APPEARANCE DATES OF ICE PHENOMENA ON THE DNIPRO CASCADE RESERVOIRS AND THE POSSIBILITY OF THEIR LONG-TERM FORECASTING

Abstract

Long-term forecasting of the appearance dates of ice phenomena on the Dnipro Cascade reservoirs, especially with the forecast lead time at least 1 month, has important practical and scientific significance. So, it is on the basis of this characteristic of the ice regime that the terms and depths of autumn operation of each of the six Dnipro reservoirs are determined. Along with this, there are no methods for long-term forecasting of the appearance date of ice phenomena on the Dnipro Cascade reservoirs. Obviously, this is due to the fact that the problem of developing reliable long-term forecasts of the ice regime of water bodies is one of the most difficult in hydrometeorology, which has not yet received a fully satisfactory solution. The objective of this of the research is to find connections between quantitative indicators of atmospheric processes in the North Atlantic and the appearance dates of ice phenomena on the Dnipro Cascade reservoirs, and to evaluate the possibilities of using the obtained results for operational long-term forecasting. In long-term forecasting of ice regime is used the statistical, correlational, regression analysis as well as the atmospheric indicators as predictors. Therefore, the minimum atmospheric pressure in the center of the Icelandic minimum, the North Atlantic Oscillation (NAO) pattern and the Scandinavian Oscillation (SCAND) pattern were used in the research. These indicators characterize the atmospheric processes of the North Atlantic. Materials of the Department of Meteorological Forecasts of the Ukrainian Hydrometeorological Center (Kyiv) and the National Oceanic & Atmospheric Administration USA (archival synoptic maps, the values of NAO and SCAND patterns) were used as initial data. The obtained results show that such indicators of the atmospheric processes of the North Atlantic as the minimum atmospheric pressure in the center of the Icelandic minimum, NAO and SCAND patterns undoubtedly have a fairly significant influence on the formation of the appearance dates of ice phenomena on the Dnipro Cascade reservoirs, but such an influence is not predominant and such indicators can only be partially used as predictors for long-term forecasting. Therefore, to obtain more acceptable results, it is necessary to expand the search for predictors of atmospheric processes both in space and time.