Research On Land Use Change Prediction Based On Improved CLUE-S Model And Landscape Pattern Analysis

Land is fundamental to human survival and development,related to people's lives with happiness,progress and socio-economic changes in the global environment,and its importance should not be overlooked.At the same time,the prediction of land type changes and the comprehensive analysis of landscape patterns are of great significance to ecological planning and construction.On November 6,2017,Changle changed from city to district and became the sixth district of Fuzhou.As a key new area of Fujian Province,Changle District is developing rapidly,and it has also had a certain impact on the spatial pattern of land use.Learning about Changle Changes in the spatial pattern since the city was withdrawn and districts were established,the development of urban land use change prediction research and landscape pattern analysis can provide corresponding technical support for land construction in Changle.This paper takes Changle District,Fuzhou City,Fujian Province as the research area,and obtains land use classification data on the basis of processing remote sensing image data in 2013,2016,and 2019.The traditional CLUE-S model is improved by introducing the neighborhood enrichment,and finally the improved CLUE-S model is used to predict the land use change of Changle District in 2025 under different scenarios of natural evolution,ecological protection,and rapid economic development.At the same time,it analyzes the changes of land types under different scenarios from the perspective of landscape pattern.The main results of the thesis are as follows:(1)Based on the Landsat TM remote sensing data in 2013,2016,and 2019,the method of support vector machine in supervised classification is used to obtain land use classification maps in three periods.On the basis of calculating the error matrix and Kappa coefficient to meet the accuracy requirements,further use Google Earth imagery to adopt the method of human-computer interaction interpretation and analysis combined with field surveys to verify the accuracy of the classification results to ensure the accuracy of the classification data.At the same time,based on the classification results,the analysis of land use transfer in Changle District from 2013 to 2019 was carried out.(2)The spatial analysis module of the CLUE-S prediction model is improved by using the Autologistic regression analysis method that introduces spatial autocorrelation factors and neighborhood abundance factors reflecting the process of land self-organization.Further improve the prediction accuracy of this model,and it is proved that the improved CLUE-S model can better simulate the future land use spatial pattern of Changle District.(3)The improved CLUE-S model was used to simulate the spatial pattern of land use in Changle District in 2025 under the scenarios of natural evolution,ecological protection,and rapid economic development.Then combined with the theory of landscape ecology,analyze the changes of the land use landscape index in Changle District from 2013 to 2025 under the three scenarios.It provides effective suggestions for the future macro-control of land use in Changle District,ecological environment protection,and overall planning of urban-rural and regional development.