Small coastal watersheds are perhaps the least well understood hydrologic systems. The Tsitika River Sediment Budget Project was initiated in 1991 to help understand the processes that effect sediment production and transport in coastal watersheds and the way in which forest harvesting and associated activities affects those processes. At present there is no firm understanding of the variability of input of water, nor the rate at which the terrestrial watershed elements convert that input into the output. However, the extent of our lack of understanding does not stop there - to add to the list, we don?t even know how to properly interpret stage records for steep gradient alluvial channels. Given these deficiencies, it is no small wonder that we do not understand the effects of forest harvesting and regeneration on streamflow in coastal watersheds! Paired watershed experiments do not account for the variability in either the input or the internal routing. If we ever hope to answer the questions and resolve the decades-old debate in the literature, we must develop a fully distributed simulation model of watershed hydrology that represents the input and transfer processes at a scale appropriate to their spatial variability. The use of such a model would allow the treatment watershed to act as its own control. A necessary first step in the development of a physically based simulation model should be to gain an understanding of the processes that govern subsurface flow by mechanisms other than Darcy?s Law. This is the subject of ongoing work by Axel Anderson at Russell Creek.