• Aagaard, K., , and E. C. Carmack, 1989: The role of sea ice and other freshwater in the Arctic circulation. J. Geophys. Res, 94 , 305311.

    • Search Google Scholar
    • Export Citation
  • Arfeuille, G., , L. A. Maysak, , and L-B. Tremblay, 2000: Simulation of the interannual variability of the wind-driven Arctic sea-ice cover during 1958–1998. Climate Dyn, 16 , 107121.

    • Search Google Scholar
    • Export Citation
  • Busch, N., , U. Ebel, , H. Kraus, , and E. Schaller, 1982: The structure of the subpolar inversion-capped ABL. Arch. Meteor. Geophys. Bioklimatol, 31A , 118.

    • Search Google Scholar
    • Export Citation
  • Curry, J., 1983: On the formation of polar continental air. J. Atmos. Sci, 40 , 22782292.

  • Deser, C., , and M. L. Blackmon, 1993: Surface climate variations over the North Atlantic Ocean during winter: 1900–1989. J. Climate, 6 , 17431754.

    • Search Google Scholar
    • Export Citation
  • Deser, C., , J. E. Walsh, , and M. S. Timlin, 2000: Arctic sea ice variability in the context of recent atmospheric circulation trends. J. Climate, 13 , 617633.

    • Search Google Scholar
    • Export Citation
  • Dickson, R. R., 1988: Great Salinity Anomaly in the northern North Atlantic 1968–1982. Progress in Oceanography, Vol. 20, Pergamon Press, 103–151.

    • Search Google Scholar
    • Export Citation
  • Dickson, R. R., and Coauthors, 2000: The Arctic Ocean response to the North Atlantic Oscillation. J. Climate, 13 , 26712696.

  • Dong, B-W., , and R. T. Sutton, 2002: Adjustment of the coupled ocean–atmosphere system to a sudden change in the thermohaline circulation. Geophys. Res. Lett.,29, 1728, doi:10.1029/ 2002GL015229.

    • Search Google Scholar
    • Export Citation
  • Gamo, M., 1996: Thickness of the dry convection and large-scale subsidence above deserts. Bound.-Layer Meteor, 79 , 265278.

  • Gao, D-Y., , and B-Y. Wu, 1998: A preliminary study on decadal oscillation and its oscillation source of sea–ice–air system in the Northern Hemisphere. Polar Meteor. Glaciol, 12 , 6878.

    • Search Google Scholar
    • Export Citation
  • Häkkinen, S., 1993: An Arctic source for the great salinity anomaly: A simulation of the Arctic ice–ocean system for 1955–1975. J. Geophys. Res, 98 , 1639716410.

    • Search Google Scholar
    • Export Citation
  • Herman, G. F., , and W. T. Johnson, 1978: The sensitivity of the general circulation to Arctic sea ice boundaries: A numerical experiment. Mon. Wea. Rev, 106 , 16491664.

    • Search Google Scholar
    • Export Citation
  • Hilmer, M., , and T. Jung, 2000: Evidence for a recent change in the link between the North Atlantic Oscillation and Arctic sea ice export. Geophys. Res. Lett, 27 , 989992.

    • Search Google Scholar
    • Export Citation
  • Honda, M., , K. Yamazaki, , Y. Tachibana, , and K. Takeuchi, 1996: Influence of Okhotsk sea-ice extent on atmospheric circulation. Geophys. Res. Lett, 23 , 35953598.

    • Search Google Scholar
    • Export Citation
  • Kahl, J. D., 1990: Characteristics of the low-level temperature inversion along the Alaska Arctic coast. Int. J. Climatol, 10 , 537548.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc, 77 , 437471.

  • Lazier, J., 1988: Temperature and salinity changes in the deep Labrador Sea 1962–1986. Deep-Sea Res, 35 , 12471253.

  • Lenderink, G., , and R. J. Haarsma, 1996: Modeling convective transitions in the presence of sea ice. J. Phys. Oceanogr, 26 , 14481467.

  • Mauritzen, C., , and S. Häkkinen, 1997: Influence of sea ice on the thermohaline circulation in the Arctic–North Atlantic Ocean. Geophys. Res. Lett, 24 , 32573260.

    • Search Google Scholar
    • Export Citation
  • Mysak, L. A., , and S. B. Power, 1992: Sea-ice anomalies in the western Arctic and Greenland–Iceland sea and their relation to an interdecaldal climate cycle. Climatol. Bull, 26 , 147176.

    • Search Google Scholar
    • Export Citation
  • Mysak, L. A., , and S. A. Venegas, 1998: Decadal climate oscillation in the Arctic: A new feedback loop for atmosphere–ice–ocean interactions. Geophys. Res. Lett, 25 , 36073610.

    • Search Google Scholar
    • Export Citation
  • Mysak, L. A., , D. K. Manak, , and R. F. Marsden, 1990: Sea-ice anomalies observed in the Greenland and Labrador Seas during 1901–1984 and their relation to an interdecadal Arctic climate cycle. Climate Dyn, 5 , 111133.

    • Search Google Scholar
    • Export Citation
  • Rigor, I. G., , J. M. Walsh, , and R. L. Colony, 2002: Response of sea ice to the Arctic Oscillation. J. Climate, 15 , 26482663.

  • Serreze, M. C., , J. D. Kahl, , and R. C. Schnell, 1992: Low-level inversions of the Eurasian Arctic and comparisons with Soviet drifting station data. J. Climate, 5 , 615629.

    • Search Google Scholar
    • Export Citation
  • Shinoda, M., , and H. Utsugi, 2001: Spring snow-disappearance timing and its possible influences on temperature fields over central Eurasia. J. Meteor. Soc. Japan, 79 , 3759.

    • Search Google Scholar
    • Export Citation
  • Skeie, P., 2000: Meridional flow variability over the Nordic Seas in the Arctic Oscillation framework. Geophys. Res. Lett, 27 , 25692572.

    • Search Google Scholar
    • Export Citation
  • Slonosky, V. C., , L. A. Mysak, , and J. Derome, 1997: Linking Arctic sea ice and atmospheric circulation anomalies on interannual and decadal timescales. Atmos.–Ocean, 35 , 333366.

    • Search Google Scholar
    • Export Citation
  • Tremblay, L-B., , and L. A. Mysak, 1997: Modeling sea ice as a granular material, including the dilatancy effect. J. Phys. Oceanogr, 27 , 23422360.

    • Search Google Scholar
    • Export Citation
  • Tremblay, L-B., , and L. A. Mysak, 1998: On the origin and evolution of sea ice anomalies in the Beaufort–Chukchi Sea. Climate Dyn, 14 , 451460.

    • Search Google Scholar
    • Export Citation
  • Vavrus, S., , and S. P. Harrison, 2003: The impact of sea-ice dynamics on the Arctic climate system. Climate Dyn, 20 , 741757.

  • Vinje, T., 2001a: Fram Strait ice fluxes and atmospheric circulation: 1950–2000. J. Climate, 14 , 35083517.

  • Vinje, T., 2001b: Anomalies and trends of sea ice extent and atmospheric circulation in the Nordic Seas during the period 1864–1998. J. Climate, 14 , 255267.

    • Search Google Scholar
    • Export Citation
  • Walsh, J. E., 1983: Role of sea ice in climate variability: Theories and evidence. Atmos.–Ocean, 21 , 229242.

  • Walsh, J. E., , and W. L. Chapman, 1990: Arctic contribution to upper-ocean variability in the North Atlantic. J. Climate, 3 , 14621473.

    • Search Google Scholar
    • Export Citation
  • Wang, J., , and M. Ikeda, 2000: Arctic Oscillation and Arctic sea-ice oscillation. Geophys. Res. Lett, 27 , 12871290.

  • Wu, B-Y., , and J. Wang, 2002: Possible impacts of winter Arctic Oscillation on the Siberian High, the east Asian winter monsoon and sea-ice. Adv. Atmos. Sci, 19 , 297320.

    • Search Google Scholar
    • Export Citation
  • Wu, B-Y., , D-Y. Gao, , and R-H. Huang, 1997: ENSO events and interannual variations of winter sea-ice in the Greenland, the Kara and the Barents Seas. Chin. Sci. Bull, 42 , 13821385.

    • Search Google Scholar
    • Export Citation
  • Wu, B-Y., , R-H. Huang, , and D-Y. Gao, 1999: Impact of variations of winter sea-ice extents in the Kara/Barents Seas on winter monsoon over east Asia. Acta Meteor. Sin, 13 , 141153.

    • Search Google Scholar
    • Export Citation
  • Wu, B-Y., , R-H. Huang, , and D-Y. Gao, 2001: Arctic sea ice bordering on the North Atlantic and interannual climate variations. Chin. Sci. Bull, 46 , 162165.

    • Search Google Scholar
    • Export Citation
  • Wu, B-Y., , R-H. Huang, , and D-Y. Gao, 2002: Numerical simulations on effects of variation of sea ice thickness and extent on atmospheric circulation. Acta Meteor. Sin, 16 , 150164.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Variations of winter (Feb–Apr) mean sea ice extent in the Greenland and Barents Seas. The dashed lines stand for climatological mean and standard deviations for the period 1953–95. Unit is 106 km2

  • View in gallery

    Winter surface wind composite (10 m): (a) the heavy ice winters and (b) the light ice winters. The gray and dark areas denote that the differences in mean northerly winds between the heavy and light ice winters (heavy minus light) exceed the 0.05 and 0.01 confidence levels, respectively. Unit is m s−1

  • View in gallery

    Sea ice extent composite for (a) the heavy ice winters and (b) the light ice winters. The solid curve denotes the location of sea ice edge (sea ice concentration = 15%)

  • View in gallery

    Same as Fig. 2 except for SLP anomalies (hPa)

  • View in gallery

    Same as Fig. 2 except for surface air temperature anomalies (°C)

  • View in gallery

    Differences in mean air temperature differences (°C; low level minus high level) between the heavy and light ice winters: (a) 925 minus 850 hPa, and (b) 850 minus 700 hPa. The gray and dark gray areas denote that differences between the heavy and light ice winters exceed the 0.01 and 0.001 confidence levels, respectively

  • View in gallery

    (a) Winter mean vertical θse gradient between 850 and 700 hPa (1958–99), (b) vertical θse gradient anomaly composite for the heavy ice winters, and (c) same as (b) but for the light ice winters. Unit is K km−1. The meaning for shading area is the same as in Fig. 2

  • View in gallery

    Variance fraction contribution of winter mean temperature advection between 850 and 700 hPa to the vertical θse gradient. The shading area denotes variance contributions of winter mean temperature advection ≥30%

  • View in gallery

    Latitude–pressure section of air temperature difference composite [°C; high minus low level, T(i, j + 1) − T(i, j), j = 1, …, 6, corresponding to 1000, 925, 850, 700, 600, 500 hPa, respectively; i = 1, … , 13, latitude points] along 15°W: (a) the heavy ice winters and (b) the light ice winters. The meaning for the shaded area is the same as Fig. 2

  • View in gallery

    The longitude–pressure section along 75°N: (a) winter mean divergence (1958–99), (b) divergence anomaly composite for the heavy ice winters, and (c) same as (b) except for the light ice winters. The gray area represents negative anomalies, and unit is 10−7 s−1

  • View in gallery

    Schematic diagram of the dynamic feedback of the (a) heavy and (b) light sea ice conditions on the local atmosphere

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 27 27 6
PDF Downloads 27 27 4

Possible Feedback of Winter Sea Ice in the Greenland and Barents Seas on the Local Atmosphere

View More View Less
  • 1 Chinese Academy of Meteorological Sciences, Beijing, China, and Institute of Marine Sciences, University of Alaska, Fairbanks, Fairbanks, Alaska
  • | 2 International Arctic Research Center, University of Alaska, Fairbanks, Fairbanks, Alaska
© Get Permissions
Full access

Abstract

Using monthly Arctic sea ice concentration data (1953–95) and the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis dataset (1958–99), possible feedbacks of sea ice variations in the Greenland and Barents Seas to the atmosphere are investigated. Winter (February–April) sea ice anomalies in the Greenland and Barents Seas display important feedback influences on the atmospheric boundary layer in terms of both thermodynamic and dynamic processes. The vertical gradients of potential pseudo-equivalent temperature (θse) between 850 and 700 hPa are greater over sea ice than over open water, implying that a more stable boundary layer forms below 700 hPa over sea ice. The effects of temperature advection are shown to account for a relatively small percentage of the temperature variance in area of sea ice feedbacks. Horizontal and vertical variations of the effects of sea ice on temperature in the Nordic and Barents Seas create the potential for dynamical influences on the atmospheric boundary layer through vertical motion induced by the pressure anomalies in the boundary layer.

Corresponding author address: Dr. Bingyi Wu, Chinese Academy of Meteorlogical Sciences, Beijing 100081, China. Email: wby@cams.cma.gov.cn

Abstract

Using monthly Arctic sea ice concentration data (1953–95) and the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis dataset (1958–99), possible feedbacks of sea ice variations in the Greenland and Barents Seas to the atmosphere are investigated. Winter (February–April) sea ice anomalies in the Greenland and Barents Seas display important feedback influences on the atmospheric boundary layer in terms of both thermodynamic and dynamic processes. The vertical gradients of potential pseudo-equivalent temperature (θse) between 850 and 700 hPa are greater over sea ice than over open water, implying that a more stable boundary layer forms below 700 hPa over sea ice. The effects of temperature advection are shown to account for a relatively small percentage of the temperature variance in area of sea ice feedbacks. Horizontal and vertical variations of the effects of sea ice on temperature in the Nordic and Barents Seas create the potential for dynamical influences on the atmospheric boundary layer through vertical motion induced by the pressure anomalies in the boundary layer.

Corresponding author address: Dr. Bingyi Wu, Chinese Academy of Meteorlogical Sciences, Beijing 100081, China. Email: wby@cams.cma.gov.cn

1. Introduction

Arctic sea ice is an important component of the climate system. On the one hand, it is an important indicator of changes in the climate system, such as a global warming trend that is likely to be amplified in high latitudes. On the other hand, observations and climate modeling have suggested that sea ice can itself be an agent of climate change (Deser et al. 2000; Honda et al. 1996; Mysak et al. 1990; Slonosky et al. 1997; Walsh 1983; Wang and Ikeda 2000; Wu et al. 1999, 2002). As shown by Kahl (1990), sea ice influences the surface and atmospheric boundary layer temperature by affecting the interface net radiation cooling and the exchange of heat between the atmosphere and the ocean. Potentially even more important for global climate is the transport of sea ice out of the Arctic from the Fram Strait (Dickson et al. 2000), which would influence the stability and variability of oceanic stratification, the deep convection, and thermohaline circulation (THC). Sea ice affects the ocean buoyancy and the THC through both thermal and freshwater impacts. The recent study of Dong and Sutton (2002) indicates that atmospheric feedbacks could spread the influence of a change in the THC around the globe much more quickly and efficiently than ocean processes alone.

The Greenland and Barents Seas are the crucial regions connecting the Arctic Ocean and the North Atlantic Ocean. Sea ice variation in these regions is closely related to freshwater output of the Arctic Ocean (Aagaard and Carmack 1989; Vinje 2001a), with its links to the North Atlantic deep water variation (Lenderink and Haarsma 1996), the thermohaline circulation adjustment (Mauritzen and Häkkinen 1997), and interdecadal climate variability (Lazier 1988; Deser and Blackmon 1993; Mysak and Venegas 1998). During the last several decades, many studies demonstrated how climate variation influences sea ice in the North Atlantic subpolar area (Dickson et al. 2000; Hilmer and Jung 2000; Wang and Ikeda 2000; Wu et al. 2001; Wu and Wang 2002; and many others) and the linkage between the atmosphere and sea ice (Deser et al. 2000; Rigor et al. 2002). Although Mysak and Power (1992) and Deser et al. (2000) suggested that sea ice variation can influence cyclone activities and sea level pressure (SLP) variations by means of diagnostic and numerical modeling methods, few studies focus on the possible feedback of sea ice variation on the atmosphere, particularly the boundary layer structure. The major reason is that feedback or other influences of sea ice anomalies on the atmosphere have been difficult to detect because of the dominance of atmospheric forcing of sea ice in the observed association (Deser et al. 2000).

A series of studies suggested that the Barents and Kara Seas are crucial regions, where variations of sea ice extent are closely related to climate variations over a large area from the North Atlantic to eastern Asia on interannual and decadal time scales (Wu et al. 1997, 1999, 2001, 2002; Wu and Wang 2002; Gao and Wu 1998). However, cause and effect are not clear yet. Fundamentally, it is necessary to consider ice–atmosphere interactions in a framework of coupled interactions or feedbacks. Otherwise, those previous studies concerning effects of sea ice on the atmosphere, particularly teleconnections to remote regions, may be seriously incomplete. The purpose of the present study is to explore the mechanisms of coupling between the atmosphere and sea ice through the boundary layer. Because sea ice extent in the subarctic seas generally reaches its maximum extent during the late winter, we focus on the February–April period.

2. Dataset

Monthly sea ice concentration data for the period of 1953–95 were obtained from the National Snow and Ice Data Center (NSIDC). This dataset contains rectangular grids of monthly sea ice concentrations at a resolution of approximately 111 km (60 n mi). The grids depict ice conditions at the end of each of 516 months. The monthly values are interpolated to the last day of each month, rather than being values for the last week of each month. For detailed information, see the Web site available online at http://nsidc.org/data/docs/noaa/g00799_arctic_southern_sea_ice/index.html. In this study, sea ice concentration values at the end of each month were interpolated to the monthly mean.

The monthly mean gridpoint atmospheric fields, including near-surface air temperature, wind fields (10 m), sea level pressure, temperature, geopotential height, divergence, and specific humidity at each standard isobaric surface from 1000 to 100 hPa, were obtained from the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis data for the period 1958–99 (Kalnay et al. 1996). In the cases of all variables (sea ice, atmosphere), we computed the average for the late-winter (February– April) period, when sea ice is close to its maximum extent.

3. Results

a. Low-level circulations and SLP anomalies

Over the past four decades, sea ice extents in the Greenland and Barents Seas experienced remarkable variations on interannual and interdecadal time scales (Fig. 1). From the early 1950s to late 1960s, the Greenland and Barents Seas experienced more-than-normal sea ice; sea ice extent reached its maximum at the end of the 1960s. A sharp increase in sea ice extent occurred between 1976 and 1977.

In order to detect possible feedback of winter sea ice extents on the atmospheric boundary layer, we chose from the past several decades those winters for which the anomaly of winter sea ice extent was >1.0 standard deviation (heavy ice winter) or <−1.0 standard deviation (light ice winter). The heavy ice winters include the winters ending in 1963, 1966, 1968, 1969, and 1977, and the light ice winters are the winters ending in 1983, 1984, 1990, 1991, 1992, 1993, and 1995. We perform a composite analysis based on these cases.

For the heavy ice winters, northerly winds prevail over the Fram Strait and the northern Barents and Norwegian Seas (Fig. 2a), which drive more sea ice out of the Arctic Ocean (Fig. 3a). Dickson (1988) pointed out that the sea ice anomaly in the Greenland Sea during the 1960s was related to the “Great Salinity Anomaly” and was explained by the local wind anomaly forcing. Skeie (2000) also suggested that ocean convection and the variability of the Arctic Ocean ice export through the Fram Strait and the Barents Seas is related to persistent anomalous northerly winds. Anomalous wind forcing not only over the Greenland Sea but also over the Arctic basin drives sea ice anomalies (Walsh and Chapman 1990; Rigor et al. 2002). Model simulation results have further confirmed that much of the variation in sea ice export out of the Arctic basin can be explained by the local and large-scale wind fields (Häkkinen 1993). Compared to the heavy ice winters, southwesterly winds prevail over the whole Barents Sea, Svalbard, and part of the Greenland Sea (Fig. 2b), opposing sea ice export southward from the Arctic basin via the Fram Strait (Fig. 3b). Vinje (2001b) suggested that variations in the ocean temperature and its positive or negative correlation with wind direction seems to be of crucial importance for the variation in the ice extent, and air temperature plays a minor role. Consequently, the local wind field anomalies produced by the atmospheric circulation and the ocean current are the principal factors in determining sea ice anomalies, with the latter especially important in the Fram Strait region.

The surface wind forcing on sea ice needs to be assessed carefully based on the statistical significant test for surface meridional wind changes between the heavy and the light ice winters (Fig. 2). Clearly, there is no statistical significance in surface meridional wind changes over the Greenland and Barents Seas. Significant changes only occurred in the Norwegian Sea and the area around Iceland. Consequently, the simultaneous surface meridional wind forcing on sea ice may not be important over the Greenland and Barents Seas during this period. The previous sea ice anomalies produced by surface wind forcing and the oceanic currents in the Arctic basin may be even more important. Arfeuille et al. (2000), based on the thermodynamic–dynamic sea ice model developed by Tremblay and Mysak (1997), first investigated the variability of the sea ice volume in the Arctic basin and the subsequent changes in the sea ice export into the Greenland Sea via the Fram Strait. Their results clearly indicated that the sea ice export anomalies are linked to prior sea ice volume anomalies in the Arctic basin. The model integration by Tremblay and Mysak (1998) also showed that sea ice anomalies originating in the western Beaufort Sea can survive a few seasonal cycles as they propagate through the Arctic basin and can account for a notable amount of anomalous ice export into the Greenland Sea. In addition, a fully atmospheric–ice interaction by Vavrus and Harrison (2003) also indicated that sea ice motion not only influences surface air temperature and SLP but also strongly affects surface winds, such that there is a greater northerly component along the boundary of the Arctic Ocean and Greenland–Iceland–Norway (GIN) Seas when the ice is mobile southward.

The distribution of SLP anomalies during the heavy (light) ice winters is similar to that during the negative (positive) phase of the North Atlantic Oscillation (NAO) (Fig. 4). Interestingly, the maximum positive anomalies of SLP are centered over the Greenland Sea (Fig. 4a) (we did not consider SLP anomalies over Greenland because of its high terrain). There are only two possible explanations for the SLP maximum anomalies: 1) SLP anomalies lead to wind anomalies and further influence sea ice, which is a traditional explanation, and 2) sea ice anomalies partly determine the local SLP anomalies; for example, Honda et al. (1996) suggested that the heavy ice gives rise to positive SLP anomalies in the Okhotsk Sea. Deser et al. (2000) also concluded that mean SLP may respond to sea ice, and the local decrease in SLP over the reduced ice extent in the Greenland Sea in their composite analysis of observational data is qualitatively consistent with the results of idealized atmospheric model experiments (Herman and Johnson 1978). In this respect, SLP anomalies over the Greenland Sea shown in Fig. 4a should partly reflect the feedback of sea ice anomalies on the atmosphere. It is well known that the frozen surfaces of the Arctic Ocean, Siberia, Alaska, Canada, Greenland, and the Beaufort Sea are the source regions of cold air masses during the Northern Hemispheric winter due to a strong radiative deficit (Curry 1983). Sea ice is partly responsible for the radiative deficit through 1) its high albedo, which influences the radiation balance, and 2) its important role as an insulator between the atmosphere and the ocean. During the light ice winter, stronger latent and sensible heating from open water to the atmosphere lead to negative SLP anomalies (Fig. 4b).

b. Air temperature anomalies

1) Surface air temperature

The surface air temperature (SAT) composite fields in Fig. 5 show that during the heavy ice winters, there are positive temperature anomalies over most of the area from Greenland to western Iceland, with negative temperature anomalies to the east (Fig. 5a). Obviously, the atmospheric circulation determines the spatial distribution of the large-scale SAT anomalies. However, over the Greenland Sea, the Fram Strait, and the Barents Sea, the more “local scale” negative anomalies of SAT increase sharply in amplitude, with maximum negative anomalies below −10°C. The centers of the negative SAT anomalies are located to the north of the anomalous sea ice extent compared to Fig. 3a. A displacement relative to the ice edge is reasonable because the anomalous sea ice boundary creates a sharp contrast with the open water, and strong advection “mixing” reduces the SAT contrast between the marginal sea ice zone and the open water. In this respect, SAT anomalies reflect the feedback of sea ice on the atmosphere. It should be pointed out that, as suggested by Deser et al. (2000), changes shown in Fig. 5a may be exaggerated because of the crude binary representation of the marginal ice zone in the NCEP–NCAR reanalysis. During the light sea ice winters an opposite scenario is evident (Fig. 5b). Differences in mean air temperature between the heavy and light ice winters are significant over the marginal sea ice zone (≥0.001 confidence level). To the south of the sea ice zone, significant air temperature differences again are indicative of strong horizontal mixing in the boundary layer.

2) Lower-tropospheric air temperature

To explore possible effects of sea ice variation on air temperature in the lower troposphere, we first calculate air temperature differences in the lower troposphere by subtracting temperature at higher pressures from those at lower pressures and then calculate differences between the heavy and light ice winters. Such calculations provide information on the combined impacts of the boundary layer radiative deficit and air temperature advection on the lapse rate of temperature. Air temperature differences are negative in the lower troposphere (Figs. 6a,b). The shaded area is mainly distributed along the Greenland and Barents Seas, indicating that the cooling effect of sea ice is stronger in the lower boundary layer than in the higher boundary layer. The shaded area shown in Fig. 6b contracts to the central Nordic and the Barents Seas, showing that the influences of sea ice on the atmosphere tend to be weaker at higher altitudes. According to a significance test, the altitude influence seems to extend upward to 700 hPa.

3) Stability of the planetary boundary layer

To detect possible influences from anomalous sea ice extent on the planetary boundary layer, we evaluate the vertical potential pseudo-equivalent temperature (θse) gradient (in degrees kelvin per kilometer) between 850 and 700 hPa,
i1520-0493-132-7-1868-eq1
where θse850 and θse700 are the θse for 850 and 700 hPa, respectively, and Z850 and Z700 geopotential heights for 850 and 700 hPa, respectively. In order to include the role of water vapor, we utilize θse rather than θ. Potential pseudo-equivalent temperature θse is defined as
i1520-0493-132-7-1868-eq2
where p, t, and q represent pressure, temperature, and specific humidity at each standard isobaric layer; td denotes the dewpoint temperature, which is obtained by an iteration method; and γ has been used to measure the development of the convective boundary layer (Gamo 1996; Shinoda and Utsugi 2001). Values of γ are positive. The larger the value of γ, the more stable the planetary boundary layer is.

Values of γ over ice-covered surfaces are greater than those over the open water (Fig. 7a). The maximum values of γ appear over northern Greenland and the Arctic basin, particularly in northern Greenland where the value of γ is close to the dry adiabatic lapse rate. This means that the layer between 850 and 700 hPa has a stronger stratification over sea ice than over the open water, reflecting the low-level diabatic cooling associated with sea ice if we assume it outweighs the effects of cold or warm air advection. When composited for the heavy ice winters, the anomalies of γ are positive over the whole Nordic and Barents Seas, with maximum positive anomalies over the eastern Greenland Sea and southwest to Svalbard, close to the marginal sea ice zone (Fig. 7b). During the light ice winters an opposite pattern can be observed (Fig. 7c). We hypothesize that the differences in Figs. 7b and 7c are manifestations of the direct influences from the underlying surface (sea ice and open water), modified by the effect of air temperature advection.

To quantify the contribution of air temperature advection to the vertical gradient of θse, the mean air temperature advection was calculated between 850 and 700 hPa. A linear regression analysis was then applied to analyze the contribution of air temperature advection to the variance of the vertical θse gradient.

Variations of air temperature advection can account for more than 30% of the total variance of the vertical θse gradient to the south of 80°N, which includes the Norwegian Sea, the Labrador Sea and Davis Strait, and northwestern Greenland (Fig. 8). Over the Greenland Sea, the Barents Sea, and most of Greenland, air temperature advection accounts for less than 10% of the total variance of the vertical θse gradients, implying that the effect of air temperature advection on the vertical θse gradient is nearly negligible. Consequently, diabatic processes such as the net radiative cooling and sensible heat flux are evidently the major factors influencing the vertical θse gradient over the sea ice zone. The diabatic effects of sea ice on the atmosphere are consistent with the spatial pattern in Fig. 8.

4) Low-level inversions

Inversions frequently form over high latitudes and the Arctic Ocean during the cold season, mainly because of strong radiative deficits at the surface. The inversions are often divided into subcategories, such as surface-based inversion and elevated inversion (Serreze et al. 1992). Inversions are also distinguished according to their formation mechanisms: radiative inversion, subsidence inversion, advection inversion, etc. Busch et al. (1982) identified radiation inversions as the most common type, comprising up to 85% of Arctic soundings from January to April.

To detect the lower-tropospheric temperature inversions over the Greenland and Barents Seas, observations with high vertical resolution are very important (Kahl 1990). However, because of the lack of high-resolution observations in the vertical direction over the Greenland and Barents Seas [no rawinsonde in situ data over the two sea regions; Historical Arctic Rawinsonde Archive consists of all available Arctic rawinsonde data from fixed-position (mostly land) stations north of 65°N], we must rely on the NCEP–NCAR reanalysis to address the variability of lower-tropospheric temperature inversion. This reliance introduces uncertainties and error due to the reanalysis model's parameterizations, especially with regard to diabatic processes. It is difficult to evaluate the NCEP–NCAR reanalysis dataset's uncertainties because no high-resolution observations can be used to make a comparison over the Greenland and Barents Seas. Moreover, Arctic temperature profiles often exhibit a complicated vertical structure that is unlikely to be completely resolved by the reanalysis output because of its coarse resolution. Subject to these caveats, we use the reanalysis output to approximately detect inversions by following the method developed by Kahl (1990). For detailed information, see Serreze et al. (1992). We calculated air temperature differences at each grid point (high level minus low level) and then constructed a latitude–pressure section of composite air temperature differences along 15°W for the heavy and light ice winters, respectively. Figure 9a shows that the isotherm of the differences is nearly vertical, with even a slight upward increase of the difference below 850 hPa over the area between 66° and 69°N, the location of the ice edge, with more frequent inversions to its north. During the light ice winters, the retreat of inversions in the difference isotherm is even stronger but is located much farther north, near 80°N (Fig. 9b). The top of the inversion reaches upward at least to 850 hPa at 80°–84°N. In addition, Fig. 9b shows an elevated inversion around 80°N, possibly because the upward fluxes of the latent and the sensible heat from the open water compensate the radiative deficit in the low layer, leading to the elevated inversion.

Significant changes in air temperature differences are mainly confined to the region below 700 hPa from 66° to 84°N, that is, within the planetary boundary layer. Moreover, the extent of the shaded area (indicating statistical significance) is horizontally consistent with the changes in sea ice extent. The significant changes within the planetary boundary layer are consistent with insulation role and high albedo of sea ice, which produces a discernible effect that is apparently greater than the thermal advection.

c. Possible dynamical roles

Sea ice extent as well as sea ice roughness can produce dynamic influences on the boundary layer process. Compared to winter mean divergence (Fig. 10a), the heavy ice winters cause the weakening of convergence and strengthening of divergence in the lower boundary layer over the Greenland and Barents Seas (Fig. 10b), consistent with positive SLP anomalies shown in Fig. 4. Simultaneously, enhanced convergence and weakened divergence occupy the high-level boundary layer (Fig. 10b). Consequently, distribution of divergence anomalies shown in Fig. 10b is in agreement with the simulation result by Honda et al. (1996). The weaker convergence anomalies between 850 and 700 hPa east of 5°E would be favorable for maintaining divergence anomalies in the low level of the boundary layer. Consequently, the anomalies of divergence could produce subcirculation anomalies in the lower troposphere. During the light ice winters (Fig. 10c), an enhanced convergence in the lower boundary layer is apparent (Fig. 10c) because of more-than-normal heat flux from open water to the atmosphere. In addition, the anomalies of divergence fields clearly show a barotropic structure in the troposphere west of 20°W and a baroclinic structure to its east. In fact, based on Ekman pumping theory, we can give a reasonable explanation for this phenomenon. According to the theory
wBhBζg
where wB is a vertical velocity at the top of the planetary boundary layer, hB is the thickness of the planetary boundary layer, and ζg is the geotrophic vorticity within the boundary layer. Because there are positive SLP anomalies corresponding to the heavy ice winters, ζg < 0, which corresponds to a divergence in horizontal direction and results in a sinking motion across the top of the boundary layer. Correspondingly, there is a horizontal convergence in the free atmosphere to compensate the sinking motion across the top of the boundary layer. The dynamic mechanism is shown in Fig. 11a. Conversely, during the light sea ice winters, there is more than normal open water in the target region, producing negative SLP anomalies due to strong sensible and latent heating from the open water to the atmosphere. Correspondingly, ζg > 0, thus, wB > 0 and the opposite situation develops (Fig. 11b).

4. Conclusions and discussion

Based on the analysis of Arctic sea ice concentration data and the NCEP–NCAR monthly mean reanalysis dataset, this study addresses possible feedbacks of interannual variation of winter sea ice extent in the Greenland and Barents Seas on the atmospheric boundary layer. The emphasis is on the vertical structure of the signals associated with sea ice. The results indicate that winter sea ice extent in the Greenland and Barents Seas shows interannual variations superimposed by a long-term trend. Prior to about 1970, there was generally more-than-normal sea ice in the Greenland and Barents Seas. Sea ice extent reached its maximum extent in the late 1960s. Since the late 1970s sea ice extent has decreased substantially.

The effect of winter (February–April average) sea ice on air temperature is significant in the Greenland–Barents Seas area. The effect seems to extend upward from the surface to 700 hPa. The results also indicate that the vertical gradient of potential temperature between 850 and 700 hPa is greater over the sea ice zone than that over open water, implying that the stability of the planetary boundary layer between 850 and 700 hPa increases over sea ice. Over sea ice, this increase is detectable despite the effect of air temperature advection. Consequently, diabatic processes and the insulation role of sea ice between the atmosphere and the ocean are major factors influencing the θse vertical gradient. Sea ice also reduces the exchange of sensible heat, latent heat, and momentum between the atmosphere and the ocean, and it also leads to enhanced net radiative cooling and stronger vertical gradients of θse over sea ice relative to open-water areas. Consequently, when there is more sea ice than normal in the Greenland and Barents Seas during winter, the planetary boundary layer is more stably stratified over the Nordic and Barents Seas. Over the Greenland and Barents Seas the baroclinicity of the atmosphere below 700 hPa increases because of the effects of both the underlying sea surface and the associated diabatic cooling.

Sea ice variations in the Greenland and Barents Seas appear to generate dynamic influences on the planetary boundary layer. The cooling associated with increased albedo and the insulation role of sea ice between the atmosphere and the ocean during the heavy ice winters can lead to a shallow surface high pressure near the surface. Horizontal and vertical variations of the effects of sea ice on temperature create the potential for dynamical influences on the atmospheric boundary layer through vertical motion anomalies induced by the pressure anomalies. On other hand, the surface high pressure and the radiative deficit in the low-level boundary layer provide favorable physical conditions for surface-based inversions over the sea ice zone. Consequently, the existence of the inversion may partly reflect a feedback of sea ice on the atmospheric boundary layer. Kahl's (1990) research supports this interpretation.

It is well known that the interaction processes between Arctic sea ice and the local atmosphere are very complicated. Because of the dominance of atmospheric forcing of sea ice, it is very difficult to detect feedbacks of the Arctic sea ice on the atmosphere. Meanwhile, the lack of observations in the lower troposphere over the Arctic sea ice and the limited knowledge of the interactions among climate systems point to the need for increased observational analyses to reveal the feedbacks of sea ice. Because we used a coarse-resolution dataset to investigate effects of sea ice on the atmospheric boundary layer, some uncertainties in analysis resulted. At the present, it is difficult to evaluate the reanalysis dataset's uncertainties because there is no credible high-resolution comparative data. Therefore, the preliminary results described here need to be confirmed by a carefully designed program and simulating with credible models in the future.

Acknowledgments

We thank two anonymous reviewers for helpful comments and constructive suggestions on an earlier version of the manuscript. This research has been supported by the International Arctic Research Center/Frontier Research System for Global Changes, University of Alaska, Fairbanks, and the National Nature Foundation of China (Grant 49905003: Arctic sea ice variations and its impacts on the east Asian monsoon). We thank the National Snow and Ice Data Center (NSIDC) for kindly providing the sea ice concentration data.

REFERENCES

  • Aagaard, K., , and E. C. Carmack, 1989: The role of sea ice and other freshwater in the Arctic circulation. J. Geophys. Res, 94 , 305311.

    • Search Google Scholar
    • Export Citation
  • Arfeuille, G., , L. A. Maysak, , and L-B. Tremblay, 2000: Simulation of the interannual variability of the wind-driven Arctic sea-ice cover during 1958–1998. Climate Dyn, 16 , 107121.

    • Search Google Scholar
    • Export Citation
  • Busch, N., , U. Ebel, , H. Kraus, , and E. Schaller, 1982: The structure of the subpolar inversion-capped ABL. Arch. Meteor. Geophys. Bioklimatol, 31A , 118.

    • Search Google Scholar
    • Export Citation
  • Curry, J., 1983: On the formation of polar continental air. J. Atmos. Sci, 40 , 22782292.

  • Deser, C., , and M. L. Blackmon, 1993: Surface climate variations over the North Atlantic Ocean during winter: 1900–1989. J. Climate, 6 , 17431754.

    • Search Google Scholar
    • Export Citation
  • Deser, C., , J. E. Walsh, , and M. S. Timlin, 2000: Arctic sea ice variability in the context of recent atmospheric circulation trends. J. Climate, 13 , 617633.

    • Search Google Scholar
    • Export Citation
  • Dickson, R. R., 1988: Great Salinity Anomaly in the northern North Atlantic 1968–1982. Progress in Oceanography, Vol. 20, Pergamon Press, 103–151.

    • Search Google Scholar
    • Export Citation
  • Dickson, R. R., and Coauthors, 2000: The Arctic Ocean response to the North Atlantic Oscillation. J. Climate, 13 , 26712696.

  • Dong, B-W., , and R. T. Sutton, 2002: Adjustment of the coupled ocean–atmosphere system to a sudden change in the thermohaline circulation. Geophys. Res. Lett.,29, 1728, doi:10.1029/ 2002GL015229.

    • Search Google Scholar
    • Export Citation
  • Gamo, M., 1996: Thickness of the dry convection and large-scale subsidence above deserts. Bound.-Layer Meteor, 79 , 265278.

  • Gao, D-Y., , and B-Y. Wu, 1998: A preliminary study on decadal oscillation and its oscillation source of sea–ice–air system in the Northern Hemisphere. Polar Meteor. Glaciol, 12 , 6878.

    • Search Google Scholar
    • Export Citation
  • Häkkinen, S., 1993: An Arctic source for the great salinity anomaly: A simulation of the Arctic ice–ocean system for 1955–1975. J. Geophys. Res, 98 , 1639716410.

    • Search Google Scholar
    • Export Citation
  • Herman, G. F., , and W. T. Johnson, 1978: The sensitivity of the general circulation to Arctic sea ice boundaries: A numerical experiment. Mon. Wea. Rev, 106 , 16491664.

    • Search Google Scholar
    • Export Citation
  • Hilmer, M., , and T. Jung, 2000: Evidence for a recent change in the link between the North Atlantic Oscillation and Arctic sea ice export. Geophys. Res. Lett, 27 , 989992.

    • Search Google Scholar
    • Export Citation
  • Honda, M., , K. Yamazaki, , Y. Tachibana, , and K. Takeuchi, 1996: Influence of Okhotsk sea-ice extent on atmospheric circulation. Geophys. Res. Lett, 23 , 35953598.

    • Search Google Scholar
    • Export Citation
  • Kahl, J. D., 1990: Characteristics of the low-level temperature inversion along the Alaska Arctic coast. Int. J. Climatol, 10 , 537548.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc, 77 , 437471.

  • Lazier, J., 1988: Temperature and salinity changes in the deep Labrador Sea 1962–1986. Deep-Sea Res, 35 , 12471253.

  • Lenderink, G., , and R. J. Haarsma, 1996: Modeling convective transitions in the presence of sea ice. J. Phys. Oceanogr, 26 , 14481467.

  • Mauritzen, C., , and S. Häkkinen, 1997: Influence of sea ice on the thermohaline circulation in the Arctic–North Atlantic Ocean. Geophys. Res. Lett, 24 , 32573260.

    • Search Google Scholar
    • Export Citation
  • Mysak, L. A., , and S. B. Power, 1992: Sea-ice anomalies in the western Arctic and Greenland–Iceland sea and their relation to an interdecaldal climate cycle. Climatol. Bull, 26 , 147176.

    • Search Google Scholar
    • Export Citation
  • Mysak, L. A., , and S. A. Venegas, 1998: Decadal climate oscillation in the Arctic: A new feedback loop for atmosphere–ice–ocean interactions. Geophys. Res. Lett, 25 , 36073610.

    • Search Google Scholar
    • Export Citation
  • Mysak, L. A., , D. K. Manak, , and R. F. Marsden, 1990: Sea-ice anomalies observed in the Greenland and Labrador Seas during 1901–1984 and their relation to an interdecadal Arctic climate cycle. Climate Dyn, 5 , 111133.

    • Search Google Scholar
    • Export Citation
  • Rigor, I. G., , J. M. Walsh, , and R. L. Colony, 2002: Response of sea ice to the Arctic Oscillation. J. Climate, 15 , 26482663.

  • Serreze, M. C., , J. D. Kahl, , and R. C. Schnell, 1992: Low-level inversions of the Eurasian Arctic and comparisons with Soviet drifting station data. J. Climate, 5 , 615629.

    • Search Google Scholar
    • Export Citation
  • Shinoda, M., , and H. Utsugi, 2001: Spring snow-disappearance timing and its possible influences on temperature fields over central Eurasia. J. Meteor. Soc. Japan, 79 , 3759.

    • Search Google Scholar
    • Export Citation
  • Skeie, P., 2000: Meridional flow variability over the Nordic Seas in the Arctic Oscillation framework. Geophys. Res. Lett, 27 , 25692572.

    • Search Google Scholar
    • Export Citation
  • Slonosky, V. C., , L. A. Mysak, , and J. Derome, 1997: Linking Arctic sea ice and atmospheric circulation anomalies on interannual and decadal timescales. Atmos.–Ocean, 35 , 333366.

    • Search Google Scholar
    • Export Citation
  • Tremblay, L-B., , and L. A. Mysak, 1997: Modeling sea ice as a granular material, including the dilatancy effect. J. Phys. Oceanogr, 27 , 23422360.

    • Search Google Scholar
    • Export Citation
  • Tremblay, L-B., , and L. A. Mysak, 1998: On the origin and evolution of sea ice anomalies in the Beaufort–Chukchi Sea. Climate Dyn, 14 , 451460.

    • Search Google Scholar
    • Export Citation
  • Vavrus, S., , and S. P. Harrison, 2003: The impact of sea-ice dynamics on the Arctic climate system. Climate Dyn, 20 , 741757.

  • Vinje, T., 2001a: Fram Strait ice fluxes and atmospheric circulation: 1950–2000. J. Climate, 14 , 35083517.

  • Vinje, T., 2001b: Anomalies and trends of sea ice extent and atmospheric circulation in the Nordic Seas during the period 1864–1998. J. Climate, 14 , 255267.

    • Search Google Scholar
    • Export Citation
  • Walsh, J. E., 1983: Role of sea ice in climate variability: Theories and evidence. Atmos.–Ocean, 21 , 229242.

  • Walsh, J. E., , and W. L. Chapman, 1990: Arctic contribution to upper-ocean variability in the North Atlantic. J. Climate, 3 , 14621473.

    • Search Google Scholar
    • Export Citation
  • Wang, J., , and M. Ikeda, 2000: Arctic Oscillation and Arctic sea-ice oscillation. Geophys. Res. Lett, 27 , 12871290.

  • Wu, B-Y., , and J. Wang, 2002: Possible impacts of winter Arctic Oscillation on the Siberian High, the east Asian winter monsoon and sea-ice. Adv. Atmos. Sci, 19 , 297320.

    • Search Google Scholar
    • Export Citation
  • Wu, B-Y., , D-Y. Gao, , and R-H. Huang, 1997: ENSO events and interannual variations of winter sea-ice in the Greenland, the Kara and the Barents Seas. Chin. Sci. Bull, 42 , 13821385.

    • Search Google Scholar
    • Export Citation
  • Wu, B-Y., , R-H. Huang, , and D-Y. Gao, 1999: Impact of variations of winter sea-ice extents in the Kara/Barents Seas on winter monsoon over east Asia. Acta Meteor. Sin, 13 , 141153.

    • Search Google Scholar
    • Export Citation
  • Wu, B-Y., , R-H. Huang, , and D-Y. Gao, 2001: Arctic sea ice bordering on the North Atlantic and interannual climate variations. Chin. Sci. Bull, 46 , 162165.

    • Search Google Scholar
    • Export Citation
  • Wu, B-Y., , R-H. Huang, , and D-Y. Gao, 2002: Numerical simulations on effects of variation of sea ice thickness and extent on atmospheric circulation. Acta Meteor. Sin, 16 , 150164.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Variations of winter (Feb–Apr) mean sea ice extent in the Greenland and Barents Seas. The dashed lines stand for climatological mean and standard deviations for the period 1953–95. Unit is 106 km2

Citation: Monthly Weather Review 132, 7; 10.1175/1520-0493(2004)132<1868:PFOWSI>2.0.CO;2

Fig. 2.
Fig. 2.

Winter surface wind composite (10 m): (a) the heavy ice winters and (b) the light ice winters. The gray and dark areas denote that the differences in mean northerly winds between the heavy and light ice winters (heavy minus light) exceed the 0.05 and 0.01 confidence levels, respectively. Unit is m s−1

Citation: Monthly Weather Review 132, 7; 10.1175/1520-0493(2004)132<1868:PFOWSI>2.0.CO;2

Fig. 3.
Fig. 3.

Sea ice extent composite for (a) the heavy ice winters and (b) the light ice winters. The solid curve denotes the location of sea ice edge (sea ice concentration = 15%)

Citation: Monthly Weather Review 132, 7; 10.1175/1520-0493(2004)132<1868:PFOWSI>2.0.CO;2

Fig. 4.
Fig. 4.

Same as Fig. 2 except for SLP anomalies (hPa)

Citation: Monthly Weather Review 132, 7; 10.1175/1520-0493(2004)132<1868:PFOWSI>2.0.CO;2

Fig. 5.
Fig. 5.

Same as Fig. 2 except for surface air temperature anomalies (°C)

Citation: Monthly Weather Review 132, 7; 10.1175/1520-0493(2004)132<1868:PFOWSI>2.0.CO;2

Fig. 6.
Fig. 6.

Differences in mean air temperature differences (°C; low level minus high level) between the heavy and light ice winters: (a) 925 minus 850 hPa, and (b) 850 minus 700 hPa. The gray and dark gray areas denote that differences between the heavy and light ice winters exceed the 0.01 and 0.001 confidence levels, respectively

Citation: Monthly Weather Review 132, 7; 10.1175/1520-0493(2004)132<1868:PFOWSI>2.0.CO;2

Fig. 7.
Fig. 7.

(a) Winter mean vertical θse gradient between 850 and 700 hPa (1958–99), (b) vertical θse gradient anomaly composite for the heavy ice winters, and (c) same as (b) but for the light ice winters. Unit is K km−1. The meaning for shading area is the same as in Fig. 2

Citation: Monthly Weather Review 132, 7; 10.1175/1520-0493(2004)132<1868:PFOWSI>2.0.CO;2

Fig. 8.
Fig. 8.

Variance fraction contribution of winter mean temperature advection between 850 and 700 hPa to the vertical θse gradient. The shading area denotes variance contributions of winter mean temperature advection ≥30%

Citation: Monthly Weather Review 132, 7; 10.1175/1520-0493(2004)132<1868:PFOWSI>2.0.CO;2

Fig. 9.
Fig. 9.

Latitude–pressure section of air temperature difference composite [°C; high minus low level, T(i, j + 1) − T(i, j), j = 1, …, 6, corresponding to 1000, 925, 850, 700, 600, 500 hPa, respectively; i = 1, … , 13, latitude points] along 15°W: (a) the heavy ice winters and (b) the light ice winters. The meaning for the shaded area is the same as Fig. 2

Citation: Monthly Weather Review 132, 7; 10.1175/1520-0493(2004)132<1868:PFOWSI>2.0.CO;2

Fig. 10.
Fig. 10.

The longitude–pressure section along 75°N: (a) winter mean divergence (1958–99), (b) divergence anomaly composite for the heavy ice winters, and (c) same as (b) except for the light ice winters. The gray area represents negative anomalies, and unit is 10−7 s−1

Citation: Monthly Weather Review 132, 7; 10.1175/1520-0493(2004)132<1868:PFOWSI>2.0.CO;2

Fig. 11.
Fig. 11.

Schematic diagram of the dynamic feedback of the (a) heavy and (b) light sea ice conditions on the local atmosphere

Citation: Monthly Weather Review 132, 7; 10.1175/1520-0493(2004)132<1868:PFOWSI>2.0.CO;2

Save