Modifying Automatic Irrigation in SWAT for Plant Water Stress scheduling

Abstract

Automatic irrigation in the Soil and Water Assessment Tool (SWAT) is triggered by using plant water stress and soil water deficit irrigation scheduling. Auto-irrigation is important to simulate the catchment’s behavior in response to climate change and water management scenarios. However, studies have identified deficiencies in the auto-irrigation algorithms in SWAT as the irrigation water amount simulated under plant water stress scheduling shows a large deviation from the simulated irrigation water amount under soil water deficit scheduling. Therefore, the current research deals with validating and modifying the auto-irrigation scheduling under plant water stress condition using SWAT. The modified SWAT model was evaluated against the Soil-Water-Atmosphere-Plant (SWAP) model as well as observed data for irrigation and crop yield at an experimental field (Hamerstorf, Lower Saxony, Germany) during the 2008–2018 cropping seasons. The two SWAT subroutines. swu and. autoirr were modified. The existing root density distribution function was replaced with the one proposed by Li et al. (1998) and also a dynamic estimation of the plant water uptake compensation factor (EPCO) was incorporated into the modified SWAT. The results revealed that SWAP and modified SWAT were able to simulate the irrigation amount and crop yield with an acceptable bias for all the crops at the experimental site. However, the overall spread of crop yield simulated (11 years) by both the models was less compared to the observed spread for most of the crops. Furthermore, the modified SWAT code was used to simulate the irrigation amount for three different agro-climatic catchments in Germany, India and Vietnam. Results showed improved irrigation simulation in terms of long-term annual amounts compared to the default SWAT under plant water stress condition.