Water temperature modeling in streams to support ecological restoration
Water temperature is a critical water quality parameter that affects salmonid survival by influencing its metabolism and growth at all life stages. Stream temperature is an especially important parameter in California rivers where it frequently limits the range of salmonids. Anthropogenic activities have increased stream temperature and degraded spawning, holding, and rearing habitats, and this has contributed to declines in salmonid populations in California. Fisheries managers have a range of analytical and empirical tools available to assess and quantify elevated stream temperature conditions, but many of these tools do not focus on water temperature conditions at the spatial and temporal scales important to salmonids. My research focuses on assessing water temperature at the watershed and upwelling hyporheic scale which are critical to salmonid survival as stream temperature approaches thermal tolerances.
I developed a model to calculate water temperature at locations throughout a watershed to provide a method to evaluate the availability and connectivity of suitable thermal habitat throughout a stream network. The model used a linear weighted average of the maximum and minimum air temperatures of the current and 4 prior days. The weighting parameter is dependent upon upstream drainage area enabling the application of the model to both small tributaries and large mainstem streams. I used historical data from the Sonoma Creek, Napa River, and Russian River watersheds to develop, test, calibrate, and partially validate the model. Model results from Sonoma Creek and Napa River indicated it was generally able to estimate daily average water temperature within 1.5 degrees C of the observed water temperature. Data from the Russian River highlighted the model was limited to streams without significant hydrologic modifications or geologic constraints that forced groundwater to the surface.
A 1-D advection dispersion heat transport model was developed to quantify the upwelling hyporheic temperature that provides cold water thermal refugia along a streambed for salmonids. I analyzed hyporheic temperature measured at five sites in a previous research program across sixteen kilometers of Deer Creek near Vina, California, to test, calibrate, and partially validate the model. At three sites, I found the 1-D advection and dispersion were the dominant heat transport mechanisms with model root mean square error less than 0.6 degrees C. At two sites, the model was not applicable because modeling results indicated that surface flow rate variations, solar radiation, and multi-day flow paths also influenced the upwelling hyporheic temperature. Modeling was valuable for highlighting the contribution of these additional processes from that of 1-D advection dispersion. The availability of monitoring data over the summer-fall period was essential for modeling upwelling temperature dynamics along a semi-natural channel.