Abstract
Understanding the spatial and temporal dynamics of vegetation is essential in drylands. In this paper, we evaluated three vegetation indices, namely the Normalized Difference Vegetation Index (NDVI), the Soil-Adjusted Vegetation Index (SAVI) and the Enhanced Vegetation Index (EVI), derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) Surface-Reflectance Product in the Xinjiang Uygur Autonomous Region, China (XUAR), to assess index time series’ suitability for monitoring vegetation dynamics in a dryland environment. The mean annual VI and its variability were generated and analyzed from the three VI time series for the period 2001–2012 across XUAR. Two phenological metrics, start of the season (SOS) and end of the season (EOS), were detected and compared for each vegetation type. The mean annual VI images showed similar spatial patterns of vegetation conditions with varying magnitudes. The EVI exhibited high uncertainties in sparsely vegetated lands and forests. The phenological metrics derived from the three VIs are consistent for most vegetation types, with SOS and EOS generated from NDVI showing the largest deviation.
Highlights
Drylands cover nearly 40 percent of global land surface [1,2]
Three vegetation indices derived from Moderate Resolution Imaging Spectroradiometer (MODIS) data were assessed in vegetation dynamic monitoring in the XUAR, China
The Enhanced Vegetation Index (EVI) time-series captured the largest inter-annual variability for the XUAR from 2001 to 2012, with 6.28% of the entire area showing variability higher than 30%
Summary
Rainfall is low and highly variable in these lands, which results in dramatic variation of temporal development and growth cycle of vegetation from year to year. Many dryland ecosystems have experienced increased threats from climate change and human-induced activities over recent decades [3,4,5,6]. Knowledge of the spatial and temporal dynamics of vegetation is essential to understand the impacts of climate change and anthropic activities on dryland ecosystems. With the capability for large spatial coverage and frequent observations, satellite data have played an important role in monitoring vegetation dynamics [7,8,9]. For dryland vegetation activity monitoring, time series of vegetation indices have often been used to detect variability of vegetation activity [12,13,14] and changes of vegetation phenology [15,16]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
