| As the rapid development of urbanization,the research on urban landscape is becoming more and more important in landscape ecology.Based on ecological theory and method as well as remote sensing and GIS software,this paper made research on the evolution of landscape pattern in the last two decades and its driving force.Using the network analysis method,this paper also build a optimazied greenbelt network system of suzhou in 2005.The research result is helpful on revealing the reason of landscape pattern evolution in city area caused by urbanization,and can also provide a new theoretical base in the biodiversity protection and plan.The major result of this research is listed below:(1)In the last two decades,the landscape pattern has changed significantly.The transfer of landscape class mostly happens in the change of farmland to urban consturction area.During 1987-2005,the main landscapce type are framland and greenbelt,the area of these two landscape type count 60% (1987),56%(1995),54%(2001)and 52%(2005)of the total area of the landscape.As the devlopment of urbanization,the area of urban construction type increases 7229ha and the area of framland and rural construction area decrease 2888ha and 7229ha.73%of the increase of the area of urban construction type comes from the decrease of area of framland and rural construction,and 27%comes from the decrease of area of greenbelt and water.(2)As the development of urbanization,the patch number increases significantly from 15425(1987)to 18184(2005)and the mean-patch area decreases from 3.48ha(1987)to 1.87ha(2005).The major landscape characteristic of the landscape is fragmentation.The landscape type fragmentation index increases from 0.28(1987)to 0.34(2005),the percentage is 16.2%.(3)The heterogenity and eveness index rises at the percentage of 6.8%and 10.2%as the development of urbanization,and the landscape shape index rises during 1987-2005. (4)Based on PCA method,the first principal component has the contribution rate of 75%,the density and the area of the road are the main factors of the first principal component which have the contribution rate of 0.951 and 0.947.The residential area factor in the first principal component also has the contribution rate of 0.895.So we draw the conclusion that the density and the area of the road as well as residential area are the main reasons for the fragmentation of landscape of Suzhou in 2005.(5)Based on the network analysis method and GIS software,the greenbelt patches lager than 10hm~2 are selected as points and the green corridor are drawn as lines.According the the common type of networks,we build 5 ecological networks of greenbelt.Using the network indices such asαβγindices,and cast ratio to evaluate the 5 ecological networks of greenbelt.The result is that the networks 1 and 2 which are built on the basis of shortest distance and least cost have low value ofα,β,rindices.One corridor which close the main greenbelt ring is added onto network 2 to build network 3 and increses theαindices 100%.The network 4 connect all of the greenbelt patches and is the most commplicated and complete network,and theα,β,γindices are increased rapidly to 0.19,1.33 and 1.47.Limited by the actual landuse situation some corridors in network 4 are impossible to build so according to the actual landuse situation we build network 5 wich has the sameα,β,γindices with the network 4 but lower cost ratio and have more operation ability.So network 5 is regarded as the optimized greenbelt ecological network.(6)The ecological network 5 consisit three main corridor system which are corridors connect the greenbelt along the Taihu Lake,corridors connect the natural and semi-natural greenbelt patchs in suburban area as well as the corridors connect the parks and greenpatches inside the urban area.Through important points like Qionglong mountain,Nanyang mountain Lingyan mountain and so on the three main corridor system are linked together to form a complete greenbelt ecological network.
|
PDF Full Text Request