Abstract

A one-dimensional daily averaged water temperature model was used to simulate Klamath River temperatures for two management alternatives under historical climate conditions and six future climate scenarios. The analysis was conducted for the Secretarial Determination on removal of four hydroelectric dams on the Klamath River. In 2012, the Secretary of the Interior will determine if dam removal and implementation of the Klamath Basin Restoration Agreement (KBRA) (Klamath Basin Restoration Agreement, 2010) will advance restoration of salmonid fisheries and is in the public interest. If the Secretary decides dam removal is appropriate, then the four dams are scheduled for removal in 2020.

Water temperature simulations were conducted to compare the effect of two management alternatives: the no-action alternative where dams remain in place, and the action alternative where dam removal occurs in 2020 along with habitat restoration. Each management alternative was simulated under historical climate conditions (1961–2010) and six 50-year (2012–2061) climate scenarios. The model selected for the study, River Basin Model-10 (RBM10), was used to simulate water temperatures over a 253-mile reach of the Klamath River located in south-central Oregon and northern California. RBM10 uses a simple equilibrium flow model, assuming discharge in each river segment on each day is transmitted downstream instantaneously. The model uses a heat budget formulation to quantify heat flux at the air-water interface. Inputs for the heat budget were calculated from daily-mean meteorological data, including net shortwave solar radiation, net longwave atmospheric radiation, air temperature, wind speed, vapor pressure, and a psychrometric constant needed to calculate the Bowen ratio. The modeling domain was divided into nine reaches ranging in length from 10.8 to 42.4 miles, which were calibrated and validated separately with measured water temperature data collected irregularly from 1961 to 2010. Calibration root mean square errors of observed versus simulated water temperatures for the nine reaches ranged from 0.8 to 1.5°C. Mean absolute errors ranged from 0.6 to 1.2°C. For model validation, a k-fold cross-validation technique was used. Validation root mean square error and mean absolute error for the nine reaches ranged from 0.8 to 1.4°C and 0.8 to 1.2°C, respectively.

Input data for the six future climate scenarios (2012–2061) were derived from historical hydrological and meteorological data and simulated meteorological output from five Global Circulation Models. Total Maximum Daily Loads or other regulatory processes that might reduce future water temperatures were not included in the simulations. Under the current climate conditions scenario, impacts of dam removal on water temperatures were greatest near Iron Gate Dam (near Yreka, California) and were attenuated in the lower reaches of the Klamath River. May and October simulated mean water temperatures increased and decreased by approximately 1–2°C and 2–4°C, respectively, downstream of Iron Gate Dam after dam removal. Dam removal also resulted in an earlier annual temperature cycle shift of 18 days, 5 days, and 2 days, near Iron Gate Dam, Scott River, and Trinity River, respectively. Although the magnitude of precipitation and air temperature change predicted by the five Global Circulation Models varied, all five models resulted in progressive incremental increases in water temperatures with each decade from 2012 to 2061. However, dam removal under KBRA appeared to delay the effects of climate change to some extent near Iron Gate Dam. With dam removal under KBRA, annual-mean water temperatures exceeded the 49-year historical mean temperature beginning in 2045; whereas with dams, annual-mean temperatures exceeded the historical mean beginning in 2025.

Potential changes in seasonal water temperatures resulting from dam removal, with or without future climate change, have a direct impact on fisheries in the Klamath Basin. Water temperature changes are of particular interest in spring (April–May) when salmon smolts out-migrate to the Pacific Ocean, and in fall (October–November) when Chinook salmon return upstream to spawn.

First posted September 21, 2011