Results of probabilistic hydrological processes research at RSHU

Currently, long-term estimates can be obtained either under the assumption of statistical stationarity of hydrometeorological processes using actual series of observations for the previous decades, i.e., in fact, by extrapolating “frozen” current probabilistic estimates to the future, or by modeling (calculation) based on equilibrium climatic scenarios under the assumption of statistical sustainability of runoff series, according to which parameterization of forecast models of runoff formation is conducted. The article considers the methodology of partially infinite hydrology, which includes sustainable forecasting of runoff and diagnostics of bifurcations of its formation, allows solving fundamentally new hydrological problems (including problems of engineering hydrology) related to the possibility of obtaining longterm estimates of probabilistic characteristics of long-term river runoff under the conditions of evolutionary changes in the runoff formation factors (climate and anthropogenic activity in catchment areas). Using the methods and patterns of partially infinite hydrology and relying only on the available hydrometeorological information (obtained at the state network of standard observations), known climatic scenarios and plans for the socio-economic development of the territory, the following main results have been obtained: 1) river basins have been diagnosed (as well as time intervals in the future), the ones in which (and when) it is possible to change the additive mechanism of the smooth evolution of the flow formation process to a bifurcation mechanism (the appearance of bifurcation foci) being identified, i.e. engineering hydrology documents can be questioned; 2) a methodology has been developed for sustainable forecasting of the probabilistic characteristics of long-term river runoff using various options for its formation models (unimodal, polymodal, one-dimensional, multidimensional, etc.).