| Cultivated land is an important foundation for agricultural development and a prere quisite for food security.As the global population continues to grow,food security has a lways been a hot issue of global concern.With the rapid development of my country’s ur banization,complex terrain,diverse climates and other factors,the phenomenon of culti vated land abandonment is more common,which has an important impact on the continu ous stability of cultivated land area and grain output.At present,remote sensing technology is still a more effective way to obtain the spa tial and temporal distribution characteristics of abandoned cultivated land in larger-scale areas.Internationally,the work of remote sensing identification of abandoned cultivated land is relatively mature,and there are few domestic studies.There is still a relatively la ck of scientific and feasible remote sensing monitoring methods for abandoned cultivate d land.Moreover,the distribution of abandoned cultivated land in my country is scattere d and fragmented.It is still difficult to extract abandoned land by remote sensing techno logy..Therefore,it is necessary to study the effective method of extracting abandoned l and with remote sensing technology,and analyze the temporal and spatial variation char acteristics of abandoned cultivated land.In addition,in order to scientifically analyze th e potential impact of cultivated land abandonment,it is also of great practical significan ce to construct a simulation and prediction model of cultivated land abandonment based on the corresponding influencing factors,which can provide a scientific basis for the for mulation of land protection policies.This study takes Jingdezhen City,Jiangxi Province as the research object,with the help of the powerful advantages of the remote sensing big data cloud computing platfor m Google Earth Engine(GEE)to efficiently acquire and process data,obtain the land us e data of Jingdezhen City from 2001 to 2020,and select random data.Forest classificati on methods for land use interpretation.On the basis of this interpretation data,according to the evolution process of cultivated land abandonment,the rules of cultivated land aba ndonment are set,a change detection method based on land use trajectory is proposed,th e abandoned land in the study area from 2001 to 2020 is extracted,and the abandoned la nd spatiotemporal evolution characteristics.LUCC model is an effective way to explain and study the dynamic characteristics o f land use.Build a simulation model of abandoned land change that couples the system d ynamics model and the CLUE-S model,and input the abandoned land area results simul ated by system dynamics into the CLUE-S model as a quantity demand file,simulating a nd predicting Jingdezhen City 2010-2025 Spatial distribution of annual abandoned land.The main research contents and conclusions are as follows:(1)Based on the definition rules of abandoned land,the abandoned land is extracte d by tracing its land use trajectory.In the past 20 years from 2003 to 2020,the average a bandonment rate in Jingdezhen was 8.37%,and the average abandoned area was 10,869.85 hectares.Overall,the abandoned cultivated land showed a periodic trend of increasin g first and then decreasing,with the most abandonment in 2016 and 2017,with an area o f more than 15,000 hectares.(2)Make full use of the advantages of system dynamics in simulating the quantitati ve changes of complex systems and the advantages of the CLUE-S model in spatial sim ulation to more accurately and intuitively analyze and predict the temporal and spatial c hanges of abandoned land.In 2025,under the three scenarios of natural development sce nario,cultivated land protection scenario and high-speed development scenario,the pred icted abandoned land areas in Jingdezhen are 19,762.3 hectares,16,568.1 hectares,and 20,671.4 hectares,respectively,and the abandonment rates are 14.67%,9.74%,and 16.28%,respectively.,the changes in the abandonment rate in 2018 were an increase of 3.17%,a decrease of 1.76% and an increase of 4.78%.Under the cultivated land protection scenario,the abandonment rate decreased by 1.76%.When the per capita arable land area is smaller and the proportion of non-agricultur al income to farmers’ income is higher,it is more likely that arable land will be abandon ed.Under the high-speed development scenario,even if the cultivated land area increase s slightly,the population growth and economic growth are more rapid,the per capita cul tivated land area decreases rapidly,the rapid development provides more employment o pportunities and higher income,and the proportion of non-agricultural income increases significantly,resulting in a larger increase in the area of abandoned arable land.Under the natural development scenario,the area of arable land decreases slightly,and the tren d of decreasing per capita arable land area and increasing the proportion of non-agricult ural income is relatively slow and tends to be stable.Under the cultivated land protectio n scenario,with the increase of cultivated land area and rural population,the per capita c ultivated land area will gradually increase,the proportion of non-agricultural income wil l gradually decrease and then stabilize,and the abandoned cultivated land area will also tend to decrease.(3)Use the constructed system dynamics and CLUE-S simulation model of abando ned land in Jingdezhen City to simulate the temporal and spatial change process of aban doned land from 2010 to 2025,and compare the simulation results in 2025 with the curr ent spatial distribution of abandoned land: Under the three scenarios,the area of abando ned land in Jingdezhen will still show an increasing trend in the future,and the abandon ment of cultivated land will mainly occur in areas with large elevation changes.In the n ortheast and south,where the terrain is the highest,there is less cultivated land and no o bvious abandonment is observed.Compared with the other two scenarios,the abandoned land under the high-speed d evelopment scenario is no longer limited to the cultivated land with poor terrain conditio ns,but also more of the high-quality cultivated land with relatively flat terrain such as th e central and southwestern regions. |
