Exploring Hydrologic Responses to Different Wildfire Spatial Patterns Through the Lens of Computational Modeling

Severe wildfire disturbances are becoming increasingly common in high-elevation forests of the western United States. These fires alter watershed hydrologic processes, threatening critical downstream water resources and aquatic ecosystems. However, watershed-scale postfire hydrologic responses and water balance changes are highly uncertain. While postfire effects on individual processes such as runoff, infiltration, evapotranspiration, and snow dynamics are relatively well known, the role of wildfire spatial patterns in governing hydrologic connectivity and interactions between water balance components is poorly understood due to challenges associated with measuring and comparing fires at large scales. This thesis aims to examine pattern-related postfire interactions between various hydrologic processes using computational modeling. Our goals are to identify the primary underlying relationships and to provide a methodological approach upon which a more comprehensive understanding of postfire watershed hydrology can be built. In Chapter 1, we briefly summarize the current knowledge base regarding postfire hydrology and introduce how hydrologic computational modeling has been used for postfire applications. Chapter 2, written as a manuscript, details the suite of modeling experiments used to explore the effects of wildfire spatial patterns on an idealized, snow-dominated mountain watershed. We used Neutral Landscape Model (NLM) algorithms to generate 150 fire mosaics with varying levels of aggregation and used a physically-based, distributed model to simulate each mosaic for a full water year. We found that each mosaic created a unique network of ow paths between the burned areas and the watershed outlet and that the size of the network controlled the timing of watershed discharge and soil water storage due to an infiltration capacity gradient between burned and unburned sites. Each fire mosaic generated the same amount of runoff from within the burned areas, but longer flow path networks resulted in more infiltration outside of the fire boundaries. However, because there was enough snow in the watershed to fully saturate the soil in every location, there was little difference in total annual discharge. While these results may be specific to snowmelt-dominated systems, they highlight the importance of considering the entire disturbance flow path network when evaluating watershed-scale postfire hydrologic responses