| This paper takes Huainan and Huaibei two typical cities in Huainan and Huaibei mining area in Anhui Province as research areas,based on RS and GIS technology,using land use dynamics index,land use transfer matrix,land use map,land use change importance index,and landscape pattern index were used to comprehensively analyze the characteristics of land use landscape pattern change in Huainan from 1985 to 2016 and Huaibei from 1985 to 2015.Logistic model was used to measure the driving effect of different driving factors on the changes of various categories.With reference to previous studies,the traditional Logistic regression model is improved by introducing spatial autocorrelation variables to form the Autologistic model,which is then applied to the CLUE-S model to simulate and predict the land use change process.The results show that:1)From 1985 to 2016,the land use change in Huainan was mainly manifested by the decrease in cultivated land,forest land and grassland area,and the increase in waters and construction land area.From 2005 to 2016,the comprehensive dynamic degree was the largest,and it was the stage where the change of each land use type was the strongest.The land change is mainly the conversion of cultivated land to construction land and waters,and the newly added waters are concentrated in the Panxie mining area.LPI and CONTAG show a downward trend,LSI shows an upward trend,the land use landscape is developing in a fragmented direction,the landscape spatial connectivity decreases,the landscape shape tends to be complex,and human interference is strong.From 1985 to 2015,the land use change in Huaibei showed that the area of cultivated land,grassland and waters decreased,and the area of forest land and construction land increased.From 2005 to 2015,the comprehensive dynamic degree of land use was the largest,among which the forest land area changed the most drastically.The main land use changes are the conversion of cultivated land and construction land,and the transfer of grassland.LPI shows a downward trend,SHDI shows an upward trend,SHEI and IJI show an upward trend and then a downward trend.The regional landscape pattern is fragmented and the landscape heterogeneity is enhanced.The area ratio of each patch type in the landscape is uneven,and the spatial relationship tends to be simplify.2)Five simulation scales of 50m×50m,100m×100m,150m×150m,200m×200m and 250m×250m were selected to calculate the binary logistic regression equations of the land use types and driving factors at the five scales respectively,and compare the goodness of fit of the models on different simulation scales,the optimal scale chosen for simulation after comparison is 150m×150m.On this basis,comparing the difference between the traditional Logistic model and the improved Autologistic model,the results show that the Autologistic model has a better fit,ROC>0.8.indicating that the selected driving factor can better explain land use spatial change driving mechanism.According to the regression coefficient ? of the results of the Autologistic model,the main driving forces for the changes in cultivated land and construction land in Huainan are the annual precipitation and GDP,respectively.The main driving force of cultivated land and grassland in Huaibei is elevation,and the main driving force of construction land is population.3)Based on the land use data of 2010,the CLUE-S model was used to simulate the spatial distribution patterns of land use in Huainan in 2016 and Huaibei in 2015,respectively.Through comparative analysis with the land use status,the overall simulation accuracy is 79.22%and 78.03%,respectively,and the Kappa index is 0.74 and 0.72,respectively,indicating that the CLUE-S model has good land use simulation capabilities in the study area.On this basis,the CLUE-S model was used to predict the spatial distribution of land use in Huainan in 2028,2034 and 2040 and Huaibei in 2025,2030 and 2035.The results show that the spatial distribution pattern of future land use in the study area is relatively stable on the whole,there is no obvious change,and the area change of each land type is relatively stableFigure 16 table 23 reference 83... |
PDF Full Text Request