Font Size: a A A

Land Use Dynamic Change Analysis And Scenario Prediction In Changchun City

Posted on:2021-05-07Degree:MasterType:Thesis
Country:ChinaCandidate:X GuoFull Text:PDF
GTID:2370330626458967Subject:Surveying and mapping engineering
Abstract/Summary:PDF Full Text Request
Changchun is one of the central cities in Northeast China.With the support of national policies,social economy developing rapidly.The exploration of LUCC and future development scenarios,simulating dynamic change process can provide relevant decision-making and suggestions for the long-term planning and development of Changchun City,which is beneficial to regional sustainable development.This paper takes Changchun City as the research area,combing RS and GIS technology and using the CART algorithm of decision tree classification to obtain land use classification data based on six landsat satellite images.On this basis,from the perspectives of quantity,transfer,level,trajectory and spatial migration,standards such as land use dynamic degree,land use transfer matrix,and land use comprehensive index,as well as models of land use trajectory and center of gravity migration are used to analyze the characteristics of land use/cover dynamic changes in Changchun qualitatively and quantitatively.A total of 11 natural factors,social economic factors and accessibility factors were extracted and the FLUS model was used to simulate the land use in Changchun in 2017.After meeting the simulation accuracy requirements,two scenarios of natural development and ecological optimization were proposed to simulate Changchun 2024 Annual land use pattern.The main findings are as follows:(1)The CART method was used to classify landsat images in 1997,2001,2005,2010,2014,and 2017.The overall accuracy of the classification was91.33%,91.17%,90.04%,90.51%,91.06% and 91.64%.Kappa coefficients are0.87,0.87,0.85,0.86,0.86 and 0.87 in order.The classification effect is good and the results are reliable,which can be used as the basic data for the analysis of land use dynamic changes in the next step.Based on the six remote sensing classification image,it can be seen that the mainly land feature types in Changchun are cultivated land,which accounts for about 74% of the total area,followed by construction land,which accounts for about 12% of the area,and the remaining land types occupy a relatively small area,less then 10%,the grassland is about 6%,woodland is about 4%,water and other land are both about 2%.From 1997 to 2017,cultivated land,grassland and woodland decreased in varying degrees,and the area of construction land increased with an annual increase rate of 1.48%.Among them,the conversion of constructionland from cultivated land and other land shows that the urban construction land is continuously expanding outward.The urbanization trajectory of cultivated land converted to construction land is mostly distributed in the south,southwest,northwest,and northeast.The southwestern part is the most densely distributed.The expansion pattern of Changchun City can be described as multi-center-driven,the main is urban center and the supplements are in the north and south of the city.(2)Based on the land use pattern of Changchun City in 2010 and combined with the FLUS model,simulating the 2017 land use pattern.The simulation result is that the overall accuracy is 90.07%,and the Kappa coefficient is 0.8811.After verification with real data,the simulation accuracy is pretty good.The land use dynamic change during the simulation period is in line with the actual situation.The model is highly applicable and can be used to simulate future land use patterns.(3)The land use scenario of Changchun City in 2024 is predicted.Under natural development scenarios,the area of cultivated land has continued to decrease,and high-quality crops have been occupied by construction land;the area of forest land and water areas has decreased,ecological land such as forests and rivers has not been protected,and some water areas have disappeared.Comprehensively consider factors such as nature and policy,this article proposed an ecological optimization scenario,which set basic farmland protection areas,river ecological corridor areas and forest protection areas,alleviated the pressure on forest land and water resources protection,optimized land use prediction results,reduced the area shrinkage of ecological land types such as cultivated land,forest land and water areas,and avoided the environmental degradation.In the end,it provided suggestions for rational planning of urban land use and sustainable regional development.
Keywords/Search Tags:Remote sensing, CART algorithm, Land use, FLUS model, Scenario prediction
PDF Full Text Request
Related items