Аннотация:Detailed long-term hydrometeorological dataset for Russian Arctic seas was created using hydrodynamic modelling via regional nonhydrostatic atmospheric model COSMO-CLM [Rockel, Hense, 2008] for a period of 1980 – 2016. In this work we present results of many test experiments, an optimal modelconfiguration choice, long-term experiments evaluation techniques and primary analysis of obtained dataset.The scheme of proposed runs based on 2-step downscaling from global reanalysis data to ~12 km grid domain and then to ~3 km grid domains over Barents, Kara and Laptev seas. Presently, the base domain (~12 km) experiments are completed only, therefore the following results are only concerned to it.Experiments were conducted for model domain including Barents, Kara and Laptev seas, with ~12 km grid. Many test experiments with different model options for periods August-September 2015 (summertime) and December-January 2012-2013 (wintertime) were evaluated to determine the best model configuration.More than 400 meteorological stations data used for experiments verification. As a result, model errors in a new model version 5.06 reduced compared to version 5.0 due to refined turbulence parameterizations, as well using the “spectral nudging” technique. Also, there are no significant differences between ERA-Interim and new detailed ERA5 reanalyses as driving conditions, therefore the final experiments were conducted using ERA-Interim reanalysis [Platonov, Varentsov, 2019].Scheme of an additional “assimilation” of soil properties from ERA-Interim global reanalysis data was suggested to avoid possible errors increment, particularly due to soil draining in the model. This technique was used worldwide for long-term experiments aimed to create hydrometeorological datasets. We have started our experiments every month using reanalysis soil data as initial and boundary conditions, i.e. reinitialized modelevery month. Final long-term consequent experiments over the base domain were simulated on MSU Supercomputer Complex “Lomonosov” during about 6 months, using 144 nodes and become more than 120 Tb data volume excluding many side files.Next, we have estimated obtained dataset comparing it with some well-known archives and reanalysescovering Russian Arctic including ERA-Interim, ERA5, ASR, satellite data. Primary assessment was calculated for surface wind and temperature characteristics. It was shown that wind speed climatology based on COSMOCLM experiments is very close to the ERA-Interim pattern, besides many details of wind speed distribution atdifferent Arctic regions is observed in COSMO-CLM data. There are well appeared sites, where some mesoscale details were reproduced by COSMO-CLM dataset compared with ERA-Interim, including east of Svalbard, Severnaya Zemlya islands, and the western coast of northern Novaya Zemlya island. The last most bright feature is associated, supposedly, with manifestation of the well-known local bora, downslope winds of mesoscale nature. At the same time, high wind speed frequencies based on COSMO-CLM data are increased compared to ERA-Interim, especially over Barents Sea, Arctic islands (Novaya Zemlya) and some seacoasts andmainland areas. Regional details are manifested in wind speed increase and marked well for large lakes (Ladoga, Onega), orography (Taymyr and Kola peninsulas, Nether-polar Ural, Eastern Siberia highlands), as well over polar region (up to 0.5 – 1 m/s). At the same time, there are mesoscale wind speed decreasing compared to ERA-Interim data over Pechora and Laptev Sea coasts, New Siberian islands.Analysis of wind speed frequencies above 17.2 and 20.8 m/s has shown that differences mentioned foraverage wind speed get more significant, especially over Svalbard, Severnaya Zemlya, Putorana plateau and Tiksi bay. Comparison of two periods (1980 – 1990 and 2010 – 2016) has shown that spatial distributions of high wind speed frequencies are very similar, but there are some detailed differences. Wind speed frequenciesabove 17.2 and 20.8 m/s has been decreased in the last decade over the Novaya Zemlya, southwest from Svalbard and northern Atlantic, middle Siberia continent; at the same time, it has been increased over between Franz Josef Land and Severnaya Zemlya, and in polar regions. These results could be interpreted as justifying manifestations of climate changes in Arctic region with some limitations.Preliminary assessment of modeling results revealed that it is promising for analysis of regional wind speed regime and severe wind speed risks estimations. The next step of this work is to run simulations at 3 km grid and collaborate with scientific community to use this dataset sufficiently. Obtained dataset could providenew, more thorough, and justified estimates of the current regional climate changes, as well as extreme weather events. The data can be used for environmental studies and the modern environmental changes researches and scientific applications, such as forcing to modeling the ocean's characteristics (wind waves and dynamics),coastal ecosystems (turbulent heat and moisture fluxes, greenhouse gases), experiments on more detailed research of individual phenomena on nested domains (extreme situations, hazardous weather events, etc.), analysis of trends in the frequency of occurrence of extreme events and features of their spatial distribution, climatology and tracking of polar mesocyclones, etc.