CALIBRATION AND PERFORMANCE OF THE SINGLE-LAYER SOIL-WATER BALANCE MODEL FOR PASTURE SITES

Two single-layer models of soil water balance were tested against fortnightly neutron probe field measurements of soil water at four North Island sites. The first model assumed constant evapotranspiration from field capacity to wilting point (the ''Veihmeyer'' approach), whereas the second model simulated a ''two-phase'' accounting of the soil moisture balance-potential evapotranspiration above a critical water deficit and declining evapotranspiration below this deficit. The Veihmeyer model was found to be the better choice for modelling soil water in layers of 0-50 and 0-70 cm, corresponding to the zone of most pasture root activity, but the two-phase ap ach had advantages when modelling the full depth of sampled soil profile (about 120 cm). The available water capacities used in the models were derived from the neutron probe measurements. Soil moisture profiles were measured fortnightly from September 1986 to September 1988. Regression of predicted moisture deficit against measured deficit explained between 88 and 97% of the variance in most model runs for three of the four sites, with soil moisture deficit errors typically in the range of 10-20 mm. Overall, at these sites the accuracy figures for the two-phase model runs were similar to those for the Veihmeyer model runs. The fourth soil profile was disturbed and therefore atypical. In general, there was little advantage to be gained in using daily evapotranspiration data over monthly mean evapotranspiration data. Results from this study suggest that for modelling pasture soil moisture on time scales of 2 weeks or more, the simpler Veihmeyer model is as good as the two-phase model and can be applied with confidence to many pastoral agricultural applications such as irrigation planning and drought assessments.