On Transformation of Vegetation Parameters in Modeling the Response of Terrestrial Heat and Water Balance Components to Possible Climate Changes

A scenario-based methodology has been developed for projecting changes in vegetation parameters used in Earth system parameterization schemes in Russia under the influence of global climate change. The methodology is based on artificial neural networks and reproduces the dependence of these parameters on climatic, edaphic, and orographic factors. For the 2073–2100 projection period and the SSP5-RCP8.5 climate scenario, changes in more than fifty vegetation parameters derived from the ECOCLIMAP physiographic database across Russia were calculated, and maps of the spatial distribution of these changes were constructed. It is shown that changes in the considered parameters by the end of the 21st century for the selected climate scenario are quite noticeable and multidirectional (in terms of the sign). The absolute deviations of the projected values of various parameters from historical ones, averaged over the territory of Russia, range from 5 to 17%.