Investigating natural drivers of vegetation coverage variation using MODIS imagery in Qinghai, China

Abstract

The climatically sensitive Qinghai province of China has been recognized as a hotspot for studies on the feedbacks of terrestrial ecosystems to global climate change. Thus, investigating vegetation coverage and its natural drivers in Qinghai is an important focus of ecosystem research. On the basis of Moderate Resolution Imaging Spectroradiometer (MODIS) Enhanced Vegetation Index (EVI) time series data, we estimated the vegetation coverage in this region using the dimidiate pixel model. Trend analyses, correlations between meteorological parameters, changes in vegetation coverage, and the temporal and spatial relationships between soil texture and vegetation coverage were used to investigate the possible drivers of vegetation coverage variations. The results indicated that the reduction of vegetation coverage slowed down in the period from 2000 to 2012. Annual mean temperature was the main climatic driver of the total extremely low and low vegetation coverage areas in Qinghai, followed by the precipitation anomalies. The extremely low and low vegetation coverage areas were mainly distributed in regions with a mean annual relative air humidity of <40% and the spatial distributions of these two area types differentiated along the 200-mm rainfall contours. The total extremely low and low vegetation coverage areas were mainly characterized by sandy clay loam soil, followed by loamy sand and sandy soil. Regions with sandy loam or loam soil have the greatest risk of vegetation coverage reductions. Knowledge of vegetation coverage variation and its natural drivers in the ecologically fragile region of Qinghai can provide scientific support for managing environmental change and desertification.