Logging effects on streamflow: Storm runoff at Caspar Creek in northwestern California

Abstract

The effects of road building and selective tractor harvesting on storm runoff were assessed for a small (424 ha) coastal watershed in northern California. Road building alone did not significantly affect the storm runoff. After road building and logging, lag time was decreased approximately 1.5 hours, and the very small storm volumes (less than 1209 m 3) and storm peaks (less than 566 L/s) were increased by about 132 and 111%, respectively. Storm volumes and peaks of large storms (occurring less frequently than eight times a year) were not significantly increased by either roads or logging, even though more than 15% of the watershed was compacted in roads, skid trails, and landings. Although a decrease in lag time showed that the average storm hydrograph was shifted forward in time, only the small storm hydrographs were changed in shape. We speculate that the rate of delivery of water to the stream channel during large channel-forming flows was governed by infiltration and subsurface flow rates on the 85% of the watershed that was unaffected by roads, landings, or skid trails. From these findings we conclude that, in a rain-dominated hydrologic environment, logging and forest road construction (as carried out in this study) are not likely to change the flow regime of a stream adversely.Res., I7(4), 907-917, 1981.

Effects of forest practices on peak flows and consequent channel response: a state-of-science report for western Oregon and Washington.

“…5b) did not follow the general trend of decreasing magnitude of change with event return period, but actually showed the highest percentage increases in peak flow for the largest storms. These anomalous results may reflect cross-transfer of water during the largest events as a result of road drainage, as discussed by Harr et al (1979), and subsequently by other researchers (Wright et al 1990). In general, results from the Coyote Creek watersheds are suspect because of the paucity of large storms during either the pre-or posttreatment periods.…”

Proceedings of the coast redwood forests in a changing California: A symposium for scientists and managers

“…2A) Past analyses of South Fork peakflows compared pre-logging correlations to North Fork peaks with those from between the onset of logging and either 1 (Ziemer 1981) or 2 years (Wright et al 1990) after logging ended, thus including 3 years of logging. No significant increase was found for moderate to large peaks.…”

“…Research results from seminal paired watershed studies in the Pacific Northwest (PNW) show that timber harvest activities do affect annual water yield (Harris 1973, Rothacher 1970, Wright et al 1990), peak flows (Harr et al 1975, Jones and Grant 1996, Rothacher 1973, Ziemer 1981, and sediment yield (Brown and Krygier 1971, Fredriksen 1970, Lewis 1998). These results, for the most part, chronicle the environmental impact associated with a bygone era.…”

Flood potential of South Prairie Creek, Pierce County, Washington

“…Harr and others (1975) found significant increases in peak flows after road building, but only when the roads occupied at least 12 percent of the watershed. Wright andothers (1990, p. 1657) found that in northwest California "storm volumes and peaks of large storms... were not significantly increased by either roads or logging even though more than 15 percent of the watershed was compacted in roads, skid trails and landings." Jones and Grant (1996, p. 970) found that road building in combination with clear-cutting in western Oregon resulted in "significant increases in all sizes of peak discharges in all seasons."…”