Ecological Response And Optimization Simulation Of Land Use Change In Chaohu Lake Basin

As a natural-economic complex,land plays an essential role in the human-land system and provides critical spatial support for ecological civilization construction.Given its response on the structure and function of the eco-economic system,land use change(LUC)has been an essential component of global environmental change and sustainable development research.At present,China's economy has shifted from rapid to high-quality development;the scientific allocation of land elements is,therefore,the key to supporting such a shift.As Chaohu Lake Basin has seen rapid social-economic growth and urbanization since 2010,a significant number of problems have arisen such as lowering spatial allocation efficiency of land resources,declining support capacity of ecological security,and increasing conversion of cultivated land into the built-up land.Also,some local areas are confronted with environmental pollution,biodiversity loss,and land degradation.In order to enhance the land use efficiency and green construction in this region,it is critically important to study the ecological response,structural optimization and spatial simulation of land use change.This thesis proposes a theoretical framework and a conceptual model of land use change response,optimization,and simulation(EROS-LUC)and includes the following research work.Firstly,land use classification of the study area was performed by the maximum likelihood classifier using Landsat image data acquired in 2000 and 2005,and HJ-1B image data in 2010 and 2015.Secondly,based on spatial autocorrelation and mathematical statistics,the spatiotemporal evolution and driving mechanism of land use were characterized from four aspects,namely structural change,degree change,transition,and landscape indices.Thirdly,in order to better quantitatively evaluate the change of eco-environmental quality,the calculation of remote sensing ecological index(RSEI)was improved using four critical ecoenvironmental elements(i.e.dryness,humidity,greenness,and heat)derived from remote sensing data,and an evaluation model of ecological response to land use change(LUCER)was built.Then,the eco-environmental quality gains and losses were introduced into the functions constraining the(quantitative)structure of land use.By using the multi-objective genetic algorithm and the multi-scenario land use structure optimization model,the land use structure of the study area was optimized for the year 2030.Lastly,the driving factors of the land use type were identified from four major categories regarding geography,soil,location transportation,and social economy of the study area.After the probability of spatial distribution was calculated by principal component factors,the spatial pattern of land use in 2030 was simulated by the DynaCLUE model under the structural and ecological constraints.The main results and conclusions of this thesis are as follows.(1)The proposed EROS-LUC framework consists of five mounting structures,namely defining the content of the ecological response,analyzing the ecological response,forming the constraints of ecological response,constructing the coordination mechanism of ecological response and implementing land use optimization simulation.It is the extension and application of the response system analysis framework.The EROS-LUC conceptual model reveals the coupling relationship and constraint mechanism of land use change model,the ecological response model,structural optimization model,and spatial simulation model.(2)The maximum likelihood-based supervised classification method results in high accuracy with the overall accuracy Kappa coefficient being over 85% and 0.83 respectively.Land use change analysis shows that the land use degree in the study area has risen with significant spatial correlation and agglomeration but decreasing hot-spots.While the land use landscape is further fragmentized,the trend of built-up land agglomeration has strengthened.The land use change is primarily explained by economic development,population growth,and urban expansion.(3)The calculation of RSEI was improved for the study area,and the model of ecological response to land use change(LUCER)was established.This model was used to quantitatively assess the eco-environmental quality gains and losses,which has proved considerably practicable.Results show that the overall eco-environmental quality of the study area was generally good despite a U-shaped trend from a rapid decline to a slow rise during the period of 2000-2015.Land use change has a significant adverse impact on the eco-environment: 1 km2 conversion from paddy field into builtup land would lead to a decline in m RSEI by 142.51.Because of such a remarkable ecological impact of paddy field,ecological constraints should be imposed in accordance with the eco-environment quality losses.(4)Given the strong ecological security and high-quality economic development in the study area,the quadrant method was adopted to set up three scenarios,i.e.,tight,moderate,and loose ecological constraints.In different scenarios,the multi-scenario land use structural optimization model was constructed with the constraint functions of the planning,ecological protection,and social development,and the objective functions of the maximum benefits of GDP(gross domestic product)and ESV(ecological service value).It is showed that when the ecological constraints are reduced,the rates of economic growth and the ecological decline are the same.The optimized quantitative structure conforms to the relevant planning and regional reality.(5)Applying a principal component analysis to calculate of the probability distribution of land use type spatial suitability not only reduces the dimension of factors and eliminates multi-collinearity,but also improves the analytical ability of the logistic regression model.The Dyna-CLUE model was used to simulate the spatial pattern of land use in the study area,resulting in a satisfactory Kappa coefficient of 0.8176.Moreover,if imposed with the constraints of the land structure optimization and the ecological suitability,the model simulation result is more scientific and reasonable.Besides,dynamic simulation of resource allocation would be achieved by setting different constraints on the model according to different scenarios.