Research Of Land Use Change And Scenario Simulation In Wuhan Based On Time Series Landsat Images

Land use/cover change is not only a fundamental research of global change and an important element of sustainable development,but also the most direct and visible signal of human alteration to the surface environment.At present,rapid urbanization has led to significant changes in regional land use,meanwhile the rough use of land resources and spatial sprawl of cities happen occasionally,which makes the conflict between people and land more acutely.Therefore,it is crucial to reveal the laws of urban land use change,explore its dynamic change mechanism,and then carry out scientific and effective modelling of future land use pattern,so as to reconcile the contradiction between human and land,thus promoting regional sustainable development.Taking Wuhan as an example,this paper first analyzed the characteristics of land use structural change,quantitative transfer characteristics,dynamic degree change and characteristic of landscape pattern in Wuhan from 1990 to 2019 from four aspects:land use structural change,transfer matrix,land use dynamic degree and landscape pattern.Then quantitatively exploring the key factors and factor interaction mechanism of land use degree changes in Wuhan,and further discussed the characteristics of land use degree change with factors.Next,the logistic regression principle was introduced and the Logistic-CA-Markov(LCM)model was constructed to simulate the distribution pattern of land use in this region from 2000 to 2019,and verified its effectiveness and practicality.Finally,three scenarios of natural development,economic priority development and sustainable development were set up to simulate the spatial distribution of land use of Wuhan in 2023 under each scenario.The main conclusions of the paper are as follows:(1)In terms of land use structure change,water,construction land,arable land and forest land in Wuhan account for a larger proportion in each year,while the proportion of grassland and unused land is less than 1%.The area of water shows a change of first increased and then decreased,and reaches the peak of 20.49%in 2010.The area of construction land increases year by year,and the average annual growth rate of construction land in the later period(2000?2019)is significantly higher than that in the earlier period(1990?2000).The area of cultivated land and forest land generally show a trend of decreasing year by year,and the increase in construction land is highly synchronized with the decrease in cultivated land and forest land.(2)In terms of land use dynamic degree,the dynamic degree of water varies greatly from year to year,with the maximum value of 0.44%from 1990 to 1995 and the minimum value of 0.01%from 2005 to 2010.The dynamic degree of construction land is higher than1%in the study period,which is at a higher level compared to other land use types.The dynamic degree of cropland and forest land have a high similarity,but the dynamic degree of cropland is generally higher than that of forest land.In addition,the correlation coefficient between the land use degree and time is 0.95,which shows a strong positive correlation.(3)In terms of landscape pattern,the largest patch index of water,construction land,and cultivated land remain basically stable with increasing grain size.The edge density of construction land is less sensitive to changes in grain size,but other landscape types are relatively more sensitive to changes in grain size.The landscape shape index of construction land basically remains unchanged with the increase of grain size,but there is a negative correlation between other landscape types and grain size.The fractal dimension of perimeter area of each landscape type has different sensitivity with the change of grain size.The perimeter area fractal dimension of other landscape types basically changing monotonously with the increase of grain except for water.It is worth noting that there is a strong negative correlation between patch cohesion index,aggregation index and grain size of each landscape type,especially in grassland and unused land.In addition,the distribution of landscape diversity has obvious spatial differences with time.And the topography has a great impact on the distribution of landscape diversity.(4)The explanatory power of each factor on the spatial distribution of land use degree varies greatly in different years.But the explanatory power values of soil type(X₉),GDP(X₂₁)and night light brightness(X₂₂)are all higher than 0.1 in each period,which indicates that soil type and economic factors have always influenced the change of land use degree in Wuhan to a greater extent.The effect of factor interactions on land use change is not a simple linear feature,but rather shows a non-linear or a double-factor enhancement,which indicates that the interaction between factors further promotes the change of land use in Wuhan.There are differences in the degree of land use within each factor classification,and the trend of change is also different as the factor level increases.Among them,the degree of land use has a strong positive correlation with the night light brightness,while a high negative correlation with the distance to residential areas,medical resources,and roads.(5)The area under the receiver operating characteristic(ROC)curve of each land use type is greater than 0.75,indicating that the regression model can better reflect the relationship between each land use type and the independent variables.The Kappa coefficients of land use simulation results of each year are all higher than 0.75,with high simulation accuracy,which further indicates that the LCM model can better simulate the spatial distribution of land use in Wuhan.Scenario simulation results show that the area of water,construction land,arable land and forest land in Wuhan are different under different development scenarios,and the area of water and forest land are the highest in the sustainable development scenario,the natural development scenario is second,and the economic priority scenario is the lowest.The area of construction land is the highest in the economic priority development scenario,the natural development scenario is second and the sustainable development scenario is the lowest.This paper has 44 figures,17 tables and 153 references.