Towards Improving Numerical Weather Predictions by Evolutionary Computing Techniques

Weather forecasting is complex and not always accurate, moreover, it is generally defined by its very nature as a process that has to deal with uncertainties. In a previous work, a new weather prediction scheme was presented, which uses evolutionary computing methods, particularly, Genetic Algorithms in order to find the most timely ‘optimal’ values of model closure parameters that appear in physical parametrization schemes which are coupled with numerical weather prediction (NWP) models. Currently, these parameters are specified manually. Our hypothesis is that the NWP model forecast skill is sensitive to the specified parameter values. And thus, by finding ‘optimal’ values of these parameters, we aim to enhance prediction quality. In this work however, the same scheme is extended by introducing different ways of prediction evaluation during the process of searching closure parameter values. To verify our new scheme, we show prediction results of an experimental case using historical data of a well known weather catastrophe: Hurricane Katrina that occurred in 2005 in the Gulf of Mexico. Obtained results provide significant enhancement in weather prediction.