Arctic Sea have low salinity water in surface layer due to existed sea-ice. The low salinity water prevent to heat exchange ocean to sea-ice and contribute to density stratification in Arctic Sea. But most Arctic models can not reproduce the low salinity water. So, this study conducted three-experiments for reproduction the low salinity water.
The first, I thought sea-ice melting water would be a factor the low salinity water, given the large sea-ice thickness initial condition. But sea-ice thickness is influenced another factor than initial condition. So, the first experiment is not simulated less than 29.5psu salinity water. However, It was confirmed that the low salinity water at the beginning of the first experiment and it spilled out of Arctic Sea during the experiment. Although, created to the low salinity water, it must be stored inside the Arctic Sea. So, It must be constituted model dynamics in the experiment.
The second, I thought non-point freshwater not be measured something, so given the large Runoff condition. It showed the effect to lower salinity in surface layer, however that was smaller in Barents Sea and Alaska coastal sea. But It can be confirmed that more low salinity water in mixed layer of Canada Basin by salinity vertical section. In addition, freshwater content pattern was more similar to PHC climatology data.
The third, I thought more non-point freshwater has effect in the low salinity water. So, The third experiment increased Runoff and precipitation. It was subjected to increased precipitation condition, sea ice and Arctic sea water influenced. Overall, Arctic Surface layer have much lower salinity water. However, Sea ice did not reproduce similar in HadISST sea ice pattern. Therefore, precipitation condition must be considerated with Runoff condition and computed geographically in Arctic Sea.
These experiments were performed using a three-dimensional ocean-sea ice model. Arctic ocean circulation simulations were carried out for 20 years by 1980 data. The model is ROMS (Regional Ocean Modeling System) 3.4 version. Horizontal grid was constructed an orthogonal curvilinear coordinate system and vertical grid was constructed S-coordinate system. Horizontal resolution is about 22-30km, vertical resolution have 50 layers.
Temperature, Salinity initial condition is interpolated using PHC (Polar science center Hydrographic Climatology) in January data. Atmospheric forcing is used ECMWF (European Centre for medium-Range Weather Forecasts) ERA interim data of 0.75° resolution. Wind data interval is 3-hour, another data interval is 12-hour. Heat and salt flux at the surface are calculated by bulk formula. Sea surface temperature was restored in the ocean and not under the sea-ice. The sea-ice standard was used OISST last 10 years data.