engleski

Experience in conditional probability application for short-time fog forecast at Zagreb airport

Long-lasting fog events at major airports can cause significant delays. Therefore, studies of fog are important in aviation meteorology, as improved forecasts can lead to considerable savings. During last cold season a simple statistical model for probabilistic short-range visibility forecasting was put into use at the operational forecasting service in the Croatian Air Navigation Service. A first-order autocorrelation process is the theoretical foundation of the model, which combines climatology and persistence. From that, a simple forecast equation for a given meteorological element (developed by Gringorten, 1971) can be defined. It links the correlation between values of the meteorological element at different time steps with the conditional probability for onset of predefined values. Hourly correlation coefficients, which describe the climatological persistence of visibility, are calculated for each month, and then used to forecast visibility 9 hours in advance. Besides the median forecast of visibility, 50 % and 80 % confidence intervals are calculated as well. The forecast by percentile is also provided for comparison. This conditional probability model was presented to the forecasters last year at a training seminar. Its use in forecasting low visibility at Zagreb Airport began in October. In February this year a short assessment was made for each forecaster. Its results show that forecasters are now able to use conditional probability and related knowledge in forecasting. There are some minor issues that can be corrected along the way, but in general, conditional probability theory can be successfully applied in improving short-range fog forecasting.

fog, Zagreb airport, conditional probability

nije evidentirano