Data description:

These files include mean climatological temperature, salinity observation during the summer season in the Kara Sea. T/S data are written in separate files. The preprocessing includes: 1) The realistic data were initially interpolated into vertical grid z=0, 5, 10, 15, 20, 25, 30, 50, 75, 100, 150, 200, 250, 300, 400, 500, 600m 2) Then all obseravtions were averaged over the bins 0.35 * 0.1 degrees over the longitude and latitude respectively. The corresponding STD estimates were derived. Data are written in the matrix format suitable to read them in Matlab. Each file includes 7 columns of following data: I , J, x, y, meat(T) or mean(S), std(T) or std(S), number of data. 999.000 is used to identify the absents of the STD for the mean data.

Features of the Kara Sea circulations:

1. Anticyclonic circulation in the central part of the KS (ROFI); 2. Northeastward current along the ENZT; 3. Weak southwestward current along the eastern NZ coast; 4. Weak eastward Persey current along the FJL;. 5. Inflow/outflow of the Atlantic water along the western/eastern sides of the St. Anna and Voronin Trough; 6. Two branches of the outflow through the eastern boundary; 7. Transport of 0.63 Sv, 1.18 Sv, 1.32 Sv, 0.52 Sv through the Kara Gate, NZ-FJL passage, northern, and eastern boundary respectively; 8. 100 % of the inflow between NZ and FJL outflows through the St. Anna Trough;. 9. 75 % of the Kara Gate inflow outflows through the Vilkitsky/Shokalsky Straits; 25% - through the northern boundary

The comparison of the optimized circulation with the results of other numerical models show:

1) The drawback of the HAMSON regional model (Harms and Karcher, 1999) is the enhanced barotropization of the model solution. 2) The drawbacks of the NPS Pan-Arctic model are: a) Absents of the explicit river discharges. b) Restoring to the climatological data and week interannual (and probably seasonal) variability.