Search

Changes in land use and landscapes have a direct impact on the regional eco-environment. It is of great importance to understand the change pattern of land use, landscapes, and their mechanism on the ecological quality, especially ecologically fragile areas. The northern sand-prevention belt (NSPB) is an important ecologically fragile area in China, which has a large influence on the ecological security of the entire country. Based on the land use data of the NSPB in 2000, 2010, and 2018, we studied the spatio-temporal characteristics of land-use change and change in landscape patterns. The ecological quality represented by the remote sensing-based desertification index (RSDI) was calculated using satellite images. The effects of land use and landscape patterns on RSDI were analyzed by geographic detector and geographically weighted regression. Important results include the following: (1) Land-use change in the study area was high during 2000–2010 but slower in 2010–2018. Grassland was the largest land-use type in the NSPB, and varied greatly in terms of total change and spatial location. The major change was the conversion between dense and moderate grass, with 64,860 km2 of dense grass turning into moderate grass, and 48,505 km2 changing the other way. (2) Among the four landscape metrics, patch density, area-weighted mean fractal dimension, and edge density increased, whereas the aggregation index decreased, which indicated that the landscape was developing towards heterogeneity, fragmentation, complexity, and aggregation. Spatially, the landscape metrics presented a strip distribution in the east of the NSPB. (3) The effects of various land-use types on ecological quality, from high to low, were unused land, woodland, dense grass, cropland, moderate grass, built-up land, sparse grass, and waterbody. The areas where the ecological quality was greatly affected by the landscape patterns were concentrated in the agro-pastoral ecotone and the forest-steppe ecotone. The results of this study reveal the trends of land use and landscape patterns in the NSPB over 18 years and can help to understand their mechanism on ecological quality, which is of significance for the management of this area. • LUC in the NSPB was active from 2000 to 2010 and weakened from 2010 to 2018. • The landscape was developing towards heterogeneity, fragmentation, complexity, and aggregation. • The effects of land use and landscape on ecological quality were studied by GD and GWR. [ABSTRACT FROM AUTHOR]

• It was found that the agricultural land expanded significantly, desert reduced from 1993 to 2019 in the oasis region in Hotan River Basin, China. • The key indicators of the coupling system of vegetation–climate, soil-adjusted vegetation index, albedo, surface temperature, and evapotranspiration were calculated and mapped their spatial-temporal dynamics. • We investigated how the different landscape patterns altered the interaction between vegetation and climate in the desert oasis ecosystem and quantified the effects with structural equation model. In the arid region, the oasis is characterized by a desert substance and a mosaic of vegetation. The biophysical processes and interactions between vegetation and the local climate in this kind of region are determined by the macroscopic structure of the ecosystems, i.e. landscape patterns. To understand how these landscape patterns impact regional hydro-heat coupling across space and evolve over time, we utilized remote sensing observational data and methods to examine the relationships among these factors. In this case, we focused on the oasis along the upstream of the Hotan River Basin in the Taklamakan Desert in Xinjiang of western China and employed the satellite imagery datasets of Landsat from 1993 to 2019 to investigate the dynamics of vegetation–climate factors. Based on the land use and cover change datasets, the landscape pattern metrics, including patch density (PD), contagion index (CONTAG), fractal dimension (FRAC), were calculated to measure the landscape features on the different aspects, i.e. the fragmentation, the connectivity, and the complexity. With the algorithm of land surface energy balance, the land surface indicators, including the soil-adjusted vegetation index (SAV I), albedo, surface irradiance temperature (T s), and evapotranspiration (ET), were calculated to represent the key process in the interaction of vegetation–climate. The temporal-spatial dynamics of the landscape patterns and the vegetation–climate metrics were mapped and demonstrated in a quantitative manner. The statistical results revealed that during the past decades, the agricultural land in the study area had significantly increased by 17 %. Grassland and shrubs had also expanded, while the desert area decreased from 82.57 % to 78.82 % of the total area, with an overall reduction rate of 1.4 %/10a. It was found the study area was getting warmer and dryer based on the general trends of T s and ET observed during the period of 1993–2019. The agricultural land had the highest PD and CONTAG , and the lowest FRAC. The agricultural land had the lowest T s and the highest ET. The results of structural equation model identified the decoupling effects of PD and CONTAG on the regional hydro-heat environment, while confirmed that FRAC had positive impact on the coupling between T s and ET. Our study bridged the landscape pattern with the regional vegetation–climate interaction and provided the suggestions for the landscape planning and management for a more sustainable arid region. [Display omitted] [ABSTRACT FROM AUTHOR]