Evaluation of Meteorological Radar Precipitation Forecast in Banten

Meteorological radars offer several advantages in representing weather phenomena, including estimating rainfall distribution with high temporal and spatial resolution and representing data in large observations. Automatic Weather Stations (AWS) and Automatic Rain Gauges (ARG) are ground station devices for the measurement of weather parameters, including the intensity of precipitation. Rain gauges suffer from high spatial and temporal variability, especially at low density. The reflectivity is a product of the radar observations. To obtain an estimate of rainfall at the site, it is processed using the Z-R Rosenfeld tropical relationship. Unfortunately, meteorological radars have certain limitations, including beam blocking from the nearest surface, beam attenuation, RF interference, and ground silence or unobserved areas. This study evaluated radar rainfall estimation compared to spatial and temporal rain gauges in Banten. This is especially useful for areas of Banten where there are no rain gauges. Evaluation metrics used in the statistical analyses, namely correlation coefficients, root mean square, and mean absolute error. The results of processing radar imagery for rainfall estimation from ARG rainfall measurements for 1 - 31 December 2022 showed a correlation of 0.80, a mean error of 3.10 and a RMSE of 9.02.