Evaluation of four gridded climate products in streamflow and drought simulations in the Kelantan River Basin, Malaysia
Du, Hongrong; Tan, Mou Leong; Chun, Kwok Pan; Fei, Zhang
Authors
Hongrong Du
Mou Leong Tan
Dr Kwok Chun Kwok.Chun@uwe.ac.uk
Lecturer in Environmental Managment
Zhang Fei
Abstract
The availability of meteorological stations is generally limited and sparsely dispersed in many regions. Gridded climate products (GCPs) are commonly utilized in hydro-climatic modelling due to their wide spatial and temporal coverage. This study aims to evaluate the capability of four GCPs, namely the Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS), the fifth generation of European Reanalysis Land Component (ERA5-Land), the Asian Precipitation Highly Resolved Observational Data Integration Towards Evaluation of Water Resources (APHRODITE), and NASA Prediction of Worldwide Energy Resources (NASA POWER) in simulating streamflow and drought from 1985 to 2015 in the Kelantan River Basin, Malaysia, using the Soil and Water Assessment Tool Plus (SWAT+) model. Then, the four GCPs are integrated with the temperature data of ERA5-Land and NASA POWER to identify the best combination to run the SWAT+ model in this tropical basin. The correlation coefficient (CC) values of CHIRPS, ERA5-Land, APHRODITE, and NASA POWER exhibit the highest performance on the monthly precipitation estimation, followed by those at the annual and daily scales. In the monthly streamflow simulations, the original GCPs tended to perform moderately with the NSE values of 0.49 (NASA POWER), 0.33(ERA5-Land), 0.61 (APHRODITE) and 0.08(CHIRPS); R2 values of 0.50 (NASA POWER), 0.40(ERA5-Land), 0.74(APHRODITE) and 0.55(CHIRPS). In general, the SWAT+ model incorporated by the temperature data from NASA POWER performs not much difference with the combination with the ERA5-Land temperature data. After conducting a bias correction using the observation data, the performance of all the four GCPs has improved dramatically, with NASA POWER and ERA5-Land performed the best, with the NSE the NSE values of 0.69 (NASA POWER), 0.67(ERA5-Land), 0.67 (APHRODITE) and 0.64(CHIRPS); R2 values of 0.79 (NASA POWER), 0.78(ERA5-Land), 0.80(APHRODITE) and 0.79(CHIRPS). The CHIRPS, APHRODITE, and ERA5-Land captured the majority of drought events as measured by Standardized Precipitation Index (SPI) and Standardized Streamflow Drought Index (SSI). SPI-1/SSI-1 captures severely droughts with more accuracy than SPI/SSI at other time scales. This study suggests the use of APHRODITE, NASA POWER ERA5-Land data for hydro-climatic simulations in this tropical basin.
