Controls and determination of resistance to overland flow on semiarid hillslopes, Walnut Gulch

ABSTRACT:

We performed 136 experiments on small runoff plots on semiarid grassland and shrubland at Walnut Gulch using trickle to generate overlandflow. The values offiiction factor f obtained were an order of magnitude less than those values recorded by Welt. et al. (1992). A second set of experiments on the grassland reveals that the dzfference between the values of f arise from the fact that we used trickle whereas Welt. et al. used simulated rainfall to generate overland flow. The higher values of f under rainfall are due not only to the retarding effect of rain on the flow but also to the overland flow generated by rain being distributed over a larger proportion of the plot than that generated by trickle. Although trickle experiments help to understand the factors controlling f, they do not yield values that can be used in modeling runoff and soil erosion. For such purposes f must be determined using rainfall-generated overland flow.