UJI KEMAMPUAN OBSERVASI DAN PRAKIRAAN MENGGUNAKAN MODEL COSMO (STUDI KASUS HUJAN LEBAT DI JAKARTA DAN LAMPUNG 20 FEBRUARI 2017)

Abstract

BMKG continues to develop various NWP models to improve the quality of weather forecast information, such as CCAM (Conformal Cubic Atmospheric Model), WRF (Weather Research and Forecasting) and most recently COSMO (Consortium for Small-Scale Modeling). This study aims to determine the ability of the COSMO model in providing extreme precipitation forecast in Indonesia with case studies of heavy rain in Jakarta and Lampung on 20 February 2017. The COSMO model runs with a resolution of 7 km, then compares with GSMaP data and three hourly rainfall observation in Jakarta and Lampung at 5 locations. The results show that GSMaP rain satellite data has a similar pattern with the observed data, so it can be used as a spatial reference to calculate the bias between the COSMO model with GSMaP. Based on COSMO spatial data output, in general, has not given good results, with an average rainfall intensity value of -3 mm in Jakarta and -10 mm for Lampung (underestimated). Thus, it can be concluded that the COSMO model run with the 7 km resolution has not been able to provide accurate information about the occurrence of extreme rain.

Keywords: COSMO, GSMaP, rainfall observation, bias