基于植被遥感的西南喀斯特退耕还林工程效果评价——以贵州省毕节地区为例  

Effect assessment of the project of grain for green in the karst region in Southwestern China:a case study of Bijie Prefecture

作　　者：李昊[1] 蔡运龙[1] 陈睿山[1] 陈琼[1,2] 严祥[1] 

机构地区：[1]北京大学城市与环境学院地表过程分析与模拟教育部重点实验室,北京100871 [2]青海师范大学生命与地理科学学院,西宁810008

出　　处：《生态学报》2011年第12期3255-3264,共10页Acta Ecologica Sinica

基　　金：国家自然科学基金资助项目(40871047);国家科技支撑计划子课题(2006BAC01A01-03)

摘　　要：中国西南喀斯特地区生态环境脆弱,人口压力和不合理的土地利用方式使得土地退化问题严峻。2000年以来,国家开始在该地区推行退耕还林等一系列生态工程。需要评估这些工程的效果,以期为进一步的生态建设工程决策提供科学依据。以贵州省毕节地区为例,利用SPOT-VGTNDVI遥感数据,以大规模开展退耕还林工程前的1998—2001年为基准,建立NDVI-气候响应模型,在此基础上结合残差法来分析2002—2008年以退耕还林工程为主的人为因素在当地生态恢复中的作用。结果表明,近年来开展的生态工程使得整个毕节地区植被条件得到了明显的改善,但在东部大方、黔西一些区域土地退化的趋势仍未扭转,需要今后进一步的政策引导和开展后续生态工程。The karst region in Southwestern China possesses fragile eco-environment. With increasing population pressure and irrational land use, the local environment had experienced severe degradation showing as soil erosion, loss of soil fertility and deterioration of vegetation cover in the past decades. Since 2000, a series of ecological restoration projects, including the project of Grain for Green （conversion cropland into forest or pasture）, have been implemented in this region. It is quite necessary and important to assess the effect of these projects so as to provide a scientific basis for decision-making of further ecological restoration projects. This paper takes Bijie Prefecture in Guizhou Province as a case study. Factors influencing the vegetation dynamics are identified as climate change and human activities. In order to verify the NDVI trends imposed by human influences such as the project of Grain for Green, this paper establishes a linear regression model between NDV1 and climate by means of SPOT-VGT NDVI data and meteorological data from 1998 to 2001. By performing lots of linear regression calculations, the NDVl-climate influence model identifies accumulated precipitations from January to June and accumulated temperature from April to July as independent variables, and the NDVI from 200d to 210d per year which is assumed to be the max value of the year as the dependent variable. The model is also calculated at the pixels level with 1000m resolution. Then the residual trends of actual NDVI and simulated NDVI from 2002 to 2008 are calculated. Positive or negative trends expressed in the residuals are interpreted as human-induced restoration or degradation. Since vegetation condition is influenced by a series of ecological restoration projects in Bijie Prefecture, the effort of the project ofGrain for Green should be distinguished from other ecological projects before evaluation. Combined with the result of residual analysis, the land-use data is also used. In particular, the paper evaluates the residu

关 键 词：植被遥感 NDVI 喀斯特 毕节地区 退耕还林工程 效果评价 

分 类 号：S721[农业科学—林木遗传育种]