Modeling of Land Surface Hydrologic Processes in Hydroclimate Modeling

The model suggested are based on the finite-element schematization of river basins which is carried out taking into account the river basin topography, soils, land use, and vegetation and include the following main blocks: 1) a detailed description of snow processes and snow cover formation in the forest and at open areas, including change of the snow depth, content of ice and liquid water, snow density, snowmelt, snow interception, sublimation, re-freezing melt water, snow metamorphism; 2) heat and moisture transfer in soil in cold periods including soil freezing and thawing, movement of the freezing front, and soil moisture transfer from the unfrozen layer of soil to the freezing front; 3) vertical heat and soil moisture transfer and evapotranspiration in the warm periods including the interaction between hydrological and carbon cycles in vegetation cover; 4) interaction between surface and groundwater; 5) overland, subsurface,ground flow; 6) channel flow (the Saint-Venant equations; the advection-diffusion equation; the kinematic wave equations). It is assumed that the snow water equivalent, the saturated hydraulic conductivity and the depth of frozen soil have subgrid stochastic variability and the ways of taking into account this variability have suggested. The structure of the models is chosen on the basis of results of sensitivity analysis and analysis of runoff generation mechanisms in given river basins. The part of parameters are the measured basin characteristics, several parameters are determined on the basis of empirical dependencies including the measured characteristics, and the rest of parameters are calibrated against runoff or other water balance component measurements. The model with some structural variations has been applied to data from more then 10 river basins with different phisiographic conditions including such large river as Vyatka, Kolyma, Seim. It has been showed that only 4–6 of most important parameters should be calibrated using comparatively short series of runoff measurements, whereas the values of the rest 10–12 parameters can be regionalized or assigned on the basis of empirical dependencies. Possibilities of using different empirical dependencies and transferability of these dependences and model parameters have been investigated. Numerical experiments have been carried out to test ability to apply the suggested physically based models of snowmelt runoff generation for the poorly gauged river basins.