Elevation-based bias correction of chirps and GSMAP rainfall in northwest vietnam (2000–2024): application to climate-resilient transport infrastructure management

Reliable rainfall data are essential for hydrological modelling and infrastructure planning, particularly in mountainous regions where gauge networks are sparse and climate risks are growing. Satellite rainfall products such as Climate Hazards Group Infrared Precipitation with Stations (CHIRPS) and Global Satellite Mapping of Precipitation (GSMaP) have been increasingly applied worldwide, yet their performance in complex terrains remains uncertain. This study evaluates CHIRPS and GSMaP against daily observations from 11 meteorological stations in Northwest Vietnam during 2000–2024 and applies an elevation-based bias correction. Statistical analyses were used to compare accuracy, including RMSE and correlation, and performance maps were generated to reveal spatial error patterns. The results show that CHIRPS achieved lower RMSE but tended to underestimate rainfall at high-elevation stations, while GSMaP yielded slightly higher correlations (R² ≈ 0.74–0.81) but often overestimated rainfall in mid-elevation valleys. Elevation-based correction reduced RMSE, narrowed bias spread across the network, and produced notable improvements at upland sites. These corrected datasets enhance reliability for hydrological simulations and flood risk assessment, especially along National Highway 6, where landslides and flash floods frequently disrupt traffic. Overall, the findings demonstrate the practical value of integrating satellite rainfall with ground observations to supplement precipitation information in ungauged regions, supporting safer and climate-resilient transport infrastructure in mountainous areas