Seasonal variation of trends and homogeneity of extreme forest fire danger index (FFDI) value

Abstract

Abstract The traditional approach of potential fire danger is determined using the forest fire danger index (FFDI). Seasonal variability of the influential variables has significant impacts on the magnitude of extreme FFDI values. In this study, the severity of FFDI is determined using different statistical approaches following various hypotheses. The application of statistical analysis requires the data to be obtained from the same population distribution. The main objective of this research was to investigate the seasonal variation of homogeneity, trend, and change points of extreme FFDI. McArthur Forest Fire Danger index technique was employed in estimating the daily forest fire danger index for 15 stations located in Tasmania, Australia. Seasonal maximum FFDI values were extracted from the estimated daily FFDI values. Seasonal variation of trend analysis was performed after performing a homogeneity test. Non-parametric Mann- Kendall trend test was applied to investigate the trend of the seasonal extreme data sets. The scale of the trend was investigated employing the commonly used Sen’s slope. The results of the analysis reveal that there are temporal and spatial variations of the increasing FFDI values across Tasmania due to climate change. An extreme data series that accepts one statistical hypothesis in one theory may reject the same hypothesis in another theory. Seasonal variation in the future trend of FFDI will help to improve the management of natural resources and sustainable planning in the region. The adoption of appropriate construction strategies in extreme fire regions can be determined from this study.