Wednesday, 17 January 2007: 4:30 PM
The relationship between drought and wildfire in the US
206A (Henry B. Gonzalez Convention Center)
There is an intuitive assumption that the more extreme the drought, the greater the potential for an active wildfire season. The 2006 fire season in Texas and Oklahoma is a recent good example of this, where extreme fire activity was attributed to extreme drought. However, it is appropriate to ask: “how much drought is needed to have fire, and does more intense drought really mean more or bigger fire?” This study is motivated in part to better understand the relationship of drought and fire business, defined here as ignitions and area burned, but also including prescribed burning occurrence - an important component of fuels management. Also, it is of interest to provide fire management with an indication of precipitation thresholds related to fire business. The ability to provide this information adds considerable value to basic precipitation indices. While there have been a small number of other studies relating precipitation indices to fire activity, our study is further motivated to examine how effectively a newly developed dataset of high-spatial resolution standardized precipitation indices (SPI) can be related to fire business. A recent project at the Climate, Ecosystem and Fire Applications (CEFA) at the Desert Research Institute (DRI) produced an 1895-2003 dataset of SPI indices for 1-, 3-, 6-, 12-, 24-, 36-, 48-, 60-, and 72-month cumulative time frames based on approximate 4km resolution PRISM precipitation data. This dataset is related to over 200,000 wildfire ignitions, area burned and prescribed burning across the US to assess quantitative thresholds of drought related to fire for the period 1985-2003. One initial result is that when all fires are considered, the median SPI index is negative (standardized precipitation deficit) for 1-, 3-, 6-, 9-, and 12-month period lengths, but is positive (standardized precipitation surplus) for longer period lengths. This is an indication of lag relationships between precipitation and fire (i.e., wet periods and likely fuel buildup prior to the drying period coincident with fire occurrence). Because of varying fuel types and seasonality, it is relevant to examine subsets of fire data and SPI values for varying times of the year, ignition cause, and geographic regions. This presentation provides results of this study and discusses the relevance of this information for decision-support in fire management.