Understanding what controls the location of the channel head in a valley network is crucial to predicting drainage density, and the response of landscapes to changes in climate, landuse, and base level. Recent studies of channel head locations in semi-arid to humid landscapes have suggested that for a given area, channel heads are found progressively farther downslope, hence draining a larger area, in gentler valleys. Theories for channel initiation are few and are severely limited by a lack of detailed field observations on runoff and erosion processes that can be used to motivate and test physics-based equations. We propose to conduct field experiments by monitoring runoff and erosion during natural and artificial rainstorms in gentle and steep gradient unchanneled valleys. We will use results from these experiments to test and extend available models for runoff generation and we will couple with these models with erosion theories to predict channel head locations. Work will be carried out at sites in California and Oregon where we have active research field programs in hydrologic and geomorphic processes. The work proposed here is also an essential stepping stone toward formulating a theory for landscape evolution by valley development.
