Accuracy of WaSim and DRAINMOD Models for Subsurface Drainage Design and Analysis in a Data-Scarce Environment

Abstract

<b><sc>Abstract.</sc></b> Simulation models are useful decision support tools for designing and analysing subsurface drainage systems in irrigated lands. The challenge, however, is how to determine soil hydraulic data inputs required by models to achieve accurate simulation of water table depths (WTDs) and drainage discharges (DDs) at various combinations of drain depth and spacing. This is particularly important for data-scarce areas e.g. middle and low income countries, that lack facilities to determine in-situ soil hydraulic properties. We evaluated the performance of WaSim and DRAINMOD models to simulate WTDs and DDs at a field scale in KwaZulu-Natal, South Africa. Hydraulic conductivity (K_sat), and Soil Water Retention Characteristics (SWRC) data were determined using the in-situ pumping test, and a pressure plate apparatus, respectively. Thereafter, these parameters were estimated by pedotransfer functions (PTFs) in the Rosetta computer program. The DRAINMOD and WaSim models were calibrated using the in-situ measured K_sat and SWRC data, while the validation exercise used the Rosetta-estimated K_sat, and SWRC data estimated by the RETention Curve (RETC) computer program as soil hydraulic inputs. The models‘ performance to simulate WTDs and DDs was assessed using Nash-Sutcliffe Model Efficiency (NSE), Coefficient of Determination (R²), and Mean Percent Error (MPE). During validation, DRAINMOD simulated WTDs with NSE, R², and MPE of 0.86, 0.82, and 6.2%, respectively, whereas for DDs, the model registered NSE, R², and MPE of 0.81, 0.80, and 4.3%, respectively. On the other hand, during the same validation period, the WaSim model simulated WTDs with NSE, R² and MPE of 0.78, 0.77, and 8.1%, respectively. Whereas for DDs, during validation, the WaSim model registered NSE, R², and MPE of 0.74, 0.72, and 7.9%, respectively. The DRAINMOD model can be used to design and analyse drainage systems with a reasonably high level of confidence. On the other hand, the WaSim model requires further calibration for it to be used as a drainage design tool with soil hydraulic inputs estimated by PTFs. We recommend this drainage design approach to be used only when reliable methods of determining inputs to the Rosetta computer program have been used.