The relationships between the wind fields of the East Asia winter monsoon (EAWM), the Siberian High (SH), and the Arctic Oscillation (AO) were investigated using reanalysis products. The winter anomalies of the wind fields were systematically examined from the Bering Sea, the Sea of Okhotsk (SoO), the Sea of Japan (SOJ), the East China Sea (ECS), and all the way to the South China Sea (SCS). The sea-level pressure (SLP) difference between the SH and the Aleutian Low (AL) determines the intensity of the EAWM.Wind field anomalies are controlled directly by the SH and indirectly by the AO that has significant impacts on the SH and AL. It is found that +SH enhances the EAWM, while +AO reduces the intensity of the EAWM by reducing the SLP difference (gradient) between the SH and AL; vice versa for the –SH and –AO, respectively. The surface air temperature (SAT) anomalies caused by the +SH result in a significant cooling in the downstream regions and a warming in the upstream regions; vice versa for the negative phase of the SH. The +AO produces a large warming in northern Eurasian and a cooling in the Bering Sea. Furthermore, using a Coupled Ice-Ocean Model (CIOM), it is found that the EAWM can produce a downwelling and dense water formation along the Siberian coast in the western Bering Sea, and also a significant surface-to-bottom convection over the Bering shelf, forming the winter shelf water, which can survive the summer as the so-called cold pool. The cold pool in the Bering Sea has significant impacts on marine ecosystems and habitat including fisheries, which has much implication to other marginal seas of East Asia.

Introduction

The East Asian winter monsoon (EAWM) system is one of the most active components of the global climate system. This EAWM is often associated with cold surges in the region, extending from the Bering Sea all the way to the East China Sea (ECS) and possibly to the South China Sea (SCS). Since surface wind, surface air temperature (SAT), and precipitation anomalies associated with EAWM interannual variability are the major physical parameters/forcings to many other subsystems such as ocean circulation, upwelling/downwelling, deep convection, and marine ecosystems, understanding these regional anomalies associated with the EAWM is very important to understanding the regional responses of ocean dynamics, thermodynamics and other subsystems to the hemispheric climate changes.

Each winter, the sea-level pressure (SLP) difference (gradient) between the Siberian High (SH) in East Asia and the Aleutian Low (AL) to its east to a great extent determines the intensity of the EAWM. Thus, the most prominent seasonal surface feature of the EAWM is characterized by strong northwesterlies along the east flank of the SH and the East Asia coast (Chen et al., 2000), possibly extending to the South China Sea (SCS) where northeasterlies prevail. The year-to-year change of the SH and the AL would directly result in the interannual variability of the wind field, SAT and precipitation of the downstream regions: the Bering Sea, the Sea of Okhotsk (SoO), Sea of Japan (SOJ), East China Sea (ECS), and SCS. Nevertheless, the key question is “what controls the year-to-year change of the SH and the AL?”

Wu and Huang (1999) and Gong et al. (2001) investigated the connection between the Arctic Oscillation (AO) and the EAWM, and pointed out that the AO influences the EAWM through the impacts on the winter SH. Although the winter AO may influence the SH, the AO accounts for only 13.0% of the variance in the SH. As two independent action centers in the boreal winter, the AO and the SH play a relatively independent role in influencing climate variability over East Asia. Wu and Wang (2002a, 2002b) further investigated different roles of the winter AO and the SH in influencing the EAWM.

The northern and northwestern Pacific Ocean, including the Bering Sea and the Gulf of Alaska, is among the most productive marine ecosystems in the world, as evidenced by large populations of marine and freshwater salmon, fish, birds, and mammals. This productivity is critical not only to the U.S. economy, since fish and shellfish from these regions constitute more than 10% of the world's and about 52% of the U.S. seafood harvest, but also to the economy of the surrounding countries. There are also rich fishery resources in the SoO, and other semi-enclosed marginal seas such as the Bering Sea, SoO, SOJ, ECS, to SCS. Their ecosystem's health and management are the key issues for the surrounding countries, since human health and activities are closely related to the health of the marine ecosystems. Thus, the investigation of the atmospheric and oceanic physical forcings to the marine ecosystems will further help not only to understand the ecosystems per se via a bottom-up effect, but also to make correct and timely decisions and policy for the management of marine resources.

Motivated by previous studies, this study aims to revisit the investigation using data from 1958 to 2009. Furthermore, we intend to investigate that the EAWM can directly cause downwelling and deep convection in the western Bering Sea, shedding light on the possible downwelling and deep convection in other downstream seas.

Data and Methods

The dataset used here, including monthly SLP, SAT (2-m), 500-hPa geopotential height for 1958–2009, was taken from the National Center for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) reanalysis dataset (Kalnay et al., 1996). The AO index is taken from NOAA ESSL. All the data were averaged over three winter months (December, January, and February). Composite analysis along with Student t-test was used to examine the significant impacts.

The model used in this study is the Coupled Ice-Ocean Model (CIOM) (Wang et al., 2002, 2005, 2009). The detailed description of development of the CIOM should refer to Yao et al. (2000) and Wang et al. (2002, 2005), which was applied to the pan-Arctic Ocean (Wang et al., 2004, 2005; Wu et al., 2004; Long et al., 2012), the Beaufort Sea (Wang et al., 2003a, 2008), and the Bering Sea (Wang et al. 2009; Hu and Wang 2010). The ocean model used is the Princeton Ocean Model (POM) (Blumberg and Mellor, 1987; Mellor, 2004), and the ice model used is a full thermodynamic and dynamics model (Hibler, 1979, 1980) that prognostically simulates sea-ice thickness, sea ice concentration (SIC), ice edge, ice velocity, and heat and salt flux through sea ice into the ocean. The configuration of this model is the same as Wang et al. (2009).

Diagnosis of East Asian Winter Monsoon, Siberian High, and Arctic Oscillation

The EAWM index is defined as the normalized SLP difference between 110°E and 160°E within 20–50°N. The AO index is defined by the first EOF mode of SLP north of 20°N in the northern hemisphere. The SH intensity index is defined as the regionally averaged SLP (40–60°N, 80–120°E) following Wu and Wang (2002a). Figure 1 shows the time series of these three indices for the 52-year study period 1958–2009 and their 0–lag correlations. SH has the largest positive correlation with EAWM (0.68), since its intensity directly determines the strength of EAWM. That is, the stronger the SH, the stronger the EAWM (i.e. northerlies), and vice versa. The AO has an opposite correlation (−0.28) or phase with the SH. This indicates that the positive (negative) phase of AO will reduce (increases) the intensity of SH, leading to a weakening (strengthening) of EAWM. This can be reflected by the out-of-phase correlation (−0.17) between the AO and EAWM. Wu et al. (2006) also show that their relationship is poor.

To further investigate the relationship of the EAWM in the northern part of the study region, the north EAWM (NEAWM) index is defined as the same as EAWM, except in the region of 50–70°N. This region overlaps the SH index (40–60°N) to the north of 70°N, or the SH's northern domain (i.e. upstream). Thus, the correlation between the SH and NEAWM reduces to 0.57; nevertheless, the correlation between the AO and NEAWM increases to −0.57. This indicates that both the AO and SH have equal direct impacts on the NEAWM, although the SH dominates over the AO on controlling the EAWM in its downstream region (AO has indirect impacts on the EAWM by changing the intensity of SH [Wu and Wang, 2002a, 2002b]). Using one standard deviation as a criterion, the winters/years with both positive and negative phases of the SH and AO were selected (Table 1) for composite analysis.

Climatology of sea-level pressure and wind field

First, the climatology of the winter SLP and surface wind field was constructed for the 52-year period (Figure 2). The remarkable features of the SLP climatology are the SH (>1035 hPa) and AL (<1000 hPa). The climatological surface wind field mimics the SLP pattern. There is a strong cyclonic circulation associated with the AL in the northern North Pacific, and a strong anti-cyclonic circulation associated with the SH in the Eurasian Continent. The most distinguishable feature is the EAWM prevailing from the Bering Sea all the way to the SCS, driven by the SLP difference between the SH and AL. The northeasterlies prevail in the Bering Sea and SCS, while the northerlies or northwesterlies prevail in the SoO, SoJ, and northern ECS. In these regions close to the SH, cold air outbreaks or cold surges frequently occur in the winter season, depending on the intensity of the EAWM or the intensity of the SLP difference between the SH and AL. The winter monsoon in the Indian Ocean and Bay of Bengal is also the significant winter feature. Another important feature is that the westerlies (jet) aloft in the mid-latitude (∼40°N) can be seen clearly in the northern North Pacific Ocean, while the surface signature over the Eurasian Continent is not obvious due to orographic effects, such as from the Himalayas. Similarly, the easterly (jet) aloft in the low-latitude (∼10°N) is also seen at the surface in the southern North Pacific Ocean, while the surface northeasterlies (EAWM) prevail in the SCS, Bay of Bengal, and India Ocean.

Influence of Siberian High

The composite anomaly between the +SH and –SH (Figure 3a) indicate the further strengthening of the SH and deepening of the AL during the positive phase of SH, leading to strengthening of the EAWM from the Bering Sea all the way to the northern SCS (Figure 3b). The regions with significant impacts over the 99% significance level (shaded area) indicate the broad influence by the SH. In particular, strong anomalous northeasterlies appear in the western Bering Sea, SoO, and northern SCS, while strong anomalous northerlies and northwesterlies appear in the SoJ and ECS, as well as in the Siberia and northern China. At the same time, the surface westerlies over the mid-latitude North Pacific Ocean are intensified. However, during the negative phase of SH, the reverse is true (Figures 3a, b). That is, both the SH (negative anomalous SLP) and AL (positive anomalous SLP) are weakened (Figure 3a), so is the EAWM. Thus, the anomalous wind vectors (directions) are reversed with the same magnitude and significance area (Figure 3b).

The consequences during the positive phase of SH are that significant cooling occurs downstream of SH in the Asian Continent with a center at 45°N, covering much of China and Japan, reaching Taiwan and Hainan in the northern SCS (Figure 3c). Cooling occurs from SoJ (∼1–1.5°C), ECS (0.5–1°C) to the SCS (∼0.5°C) with a maximum in the SoJ and northern China seas including Bohai Sea. Upstream of SH, a warming occurs in northern Russia extending to eastern Siberia and SoO. Warming occurs on the northeastern North Pacific side (west coast of America, Gulf of Alaska, Alaska, and eastern Bering Sea), while cooling occurs on the western Bering Sea and eastern Siberian.

Influence of Arctic Oscillation

The SLP difference between the +AO and –AO phases (Figure 4a) shows a positive anomaly in the northern North Pacific, leading to a weakening of the AL. More interestingly, the negative anomalies occupy the northwestern domain of the SH (leading to a weakening of the SH), while positive anomalies occupy the southeastern part of the SH (strengthening of the SH). This is why the correlation between the AO and SH is only −0.28 (Figure 1). The overall effects of the AO weakens the EAWM by reducing the SLP gradient between the SH and AL.

The anomalous wind field associated with +AO and –AO is that there are anomalous southwesterlies in the Bering Sea and the SoO, anomalous southeasterlies over the SoJ and ECS, and weak southerlies in the SCS. All these anomalous winds are unfavorabale to the climatological EAWM, leading to a significant reduction in the intensity of the EAWM.

As a result, during the positive AO phase, the entire northern Eurasian Continent is covered by a large positive SAT anomaly, i.e. warming (Figure 4b). In particular, in northern China and the Far East including SoJ and Japan, the warming can be as much as 2–5°C, and can cover the entire Yangzi River watershed and much of China, except a small area of cooling located in the northern Bay of Bengal. In the ECS and SCS, a weak warming (∼0.5°C), but not statistically significant, is detected. Interestingly, the entire Bering Sea experiences a significant cooling as large as −4°C due to the weakening of the AL. The weakened AL, similar to the effect of the La Nina event, hinders the advection of warm, moist North Pacific (current, the extension of warm Kuroshio) water into the Gulf of Alaska and the Bering Sea, leading to the cooling.

Modeling of downwelling caused by the East Asian Winter Monsoon in the Western Bering Sea

A typical downwelling caused by the EAWM is demonstrated using the CIOM. The study region is the same as in Wang et al. (2009), covering the entire Bering Sea. Wang et al. (2009) demonstrated that the summer southwesterly monsoon in the Bering Sea produces a distinguishable, narrow, productive upwelling band along the Siberian coast; i.e. in the western Bering Sea using the same CIOM, which was confirmed by the satellite-measured SST and chlorophyll-a. The CIOM was driven by daily climatological NCEP reanalysis atmospheric forcing (see Figure 2b for wind field for example) over a seasonal cycle for five years. The results from the fifth year were used for this analysis.

Figure 5 shows the CIOM-simulated January surface ocean circulation under the prevailing northeasterlies over the Bering Sea (Figure 2b). The surface current on the Bering Shelf is westward due to Ekman drift. The significantly distinguishable reversal of the Anadyr Current to the south along the Siberian coast was detected in the winter. This feature implies that possible downwelling occurs during winter due to deep convection caused by cooling and salt injection from ice formation.

Figure 6 shows the CIOM-simulated salinity and temperature distribution along the 62.5°N-transect (see Figure 5 for the location) from January to March. A remarkable feature in January is that the dense (high salinity and cold temperature) water formation occurs along the Siberian coast, descending all the way to the bottom (Wang et al., 2003b). At the same time, vertical convection also occurs on the Bering Shelf, forming the winter shelf water or the so-called “cold pool” (Hu and Wang, 2010). In February to March, although the dense water formation along the Siberian coast weakens, the vertical convection on the shelf increases, leading to an accumulation of saline and cold water. This cold pool has a significant impact on fisheries habitat, since species such as Arctic Cod prefer temperatures less than 2°C, while Pollock prefer temperatures higher than 2°C, staying away from the cold pool.

Discussion and Conclusions

We examine the impacts of the SH and AO on the EAWM using a NCEP reanalysis dataset from 1958–2009. Furthermore, we investigate an example of the impacts of EAWM on the downwelling and deep convection in the western Bering Sea and its shelf using the CIOM. Based on the above investigation, the following conclusions can be drawn:

  1. The SH has a direct positive correlation with the EAWM, i.e. the stronger the positive SH, the more severe the EAWM, and vice versa. The intensified EAWM causes significant negative SAT anomalies (the severe ones sometimes are called cold surges or cold air outbreaks) over its downstream regions all the way to the South China Sea with a significant area in northern China and Japan (with a center along the 40°N). Conversely, the positive SH causes strong positive SAT anomalies all the way to the Arctic Ocean including a strong warming of as much as 2.5°C in the Sea of Okhotsk. At the same time, the western Bering Sea including Eastern Siberia experiences a strong negative anomalous SAT, while the eastern Bering Sea including Alaska experiences positive SAT anomalies.

  2. The AO has an indirect, negative (out of phase) correlation to the EAWM by increasing/reducing the intensity of the SH and the AL. The positive AO exerts a positive SLP anomaly to the AL region, reducing the strength of the AL, while a negative AO exerts a negative SLP anomaly and a positive SLP anomaly to the northwestern and southeastern parts of the SH, respectively, with an overall reduction in SH intensity. Therefore, the positive (negative) AO reduces (increases) the SLP gradient between the SH and the AL, leading to weakening (strengthening) of the EAWM. The SAT anomalies caused by the +AO result in the significant warming in the Eurasian Continent and significant cooling in the Bering Sea (this scenario can be seen in the early 1990s). Conversely, the –AO can cause cooling in the Eurasian Continent and warming in the Bering Sea (this scenario can be seen during the period from the late 1990s to the early 2000s).

  3. In the western Bering Sea, the EAWM drives the Anadyr Current (which is northward flow in other seasons) toward the south and the shelf water toward the west due to Ekman drift. A prevailing downwelling along the Siberain coast (i.e. in the western Bering Sea) is induced by the strong northeasterly EAWM, accompanied by a cold, dense saline-rich water formation. The vertical convection is reproduced on the broad Bering shelf, forming the winter shelf water with similar properties‥ This locally-formed winter water is the source of the summer cold pool on the bottom of the Bering shelf (Hu and Wang 2010). The cold pool in the Bering Sea favors species such as Arctic Cod, but is a barrier to Pollock that prefer water temperatures warmer than 2°C.

The findings from this study have significant implications, both physical and biological, in these semi-enclosed, marginal seas of East Asia, which are under the forcing of the EAWM. The year-to-year variability of both the SH intensity (Wu and Wang 2002a, 2002b) and the phase of the AO (Wang and Ikeda, 2000) significantly changes the intensity and timing/onset of the EAWM and SAT distribution. The impacts of these changes can be reflected by not only climate changes (terrestrial changes) and economy (agriculture, transportation, etc.) per se, but more importantly also by the marine physical and biological responses. The winds and SAT associated with the EAWM are direct drivers to these marine subsystems. The example of the downwelling and deep convection (dense water formation) in the western Bering Sea can be generally applied to all of these marginal seas of East Asia. For example, under forcing of the EAWM with its cooling, the downwelling and deep convection (or deepening of the mixed layer) are expected along the western coasts in the Sea of Okhotsk, the Sea of Japan, northern China seas, the East China Sea, and the South China Sea (along Vietnam coast), including east coasts of Hainan and Taiwan. The deep convection on the shelves of these seas would produce local winter water masses such as in the Sea of Okhotsk and ECS, which can be observed in the summer as the cold water masses, such as the ECS cold water mass. The winter upwelling is expected along the west coasts of Japan, Taiwan, and Hainan.

The locally-formed winter cold water masses accompanied by the downwelling have significant effects on fisheries habitat throughout the year. The downwelling enhances the deposit of nutrients, pelagic phytoplankton, and zooplankton that enrichs the benthic biomass, which is the major supplier to the upper eutrophic layer in the following summer. Some species prefer cold water masses, while other species avoid these cold water masses, depending on other environmental and physical conditions in each sea. More importantly, these downwelling regions in winter, under forcing of the East Asia Summer Monsoon (EASM), shift to upwelling regions in summer under forcing of the EASM, such as in the western Bering Sea (Wang et al., 2009). These upwellings bring nutrients in the subsurface and benthos up to the surface eutrophic zone (Mizobata et al., 2006, 2008; Wang et al., 2009), feeding phytoplankton that are fed to zooplankton and fisheries. Therefore, the impacts of the EAWM and EASM on marine physical and biological components, including fisheries, are important, and further research is needed for sustainable development in these marginal seas of East Asia.

Acknowledgements

We thank supports from the NOAA RUSALCA (Joint Russian-American Long-Term Census of the Arctic) International Polar Year modeling project. We want to thank two anonymous reviewers for their very constructive comments, which helped significantly improve this paper. Thanks also go to Cathy Darnell of NOAA GLERL for editing this paper. This is also GLERL Contribution No. 1617.

References

Blumberg, A. F. and Mellor, G. L.
1987
. “
A description of 3-D coastal ocean circulation model
”. In
Coastal and Estuarine Sciences 4: 3-D Coastal Ocean Models
, Edited by: Heaps, N. S.
1
16
.
Washington DC
:
American Geophysical Union
.
Chen, W., Graf, H.-F. and Huang, R.-H.
2000
.
The interannual variability of East Asian Winter Monsoon and its relation to the summer monsoon
.
Adv. Atmos. Sci.
,
17
(
1
):
48
60
.
Gong, D. Y., Wang, S. W. and Zhu, J. H.
2001
.
East Asian winter monsoon and Arctic Oscillation
.
Geophy. Res. Lett.
,
28
(
10
):
2073
2076
.
Hibler, W. D. III.
1979
.
A dynamic and thermodynamic sea ice model
.
J. Phys. Oceanogr.
,
9
15,959-15,969
Hibler, W. D. III.
1980
.
Modeling a variable thickness sea ice cover
.
Mon. Wea Rev.
,
108
:
1943
1973
.
Hu, H. and Wang, J.
2010
.
Modeling effects of tidal and wave mixing on circulation and thermohaline structures in the Bering Sea: Process studies, J
.
Geophys. Res.
,
115
:
C01006
doi:
Kalnay, E., Kanamitsu, M., Kistler, R., Collins, W., Deaven, D., Gandin, L., Iredell, , Saha, S., White, G., Woollen, J., Zhu, Y., Leetmaa, A., Reynolds, R., Chelliah, M., Ebisuzaki, W., Higgins, W., Janowiak, J., Mo, K. C., Ropelewski, C., Wang, J., Jenne, R. and Joseph, D.
1996
.
The NCEP/NCAR 40-year Reanalysis Project
.
Bull. Am. Meteorol. Soc.
,
77
:
437
471
.
Long, Z., Perrie, W., Tang, C. L., Dunlap, E. and Wang, J.
2012
.
Simulated interannual variations of freshwater content and sea surface height in the Beaufont Sea
.
J. Clim.
,
doi:10.1175/2011JC114121.1
Mellor, G. L.
2004
.
Users guide for a 3-D, primitive equation, numerical ocean model
,
NJ
:
Atmospheric and Oceanic Sciences Program, Princeton Univ.
.
Mizobata, K., Wang, J. and Saitoh, S.
2006
.
Eddy-induced cross-slope exchange maintaining summer high productivity of the Bering Sea shelf break
.
J. Geophys. Res.
,
111
:
C10017
doi:
Mizobata, K., Saitoh, S. and Wang, J.
2008
.
Interannual variability of summer biochemical enhancement in relation to the mesoscale eddy at the shelf break in the vicinity of the Pribilof Islands, Bering Sea
.
Deep Sea Res.
, doi:
Wang, J. and Ikeda, M.
2000
.
Arctic Oscillation and Arctic Sea-Ice Oscillation
.
Geophys. Res. Lett.
,
27
(
9
):
1287
1290
.
Wang, J., Liu, Q. and Jin, M.
2002
.
A user's guide for a Coupled Ice-Ocean Model (CIOM) in the Pan-Arctic and North Atlantic Oceans. International Arctic Research Center-Frontier Research System for Global Change
.
Tech. Rep. 02-01 (available at Jia.Wang@noaa.gov)
Wang, J., Kwok, R., Saucier, F. J., Hutchings, J., Ikeda, M., Hibler, W., Haapala, J., Coon, M. D., Meier, H. E.M., Eicken, H., Tanaka, N., Prentki, R. and Johnson, W.
2003a
.
Working towards improved small-scale sea ice and ocean modeling in the Arctic seas
.
EOS, AGU
,
84
(
34
):
329
330
.
325
Wang, J., Ikeda, M. and Saucier, F.
2003b
.
A theoretical, two-layer, reduced-gravity model for descending dense water flow on continental slopes
.
J. Geophys. Res.
,
108
(
C5
):
3161
doi:
Wang, J., Wu, B., Tang, C., Walsh, J. E. and Ikeda, M.
2004
.
Seesaw structure of subsurface temperature anomalies between the Barents Sea and the Labrador Sea
.
Geophys. Res. Lett.
,
31
:
L19301
doi:
Wang, J., Liu, Q., Jin, M., Ikeda, M. and Saucier, F. J.
2005
.
A coupled ice-ocean model in the pan-Arctic and the northern North Atlantic Ocean: Simulation of seasonal cycles
.
J. Oceanogr.
,
61
:
213
233
.
Wang, J., Mizobata, K., Hu, H., Jin, M., Zhang, S., Johnson, W. and Shimada, K.
2008
.
Modeling seasonal variations of ocean and sea ice circulation in the Beaufort and Chukchi Seas: A model-data fusion study
.
Chinese Journal of Polar Science
,
19
(
2
):
168
184
.
Wang, J., Hu, H., Mizobata, K. and Saitoh, S.
2009
.
Seasonal variations of sea ice and ocean circulation in the Bering Sea: A model-data fusion study
.
J. Geophys. Res.
,
114
:
C02011
doi:
Wu, B and Huang, R.
1999
.
Effects of the extremes in the North Atlantic Oscillation on the East Asia winter monsoon
.
Chinese J. Atmos. Sci.
,
23
(
3
):
226
236
.
Wu, B. and Wang, J.
2002a
.
Winter Arctic oscillation, Siberian High and the East Asia winter monsoon
.
Geophys. Res. Lett.
,
29
(
19
):
1897
1900
.
Wu, B. and Wang, J.
2002b
.
Possible impacts of winter Arctic Oscillation on Siberian High and the East Asia winter monsoon
.
Advances in Atmospheric Sciences
,
19
:
297
320
.
Wu, B., Wang, J. and Zhang, R.
2004
.
Effects of intraseasonal variations of the Arctic Oscillation on the Barents Sea
.
Polar Meteorolo. Glaciol.
,
18
:
82
95
.
Wu, B, Zhang, R, Ding, Y. and D’Arrigo, R.
2006
.
Distinct modes of the East Asian winter monsoon, Mon
.
Wea. Rev.
,
134
:
2165
2179
.
Yao, T., Tang, C. L. and Peterson, I. K.
2000
.
Modeling the seasonal variation of sea ice in the Labrador Sea with a coupled multi-category ice model and the Princeton Ocean Model
.
J. Geophys. Res.
,
105
:
1153
1165
.