|Abstract or Summary
- The western United States is experiencing significant changes in wildfire and snow regimes as a result of warming temperatures. An amplification of wildfire activity and reduction in snow water equivalent, snow covered area, and earlier spring snowmelt are documented trends that are projected to continue into the future. With an increase in wildfire activity, it is important to understand how a reduction in snow will impact regenerating vegetation in the western United States. The first objective of this study was to assess summer vegetation biomass response to antecedent winter snow on a local scale by determining the physiographic characteristics that influence the relationship between snow and vegetation in the case of the 2002 Biscuit Fire. The second objective was to assess the broad scale regional patterns of regenerating vegetation response to snow, by comparing the correlation between summer vegetation biomass and antecedent winter snow before and after large wildfires across the western United States. Remote sensing data and spatial-temporal statistics were used to analyze the relationship between snow and vegetation. In the local scale analysis, the 2002 Biscuit Fire was analyzed, which burned over 2,000 km² in southwest Oregon and northern California. Nonparametric Multiplicative Regression (NPMR) was used to explore the complex relationships between multiple predictor variables (winter snow frequency, elevation, slope, aspect, and burn severity) and the summer vegetation response variable (enhanced vegetation index, or EVI), before and after the Biscuit Fire burned. The burned area was subset by soil type to determine how soil texture influenced the snow and vegetation relationship. In the regional scale analysis, the Pearson's Correlation Coefficient was calculated to analyze the relationship between winter snow frequency and summer EVI before and after 23 wildfires across the western United States. In the case of the Biscuit Fire, summer EVI responded negatively to snow before the fire, and responded positively to snow after the fire. EVI in coarse-textured skeletal soils exhibited the clearest shift to a positive response to snow after the fire burned, while EVI in fine-textured clay soils did not exhibit this type of shift. The regional analysis proved that wildfire disturbances affect the relationship between snow and vegetation differently across the western United States. Seven fires clustered near the Biscuit Fire in northern California and southwestern Oregon behaved similar to the Biscuit Fire, shifting from a negative pre-fire snow and EVI correlation to a less negative or positive post-fire snow and EVI correlation. The majority of these fires had relatively low average elevations (430 to 1708 m) with greater than 80% forest land cover. Ten fire areas exhibited a significant positive pre and post-fire snow and EVI correlation. The majority of these fires had relatively high average elevations (1612 to 2291 m) and consisted of greater than 50% shrub, scrub, and grass land cover. The local scale analysis suggests that the condition of the vegetation (undisturbed vs. regenerating) and the soil texture in which it grows affects its response to winter snow. The low water holding capacity of coarse-textured soils and the short root-lengths of regenerating vegetation may result in greater dependence on snow as a water resource. Regionally, vegetation type and elevation may affect the vegetation's response to snow; short-rooted shrubs at higher elevations above the transient snow zone may be more dependent on snow as a water resource. These results suggest that the relationship between snow and vegetation is not constant, depending on the condition of the vegetation. Increases in wildfire activity and a reduction of snow in the future may impact successional trajectories in certain regions where vegetation may have historically relied on snowmelt to regenerate.