FORECASTING THE VOLUME OF FLOOD RUNOFF OF THE YESIL RIVER USING THE PATTERN

Abstract

The research presents the results of forecasting the volume of flood flow of the Yesil River by pattern recognition method for 1th february and 1th march 1. The calculations used daily data of water consumption, the volumes at the hydrological post of the river Yesil gauge station Turgen, also as predictors of minimum air temperature, precipitation, decadal data on water reserves in the snow cover for the long-term period 1980...2020. The results showed a satisfactory quality of the forecast in terms of efficiency and correlation. Statistical analysis showed a good correlation between the observed and predicted values: 0.76 according to the forecast for February 1 and 0.80 on March 1. The result of the research, it revealed that the prediction using the image recognition method based on the data for March 1 showed more accurate results in terms of quality.