Mathematical models of environmental forecasting

A technique of environmental forecasting with allowance for climate-related factors is proposed. For this purpose, a set of subspaces ranked in terms of disturbance scales is chosen by means of orthogonal decomposition from multidimensional multicomponent databases containing information on the functions of state that describe atmospheric processes for a long period. The leading part of subspaces that take into account climate-scale processes composes an informative basis for hydrodynamic background formation in calculating the forecast scenarios of changes in atmosphere quality. Calculated scenarios with estimates of atmosphere pollution in the Far East of Russia and at the adjacent territories of China and Korea are presented.