ИСТИНА

Spatial resolution of model/reanalysis plays a key role in forecating/analysis of spatial structure and intensity of orographic winds. Comparison of reanalysis datasets and simulations with mesoscale model WRF-ARW with different resolution with the satellite measurements from Synthetic Aperture Radars (SAR) for several cases of strong wind allowed us to estimate the influence of spatial resolution on reproducibility of orographic winds. It was found that almost any spatial resolution is suitable for tasks where the average spatial and temporal values of wind speed are required. At the same time, it is necessary to use a grid spacing 3-5 km or smaller for tasks related to the analysis of the fine structure of orographic winds (such as compiling a detailed regional climatology, analyzing extreme values, and assessing the risks of such dangerous phenomena as clear air turbulence or dangerous sea waves). Therefore, the use of reanalysis for such tasks is impossible. However, improved resolution often leads to poor simulation results when compared with observational data. We found that the reason of wind underestimation in numerical experiments with the WRF-ARW model is the unrealistically high surface roughness used in the model, which causes the so-called separation of the boundary layer. The correction of the roughness length was carried out in accordance with the observational data on the meteorological mast installed in Tiksi, where strong downslope windstorms are observed. The use of the observed roughness length in the model leads to a 2–3-fold reduction in the wind speed error. This work was funded by the Russian Foundation for Basic Research (project No. 18-05-80065) and the Russian Science Foundation (project No. 18-47-06203)