【摘要】 利用空间二元统计和多项Logistic回归两种方法分析耕地变化的驱动力及作用方式,在此基础上对未来耕地演变进行了预测。结果表明,青岛市土地利用变化特点是耕地消失和城乡建设用地扩张,主要原因是各项人工建设,自然条件起到了一定的限制作用。空间二元统计分析方法可直观得到空间要素不同条件下耕地变化密度分布,并为多项Logistic回归参数选择提供依据。采用多项Logistic回归进行土地利用/覆盖演变分析,在分析精度与驱动力解释性方面具有明显优势,并为预测未来演变趋势提供了准确的参数。更多还原

【Abstract】 This paper uses spatial bivariate analysis and multinomial logistic analysis to find the driving forces of arable land change in Qingdao city and predicts the change of arable land in the future.The results show that(1) the characteristics of land use change in the study area is the disappearance of much arable land and the expansion of built-up area,and(2) the main force making the arable land lost is artificial constructions,and natural conditions play a constrain role.The method of spatial bivariate analysis can get the changing density distribution under different condition of spatial features and support the decision of which variables to integrate for multinomial logistic regression.The method of multinomial logistic regression has obvious advantages in terms of accuracy of analysis results and explanation of the driving forces,and provides accurate parameters for the prediction of changes in the future.更多还原