Spatial and Temporal Climate Variations Influencing Medium-Range Temperature Predictions Over South-Central European Russia

To issue medium-range forecasts, forecasters should integrate climatology, synoptic meteorology, statistics, local topography, computer science, human experience and judgment. Often, if forecasts extend beyond 72 hours, a systematic approach is abandoned for a more haphazard one. The aim of this research is to reduce a forecasters guesswork and make the forecast method more objective. A general climatology was developed for south-central European Russia. Monthly and diurnal trends are easily noted. One station, Astrakhan, was selected for a more detailed climatology. A monthly overview for 9 stations in the region is given for the weather elements as stated above. Pentad and 3- hourly averages are also included. Detailed climatology is helpful in orienting a new weather forecaster to the peculiarities of a novel region. Regression equations were developed using lag functions of 1-5 days. Low and high temperatures for the forecast period were regressed back to day zero. Low and high temperatures were predicted for 1-5 days past the initial forecast day. The regression equations developed from 24 years of data have a higher percentage of correctly predicting temperatures at the times mentioned than do climatology, persistence, or combinations thereof. MM