Professor North Carolina State University, United States
Abstract: The effectiveness of fuel reduction treatments on reducing fire size and severity depends upon the extent to which severe wildfires are driven by fuel accumulation, versus climatic factors such as fuel moisture or wind speed. It is likely that the same processes do not equally affect different regions, and that fire size and fire severity might be associated with different drivers. In this study, we assessed the relative contribution of weather and fuel characteristics to fire spread and severity, and explored how these relationships varied temporally and spatially. We used polygons of daily fire size to estimate probabilities of fire spread, which we modeled as a function of weather and proxies for fuel loads and fuel moisture. From remote sensing and climate products, we obtained data for over 4 million cells at a daily temporal resolution and 180-m spatial resolution. Our preliminary results find that fire spread (i.e., the probability of spreading from one cell to a neighboring cell) is largely dependent upon effective windspeed, with fire spread about a third as sensitive to fine fuels as to windspeed. There were also large interactions between fuel load and weather: fire spread was more sensitive to fuel load under drier conditions. There was extremely large spatial variation in the importance of different factors; for example, fire spread in the Tahoe/Central Sierra region was very sensitive to fire weather index, while other regions, such as the eastern slope of the Sierra, were comparatively insensitive to fire weather. We also found that fuel has become increasingly important over time, perhaps suggesting a nonlinear response to fuel loads. Our results highlight the difficulty of generalizing fire behavior – and thus fire management – across broad ecoregions. The challenges of managing resilient forests in the coming decades likely require approaches tailored to the ecology and history of individual forests.
