文献类型：期刊文献

中文题名：基于TM遥感影像丰宁县森林地上生物量估测研究

英文题名：Estimation of Forest Biomass in Fengning County by TM Satellite Images

作者：王红岩[1] 高志海[1] 王琫瑜[1] 白黎娜[1] 吴俊君[1]

第一作者：王红岩

机构：[1]中国林业科学研究院资源信息研究所

年份：2010

期号：32

起止页码：18472-18474

中文期刊名：安徽农业科学

外文期刊名：Journal of Anhui Agricultural Sciences

收录：北大核心:【北大核心2008】；

基金：国家高技术研究计划(863)重点项目(2006AA120108)

语种：中文

中文关键词：森林生物量;TM影像;植被指数(VI);丰宁县

外文关键词：Forest biomass; TM image; Vegetation index（VI）; Fengning County

分类号：S127

摘要：利用TM遥感数据以及同期获得的野外调查样地数据,研究了河北省丰宁满族自治县森林地上生物量的遥感估测技术。提取TM遥感影像6个波段反射率及DVI、NDVI、PVI、RVI、VI3、SLAVI和SAVI 7个植被指数,分析了森林样地地上生物量与各个因子间的关系,得出相关系数较小(均小于0.400);因此采用Stepwise逐步回归法,建立了多元回归模型。结果表明,ρ2、ρ3、ρ4、ρ54波段反射率和有效叶面积植被指数(SLAVI)结合建立的多元回归模型,可用于森林生物量的遥感估测,估测的R2值达0.730,留一交叉验证均方根误差RMSE最小,达33.712。利用2008年的全覆盖TM影像,结合丰宁遥感分类图像,获得了丰宁县2008年森林地上生物量分布图,森林植被总生物量为1.805×107t。
By using TM images and filed sample measurement data,the paper studied the remote sensing techniques for estimating the forest aboveground biomass in Fengning Man Autonomous County,Hebei Province,selected six bands of reflectance values and seven kinds of vegetation index including DVI,NDVI,PVI,RVI,VI3,SLAVI and SAVI,analyzed the correlation between the aboveground biomass and each factor,and concluded that their correlation coefficient was small,all smaller than 0.400;thus,the Stepwise regression method was adopted to establish multiple regression model,results indicated that the multiple regression model established by four brand reflectivity of ρ2,ρ3,ρ4 and ρ5 spectral reflectance and Effective leaf area vegetation index（SLAVI） could be used for estimating forest biomass,the coefficient of determination（R2） reached 0.730,the root mean square error RMSE was 33.712,which was the smallest.The distribution map of forest biomass in Fengning Coutry was made by using TM images and the vegetation classification map acquired in 2008.The map reflected that the total aboveground biomass in 2008 in Fengning is 1.805×107 t.