SOIL-MOISTURE PREDICTION UNDER MAIZE IN SANDY LOAM

Soil-Plant-Atmosphere-Water (SPAW) model has been used to provide dyna mic soil water estimates under irrigated maize at Delhi for six years, 1985 to 1991, excepting 1987. Simulated soil moisture values are comp ared with gravimetrically observed data for three layers, viz., 0-2.5 cm, 2.5-30 cm and 30-60 cm. Simulated runs mimic the observed fluctuat ions and trend in moisture content of different layers of soil profile . Deeper soil layers exhibit better agreement between simulated and me asured soil moisture content compared to skin layer of profile. Output from six runs were used to make comparisons at 40 points for each lay er. Correlations with standard errors (cm) for first three layers are 0.59 (0.3), 0.62 (1.3) and 0.62 (1.1), respectively. For underlining t he utility for model for irrigation Scheduling in real time. observed and simulated values (integrated up to 60 cm depth) were pooled for si x years. High correlation coefficient (0.89) was observed with SE 1.0 cm. The results established the usefulness of model in irrigation sche duling for maize crop by predicting moisture content.