WRF model input for improved radar rainfall estimates using Kalman Filter

The indirect measurement of rain through radar reflectivity is associated with various sources of errors such as ground clutter, partial beam occultation, beam blockage and attenuation effects. Removing the systematic error (bias) and enhancing the precision and limitations of radar data sources are the main focus in enhancing radar rainfall accuracy. This research work was to reduce radar rainfall bias due to the process and measurement noises using Kalman Filter with a multivariate analysis technique. The implementation of this technique involved numerical weather prediction (NWP) namely the Weather Research Forecasting (WRF) model data output parameters such as temperature and relative humidity. The study found that filtering technique using Kalman Filter with multivariate analysis applying the WRF model output has satisfactorily improve radar rainfall estimates.