Response of vegetation to different time-scales drought across China: Spatiotemporal patterns, causes and implications Academic Article uri icon

abstract

  • © 2017 Elsevier B.V. Based on Normalized Difference Vegetation Index (NDVI) and Standardized Precipitation Evapotranspiration Index (SPEI), we investigated vegetation response to different time-scales drought across different vegetation types and homogeneous clusters in China, by annual maximum Pearson correlation (R max ) and the corresponding time-scales of drought. Results showed that: (1) 8 subregions with homogeneous climate-vegetation conditions were identified using Fuzzy C-Means algorithm; (2) SPEI and NDVI's annual R max were in significantly positive correlation in most regions of China, indicating that vegetation biomass were influenced mainly by the spatiotemporal characteristics of the water availability. The southeastern Yangtze River basin and the lower Pearl River basin are dominated by abundant precipitation, and vegetation is not sensitive to droughts in these regions. The northeastern Heilongjiang province, the Changbai Mountains and western Sichuan province are characterized by weak NDVI versus SPEI relations, indicating a relatively small effect of drought on vegetation; (3) The effects of annual average water balance, annual average annual precipitation, annual average effective accumulative temperature, and annual average daily sunshine hours on the NDVI versus SPEI correlation show that the annual average water balance is the key factor behind the change of vegetation vigor. It can therefore be concluded that the change of water availability is the key factor behind the change of vegetation activity and biomass. Regional precipitation or water balance was significantly related to the correlation between SPEI and NDVI. Vegetation in the regions with longer sunshine hours is more sensitive to droughts. In general, the sensitivity of grassland to droughts is the largest, followed by the sensitivity of shrubs and forests to droughts.

altmetric score

  • 0.25

author list (cited authors)

  • Zhang, Q., Kong, D., Singh, V. P., & Shi, P.

citation count

  • 83

publication date

  • May 2017