Assessment of Hydrus Model Accuracy in Simulating Soil Moisture under Drip Irrigation for Maize

Abstract

A Field experiment was conducted in a split plot design consisting of three irrigation levels (0.6 (I1), 0.8 (I2) and 1.0 (I3) crop evapotranspiration (ETc) and four nitrogen levels.  Nitrogen levels includes (80% (N1), 100% (N2), 120% (N3) RDN and Manual application (N4). The experiment was carried out on sandy clay loam soil during the kharif 2018, rabi 2018–19, and kharif 2019 seasons at Dr. N.T.R. College of Agricultural Engineering, Bapatla, Andhra Pradesh. The primary objective of the study was to calibrate and validate the HYDRUS-2D model for simulating soil moisture in a drip irrigated maize crop. The water flow equation used in the Hydrus-2D is in the form of the Richard's equation. The model was calibrated for hydraulic conductivity and dispersivity values for sandy clay loam soil. After calibration, model was validated with the seasonal observed data to examine its predictability. Model performance was evaluated using dimensionless statistical performance criteria, viz., coefficient of determination (R2) and Root Mean Square Error (RMSE), Relative error (%) and Model efficiency. Results of the study showed that, calibrated values of longitudinal and transverse dispersivities for sandy clay loam soil are 0.5 and 0.05 cm, respectively. The simulated and observed soil moisture content upto a depth of 100 cm in sandy clay loam soil at 50 days after sowing for 3h, 6h and 24h after irrigation at emitter, 15 cm from emitter and 30 cm from emitter clearly showed good agreement between observed and simulated values with R2 in between 0.992 - 0.996. These results indicate that the HYDRUS-2D model effectively simulates soil moisture distribution in the root zone under drip irrigation and can be reliably applied to other crops and similar soil conditions.