基于Markov和Logistic模型的莆田市土地利用变化及林地转出空间模拟

利用莆田市1995、2005和2015年3期遥感影像、数字高程模型及社会经济统计数据,从面积变化和空间转换2个方面分析莆田市土地利用变化的时空特征,运用Logistic回归分析土地利用空间变化与自然、社会经济驱动因子之间的定量关系,基于Markov模型对莆田市未来各土地利用类型面积变化进行预测,采用Markov模型与Logistic回归模型相结合对未来林地空间格局进行模拟与预测,结果表明:1995—2015年,林地大面积减少,主要转为农用地;农用地大面积减少,主要转为林地、建设用地和裸露地;建设用地面积大幅度增加,主要来自裸露地、农用地和水体;水体面积持续减少;裸露地面积先减少后增加。坡向、高程、距裸露地距离、距农用地距离、距林地距离、人口密度和地区生产总值7个因子对林地转出具有显著的影响。根据预测结果,至2035年,林地的比例将减少至40.37%;2015—2035年,林地转出的位置主要分布在仙游县和涵江区。

Study on land use change in Putian City and spatial simulation of forestland transition based on Markov model and Logistic regression model

Taking Putian City as the study area,using remote sensing images of 1995,2005 and 2015,DEM,and social-economic statistical data,the characteristics of temporal and spatial change of land use from two aspects of area change and space transition were analyzed by using Logistic regression model.Aims of this study were to explain the geographical and social-economic driving factors that caused land use change,predict the area of each land use type in the future based on Markov model,and simulate the spatial pattern of forestland in the future based on Markov and Logistic regression model.The result showed that forestland,which mainly transited to agricultural land,decreased sharply from 1995 to 2015;Agricultural land decreased quickly,which mainly transited to forestland,construction land,and bare land;Construction land increased significantly,which originated from bare land,agricultural land and water area;Water area decreased steadily;Bare land decreased first and then increased.Aspect,elevation,population density,gross domestic product,and distances to bare land,agricultural land and forestland,were important factors driving the spatial change of forestland.In conclusion,it is estimated that the proportion of forestland will be reduced to 40.37% in 2035 and the locations of forestland transition are mainly distributed in Xianyou County and Hanjiang Area from 2015 to 2035.