摘要：Water is one of the most critical resources derived from natural systems. While it has long been recognized that forest disturbances like fire influence watershed streamflow characteristics, individual studies have reported conflicting results with some showing streamflow increases post-disturbance and others decreases, while other watersheds are insensitive to even large disturbance events. Characterizing the differences between sensitive (e.g. where streamflow does change post-disturbance) and insensitive watersheds is crucial to anticipating response to future disturbance events. Here, we report on an analysis of a national-scale, gaged watershed database together with high-resolution forest mortality imagery. A simple watershed response model was developed based on the runoff ratio for watersheds (n = 73) prior to a major disturbance, detrended for variation in precipitation inputs. Post-disturbance deviations from the expected water yield and streamflow timing from expected (based on observed precipitation) were then analyzed relative to the abiotic and biotic characteristics of the individual watershed and observed extent of forest mortality. The extent of the disturbance was significantly related to change in post-disturbance water yield (p < 0.05), and there were several distinctive differences between watersheds exhibiting post-disturbance increases, decreases, and those showing no change in water yield. Highly disturbed, arid watersheds with low soil: water contact time are the most likely to see increases, with the magnitude positively correlated with the extent of disturbance. Watersheds dominated by deciduous forest with low bulk density soils typically show reduced yield post-disturbance. Post-disturbance streamflow timing change was associated with climate, forest type, and soil. Snowy coniferous watersheds were generally insensitive to disturbance, whereas finely textured soils with rapid runoff were sensitive. This is the first national scale investigation of streamflow post-disturbance using fused gage and remotely sensed data at high resolution, and gives important insights that can be used to anticipate changes in streamflow resulting from future disturbances.
