Abstract:The paper assesses spatio-temporal patterns of land surface temperature (LST) and fire severity in the Las Hurdes wildfire of Pinus pinaster forest, which occurred in July 2009, in Extremadura (Spain), from a time series of fifteen Landsat 5 TM images corresponding to 27 post-fire months. The differenced Normalized Burn Ratio (dNBR) was used to evaluate burn severity. The mono-window algorithm was applied to estimate LST from the Landsat thermal band. The burned zones underwent a significant increase in LST after fire. Statistically significant differences have been detected between the LST within regions of burn severity categories. More substantial changes in LST are observed in zones of greater fire severity, which can be explained by the lower emissivity of combustion products found in the burned area and changes in the energy balance related to vegetation removal. As time progresses over the 27 months after fire, LST differences decrease due to vegetation regeneration. The differences in LST and Normalized Difference Vegetation Index (NDVI) values between burn severity categories in each image are highly correlated (r = 0.84). Spatial patterns of severity and post-fire LST obtained from Landsat time series enable an evaluation of the relationship between these variables to predict the natural dynamics of burned areas.

Abstract. This paper explores spatial-temporal dynamics in fire regime features, such as fire frequency, burnt area, large fires and natural-and human-caused fires, as an essential part of fire regime characterization. Changes in fire features are analysed at different spatial -regional and provincial/NUTS3 -levels, together with summer and winter temporal scales, using historical fire data from Spain for the period 1974-2013. Temporal shifts in fire features are investigated by means of change point detection procedures -Pettitt test, AMOC (at most one change), PELT (pruned exact linear time) and BinSeg (binary segmentation) -at a regional level to identify changes in the time series of the features. A trend analysis was conducted using the Mann-Kendall and Sen's slope tests at both the regional and NUTS3 level. Finally, we applied a principal component analysis (PCA) and varimax rotation to trend outputs -mainly Sen's slope values -to summarize overall temporal behaviour and to explore potential links in the evolution of fire features.Our results suggest that most fire features show remarkable shifts between the late 1980s and the first half of the 1990s. Mann-Kendall outputs revealed negative trends in the Mediterranean region. Results from Sen's slope suggest high spatial and intra-annual variability across the study area. Fire activity related to human sources seems to be experiencing an overall decrease in the northwestern provinces, particularly pronounced during summer. Similarly, the Hinterland and the Mediterranean coast are gradually becoming less fire affected. Finally, PCA enabled trends to be synthesized into four main components: winter fire frequency (PC1), summer burnt area (PC2), large fires (PC3) and natural fires (PC4).

No abstract

View related articles View Crossmark data Citing articles: 6 View citing articles Modelling temporal variation of fire-occurrence towards the dynamic prediction of human wildfire ignition danger in northeast Spain Yago Mart ın a , Mar ıa Z uñiga-Ant on d and Marcos Rodrigues Mimbrero b,c