Weather type classification of wildfire ignitions in the central Gulf Coast, United States
Additional Document Info
2014 Royal Meteorological Society. Limited research has been performed examining the relationships between southeast US wildfire and weather type patterns using modern techniques and data sets. The purpose of this article is to examine the relationship between wildfire ignitions in the central Gulf Coast, United States and weather type occurrences in an effort to identify regional patterns associated with wildfire ignitions. Wildfire occurrence data are obtained for nine national forests in the central Gulf Coast for the period 1970 to 2011. The Spatial Synoptic Classification scheme (SSC) is used to classify daily weather type variations for each national forest. Years with high numbers of ignitions experience statistically higher numbers of dry tropical (DT) weather types than years with low numbers of ignitions in five of nine national forests. These differences range from 17 to 160% more DT days during periods of high ignitions. Statistically significant positive Spearman rank correlations exist between annual number of ignitions and annual number of DT days in a given forest. While understanding such relationships on an annual scale is useful, it says nothing about the actual timing of weather types leading up to an ignition. An analysis of weather type activity 30, 90, and 180 days prior to ignitions reveals that the median ratio of DT days to moist moderate (MM) and moist tropical (MT) days is statistically higher for periods leading up to a fire. Positive phases of the North Atlantic Oscillation (NAO) result in an increased number of DT days. This confirms previous studies that have suggested the strength of the westerlies increases during the positive phase of the NAO, driving DT weather types eastward from their southwest US/Mexico source region. We can therefore also conclude that the positive phase of the NAO is linked with increased wildfire activity in the central Gulf Coast, United States.