Abstract
The DRAINMOD model is a superior tool used to predict the changes in farmland water balance under different agricultural drainage layouts, fields, weather conditions, and management practices. In the present study, we assessed the sensitivity of the DRAINMOD predictions in farmland water balance to the time step (hourly or daily) in daily evapotranspiration (ET0) computations for 12-hectares of farmland located at the lower reaches of the Yangtze River basin. The model was calibrated and validated and then was applied twice under two sets of daily ET0 values, computed using the standardized ASCE Penman-Monteith model (one using the hourly time step (HTS) and the other using the daily time step (DTS)). Regarding daily computed ET0 values, results show that abrupt diurnal changes in the weather always result in significant differences between daily ET0 values when computed based on DTS and HTS. DRAINMOD simulations show that such differences between daily computed ET0 values affected the model's predictions of the "water fate" in the study area; e.g., adopting HTS rather than DTS resulted in a 4.8% increase, and a 3.1% and 1% decrease in the models' cumulative predictions of runoff, drainage, and infiltration, respectively. Therefore, for a particular study area, it is critical to pay attention when deciding the best time step in ET0 computations to ensure accurate DRAINMOD simulations, thereby ensuring better utilization of agricultural water alongside high agricultural productivity.