SKILL OF 6HR AND 12HR THUNDERSTORM FORECASTS BASED ON ASSESSMENT OF LARGER-SCALE INFORMATION

The relationship between information, contained in aerological data fr om the European area, and a thunderstorm occurrence in the area of the Czech Republic was investigated with input data from the period of Ma y-September 1989-1991. SYNOP reports from Czech ground stations were u tilized to assess event occurrence, TEMP 00UTC and TEMP 12UTC reports from European stations were used to determine potential diagnostic pre dictors, and the TEMP00 data served as the input data set for the 12hr mesoscale model forecast to gain prognostic predictors. Each of the t wo diagnostic data sets from 00UTC and 12UTC and of the prognostic dat a set comprised about 400 predictors/predictand elements, The categori cal forecast of thunderstorm occurrence, based on the application of l inear regression and a simple version of pattern recognition, is discu ssed. The critical success index was determined for every type of fore cast and used to assess forecast skill.