Assessing Relationships between Forest Spatial Patterns and Fire History with Fusion of Optical and Microwave Remote Sensing

Abstract In this paper, we tested the use of active and passive sensor fusion for relating forest fire history to landscape spatial patterns. Principal Components Analysis (PCA) was implemented to combine Landsat Thematic Mapper (TM) and Shuttle Imaging Radar (SIR-C) data from October 1994. Resulting PCs were converted to landscape patch maps. Plots with known fire history were delineated using a fire atlas of the study area. These plots came from four fire history categories: unburned (nine plots), once burned (three plots), twice burned (three plots), and multiple burned (three plots). Landscape metrics were calculated for each plot, including a shape index, mean patch size, Shannon's Diversity Index, and Shannon's Evenness Index. Spearman's Rank Correlation Analysis was used to compare the patch map‐derived landscape metrics to fire history characteristics, such as average fire‐free interval and number of fire‐free years in different time periods. Results showed that landscape patterns derived from fused data were significantly (p < 0.05) related to fire history and typically performed better than SIR-C data (a greater number of significant correlations), but not as well as TM data.