Publications

Yang, J; Pan, SF; Dangal, S; Zhang, BW; Wang, SY; Tian, HQ (2017). Continental-scale quantification of post-fire vegetation greenness recovery in temperate and boreal North America. REMOTE SENSING OF ENVIRONMENT, 199, 277-290.

Abstract
Knowledge and information of post-fire vegetation recovery are essential for our understanding of ecosystem stability and resilience in response to present and future disturbances. Although previous studies have examined the post-fire vegetation recovery at landscape and regional levels, the lack of continent-wide analysis limits our understanding of fire effect on vegetation recovery across ecoregions and climate zones. With advanced technology in satellite observations as well as large-scale database development, it is possible for us to explore the large-scale effect of concurrent and lagged responses of terrestrial ecosystems to fire disturbances. In this study, we investigated the recovery of vegetation greenness in the early post-fire period in North America by integrating two national-level fire perimeter datasets and multiple Moderate Resolution Imaging Spectroradiometer (MODIS) products. We developed a new algorithm to estimate fire effect on the Normalized Difference Vegetation Index (NDVI) by comparing fire pixels with their surrounding unburned pixels. Our results showed the fire effect on needleleaf trees was stronger than on other vegetation types, and the high-latitude ecoregions had larger NDVI reduction and longer recovery time than the mid-latitude ecoregions. We also found the variations in vegetation composition and fire severity can explain a large fraction of post-fire vegetation greenness recovery. Our satellite-based analysis showed a longer NDVI recovery time than the existing estimates, which indicates a stronger fire effect on ecosystems than previously thought. Findings in this study indicate that more sophisticated parameterization schemes of fire severity and post-fire vegetation recovery are needed for the vegetation-fire models to better simulate the terrestrial carbon cycling and climate-ecosystem feedbacks.

DOI:
10.1016/j.rse.2017.07.022

ISSN:
0034-4257