Performance evaluation of satellite precipitation estimation with ground monitoring stations over southern Himalayas in Bhutan

• Published in: Acta geophysica 2020-06, Vol.68 (3), p.933-943
• Main Authors: Adhikari, Kirtan, Choden, Yeshi, Cheki, Tshering, Gurung, Lily, Denka, Tshering, Gupta, Vivek
• Format: Article
• Language: English
• Subjects: Bias; Confidence intervals; Correlation; Earth and Environmental Science; Earth Sciences; Error correction; Geophysics/Geodesy; Geotechnical Engineering & Applied Earth Sciences; Ground stations; Hydrologic data; Hydrometeorology; Meteorological data; Performance evaluation; Precipitation; Precipitation data; Precipitation estimation; Rain gauges; Rainfall; Rainy season; Remote sensing; Research Article - Atmospheric & Space Sciences; Root-mean-square errors; Satellite data; Satellite observation; Satellites; Seasons; Structural Geology; TRMM satellite; Tropical rainfall; Tropical Rainfall Measuring Mission (TRMM)
• ISSN: 1895-6572; 1895-7455
• DOI: 10.1007/s11600-020-00434-z

High-resolution remote sensing-based hydro-meteorological data products are being increasingly used for various scientific studies throughout the world. As such, it is important to evaluate the quality of the data retrieved by the means of remote-sensing especially for the regions characterized by the drastic variation of topography such as the Himalayas. This work focuses on the comparison and evaluation of the Tropical Rainfall Measuring Mission (TRMM) Multi-satellite Precipitation Analysis (TMPA–3B43v7) with point-based ground observations recorded by Department of Hydro-Met Services (DHMS) in Bhutan. For this study, 32 rain gauge stations over Southern Himalayas in Bhutan (SHB) were selected and precipitation data for a period of 19 years (from 1998 to 2017) were compared with the TRMM precipitation product. The comparison revealed that the coefficient of correlation between satellite data and ground observation is statistically significant at a 95% confidence level. Furthermore, the coefficient of correlation is near unity in some stations and an average of 0.814 over the entire SHB region for 19 years. This finding imparts that the TRMM can capture the rainfall trend over the SHB. As per the findings, the average root-mean-square error was 219.1 mm per month considering the entire duration and 344.7 mm per month considering only the rainy season. Similarly, the average Bias was computed at 0.27 for all seasons and 0.32 for the rainy season, indicating TRMM underestimates the precipitation over SHB. The satellite estimate provides a piece of good information on the distribution of rainfall over the SHB. Nevertheless, it is still advisable to correct the bias of the satellite product, as the relative root-mean-square error is larger than 50% for 96% of the evaluated stations.