Simulation Of Spatio-temporal Evolution Of Land Use In Central Yunnan Urban Agglomeration Based On Markov-PLUS Model

Land use/cover change（LUCC）affects global climate change,local ecosystem cycles and human survival,and has become one of the current hot issues in global change and sustainability research.As an important part of LUCC research,simulating and predicting future spatial and temporal land use changes has been the focus of research in land change science and sustainability science.It is a difficult point.Simulation models have become an important tool for exploring land use change mechanisms and predicting future spatial distribution patterns of LUCC.However,with the rapid socio-economic development,the rate of urban expansion has accelerated,the ecological space has been compressed,and the contradiction between supply and demand of land resources and the relationship between human and land within urban agglomerations has become prominent.Moreover,few studies have been conducted at home and abroad to fine-tune the spatial land use type simulation of urban agglomerations at the regional scale.In the face of the complex spatial and temporal changes of urban agglomerations,how to simulate spatial LUCC dynamically and finely,and how to predict and optimize the allocation of land resources during the development of urban agglomerations scientifically? This is a scientific problem that must be solved at present.Therefore,this paper takes the Central Yunnan Urban Agglomeration as the research area,and analyzes the evolution characteristics of the land use pattern of the Central Yunnan Urban Agglomeration from 1980 to 2020 qualitatively and quantitatively in time and space by means of spatial superposition analysis,land dynamics model,transfer matrix,and landscape pattern index,and explores the land use quantity structure,land use change rate,land class increase and decrease rate,land use dynamics,land use change status,land class transfer index,and land use change status of the Central Yunnan Urban Agglomeration.The land use change state,land class transfer change,land class exchange,and land use landscape pattern change characteristics are discussed,and the current problems of land use development in the Central Yunnan Urban Agglomeration are summarized.To address the existing problems,the land use expansion analysis strategy is used to analyze the contribution of land use types to the driving factors and to explore the driving mechanism of land use change,based on which the Markov-PLUS model simulates and predicts and analyzes the spatial pattern of LUCC under three different development scenarios: natural development scenario,arable land conservation scenario and ecological conservation scenario in 2030,2040 and 2050.The main findings are as follows:（1）Spatial and temporal characteristics of historical land use changes in the Central Yunnan Urban Agglomeration.from 1980 to 2020,the natural land surface dominates in land use and constitutes a land use structure characteristic of agroforestry complex.The active degree of land use change varies in different periods,and the overall trend is increasingly active,with construction land being the most active land type.The process of interconversion of land use types is complex,and the performance varies greatly in different periods.There are differences in the net change and exchange change dominance of land types in Central Yunnan Urban Agglomeration in different periods.Among them,arable land,forest land,grassland and water area mainly show exchange changes,while construction land is a net change.In the past 40 years,human activities in the Central Yunnan Urban Agglomeration have gradually increased the disturbance of land use,and the fragmentation of land class landscape by human influence has increased.The aggregation of land type patches gradually decreases,and the fragmentation of land type landscape is increasing.Among them,the fragmentation degree of internal arable land is the highest,and the interference and influence of human activities on arable land is the greatest.（2）Analysis of the contribution of driving factors of land class expansion in the Central Yunnan Urban Agglomeration.The main driving factors for the expansion of different land types are different.The main drivers of arable land expansion are distance to the countryside and DEM and GDP.The biggest influence on woodland expansion is DEM and slope,and the higher the altitude,the more forest land is distributed.The largest contributor to grassland expansion is slope direction.The drivers of watershed expansion are distance to lake,distance to river,and DEM,with watershed expansion occurring in low-lying,gently sloping,or high mountain valley areas.The main driving factors for the expansion of built-up land are distance to food and beverage areas,GDP,and distance to primary and tertiary roads,and built-up land is mostly located in areas with intensive human economic activities and dense transportation routes.The correlation between the expansion of unused land and DEM is very high and much greater than the contribution of other factors.Among all factors,slope orientation contributes less to all land use types except for the expansion of forest land,while DEM has a higher influence on the expansion of each land type.（3）The coupled Markov-PLUS model can accurately predict the land use/cover changes in the Central Yunnan Urban Agglomeration.By comparing the actual land use data in 2020 with the simulation results in 2020,the overall accuracy of the simulation results is 0.96,the Kappa coefficient is 0.93,and the Fo M coefficient is 0.58,which verifies the advantages of the Markov-PLUS model in simulating LUCC at the urban cluster scale,and the use of the Markov-PLUS model for future LUCC scenario simulation has high reliability.（4）Dynamic simulation results of the future land use of the central Yunnan urban agglomeration under multiple scenarios.Three land use development patterns,namely,natural development scenario,arable land protection scenario and ecological protection scenario,are set by the dual quantitative-spatial constraint method.The simulation results show that the future land use changes under different scenarios differ significantly in terms of quantitative and spatial changes,and the future land use rules of the urban agglomeration can be selected according to different needs.The research results help decision makers to evaluate the land use development patterns and provide a reference for the formulation of regional sustainable development policies.