Evaluation of GSMaP and TRMM TMPA Rainfall with Surface Rainfall on Jakarta – Bogor

Abstract

Global Satellite Mapping of Precipitation (GSMaP) and Tropical Rainfall Measuring Mission Multi-Satellite Precipitation Analysis (TRMM TMPA) are global measured precipitation projects using satellite combination (blended) technique of infrared sensor and microwave sensor. This research having purpose to evaluate GSMaP and TRMM data with Surface rainfall data in various topographies (beach, level-land, mountainous region, and The whole area) and temporal variation (daily, 10days, and monthly) in Jakarta-Bogor area. The evaluation is done using visual and statistical comparison method (correlation coefficient, MAE, RMSE, and Mann-Whitney Test). The evaluation result of daily and 10days of GSMaP in all area studies showing a pattern didn’t follow the Surface rainfall (r<0.20), while TRMM daily rainfall data in beach area and The whole having r>0.60 and the Mann-Whitney test shows TRMM doesn’t have significant difference with surface rainfall data. 10days TRMM rainfall in all region, except mountainous region, shows TRMM rainfall can follow surface rainfall pattern (r>0.77). Monthly GSMaP data comparison to surface rainfall showing correlation more than 0.60 on beach and land area can be applied, by using correction equation, because GSMaP rainfall data always lower than surface data. The monthly TRMM rainfall data can be applied in all studies area because the pattern is suitable to surface rainfall data (maximum correlation is 0.98). Error identification shows that there is warm rain and high variation on land surface emission especially in mountainous area causing either GSMaP and TRMM cannot follow the surface rainfall data. GSMaP correction equation in beach region, shows the parameter that affect error are surface temperature and specific humidity in 850mb, while at region II, III, and The Whole there is no local parameter which has influence. TRMM correction equation in region III(mountainous area), shows local parameter which has an influence to accurate of this TRMM is surface pressure, wind 850mb, interaction of meridional wind and surface slope, air temperature 850mb, soil humidity in 10-200cm, and skin temperature.