Chirps Rainfall Estimates form Monthly Bias Adjustment

To address the bias inherent in CHIRPS precipitation data, a monthly multiplicative correction was applied following the methodology of Piani et al. (2010), using in situ rainfall measurements from 62 meteorological stations. For each station, the geographically nearest CHIRPS grid point was identified based on latitude and longitude. A monthly bias ratio was then calculated as the quotient of observed to CHIRPS-estimated precipitation values, as shown in Equation: \[\text{Bias Ratio}_{i,m} = \frac{P_{\text{obs},i,m}}{P_{\text{CHIRPS},i,m}}\] Where \( P_{\text{obs},i,m} \) represents the rainfall observed at station i during month m, and \(P_{\text{CHIRPS},i,m}\) denotes the corresponding CHIRPS pixel value. These ratios were computed for each month and station across the full study period (1983–2022). The resulting bias values were spatially interpolated using inverse distance weighting (IDW) to generate continuous monthly bias surfaces over the CHIRPS grid. Subsequently, each CHIRPS estimate was adjusted by multiplying it by the interpolated bias ratio for the corresponding location and month, as indicated in Equation: \[P_{\text{corrected},j,m} = P_{\text{CHIRPS},j,m} \times \overline{\text{Bias Ratio}_{j,m}}\] This correction process refined the satellite-derived precipitation estimates to more accurately reflect observed rainfall patterns across the study area, thereby enhancing the dataset’s reliability for regionalization and climatological assessments. The corrected CHIRPS dataset resulting from this procedure is hereafter referred to as “CHIRPS corrected.”These data correspond to the Orinoquía region of Colombia, a diverse landscape characterized by mountains, foothills, and vast plains. The region is located to the east of the Andes mountain range, in the northern part of South America.