The trend and magnitude of annual runoff response to changes in amount of forest cover in the Sierra Nevada were estimated with a simple procedure. The use of this method is strictly limited to reconnaissance level assessments of the potential effects of broad-scale forest management scenarios. A model relating annual runoff response to timber harvest was developed from a simple linear regression of 31 catchment experiment results from the western United States. Correlations of runoff response following timber harvest with mean annual precipitation, mean annual runoff, and the ratio of mean annual runoff to mean annual precipitation indicated that the model would benefit from stratification by one or more of these variables. Because mean annual precipitation data are generally easier to obtain, a model was developed by stratifying the data set to a mean annual precipitation range representative of the Sierra Nevada conifer forest zone.

The simple linear regression indicates that a ten percent increase in timber harvest distributed evenly across the Sierra Nevada conifer forest zone may result in a 0 to 26-mm increase in mean annual runoff. Stratification of the data set to regions of above and below average mean annual precipitation indicated that a ten percent reduction in forest may result in a 10 to 52-mm and a -1 to 18-mm increase in annual runoff, respectively. Furthermore, these results indicate that trend and magnitude of changes in runoff following forest reduction are much more difficult to predict in drier regions.

The results of the stratified regressions were applied in two examples of projected changes in forest, one forest-wide and one Sierra Nevada-wide. The results indicate that annual runoff would be minimally affected by projected trends in forest reduction alone. However, these results do not include further effects from logging road construction, skid trails, or any other aspect of multiple-use management. Furthermore, the models and results are limited to annual runoff only.

A paired catchment technique was used to assess historical trends in channel-forming peak flows in response to long-term watershed conversion to a logging-based ecosystem in the southern Sierra Nevada. The "treated" watershed, the South Fork Tule River, was subject to cumulative logging and road construction from approximately 1950 to 1989. By 1984, 58% of the forested area and 21% of the entire watershed had been logged. Double-mass plots of the treated and control watersheds over the period 1940-1989 indicated that an inflection point occurred at water years 1967-1969, about the same time as a significant increase in the land-conversion. Separation of the data set at 1967 produced a post-conversion slope twice as steep as the pre-conversion period, implying that channel-forming peak flows increased in response to cumulative canopy reduction and road construction. One possible cause of the increase in peak flows is an increase in snowpack and exposure to latent and sensible heat flux in clearcuts. Two hypothetical scenarios were developed to assess the increase in water available for runoff from the clearcut watershed. Increases of 15% and 11% were found for the forested area and the whole watershed, respectively.

Related SNEP chapter:
Marvin, S. Possible changes in water yield and peak flows in response to forest management. Vol. 3, chapter 4, pp. 153-199.