Using variable column precipitable water as a predictor for large fire potential

Drought and extended periods of low values of relative humidity have long been known to be precursors to large wildland fires. However, little focus has been given to the entire tropospheric moisture column, or subsets of this column, as related to wildland fires. This study utilizes a precipitable water database from the upper-air observing site at Rapid City, South Dakota— in conjunction with the wildland fire database for western South Dakota—to examine large wildland fires. Comparisons were made between large fire occurrence/growth and deviation from average values of precipitable water over various atmospheric layers. Additionally, precipitable water was then introduced as a moisture variable in a Haines Index-type formulation, and resulting data were compared to the original Haines Index in an effort to increase the robustness of the fire potential index. Preliminary results indicate that using a layer precipitable water deviation is, at times, more relevant to wildland fire potential than the dewpoint depression at a single level.