Submitted to: Catena
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/22/2011
Publication Date: 10/18/2011
Citation: Zhang, Y., Nearing, M.A., Wei, H. 2011. Effects of antecedent soil moisture on runoff modeling in small semiarid watersheds of southeastern Arizona. Hydrol. Earth Syst. Sci., 15: 3171–3179. Interpretive Summary: The amount of water in the soil prior to a rainstorm can greatly affect the amount of water that infiltrates and the amount of water that runs off the surface of the soil. This effect can be important to understand when we apply runoff and soil erosion models in order to make assessments of soil condition and the effects of rainstorms, including soil erosion and flooding. In this study we took measured soil moisture, rainfall, and runoff data from watersheds in dry-lands of Arizona and applied a model to determine how important it is to know the soil moisture in order to obtain good estimates off runoff from storms. Our results indicate that the overall effect in terms of quantifying runoff in the long term were relatively minor. The reason for this is that in arid environments the soils are generally very dry nearly all of the time, and they dry very quickly after any rain that falls in days immediately prior to the rainstorms being studied. We hypothesize that the effect could be much more important in humid areas where soil moisture levels are generally higher much of the time.
Technical Abstract: Antecedent soil moisture prior to a rain event influences the rainfall-runoff relationship, but its degree of importance and influence on overall model prediction capability for runoff modeling in semiarid areas is not clear. This study aimed to examine the sensitivity of runoff response to antecedent soil moisture and to test the predictability of runoff at a small catchment scale by comparing the use of antecedent vs. long term average soil water content in setting the model initial conditions for the modified Green-Ampt Mein-Larson model within the Rangeland Hydrology and Erosion Model (RHEM). A nine year period of measured rainfall, runoff, and soil moisture data from five semiarid rangeland watersheds ranging in size from 0.34 to 4.53 ha on the Walnut Gulch Experimental Watershed in southeastern Arizona, USA were used to calibrate RHEM and to test hypotheses. Results showed that there was no significant correlation between measured runoff ratio and antecedent soil moisture in any of the five watersheds. Using antecedent soil moisture ranging from 0% to daily maximum record, we found that the average sensitivity of simulated runoff amounts and peaks to antecedent soil moisture were 0.05mm and 0.17 mm h-1, respectively, with each 1% increase in antecedent soil moisture. The statistics showed that there was no significant difference between runoff amounts and peaks simulated with long term average soil moisture and those simulated with measured antecedent soil moisture, indicating that runoff characteristics could be adequately simulated with long term average soil moisture as the model initial condition in these semiarid rangeland watersheds.