基于DBEST的2003—2012年济南市植被变化研究  

Study on Vegetation Change in Jinan During 2003—2012Based on DBEST

作　　者：刘吉彤 安如[1] Liu Jitong;An Ru(School of Earth and Engineering,Hohai University,Nanjing 211100,China)

机构地区：[1]河海大学地球科学与工程学院,江苏南京211100

出　　处：《甘肃科学学报》2018年第4期17-23,共7页Journal of Gansu Sciences

基　　金：国家自然科学基金(41271361;41271420);国家科技支撑计划资助项目(2013BAC03B0401);国家973计划资助项目(2009CB421105);江苏省"六大人才高峰"资助项目(XXRJ-011)

摘　　要：基于2003—2012年MODISNDVI数据,以济南市作为研究区域,利用DBEST算法对其植被变化进行研究,并利用Landsat数据进行植被变化的对比验证。利用DBEST算法对济南地区10a的MODISNDVI时间序列进行分析,结果表明其呈现一种先减少后增加的交替变化上升趋势,上升幅度大约为0.06。对其所有像元进行遍历,发现其主要变化大多发生在2003—2007年,其总体的植被变化幅度在-0.19~0.27之间,95%以上像元是恢复的。利用Landsat数据对探测结果进行验证发现,其精度可以达到77.3%,说明DBEST算法在探测植被变化方面有较好的应用。According to the MODIS NDVI data,a research was carried out for the vegetation changes by using DBEST algorithm,and vegetation was contrasted and verified by Landsat data.After study,it found that time series trend had an uptrend that alternately changes with decreasing first then increasing by applyingDBEST algorithm to analyzing the time series of MODIS NDVI during 10 years in Jinan district,and accessional range was approximately 0.06.The traverse was made for its all pixels,and it was found that most changes happened during 2003—2007,and change range of overall vegetation was within-0.19—0.27,and more than 95% pixels were recovered.The found results was verified by using Landsat data,and in the verification,its accuracy was up to 77.3%,which shows that DBEST algorithm has a good application in the exploration of vegetation changes.

关 键 词：DBEST算法 植被变化 时间序列分析 

分 类 号：P237[天文地球—摄影测量与遥感]