Abstract
Climate change and a long history of fire suppression have contributed to an increase in the frequency and extent of high-severity wildfires across the western United Sates. Forest treatments such as mechanical thinning and prescribed burning are commonly used strategies to reduce wildfire severity and improve forest resilience. Because of the importance of vegetation cover on the timing and quantity of snow accumulation and melt, both forest treatment and wildfire can affect snowpack behavior in important ways. We developed SNOW-17(VEG), an adaptation of the widely used and validated SNOW-17 temperature index model that incorporates vegetation structure. Statistical evaluation of SNOW-17(VEG) indicated strong agreement between simulated and observed snowpack conditions in a seasonal snowpack when comparing dense forest to a high severity burn scar in Colorado (NSE: 0.95, 0.94; Pbias: 3.4, -2.6; R: 0.97, 0.94). In an ephemeral snowpack in New Mexico, while performance was weaker in terms of absolute magnitude, the model reproduced key differences in snow accumulation and melt timing among dense forest, thinned forest, and high severity burn scar conditions (NSE: 0.84, 0.86, 0.92; Pbias: 56.3, 57.7, 26.7; R: 0.91, 0.90, 0.95). We also developed an automated methodology for delineating vegetation-density zones, integrating stochastic temperature and precipitation scenarios, and operationalizing the model through a graphical interface. Model calibration shows strong agreement with observed snowpack, especially in the snow-dominated portions of the watershed (NSE: 0.96 , Pbias: 8.8, R: 0.97). Projected increases in temperatures reduced snowpack accumulation and accelerated snowmelt across simulation scenarios. SNOW-17(VEG) provides a low-data framework for evaluating the influence of forest management and wildfire disturbance on snow accumulation and melt in mountain watersheds with available temperature and precipitation observations.
Comments
There are currently no comments
New Comment