Recent advances in Arctic Ocean studies employing models from the Arctic Ocean model intercomparison project

Observational data show that the Arctic Ocean has significantly and rapidly changed over the last few decades, which is unprecedented in the observational record. Air and water temperatures have increased, sea ice volume and extent have decreased, permafrost has thawed, storminess has increased, sea level has risen, coastal erosion has progressed, and biological processes have become more complex and diverse. In addition, there are socio-economic impacts of Arctic environmental change on Arctic residents and the world, associated with tourism, oil and gas exploration, navigation, military operations, trade, and industry. This paper discusses important results of the Arctic Ocean Model Intercomparison Project, which is advancing the role of numerical modeling in Arctic Ocean and sea ice research by stimulating national and international synergies for high-latitude research.

z-coordinate models that use the original code of Bryan (1969) are the most common. Several AOMIP groups employ different variants of isopycnic (same density; Bleck and Boudra, 1981) and sigma coordinate (topographic; Blumberg and Mellor, 1987) Figure 1. illustration of synthesis and integration activities employing a modeling approach.
parameterization of first-year ice formation rates and multiyear ice melting rates, and include tidal forcing as well.
Gerdes and Koeberle (2007) where f is the Coriolis force, V is velocity, and ∇D is the gradient of total depth).
Topostrophy ( The greater computing resources available in recent years has allowed simulations from fine-grid, eddypermitting AOMIP models that seem to be approaching more realistic cyclonic rim-current circulations similar to those shown in Figure 2. A resolution-dependence study using the Massachusetts Institute of Technology global climate model ECCO2 is particularly revealing (Holloway et al., in press) because it shows that τ increases when the grid size is reduced. Specifically, in ECCO2, τ increases by 60% when the grid size is reduced from 18 km to 9 km, but it increases by only 13% when the grid is refined from 9 km to 4.5 km. Hence, this study suggests that the model may be improving in the sense of better representation of topographically trapped rim current with respect to further grid refinement.

pacific waters
The Canada Basin PW layer occurs at depths between approximately 50 and 150 m (Steele et al., 2004). It originates in ~1 Sv (1.0 x 10 6 m 3 s -1 ) of northward flow through Bering Strait, driven by approximately 1 m of sea level difference between the Pacific and Atlantic Oceans (Coachman and Aagaard, 1966 Project, 1997Project, , 1998 (Proshutinsky et al., 2005).  indicate that high-resolution models better represent the bathymetry of the region, and thus may more realistically represent flow through the strait.

Fre ShwaTer dYNaMiCS
However, in terms of fluxes and mean properties, high-resolution models are not always the most accurate. We find that all models achieve the correct order of magnitude for volume flux and correlate significantly with observations; however, there is still room for improvement, especially in terms of heat and salt fluxes. At the same time, additional measurements with better spatial coverage are needed to minimize uncertainties and better constrain models.
Currently, available observational data contain little information on the upper water column and near the coasts.  Mechanically, BG releases FW and FW gradients smooth. But sea ice melts, river runo increases, and FW content in the BG increases. Figure 7. Conceptual mechanisms of freshwater (Fw) accumulation and release in the beaufort gyre (bg) during an annual cycle. Freshwater content in summer and winter is shown in meters (isolines) calculated relative to salinity 34.8. Slp = sea level atmospheric pressure. The bottom panels show salinity distribution along sections in the center of the beaufort gyre region. it is hypothesized that in winter, the wind drives the ice and ocean in an anticyclonic (clockwise) sense so that the beaufort gyre accumulates freshwater mechanically through deformation of the salinity field (ekman convergence and subsequent downwelling; bottom left panel). in summer, anticyclonic winds are weaker (and may even reverse to be cyclonic), and the resultant summer anomaly in ekman convergence releases freshwater, thereby relaxing salinity gradients (bottom right panel) and reducing beaufort gyre freshwater content. From Proshutinsky et al. (2009)

eCOSYSTeM QueSTiONS aNd MOdeliNg
Physical factors play a disproportionately significant role in plankton productivity in the Arctic Ocean compared with the rest of the world ocean (Smith and Niebauer, 1993;Carmack et al., 2006;Popova et al., 2010). Light and nutrients dominate the control of Arctic primary  (Popova et al., 2010).
Recognizing that marine ecosystem modeling is complex and that ecosystems come in many forms, even in the