time mean LHTFL from this study with the
243
time-mean LHTFL provided by alternative analyses (NCEP/NCAR Reanalysis, WHOI
244
OAFlux, and ICOADS) indicates reasonable correspondence of spatial patterns (Figs. 1
245
a-d).
246
247
Variations of LHTFL closely follow variations of the product of sea-air humidity
248
difference and wind speed qw (1). This product accounts for a major portion of LHTFL
249
variability as the Dalton number, C E , has weak dependence on wind speed (for winds
250
ranging from 4 ms-1 to 14 ms-1, Large and Pond, 1982). For the four products shown in
251
Fig.1 the time mean LHTFL is dominated by evaporation in the trade wind regions and
252
resembles the time average wind speed in the 30o S to 30o N belt (Fig. 2a). In the tropical
253
belt the humidity difference, q  qs  qa , is high (Fig. 2b) and has relatively weak
254
meridional variations (except in the eastern Pacific and Atlantic). Hence, LHTFL
255
variability is explained by winds.
256
257
SST impacts are evident in the equatorial eastern Pacific and Atlantic where the mean
258
LHTFL weakens due to the presence of cold tongues of SST maintained by the equatorial
259
upwelling. Local minimum of evaporation over the cold tongue regions is explained by
260
direct impact of cool SST on the sea surface and air humidities (Fig. 2b) as well as by
261
indirect impact of SST on the near surface atmospheric boundary layer that tends to
262
decelerate over cold water (Wallace et al., 1989). A stronger impact of SST on LHTFL is
263
evident across the subtropical fronts where temperature sharply decreases with latitude.
264
Poleward decrease in SST is accompanied by decrease in Ta because the atmosphere
11
265
boundary layer stability is close to neutral over much of the interior ocean. Hence, the
266
humidity difference (  q ) also decreases sharply poleward of 30o S and 30o N (Fig. 2b)
267
because both q a and qs decrease with temperature in accordance with the Clausius-
268
Clapeyron law. These meridional changes of  q explain weak LHTFL in the
269
extratropical oceans (Figs. 1a – 1d) in spite of rather strong winds in the northern and
270
especially southern hemisphere storm track corridors (Fig. 2a).
271
272
Regardless of the good correspondence of the geographical distribution of the time mean
273
LHTFL, the four analyses are somewhat different in magnitude. In the current analysis
274
LHTFL (Fig. 1a) has higher values in the trade wind regions than in the other three
275
analyses. This analysis is closer to in-situ ship observations from the ICOADS (Fig. 1d)
276
and the NCEP/NCAR reanalysis (Fig. 1b), but exceeds the WHOI OAFlux estimates by
277
20 to 40 Wm-2 (Fig. 1c). As a result this study suggests the highest estimate of the
278
globally averaged LHTFL and evaporation summarized in Table 1.
279
280
Table 1. Globally and time averaged LHTFL and evaporation.
Global mean
This Study
NCEP/NCAR
WHOI OAFlux
ICOADS 2.4
Reanalysis
LHTFL (Wm-2)
107
97
92
105
Evaporation
11.2
10.2
9.6
11.0
(cm month-1)
281
12
282
Strong time-mean latent heat loss (exceeding 80 Wm-2) is drawn from the warm Gulf
283
Stream waters off the east coast of the United States. Similarly, strong time-mean LHTFL
284
is observed near Japan over the warm Kuroshio (Fig. 1 a-d). In both of these regions the
285
LHTFL experiences the strongest annual variation peaking during the winter, when cold
286
dry continental air off-shore of North America and Japan crosses the Gulf Stream north
287
wall in the Atlantic or the Kuroshio SST front in the Pacific, respectively (Fig. 1e, 1f).
288
Semiannual LHTFL variations are prominent in the Arabian Sea and Bay of Bengal due
289
to annual reversal of winds forced by the South Asian Monsoon (Figs. 1g, 1h). The
290
monsoon flow in the Arabian Sea low-level westerly jet intensifies in boreal summer
291
while northeasterly winds spread over the region in boreal winter when the monsoon
292
ceases. Weaker semiannual variability is observed in the Caribbean low-level jet where
293
the easterly winds also intensify twice a year in February and again in July (Munoz et al.,
294
2008).
295
296
Magnitude of intraseasonal variation
297
As expected from (1), the intraseasonal variability of LHTFL is defined by intraseasonal
298
variability of winds (Fig. 2c) and sea-air humidity difference (Fig. 2d). But the spatial
299
distribution of the intraseasonal LHTFL (Fig. 2e) bears only partial correspondence to the
300
spatial distribution of intraseasonal winds or sea-air humidity difference. In particular, the
301
decrease in variance of intraseasonal LHTFL towards the equator reflects relatively weak
302
variability of intraseasonal wind at low latitudes. In contrast to low latitudes, the
303
intraseasonal variability of LHTFL decreases at high latitudes despite stronger wind
304
variability there. This behavior is explained by the spatial distribution of the time-mean
13
305
humidity difference that is weak over cold SSTs (and cold Ta ) of each hemisphere (Fig.
306
2b).
307
308
Although linear decomposition of the intraseasonal LHTFL suggests that wind
309
component (~ q  STD ( w' ) , Fig. 2f) accounts for a major portion of variability of the
310
intraseasonal LHTFL, neither component dominates globally. In particular, the air
311
humidity variability component (~ w  STD ( qa ' ) , Fig. 2g) peaks along the major SST
312
fronts and reflects an impact of moisture transport across the ocean SST fronts by
313
synoptic weather systems. SST itself (~ w  STD( qs ' ) , Fig. 2h) also impacts the
314
intraseasonal LHTFL along the major western boundary current fronts and in the Agulhas
315
current area. Both the mean LHTFL (Fig. 1a) and its variability (Fig. 2e) weaken over
316
cold SSTs where mean values of LHTFL are also low. This is particularly evident in the
317
cold sector of the Gulf Stream, in the Brazil-Malvina confluence region, in the subpolar
318
north Pacific, and in the Southern Ocean.
319
320
Intraseasonal LHTFL in mid-latitudes
321
The strongest variability of the intraseasonal LHTFL occurs in mid-latitudes where the
322
regional maxima are linked to areas of major SST fronts (Fig. 2e). In particular, in the
323
Atlantic sector the highest intraseasonal variance is observed along the Gulf Stream front.
324
Similarly high intraseasonal variability is observed in the Agulhas Current and in the
325
Brazil-Malvina confluence region. This suggests that the stratified atmospheric boundary
326
layer plays an important role in amplifying intraseasonal air-sea interactions. The
14
327
intraseasonal LHTFL variance changes seasonally and peaks in winter (not shown)
328
suggesting an association with midlatitude storms which also intensify in the cold season.
329
330
We next identify weather systems that are responsible for strong intraseasonal variability
331
of LHTFL in these regional maxima areas by projecting the intraseasonal LHTFL time
332
series spatially averaged over a particular index area onto atmospheric parameters
333
elsewhere. This regression analysis reveals correspondence between strengthening of
334
intraseasonal LHTFL in the Gulf Stream region and midlatitude storm systems in the
335
Atlantic (Fig. 3a). Increase in LHTFL drawn from the Gulf Stream region is associated
336
with the area of mean sea level pressure low and corresponding cyclonic anomalous
337
winds centered east of the region. The air pressure pattern is similar to that deduced by
338
Zolina and Gulev (2003) and by Foltz and McPhaden (2004) in their analyses of the
339
intraseasonal variability of the Atlantic winds. In fact, the anomalous wind in Fig. 3b
340
decelerates the northern flank of the northeasterly trades (where anomalous LHTFL is
341
somewhat weaker) and significantly accelerates off-shore winds over the Gulf Stream
342
(Fig. 3b). Maximum increase in wind speed is observed over the warm sector of the Gulf
343
Stream where winds further accelerate due to the atmospheric boundary layer adjustment
344
(Beal et al., 1997). In addition to intensification of mean winds, the anomalous
345
northwesterly wind outbreaks bring cold and dry continental air towards the sea.
346
Spreading of dry continental air lowers air humidity thus increasing the air-sea humidity
347
contrast (Fig. 3c). This, in turn, compliments the LHTFL increase due to stronger winds.
348
The ocean responds to continental air outbreaks by cooling SST north of the Gulf Stream
349
northern wall that is seen in decreasing values of qs (Fig. 3c). Intraseasonal winds have a
15
350
weak impact on SST south of the Gulf Stream temperature front where the ocean mixed
351
layer is deep and its thermal inertia is relatively strong.
352
353
Similar correspondence between intraseasonal LHTFL and atmospheric synoptic systems
354
has been observed by Bond and Cronin (2008) in the Kuroshio extension region in late
355
fall through early spring when cold air outbreaks associated with synoptic events lead to
356
intense regional episodes of LHTFL and sensible heat loss. Here we focus on regions of
357
locally strong intraseasonal LHTFL observed in the South Atlantic in the Agulhas current
358
south of the Cape of Good Hope (Fig. 2e). Similarly to what occurs in the Gulf Stream
359
area, increase of LHTFL over the warm Agulhas Current is linked to passing storms (Fig.
360
4). When the storm center locates to the east of the index area the anomalous southerly
361
winds bring cold and dry sub-Antarctic air northward. This amplifies the latent heat loss
362
due to increasing wind speed and increasing air-sea surface humidity difference.
363
Although storm systems are generally strong as they propagate around the globe in the
364
South Atlantic and the Southern Oceans, the intraseasonal LHTFL is stronger in the
365
Agulhas region and in the Brazil-Malvina confluence (Fig. 2e) in comparison to values
366
observed at similar latitude in the ocean interior. Both these areas host sharp SST fronts
367
that promotes higher  q and stronger LHTFL. It is interesting to note that the regression
368
analysis in Fig. 4 reveals a sequence of zonally propagating storm systems over open
369
spaces of the South Atlantic and South Oceans. The mean sea level pressure troughs in
370
the regression pattern are separated by approximately 90o in longitude suggesting the
371
zonal wavenumber of 4.
372
16
373
Intraseasonal surface fluxes and SST
374
The variability of LHTFL and SST is related. Exploring the relationships between the
375
two offers an additional tool to evaluate the consistency of the flux product. In fact,
376
LHTFL affects SST by affecting the net ocean surface heat balance. But, SST also affects
377
LHTFL directly through qs and indirectly by affecting near-surface winds that accelerate
378
over warmer SSTs. We next characterize the interplay between intraseasonal variations of
379
LHTFL and SST.
380
381
As expected, the LHTFL response to underlying anomalous SST is generally positive
382
(Fig. 5a), i.e., LHTFL increases in response to increased SST. This suggests a damping of
383
the underlying SST anomalies, although there are considerable geographical variations
384
(Park et al., 2005). The feedback exceeds 20 Wm-2/oC in the regions around 20oS and
385
20oN, but decreases at high latitudes and in the eastern tropical Pacific and Atlantic
386
where the time average LHTFL is also weak.
387
388
In contrast to the LHTFL response to underlying SST that is positive, the SST response to
389
intraseasonal variation of LHTFL is negative over much of the ocean (Fig. 5b) suggesting
390
cooling down of SST in response to increasing surface heat lost. But, in several regions
391
SST warms up in response to LHTFL increase. In particular, this behavior occurs in the
392
cold tongue regions of the eastern tropical Pacific and Atlantic Oceans. The relationship
393
between intraseasonal LHTFL and SST depends on the relative role the LHTFL plays in
394
the mixed layer heat balance. If this balance is local and governed by the surface flux, the
395
SST cools down in response to increasing latent heat loss (negative correlation when
17
396
LHTFL leads). This negative relationship dominates away from the cold tongue regions
397
and strong currents. In contrast, in the cold tongue regions the mixed layer temperature
398
balance is governed primarily by the vertical (upwelling) or horizontal (Tropical
399
Instability Waves, e.g., Grodsky et al., 2005) heat transports. Here the positive correlation
400
between LHTFL and SST is explained by the stratified atmospheric boundary layer
401
adjustment and associated wind acceleration over warm SSTs. Therefore, in the cold
402
tongue regions the LHTFL increases in response to increasing winds and SST rather than
403
SST responses to change in LHTFL.
404
405
Over the regions where the surface heat flux dominates the mixed layer heat budget, the
406
variations of LHTFL force variations of the mixed layer temperature and, thus, should be
407
negatively correlated with the SST rate of change, T / t , as seen in Fig. 6a1. The time
408
correlation of intraseasonal LHTFL and T / t is statistically significant over much of
409
the ocean2. It decreases at high latitudes where the upper ocean stratification is weak, the
410
mixed layer is deep, and SST response is weak. The time correlation is also weak in the
411
tropical Pacific and Atlantic Oceans in the regions where vertical and horizontal ocean
412
heat transports (rather than surface flux) dominate the mixed layer heat budget. Similar
413
but weaker correlation is found for the short wave radiation, SWR, (Fig. 6b)3. In fact the
414
SST rate of change is driven by the net heat flux across the air-sea interface, for which
415
the LHTFL and SWR are the major components. If these two components of the surface
416
flux are combined to better represent the ocean heat lost (LHTFL-SWR), the correlation
1
LHTFL is positive if the ocean loses heat.
The 99% confidence level of zero correlation is 0.1.
3
SWR is inversed to be consistent with the sign of LHTFL
2
18
417
increases (Fig. 6c) suggesting reasonable correspondence of intraseasonal flux variations
418
with intraseasonal SST.
419
420
Intraseasonal LHTFL and SWR
421
Both, the intraseasonal LHTFL and SWR agree reasonably with independent
422
measurements of the rate of change of intraseasonal SST. We next explore the global
423
correspondence between intraseasonal variations of the two surface flux components.
424
They are weakly correlated over much of the global ocean with an exception of the
425
tropical Indian Ocean and the western tropical Pacific where the intraseasonal LHFL and
426
SWR are negatively correlated (Fig. 7a). Coherent variations of clouds and winds are
427
evident in the Indo-Pacific warm pool where the eastward propagating Madden-Julian
428
oscillations are the most pronounced (Shinoda and Hendon, 1988). Lagged correlation
429
indicates that LHTFL increases in phase with decrease in SWR, suggesting stronger
430
latent heat loss and evaporation just below convective systems. Negative zero-lag
431
correlation of SWR and LHTFL is observed with both satellite flux data and in-situ
432
TAO/TRITON mooring data (Fig. 7a, inlay). It is consistent with the Zhang and
433
McPhaden (2000) analysis of the TAO/TRITON surface fluxes who also have found near
434
in-phase relationships among maxima in latent heat flux and minima in solar radiation
435
during passage of the MJO events. This phase relationship remains in place if higher
436
temporal resolution (i.e. daily) LHTFL (WHOI OAFlux by Yu et al., 2004) and daily
437
SWR are used instead of weekly averaged data (Fig. 7b). It may be noted that lagged
438
correlations in Figs. 7a and 7b have positive peaks at approximately 2 to 3 week lags
19
439
(also seen in Fig. 9) that reflect an oscillatory behavior of lagged correlation for periodic
440
processes like the MJO.
441
442
On one hand, the out-of-phase variations of intraseasonal LHTFL and SWR support a
443
hypothesis that the evaporation affects the humidity and therefore the cloudiness and thus
444
solar radiation at the sea surface. But theoretical considerations (see Zhang and
445
McPhaden, 2000 for a summary of existing approaches) suggest a lagged relationship
446
between intraseasonal LHTFL and SWR forced by MJO. In particular, in the Neelin et al.
447
(1987) model the maximum LHTFL is shifted to the east of the convective center, if the
448
mean wind is easterly. Explanation of the phase relationship between LHTFL and SWR
449
variations on the intraseasonal timescales is not clear. Next presented are observed
450
relationships between intraseasonal variations of SWR, LHTFL, and parameters affecting
451
LHTFL.
452
453
Coherent variations of the intraseasonal LHTFL lows and SWR highs (and vice versa)
454
are apparent in the time-longitude diagrams in Fig. 8. These accorded intraseasonal
455
variations propagate eastward between 60° E and the dateline at speeds ranging from 4.5
456
ms-1 to 7.5 ms-1, typical of the MJO. East of the dateline the correlation between LHTFL
457
and SWR is weak (Fig. 8). This zonal change of correlation is explained by the lack of
458
cloudiness east of the dateline that is the only major source of SWR variability.
459
460
Intraseasonal LHTFL variations are mostly driven by the intraseasonal variations of wind
461
speed (Araligidad and Maloney, 2008). Wind strengthens just below convective clusters
20
462
where SWR is low (Fig. 9a). Coherent variations of intraseasonal winds and SWR occur
463
mostly west of the dateline with a gap in correlation over the maritime subcontinent. As
464
expected, LHTFL strengthens in phase with winds that leads to out-of-phase relationship
465
between intraseasonal LHTFL and SWR (Fig. 9b). Evaluation of q a indicates that
466
intraseasonal wind fluctuations are not the only forcing of intraseasonal LHTFL. In fact,
467
q a also varies in accord with intraseasonal SWR (Fig. 9c). It may be suggested that
468
specific air humidity decreases below convective systems in response to cooling of the
469
near-surface atmosphere while the sea surface saturated humidity doesn’t change much
470
because of the thermal inertia of the ocean mixed layer. Difference in responses of q a
471
and qs leads to an increase in the vertical gradient of air humidity below convective
472
cloud clusters that, in turn, further enhances anomalous evaporation and LHTFL
473
produced by wind speed anomaly.
474
475
Intraseasonal and longer period variability of LHTFL
476
The interannual evolution of the ocean surface fluxes has been extensively studied. But, it
477
appears that amplitudes of intraseasonal fluxes are not stationary and experience
478
significant modulations by longer period variability. Noting that our dataset is only 16
479
years long, our study is limited to the tropical Pacific Ocean that hosts the ENSO and,
480
thus, displays significant interannual variability that can be resolved by relatively short
481
records. Interannual SWR anomaly is modulated by ENSO through zonal displacements
482
of convection. These interannual displacements of convection between the western
483
tropical Pacific and the central tropical Pacific produce SWR anomalies that are well
484
detected by satellite techniques (Rodriguez-Puebla et al., 2008). Because clouds are the
21
485
only physical mechanism driving the intraseasonal SWR, the amplitude of intraseasonal
486
SWR also shifts zonally following anomalously low SWR. In the central equatorial
487
Pacific the magnitude of intraseasonal SWR increases in-phase with warming of the
488
Nino3 SST (Fig. 10a, inlay). Here, the standard deviation of intraseasonal SWR increases
489
by up to 5 Wm-2 in response to a 1oC rise of SST in the Nino3 region (Fig. 10a). As such,
490
interannual variation of the amplitude of intraseasonal SWR reaches 15 Wm-2 during a
491
mature phase of El Niño when anomalous Nino3 SST warms up by 3oC. This interannual
492
modulation of amplitude of the intraseasonal SWR is comparable to the characteristic
493
amplitude of SWR variation by the MJO (Shinoda and Hendon, 1998).
494
495
In contrast to the amplitude of intraseasonal SWR that varies in-phase with El Niño, the
496
magnitude of intraseasonal LHTFL doesn’t have a similar significant in-phase variation.
497
Impact of El Niño on the intraseasonal LHTFL differs from its impact on the total
498
anomalous LHTFL that is enhanced in the eastern tropical Pacific, around the Maritime
499
Continent, and the equatorial Indian Ocean (Mestas-Nunez et al., 2006). In contrast, the
500
magnitude of intraseasonal LHTFL amplifies over the western tropical Pacific
501
approximately 8 months in advance of the mature phase of El Niño (Fig. 10b and inlay).
502
This amplification reflects impacts of the westerly wind bursts that often precede the
503
onset of El Niño, which were evident in advance of the 2002/03 El Niño and particularly
504
noticeable in advance of the 1997-98 event (McPhaden, 2004).
505
506
4. Conclusions
22
507
Although the major portion of the intraseasonal variability of LHTFL is accounted for by
508
winds, no one component (wind, air humidity, or sea surface humidity) dominates the
509
variability globally. In particular, contributions of q a and qs are significant along major
510
SST fronts due to moisture transport across the ocean SST fronts by synoptic weather
511
systems. Both the mean LHTFL and its intraseasonal variability weaken over cold SSTs
512
due to low air-sea humidity difference. In contrast, the strongest intraseasonal LHTFL is
513
observed over the warm sectors of SST fronts.
514
515
The strongest variability of the intraseasonal LHTFL (in excess of 50 Wm-2) occurs at
516
mid- latitudes where the regional maxima are linked to areas of major SST fronts. In
517
particular, in the Atlantic sector the highest intraseasonal variance is observed along the
518
Gulf Stream. Similarly high variability is observed in the Agulhas Current and in the
519
Brazil-Malvina confluence. Coincidence of the regional maxima of intraseasonal LHTFL
520
with SST fronts suggests the important role the stratified atmospheric boundary layer
521
plays in amplifying intraseasonal air-sea interactions. Temporal variations of the
522
intraseasonal LHTFL in these regional maxima are linked to passing midlatitude storms.
523
The intraseasonal variability of LHTFL forced by these passing storms is locally
524
amplified by unstable atmospheric stratification over warm SSTs.
525
526
Although weaker in amplitude but still significant intraseasonal variability of LHTFL
527
(standard deviation of 20 to 30 Wm-2) is observed in the tropical Indian and Pacific
528
Oceans. This variability is linked to the eastward propagating Madden-Julian
529
Oscillations. In this tropical region the intraseasonal LHTFL and incoming solar radiation
23
530
vary out-of-phase, i.e. evaporation enhances just below the convective clusters. The out-
531
of-phase relationship between the intraseasonal LHTFL and SWR is observed west of the
532
dateline, while east of the dateline both intraseasonal LHTFL and SWR are weak and
533
their relationship is not significant. Both intraseasonal variations of wind speed and q a
534
contribute to this out-of-phase relationship. Intraseasonal wind speed amplifies below
535
convective clusters where SWR is low. Specific air humidity decreases below convective
536
clusters following cooling of the near-surface atmosphere while qs doesn’t change much
537
because of the ocean thermal inertia. Difference in responses of q a and qs increases the
538
vertical gradient of air humidity below convective cloud clusters and thus enhances
539
anomalous evaporation and LHTFL produced by anomalous wind speed.
540
541
Amplitudes of intraseasonal LHTFL and SWR display significant interannual variations
542
in the tropical Pacific Ocean. Amplitudes of intraseasonal SWR increase in the central
543
equatorial Pacific by 15 Wm-2 during mature phase of El Niño following the eastward
544
shift of convection. In contrast to the amplitude of intraseasonal SWR that varies in phase
545
with El Niño, the amplitude of intraseasonal LHTFL doesn’t exhibit similar significant
546
in-phase variation. In contrast, the intraseasonal LHTFL amplifies over the western
547
tropical Pacific approximately 8 months in advance of the mature phase of El Niño. This
548
amplification reflects impacts of the westerly wind bursts that often precede the onset of
549
El-Nino.
550
551
Over much of the interior ocean where the ocean-atmosphere heat exchange drives the
552
ocean mixed layer balance, SST cools down in response to anomalously strong LHTFL.
24
553
There are considerable geographical variations in magnitude of this response that are
554
related in part to the spatial variations of oceanic mixed layer depth and its thermal inertia
555
that mitigates the impact of surface fluxes. Moreover, in the eastern tropical Pacific and
556
Atlantic cold tongues the SST warms up in response to LHTFL strengthening. In these
557
equatorial upwelling areas the SST is strongly affected by the ocean advection and
558
LHTFL responds to this rather than driving SST.
559
560
Appendix
561
Comparisons of in-situ LHTFL with satellite-derived LHTFL in the intraseasonal band is
562
shown in Fig. 11. This comparison is based on in-situ buoy measurements in the tropics
563
including 68 TAO/TRITON buoys in the Pacific, 21 PIRATA buoys in the Atlantic, and
564
10 RAMA buoys in the Indian Ocean. During 1992-2007 the data set has 30592
565
concurrent buoy-satellite weekly measurements in the Pacific, 3044 weeks of data in the
566
Atlantic, and 318 weeks of concurrent buoy and satellite data in the Indian Ocean. The
567
aggregate time series of buoy and satellite LHTFL have statistically significant
568
correlation around 0.6. The 99% confidence level of zero correlation is corr99% <0.1 for
569
the Pacific and Atlantic while it is slightly higher corr99% =0.14 for the Indian Ocean due
570
to shorter time series. Time series of intraseasonal LHTFL at each buoy location also
571
indicate significant correlation (Figs. 11a, 11c). Time correlation (TCORR) exceeds 0.6
572
over much of the tropical Pacific where average length of the LHTFL time series at
573
particular buoy is around 450 weeks ( corr99% =0.12). TCORR increases towards the west
574
following the westward increase of the mean LHTFL in the tropical Pacific (Fig. 1a). In
575
contrast, somewhat weaker TCORR is observed along 5° N where LHTFL is weaker due
25
576
to weaker winds and higher specific humidity in the ITCZ. The impact of the ITCZ is
577
better seen in the Atlantic where TCORR decreases below 0.5 in the ITCZ area (Fig.
578
11c). These comparisons suggest that the LHTFL retrieval should be rectified in the
579
ITCZ area. The air relative humidity has regional maximum in the ITCZ area. Therefore,
580
the meridional displacement of the ITCZ could produce variations of the relative
581
humidity strong enough that need to be accounted for in the Konda et al. (1996) Bowen
582
ratio approach.
583
584
Satellite intraseasonal LHTFL compares well with in-situ LHTFL to within the scatter of
585
the data (Figs. 11b, 11d, 11f). However, the magnitude of satellite intraseasonal LHTFL
586
is weaker than in-situ data. This bias is more evident in the Pacific where the
587
intraseasonal variations of satellite LHTFL are 15% to 20% weaker than those from
588
buoys, while this bias is less than 10% in the Atlantic and is not evident in the Indian
589
Ocean. We attribute this bias to the spatial and temporal smoothing of satellite data that
590
inevitably results in losing of a portion of variance observed at fixed location and high
591
temporal resolution.
592
593
Acknowledgements. SAG and JAC acknowledge support from the NASA Ocean Wind
594
Vector Science Team (OWVST). AB is grateful for the technical and financial supports
595
provided by CERSAT/IFREMER and MERCATOR. RTP gratefully acknowledges
596
support by NASA grant NNG04GD65G from the Radiation Sciences Program and NSF
597
grant ATM0631685 to the University of Maryland. The ISCCP data were obtained from
598
the NASA Langley Research Center EOSDIS Distributed Active Archive Center.
26
599
600
References:
601
Araligidad, N. M., and E. D. Maloney, 2008: Wind-driven latent heat flux and the
602
intraseasonal oscillation. Geophys. Res. Lett., 35, L04815,
603
doi:10.1029/2007GL032746.
604
Beal, R. C., V. N. Kudryavtsev, D. R. Thompson, S. A. Grodsky, D. G. Tilley, V.A.
605
Dulov, and H. C. Graber, 1997: The influence of the marine atmospheric boundary
606
layer on ERS 1 synthetic aperture radar imagery of the Gulf Stream. J. Geophys.
607
Res., 102, 5799-5814.
608
609
Bentamy A., Y. Quilfen, F. Gohin, N. Grima, M. Lenaour, and J. Servain, 1996: Global
Atmos. Ocean Sys., 4, 1-29.
610
Bentamy, A., K.B. Katsaros, A.M. Mestas-Nuñez, W.M. Drennan, E.B. Forde, and H.
611
Roquet, 2003: Satellite Estimates of Wind Speed and Latent Heat Flux over the
612
Global Oceans. J. Climate, 16, 637–656.
613
Bentamy, A., L-H. Ayina, W. Drennan, K. Katsaros, A. M. Mestas-Nuñez, R. T. Pinker,
614
2008: 15 Years of Ocean Surface Momentum and heat Fluxes from Remotely
615
Sensed Observations, FLUXNEWS, 5, 14-16. Available online at
616
http://sail.msk.ru/newsletter/fluxnews_5_final.pdf
617
Bond, N.A., and M.F. Cronin, 2008: Regional Weather Patterns during Anomalous Air–
618
Sea Fluxes at the Kuroshio Extension Observatory (KEO). J. Climate, 21, 1680–
619
1697.
620
621
Bourlès, B., R. Lumpkin, M.J. McPhaden, F. Hernandez, P. Nobre, E. Campos, L. Yu, S.
Planton, A. Busalacchi, A.D. Moura, J. Servain, and J. Trotte, 2008: The Pirata
27
622
Program: History, Accomplishments, and Future Directions. Bull. Amer. Meteor.
623
Soc., 89, 1111–1125.
624
Chou, S.H., E. Nelkin, J. Ardizzone, R.M. Atlas, and C.L. Shie, 2003: Surface Turbulent
625
Heat and Momentum Fluxes over Global Oceans Based on the Goddard Satellite
626
Retrievals, Version 2 (GSSTF2). J. Climate, 16, 3256–3273.
627
da Silva, A., C. C. Young, and S. Levitus, 1994: Algorithms and Procedures. Vol. 1,
628
Atlas of Surface Marine Data 1994, NOAA Atlas NESDIS 6, 83 pp.
629
Fairall, C.W., E.F. Bradley, J.E. Hare, A.A. Grachev, and J.B. Edson, 2003: Bulk
630
Parameterization of Air–Sea Fluxes: Updates and Verification for the COARE
631
Algorithm. J. Climate, 16, 571–591.
632
Foltz, G. R., and M. J. McPhaden, 2004: The 30–70 day oscillations in the tropical
633
Atlantic. Geophys. Res. Lett., 31, L15205, doi:10.1029/ 2004GL020023.
634
Grodsky, S. A., J. A. Carton, C. Provost, J. Servain, J. A. Lorenzzetti, and M. J.
635
McPhaden, 2005: Tropical instability waves at 0N, 23W in the Atlantic: A case
636
study using Pilot Research Moored Array in the Tropical Atlantic (PIRATA)
637
mooring data. J. Geophys. Res., 110, C08010, doi:10.1029/2005JC002941.
638
Han W., D. Yuan, W. T. Liu, D. J. Halkides, 2007: Intraseasonal variability of Indian
639
Ocean sea surface temperature during boreal winter: Madden-Julian Oscillation
640
versus submonthly forcing and processes. J. Geophys. Res., 112, C04001,
641
doi:10.1029/2006JC003791.
642
Jones, C., and B. C. Weare, 1996: The role of low-level moisture convergence and ocean
643
latent heat fluxes in the Madden-Julian oscillation: An observational analysis using
644
ISCCP data and ECMWF analyses. J. Climate, 11, 1057– 1072.
28
645
646
647
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-year reanalysis project, Bull.
Amer. Meteorol. Soc., 77, 437-471.
Konda, M., N. Imasato, and A. Shibata, 1996: A new method to determine near-sea
648
surface air temperature by using satellite data, J. Geophys. Res., 101(C6), 14349-
649
14360.
650
651
652
653
654
655
Krishnamurti, T., D. Oosterhof, and A. Mehta, 1988: Air–Sea Interaction on the Time
Scale of 30 to 50 Days. J. Atmos. Sci., 45, 1304–1322.
Large, W., and S. Pond, 1982: Sensible and Latent Heat Flux Measurements over the
Ocean. J. Phys. Oceanogr., 12, 464–482.
Liu, H. and R. T. Pinker, 2008: Radiative fluxes from satellites: Focus on aerosols. J.
Geophys. Res., 113, D08208, doi:10.1029/2007JD008736.
656
Liu, W. T., K. B. Katsaros, and J. A. Businger, 1979: Bulk parameterization of air-sea
657
exchanges of heat and water vapor including the molecular constraints at the sea
658
surface. J. Atmos. Sci., 36, 1722-1735.
659
Liu, W.T., 2002: Progress in scatterometer application, J. Oceanogr., 58(1), 121-136.
660
Madden, R.A., and P.R. Julian, 1972: Description of Global-Scale Circulation Cells in
661
662
663
664
665
666
667
the Tropics with a 40–50 Day Period. J. Atmos. Sci., 29, 1109–1123.
Maloney, E. D., and A. H. Sobel, 2004: Surface fluxes and ocean coupling in the tropical
intraseasonal oscillation, J. Climate., 17, 4368–4386.
McPhaden, M.J. , and Coauthors, 1998: The Tropical Ocean-Global Atmosphere observing
system: A decade of progress, J. Geophys. Res., 103(C7), 14,169-14,240.
McPhaden, M. J., G. Meyers, K. Ando,Y. Masumoto, V. S. N. Murty, M. Ravichandran,
F. Syamsudin, J. Vialard, L. Yu, W. Yu, 2008: RAMA: The Research Moored
29
668
Array for African-Asian-Australian Monsoon Analysis and Prediction, subm. Bul.
669
Amer. Meteor. Soc., http://www.pmel.noaa.gov/tao/oceansites/RAMA_BAMS.pdf
670
671
McPhaden, M.J., 2004: Evolution of the 2002/03 El Niño. Bull. Amer. Meteor. Soc., 85,
677–695.
672
Mestas-Nuñez, A.M., A. Bentamy, and K.B. Katsaros, 2006: Seasonal and El Niño
673
Variability in Weekly Satellite Evaporation over the Global Ocean during 1996–98.
674
J. Climate, 19, 2025–2035.
675
676
677
678
Muñoz, E., A.J. Busalacchi, S. Nigam, and A. Ruiz-Barradas, 2008: Winter and Summer
Structure of the Caribbean Low-Level Jet. J. Climate, 21, 1260–1276.
Neelin, J.D., I.M. Held, and K.H. Cook, 1987: Evaporation-Wind Feedback and LowFrequency Variability in the Tropical Atmosphere. J. Atmos. Sci., 44, 2341–2348.
679
Park, S., C. Deser, and M.A. Alexander, 2005: Estimation of the Surface Heat Flux
680
Response to Sea Surface Temperature Anomalies over the Global Oceans. J.
681
Climate, 18, 4582–4599.
682
Pinker, R. T., and I. Laszlo, 1992: Modeling surface solar irradiance for satellite
683
applications on a global scale. J. Appl. Meteor. Clim., 31 (2), 194-211.
684
Qiu, B., S. Chen, and P. Hacker, 2004: Synoptic-Scale Air–Sea Flux Forcing in the
685
Western North Pacific: Observations and Their Impact on SST and the Mixed
686
Layer. J. Phys. Oceanogr., 34, 2148–2159.
687
Reynolds, R. W., T. M. Smith, C. Liu, D. B. Chelton, K. S. Casey and M. G. Schlax,
688
2007: Daily High-resolution Blended Analyses for sea surface temperature. J.
689
Climate, 20, 5473-5496.
30
690
Rodriguez-Puebla, C., R. T. Pinker, and S. Nigam, 2008: Relationship between
691
downwelling surface shortwave radiative fluxes and sea surface temperature over
692
the tropical Pacific: AMIP II models versus satellite estimates, Ann. Geophys., 26,
693
785–794.
694
695
Rossow, W. B., and R. A. Schiffer, 1999: Advances in understanding clouds from
ISCCP. Bull. Amer. Meteor. Soc., 80, 2261-2287.
696
Shinoda, T., and H.H. Hendon, 1998: Mixed Layer Modeling of Intraseasonal Variability
697
in the Tropical Western Pacific and Indian Oceans. J. Climate, 11, 2668–2685.
698
Wallace, J.M., T.P. Mitchell, and C. Deser, 1989: The influence of sea surface
699
temperature on surface wind in the eastern equatorial Pacific: seasonal and
700
interannual variability. J. Climate, 2, 1492-1499.
701
702
703
Worley, S.J. S. D. Woodruff, R. W. Reynolds, S. J. Lubker, and N. Lott, 2005: ICOADS
release 2.1 data and products, Int. J. Climate, 25, 823-842, doi:10.1002/joc.1166
Yu, L., R.A. Weller, and B. Sun, 2004: Improving Latent and Sensible Heat Flux
704
Estimates for the Atlantic Ocean (1988–99) by a Synthesis Approach. J. Climate,
705
17, 373–393.
706
Yu, L., X. Jin, and R.A. Weller, 2006: Role of Net Surface Heat Flux in Seasonal
707
Variations of Sea Surface Temperature in the Tropical Atlantic Ocean. J. Climate,
708
19, 6153–6169.
709
Zhang, B. L., R. T. Pinker, and P. W. Stackhouse, 2007: An empirical orthogonal
710
function iteration approach for obtaining homogeneous radiative fluxes from
711
satellite observations. J. Appl. Meteor. Clim., 46, 435-444.
31
712
713
714
715
Zhang, C., and M.J. McPhaden, 2000: Intraseasonal Surface Cooling in the Equatorial
Western Pacific. J. Climate, 13, 2261–2276.
Zolina, O., and S.K. Gulev, 2003: Synoptic Variability of Ocean–Atmosphere Turbulent
Fluxes Associated with Atmospheric Cyclones. J. Climate, 16, 2717–2734.
716
32
717
Tables.
718
Table 1. Globally and time averaged LHTFL and evaporation.
Global mean
This Study
NCEP/NCAR
WHOI OAFlux
ICOADS 2.4
Reanalysis
LHTFL (Wm-2)
107
97
92
105
Evaporation
11.2
10.2
9.6
11.0
(cm month-1)
719
720
33
721
Figure legends.
722
723
Figure 1. 1992-2007 mean LHTFL (Wm-2) from (a) this study, (b) NCEP/NCAR
724
reanalysis, (c) WHOI OAFlux, and (d) ICOADS. The means are based on whatever part
725
of this time interval is available. Annual harmonics (e) magnitude and (f) phase.
726
Semiannual harmonics (g) magnitude and (h) phase. Phase is in month.
727
728
Figure 2. Time mean (a) wind speed, w , (b) sea-air specific humidity difference,
729
q  qs  qa . Standard deviation of intraseasonal (c) wind speed, w , (d) humidity
730
difference,  q . (e) Standard deviation of intraseasonal LHTFL and contribution to it
731
from intraseasonal variation of (f) wind speed, (e) specific humidity, and (h) saturated
732
near surface humidity. The prime symbol denotes the intraseasonal component.
733
734
Figure 3. Time regression of intraseasonal LHTFL index averaged over the Gulf Stream
735
area onto (a) mean sea level pressure and winds, (b) LHTFL elsewhere and wind speed,
736
(c) saturated near surface humidity and humidity. The index area is defined as the area
737
where STD of intraseasonal LHTFL exceeds 40 Wm-2 (see Fig. 2e) and is dotted in panel
738
(a). The index is defined as the index area average intraseasonal LHTFL normalized by
739
its standard deviation (25 Wm-2). Only wind arrows exceeding 0.4 ms-1 are shown in
740
panel (a).
741
742
Figure 4. Time regression of intraseasonal LHTFL index averaged over the Agulhas
743
Current area onto intraseasonal mean sea level pressure (contours) and winds (arrows).
34
744
The index is defined as an area average intraseasonal LHTFL normalized by its standard
745
deviation (37 Wm-2). The index area is shown by the shaded rectangle. Only wind arrows
746
exceeding 0.4 ms-1 are shown.
747
748
Figure 5. Lagged regression of intraseasonal LHTFL and SST. (a) SST leads LHTFL by
749
1 week, (b) LHTFL leads SST by 1 week. Areas where time correlation exceeds the 99%
750
confidence level of zero correlation are dotted. Inlay in panel (b) shows lagged
751
correlations of LHTFL and SST time series spatially averaged over the equatorial east
752
Pacific (black) and the midlatitude Pacific (red). Negative lags are LHTFL lead time in
753
weeks.
754
755
Figure 6. Time correlation of the rate of change of intraseasonal SST ( T / t ) with (a)
756
intraseasonal LHTFL, (b) inversed intraseasonal short wave radiation (-SWR), and (c)
757
sum of the two (LHTFL-SWR). Correlation exceeding 0.1 is significant at the 99%
758
confidence level.
759
760
Figure 7. Time correlation of (a) weekly average and (b) daily average intraseasonal
761
LHTFL and SWR. Inlay in each panel shows lagged correlation of LHTFL and SWR
762
(solid) averaged over the equatorial Indian Ocean area marked by the rectangle and
763
(dashed) from the TAO/TRITON mooring at 0°N, 165°E (shown by the cross). Lags are
764
in weeks. Positive lags imply that SWR leads LHTFL. Weekly average LHTFL is from
765
this study, daily average LHTFL is the Woods Hole Institution OAFlux by Yu et al.
766
(2004).
35
767
768
Figure 8. Time-longitude diagrams of intraseasonal (a) LHTFL and (b) SWR averaged
769
5oS to 5oN in the Indian and western Pacific Oceans. Slope lines mark the same events
770
and correspond to eastward propagation at 4.5 to 7.5 ms-1.
771
772
Figure 9. Lagged correlation along the equator of intraseasonal (a) wind speed (W) and
773
SWR, (b) LHTFL and SWR, and (c) specific air humidity (Qa) and SWR. All variables
774
are averaged over the 5oS to 5oN belt. Land points are vertical boxes shaded in gray.
775
776
Figure 10. Time regression of anomalous Nino3 SST (210°E-270°E, 5°S-5°N) with
777
running standard deviation,  , of intraseasonal (a) SWR and (b) LHTFL. Correlation for
778
SWR is instantaneous, while LHTFL is correlated with Nino3 SST that lags it by 35
779
weeks. Inlays show lagged correlation of Nino3 SST with the intraseasonal variance of
780
flux spatially averaged over the rectangle shown in each panel. Positive lags imply flux
781
leading Nino3 SST. Areas where the time regression is significant at the 99% level are
782
cross-hatched.
783
784
Figure 11. Comparison of intraseasonal buoy and satellite-derived LHTFL. Spatial maps
785
of time correlation (left) and scatter diagrams (right) for (a,b) tropical Pacific, (c,d)
786
tropical Atlantic, and (e,f) tropical Indian Ocean. Gray shading in (b,d) shows the
787
standard deviation of satellite intraseasonal LHTFL in 2 Wm-2 intervals of in-situ data.
788
TCORR is the time correlation evaluated from the aggregate satellite/buoy comparisons
36
789
for each basin; NUM is the length of the aggregate record (in weeks). Buoy locations are
790
shown by closed circles in (a,c,e).
791
792
37
793
794
795
796
797
Figure 1. 1992-2007 mean LHTFL (Wm-2) from (a) this study, (b) NCEP/NCAR
reanalysis, (c) WHOI OAFlux, and (d) ICOADS. The means are based on whatever part
of this time interval is available. Annual harmonics (e) magnitude and (f) phase.
Semiannual harmonics (g) magnitude and (h) phase. Phase is in month.
38
798
799
800
801
802
803
804
Figure 2. Time mean (a) wind speed, w , (b) sea-air specific humidity difference,
q  qs  qa . Standard deviation of intraseasonal (c) wind speed, w , (d) humidity
difference,  q . (e) Standard deviation of intraseasonal LHTFL and contribution to it
from intraseasonal variation of (f) wind speed, (e) specific humidity, and (h) saturated
near surface humidity. The prime symbol denotes the intraseasonal component.
39
805
806
807
808
809
810
811
812
813
814
Figure 3. Time regression of intraseasonal LHTFL index averaged over the Gulf Stream
area onto (a) mean sea level pressure and winds, (b) LHTFL elsewhere and wind speed,
(c) saturated near surface humidity and humidity. The index area is defined as the area
where STD of intraseasonal LHTFL exceeds 40 Wm-2 (see Fig. 2e) and is dotted in panel
(a). The index is defined as the index area average intraseasonal LHTFL normalized by
its standard deviation (25 Wm-2). Only wind arrows exceeding 0.4 ms-1 are shown in
panel (a).
40
815
816
817
818
819
820
821
822
Figure 4. Time regression of intraseasonal LHTFL index averaged over the Agulhas
Current area onto intraseasonal mean sea level pressure (contours) and winds (arrows).
The index is defined as an area average intraseasonal LHTFL normalized by its standard
deviation (37 Wm-2). The index area is shown by the shaded rectangle. Only wind arrows
exceeding 0.4 ms-1 are shown.
41
823
824
825
826
827
828
829
830
Figure 5. Lagged regression of intraseasonal LHTFL and SST. (a) SST leads LHTFL by
1 week, (b) LHTFL leads SST by 1 week. Areas where time correlation exceeds the 99%
confidence level of zero correlation are dotted.
Inlay in panel (b) shows lagged correlations of LHTFL and SST time series spatially
averaged over the equatorial east Pacific (black) and the midlatitude Pacific (red).
Negative lags are LHTFL lead time in weeks.
42
831
832
833
834
835
836
Figure 6. Time correlation of the rate of change of intraseasonal SST ( T / t ) with (a)
intraseasonal LHTFL, (b) inversed intraseasonal short wave radiation (-SWR), and (c)
sum of the two (LHTFL-SWR). Correlation exceeding 0.1 is significant at the 99%
confidence level.
43
837
838
839
840
841
842
843
844
845
846
847
Figure 7. Time correlation of (a) weekly average and (b) daily average intraseasonal
LHTFL and SWR. Inlay in each panel shows lagged correlation of LHTFL and SWR
(solid) averaged over the equatorial Indian Ocean area marked by the rectangle and
(dashed) from the TAO/TRITON mooring at 0oN, 165oE (shown by the cross). Lags are
in weeks. Positive lags imply that SWR leads LHTFL. Weekly average LHTFL is from
this study, daily average LHTFL is the Woods Hole Institution OAFlux by Yu et al.
(2004).
44
848
849
850
851
852
Figure 8. Time-longitude diagrams of intraseasonal (a) LHTFL and (b) SWR averaged
5oS to 5oN in the Indian and western Pacific Oceans. Slope lines mark the same events
and correspond to eastward propagation at 4.5 to 7.5 ms-1.
45
853
854
855
856
857
Figure 9. Lagged correlation along the equator of intraseasonal (a) wind speed (W) and
SWR, (b) LHTFL and SWR, and (c) specific air humidity (Qa) and SWR. All variables
are averaged over the 5oS to 5oN belt. Land points are vertical boxes shaded in gray.
46
858
859
860
861
862
863
864
865
866
867
Figure 10. Time regression of anomalous Nino3 SST (210E-270E, 5S-5N) with running
standard deviation,  , of intraseasonal (a) SWR and (b) LHTFL. Correlation for SWR is
instantaneous, while LHTFL is correlated with Nino3 SST that lags it by 35 weeks.
Inlays show lagged correlation of Nino3 SST with the intraseasonal variance of flux
spatially averaged over the rectangle shown in each panel. Positive lags imply flux
leading Nino3 SST. Areas where the time regression is significant at the 99% level are
cross-hatched.
47
868
869
870
871
872
873
874
875
876
Figure 11. Comparison of intraseasonal buoy and satellite-derived latent heat flux. Spatial
maps of time correlation (left) and scatter diagrams (right) for (a,b) tropical Pacific, (c,d)
tropical Atlantic, and (e,f) tropical Indian Ocean. Gray shading in (b,d) shows the
standard deviation of satellite intraseasonal LHTFL in 2 Wm-2 intervals of in-situ data.
TCORR is the time correlation evaluated from the aggregate satellite/buoy comparisons
for each basin, NUM is the length of the aggregate record (in weeks). Buoy locations are
shown by closed circles in (a,c,e).
48