Abstract:Animal treading can change soil physical properties, and thus is an important factor in hydrological modelling. We investigated the impacts of animal treading on infiltration by using a series of rainfall simulation experiments at Whatawhata Research Center, Waikato, New Zealand. The study identified significant variables for estimating soil steady-state infiltration at a microsite (0Ð5 m 2 ) and fitted the Green and Ampt equation by modifying or including variables for soil and water parameters and animal activities on grazing paddocks. A regression function for estimating steady-state infiltration rate was created for each of four scenarios: between tracks (inter-track), track, easy slope with ash soil, and easy slope with clay soil. Significant variables included the number of days after treading, antecedent soil moisture, field capacity, percentage of bare ground, bulk density, and the high degree of soil damage (damage not compacted). Regression models explained more than 71% of the variance in steady-state infiltration for three scenarios, but only 53% for the easy slope with clay soil. The remodified Green and Ampt equation provided satisfactory estimation of infiltration for all scenarios (accuracy >80%), and thus enables us to use the modified model for Waikato hill country pastures of different topography, soil physical condition, season and grazing management.