Analysis of the water balance in the Caledon river catchment using the soil and water assessment tool (swat) model

Abstract

The Caledon catchment is one of the primary sub-catchments in the Orange Senqu River basin and a crucial hydrological area in both South Africa and Lesotho. The sustainable water resources planning of the Caledon can be achieved through hydrological modelling, given its status as a transboundary catchment shared by Lesotho and South Africa. This study was initiated to critically analyse the water balance components (WBCs) and understand the rainfall-runoff behaviour of the catchment, which are essential for sustainable water resources development and management. This study presents an application of the SWAT model (Soil and Water Assessment Tool) in the mountainous-agricultural Caledon catchment, which spans Lesotho and parts of South Africa, covering an area of 13 370 km². The watershed area was divided into 31 sub-watersheds, predominantly occupied by agricultural land (42.44%), grassland (42.29%), urban development (5.05%), shrubs (5.74%), and water bodies (4.47%), with the remaining 0.01% being range land. The catchment ranges in longitude from 26°34’46” to 28°49’55” and in latitude from -28°22’53” to -29°43’58”, with an elevation ranging from 1413 m to 3158 m. The model was set up using observed rainfall from the Lesotho Meteorological Services (LMS) and observed temperature. The SWAT model was calibrated and validated manually by adjusting the curve number (CN2) for the period from 2010 to 2018. Model performance was evaluated using statistical performance indicators such as Nash–Sutcliffe model efficiency (NSE), the ratio of the root mean square error to the standard deviation of measured data (RSR), and the coefficient of determination (R²). The model, set up with observed rainfall from LMS, produced satisfactory results for both calibration and validation, demonstrating a high capability for estimating runoff in the Caledon catchment when coupled with observed rainfall. The outputs of this study will be used by engineers, scientists, and decision-makers to manage water resources at the sub-watershed level. Finally, the developed SWAT model analysed important water balance components, including rainfall, direct runoff, evapotranspiration, surface runoff, and water yield at the catchment level for the nine-year period from 2010 to 2018. A drought occurred between 2015 and 2016 due to low precipitation, but there was a significant increase in precipitation thereafter. The monthly rainfall range over the nine years of simulation was from 0 to a maximum of 88.83 mm, evapotranspiration (ET) ranged from 1.44 to 78.00 mm, direct runoff from 0 to 22.51 mm, and water yield from 0.09 to 21.13 mm. It was observed that ET and water yield were the only water balance components consistently above zero, possibly due to groundwater contributing to water yield, resulting in perennial flow throughout the years. Overall, the performance of the SWAT model in simulating the water balance in the Caledon catchment is rated as good, and the calibrated model could be used for runoff simulation in this agriculture-dominated watershed. The SWAT model application, supported by GIS technology, proved to be a flexible and reliable tool for water resource decision-making.