In order to explore the changing trend and response mechanism of karst carbon sinks under the background of climate change and rocky desertification control engineering, this paper estimates the temporal and spatial pattern of karst carbon sinks in Guangxi from 2005 to 2020 based on Guangxi lithologic geological data and meteorological data, and further The effects of climate change and rocky desertification control on karst carbon sinks are analyzed. The results show that from 2005 to 2020, the temporal and spatial differences of karst carbon sinks in Guangxi are significant, showing a trend of stable fluctuation and rising with an average annual increase of 8.1%. Precipitation has the most significant impact on the changes of karst carbon sinks while temperature and evapotranspiration have little effect. The responses to the control measures are different. As the most common control measures for rocky desertification, closing the mountain for management and protection has the most significant role in promoting carbon sequestration and increasing sinking with an average annual growth contribution rate of about 28%. 

Based on meteorological data, MODIS remote sensing image data and NCEP reanalysis data, the forward trajectories of dust storms in different seasons at different altitudes (500, 1000, and 1500 m) over the Mu Us Desert from 2000 to 2019 were simulated by HYSPLIT model, and was compared with the dust information identified by remote sensing image to determine the influence range of dust storm in the Mu Us Desert. The results show that: 1) In the past 20 years, the number of dust storms in the Mu Us Desert decreased slowly, with 143 dust storms months and 387 dust storms days. Since 2004, the frequency of dust storms has gradually decreased. 2) Seasonal climatic conditions have great influence on the moving path of dust storms in the Mu Us Desert. The frequency of dust storms is higher in spring and lowest in autumn. 3) The main areas affected by dust storms in the Mu Us Desert include northern Shaanxi Province, northern Shanxi Province, northern and central Hebei Province, northern and central Shandong Province, Beijing, Tianjin, eastern Inner Mongolia Autonomous Region and Northeast Three Provinces. The affected areas abroad include North Korea, North Japan and the border between Russia and China. The affected sea areas include Bohai Sea, Northern Yellow Sea and sea of Japan Northwest. The comparison between remote sensing image and HYSPLIT model provides an effective tool to verify and enhance the understanding of dust transport, diffusion and deposition process, so as to provide scientific reference for disaster prevention and mitigation decision-making and sand source control.

Based on Landsat TM data from 1992 to 2017, the second-class survey data of Jianshui County forest resources and the monitoring data of rocky desertification in Yunnan Province, with Mann-Kendall mutation analysis and trend analysis, the temporal and spatial variation characteristics of vegetation and the restoration of vegetation in different rocky desertification restoration models in Jianshui County of Yunnan Province in the past 25 years were discussed. The study found that during the 25 years, the vegetation cover of Jianshui County showed an overall upward trend with an average growth rate of 0.065%/10a. Since 2008, the growth rate increased from 0.077%/10a to 0.475%/10a. The area of vegetation cover in Jianshui County showed an increasing trend, no significant change, and the proportion of reduction trend was 55.32%, 34.44% and 10.24% respectively. For different restoration models, within the restoration time range of ten years, there was not much difference in the effect of vegetation restoration under the pure ecological forest mode and the mixed ecological forest mode. Within the restoration time range of 25 years, the mixed ecological forest mode vegetation restoration rate was obviously better than pure ecological forest restoration mode, and the advantages were mainly displayed in the later stage of restoration. In terms of species selection of pure ecological forest model, broad-leaved forests and alders had better restoration effects in the short term; in the long run, Yunnan pine and Che sang-tzu were better restoration species. In the mixed ecological forest model, the coniferous mixed forest restoration rate was the best both in the range of ten and twenty-five years, indicating mixed ecological forest model is a better mixed model. The study revealed the effects of different governance models on vegetation restoration, and provided a theoretical basis for optimizing

ecological restoration methods in rocky desertification areas.

In order to objectively evaluate the impact of land use on the ecological environment in faulted basins, and reflect the complex relationship between people and land at a deeper level from changes in ecosystem services, based on multi-time series land use data and terrestrial vegetation net productivity (NPP) in faulted basins, the temporal and spatial variation characteristics of NPP in faulted basins are clarified, and the characteristics of landlevel changes in land use change intensity over many years is revealed. The life cycle assessment method is used to quantitatively analyze the impact of land occupation and land conversion on the net productivity of terrestrial vegetation in rocky desertification areas, and explores the ecological benefits of rocky desertification control. The results show that 1) the change of net primary productivity (NPP) in most areas of the study area was not significant between 2000 and 2015, NPP showed significant changes in a small part of the region with an upward trend as the main trend; 2) among the six land types, the change of forest land is stable, the growth trend of construction land is active and strong, and the grassland reduction trend is active and strong; 3) the difference between the potential vegetation productivity of forest land and the existing NPP gradually increases from west to east, and has a consistent spatial distribution with the reduction of forest land occupation, indicating that forest land distribution will affect the net productivity of forest land vegetation; and 4) the natural vegetation growth in the northern part of the study area is good. The forest vegetation in the rocky desertification areas of Liupanshui City, Pu’an County and Luliang County in the eastern region has a certain degree of natural degradation, and the ecological restoration effect of forest land is general; 5) 80% of land conversion types in the research area makes NPP gain. The common way to makes NPP significantly improved is to convert grassland and cultivated land into forest land; the conversion of forest land to construction land causes the most damage to NPP.

In order to reveal the changing rules of regional water system connectivity under the development of urbanization, this paper takes Qingshui River, a tributary of the Yongding River, as an object, integrates remote sensing imagery, statistical data, and land use data, uses the comprehensive evaluation system for river connectivity, combining the longitudinal connectivity of rivers based on the cumulative effects of barriers and the lateral connectivity of rivers based on the effects of catchment fragmentation, and systematically studies the connectivity changes of water systems in the Qingshui River Basin from 1980 to 2015. The results of the study indicate that based on expert knowledge and combined with the actual situation of the Qingshui River, the barriers of the Qingshui River can be divided into four types: reservoirs, sluice gates, diffuse bridges, and river accumulations. During the study period, the number of barriers continues to increase. In 2000, the number of barriers increases by 10.4% in comparison with 1980 and increases by 23.9% in 2015. From 1980 to 2015, the longitudinal connectivity of the Qingshui River Basin as a whole shows an upward trend. The proportion of the watershed with poor vertical connectivity gradually decreases from 40% in 1980 to 14% in 2015. The areas with increased vertical connectivity are mainly concentrated in the eastern and central parts of the basin, and the connectivity in the southwestern part of the catchment area deteriorates. The overall change in river lateral connectivity is not significant, with the western part of the basin decreasing and the eastern part having improved. From 1980 to 2015, the comprehensive connectivity of the Qingshui River Basin shows an overall upward trend. The proportion of catchments with poor comprehensive connectivity is from 26% in 1980, 17% in 2000 and 17% in 2015, respectively. The areas with increased connectivity are mainly concentrated in the east of the basin, while the connectivity in some areas in the southwest of the basin is always poor. The study reveals the law of connectivity changes in the basin during the urbanization process in Beijing, Tianjin and Hebei, and can provide a scientific reference for regional flood prevention and disaster reduction and river ecological rehabilitation.