News Release

Measured soil moisture improves wildfire prediction

Use of high quality soil moisture data proposed

Peer-Reviewed Publication

American Society of Agronomy

Wildfire in OK

image: Smoke and pyrocumulus cloud from a 2,800 ha fire that destroyed 23 homes near Stillwater Oklahoma in August of 2012. FAW at the time of the fire was only 0.10, or 10 percent of its possible maximum value, and KBDI was 744. This extremely low soil moisture desiccated herbaceous and woody vegetation (photo foreground), leading to increased wildfire probability. view more 

Credit: J.D. Carlson.

Despite the known connection between soil moisture and wildfire danger, measured soil moisture is conspicuously absent from the list of variables commonly used in wildfire danger assessments. Instead, assessments enlist the help of the decades-old Keetch-Byram Drought index (KBDI), a soil moisture surrogate calculated from precipitation and estimated evapotranspiration. In the absence of measured soil moisture data, the reliance upon KBDI as a surrogate to assess wildfire danger is understandable. But is the continued reliance on KBDI justified when high quality soil moisture data are available?

According to recent work published in Soil Science Society of America Journal, the answer is no. Researchers in Oklahoma compared the relationships of measured soil moisture, as fraction of available water capacity (FAW), and KBDI with wildfire occurrence statewide and found that FAW consistently outperformed KBDI. Soil moisture conditions conducive to large growing-season wildfires were more narrowly defined by FAW, regression models based on FAW correctly classified days with large fires at a higher rate, and FAW provided earlier warning of extreme wildfire potential. Based on these findings, the authors call for the replacement of KBDI with FAW in growing-season wildfire danger assessments in Oklahoma and regions with similar climate and vegetation types.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.