Predicting post-wildfire overland flow using remotely sensed indicators of forest productivity

Wildfire can induce an increase in infiltration excess overland flow, which varies from barely detectible to extreme. Soil properties are an important contributor to this variability. Several studies found that a landscape's aridity (the balance of energy and water) is strongly correlated with an increased quantity of post-wildfire overland flow, with burnt forests of higher aridity producing higher peak flow compared to wetter forests. Related process-based studies suggested that this relationship can be explained by an interaction between inherent soil macroporosity (varies inversely with aridity) and fire induced water repellency; soils of higher macro-porosity maintain high infiltration rates despite fire-enhanced water repellence. Although the observed post-wildfire runoff/aridity relationship has proven useful in mapping hydrological risk, its transferability is potentially limited for areas where other soil formation factors have greater influence. Here we propose that measurements related to landscape productivity may provide a mappable landscape metric that is based on soil formation processes that influence post-fire infiltration capacity and may provide a more robust and transferable proxy of fire induced hydrologic change. To test this hypothesis, post-wildfire runoff data from three new experimental sites with a coastal climate influence were combined with previously published data collected inland across a gradient of aridity. The results showed that post-wildfire runoff was better correlated with measures of productivity than with aridity, supporting the hypothesis. It is therefore proposed that long-term average productivity may be a more robust and transferable proxy in estimating the magnitude of increase in overland flow after fire than aridity. This most likely results from a stronger causal link between productivity and soil macroporosity, though this was not measured in this study. As with aridity, this proxy is mappable at high resolution using climate and remotely sensed data, enabling the application of landscape productivity metrics when predicting post-wildfire hydrologic risk.