USGS Home
Contact USGS
Search USGS

U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2007–5164

Back to Table of Contents

Introduction

The McKenzie River originates in the High Cascades geologic province and flows westward through the Western Cascades (McKee, 1972; Tague and Grant, 2004) before reaching its confluence with the Willamette River near the cities of Springfield and Eugene, Oregon. Located on the South Fork McKenzie River, Cougar Reservoir is one of two large flood-control reservoirs in the basin operated by the U.S. Army Corps of Engineers (USACE) (fig. 1). In order to adhere to water-temperature requirements for salmonids in the South Fork McKenzie River and downstream on the mainstem McKenzie River, a construction project began in February 2002 to modify the reservoir’s control tower and intake structure to allow withdrawal from multiple depths (U.S. Army Corps of Engineers, 2003).

The construction at Cougar Reservoir involved a drawdown of the reservoir pool to elevations well below (200–300 feet) its normal winter low-pool level (fig. 2) (U.S. Army Corps of Engineers, 2003). On February 23, 2002, a diversion tunnel beneath the dam was reopened, and beginning April 1, 2002, the reservoir’s pool elevation was lowered. The pool remained at a lowered elevation until December 2004, at which point construction was complete and the reservoir began to refill and resume normal winter operations. Additional details on the construction are provided by the U.S. Army Corps of Engineers (2003).

As a result of the initial drawdown, several miles of the upper reaches of the reservoir pool were exposed to erosion from scouring flows during rainstorms, beginning in spring 2002 (U.S. Army Corps of Engineers, 2003) and continuing until completion of the project in December 2004. Deltaic sediments that had been deposited over almost 40 years of previous reservoir operation were thus subject to mobilization and downstream transport. The resulting elevated turbidity in downstream reaches of the South Fork (fig. 3) and mainstem of the McKenzie River caused increased local concern for potential negative effects on aquatic biota, including degradation of salmonid spawning habitat, from deposited sediments (U.S. Army Corps of Engineers, 2003). Additionally, the drinking-water facility for the city of Eugene, which draws its water from the McKenzie River, reported increased treatment costs when turbidities were elevated (K. Morgenstern, Eugene Water and Electric Board, written commun., 2002).

Turbidity is defined as an expression of the optical properties of a liquid that causes light rays to be scattered and absorbed rather than transmitted in straight lines through a sample (ASTM International, 2003). Turbidity is caused by the presence of suspended and dissolved matter, such as clay, silt, finely divided organic matter, plankton and other microscopic organisms, organic acids, and dyes (Anderson, 2004). Although not strictly a measure of particle concentration, turbidity commonly is used as a surrogate for suspended-sediment concentration (SSC), obtained by using locally derived regressions (Lewis, 1996; Sun and others, 2001; Gray and Glysson, 2003), often with relatively robust results. Recently, the use of logging turbidimeters that can be deployed in streams to obtain a record of nearly continuous turbidity, together with discharge and site-specific regressions between turbidity and SSC, has allowed the estimation of nearly continuous concentrations and loads of suspended sediment (Christensen and others, 2000; Uhrich and Bragg, 2003).

Initial transport of suspended sediment during 2002 was estimated by Grant and others (2002) and the U.S. Army Corps of Engineers (2003) using relations between turbidity and SSC developed for the adjacent North Santiam River basin (Uhrich and Bragg, 2003). Nearly continuous (half-hourly) turbidity data were available, beginning in 2000, at gaging stations in the South Fork McKenzie River immediately upstream and downstream of Cougar Reservoir. However, only a few SSC samples with concurrent turbidity readings were available from these two stations, and it was unclear to what extent turbidity-SSC relations that had been developed for the North Santiam River were directly transferable to the McKenzie River basin.

Adding to local concerns about effects on aquatic biota, the USACE determined through routine testing that sediment from the reservoir and surrounding upland soils and downstream bank sediments was contaminated with low to moderate concentrations of the legacy organochlorine pesticide dichloro-diphenyl-trichloroethane (DDT) and its metabolites, dichloro-diphenyl-dichloroethane (DDD) and dichloro-diphenyl-dichloroethylene (DDE) (the sum of these terms is henceforth referred to as ΣDDx) (U.S. Army Corps of Engineers, 2003). The source for ΣDDx was presumed to be residual from forest spraying of DDT in the 1950s and 1960s to control spruce budworm and other pests in the South Fork and much of the upper McKenzie River drainage basins (Dolph, 1980; Moore and Loper, 1980). It was unknown to what extent this ΣDDx was being mobilized and transported downstream with the elevated suspended sediment following the reservoir drawdown and construction; however, there was concern that the ΣDDx could be sorbed to fine sediment and deposited in salmonid spawning areas, potentially harming developing fry.

On the basis of these concerns, the U.S. Geological Survey (USGS) was asked to help the USACE determine the contribution of sediment from the Cougar Reservoir drawdown and construction project to downstream sediment transport and deposition during 2002–04, including the potential for transport and deposition of ΣDDx.

Purpose and Scope

This report presents the results of the 2002–04 cooperative USACE-USGS study. Estimated suspended-sediment loads are mass balanced to evaluate overall sources and sinks in the McKenzie River from the South Fork down to the Vida reach, and the role of the construction project at Cougar Reservoir on downstream sediment transport and deposition. The report also evaluates the effect of the construction on transport of ΣDDx in water and deposition in fine sediment downstream of Cougar Reservoir.

Turbidity data used in the estimation of continuous SSC were collected from five stations on the South Fork McKenzie, Blue River, and mainstem McKenzie River and were previously published (U.S. Geological Survey, 2002, 2003, 2004). Individual sample data on SSC and ΣDDx, and deposition of fine sediment and ΣDDx, are presented in this report, along with daily loads of SSC as estimated from the product of site-specific regressions of turbidity and SSC, and discharge.

Approach

The approach for this study was to determine locally specific relations between turbidity and SSC and, where possible, use those relations to estimate instantaneous SSC concentrations, using hourly turbidities at selected monitoring locations in the basin. During a few high-flow events, suspended sediment was analyzed for ΣDDx to determine if it was being transported and if additional sampling was warranted. Deposition of fine sediment in spawning gravels was estimated using infiltration bags, as described by Lisle and Eads (1991), modified to allow subsampling for organic contaminants such as ΣDDx, and deployed from August 2003 to July 2004. For this report, the fine-fraction of sediment is defined as silt and clay particles having a diameter of less than 0.063 millimeter (mm).

SSC and Turbidity Relations

In the McKenzie River basin, turbidity data were collected continuously in the South Fork McKenzie River at stations upstream and downstream of Cougar Reservoir beginning in December 2000. Additional turbidity-monitors were installed in January 2003 on the McKenzie River above South Fork McKenzie River, near Rainbow; Blue River at Blue River downstream of Blue River Reservoir, and the McKenzie River near Vida downstream of all other stations (fig. 1, table 1). All turbidity-monitors were collocated with preexisting streamflow stations, or in one case installed concurrently with streamflow instrumentation, and the data were published annually (U.S. Geological Survey, 2000, 2001, 2002, 2003, 2004). However, suspended-sediment samples were collected only sporadically during the drawdown period in 2002, and the number of samples was insufficient to perform a statistically significant SSC–turbidity regression. Suspended-sediment sampling was initiated at all five gaging stations in January 2003 and continued through May 2004. Data collected was used to develop site-specific relations between turbidity and SSC.

The role of Cougar Reservoir in transporting suspended sediment downstream raises questions about the larger role of streamflow regulation in the basin. Other reservoirs in the McKenzie River system include, from the headwaters moving downstream, the Eugene Water and Electric Board (EWEB) hydroelectric facilities at Trail Bridge Reservoir, at the mouth of the Smith River at about river mile (RM) 81.9, and Carmen Smith Reservoir, about 2 miles (mi) upstream on the Smith River. The South Fork joins the McKenzie River at RM 59.7, just upstream of Blue River (RM 57.0), with Blue River Reservoir located 1.7 mi upstream on Blue River. Farther downstream of the South Fork, EWEB also operates Leaburg Reservoir (RM 38.8), which is primarily a diversion structure for additional power-generation facilities in the lower river. Of these water bodies, the flood-control reservoirs (Cougar and Blue River) experience the largest annual pool-elevation fluctuations under normal operating conditions, as pool levels are drawn down 100–200 feet (ft) in winter in order to accommodate large winter storms, and residence times are on the order of 3-4 months (Johnson and others, 1985). The reservoir pools at Trail Bridge and Carmen Smith hydroelectric facilities may have moderate, frequent changes in pool elevations (5–12 ft) but are largely designed to let large flows pass through them with minimal effect.

From the standpoint of flow regulation, therefore, Cougar and Blue River Reservoirs have a much larger effect on the hydrology of the McKenzie River than do Carmen Smith and Trail Bridge Reservoirs. In this report, “regulated” sites are those directly downstream of flood-control reservoirs; the South Fork McKenzie River near Rainbow (CGRO), located downstream of Cougar Reservoir and Blue River at Blue River (BLUE). Unregulated sites, which also serve as reference sites for evaluation of sediment transport and deposition, are those that are completely or almost unaffected by reservoir operation, including the South Fork McKenzie River above Cougar Reservoir, near Rainbow (SFCO) and the McKenzie River above the South Fork, near Rainbow (MRBO). The status of the McKenzie River near Vida (VIDA), downstream of the mouths of Blue River and the South Fork McKenzie River, is evaluated separately for the differential effects of regulation by the upstream flood-control reservoirs (Blue River and Cougar Reservoirs) on flow and suspended-sediment concentrations.

Deposition

Deposition of fine-grained sediment released from Cougar Reservoir into spawning gravels also was a concern following reservoir drawdown. Fine sediment can clog pore spaces in the streambed, decreasing subsurface flow (Brunke and Gonser, 1997) and potentially decreasing the dissolved-oxygen (DO) concentration in the pore water. Such a decrease in pore-water DO concentration could negatively affect developing eggs of salmonids (Meyer, 2003), such as those present in the McKenzie River basin, including several threatened or endangered species. Additionally, the potential that deposited fine sediment could be contaminated with residual DDT or its metabolites from historical spraying practices caused additional concern for toxicity to salmonid larvae as they developed within the spawning gravels.

During the summer of 2002 immediately following the drawdown, Gregory Stewart and others (Oregon State University, unpub. data, 2002) used freeze-core techniques to sample for fine-grained sediment throughout a depth of 0–40 centimeter (1.3 ft) in spawning gravels at approximately the same five locations upstream and downstream of Cougar Reservoir and in the mainstem McKenzie River as sampled for this study. The freeze cores allow the examination of native, undisturbed sediment in “gravel popsicles,” including profiles with depth. Although the freeze cores provide an indication of the total amount of recent and historical sediment deposition at a site, they do not by themselves provide a means to interpret the time frame of deposition or estimate recent deposition. Data from the freeze cores indicated an increased percentage of fine sediment in cores collected downstream of Cougar and Blue River Reservoirs; however, only the middle 20 cm of the cores could be used because of downstream sloughing or erosion of fine materials around the perimeter of the frozen core cylinder, primarily in the top and bottom 10 cm, upon removal of the core from the streambed. The freeze-coring technique also did not allow subsampling for analysis of contaminants within the sediment matrix.

As a result of the unknowns remaining following the freeze-core sampling, infiltration bags (Lisle and Eads, 1991) were used during 2003–04 to discern the effect of Cougar Reservoir release of fine sediment on deposition in spawning gravels. The infiltration bags involved the burial of a collapsed bag under experimental bed material (gravel or cobbles that approximate the native material) for a defined period of time. During retrieval, the bag was extended upward past the sediment surface, capturing the entire column of sediment. This technique allows for collection of fine sediment throughout the depth of the buried infiltration bag, mimicking the sediment-accumulation processes in a native streambed, and for extrapolation to an overall deposition rate given the known period of deployment; however, no information on historical deposition rates can be learned from them. In this study the infiltration bags were modified to allow clean sampling for DDT and metabolites associated with the fine sediment in the bags. Specific methods are described in the section, “Methods.”

Back to Table of Contents