Persistence of effects of high sediment loading in a salmon-bearing river, northern California

Many rivers in the United States have been disturbed by increases in sediment loads from activities such as mining, grazing, agriculture, logging, and road construction. Too much sediment in rivers can diminish the quality of drinking water and damage aquatic habitat. Since 1955, in Redwood Creek, north coastal California (Figure 1), several large floods resulted in widespread landsliding and gullying from previously logged hillslopes, as well as channel sedimentation and burial of streamside redwood trees (Figure 2). Because of the impact of this excess sedimentation on salmon habitat, Redwood Creek is currently listed as "sediment impaired" under the Federal Clean Water Act.

To track the fate of fresh flood deposits in the Redwood Creek basin, the U. S. Geological Survey (USGS) and the National Park Service (NPS) established a channel monitoring program in the mid-1970's. Over the next three decades they measured river channel changes through field surveys of channel width and streambank erosion, river bed elevation, number and depth of pools, and analyses of the gravel particle size on the river bed.

The history of the flood sediments differed between upstream and downstream reaches of the river (Figure 3). Channel aggradation (the filling in of a channel bed with sediment) was especially severe in the headwaters of the Redwood Creek basin, but by the mid-1980's the river had cut down to pre-flood levels, the size of gravel in the streambed had doubled, and remnants of flood deposits along valley walls were becoming well vegetated and stabilized. In contrast, in the downstream portion of Redwood Creek, channel aggradation (filling) continued through the 1980's as sediment was flushed from upstream to downstream. Here, the channel bed is still in the process of downcutting, and the size of streambed gravel is slowly increasing. This process is illustrated in Figure 4, where the river is cutting through sandy flood deposits down to a gravel and boulder river bed.

As the river has cut through previous flood deposits, pool frequencies and depths have increased. The number, location, and depths of pools are quantified through longitudinal profile surveys, which measure the channel bed elevation along a length of stream (Figure 5). More pools exist in the upstream reach, which has been recovering for a longer time (30 years) than in a reach 20 km downstream, which has only started to recover 17 years ago (Figure 6). In 1997 a 10-year flood (a flood that would be expected to occur about once in 10 years) again resulted in some sedimentation in the pools, reversing the trend of recovery for a few years. By 2006 the number of pools was again increasing in both reaches, but has not reached the pre-flood level seen in 1995.

Salmon depend on clean gravel and deep pools for spawning and rearing. The duration of sediment impacts in Redwood Creek has affected multiple life cycles of salmon and steelhead. The persistence of sediment impacts has implications for river restoration. Once sediment enters a river channel, it can take decades to be routed downstream to the ocean. In steep, confined mountain channels, in-channel restoration opportunities are limited by access. In the case of Redwood Creek, the NPS has focused restoration work on addressing hillslope erosion problems rather than attempting to modify the river bed. The goal of hillslope erosion-control work is to reduce sediment supply to Redwood Creek in future storms, but the effectiveness of this work in preventing road failures, landslides, and gullying has not yet been tested by a large (25-year) event.

Implications:
Land managers in downstream portions of a watershed may have to deal with sediment impacts decades after sediment entered the river in upstream areas.

Long-term (> 30 years) monitoring has value in revealing a story of river recovery that would not have been possible with a short-term effort.

The effectiveness of recent hillslope erosion-control work in protecting rivers from sedimentation has not yet been tested by a large storm.

• Madej, M.A. and V. Ozaki. 1996. Channel response to sediment wave propagation and movement, Redwood Creek, California, USA. Earth Surface Processes and Landforms. Vol. 21, p. 911-927.
• Madej, M.A., and Ozaki, V., 2009, Persistence of effects of high sediment loading in a salmon-bearing river, northern California, in James, L.A., Rathburn, S.L., and Whittecar, G.R., eds., Management and Restoration of Fluvial Systems with Broad Historical Changes and Human Impacts: Geological Society of America Special Paper 451, p. 43-55, Boulder, CO. doi: 10.1130/2008.2451(03).
• Madej, M.A. 1999. Temporal and spatial variability in thalweg profiles of a gravel bed river. Earth Surface Processes and Landforms. Vol. 24, p. 1153-1169.