Аннотация:The application of a predictive intellectual system previously developed for the Northern Dvina River is considered for a new region—the basin of the Lena River. The use of this technology under conditions of another region becomes possible due to the similar formulation of the problem of forecasting and publish- ing new open sets of hydrological and meteorological data for the period of 1985–2019. Based on the results of observations at gauging and meteorological stations, the system makes it possible to form a short-term fore- cast of the formation of powerful ice jams in river sections under conditions of incompleteness and data gaps. Interpolation methods based on machine learning are used to prepare the initial data and eliminate gaps. Cal- culations have shown the efficiency of the predictive system. The estimated accuracy of forecasting is 76%. The assessment of the importance of the factors have shown the common influence of groups of factors in different regions on the final result of the ice jamming process.
