Abstract
Frequent droughts in a warming climate tend to induce the degeneration of vegetation. Quantifying the response of vegetation to variations in drought events is therefore crucial for evaluating the potential impacts of climate change on ecosystems. In this study, the standardized precipitation index (SPI) was calculated using the precipitation data sourced from the China Meteorological Forcing Dataset (CMFD), and then the drought events in southern Tibet from 1982 to 2015 were identified based on the SPI index. The results showed that the frequency, severity, and intensity of drought events in southern Tibet decreased from 1982 to 2015, and the highest frequency of drought was found between 1993 and 2000. To evaluate the impact of drought events on vegetation, the vegetation characteristic indexes were developed based on the normalized difference vegetation index (NDVI) and the drought characteristics. The assessment of two drought events showed that the alpine grasslands and alpine meadows had high vegetation vulnerability (AI). The assessment of multiple drought events showed that responses of vegetation to drought were spatially heterogeneous, and the total explain rate of environmental factors to the variations in AI accounted for 40%. Among the many environmental factors investigated, the AI were higher at middle altitudes (2000–3000 m) than low altitudes (<2000 m) and high altitudes (3000–4500 m). Meanwhile, the silt soil fraction in the upper soil layer (0–30 cm) had the greatest positive correlation with AI, suggesting that areas with a high silt soil fraction were more sensitive to drought. The relative contribution rates of environmental factors were predicted by a multivariate linear regression (MLR) model. The silt soil fraction was found to make the greatest relative contribution (23.3%) to the changes in AI.
Highlights
Drought is considered to be one of the most complex and severe natural disasters affecting water resources, agricultural production, and natural ecosystems [1,2]
Based on the relationships between R-drought severity (DS) and R-normalized difference vegetation index (NDVI) of different vegetation types, we found that the mean and median of the R-NDVI of alpine grassland increased with the increase in the R-DS, suggesting that alpine grassland was more sensitive to drought (Figure 9a)
This study evaluated the space–time characteristics of drought events in southern Tibet based on a three-dimensional clustering algorithm
Summary
Drought is considered to be one of the most complex and severe natural disasters affecting water resources, agricultural production, and natural ecosystems [1,2]. Several methods have been used to characterize drought including the threshold method [9], the runs theory [10,11], the empirical orthogonal method [12,13], the wavelet analysis [14], and principal component analysis [15] These methods usually quantify the characteristics of drought based on the time evolution in a fixed region or the spatial pattern over a fixed duration. This could limit the comprehensive understanding the spatio-temporal structure of drought events [16]. The application of this method could improve our understanding of the development of drought events, the changes in drought characteristics, and the potential effects of drought on vegetation health
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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
