Estimating and modeling the effects of grass growth on surface runoff through a rainfall simulator on field plots

Seasonal changes in grass cover impact the generation of surface runoff due to the effects of grass roots on soil hydrologic properties and processes (i.e., infiltration). Using a rainfall simulator in a grass field site, we broadly investigated the influence of different initial conditions of soil moisture and grass growth stages on rainfall–runoff transformations. To parameterize the stages of grass growth, we used the height of the vegetation hveg, which is related to the leaf area index. Surprisingly, typical characteristics of runoff formation (peak flow and time to peak flow) were conditioned mainly by hveg. The runoff coefficient decreased about 40% when grass reached its maximum growth and was inversely and significantly related to the height of grass in general. Using the rainfall simulator experiments, we estimated the saturated soil hydraulic conductivity ks, a key parameter of infiltration models. We found strong relationships between ks and hveg when the Philip infiltration model was used, and we proposed a linear relationship between ks and hveg, making ks vary in time with grass growth (i.e., hveg). We compared predictions of hydrologic models at plot scale using ks varying with grass growth with predictions using a constant ks, as hydrological models commonly assume. Neglecting ks variability with grass growth can lead to errors up to 100% in surface runoff predictions at an event time scale and up to 87% at a monthly time scale. Ecohydrological models for runoff predictions should take into account the influence of grass growth dynamics on soil infiltration parameters.