| Land desertification is one of the four most serious natural disasters of human beings,which not only affects the current human production and life,but also seriously threatens the sustainable development of ecological environment and social economy.The scientific decision support for the realization of regional desertification control can be provided by researching on the dynamic changes of desertification and its driving factors.Northern China is mainly located in arid and semi-arid areas,where land desertification is widely distributed.In this paper,the distribution and dynamic changes of desertification in different periods are analyzed in the whole northern region.Firstly,Normalized Vegetation Index(NDVI)data are used to estimate the vegetation coverage based on the Pixel Dichotomy Model as an evaluation indicator of land desertification.In order to avoid the deviation of NDVI data in a single year,the spatial-temporal distribution and dynamic changes of land desertification in three periods of 1991-2000,2001-2010 and 2011-2019 were analyzed.Secondly,analyzed the dynamic change of vegetation and the impact of human activities on land desertification in the study area after a series of ecological restoration projects such as "Grain for Green Program","Natural Forest Protection Project","Beijing-Tianjin Sandstorm Source Treatment Project".Finally,the Long Short-Term Memory(LSTM)neural network is used to analyze the future desertification characteristics of Horqin area with the data of future temperature and precipitation.The main results are summarized as follows:(1)The spatial and temporal dynamics of land desertification in northern China from 1991 to 2019.In terms of spatial distribution,the northwestern part of the study area is covered by a large area of desertification land,mainly distributed in Xinjiang,Qaidam Basin,central and western Inner Mongolia Autonomous Region,northwestern Gansu and the Loess Plateau.The severity of land desertification gradually increased from the non-desertification area in the southeast to the northwest area.In terms of dynamic changes,the area of desertification in the study area in 1991-2000 accounted for 63.01%,and the area of non-desertification accounted for 36.91%;the area of desertification in 2000-2010 accounted for 59.74%,and the area of non-desertification accounted for 40.23%;the area of desertification in 2011-2019 accounted for 55.51%,and the area of non-desertification accounted for 44.46%;the total area of desertification in the study area display a trend of decreases year by year.Significant changes occurred in the proportion of land desertification in different degrees.Form 1991-2000 to 2011-2019,the extremely severe desertification increased from 21.72%to 24.25%;and severe desertification,moderate desertification,slight desertification decreased from 19.04%,9.45%,12.80%to 13.74%,8.06%,9.46%,respectively;non-desertification area increased from 36.91%to 44.46%.Overall,the extremely severe land desertification increased,while other types of desertification decreased in study area.Therefore,land desertification in northern China showing a trend of "overall decrease and local aggravation".(2)The Sen trend analysis method and the Mann-Kendall trend test method were used to analyze the NDVI variation trend and its significance in the study area.The areas of vegetation NDVI with extremely significant decrease(p<0.01)and significant decrease(0.01<p<0.05)accounted for 0.94%and 3.90%,respectively,which mainly distributed in the desert and gobi regions in the northwest.The areas of vegetation NDVI with extremely significant increas(p<0.01)and significant increas(0.01<p<0.05)accounting for 21.97%and 19.86%,respectively,which mainly distributed in loess plateau,Inner Mongolia grassland and northeast of China.The areas of vegetation NDVI with no significant variation trend accounted for 53.33%,which mainly distributed in Xinjiang,western Inner Mongolia,northwest Gansu and Tarim basin.The residual trend method was used to analyze the effect of human activities on vegetation restoration and desertification control in the northern region.The results show that the period from 2000 to 2019,the area of vegetation change with human activities in the study area showed significant positive correlation(p<0.05)and positive correlation(0.1<p<0.05)accounted for 21.36%;the area of significant negative correlation(p<0.05)and negative correlation(0.1<p<0.05)accounted for 6.49%.Overall,human activities have a positive impact on vegetation change in the study area.A series of ecological environmental protection measures and sand control projects have gradually reversed the land desertification in the study area.(3)LSTM neural network be well-suited for processing time series prediction,and the short-term NDVI prediction results for Horqin region show that vegetation NDVI will continue to increase from 2020 to 2030.During the period of 2020-2030,land desertification in Horqin area is mainly distributed in the central and southwest regions.In the next 10 years,land desertification in Horqin area will be reversed. |
PDF Full Text Request