Evaluasi Kinerja Teknik Koreksi Bias untuk Meningkatkan Akurasi Global Precipitation Measurement (GPM) Di Daerah Aliran Sungai Rejoso, Indonesia

Abstract

Enhancing the accuracy of satellite rainfall data from the Global Precipitation Measurement (GPM) satellite is essential to support its vital role in hydrological applications and disaster management. This study aims to identify biases and inaccuracies in GPM data, particularly in regions characterized by complex terrain and extreme rainfall, which significantly impact the reliability. To address this, various bias-correction techniques, including Linear Scaling, Regression Analysis, and Correction Factor Methods, were applied and evaluated using metrics such as the Nash-Sutcliffe Efficiency (NSE) and correlation coefficients. The results indicate that the Linear Scaling method outperforms the others, achieving the greatest NSE of 0.906 and a correlation coefficient of 0.904. Simultaneously, the Regression and Correction Factor methods exhibited robust performance with NSE values ranging from 0.79 to 0.80. The application of the correction arises from the merging of the Linear Scaling method with the IDW-based spatial interpolation technique, facilitating the validation of the spatial distribution of adjusted rainfall data. This hybrid methodology enhances comprehension of spatial-temporal variability and extreme phenomena. This study enhances the precision of satellite rainfall data and lays a foundation for hydrological modeling and disaster risk mitigation.