Water temperature is a key factor regulating physical and biological processes in small streams, and streamside logging often results in increased summer temperatures that can be detrimental to resident fish. Recently, Mellina et al. (2002) (Can. J. Fish. Aquat. Sci. 59: 1886-1900) developed a model to predict downstream temperature trends in small headwater and lake-headed streams using easily measured predictors, and during this study summertime stream temperature and canopy cover data collected from 21 lake-headed and headwater streams in north-central B.C. were used to subject the model to an independent test to assess its predictive capabilities across space and time. The temperature data revealed that downstream cooling was widespread among the lake-headed streams, and downstream warming was prevalent in the headwater streams, regardless of whether or not the riparian zones were harvested. The data also suggested that temperature increases following streamside timber harvesting around lake-headed streams were more modest when compared to headwater streams, and that these increases were likely mitigated by a combination of warm outlet temperatures (promoted by the presence of the lakes) and cold groundwater inputs. Using complete summertime data, the cooling model predicted with relative accuracy the average daily downstream cooling or warming in the study streams (with deviations ranging from 0.01 ? 2.02 °C), and it continued to be relatively accurate when using data from restricted summertime intervals (with deviations ranging from 0.001 ? 2.82 °C). The cooling model has the potential to become a powerful and cost-effective risk-management tool that may help managers plan appropriate timber harvesting activities that will minimize the potential impacts on stream-dwelling fish, and the independent testing provided by this study will likely increase confidence in its application to management issues.
Scott Hinch, Eric Mellina, and R. Dan Moore
Hinch, Scott G., Mellina, Eric; Moore, R. Dan. 2004. Stream temperature responses to clear-cut logging in the central interior of British Columbia: test of the predictive model developed by Mellina et al. (2002). Forest Investment Account (FIA) - Forest Science Program. Forest Investment Account Report