An emerging paradigm in forest ecology is that spatial heterogeneity in vegetation structure increases forest resilience to disturbance. However, resilience of systems spanning large geographic extents and comprising long-lived species is notoriously difficult to quantify. One component of resilience is resistance– how hard it is to change the system state. In mixed conifer forests of California’s Sierra Nevada mountain range, wildfire disturbance is most frequently associated with landscape-scale changes in system state. Wildfire severity describes the magnitude of these changes, with “high severity” signifying a stand-replacing event. Thus, a resistant forest system should generally experience lower wildfire severity when a fire inevitably occurs. Previous research has shown that a locally heterogeneous forest can modulate fire behavior and be less likely to burn at high severity. It is unclear whether this locally-observed phenomenon is a general feature of the forest system, and it isn’t resolved at what scale heterogeneity in forest structure is meaningful for resilience.
We use a new remote sensing approach to calculate wildfire severity across broad spatial extents as well as image texture analysis to ask: does spatial variability in forest structure make California mixed conifer forests more resilient by reducing the severity of wildfires when they occur?
Results/Conclusions
To generate the most comprehensive database of wildfire severity in the Sierra Nevada ecoregion, we apply our new approach of measuring remotely-sensed wildfire severity to the Fire and Resource Assessment Program fire perimeters (FRAP; http://frap.fire.ca.gov/projects/fire_data/fire_perimeters_index) between 1984 and 2016.
We find that canopy density (effect size = 2.9e-2) and potential annual heat load (effect size = 1.4e-2) have strong positive relationships with fire severity, and that 100-hour fuel moisture (effect size = -1.9e-2) and heterogeneity of forest structure (-2.4e-3) have significant negative relationships with fire severity. The effect size of heterogeneity of forest structure was approximately 10% that of canopy density– the strongest predictor. Further, a dampening interaction between forest heterogeneity and 100-hour fuel moisture (effect size = 8.3e-4) suggests that greater variability in forest structure may only affect wildfire severity under non-extreme fuel moisture conditions.
While the severity of a wildfire in any given place may be idiosyncratic and controlled by many variables, it is clear that heterogeneous forest structure generally makes mixed conifer forest in the Sierra Nevada more resistant to this inevitable disturbance. Because a resistant forest is a resilient forest, heterogeneity in forest structure may increase the probability of long-term forest persistence.