USGS Home
Contact USGS
Search USGS

U.S. GEOLOGICAL SURVEY
Scientific Investigations Report 2007–5185

Back to Table of Contents

Summary and Conclusions

The Willamette River flow and temperature models were used to explore and quantify the thermal effects of point-source discharges, riparian shading, and upstream dam operations on water temperature in the Willamette River and portions of its largest tributaries. The models, which form the basis of the Willamette temperature Total Maximum Daily Load (TMDL), were reviewed prior to their use, and several modifications were made to correct errors, remove instabilities, and make the models and their results more usable. Model results were evaluated using methods very similar to those used in the construction of the TMDL, in which the results of a model scenario were compared to model results from a baseline scenario and the 95^th percentile of the 7-day moving average of the daily maximum (7dADM) water-temperature difference was calculated as a function of location along the modeled river reaches. In this way, the cumulative heating effects of one or more point sources or the thermal effect of restoring riparian shade along a particular river reach, for example, could be quantified relative to a Natural Thermal Potential baseline condition. Because the results of each model run were evaluated as a difference between that run and a baseline model run, any errors that occurred in both were largely eliminated via that subtraction. The predicted temperature differences, therefore, were expected to be far more precise than the modeled prediction of actual river conditions, meaning that predicted temperature differences as small as hundredths of a degree Celsius might be real and useful.

The flow and temperature models first were used to examine the cumulative heating effects of the 27 point sources included in the TMDL. The models were run for each point source separately to quantify the “heating signature” of each source. These signatures then were incorporated into a spreadsheet-based screening tool in which each signature could be linearly increased or decreased using a multiplicative “strength factor.” The sum of the strength-weighted point-source heating signatures was used as an estimate of the cumulative point-source heating effect, such that when the strength factor for each point source was set to 1.0, the sum equaled the results from one model run that included all point sources at their full TMDL heat allocation. This screening tool allows users to quickly evaluate the effects of potential changes to point-source heat allocations, including the trading of such allocations among the point sources. Graphs and various metrics are calculated to facilitate the evaluation of potential heat-load trades. Once a potential trade is identified through the use of the screening tool, the modified point-source conditions can be evaluated with the full suite of flow and temperature models. Four test trades were evaluated, and the screening tool was shown to accurately predict the resulting cumulative point-source heating effects when compared to the exact same conditions simulated with the Willamette flow and temperature models, with mean absolute errors on the order of 0.002 degrees Celsius (°C) and maximum prediction errors of roughly plus-or-minus 0.005°C.

The cooling effects of riparian shading were quantified with the Willamette River flow and temperature models. Model results showed that restoring all riparian shading along the Long Tom River could cool the Willamette River at its point of inflow by approximately 0.03°C, which is small but potentially useful for heat-load trades with some of the point sources upstream of Albany. Shade restoration along selected 5-mile reaches of the Willamette River upstream of Albany showed cooling effects as large as 0.19°C at certain locations. A 5-mile reach can be traversed in a few hours, however, which caused the cooling effects to exhibit a nodal pattern with downstream distance. The cooling effect was minimal at one-half day of travel time downstream of the restored reach because the water passed by that reach during the night when shading does not particularly affect the heat budget of the river. This pattern in downstream cooling effects has potentially important ramifications for heat-load trading and the siting of riparian restoration projects.

Finally, the downstream temperature effects of changed operations at Cougar Dam on the South Fork McKenzie River were evaluated using the TMDL models. Since 2005, temperature releases from Cougar Dam have been controlled through the use of a selective withdrawal tower; this change occurred after the 2001-2002 time period modeled for the temperature TMDL. The downstream thermal effects were estimated by substituting 2006 measured temperatures downstream of Cougar Dam for the measured 2001 and 2002 temperatures used in the TMDL models. The model results showed that the modified operations at Cougar Dam have an important effect on downstream temperatures, with warming in mid-summer and cooling in autumn. Predicted 7dADM temperature shifts in the South Fork McKenzie River were greatest, with changes as large as 6.0°C. Downstream, the models showed 7dADM temperature changes in the McKenzie River of close to 2.0°C at times. In the Willamette River upstream of the Santiam River confluence, predicted 7dADM temperature changes were as large as 0.5°C, while downstream of the Santiam River the changes were less than 0.3°C. Although large temperature shifts were apparent as a result of the change in operations at Cougar Dam, if those changes also were incorporated into the Natural Thermal Potential baseline conditions, the modeled cumulative point-source heating effect was not significantly altered.

This investigation has helped to quantify the heating effects of several important influences on water temperature in the Willamette River and several of its major tributaries. Only through a detailed understanding of the river’s heat budget and the factors that influence it can scientists, resource managers, and regulators construct defensible plans for optimizing the river’s thermal regime to protect its beneficial uses and aquatic resources. The results of this investigation, and the tools produced by it, should prove useful as these managers and regulators move forward to implement the Willamette temperature TMDL.

Back to Table of Contents