T H E F U T U R E O F O C E A N O G R A P H Y
NUMERICAL SIMULATION OF ARCTIC SEA-ICE AND OCEAN CIRCULATION
By David M. Holland
C U R R E N T L Y THERE IS c o n s i d e r a b l e i n - t e r e s t
in improving our understanding of the interactions between the compo- nents of the earth's climate system. The polar regions, which form an integral yet distinct part of this system, are be- lieved to be particularly sensitive to an- thropogenically induced global warm- ing and therefore are worthy of special study. The polar climate s u b s y s t e m consists of the atmosphere, the oceans, and the sea ice. The presence of sea ice on the ocean surface drastically alters the interaction between the atmosphere and the ocean at high latitudes.
Our u n d e r s t a n d i n g of the interac- tions in the polar r e g i o n s has been hampered by a shortage of basin-wide observations due to the harsh climate.
Those that do exist up to the late 1980s are reviewed by Carmack (1990). Nu- merical modeling thus provides an im- portant tool for investigating the com- plex mechanisms and interactions that control polar climate. The northern and southern polar regions are very differ- ent from one another. The Arctic con- sists of an ocean surrounded by conti- nental land masses, whereas the Antarctic has a continent surrounded by a h i g h - l a t i t u d e ocean. This study fo- cuses on a numerical simulation of the Arctic using the r e c e n t l y d e v e l o p e d Oberhuber (1993) ocean model.
The Arctic Ocean is essentially con- tained within a closed basin. It has the largest continental shelves in the world oceans. These shallow and broad mar- gins cover about one-third of the sur- face area of the Arctic Ocean. The basin is divided into two subbasins, namely the Canadian Basin and the Eurasian Basin. These basins are sepa-
D.M. Holland, Department of Atmospheric and Oceanic Sciences and Center for Climate and Global Change Research, 805 Sherbrooke Street West, McGill University, Montreal, Quebec, Canada H3A 2K6; Ph.D. 1993, McGill University (supervisor: Lawrence A. Mysak).
rated by the L o m o n o s o v Ridge, which extends from Siberia across the Arctic Ocean to Greenland. The waters of the basin c o m m u n i c a t e with the Pacific Ocean through the narrow and shallow Bering Strait and with the Atlantic Ocean via the relatively wider and deeper Fram Strait as well as the wide opening between Spitsbergen and Nor- way. There is also a flow from the western Arctic through the Canadian Arctic A r c h i p e l a g o into Baffin Bay.
Freshwater is received along the periph- ery of the basin from numerous rivers.
The Arctic Ocean is well stratified because of the annual cycle of sea-ice growth and melt and also because of river input. There is a strong pycnocline at - 3 0 0 m depth, which separates the cold, fresh surface waters from the rela- tively warm, salty deeper waters. The general circulation of the waters within this basin is not well known. It is as- sumed that the surface mixed layer is dragged along by the moving ice sur- face and thus follows an a n t i c y c l o n i c path in the Canadian basin and then exits the basin through Fram Strait. The deeper waters are thought to flow in the opposite (i.e., c y c l o n i c ) sense. The deeper waters may be strongly con- strained by t o p o g r a p h y and thus flow along the contours of bottom topogra- phy.
The present modeling work was moti- vated in part by two previous modeling studies of the Arctic Ocean, namely those of Hibler and Bryan (1987) and Semtner (1987). Both of these studies used a cou- pled sea-ice-ocean model with specified atmospheric forcing to simulate the gen- eral circulation of the Arctic sea-ice and ocean. Hibler and Bryan were not able to investigate the deeper circulation because they had imposed an artificial damping of the salinity and temperature fields back to long-term averages (i.e., a relaxation constraint) at all depths below the top level of their ocean model. Semtner
(1987) removed the relaxation constraint of Hibler and Bryan and simulated an an- ticyclonic circulation at all depths in the Canadian Basin while obtaining a cy- clonic flow below 200 m depth in the Eurasian Basin. This anticyclonic flow pattern in the Canadian Basin is inconsis- tent with observations that suggest a cy- clonic flow in that basin.
The Oberhuber model used here is a general circulation model based on the following ideas: isopycnals are used as Lagrangian vertical coordinates, a real- istic equation of state is included, the primitive equations together with the hydrostatic approximation are applied, and a surface mixed layer and a snow and sea-ice model are c o u p l e d to the interior ocean. The sea-ice model in- c o r p o r a t e s both t h e r m o d y n a m i c s and d y n a m i c s . A f u n d a m e n t a l d i f f e r e n c e from other sea-ice models is that Ober- huber uses a different numerical scheme (i.e., implicit) and also solves the sea-ice equations written in spheri- cal c o o r d i n a t e s . The work presented here is the first application of the Ober- huber model to the Arctic Ocean; Ober- huber (1993) has previously applied the model to the North Atlantic.
The so-called Atlantic layer of the Arctic O c e a n is the result of warm, saline water penetrating into the Arctic Ocean via the Fram Strait (and also partly from the Barents Sea). This water is heavier than the surface Arctic waters and subducts as it travels north of Fram Strait, o c c u p y i n g the water c o l u m n b e t w e e n 300 and 1,000 m below the surface. The model simula- tion of the velocities and temperatures at a depth of 500 m are shown in Fig- ure 1. The v e l o c i t y pattern indicates that the waters there are basically con- strained to follow the topography. The m o d e l e d circulation consists of three principal cyclonic gyres (see the num- bers 1, 2, and 3 in Fig. 1), which sim- ply m i m i c the pattern o f the ocean
OCEANOGRAPHY'VoI. 7, No. 1"1994
27
t o p o g r a p h y . This c i r c u l a t i o n can be compared with the circulation inferred f r o m earlier t e m p e r a t u r e and salinity m e a s u r e m e n t s , which show a single b a s i n - w i d e g y r e as o p p o s e d to the multigyre structure shown here. How- ever, an a n a l y s i s of recent data col- lected during the 1991 Oden cruise suggests the existence of a multigyre structure (Rudels et al., 1993).
The m o d e l e d t e m p e r a t u r e shows a pattern consistent with the circulation.
It shows the warm, saline Atlantic water e n t e r i n g via Fram Strait. The temperature decreases as the water cir- culates c y c l o n i c a l l y in the E u r a s i a n Basin. This c o o l i n g of the Atlantic layer water is due to mixing and diffu- sion of the Atlantic l a y e r w a t e r with the waters a b o v e and below it. Simi- larly, the temperature pattern suggests a cyclonic circulation in the Canadian Basin. The Atlantic layer water enters the C a n a d i a n Basin by flowing o v e r
the L o m o n o s o v Ridge near where the ridge intersects the continental slope.
Furthermore, the modeled temperature is consistent with the pattern of the ob- served field; h o w e v e r , the a b s o l u t e t e m p e r a t u r e s in the s i m u l a t i o n are everywhere -0.5°C too warm.
Overall, the study has validated the usefulness of the model for simulating the general circulation of the sea-ice and ocean in the Arctic. Although not discussed here, the simulation of the sea-ice thickness, compactness, and ve- locities is in good agreement with ob- servations (Holland, 1993). There are limitations of the model that prevent a direct a p p l i c a t i o n of the results pre- sented here to that of the global climate system. First, a global model domain was not used; rather, the simulation was p e r f o r m e d using only the Arctic Ocean and the North Atlantic Ocean. In reality, the Arctic Ocean is intricately connected to the global ocean. Varia-
1.1 1.2 1.3 1,4 1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4
UNIT! Ce,!# (us
Fig. 1: Simulated temperature and circulation of the Atlantic layer (i.e., depth of 500 m) in the Arctic Ocean Orbr January). The underlying contours give the temperature field, which ranges from hot temperatures in excess of 2.4°C in Fram Strait to cold temperatures less than 1 °C in the center of the Canadian Basin. The overlying vectors illustrate a multigyre cyclonic circulation with maximum speeds o f - 3 cm/s. The solid green color indicates land. The thin white lines represent lines of constant longitude spanning out.from the North Pole, which is located near the center of the image. The numbers' 1, 2, and 3 in the figure show the centers of three cyclonic gyres.
tions in the global thermohaline circu- lation could have impacts on the Arctic Ocean and vice versa. Secondly, a major limitation of the model is that it consists only of a c o u p l e d i c e - o c e a n model with the atmosphere specified. In reality, the polar c l i m a t e s y s t e m in- cludes three freely interacting systems, n a m e l y the a t m o s p h e r e , sea-ice, and ocean, with various feedback processes.
Only the sea-ice and ocean are coupled together here; consequently, there are no feedbacks operating between the at- mosphere and the sea-ice or ocean.
The results of this work, which are f u r t h e r d e s c r i b e d in H o l l a n d (1993), indicate that f u r t h e r studies of the polar regions can be profitably carried out using the Oberhuber model. Future work will involve studying interannual variability of the Arctic sea-ice cover and simulating the general circulation of the Arctic Ocean and its connection to the global circulation. The m o d e l will likewise be used to investigate the c i r c u l a t i o n in the Antarctic to m a k e c o m p a r i s o n s with o b s e r v a t i o n s and other modeling studies.
Acknowledgments
This project was carried out with the assistance of Prof. Lawrence A. Mysak and Dr. Josef M. Oberhuber. Computing resources were provided by Cray Research Inc. and the Arctic Region Supercomput- ing Center of the University of Alaska.
References
Carmack, E.C., 1990: Large-scale physical ocean- ography of polar oceans. In: Polar Ocean- ography, Part A, Physical Science, W.O.
Smith, ed. Academic Press, New York, 171-222.
Hibler, W.D. III and K. Bryan, 1987: A diagnostic ice-ocean model. J. Phys. Oceanogr., 17, 987-1015.
Holland, D.M., 1993: Numerical simulation of the Arctic sea ice and ocean circulation. Ph.D.
thesis, McGill University, 203 pp.
Oberhuber, J.M., 1993: Simulation of the Atlantic circulation with a coupled sea ice-mixed layer-isopycnal general circulation model.
Part I: model description. J. Phys. Ocean- ogr., 23, 808-829.
Rudels, B., E.P. Jones, L.G. Anderson, and G.
Kattner, 1994: On the origin and circulation of the Atlantic Layer and the intermediate depth waters of the Arctic Ocean. In: The Role of the Polar Oceans in Shaping the Global Environment, R. Muench and O.M.
Johannesen, eds. Amer. Geophys. Union, Washington D.C. In Press.
Semtner, A.J., Jr., 1987: A numerical study of sea ice and ocean circulation in the Arctic. J.
Phys. Oceanogr., 17, 1077-1099. {_]
