1
Low-cloud, boundary layer, and sea ice interactions over the Southern Ocean
2
during winter
3
4
Casey J. Wall, Tsubasa Kohyama, and Dennis L. Hartmann
5
Department of Atmospheric Sciences, University of Washington, Seattle, Washington
6
7
Corresponding Author:
8
Casey J. Wall,
[email protected]
9
Department of Atmospheric Sciences, University of Washington
10
408 ATG Building, Seattle, WA 98195-1640
11
12
Journal of Climate
13
Submitted June 30, 2016
14
15
16
Revision submitted January 19, 2017
1
17
Abstract
18
19
layer structure across the Antarctic sea ice edge is seen in ship-based measurements
20
and in active satellite retrievals from Cloud-Aerosol Lidar and Infrared Pathfinder
21
Satellite Observation (CALIPSO), which provide an unprecedented view of polar
22
clouds during winter. Sea ice inhibits heat and moisture transport from the ocean to
23
the atmosphere, and, as a result, the boundary layer is cold, stable and clear over sea
24
ice, and warm, moist, well-mixed and cloudy over open water. The mean low-cloud
25
fraction observed by CALIPSO is roughly 0.7 over open water and 0.4-0.5 over sea
26
ice, and the low-cloud layer is deeper over open water. Low-level winds in excess of
27
10 ms-1 are common over sea ice. Cold advection off of the sea ice pack causes
28
enhanced low-cloud fraction over open water, and thus an enhanced longwave
29
cloud radiative effect at the surface. Quantitative estimates of the surface longwave
30
cloud radiative effect contributed by low-clouds are presented. Finally, ten state-of-
31
the-art global climate models with satellite simulators are compared to
32
observations. Near the sea ice edge, seven out of ten models simulate cloudier
33
conditions over open water than over sea ice. Most models also underestimate low-
34
cloud fraction both over sea ice and over open water.
35
1. Introduction
During austral winter, a sharp contrast in low-cloud fraction and boundary
36
Sea ice, low-clouds, and the atmospheric boundary layer modulate the
37
climate of the Southern Ocean by influencing surface heat fluxes. During winter, sea
2
38
ice insulates the ocean from the cold atmosphere above, reducing the rate of ocean
39
heat loss at the surface by a factor 10-100 [Gordon, 1991]. Low-clouds and moisture
40
emit longwave (LW) radiation downward and heat the surface, and low-level winds
41
control the surface turbulent heat and moisture fluxes. When sea ice forms, brine is
42
rejected, adding salt to the near-surface waters. These processes modify the
43
buoyancy of surface waters and are responsible for deep and intermediate water
44
formation. Roughly two-thirds of the deep water in the global ocean is formed in the
45
Southern Ocean [Johnson, 2008], making it a region of critical importance for the
46
global overturning circulation of the ocean [Marshall and Speer, 2012; Talley, 2013].
47
Surface fluxes of heat and moisture in the polar regions are intimately linked to the
48
atmospheric boundary layer and to sea ice, are poorly observed, and are a topic of
49
high priority for improving our understanding of polar climate and climate change
50
[Bourassa et al., 2013].
51
Interactions between sea ice and boundary layer clouds have previously
52
been studied, but focus on this topic has generally been on the Arctic. Across the
53
Arctic basin during summer and early fall, low-clouds are more abundant and
54
optically thicker over open water than over sea ice when viewed from active
55
satellite remote sensing products [Kay and Gettelman, 2009; Palm et al., 2010] and
56
from surface observers [Eastman and Warren, 2010]. On the other hand, Schweiger
57
et al. [2008] used passive satellite retrievals and found that, during fall, regions of
58
low sea ice concentration coincide with enhanced mid-level cloudiness and reduced
59
low-cloud cover. Barton et al. [2012] found that the sensitivity of Arctic low-cloud
60
fraction to variations in sea ice concentration depends on synoptic regime. For
3
61
stable regimes, which support low-clouds, a significant but weak covariance
62
between sea ice concentration and cloud properties occurs during most seasons
63
[Taylor et al., 2015]. Near the sea ice edge, cold, off-ice advection is known to cause
64
enhanced low-cloud cover, but, due to a lack of observations, previous work has
65
focused on case studies of extreme events [e.g. Walter, 1980; Renfrew and Moore,
66
1999; Petersen and Renfrew, 2007]. It has also been argued that boundary layer
67
moisture, or lack thereof, triggers the onset of sea ice melt and freeze-up when
68
advected over Arctic sea ice [Kapsch et al., 2013] and that a warmer, moister
69
atmospheric boundary layer has amplified Arctic sea ice decline in recent years
70
[Serreze et al., 2009; Screen and Simmonds, 2010; Boisvert and Stroeve, 2015].
71
Unlike the Arctic, interactions between Antarctic sea ice and boundary layer
72
clouds have been given relatively little attention. Bromwich et al. [2012] first
73
pointed out that the total cloud fraction observed from active satellite retrievals is
74
about 0.1-0.2 lower in sea ice covered regions of the Southern Ocean than over open
75
water to the north. Fitzpatrick and Warren [2007] used ship-based measurements
76
of downwelling solar radiation over the Southern Ocean to show that, during austral
77
spring and summer, clouds tend to be optically thicker over open water than over
78
sea ice. It is not clear if the relationship between low-clouds and sea ice in the Arctic
79
is similar to that in the Southern Ocean.
80
In this study we describe the boundary layer properties and low-cloud
81
fraction across the Southern Ocean during winter, but with an emphasis on the
82
marginal sea ice zone. We use satellite-based active retrievals of clouds, which
83
provide an unprecedented view of polar clouds during winter, as well as ship-based
4
84
measurements. We also use a radiative transfer model to compute estimates of the
85
downward flux of LW radiation near the sea ice edge and its sensitivity to low-level
86
warm and cold advection. Finally, we evaluate ten state-of-the-art climate models.
87
This paper is organized as follows: datasets and the methodology of the radiative
88
transfer calculations are described in Section 2, results are given in Section 3, and
89
conclusions are given in Section 4.
90
2. Data and Methods
91
a. Datasets
92
Cloud observations are taken from the Cloud-Aerosol Lidar with Orthogonal
93
Polarization (CALIOP) instrument onboard the Cloud-Aerosol Lidar and Infrared
94
Pathfinder Satellite Observation (CALIPSO) satellite. CALIOP is a lidar that measures
95
high vertical resolution profiles of backscatter from which estimates of cloud
96
properties are derived [Winker et al., 2007]. Because CALIOP is an active
97
instrument, retrievals are not affected by lack of sunlight or near-surface inversions
98
– two conditions that are common at high latitudes during winter and are
99
problematic for passive satellite retrievals of low-clouds. As an example of the
100
challenge of cloud detection with passive instruments over the polar regions, Liu et
101
al. [2004] found that, during polar night, about 40% of all clouds went undetected
102
by the cloud mask algorithm of the moderate-resolution imaging spectroradiometer
103
(MODIS) used at that time. The algorithm has since been improved [Baum et al.,
104
2012], but detection of low-clouds over the polar regions remains a major challenge
5
105
for passive instruments [Ackerman et al., 2008]. Unlike passive instruments, the
106
signal-to-noise ratio of CALIOP is maximized in the absence of sunlight, making it
107
well suited for studying clouds during polar night.
108
We use the General Circulation Model-Oriented CALIPSO Cloud Product
109
(CALIPSO-GOCCP) version 2.9 [Chepfer et al., 2010; CALIPSO, 2015]. CALIPSO-
110
GOCCP provides cloud fraction on a 2° longitude, 2° latitude and 480 m height grid.
111
It also provides low (below 3.2 km), middle (3.2-6.5 km) and high (above 6.5 km)
112
cloud fraction, and estimates for how they are partitioned between liquid and ice.
113
For a given grid box and time interval, cloud fraction is defined as the number of
114
scenes in which a cloud was positively identified divided by the number of scenes in
115
which the lidar was not fully attenuated in the grid box. The lidar beam becomes
116
fully attenuated at an optical depth of ~3 [Winker et al., 2007], so CALIOP often does
117
not measure the bottom portion of low-clouds [Cesana et al., 2016]. The vertical
118
resolution of CALIOP is 30 m below 8 km and 60 m above 8 km, with a total of 583
119
vertical levels. Using the relatively coarse GOCCP vertical grid, which has only 40
120
levels, significantly increases the signal-to-noise ratio and provides a grid that is
121
better for comparison with global climate models [Chepfer et al., 2010].
122
123
National Oceanic and Atmospheric Administration/National Snow and Ice Data
124
Center Climate Data Record of Passive Microwave Sea Ice Concentration dataset
125
[Peng et al., 2013; NOAA/NSIDC, 2015], and cloud liquid water path from the Multi-
126
Sensor Advanced Climatology of Liquid Water Path dataset [Elsaesser et al., 2015].
127
Cloud liquid water path is defined as the total mass of cloud liquid water above a
We also use satellite-based observations of sea ice concentration from the
6
128
unit area of the earth’s surface. The latitude of the sea ice edge, which we define as
129
the northernmost point at which the sea ice concentration is 0.35, is computed from
130
the sea ice data. Our main conclusions are not sensitive to choosing a threshold of
131
0.50 or 0.25 for this definition. We also use temperature, specific humidity, and
132
wind fields from the European Center for Medium-Range Weather Forecasts
133
Reanalysis (ERA Interim; [Dee et al., 2011; ECMWF, 2015]). Boundary layer fields in
134
reanalysis products are poorly constrained by observations over the Southern
135
Ocean, and should be interpreted with caution. However, ERA-Interim is
136
consistently ranked among the most reliable reanalysis products in the high
137
southern latitudes [e.g. Bromwich et al., 2011; Bracegirdle and Marshall, 2012; Jones
138
et al., 2016].
139
All satellite and reanalysis data are analyzed on either monthly- or daily-
140
mean timescales during the months of June, July and August (JJA) from 2006
141
through 2014. ERA-Interim reanalysis data are available on monthly-mean and
142
instantaneous 6-hourly time resolutions, and daily-means are computed from the
143
instantaneous 6-hourly data. Monthly-mean fields are analyzed unless stated
144
otherwise.
145
146
speed from soundings, and cloud-base height measured by a ceilometer.
147
Measurements were made on two cruises that traversed the Weddell Sea during
148
June-August 2013 and May-August 1992 [König-Langlo et al., 2006; König-Langlo,
149
2013; König-Langlo, 2005]. The cruise tracks are shown in Figure 1. On the 2013
150
cruise, soundings were launched once per day and have a vertical resolution of
Additionally, we use ship-based observations of air temperature and wind
7
151
about 30 m in the lower troposphere, while, on the 1992 cruise, soundings were
152
launched four times per day and have a vertical resolution of about 60 m. Sounding
153
data are linearly interpolated to a vertical grid with a spacing of 30 m and 60 m on
154
the 2013 cruise and 1992 cruise, respectively. A total of 57 and 161 soundings were
155
taken poleward of 55°S on the 2013 cruise and 1992 cruise, respectively. Wind
156
speed measurements used global positioning system (GPS) technology on the 2013
157
cruise and the OMEGA radio navigation system on the 1992 cruise. As a result,
158
higher quality wind speed measurements were made on the 2013 cruise [König-
159
Langlo et al., 2006]. Measurements of cloud-base height were made on the 1992
160
cruise but not the 2013 cruise. These ship-based observations complement the
161
satellite observations because the soundings can resolve the vertical structure of the
162
boundary layer and the ceilometer measurements can reliably detect the cloud-base
163
height. The main weakness of the ship-based observations is that measurements are
164
sparse.
165
166
Coupled Model Intercomparison Project Phase 5 (CMIP5) [Taylor et al., 2012],
167
including output from the CALIPSO simulator [Chepfer et al., 2008]. Models in fully-
168
coupled and atmosphere-only configurations are evaluated, and the first ensemble
169
member for each model is used.
170
b. Radiative transfer modeling
171
Finally, we use output from ten global climate models that participated in the
One goal of this study is to quantify the downward flux of LW radiation at the
172
ocean surface, which we call “surface 𝐿𝑊↓ ”, and its dependence on low-cloud cover.
173
Because direct observations of surface 𝐿𝑊↓ are not available, we compute it using a
8
174
radiative transfer model. The advantage of using a radiative transfer model
175
compared to using reanalysis data is that we are able to vary low-clouds while
176
holding middle- and high-cloud cover fixed to zero. This method isolates the
177
contribution to surface 𝐿𝑊↓ made by low-clouds.
178
We use the Rapid and Accurate Radiative Transfer Model for Global Climate
179
Models [Mlawer et al., 1997; Clough et al., 2005; Iacono et al., 2008]. This model is
180
one-dimensional and takes vertical profiles of temperature, humidity, cloud liquid
181
and ice water content as inputs and computes surface 𝐿𝑊↓ . The temperature and
182
humidity profiles are taken from the ERA-Interim reanalysis, and for each profile
183
the surface 𝐿𝑊↓ is computed with a clear-sky and with low-cloud completely
184
obscuring the sky. We refer to these values as 𝐿𝑊↓,!"#$% and 𝐿𝑊↓,!"#$%&'( ,
185
respectively. A best estimate for the true value of the flux of downward LW
186
radiation at the surface, which we call 𝐿𝑊↓,!""!!"# , is also computed: 𝐿𝑊↓,!""!!"# = 𝐿𝑊↓,!"#$% (1 − 𝐿𝐶𝐹) + 𝐿𝑊↓,!"#$%&'( 𝐿𝐶𝐹
187
where 𝐿𝐶𝐹 is the low-cloud fraction observed by CALIPSO. Daily-mean data are used
188
because it is the shortest time resolution for which CALIPSO-GOCCP cloud
189
observations are available.
190
The radiative transfer calculations are done only for regions of open water
191
near the sea ice edge. We focus on scenes between 1°-3° equatorward of the ice
192
edge, where the sea ice concentration is approximately zero. This restriction is made
193
because microwave, satellite-based retrievals of cloud liquid water path are not
194
available over sea ice. Although the domain is restricted to open water scenes only,
9
195
the surface heat budget over open water is of interest because extreme air-sea heat
196
fluxes can occur there [e.g. Papritz et al., 2014].
197
In the radiative transfer calculations, several assumptions about low-clouds
198
are made that are based on observations presented in Table 1. First, low-clouds are
199
assumed to consist entirely of supercooled liquid. CALIPSO-GOCCP observations of
200
cloud fraction partitioned by phase are presented in Table 1 and show that low-
201
clouds over the Southern Ocean are frequently composed of liquid, consistent with
202
previous studies [Hu et al., 2010; Morrison et al., 2011]. Cesana et al. [2016] suggest
203
that low-clouds with unclassified phase in the CALIPSO-GOCCP dataset are mostly
204
mixed-phase clouds. Even so, at least three-quarters of the low-clouds detected by
205
CALIPSO are liquid (low-cloud fraction is 0.50 for liquid clouds and 0.69 for all
206
phases). Second, all cloud liquid water is assumed to reside in the lowest 3.2 km of
207
the atmosphere. Under this assumption the column-integrated cloud liquid water
208
path is equal to the total liquid water contained in low-clouds. This assumption is
209
justified by the CALIPSO phase observations (Table 1), which show that liquid
210
clouds are usually found at low levels. Third, the liquid water path is assumed to be
211
60 gm-2, which is near the median of satellite observations (Table 1). Fourth, the
212
height of low-cloud base and top are set to 500 m and 1000 m, respectively. The
213
cloud-base height value comes from measurements from the 1992 cruise taken
214
when surface observers reported open water or open pack ice near the ship (Table
215
1). During this cruise, observed cloud-base height was distributed nearly uniformly
216
between the surface and 1000 m. Fifth, low-clouds are assumed to have a droplet
10
217
effective radius of 16 𝜇m, which is close to the observed wintertime mean over the
218
Southern Ocean [McCoy et al., 2014, their Figure 9].
219
220
about low-cloud properties. In order to test the sensitivity of the radiative transfer
221
calculations to the assumptions about low-cloud properties, runs were performed
222
with cloud-base height set to 0 m and 1000 m, with liquid water path halved to 30
223
gm-2 and with cloud effective radius doubled to 32 𝜇m and halved to 8 𝜇m. The
224
results are not sensitive to modest changes in liquid water path because liquid
225
clouds are nearly opaque to LW radiation for liquid water path values greater than
226
~20 g/m2 [Hartmann, 2016], which is much lower than observed values (Table 1).
227
In fact, each modification of the assumptions resulted in a change in 𝐿𝑊↓,!"#$%&'( of
228
about 7 Wm-2 or less, which is small compared to the contribution to surface 𝐿𝑊↓
229
made by low-clouds during overcast conditions (~80 Wm-2 – discussed in the
230
Results section). In other words, to leading order, low-cloud fraction controls the
231
surface LW radiative effect from low-clouds. The model error is less than 1 Wm-2
232
[Mlawer et al., 1997], which is much smaller than the uncertainty due to the five
233
assumptions made about cloud properties.
234
3. Results
235
a. Satellite observations of sea ice and low-cloud over the Southern Ocean
236
237
shows July-average sea ice concentration over the Southern Ocean. Two contours of
Our results are insensitive to modest changes in these five assumptions
We start with a brief description of Antarctic sea ice during winter. Figure 1
11
238
sea ice concentration are shown: 0.35 and 0.95. These contours can be thought of as
239
marking the boundaries between open water, fragmented sea ice, and a sea ice pack
240
that covers the surface nearly completely. Throughout most of the Eastern
241
Hemisphere, sea ice concentration rarely exceeds 0.95. This could be a result of the
242
coastline extending equatorward and forcing the sea ice closer to the Antarctic
243
Circumpolar Current. In regions where the coastline cuts poleward, like the Weddell
244
and Ross Seas, sea ice concentrations greater than 0.95 are much more common.
245
Average sea ice concentrations in June and August are similar in this regard (not
246
shown). Wadhams et al. [1987] describe the winter sea ice pack in the Weddell Sea
247
as observed from a cruise. They found the marginal sea ice zone to be a ~250 km
248
band of fragmented pancake ice with pockets of exposed seawater. Farther south,
249
they found sea ice organized into vast floes that covered the ocean surface nearly
250
completely. We recommend viewing photographs of these features in Wadhams et
251
al. [1987, their Figure 12].
252
Now, consider low-cloud fraction over the Southern Ocean. The 2006-2014
253
winter climatology of low-cloud fraction and the latitude of the sea ice edge are
254
shown in Figure 2. The interannual standard deviation of the latitude of the sea ice
255
edge ranges between about 0.5° to 1.5° latitude. One standard deviation on either
256
side of the mean position of the sea ice edge is shaded in Figure 2 to show that the
257
effects of interannual variability of the location of the ice edge are likely small.
258
Throughout the Southern Ocean, cloudier conditions are seen over open water than
259
over sea ice. Near the sea ice edge, low-cloud fraction is about 0.7 over open water
260
and 0.5 over sea ice. The gradient of low-cloud fraction across the sea ice edge is
12
261
weakest in the Southern Indian and Western Pacific Oceans (20°E – 160°E). This
262
weak gradient is likely because the sea ice pack is more fragmented in this region
263
than elsewhere in the Southern Ocean (Figure 1). In this region, the low-cloud
264
fraction is more variable over sea ice than over open water because gaps in the sea
265
ice pack are found throughout the ice pack, but little sea ice is found equatorward of
266
the sea ice edge. In the Weddell and Ross Seas, where the sea ice pack covers the
267
surface nearly completely, the low-cloud fraction is about 0.4 or less and the
268
gradient in low-cloud fraction across the sea ice edge is sharp.
269
The relationship between sea ice concentration and low-cloud properties
270
near the sea ice edge is made clearer by stratifying the observations based on
271
distance from the sea ice edge. For each grid point and time (monthly-means from
272
June, July and August between 2006-2014 are considered), the meridional distance
273
between the grid point and the ice edge is computed. Data are then composited by
274
meridional distance from the ice edge, using a bin width of 0.5° latitude, and
275
averaged. We analyze data from the Weddell and Ross Seas (defined as 50°W-0°E
276
and 130°W-170°E, respectively; Figure 1), two regions where the sea ice pack
277
covers the surface nearly completely and where the sea ice edge is located far
278
offshore (Figure 1). This procedure was also done on the JJA-mean of each year, and
279
the main conclusions are the same using either monthly- or seasonal-averages.
280
281
troposphere over the Weddell Sea as a function of meridional distance from the sea
282
ice edge, and Figure 3d is similar but for the Ross Sea. On average, low-clouds
283
extend deeper and are more prevalent equatorward of the sea ice edge. The mean
Figure 3a shows the vertical profile of mean cloud fraction in the lower
13
284
low-cloud fraction and sea ice concentration are shown in Figure 3b and 3e, and the
285
method for deriving the confidence interval for the mean is described in the
286
Appendix. The domain can be split into three regions based on sea ice
287
concentration: an “ice” zone where sea ice concentration is ~1 that is located
288
poleward of 2° south of the ice edge, an “open water” zone where sea ice
289
concentration is ~0 that is located equatorward of 1° north of the ice edge, and a
290
“transition” zone between. Within the ice zone the mean low-cloud fraction is nearly
291
uniform at around 0.5, and within the open water zone the mean low-cloud fraction
292
is nearly uniform at around 0.7. The mean low-cloud fraction is significantly larger
293
in the open water zone than the ice zone. From south to north across the transition
294
zone, the low-cloud fraction increases smoothly as sea ice concentration decreases.
295
Figures 3c and 3f show vertical profiles of mean potential temperature and
296
specific humidity from reanalysis data as a function of meridional distance from the
297
ice edge. In current reanalysis data, the surface heat budget and the atmospheric
298
boundary layer over the Southern Ocean are poorly constrained by observations,
299
and therefore these data should be interpreted with caution. Nevertheless, the data
300
suggest several differences between the boundary layer over sea ice and over open
301
water. The lower troposphere is more stable over sea ice than over open water, as
302
can be seen by the vertical spacing in the potential temperature contours. Over open
303
water, near-surface temperatures are close to the freezing temperature of seawater,
304
and across the sea ice edge, near-surface temperatures drop rapidly. Boundary layer
305
specific humidity values are also nearly a factor of two larger over open water than
306
over sea ice.
14
307
b. Boundary layer structure from ship-based observations
308
309
further insight into the physical processes at work. In this section, sounding data are
310
represented by probability distributions. For each height measured by the
311
soundings, the probability distributions are computed by binning the data,
312
computing the number of observations in each bin, and normalizing by the total
313
number of soundings. Data are composited into measurements made between 55°-
314
65°S and poleward of 65°S. Because the sea ice edge is typically located between
315
60°-65°S in the Weddell Sea during winter, it is likely that most of the soundings
316
poleward of 65°S were taken over consolidated pack ice. Meanwhile, soundings
317
between 55°-65°S are likely a mixture of some taken over consolidated pack ice and
318
some taken where open water was exposed to the atmosphere.
319
Soundings resolve the vertical structure of the boundary layer and provide
Figure 4 shows the probability distribution of temperature at each height
320
between 10-1500 m. The 2013 and 1992 cruises are shown separately in Figure 4a-
321
b and 4c-d, respectively, because the cruises used different sounding technologies
322
[König-Langlo et al., 2006] and had different times between successive launches.
323
Two boundary layer regimes are seen: a warm and a cold mode. The warm mode is
324
characterized by having near-surface temperatures close to the freezing
325
temperature of seawater and by a moist adiabatic lapse rate above. In this regime
326
the boundary layer is well-mixed and moist. The cold mode is characterized by
327
typical near-surface temperatures of about -15°C to -25°C and by a low-level
328
inversion. Poleward of 65°S, the cold mode dominates (Figure 4b and 4d). Between
329
55°-65°S, both the warm and the cold modes are seen, albeit with different
15
330
likelihoods between the two cruises (Figure 4a and 4c). Differences in the relative
331
occurrence of the warm and cold mode in the 55°-65°S composite between the two
332
cruises could be a result of different weather events. The latitudinal distribution of
333
the warm and cold modes suggests that the cold mode forms over consolidated pack
334
ice, and the warm mode forms over open ocean or gaps in the sea ice.
335
The soundings also measured wind speed, and this is shown in Figure 5. The
336
probability distribution of wind speed as a function of height is shown for all
337
soundings taken poleward of 55°S. Sounding data are not composited by latitude
338
here, but doing so results in composites that resemble Figure 5 but are noisier (not
339
shown). The soundings reveal that wind speeds of 10 ms-1 or more are common at
340
heights of 200-600 m. For both cruises, the average wind speed between 200-600 m
341
is 10 ms-1 or more for 60-70% of the soundings. For the 2013 cruise, the strong low-
342
level winds are often associated with a low-level jet. On this cruise, the modal value
343
of wind speed is ~12-15 ms-1 at heights of 200-400 m and decreases with height to
344
~8 ms-1 at heights of 800-1000 m (Figure 5a). Data from both cruises show that
345
strong low-level winds are common during winter.
346
Low-level jets are of interest because they indicate the presence of a stable
347
boundary layer. Low-level jets exist at the top of stable boundary layers and, at least
348
in temperate latitudes, are initiated when the boundary layer transitions from
349
convective to stable. During this transition, the sudden shoaling of the boundary
350
layer causes a reduction in drag from turbulent momentum flux, and therefore a
351
sudden increase in wind speed, at heights above the stable boundary layer but
352
below the top of the former convective boundary layer. The stable boundary layer
16
353
limits drag on the winds above and allows the jet to persist and follow an inertial
354
oscillation [Blackadar 1957; Thorpe and Guymer, 1977]. The mechanisms that
355
initiate low-level jets over Antarctic sea ice during winter are not fully understood.
356
One possible mechanism is warm advection from open water to sea ice covered
357
regions, which temporarily deepens the boundary layer and then allows a new jet to
358
form when the boundary layer collapses to a stable profile [Andreas et al., 2000].
359
Another possible mechanism is motions arising from baroclinic instability
360
associated with the thermal contrast between sea ice and open ocean.
361
We emphasize that a weakness of this study is the short time span of
362
sounding data. Soundings were taken over a total of 50 days between 55°-65°S and
363
59 days poleward of 65°S. Despite this drawback, the main conclusions are robust:
364
in both cruises, a warm and a cold boundary layer regime are seen, and low-level
365
wind speeds in excess of 10 ms-1 are common.
366
c. Advection across the sea ice edge
367
368
sea ice edge, and vice versa? We start with an investigation of low-cloud fraction and
369
its sensitivity to advection across the sea ice edge. Cold air outbreaks, in which air is
370
advected from a cold land or ice surface to a warmer ocean, are known to cause the
371
development of low-clouds [e.g. Walter, 1980; Liu et al., 2006]. When the cold air
372
mass is heated from below by the warm ocean surface, convection occurs and low-
373
clouds form. Once formed, low-clouds are maintained by radiative cooling at cloud-
374
top, radiative heating at cloud-base, and the moisture source of the ocean.
How do clouds respond when cold air is advected equatorward, across the
17
375
376
distance from the sea ice edge, stratified by low-level advection across the sea ice
377
edge. As a metric for low-level advection across the sea ice edge, the meridional
378
wind at 1000 hPa is linearly interpolated to the latitude of the sea ice edge. We refer
379
to this value as 𝑣!"# !"#! . Data are composited into scenes in which 𝑣!"# !"#! is less
380
than −0.5𝜎 ≈ −3 ms-1 and greater than 0.5𝜎 ≈ 3 ms-1, where 𝜎 is the standard
381
deviation of 𝑣!"# !"#! . These composites correspond to on-ice flow and off-ice flow,
382
respectively. These composites are made using daily-mean data over the Weddell
383
Sea. The mean low-cloud fraction equatorward of the sea ice edge is significantly
384
larger during periods of off-ice flow than periods of on-ice flow. The peak in low-
385
cloud fraction during periods of off-ice flow is located at about 2° equatorward of
386
the ice edge, suggesting that low-clouds formed by cold advection can persist well
387
away from the sea ice edge. The fact that the peak in low-cloud fraction is about 2°
388
latitude equatorward of the sea ice edge may be a result of the predominant low-
389
cloud type transitioning from roll clouds near the sea ice edge to cellular convection
390
downstream [Walter, 1980]. This hypothesis is also consistent with the composites
391
in Figure 3a and Figure 3d, which show that, near the sea ice edge, the low-cloud
392
layer deepens toward the equator. Finally, for latitudes 2° south of the ice edge and
393
poleward, where sea ice covers the surface nearly completely (Figure 3b), there is
394
either no significant difference, or a very small difference, in low-cloud fraction
395
between the on-ice flow and off-ice flow composites. This result suggests that low-
396
clouds over open water are coupled to the surface and require the moisture source
397
of the open ocean to exist, and therefore dissipate when separated from open water.
Figure 6 shows the mean low-cloud fraction as a function of meridional
18
398
d. Impact of low-level advection on the surface heat budget
399
We have seen evidence of a warm and a cold boundary layer regime, and that
400
cold, low-level advection off of the sea ice pack causes low-clouds to form over open
401
water. How do low-level advection and the resulting boundary layer and low-cloud
402
changes impact the surface heat budget? To address this question we use a radiative
403
transfer model to compute surface 𝐿𝑊↓ over open water near the sea ice edge and to
404
estimate the contribution made by low-clouds. The Weddell Sea is again used as the
405
region of study. Recall that estimates of surface 𝐿𝑊↓ are computed for a clear sky,
406
with low-cloud completely covering the sky, and using low-cloud fraction observed
407
by CALIPSO. These values will be called 𝐿𝑊↓,!"#$% , 𝐿𝑊↓,!"#$%&'( , and 𝐿𝑊↓,!""!!"#
408
respectively. The 𝐿𝑊↓,!""!!"# values are the best estimate for the real world, while
409
the 𝐿𝑊↓,!"#$% and 𝐿𝑊↓,!"#$%&'( values help with interpretation. Also, recall that there
410
are no middle- or high-clouds in these calculations, so the radiative effects of low-
411
clouds are isolated here.
412
First, consider the average values of surface 𝐿𝑊↓ . The average values of
413
𝐿𝑊↓,!"#$% and 𝐿𝑊↓,!"#$%&'( are about 210 Wm-2 and 290 Wm-2, respectively. In other
414
words, if a point at the ocean surface was located under a clear sky, and a low-cloud
415
passed overhead, then the downward flux of LW radiation would suddenly increase
416
by about 80 Wm-2, a 40% increase from the clear-sky value. The average value of
417
𝐿𝑊↓,!""!!"# is around 270 Wm-2. The surface LW cloud radiative effect, defined as
418
𝐿𝑊↓,!""!!"# − 𝐿𝑊↓,!"#$% , is about 50-60 Wm-2. During winter, low-clouds warm the
419
ocean surface by about 50-60 Wm-2 on average.
19
420
Furthermore, surface 𝐿𝑊↓ depends on the strength of warm or cold
421
advection at low-levels. In the calculations of 𝐿𝑊↓,!"#$% and 𝐿𝑊↓,!"#$%&'( , temperature
422
and specific humidity are varied but low-cloud fraction is held fixed, and therefore
423
the surface LW cloud radiative effect is nearly constant. In the calculation of
424
𝐿𝑊↓,!""!!"# , temperature, humidity, and low-cloud fraction are all varied. Thus, by
425
comparing data from the 𝐿𝑊↓,!""!!"# , 𝐿𝑊↓,!"#$% and 𝐿𝑊↓,!"#$%&'( calculations, the
426
sensitivity of surface 𝐿𝑊↓ to low-cloud variations can be separated from the effects
427
of temperature and humidity variations.
428
Figure 7 shows surface 𝐿𝑊↓ plotted as a function of 𝑣!"# !"#! . In the 𝐿𝑊↓,!"#$%
429
and 𝐿𝑊↓,!"#$%&'( calculations, where low-cloud fraction is held fixed, the data are
430
anticorrelated with 𝑣!"# !"#! (𝑟 = −0.61 and 𝑟 = −0.64, respectively). This happens
431
because air masses that form over the sea ice pack are cold and have low specific
432
humidity, and these features of the air cause it to radiate relatively weakly to the
433
surface when advected over open water (when low-cloud fraction is held fixed).
434
However, in the 𝐿𝑊↓,!""!!"# calculation, where low-cloud fraction is varied according
435
to CALIPSO observations, data are weakly anticorrelated with 𝑣!"# !"#! (scatterplot
436
in Figure 7). When the 𝐿𝑊↓,!""!!"# data are binned by 𝑣!"# !"#! and averaged, the
437
result agrees well with a linear regression (compare the blue and red lines in Figure
438
7). Because cold advection causes cloudy conditions, 𝐿𝑊↓,!"!!!"# data approach the
439
𝐿𝑊↓,!"#$%&'( regression line for large positive values of 𝑣!"# !"#! . The surface LW
440
cloud radiative effect, seen in Figure 7 by the difference between 𝐿𝑊↓,!""!!"# and
441
𝐿𝑊↓,!"#$% , increases by 1.1 ± 0.1 Wm-2 per 1 ms-1 increase in 𝑣!"# !"#! . Put another
20
442
way, typical values of the average surface LW cloud radiative effect, estimated by
443
the regression, range from 43 Wm-2 to 65 Wm-2 for 𝑣!"# !"#! = −10 ms-1 to
444
𝑣!"# !"#! = 10 ms-1, respectively. As a result, the regression coefficient of 𝐿𝑊↓,!""!!"#
445
on 𝑣!"# !"#! (−0.7 ± 0.1 Wm-2 per 1 ms-1 increase in 𝑣!"# !"#! ) is significantly smaller
446
in magnitude than the regression coefficients of 𝐿𝑊↓,!"#$% and 𝐿𝑊↓,!"#$%&'( (compare
447
slopes in Figure 7). Therefore, when low-cloud fraction, temperature, and humidity
448
are all allowed to vary, as they are in the real world, then surface 𝐿𝑊↓ is much less
449
sensitive to warm or cold advection than when low-cloud fraction is held fixed. Low-
450
clouds warm the surface most strongly during cold advection events, and therefore
451
act to reduce the sensitivity of surface 𝐿𝑊↓ to cold advection.
452
We emphasize that these calculations are only able to capture one term of the
453
surface heat budget: surface 𝐿𝑊↓ . Surface turbulent heat fluxes are likely very
454
important as well. Over open water in the Southern Ocean during winter, average
455
values of surface turbulent fluxes of sensible and latent heat are around 30 Wm-2
456
and 50 Wm-2 respectively – on the order of the average surface LW cloud radiative
457
effect – but surface turbulent heat fluxes can be several hundred watts per square
458
meter during extreme cold air outbreaks [Papritz et al., 2014]. These turbulent heat
459
flux values come from reanalysis data, and should therefore be interpreted with
460
caution. Our work is progress towards constraining the Southern Ocean surface heat
461
budget, but a complete understanding also requires knowledge of the surface
462
turbulent heat flux.
21
463
e. Evaluation of Global Climate Models
464
We have seen that the surface LW cloud radiative effect from low-clouds is
465
about 50-60 Wm-2 on average. Therefore, low-cloud biases in global climate models
466
(GCMs) have the potential to significantly bias the modeled surface heat budget.
467
Accurate representation of marine boundary layer clouds and their radiative effects
468
are a major challenge for GCMs [Dufresne and Bony, 2008; Trenberth and Fasullo,
469
2010], and low-clouds are especially poorly represented in GCMs during polar night
470
[Karlsson and Svensson, 2011]. Here we evaluate the climatology of low-cloud
471
fraction near the sea ice edge in ten GCMs. Low-cloud fraction in the models is
472
computed by a CALIPSO simulator that estimates what CALIPSO would observe if it
473
were flying above the atmosphere in the model. Both fully-coupled and atmosphere-
474
only (AMIP) configurations are analyzed between 1990-2004 [Taylor et al., 2012;
475
Gates, 1992]. Fully-coupled models have prescribed atmospheric greenhouse gas
476
concentrations from observations, while atmosphere-only models have prescribed
477
sea surface temperature and sea ice concentration from observations. Most
478
importantly, interactions between the ocean, sea ice, low-clouds and the
479
atmospheric boundary layer are active in fully-coupled configuration and are
480
suppressed in atmosphere-only configuration. Model climatologies are compared to
481
the observed climatology between 2006-2014.
482
Figure 8 shows the July-mean low-cloud fraction observed by CALIPSO and
483
simulated by six GCMs that have output from both atmosphere-only and fully-
484
coupled configuration. Model behavior is quite diverse. Starting with the
485
atmosphere-only runs, HadGEM2, MIROC5 and MRI-CGCM3 have cloudier
22
486
conditions over open water than over sea ice. However, the gradient in low-cloud
487
fraction across the sea ice edge is weak in MIROC5. The IPSL models and MRI-
488
CGCM3 significantly underestimate low-cloud cover over open water. Compared to
489
the atmosphere-only runs, models in fully-coupled configuration generally have
490
similar low-cloud fractions, and the MIROC5 and MPI-ESM-LR models have a sea ice
491
edge that is located too close to the coastline. The models do not capture the weaker
492
low-cloud fraction gradient across the sea ice edge in the Eastern Hemisphere that is
493
seen in observations (Figure 2). Other than that, the modeled low-cloud fraction
494
does not have any systematic geographic bias.
495
Model bias in low-cloud fraction is quantified and shown in Figure 9. As a
496
metric for the low-cloud fraction near the sea ice edge, the mean low-cloud fraction
497
is computed over open water and sea ice, again defined as 1°-3° equatorward and
498
2°-4° poleward of the ice edge, respectively. These will henceforth be referred to as
499
LCFopen water and LCFsea ice. One common feature between models is a bias of too little
500
low-cloud fraction over open water, consistent with previous work [e.g. Zhang et al.,
501
2005]. In fact, over open water, nine out of ten models underestimate low-cloud
502
fraction, one model (GFDL-CM3) agrees with observations at 95% confidence, and
503
no models overestimate low-cloud fraction. This can be seen by noting that LCFopen
504
water for all of the models appears to the left of the observations in Figure 9a. The
505
magnitude of the largest model bias is about 0.35 (IPSL models), meaning that low-
506
cloud fraction in these models is roughly half of the observed value. Model bias in
507
LCFsea ice is more diverse and ranges from +.15 to -.26. However, out of the eight
508
fully-coupled models, all but one model underestimate LCFsea ice, one model agrees
23
509
with observations (MPI-ESM-LR), and no models overestimate LCFsea ice. This can be
510
seen by noting that all but one of the fully-coupled models are below the
511
observations in Figure 9a. Again, models that underestimate low-cloud fraction the
512
most have low-cloud fractions that are roughly half of the observed value. Most
513
models underestimate low-cloud fraction over open water and over sea ice.
514
Having established the mean low-cloud fraction bias in the models, we now
515
examine the difference of low-cloud fraction between the open water and sea ice
516
regions, which is shown in Figure 9b. Seven out of ten models have significantly
517
cloudier conditions over open water than over sea ice. Although most models
518
capture the correct sign of LCF!"#$ !"#$% − LCF!"# !"# , the magnitude varies between
519
0.38 in the MIROC-ESM and MIROC-ESM-CHEM models and -0.11 in the IPSL
520
models, while the observed value is LCF!"#$ !"#$% − LCF!"# !"# = 0.17 ± 0.01.
521
Finally, it is interesting to consider how low-clouds change with model
522
resolution and configuration. Models that have both atmosphere-only and fully-
523
coupled output provide an opportunity to compare low-clouds when the ocean and
524
sea ice are prescribed and when they are interactive. Of these six models that have
525
both atmosphere-only and fully-coupled output, three have little or no difference
526
between low-cloud fraction in fully-coupled and atmosphere-only configuration
527
(IPSL-CM5A-LR, IPSL-CM5A-MR and MRI-CGCM3). However, these models also
528
underestimate low-cloud fraction the most (Figure 8 and Figure 9a). The MIROC5
529
and MPI-ESM-LR models have more realistic values of LCF!"#$ !"#$% − LCF!"# !"# in
530
the atmosphere-only configuration than in the fully-coupled configuration (Figure
531
9b), but this may be due to the fact that the sea ice edge in these models is much
24
532
closer to shore, and therefore more exposed to cold continental air, in fully-coupled
533
configuration (Figure 8). Finally, the IPSL models make for an interesting
534
comparison because they differ only in resolution: IPSL-CM5A-LR and IPSL-CM5A-
535
MR have horizontal resolutions of 1.9°×3.75° and 1.25°×2.5°, respectively. In this
536
model, finer resolution does not improve the low-cloud bias.
537
4. Summary and conclusions
538
During austral winter, active satellite retrievals from CALIPSO and ship-
539
based measurements show a strong contrast in low-cloud fraction and boundary
540
layer structure over Antarctic sea ice and the adjacent open ocean. Low-cloud
541
fraction is roughly 0.7 over open water and 0.4-0.5 over sea ice, and the low-cloud
542
layer is much deeper over open water. The boundary layer is cold, stable, dry and
543
clear over consolidated sea ice and warm, moist, cloudy and well-mixed over open
544
water. At heights of 200-600 m, wind speeds in excess of 10 ms-1 are common over
545
sea ice, and are often associated with a low-level jet. During periods of cold, off-ice
546
advection, low-cloud fraction and the surface LW cloud radiative effect are
547
enhanced over open water. This enhanced cloud radiative effect acts to substantially
548
slow the rate of LW cooling of the ocean mixed layer compared to what would
549
happen if low-cloud fraction were uncorrelated with warm or cold advection. Low-
550
cloud fraction over sea ice is similar for on-ice and off-ice advection conditions,
551
indicating that low-clouds that form over the open ocean are coupled to the surface
552
and do not survive when separated from the moisture source provided by open
553
water.
25
554
These results support the hypothesis of two-way interactions between the
555
ocean surface and the atmospheric boundary layer during winter over polar oceans.
556
Regions of open water have relatively warm surface temperatures and large surface
557
fluxes of heat and moisture to the atmosphere. Moist and warm boundary layers
558
with a strong greenhouse effect form over open water and thus reinforce the warm
559
surface temperatures there. If such a region were to become covered by sea ice then
560
the surface heat and moisture fluxes would reduce and the boundary layer would
561
cool, dry, and become less cloudy, causing the greenhouse effect of the boundary
562
layer to weaken. This would reinforce the cool surface temperatures and help the
563
sea ice persist.
564
565
output were examined. Seven out of ten models simulate a larger low-cloud fraction
566
over open water than over sea ice. Nine out of ten models underestimate low-cloud
567
fraction over open water, and seven out of eight fully-coupled models
568
underestimate low-cloud fraction over sea ice. The observed low-cloud and
569
boundary layer properties shown in this work can be used as a test in future model
570
intercomparison projects when CALIPSO simulator output for more models is
571
available.
Additionally, ten state-of-the-art climate models with CALIPSO simulator
26
572
Acknowledgements
573
574
Terra and Aqua Science. Tsubasa Kohyama was supported by the National Science
575
Foundation (NSF) under grant AGS-1549579 and AGS-0960497, and the Takenaka
576
Scholarship Foundation. We are grateful to Steve Warren for helpful discussion,
577
Peter Blossey for sharing code for the radiative transfer model, Gregory Elsaesser
578
for providing the MAC-LWP data, and three anonymous reviewers for their
579
thorough and constructive feedback. We also thank the CALIPSO team and the
580
authors of [Chepfer et al., 2010] for creating the CALIPSO-GOCCP dataset. We
581
acknowledge the World Climate Research Programme's Working Group on Coupled
582
Modelling, which is responsible for CMIP, and we thank the climate modeling groups
583
for producing and making available their model output. For CMIP the U.S.
584
Department of Energy's Program for Climate Model Diagnosis and Intercomparison
585
provides coordinating support and led development of software infrastructure in
586
partnership with the Global Organization for Earth System Science Portals.
This research was conducted with support from NASA grant NNX14AJ26G,
27
587
Appendix
588
Derivation of 95% confidence interval for the mean low-cloud fraction
589
Figure 3b and Figure 3e show the mean low-cloud fraction as a function of
590
meridional distance from the sea ice edge in the Weddell Sea and Ross Sea,
591
respectively. The 95% confidence interval for mean low-cloud fraction is computed
592
assuming that low-cloud fraction measurements at each grid-cell and time are
593
independent. The low-cloud fraction data is available as monthly averages on a 2°
594
longitude by 2° latitude grid. To justify the assumption that measurements of low-
595
cloud fraction at each grid cell and time are independent we must assess spatial and
596
temporal autocorrelation.
597
598 •
Serial correlation in the meridional dimension
599
Our goal is to bin the data by meridional distance from the ice edge and
600
compute the mean and 95% confidence interval for the mean of each bin. The
601
CALIPSO grid is resolved in 2° latitude grid cells, and when compositing by
602
meridional distance from the ice edge we use bins of width 0.5° latitude. Therefore,
603
for any given time, no two grid cells of the same longitude and neighboring latitudes
604
can be assigned to the same bin. We therefore do not need to consider serial
605
correlation in the meridional dimension when estimating the effective degrees of
606
freedom of each bin.
607
608 •
Serial correlation in the zonal dimension
28
609
We estimate a lower bound for the number of degrees of freedom in the
610
zonal dimension by computing the correlation length scale 𝐿 for each latitude and
611
time, and comparing it to the resolution of the CALIPSO grid. Following [Taylor,
612
1921; Keller, 1935], we define the correlation length scale as
613
∞
𝐿=
𝑟 𝜏 𝑑𝜏 !
614
615
where 𝑟(𝜏) is the spatial autocorrelation function of low-cloud fraction in the zonal
616
dimension and 𝜏 is the separation distance. The distance between independent
617
points in the zonal dimension can be thought of as 2𝐿. We computed 2𝐿 for each
618
time and latitude within 10 degrees of the sea ice edge, and the maximum value was
619
an order of magnitude smaller than the 2° longitude resolution of the CALIPSO grid.
620
We therefore treat each grid cell as an independent measurement of low-cloud
621
fraction in the zonal dimension.
622
623 •
Serial correlation in the time dimension
624
Each longitude and latitude grid-cell contains a timeseries of low-cloud
625
fraction observations. The number of effective degrees of freedom of the low-cloud
626
fraction timeseries (𝑁!"" ) is related to the number observations of low-cloud
627
fraction (𝑁) and the lag-1 autocorrelation of low-cloud fraction (𝑟! ) by the following
628
equation [Bretherton et al., 1999]:
629
29
𝑁!"" 1 − 𝑟! = 𝑁 1 + 𝑟! 630
631
Our goal is to bin the observations of low-cloud fraction by their meridional distance
632
from the ice edge and then to use this equation to estimate the effective degrees of
633
freedom for each bin. For each longitude and latitude grid cell we compute 𝑟! over
634
the entire timeseries. Then, for a given bin, say “bin A,” and a given grid cell, say
635
“grid cell B,” we keep track of the number of times that grid cell B is assigned to bin
636
A and then compute an estimate for the number of effective degrees of freedom
637
contributed to bin A by grid cell B by scaling by the right hand side of the above
638
equation. This procedure is done for every grid cell and every bin.
639
The total number of effective degrees of freedom for each bin estimated by this
640
procedure is slightly greater than if one were to assume each observation of low-
641
cloud fraction is independent in the time dimension. This happens because the lag-1
642
autocorrelation of low-cloud fraction over the domain tends to be slightly negative,
643
probably because of random sampling variability. We therefore assume that each
644
estimate of low-cloud fraction is independent in the time dimension.
30
645
References
646
Ackerman, S., et al. (2008). Cloud detection with MODIS. Part II: Validation. Journal of Atmospheric and
647
Oceanic Technology, 25(7), 1073–1086. doi:10.1175/2007JTECHA1053.1
648
Andreas, E. L., Claffey, K. J., & Makshtas, a. P. (2000). Low-level atmospheric jets and inversions over the
649
western Weddell Sea. Boundary-Layer Meteorology, 97(3), 459–486. doi:10.1023/A:1002793831076
650
Barton, N. P., Klein, S. a., Boyle, J. S., & Zhang, Y. Y. (2012). Arctic synoptic regimes: Comparing
651
domain-wide Arctic cloud observations with CAM4 and CAM5 during similar dynamics. Journal of
652
Geophysical Research Atmospheres, 117(15), 1–19. doi:10.1029/2012JD017589
653
Baum, B., et al. (2012). MODIS cloud-top property refinements for collection 6. Journal of Applied
654 655
Meteorology and Climatology, 51(6), 1145–1163. doi:10.1175/JAMC-D-11-0203.1
Blackadar, A. K. (1957). Boundary Layer Wind Maxima and their Significance for the Growth of
656 657
Nocturnal Inversions, Bull. Amer. Meteorol. Soc. 38, 283–290.
Boisvert, L. N., & Stroeve, J. C. (2015). The Arctic is becoming warmer and wetter as revealed by the
658
Atmospheric Infrared Sounder, Geophysical Research Letters 4439–4446.
659
doi:10.1002/2015GL063775
660
Bourassa, M. A., et al. (2013). High-latitude ocean and sea ice surface fluxes: Challenges for climate
661
research. Bulletin of the American Meteorological Society, 94(3), 403–423. doi:10.1175/BAMS-D-
662
11-00244.1
663
Bracegirdle, T. J., & Marshall, G. J. (2012). The Reliability of Antarctic Tropospheric Pressure and
664
Temperature in the Latest Global Reanalyses. Journal of Climate, 25(20), 7138–7146.
665
doi:10.1175/JCLI-D-11-00685.1
31
666
Bretherton, C. S., Widmann, M., Dymnikov, V. P., Wallace, J. M., & Blade, I. (1999). Effective number of
667
degrees of freedom of a spatial field. Journal of Climate, 12(1969), 1990–2009. doi:10.1175/1520-
668
0442(1999)012<1990:Tenosd>2.0.Co;2
669
Bromwich, D. H., Nicolas, J. P., & Monaghan, A. J. (2011). An Assessment of Precipitation Changes over
670
Antarctica and the Southern Ocean since 1989 in Contemporary Global Reanalyses. Journal of
671
Climate, 24(16), 4189–4209. doi:10.1175/2011JCLI4074.1
672
Bromwich, D. H., et al. (2012). Tropospheric clouds in Antarctica. Reviews of Geophysics, 50(1), 1–40.
673 674
doi:10.1029/2011RG000363
CALIPSO, 2015: GOCCP, Subset: Monthly, June, 2006-August, 2014. CFMIP-Obs: Cloud Observations
675
for Model Evaluation. Accessed 1 January, 2016. [available online at
676
http://climserv.ipsl.polytechnique.fr/cfmip-obs/]
677
Cesana, G., et al. (2016). Using in situ airborne measurements to evaluate three cloud phase products
678
derived from CALIPSO. Journal of Geophysical Research: Atmospheres, 121(10), 5788–5808.
679
doi:10.1002/2015JD024334
680
Chepfer, H., et al. (2008). Use of CALIPSO lidar observations to evaluate the cloudiness simulated by a
681 682
climate model. Geophysical Research Letters, 35(15), 1–6. doi:10.1029/2008GL034207
Chepfer, H., et al. (2010). The GCM-oriented CALIPSO cloud product (CALIPSO-GOCCP). Journal of
683 684
Geophysical Research Atmospheres, 115(5), 1–13. doi:10.1029/2009JD012251
Clough, S.A., et al. (2005). Atmospheric radiative transfer modeling: a summary of the AER codes, J.
685 686
Quant. Spectrosc. Radiat. Transfer., 91, 233-244
Dee, D. P., et al. (2011). The ERA-Interim reanalysis: Configuration and performance of the data
687
assimilation system. Quarterly Journal of the Royal Meteorological Society, 137(April), 553–597.
688
doi:10.1002/qj.828
32
689
Dufresne, J. L., & Bony, S. (2008). An assessment of the primary sources of spread of global warming
690
estimates from coupled atmosphere-ocean models. Journal of Climate, 21(19), 5135–5144.
691
doi:10.1175/2008JCLI2239.1
692
Eastman, R., & Warren, S. G. (2010). Interannual variations of arctic cloud types in relation to sea ice.
693 694
Journal of Climate, 23(15), 4216–4232. doi:10.1175/2010JCLI3492.1
ECMWF, 2015: ERA-Interim Project. Subset: Monthly, June 2006-August 2014. Accessed 1 January,
695
2016. [available online at http://apps.ecmwf.int/datasets/]
696
Elsaesser, G., C. O'Dell, and J. Teixeira (2015). Algorithm Theoretical Basis Document (ATBD) Version
697
1: Multi-Sensor Advanced Climatology of Liquid Water Path (MAC-LWP). An Enhanced Climate
698
Data Record (CDR) for MEaSUREs Project “A Data Record of the Cloudy Boundary Layer”. 20
699
pps.
700
Fitzpatrick, M. F., & Warren, S. G. (2007). The Relative Importance of Clouds and Sea Ice for the Solar
701 702
Energy Budget of the Southern Ocean. Journal of Climate, 20(6), 941–954. doi:10.1175/JCLI4040.1
Gates, W. L. (1992), AMIP: The Atmospheric Model Intercomparison Project. Bull. Amer. Meteor. Soc.,
703 704
73, 1962-1970
Gordon, A. L. (1991), Two stable modes of Southern Ocean winter stratification, in Deep Convection and
705
Deep Water Formation in the Oceans, Elsevier Oceanography Series, vol. 57, edited by P. C. Chu
706
and J. C. Gascard, pp. 17–35, Elsevier, New York.
707
Hartmann, D. L. (2016). Global Physical Climatology, Second Ed., 77 pp., Academic Press, Amsterdam
708
Hu, Y., et al. (2010). Occurrence, liquid water content, and fraction of supercooled water clouds from
709
combined CALIOP/IIR/MODIS measurements. Journal of Geophysical Research: Atmospheres,
710
115(19), 1–13. doi:10.1029/2009JD012384
33
711
Iacono, M.J., J.S. Delamere, E.J. Mlawer, M.W. Shephard, S.A. Clough, and W.D. Collins, (2008).
712
Radiative forcing by long-lived greenhouse gases: Calculations with the AER radiative transfer
713
models, J. Geophys. Res., 113, D13103, doi:10.1029/2008JD009944
714
Johnson, G. C. (2008). Quantifying Antarctic Bottom Water and North Atlantic Deep Water volumes.
715
Journal of Geophysical Research: Oceans, 113(5), 1–13. doi:10.1029/2007JC004477
716
Jones, R. W., I. A. Renfrew, A. Orr, B. G. M. Webber, D. M. Holland, and M. A. Lazzara (2016),
717
Evaluation of four global reanalysis products using in situ observations in the Amundsen Sea
718
Embayment, Antarctica, J. Geophys. Res. Atmos., 121, doi:10.1002/2015JD024680.
719
Kapsch, M.-L., Graversen, R. G., & Tjernström, M. (2013). Springtime atmospheric energy transport and
720
the control of Arctic summer sea-ice extent. Nature Climate Change, 3(8), 744–748.
721
doi:10.1038/nclimate1884
722
Karlsson, J., & Svensson, G. (2011). The simulation of Arctic clouds and their influence on the winter
723
surface temperature in present-day climate in the CMIP3 multi-model dataset. Climate Dynamics,
724
36(3), 623–635. doi:10.1007/s00382-010-0758-6
725
Kay, J. E., & Gettelman, A. (2009). Cloud influence on and response to seasonal Arctic sea ice loss.
726 727
Journal of Geophysical Research Atmospheres, 114(18), 1–18. doi:10.1029/2009JD011773
Keller, L. B., 1935: Expanding of limit theorems of probability theory on functions with continuous
728 729
arguments (in Russian). Works Main Geophys. Observ., 4, 5–19. Kikkawa,
König-Langlo, Gert (2005): Meteorological observations during POLARSTERN cruise ANT-X/4. Alfred
730
Wegener Institute, Helmholtz Center for Polar and Marine Research, Bremerhaven,
731
doi:10.1594/PANGAEA.269526 [available online at
732
https://pangaea.de/PHP/CruiseReports.php?b=Polarstern]
34
733
König-Langlo, Gert (2013): Meteorological observations during POLARSTERN cruise ANT-XXIX/6
734
(AWECS). Alfred Wegener Institute, Helmholtz Center for Polar and Marine Research,
735
Bremerhaven, doi:10.1594/PANGAEA.819610 [available online at
736
https://pangaea.de/PHP/CruiseReports.php?b=Polarstern]
737
König-Langlo, G., Loose, B., & Bräuer, B. (2006). 25 Years of Polarstern Meteorology. WDC-MARE
738
Reports. doi:10.2312/wdc-mare.2006.4
739
Liu, Y., Key, J. R., Frey, R. a., Ackerman, S. a., & Menzel, W. P. (2004). Nighttime polar cloud detection
740
with MODIS. Remote Sensing of Environment, 92(2), 181–194. doi:10.1016/j.rse.2004.06.004
741
Liu, a. Q., Moore, G. W. K., Tsuboki, K., & Renfrew, I. a. (2006). The effect of the sea-ice zone on the
742
development of boundary-layer roll clouds during cold air outbreaks. Boundary-Layer Meteorology,
743
118(3), 557–581. doi:10.1007/s10546-005-6434-4
744
Marshall, J., & Speer, K. (2012). Closure of the meridional overturning circulation through Southern Ocean
745 746
upwelling. Nature Geoscience, 5(3), 171–180. doi:10.1038/ngeo1391
Mccoy, D. T., Hartmann, D. L., & Grosvenor, D. P. (2014). Observed Southern Ocean Cloud Properties
747
and Shortwave Reflection Part I : Calculation of SW flux from observed cloud. Journal of Climate,
748
27, 8836–8857. doi:10.1175/JCLI-D-14-00287.1
749
Mlawer, E.J., S.J. Taubman, P.D. Brown, M.J. Iacono and S.A. Clough (1997). RRTM, a validated
750 751
correlated-k model for the longwave. J. Geophys. Res., 102, 16,663-16,682
Morrison, A. E., Siems, S. T., & Manton, M. J. (2011). A three-year climatology of cloud-top phase over
752
the Southern Ocean and North Pacific. Journal of Climate, 24, 2405–2418.
753
doi:10.1175/2010JCLI3842.1
754
Nicolas, J. P., & Bromwich, D. H. (2011). Climate of West Antarctica and influence of marine air
755
intrusions. Journal of Climate, 24(1), 49–67. doi:10.1175/2010JCLI3522.1
35
756
NOAA/NSIDC Climate Data Record of Passive Microwave Sea Ice Concentration, Version 2, 2015.
757
Subset: Monthly, June 2006-August, 2014. National Snow and Ice Data Center, accessed 1 January,
758
2016. [available online at https://nsidc.org/data]
759
Palm, S. P., Strey, S. T., Spinhirne, J., & Markus, T. (2010). Influence of Arctic sea ice extent on polar
760
cloud fraction and vertical structure and implications for regional climate. J. Geophys. Res.,
761
115(D21), D21209–. doi:10.1029/2010JD013900
762
Papritz, L., Pfahl, S., Sodemann, H., & Wernli, H. (2014). A climatology of cold air outbreaks and their
763
impact on air-sea heat fluxes in the high-latitude South Pacific. Journal of Climate, 28(1), 342–364.
764
doi:10.1175/JCLI-D-14-00482.1
765
Peng, G., Meier, W. N., Scott, D. J., & Savoie, M. H. (2013). A long-term and reproducible passive
766
microwave sea ice concentration data record for climate studies and monitoring. Earth System
767
Science Data, 5(2), 311–318. doi:10.5194/essd-5-311-2013
768
Peterson, G. N., & Renfrew, I. A. (2007). Aircraft-based observations of air-sea fluxes over Denmark Strait
769
and the Irminger Sea during high wind speed Conditions. Quarterly Journal of the Royal
770
Meteorological Society, 133(October), 937–948. doi:10.1002/qj.355
771
Renfrew, I. a., & Moore, G. W. K. (1999). An Extreme Cold-Air Outbreak over the Labrador Sea: Roll
772
Vortices and Air–Sea Interaction. Monthly Weather Review, 127(10), 2379–2394. doi:10.1175/1520-
773
0493(1999)127<2379:AECAOO>2.0.CO;2
774
Schweiger, A. J., Lindsay, R. W., Vavrus, S., & Francis, J. a. (2008). Relationships between Arctic sea ice
775
and clouds during autumn. Journal of Climate, 21(18), 4799–4810. doi:10.1175/2008JCLI2156.1
776
Taylor, G. I., 1921: Diffusion by continuous movement. Proc. London Math. Soc., 20 (2), 196–212.
777
Taylor, K. E., Stouffer, R. J., & Meehl, G. a. (2012). An overview of CMIP5 and the experiment design.
778
Bulletin of the American Meteorological Society, 93(4), 485–498. doi:10.1175/BAMS-D-11-00094.1
36
779
Taylor, P. C., Kato, S., Xu, K. M., & Cai, M. (2015). Covariance between Arctic sea ice and clouds within
780
atmospheric state regimes at the satellite footprint level. Journal of Geophysical Research:
781
Atmospheres, 120(24), 12656–12678. doi:10.1002/2015JD023520
782
Talley, L.D. (2013). Closure of the global overturning circulation through the Indian, Pacific, and Southern
783
Oceans: Schematics and transports. Oceanography 26(1):80–97,
784
http://dx.doi.org/10.5670/oceanog.2013.07.
785
Thorpe, A. J. and Guymer, T. H. (1977). ‘The Nocturnal Jet’, Quart. J. Roy. Meteorol. Soc. 103, 633–653.
786
Trenberth, K. E., & Fasullo, J. T. (2010). Simulation of present-day and twenty-first-century energy
787
budgets of the southern oceans. Journal of Climate, 23, 440–454. doi:10.1175/2009JCLI3152.1
788
Wadhams, P., Lange, M. a, & Ackley, S. F. (1987). The ice thickness distribution across the Atlantic sector
789
of the Antarctic Ocean in midwinter. Journal of Geophysical Research: Oceans (1978--2012),
790
92(C13), 14535–14552.
791
Walter, B. A. (1980). Wintertime observations of roll clouds over the Bering Sea. Monthly Weather
792 793
Review.
Winker, D. M., Hunt, W. H., & McGill, M. J. (2007). Initial performance assessment of CALIOP.
794 795
Geophysical Research Letters, 34(19), 1–5. doi:10.1029/2007GL030135
Zhang, M. H., et al. (2005). Comparing clouds and their seasonal variations in 10 atmospheric general
796
circulation models with satellite measurements. Journal of Geophysical Research D: Atmospheres,
797
110(15), 1–18. doi:10.1029/2004JD005021
37
798
Table Captions
799
Table 1. Summary of observations used to guide the radiative transfer calculations.
800
(top) Cloud fraction partitioned by cloud height and cloud thermodynamic phase
801
from CALIPSO-GOCCP observations. (middle) Summary of the distribution of cloud
802
liquid water path from MAC-LWP observations. Cloud phase and liquid water path
803
are from satellite observations taken during JJA from 2006-2014 over open water
804
near the sea ice edge in the Weddell Sea. (bottom) Cloud-base height measurements
805
from a cloud ceilometer onboard the 1992 cruise. Statistics of cloud-base height are
806
shown for all scenes in which a cloud with a base below 1500 m was detected and
807
surface observers reported open water or open pack ice near the ship. The
808
probability density function shown in parenthesis was computed by dividing the
809
probability of each bin by the width of the bin.
810
Figure Captions
811
Figure 1. Map of Antarctica and the Southern Ocean showing average sea ice
812
concentration during July from 2006-2014 from satellite observations. Two
813
contours of sea ice concentration are shown: 0.35, which we use as a metric for the
814
sea ice edge, and 0.95. Boundaries for the Weddell (50°W-0°) and Ross (130°W-
815
170°E) Seas are shown in the black dashed lines, and tracks for the cruises to the
816
Weddell Sea are shown as colored lines. The cruises started near 0° longitude and
817
finished near the Antarctic Peninsula.
38
818
Figure 2. 2006-2014 climatology of June, July and August low-cloud fraction (color)
819
from CALIPSO-GOCCP observations and the position of the sea ice edge. The red line
820
shows the average position of the sea ice edge, and the red shading shows one
821
standard deviation on either side of the mean.
822
Figure 3. Wintertime cloud fraction, temperature and humidity in the lower
823
troposphere plotted as a function of meridional distance from the sea ice edge. (a)
824
Vertical profile of mean cloud fraction, (b) mean sea ice concentration and low-
825
cloud fraction, with error bars showing the 95% confidence interval of the mean,
826
and (c) mean potential temperature (contours) and specific humidity (color) over
827
the Weddell Sea. (d-f) as in (a-c) but for the Ross Sea. Cloud and sea ice fields come
828
from satellite observations, and temperature and humidity come from ERA Interim
829
reanalysis data. The boundaries for the Weddell Sea and Ross Sea are shown in
830
Figure 1.
831
Figure 4. Temperature profile of the lower troposphere over the Weddell Sea from
832
soundings. For each height, color shows the probability density function of air
833
temperature. Data are composited into soundings taken poleward of 65°S and
834
between 55°-65°S. The number of days in which soundings were collected is shown
835
in the top right corner of each panel. Data from the 2013 cruise are shown in (a-b),
836
and from the 1992 cruise in (c-d). Bins of width 2°C are used in the calculation. The
837
black dashed line shows a profile with a surface temperature of -1.8°C, which is
838
about the freezing temperature of seawater in the Southern Ocean, and a moist
839
adiabatic lapse rate. Note that two boundary layer regimes are seen: a warm mode
840
with near-surface temperatures close to the freezing temperature of seawater and
39
841
with a most adiabatic lapse rate, and a cold mode with near-surface temperatures
842
from -15°C to -25°C and with a low-level inversion.
843
Figure 5. Vertical profile of wind speed from soundings in the Weddell Sea region
844
poleward of 55°S. For each height, color shows the probability density function of
845
wind speed. Bins of width 3 ms-1 are used in the calculation. Data from the 2013 and
846
1992 cruises are shown in the top and bottom panels, respectively. Data were
847
collected over 53 days on both cruises. Note that wind speeds of 10 ms-1 or more are
848
common at heights between 200-600 m, and that the signature of a low-level jet can
849
be seen in the measurements from the 2013 cruise.
850
Figure 6. Mean low-cloud fraction observed by CALIPSO as a function of meridional
851
distance from the sea ice edge, and its dependence on low-level warm or cold
852
advection. Observations are composited into periods of poleward, on-ice flow at
853
low-levels (𝑣!"# !"#! < −0.5𝜎 ≈ −3 ms-1, where 𝜎 is the standard deviation of
854
𝑣!"# !"#! ) and periods of equatorward, off-ice flow at low-levels (𝑣!"# !"#! > 0.5𝜎 ≈ 3
855
ms-1). Averages are computed using daily-mean data, and error bars show the 95%
856
confidence interval of the mean. Over open ocean, cloudier conditions are seen
857
during periods of off-ice advection. Over sea ice, low-cloud fraction is similar for
858
periods of on-ice and off-ice advection.
859
Figure 7. Computed surface 𝐿𝑊↓ over open water near the sea ice edge plotted as a
860
function of near-surface meridional wind at the sea ice edge (𝑣!"# !"#! ). The dots
861
show individual 𝐿𝑊↓,!""!!"# values, and the blue line shows 𝐿𝑊↓,!""!!"# binned by
862
𝑣!"# !"#! and averaged. Error bars on the blue line are the 95% confidence interval of
863
the mean. The red, solid black, and dashed black lines show linear regressions of
40
864
𝐿𝑊↓,!""!!"# , 𝐿𝑊↓,!"#$%&'( , and 𝐿𝑊↓,!"#$% on 𝑣!"# !"#! , respectively. The red shading is
865
the 95% confidence interval for the regression slope of 𝐿𝑊↓,!""!!"# . Note that surface
866
LW cloud radiative effect, seen in the figure as the difference between 𝐿𝑊↓,!""!!"#
867
and 𝐿𝑊↓,!"#$% , increases with 𝑣!"# !"#! . As a result, surface LW cloud radiative effect
868
is largest during periods of strong off-ice flow.
869
Figure 8. July-mean low-cloud fraction and latitude of the sea ice edge in climate
870
models and observations. Low-cloud fraction is shown in blue, and the sea ice edge
871
is shown in red (the red line shows the mean, and red shading shows one standard
872
deviation on either side of the mean). Each model is shown with output from fully-
873
coupled and atmosphere-only configuration. Low-cloud fraction in the models was
874
computed by a CALIPSO simulator, and observations are from CALIPSO-GOCCP.
875
Figure 9. Evaluation of wintertime low-cloud fraction near the sea ice edge in
876
climate models. (a) CALIPSO observed and simulated mean low-cloud fraction over
877
open water (1°-3° equatorward of the sea ice edge) and sea ice (2°-4° poleward of
878
the sea ice edge). The gray plus sign shows the 95% confidence interval for the
879
observed value. Models in both atmosphere-only and fully-coupled configurations
880
are shown. Six models have both atmosphere-only and fully-coupled output, and
881
these models are labeled with bold text in the legend. For these models, the
882
atmosphere-only and fully-coupled data points are connected by dashed lines, but
883
for some models the difference is small and the dashed line is not visible. Note that
884
all but one model (GFDL-CM3) underestimate low-cloud fraction over open water,
885
and all but one fully-coupled model (MPI-ESM-LR) underestimate low-cloud fraction
41
886
over sea ice. (b) Difference between mean low-cloud fraction over open water and
887
over sea ice near the sea ice edge. Error bars show the 95% confidence interval.
888
42
889
Tables Cloud Phase
Level
Cloud Fraction
Total
Liquid
Ice
Unclassified
High
0.29
0
0.28
0.02
Middle
0.27
0.09
0.16
0.04
Low
0.69
0.50
0.05
0.14
Liquid Water Path
Percentile
5
25
50
75
95
Liquid
43.1
54.1
61.4
70.1
79.1
water path (g/m2) Cloud-Base Height Height Range
0-50
(m) Counts (Probability
12 (.10)
50-
100-
200-
100
200
300
11
22
36
(0.09)
(0.09)
(0.15)
300-600
58 (0.08)
600-
1000-
1000
1500
67
34
(0.07)
(0.03)
density ×10!! (m-1)) 890
891
Table 1. Summary of observations used to guide the radiative transfer calculations.
892
(top) Cloud fraction partitioned by cloud height and cloud thermodynamic phase
43
893
from CALIPSO-GOCCP observations. (middle) Summary of the distribution of cloud
894
liquid water path from MAC-LWP observations. Cloud phase and liquid water path
895
are from satellite observations taken during JJA from 2006-2014 over open water
896
near the sea ice edge in the Weddell Sea. (bottom) Cloud-base height measurements
897
from a cloud ceilometer onboard the 1992 cruise. Statistics of cloud-base height are
898
shown for all scenes in which a cloud with a base below 1500 m was detected and
899
surface observers reported open water or open pack ice near the ship. The
900
probability density function shown in parenthesis was computed by dividing the
901
probability of each bin by the width of the bin.
902
44
903
Figures Weddell! Sea
0°
0.95 90°W
90°E 60°S
0.35
70°S
sea ice! concentration
50°W
130°W
Ross! Sea
904
170°E
905
Figure 1. Map of Antarctica and the Southern Ocean showing average sea ice
906
concentration during July from 2006-2014 from satellite observations. Two
907
contours of sea ice concentration are shown: 0.35, which we use as a metric for the
908
sea ice edge, and 0.95. Boundaries for the Weddell (50°W-0°) and Ross (130°W-
909
170°E) Seas are shown in the black dashed lines, and tracks for the cruises to the
910
Weddell Sea are shown as colored lines. The cruises started near 0° longitude and
911
finished near the Antarctic Peninsula.
912
45
June
July
August 0.7 0.6 0.5
914
70°
70°
70°
60°
60°
60°
50°
50°
50°
0.4
low-cloud fraction
913
915
Figure 2. 2006-2014 climatology of June, July and August low-cloud fraction (color)
916
from CALIPSO-GOCCP observations and the position of the sea ice edge. The red line
917
shows the average position of the sea ice edge, and the red shading shows one
918
standard deviation on either side of the mean.
46
Ross Sea 0.3 .3
2 2
0.25
1.5 1.5
0.2 .2
1 1
0.15
0.5 0.5
0.1 .1
-4
distance from ice edge (° lat)
920
0
2
2
4
6 4
0.675 0.675 0.6 0.6
0.45 0.45 -6 -6
-4 -4
-2 -2
0 0
2 2
33
280
2.5
22
900 265
1.5
270
0.5 0.5 0.25 0.25
11
260 -6
0.75 0.75
3.5
275
950
1 1
0 6 0 6
850
south
6
4 4
Potential Temp./q
800
1000
north
0
0.525 0.525
f
south
-2
LCF/SIC LCF/SIC
750
c
-2
sea ice conc.
e
LowCloud CloudFraction Fraction Low
low-cloud frac.
sea ice conc.
b
-4
0.75 0.75
Pressure (hPa)
pressure (hPa)
low-cloud frac.
-6 -6
SeaIce IceConcentration Concentration Sea
d
cloud fraction
Cloud Incidence Cloud Incidence
2.5 2.5
a
Height(km) (km) Height
height (km)
Weddell Sea
specific! humidity (g/kg)
919
-4
-2
0
2
4
Distance from Ice Edge (degrees)
6
north
distance from ice edge (° lat)
921
Figure 3. Wintertime cloud fraction, temperature and humidity in the lower
922
troposphere plotted as a function of meridional distance from the sea ice edge. (a)
923
Vertical profile of mean cloud fraction, (b) mean sea ice concentration and low-
924
cloud fraction, with error bars showing the 95% confidence interval of the mean,
925
and (c) mean potential temperature (contours) and specific humidity (color) over
926
the Weddell Sea. (d-f) as in (a-c) but for the Ross Sea. Cloud and sea ice fields come
927
from satellite observations, and temperature and humidity come from ERA Interim
928
reanalysis data. The boundaries for the Weddell Sea and Ross Sea are shown in
929
Figure 1.
47
poleward of 65°S
65°-55°S (25)
b
(28)
c
(34)
d
(22)
probability
1992
Height (m)
2013
a
930
Temperature (°C)
931
Figure 4. Temperature profile of the lower troposphere over the Weddell Sea from
932
soundings. For each height, color shows the probability density function of air
933
temperature. Data are composited into soundings taken poleward of 65°S and
934
between 55°-65°S. The number of days in which soundings were collected is shown
935
in the top right corner of each panel. Data from the 2013 cruise are shown in (a-b),
936
and from the 1992 cruise in (c-d). Bins of width 2°C are used in the calculation. The
937
black dashed line shows a profile with a surface temperature of -1.8°C, which is
938
about the freezing temperature of seawater in the Southern Ocean, and a moist
939
adiabatic lapse rate. Note that two boundary layer regimes are seen: a warm mode
940
with near-surface temperatures close to the freezing temperature of seawater and
941
with a most adiabatic lapse rate, and a cold mode with near-surface temperatures
942
from -15°C to -25°C and with a low-level inversion.
48
2013
height (m)
1400
1000
600 0.3
5
10
15
20
1992
0.2
0.1
probability
200
1400
height (m)
0 1000
600
200 5
10
15
20
wind speed (ms-1)
943
944
Figure 5. Vertical profile of wind speed from soundings in the Weddell Sea region
945
poleward of 55°S. For each height, color shows the probability density function of
946
wind speed. Bins of width 3 ms-1 are used in the calculation. Data from the 2013 and
947
1992 cruises are shown in the top and bottom panels, respectively. Data were
948
collected over 53 days on both cruises. Note that wind speeds of 10 ms-1 or more are
949
common at heights between 200-600 m, and that the signature of a low-level jet can
950
be seen in the measurements from the 2013 cruise.
49
low-cloud fraction
0.8 0.8
off-ice flow
0.7 0.7
0.6 0.6
on-ice flow
0.5 0.5
-6 -6
-4 -4
-2 -2
0 0
2 2
4
south distance from ice edge (° latitude)
951
6 6
north
952
Figure 6. Mean low-cloud fraction observed by CALIPSO as a function of meridional
953
distance from the sea ice edge, and its dependence on low-level warm or cold
954
advection. Observations are composited into periods of poleward, on-ice flow at
955
low-levels (𝑣!"# !"#! < −0.5𝜎 ≈ −3 ms-1, where 𝜎 is the standard deviation of
956
𝑣!"# !"#! ) and periods of equatorward, off-ice flow at low-levels (𝑣!"# !"#! > 0.5𝜎 ≈ 3
957
ms-1). Averages are computed using daily-mean data, and error bars show the 95%
958
confidence interval of the mean. Over open ocean, cloudier conditions are seen
959
during periods of off-ice advection. Over sea ice, low-cloud fraction is similar for
960
periods of on-ice and off-ice advection.
50
surface LW↓ (W/m2)
on-ice! flow
961
vice edge (ms-1)
off-ice! flow
962
Figure 7. Computed surface 𝐿𝑊↓ over open water near the sea ice edge plotted as a
963
function of near-surface meridional wind at the sea ice edge (𝑣!"# !"#! ). The dots
964
show individual 𝐿𝑊↓,!""!!"# values, and the blue line shows 𝐿𝑊↓,!""!!"# binned by
965
𝑣!"# !"#! and averaged. Error bars on the blue line are the 95% confidence interval of
966
the mean. The red, solid black, and dashed black lines show linear regressions of
967
𝐿𝑊↓,!""!!"# , 𝐿𝑊↓,!"#$%&'( , and 𝐿𝑊↓,!"#$% on 𝑣!"# !"#! , respectively. The red shading is
968
the 95% confidence interval for the regression slope of 𝐿𝑊↓,!""!!"# . Note that surface
969
LW cloud radiative effect, seen in the figure as the difference between 𝐿𝑊↓,!""!!"#
970
and 𝐿𝑊↓,!"#$% , increases with 𝑣!"# !"#! . As a result, surface LW cloud radiative effect
971
is largest during periods of strong off-ice flow.
51
IPSL-CM5A-LR
MPI-ESM-LR
MRI-CGCM3
IPSL-CM5A-MR
MIROC5
atmosphere-! only
fully-! coupled
HadGEM2
fully-! coupled
Observations
0.6
atmosphere-! only
0.5 0.4 70°
low-cloud fraction
0.7
0.3
60°
972
973
Figure 8. July-mean low-cloud fraction and latitude of the sea ice edge in climate
974
models and observations. Low-cloud fraction is shown in blue, and the sea ice edge
975
is shown in red (the red line shows the mean, and red shading shows one standard
976
deviation on either side of the mean). Each model is shown with output from fully-
977
coupled and atmosphere-only configuration. Low-cloud fraction in the models was
978
computed by a CALIPSO simulator, and observations are from CALIPSO-GOCCP.
52
LCFsea ice
0.7 0.7
1:1
a
0.7
0.6 0.6 0.5 0.5
0.6
0.4 0.4 0.3 0.3
0.5
0.3
0.3
0.4
0.4
0.5
0.5
0.6
0.6
0.4
0.3
b
0.3
0.2 0.2 0.3
0.4
0.5
0.6
0.1
00 -0.1 -0.2 -0.2
HH aadd GG IIPP EEMM SSLL 22 --CC MM IIPP 55AA SSLL --LL RR --CC MM 55AA --MM RR MM IIRR O O MM CC5 PPII 5 --EE SSMM --LL MM R RRI I--CC R GGC CM MM M IIRR 33 OOC C-EE S SMM --CC HHE EMM MM IIRR OOC C-EE SSMM CCN NR RM M-CC MM5 GGF 5 FDD LL-CCM M3 3
979
0.7
0.7
0.4 0.4
oobb ssee rrvv aatt ioio nnss
LCFopen water - LCFsea ice
LCFopen water
HadGEM2 IPSL-CM5A-LR IPSL-CM5A-MR MIROC5 MPI-ESM-LR MRI-CGCM3 MIROC-ESM-CHEM MIROC-ESM CNRM-CM5 GFDL-CM3 observations atmosphere-only fully-coupled
980
Figure 9. Evaluation of wintertime low-cloud fraction near the sea ice edge in
981
climate models. (a) CALIPSO observed and simulated mean low-cloud fraction over
982
open water (1°-3° equatorward of the sea ice edge) and sea ice (2°-4° poleward of
983
the sea ice edge). The gray plus sign shows the 95% confidence interval for the
984
observed value. Models in both atmosphere-only and fully-coupled configurations
985
are shown. Six models have both atmosphere-only and fully-coupled output, and
986
these models are labeled with bold text in the legend. For these models, the
987
atmosphere-only and fully-coupled data points are connected by dashed lines, but
988
for some models the difference is small and the dashed line is not visible. Note that
989
all but one model (GFDL-CM3) underestimate low-cloud fraction over open water,
990
and all but one fully-coupled model (MPI-ESM-LR) underestimate low-cloud fraction
53
0.7
991
over sea ice. (b) Difference between mean low-cloud fraction over open water and
992
over sea ice near the sea ice edge. Error bars show the 95% confidence interval.
54
