USGS Home
Contact USGS
Search USGS

U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2008–5093

Back to Table of Contents

Sediment-Transport Characteristics

Sediment-transport models also require data on the variation of total sediment discharge (suspended-sediment discharge [fine and sand] plus bedload) with changes in stream discharge at the upstream model boundary. Measurements of suspended-sediment discharge also are required to test model performance.

The available supply of fine-grained particles typically controls the suspended-sediment discharge of fine-grained particles in a stream because the supply often is less than the stream can transport (Colby, 1956). These fine-grained sediments move downstream at about the same velocity as the water.

In contrast, the supply of coarser grained sediments in streams generally is greater than the stream can transport; therefore the transport of coarser grained sediments as bedload typically is controlled by the ability of the stream to transport them (Guy, 1970). Bedload is sediment that moves on or near the streambed by sliding, rolling, or bouncing (Edwards and Glysson, 1999). Most bed sediment moves occasionally, but remains at rest much of the time, especially in gravel/cobble/boulder streams. Because of these differences in transport mechanisms, large variations can be expected in the concentrations and grain-size characteristics of sediments at different locations in a stream and with changes in stream discharge. Fine- and coarse-grained sediment concentrations in a stream increase with increasing stream discharge (Edwards and Glysson, 1999).

Suspended Sediment

Since the study of Clark and Woods (2001) ended in April 2000, additional suspended-sediment samples have been collected at three sites in the study reach: NF Coeur d’Alene River at Enaville (12413000), SF Coeur d’Alene River near Pinehurst (12413470), and Coeur d’Alene River near Harrison (12413860). The first two sites are at the upstream boundary of the study area, and the last site (Harrison) is at the downstream boundary. Clark and Woods (2001) collected samples only at the Rose Lake gaging station (12413810) and provide a complete description on the collection, quality control, and processing of suspended sediments.

Clark and Woods (2001) developed relations between suspended sediment and river discharge from data collected between February 1999 and April 2000. They used a power function to develop transport curves for sediment discharge. First they converted suspended-sediment data from milligrams per liter to tons per day using Qs = Q × Cs × k, where Qs is suspended-sediment discharge, in tons per day; Q is stream discharge, in cubic feet per second; Cs is suspended-sediment concentration, in milligrams per liter; and k is a conversion factor of 0.0027. A relation between Qs and Cs then was developed for each gaging station. The data pairs for the Enaville and Pinehurst gaging stations (Clark and Woods, 2001) showed a strong correlation (r = 0.974 and r = 0.945, respectively); however, the Rose Lake and Harrison gaging stations showed a weaker correlation (r = 0.888 and r = 0.829, respectively). The r value increased as distance increased from Coeur d’Alene Lake probably due to the effects of backwater in the river.

The relations between suspended sediment and river discharge were then updated using the same procedures as Clark and Woods (2001). The old and new regression curves and new equations and correlation coefficients for the Enaville, Pinehurst, and Harrison gaging stations for suspended sand discharges are shown in figure 10.

The old and new regression curves and new equations for suspended fines (silt and clay) discharges, and the sampled data pairs are shown with the relations in figure 11. No updated curve was available for the Rose Lake gaging station because no suspended-sediment samples were collected since Clark and Woods’ (2001) study. The data pairs for these sites had a strong correlation (r ≥ 0.90). The highest r value of 0.964 occurred at the Enaville gaging station for suspended sand discharge, and the lowest r value of 0.906 occurred at the Harrison gaging station for suspended fine discharge. For these sites, the scatter amount about the regression line is minimal. However, for the Pinehurst gaging station, a fair amount of scatter is around the regression line in suspended fine (silt and clay) and sand discharges. Clark and Woods (2001) attributed the scatter to hysteresis.

Total Sediment Discharge

Suspended-sediment discharge (fines and sands) plus bedload data were used to determine total sediment discharge. Bedload samples were collected during Clark and Woods’ (2001) study, and since then, no bedload samples have been collected. The bedload curves from Clark and Woods’ (2001) were used without modification. Total sediment discharge (Q_T) and total suspended-sediment discharge curves for the Enaville, Pinehurst, Rose Lake, and Harrison gaging stations are shown in figure 12. The bedload contribution to total sediment discharge is small; typically less than 15 percent (Knighton, 1998). Emmett (1975) indicated that bedload contribution ranged from 1 to 10 percent for streams in the upper Salmon River basin. Clark and Woods (2001) indicated the contribution from bedload from Enaville and Pinehurst gaging stations was less than 10 percent. For the Rose Lake and Harrison gaging stations, the bedload contribution was less than 1 percent. Although bedload was sampled several times at the Harrison gaging station, no sediments were collected in the samplers even at a discharge of 24,500 ft³/s (April 16, 2002). Thus, an assumption of zero bedload was made at the Harrison gaging station. Total sediment discharges (Q_T) for the Enaville and Pinehurst gaging stations were used to estimate the total sediment-transport discharge for any river discharge at the upstream boundary of the sediment-transport model. Sediment discharge at the Harrison gaging station was used to test the sediment-transport model.

Comparison of total sediment discharges (Q_T) from the Enaville and Pinehurst gaging stations indicates that on an equal discharge basis, the contribution of total sediment discharge (Q_T) is greater from the South Fork than from the North Fork. Total sediment discharge of the South Fork near Pinehurst is about 10 times greater than from the North Fork at Enaville for river discharges ranging from 100 to 10,000 ft³/s. For river discharges ranging from 1,000 to 5,000 ft³/s, total sediment discharge on the South Fork near Pinehurst is about 15 times greater than the North Fork at Enaville. Because water discharge is much greater from the North Fork than the South Fork, the annual contribution of sediment discharge to the main stem is greater from the North Fork than the South Fork. For example, in water year (WY) 1996, Q_T for the North Fork at Enaville was 381,000 ton/d and for the South Fork near Pinehurst was 73,400 ton/d, about one-fifth of Enaville. On average in WY 1996, river discharge at the Enaville gaging station was four times greater than the river discharge of Pinehurst.

Back to Table of Contents