武夷山市土地利用结构信息熵动态研究

Dynamic study on information entropy of land use structure in Wuyishan City

应用 Shannon等人提出的 Shannon熵理论 ,研究武夷山市土地利用结构信息熵的动态演变规律及其成因。结果表明 ,该市土地利用结构信息熵值较全国水平低 ,具有时间上的动态性 ,在研究年限内呈单调递增趋势 ,土地利用系统向无序方向发展 ;林地、交通用地、居民点及工矿用地、水利设施用地面积的高低是影响该市土地利用结构信息熵值大小的决定性因素 ;整体上看 ,武夷山市耕地、林地和未利用地面积所占比重呈递减趋势 ,园地、居民点及工矿用地。

The present paper aims to introduce the authors' use of the method initiated by Shannon to analyze the land use structure in Wuyishan City so as to reflect the dynamic changes and transformation degree of land use type. If the value of entropy is higher, it indicates that the order of the system is worse, and, in turn, its structure will be proved illogic, too. As is well known, land is an integrated system, which has structural and functional dynamic and serial character, which can be shown by the information entropy. To define the serial character, a method was found by Chanon to analyze and reflect the dynamic changes and transformation degree of land use type, for example, the land use structure in Wuyishan City. As a matter of fact, the value of information entropy of the land use structure in Wuyishan City is rather lower in comparison with the level of our whole country, namely, took on the dynamically temporary nature. That is to say, its present flat increase is done by degrees in a fixed rate from 1991 to 2001, whose land use system is proved in the bad serial. The result shows that the amount of woods and the constructive land area are the key factors influencing the information entropy of land use structure in the city. However, on the whole, the land with vegetation, the land covered with trees and uncultivated areas have been decreasing; the garden plots, land for traffic use, residential areas, industrial land and mineral land areas have been increasing from 1991 to 2001. With the correlation coefficient to analyze the correlative degree of the seven land use types and information entropy, the result shows that wood land has the greatest correlative degree with information entropy, next being the traffic land. And finally, water covered land areas are the least correlative degree with the information entropy.