China’s deserts greening and response to climate variability and human activities

Fig 4 also shows the relative frequency distributions of the greening and browning trends over longitude and latitude. The area with greening trends was positively correlated with the desert area. Although the area with browning trends was very small, it was distributed over longitudes and latitudes with large desert areas. The NDVI trends differed for different deserts (Table 1). The proportion of greening in the Badain Jaran Desert was the highest among the deserts at 88.26%, which increased signifi- cantly, followed by the Kumtag Desert (75.26%) and Tengger Desert (70.07%). While the pro- portion of greening in the Gurbantunggut Desert was only 31.22%, the lowest of the deserts. Fig 4. Spatial distribution of greening and browning trends and statistical results based on the longitude and latitude of China’s desert areas from 2000 to 2017. Gray areas represent desert regions with no significant trends at the 5% confidence level. The green and brown areas indicate that the pixel has a significant greening and browning trend, respectively. https://doi.org/10.1371/journal.pone.0256462.g004 Table 1. The NDVI trends in different deserts. Desert Significant Increase (%) Increase but not significant (%) Significant decrease (%) Decrease but not significant (%) Gurbantunggut Desert 31.22 62.50 0.17 6.12 Kumtag Desert 75.26 23.30 0.07 1.38 Qaidam Basin Desert 63.60 30.46 1.25 4.68 Kubuqi Desert 60.31 33.67 0.67 5.35 Badain Jaran Desert 88.26 11.14 0.05 0.55 Ulan Buh Desert 40.41 50.40 0.30 8.89 Tengger Desert 70.07 27.67 0.11 2.15

Taklimakan Desert 65.82 29.69 0.84 3.65 https://doi.org/10.1371/journal.pone.0256462.t001 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 8 / 20