Abstract:

A case study of Kunshan using data of farmland area and other nine social economic indices,such as total population,nonagricultural population,employments in primary industry,gross domestic product,per capita gross domestic product,value added of primary industry,value added of secondary industry,grain output and social fixed asset investment from 1985 to 2006,was held to research the farmland loss and its driving factors.In order to decrease the disadvantages in the application of mathematical methods of direct correlation analysis and principal component analysis (for example,the variations of social economic indices with time were similar,and it was difficult to get useful information by direct correlation analysis,and some researchers often mistaken principal component analysis for a method distinguishing relative importance of driving factors),new methods,such as correlation analysis on the changing rate of indices and segmentation of farmland loss process by principal component analysis were taken.Results showed that the new methods had distinct advantages in main driving factor analysis and process segmenting of farmland loss,and the conclusions were proved to be credible and reliable by testing the regression equations.Farmland loss in Kunshan was primarily due to the rapid development of social economics represented by the index of per capita gross domestic product,and was hastened by the development of the second industry and the increase of total population.The farmland loss process of Kunshan should be divided into three stages according to whether farmland loss handicapped the development of agriculture or not: not handicapped stage from 1985 to 1994,balanced stage from 1995 to 1997 and handicapped stage from 1998 to 2006.〖