Impact of VAISALA AWS surface observations on the rainfall forecast in the Philippines

Abstract

The impact of ground data assimilation on the Numerical Weather Prediction in the Philippines was investigated using observations from 38 Automatic Weather Stations(AWS). Two sets of experiments were conducted - with and without data assimilation. The forecasts were generated using the Weather Research and Forecasting model with initializations from June 10 to June 19 at 12 UTC. The Mean Absolute Error difference of the forecasts were computed and mapped to show the performance of both experiments. Results show that the positive impact clusters in areas around the AWS. The sparse number of AWS observations has limited the impact of data assimilation.