Performance Assessment of Satellite-based Rainfall Product over the Oromia Region of Ethiopia

Rainfall data is vital for agricultural planning, water resource management, and climate research in areas like Oromia, Ethiopia, where agriculture supports a large part of the population. However, ground-based observations are often limited and inconsistent. This study assesses how well satellite-derived rainfall estimates match ground observations in Oromia. In the analysis of the ENACT, CHIRPS, and TAMSAT satellite products against ground-based rainfall data, all three exhibited a consistent value sequence but did not accurately reflect the observed declining rainfall trend from 1470 mm to 30 mm. While they showed correlation in pattern, each product indicated an increasing trend that contradicted ground measurements, suggesting issues in remote precipitation sensing due to varied algorithms and data sources. CHIRPS demonstrated a perfect Probability of Detection (POD) and the correlation value (CORR, 0.717), making it preferable despite a higher Mean Absolute Error (MAE, 55.678) compared to ENACT (48.195), which had a smaller Root Mean Square Error (RMSE, 88.753). ENACT (CORR, 0.736) displayed the strongest correlation with ground truth observations in Oromia, proving its reliability for temporal and geographical rainfall patterns, highlighting the need for ongoing validation against ground truth for improved precipitation estimates crucial for hydrological and agricultural purposes. These results imply that satellite products can be reliable alternatives in areas like Oromia, where data is limited, although regional differences in topography and rainfall patterns affect accuracy.