Author: Elizabeth Myers Toman

Publisher:

Published: 2004

Total Pages: 156

ISBN-13:

The management of existing forest road systems is an issue of growing importance and public debate. Roads can alter the hydrologic processes in a watershed especially at stream crossing culverts where road ditches channel runoff directly into the stream. The objective of this study was to determine how surface runoff from roads augments natural stream flow at stream crossing culverts. This study took place within an 824 ha watershed in the foothills of the Oregon Coast Range approximately three miles west of Corvallis, Oregon. Sixteen stream crossing culverts were selected for study. Discharge was measured from October 2002 through May 2003 at each stream and at the adjoining ditch(es). Hydrographs for both stream flow and ditch flow were analyzed for five storms that occurred during the winter 2002-2003. The interaction of the road with subsurface flow from the hillslope caused the hydrology of the road segment to be classified as either "intermittent" or "ephemeral." Peak flow and total runoff at the stream crossing culverts was compared with the magnitude and timing of peak flow and total runoff in the adjoining ditch(es). Forest roads were found to alter the flow paths of water through the Oak Creek watershed. The road altered storm runoff and peak flow at the stream crossing culverts seventy-four times out of seventy-eight opportunities during five storms. The amount of the change depended primarily on whether or not the road cutslope intercepted subsurface flow. Contributions of intercepted subsurface runoff to the stream were greater than contributions of surface runoff by an order of magnitude. In the Oak Creek watershed, 56 percent of the road cutslopes adjacent to streams intercepted subsurface flow. Intercepted subsurface flow was more connected in time to stream flow than surface runoff. Ditch flow, on average, contributed the most volume on the rising limb of the stream hydrograph at the culvert. Flow responses at individual culvert locations were highly variable and could not be predicted using traditional topographic variables.