Driving Forces Analysis Of Land Use/Cover Change Based On Logistic Regression Model In Beijing

Based on the land use/cover maps of Beijing in 2000 and 2010 with scale 1:100000,the quantitative and spatial characteristics of land use change and unstable patches were analyzed with the help of GIS and SPSS.During the process,the arable land spatial change,as the dependent variable,and other 20 explanatory variables chosen from natural factors,socio-economic factors and integrated factors were analyzed by the multinominal Logistic regression model.The results are as follows:(1)Forest land and arable land were the main land use types in Beijing.During the study period,the area of arable land and water changed most.Unused land and the water area had larger dynamic degree.The result of single dynamic degree on the scale of the zoning showed that the amount stability of a certain land use type in different partitions had a strong positive correlation with the area proportion.Forest land,arable land and construction land kept higher spatial stability in Beijing.Both of the spatial stability of a certain land use type in different partitions and the spatial stability of different land use types in the same partition were also positively correlated with the area proportion.The results of landscape pattern showed that the degree of fragmentation of the construction land decreased,while the degree of fragmentation of the others increased,with 30×30m resolution.Among them,arable land had the smallest degree of fragmentation,while the degree of fragmentation of forest land,grassland,water area and unused land increased greatly.Further studies showed that the patches whose single patch area was less than 10hm2 devoted most to the fragmentation of the forest land.The mutual transition among arable land,forest land and grass land,the transition between arable land and construction land and the transformation of water area to arable land explained 92.95% of the land use change.Compared with the position of land use type in 2000,the centroid of forest land and unused land shifted away from the urban area in 2010,while the centroids of the others were more close to the urban area.(2)The change of land use in Beijing was the result of macro-demand and spatial characteristics.Macro factors,such as policy,urban planning,landscape engineering,increase of population,development of society and economy,adjustment of agriculture,the Olympic Games and other events,had a siginificant impact on the land use change.(3)The results of the driving forces were as follows.Expect neiborhood factors,the driving factor which had a positive effect on the conversion of a certain land use type to another type generally had a negative impact on the conversion to the certain land use type.In areas with poor natural conditions and human accessibility,the land use types were more likely to be transferred to the land use type which was less influnced by human beings and was more friendly to the environment.Except the amount of water area in the neighborhood,other neighborhood factors promoted the transformation of the primary land use type to the type whose human impact strength was more close to the neighborhood type's.(4)The main factors which determined the distribution of land use change vastly.In general,the land use changed in a positive direction.However,forest land was still suffering damage from humans.The contribution rate of the conversion of forest land to grass land was up to 14.03%,and it scattered in the mountainous area of Yanshan.Therefore,the protection of forest land in the region should be paid more attention.The results indicated that the expansion of construction land could be guided by road construction and population growth,which could be used to optimize the development of the city.