China’s deserts greening and response to climate variability and human activities
pmc-PMC8405022.pdf · Phiên bản 1 · 20 trang
The NDVI is an effective indicator of conditions affecting vegetation growth and can reflect the impacts of environmental and anthropogenic factors. Previous studies have paid more attention to NDVI dynamics and influential factors in different regions (e.g., Hungary [38], Central Asia [21], and Canada [39]), as well as different land cover types (e.g., grassland [40], forests [41], and cropland [42]). By contrast, few studies have considered NDVI variations and driving factors at both the interannual and intra-annual scales in regions with low vegetation cover, especially arid zones and deserts, which are more sensitive to outside factors. Combined with existing studies, it has been found that vegetation in most arid zones around the world has shown a gradual recovery trend recently [8,43]. As a country with vast desert areas, China has been under the attack of wind and sand for years, the government attached great importance to it and made great practices in desert control, especially since the 21st century, ecological protection has been highly emphasized in China. However, the detailed studies on the vegetation dynamics of all deserts in China and the extent to which they are affected by climatic factors and human activities are still lacking. This study aims to analyze NDVI variations and the influential factors in China’s deserts from 2000 to 2017. Specifically, this study seeks to (1) use monthly, seasonal, and annual data to explore trends in the NDVI and climatic factors temporally and spatially; (2) analyze the
correlations between the NDVI and temperature and precipitation at the monthly, seasonal, and annual scales to better understand the relationships between the NDVI and climatic fac- tors and to compare their influence at different time scales; and (3) use residual analysis to dis- tinguish greening regions caused by human activities from 2000 to 2017. Materials and methods Study area Deserts in China are widely distributed in the north, especially northwestern China, which is located between 75˚-115˚E and 35˚-47˚N, with an area of approximately 0.98 million km2. Most part of the deserts is distributed in arid areas, and the remaining part is distributed in semi-arid areas. The annual average temperature of China’s deserts ranges from -10˚C to 15˚C, and the annual accumulated precipitation in most of the areas is less than 200 mm and mostly distributed in summer. Eight deserts are located in the study area: the Gurbantunggut Desert, Taklimakan Desert, Kumtag Desert, Qaidam Basin Desert, Badain Jaran Desert, Teng- ger Desert, Ulan Buh Desert, and Kubuqi Desert (Fig 1(a)). Data sources Distribution of deserts. The desert distribution data used in this study were mainly obtained from the 1:100000 desert (sand) distribution dataset in China [44]; the data were interpreted, extracted, and edited from Thematic Mapper (TM) digital images and the China land-use map for 2000. The range of deserts was corrected by using a topographic map of China and the map of desert distribution at 1:2000000 in China [45].
NDVI data. In this study, the NDVI from MOD13A2 was used as an indicator of vegeta- tion activity. The NDVI dataset was sourced from the MODIS (Moderate Resolution Imaging Spectroradiometer) vegetation index product data on the NASA website (https://lpdaac.usgs. gov/products/mod13a2v006/) [46] and had already been corrected for the effect of atmo- spheric gases, thin cirrus clouds, and aerosols, with a spatial resolution of 1 km and a temporal resolution of 16 days. A total of 411 images were used for the period between December 2017 and February 2000. Data processing, such as image merging, map projection transformation, resampling, and masking by desert areas, was performed on the original downloaded NDVI data. The maximum value composite (MVC) method was applied to calculate the monthly and PLOS ONE China’s deserts greening and response to climate variability and human activities PLOS ONE | https://doi.org/10.1371/journal.pone.0256462 August 30, 2021 3 / 20