Evaluation of ground-based, daily, gridded precipitation products for Upper Benue River basin, Nigeria

Abstract

In-situ rain gauge stations are not adequately covering the total area of Upper Benue basin in Nigeria. An accurate gridded precipitation dataset is therefore required for climate study, hydrological modelling and water resource management. For this purpose, three candidates of global, gauge-based, daily, gridded precipitation products were evaluated and compared with the monitoring data from eight stations spanning 25 years period from 1982 to 2006. The three products are Climate Research Unit (CRU), the Climate Prediction Centre (CPC), and the Global Precipitation Climatology Centre (GPCC). The evaluations covered spatial analyses and statistical validations for daily, monthly, seasonal, and extreme rainfall data. Results show that all the three datasets captured well the spatial rainfall characteristics and were able to replicate the rainfall gradient and orographic conditions of the study area, thoughthe GPCC dataset outperformed the others. Line plots and scatter plots were used to compare the gridded datasets with the observation data formonthly data. The GPCC and CRU datasets are both better than that of the CPC. We also used the mean absolute error, the mean bias error, the correlation coefficient, and the modified index of agreement to compare daily data, and found that the GPCC set isthe most agreeable with the observation and the CPC set is the least. The daily gridded data products were also compared using probability distribution similarity, with the in-situ data by KS and AD tests. We found the distributions of GPCC and CRU datasets to follow the same distribution as that of the rain gauge data but that of the CPC does not. The results of the extreme events namely annual maximum numbers of consecutive dry-, wet-day, and total rainfall depth show that the dataset of GPCC is the most compatible with the observation and the CPC set is again the least. The most suitable daily ground-based gridded precipitation dataset for hydrological study and modelling in the Upper Benue basin is therefore the GPCC.

Keywords

• accuracy
• evaluation
• gridded precipitation data
• hydro-climatology
• spatiotemporal
• upper benueriver basin