Temporal and spatial monitoring of post-fire forest dynamics using time-series MODIS data

摘要

Forest fire is one of the dominant disturbance regimes, especially in boreal forests. Identifying post-fire forest dynamics is crucial to both ecological research and forest monitoring and management. In this study, taking the most serious forest fire in the history of P. R. China as an example, we proposed to monitor the temporal and spatial characteristics of post-fire forest dynamics using MODIS Land Cover Type data and investigate their responses to MODIS Vegetation Indices values. The results indicated that the “Forests” area increased from 50.29% in 2001 to 89.35% in 2012 while that of “Cropland/Natural vegetation mosaic” decreased from 30.46% to 5.94%. The regenerated forest mainly distributed in the severe and continuous burned area. The four types of forests and three classes of land cover vegetation can be separated by Normalized Difference Vegetation Index (NDVI) while Enhanced Vegetation Index (EVI) is not applicable for the separation between forests and other vegetation types. Both the NDVI and EVI of burned forests were significantly higher than that of unburned forests, suggesting the burned forests were more productive and wit active herbaceous and new growth which were much greener than mature unburned forests that were in more of steady state, but old.