Evaluate the Spawning of Fall Chinook and Chum
Salmon Just Below the Four Lowermost
Columbia River Mainstem Dams

The Problem

Spawning of fall chinook and chum salmon in the mainstem Columbia River has been documented just below Bonneville Dam and may also be occurring below The Dalles, John Day, and McNary dams. Lower Columbia River chum salmon are currently listed as endangered under the Endangered Species Act. Little is known about the existence and/or size of these populations, and their associated genetic composition, spawning timing, emergence timing, juvenile rearing duration, and juvenile to adult survival rates. Additionally, little is known about the habitat requirements for these populations and the effect changing flows and water temperatures have on spawning location and success. Chum and fall chinook salmon spawn below Bonneville Dam in the vicinity of Ives and Pierce islands at a time of the year when flows are at a seasonal low. Fishery managers are often faced with the difficult challenge of providing adequate spawning and incubation flows in the face of uncertain flow and precipitation forecasts that must be balanced with need to operate the hydroelectric system for power and management of other species.

Objectives

The objective of this study is to determine habitat requirements that are necessary for successful fall chinook and chum salmon spawning and rearing in the lower Columbia River so that the hydrosystem can be managed to protect and enhance these spawning populations. Results will be used by fishery managers to determine outflows from Bonneville Dam that will both create and protect fall chinook and chum salmon spawning and rearing habitat.

Methodology

Chinook and chum spawning habitat was assessed by estimating physical habitat parameters in our study area and incorporating that information in a predictive statistical model and a GIS- based analysis. We first conducted surveys to collect detailed bathymetric information for our study. This served as a platform

Estimated water velocities (red is high, blue is low).

for conducting two- dimensional hydrodynamic modeling. We used the River_2D model to estimate water velocities, depths, and shorelines at steady-state flows ranging from 115-160 kcfs in 5-kcfs increments. We also collected habitat surveys of spawning habitat use, and nonuse, for both chum and fall chinook salmon. We analyzed these data using logistic regression to develop models to predict the probability of redd presence in our study area. Estimates of physical habitat data produced from the hydrodynamic model and substrate surveys were analyzed using our logistic regression model in a GIS to make predictions of suitable habitat at different flows.

Highlights and Key Findings

Our predictive habitat model generally showed good agreement between actual and predicted chum and fall chinook salmon redd locations. We found that at Columbia River discharges less than 120 kcfs, an important chum salmon spawning area below the mouth of Hamilton Creek could only be supported by discharge for Hamilton Creek. At flows greater than this, the Columbia River flows could support chum salmon spawning in this area. Chum salmon did not appear to spawn in proportion to habitat availability,

Predicted chum salmon spawning habitat.

however our predictive model did not include all variables known to be important to chum salmon redd-site selection. As flows increased, the locations of chum salmon spawning habitat shifted but not the net amount of available habitat. Fall chinook salmon spawning habitat was not as sensitive to flow but the amount of habitat did increase slightly with increases in flow. The main channel of the Columbia River along Pierce Island was predicted to contain ample habitat at all flows modeled.

Where Are We Headed In 2003

In the coming year, we will begin an assessment of juvenile fall chinook salmon rearing habitat below Bonneville Dam near Ives and Pierce islands. The approach will be similar to that described above. This will be a multi-year effort and will involve a variety of data collection techniques and two-dimensional hydrodynamic modeling. A technical report summarizing our adult salmon spawning habitat will be published in the spring of 2003.

Project Contact

Kenneth Tiffan
U.S. Geological Survey
Western Fisheries Research Center
Columbia River Research Laboratory
5501-A Cook-Underwood Rd.
Cook, WA 98605-9717

Email: ken_tiffan@usgs.gov
Phone: 509-538-2299
Fax: 509-538-2843

Publications