ИСТИНА

To assess the effect of observed climate variability on surface and subsurface runoff, a physically-based basin-scale model of the transformation of precipitation into runoff has been developed. The model consists of two submodels. The first submodel SURFBAL is the model of transformation of precipitation on the surface within a representative element of the landscape. This model considers the accumulation of snow cover on the surface of the earth, its snowmelt during the winter thaws and the spring period, the formation of surface runoff, taking into account the freezing and thawing of the topsoil zone. This model transmits to the second submodel, i.e., groundwater flow model the potential infiltration and potential evapotranspiration as the upper boundary conditions on the surface for the each landscape zones. The second model is based on MODFLOW 2005 with using UZF package. This submodel calculates the flow in unsaturated zone, groundwater recharge and ET and the formation of surface and groundwater runoff within the basin. Results of reanalysis using the described simulation system with data on the dynamics of the observed precipitation and temperature over the past 70 years on the territory of the studied river basin showed a long-term the non-stationary of the total runoff formation processes, which is in good agreement with observations for the flow of the river. This non-stationarity is associated with an increase in river flow and groundwater recharge during the winter period in the last 40-30 years due to an increase in winter temperatures and the formation of surface runoff and infiltration during the more frequent thaw events in comparison with pre 1960 years. In this case, the peak flow decreases in the spring period. Predictive calculations for 50 years ahead are performed using the LARS - WG generator , which forms time series of precipitation and daily resolution temperatures based on global circulation forecasts. The use of these series as inputs to the basin model showed that in the projection of 50 years the annual total runoff in the basin will change little, since the predicted increase in temperatures is compensated by the predicted increase in precipitation. However, model intra-annual hydrographs of surface runoff indicate a continuation of the decline in the peak spring runoff of the river in the spring period and an increase in winter runoff due to active melting of snow in winter. Groundwater discharge to the river based on the results of predictive modeling proved to be more stable in the long-term and intra-annual sections than the surface runoff. This work was supported by the Russian Science Foundation via grant 16-17-10187