机构地区: 华南师范大学地理科学学院
出 处: 《水土保持通报》 2011年第1期238-243,F0003,共7页
摘 要: 基于遥感、GIS和地统计学分析方法,通过引入景观指数、景观变化动态度指数和景观变化动力梯度指数,定量分析了佛山市土地利用景观格局的时空变化及其综合驱动力的空间分异。结果表明,佛山市景观的数量与格局都发生了剧烈的变化,且差异显著。1991—2005年,佛山市景观的数量变化主要以耕地和城镇景观为主,前者大规模减少,而后者剧烈增加;耕地与园地景观受干扰的程度最大,景观破碎程度较高;而林地尽管受到一定干扰,但破碎程度不高,景观趋于均匀;城镇景观的斑块集中连片分布趋势明显,但依然显示出强烈的破碎状态。对驱动力因子的空间梯度分析表明,1991—2005年,佛山市景观格局变化的动力因素呈现出显著的空间梯度分异,且与佛山市土地利用景观的数量和空间变化形成了较好的对应关系。从空间上度量景观格局变化驱动力的空间梯度分异,有利于寻找驱动景观格局形成过程与格局本身之间的对应关系,进而加深对区域景观格局变化的深入理解。 Great attentions have been paid on landscape pattern changes and the driving forces for rapidly urbanizing areas in recent years.Based on TM images of Foshan City in 1991 and 2005,landscape metrics,landscape change driving forces index(LCDI) and landscape gradient index(LCGI) were selected to analyze the changes of landscape pattern and spatial gradient heterogeneity of driving forces with GIS and geostatistics.The results indicate that the landscape pattern changed remarkably from 1991 to 2005.Changes of cultivated lands and constructional lands were dominant in the whole landscape change.The area of cultivated lands reduced greatly from 37.50% to 15.09%.Meanwhile,the total area of construction lands increased 98 570 hm2 during 1991—2005,with the highest LCDI value.According to the results of landscape metrics analysis,the fragmentation degree of cultivated lands and garden lands were relatively high,due to the intense human disturbances in these lands.Woodlands showed symmetrical spatial characteristics.Construction lands had the highest fragmentation degree of all the landscapes in the city,although strong spatial distribution centrality was found in constructional lands.Through spatial analysis on synthetic driving forces of the landscape changes,the spatial gradient heterogeneity became increasingly evident from 1991 to 2005,closely related to landscape patterns and driving forces.Spatial gradient analysis on driving mechanism of landscape pattern change is proven helpful in identifying the relationship between landscape patterns and processes,which could potentially enhance our understanding of the pattern and direction heterogeneity of landscape pattern changes.