RT info:eu-repo/semantics/article
T1 Effects of wildfires on environmental variability: a comparative analysis using different spectral indices, patch metrics and thematic resolutions
A1 Lozano, F. Javier
A1 Suárez Seoane, Susana
A1 Luis Calabuig, Estanislao
A2 Ecologia
K1 Ecología. Medio ambiente
K1 Diversity
K1 Richness
K1 Evenness
K1 Landsat
K1 Disturbance
K1 Post-fire recovery
AB Knowledge on environmental variability and how it is affected by disturbances is crucial for understanding patterns of biodiversity and determining adequate conservation strategies. The aim of this study is to assess environmental variability in patches undergoing post-fire vegetation recovery, identifying trends of change and their relevant drivers. We particularly evaluate: the value of three spectral indices derived from Landsat satellite data [Normalized Burn Ratio (NBR), Normalized Difference Vegetation Index (NDVI) and Wetness Component of the Tasseled Cap Transformation (TCW)] for describing secondary succession; the effectiveness of three metrics (diversity, evenness and richness) as indicators of patch variability; and how thematic resolution can affect the perception of environmental variability patterns. While the system was previously characterised as highly resilient from estimations of vegetation cover, here we noted that more time is required to fully recover pre-fire environmental variability. Using mean diversity as indicator of patch variability, we found similar patterns of temporal change for the three spectral indices (NBR, NDVI and TCW). Analogous conclusions could be drawn for richness and evenness. Patch variability, measured as diversity, showed consistent patterns across thematic resolutions, although values increased with the number of spectral classes. However, when the variance of diversity was plotted against thematic resolution, different scale dependencies were detected for those three spectral indices, yielding a dissimilar perception of patch variability. In general terms, NDVI was the best performing spectral index to assess patterns of vegetation recovery, while TCW was the worst. Finally, burned patches were classified into three classes with similar trends of change in environmental variability, which were strongly related to fire severity, elevation and vegetation type.
PB Springer
YR 2019
FD 2019-04-17
LK http://hdl.handle.net/10612/10305
UL http://hdl.handle.net/10612/10305
NO Landscape Ecology, 2010,  vol. 25, n.5
NO P. 697-710
DS BULERIA. Repositorio Institucional de la Universidad de León
RD 17-may-2024