ИСТИНА

Wind velocity at the boundary layer has strong spatio-temporal variability, therefore its cor-rect simulation is still associated with significant errors. It is especially relevant to Arctic seashores with complicate surface and to poor data areas. One of the ways to reduce errors is the mesoscale modeling. Many studies [6, 4, 5, 8] has shown definite advantage of mesoscale atmospheric mod-els in the reproduction of wind velocity pattern against the most modern reanalyses. In [11] many interesting features of wind velocity statistical distribution were revealed. Wind velocity extremes are belonging to different general samples, each described by Weibull dis-tribution. These sets of extremes were called as ‘black swans’ and ‘dragons’ [8, 9], at that ‘drag-ons’ exceeds ‘swans’ 10 – 30 % (by the same quantile 0,99). These extremes and specifications could be reproduced by the mesoscale modeling only. In this study, the climate version of COSMO model (COSMO-CLM) was used. It is well-known non-hydrostatic regional atmospheric model developed by German Weather Service (DWD) and CLM-Community [1]. The COSMO-CLM model was applied for many case-studies simulation of the most extreme winds observed over the Russian Arctic basin during the last 15 years. Many cases were presented for further modeling: 15 – 17.12.1997, 29 – 30.10.2000, 26.01.2002, 05.02.2003, 12.12.2013. Model runs were performed for the unified ‘large’ domain with spatial resolution of 0.120, covered the Barents Sea, part of Kara Sea, northern European territory of Russia and the surround-ing water areas. Driving conditions came from ERA-Interim reanalysis (~0.750 resolution). After that, the downscaling technology was performed for the different ‘small’ domains (resolution of ~2.8 km), inside the ‘large’ domain. Standard configuration of COSMO-CLM model (version 5.0) was applied: Runge-Kutta integration scheme with 5th advection order; 40 vertical levels; prognos-tic TKE-based scheme for turbulence, etc. [2]. Runs continued for a week for the most cases in-cluding extreme situations observed near the middle of the period. Analyze has shown that model reproduces the synoptic-scale dynamics and general synoptic-scale wind velocity patterns well as both with the 0.120 km, and 2 – 3 km resolutions. Model with 2.8 km resolution succeed to reproduce detailed spotty wind pattern, caused by local orography or/and dynamic factors. On the one hand, the model underestimates observed mean values and wind gusts over seashores up to 4 – 5 m/s systematically. On the other hand, it could be interpret-ed as follow: such extreme speeds of air particles (15 – 20 m/s and more) does not make much physical sense to focus on wind velocity at a certain point. Therefore, we can consider wind veloc-ity values for some surrounding area, according to the distance, corresponding to wind velocities. Taking into account these reasons, we can ascertain, that COSMO-CLM model reproduces wind velocity pattern quite adequately, but using the resolution 5 km and less, only. With respect to revealing many differences between ‘swans’ and ‘dragons’ situatons, there were found out no clear distinctions. It could be associated with a poor sample of cases. We can assume it caused by the rare overlay of large-scale synoptic factors and many local meso- and mi-croscale factors (surface, coastline configuration etc.). In the future, further studies of the extreme wind speeds genesis in the Arctic, such as the ‘black swans’ and ‘dragons’, necessary to focus on nonhydrostatic high-resolution modeling using downscaling techniques.