Validasi data hujan satelit TRMM dengan data hujan observasi dalam analisis hujan rancangan Daerah Aliran Sungai (DAS) Serang

The accuration of rainfall data is essential for hydrological analysis, while observation station data are often limited. The Tropical Rainfall Measuring Mission (TRMM) satellite can serve as an alternative data source; however, validation against ground observations is required. This study aims to evaluate the accuracy of TRMM data, examine its correlation with ground-based rainfall data, and compare the design rainfall results for various return periods. The research was conducted in the Serang Watershed (DAS Serang), located in Kulon Progo Regency, Special Region of Yogyakarta. Rainfall data from 2008 to 2019 were used and tested for consistency using the RAPS method and stationarity tests. The data were then calibrated and verified using regression models, and validated through statistical parameters including RMSE, NSE, KR, and the correlation coefficient (R). The design rainfall frequency analysis was carried out using four probability distributions: Gumbel, Normal, Log-Normal, and Log Pearson Type III. The results show that the corrected TRMM data has good accuracy based on 11 years of calibration and 1 year of verification using a polynomial equation. The evaluation yields a relatively small RMSE, an NSE within the acceptable category, 'very accurate' Relative Error (KR), and a very strong correlation. The highest correlation occurs in the 1-year period (R = 0.966; R² = 0.9522). Frequency analysis reveals that TRMM-calculated design rainfall is 30%–40% lower than observed data, though the upward trend remains consistent. In conclusion, TRMM is a viable alternative for rainfall data, especially in areas with limited stations, although further calibration for large return periods is necessary.