Evaluation of DRAINMOD for simulating phosphorus and nitrogen loss from a subsurface drained field in northcentral Ohio, USA

A recent version of the widely used water management model, DRAINMOD, was developed for simulating phosphorus (P) dynamics and transport in artificially drained agricultural land. This model version is fully integrated with the earlier nitrogen (N) version of DRAINMOD, which makes it ideal for simultaneously simulating both P and N dynamics in drained croplands as affected by weather, soil, crop, and drainage related factors. The primary objective of this study was to test the newly developed P component of the model and the secondary objective was to assess the model’s capability of simultaneously simulating both N and P dynamics in drained cropland. Measured data (2017–2020) from an artificially drained agricultural field in northcentral Ohio was used to parameterize, calibrate, and evaluate simulated subsurface drainage discharge as well as losses of nitrate (NO 3 - -N), dissolved reactive P (DRP), and total P (TP) via drainage water. The overall model performance was characterized as “good” for simulated monthly discharge (Nash-Sutcliffe Efficiency (NSE) of 0.75), NO 3 - -N load (NSE of 0.63), and DRP load (NSE of 0.63) through subsurface discharge. However, the model performance was poor in simulating monthly TP load via subsurface discharge (NSE = 0.00), resulting in an overall underprediction of about 19 %. These findings illustrate that DRAINMOD can reasonably simulate subsurface discharge and associated soluble nutrients. However, model improvements are required to simulate sediment-bound P transport. Additionally, further validation of the model’s P component across a range of soils, drainage characteristics, and climates are needed and would advance the model’s utility. • DRAINMOD was used to simulate N and P dynamics at a field in northcentral Ohio. • The model reasonably simulated discharge and associated soluble nutrients. • DRAINMOD largely underpredicted simulated total P (TP). • Further model testing is required with focus on processes controlling TP dynamics.