基于遥感数据定位老龄树群  

Locating old trees based on remote sensing data

作　　者：王博闻 史江峰[1,2] 史逝远 张伟杰 马晓琦 赵业思 Wang Bowen;Shi Jiangfeng;Shi Shiyuan;Zhang Weijie;Ma Xiaoqi;Zhao Yesi(School of Geography and Ocean Science,Nanjing University,Nanjing,210023,China;Laboratory of Tree-RingResearch,School of Environment,The University of Arizona,Tucson,AZ85721,USA)

机构地区：[1]南京大学地理与海洋科学学院,南京210023 [2]亚利桑那大学树木年轮研究实验室,Tucson,AZ85721,USA

出　　处：《南京大学学报（自然科学版）》2019年第4期699-707,共9页Journal of Nanjing University（Natural Science）

基　　金：江苏省科技厅自然科学基金(BK20161394);国家自然科学基金(41671193);国家留学基金委青年骨干教师出国研修(201806195033);国家重点研究发展计划(2016YFA0600503)

摘　　要：在野外找到老龄树群,是树木年轮气候学研究的一个关键环节.目前还没有在大尺度空间上连续的高精度的树龄数据可供使用,尝试建立一种基于遥感数据定位老龄树群的方法.以30m分辨率的卫星Landsat8OLI(Operational Land Imager)遥感影像一景为例,首先在该影像范围内收集了22个已发表的树轮宽度年表长度数据,根据经纬度定点提取年表对应的归一化植被指数(Normalized Difference Vegetation Index,NDVI)值,然后用R语言建立树轮宽度年表的长度与NDVI之间的一元线性回归模型,利用遥感影像数据良好的监测地表空间异质性的能力,实现对树轮年表长度在空间上连续的高分辨率的估算.将该信息作为树龄的一种近似替代,可以辅助树木年轮工作者快速、准确、定量地寻找到老龄树群。In dendroclimatological studies,old trees should be sampled in order to reconstruct longer past climate. It has always taken a lot of efforts for dendrochronologists to find old trees in the fields,especially where old trees were cut for different purposes. However, there are not continuous and high - esolution data of tree ages in a large spatial scale at present in China. Therefore, to find a way to quickly locate old tree groups is urgently needed in tree - ring and even ecological communities. In this paper a method is proposed to locate old tree groups that uses the relationship between remote sensing data and published tree age data. The method is based on the well demonstrated recognition that some indicator calculated from remote sensing data and stand ages often have a significant relationship. A Landsat 8 OLI (Operational Land Imager / Thermal Infrared Sensor) 131 / 38 remoting sensing image, acquired on 3 October 2016, is used to calculate NDVI (Normalized Difference Vegetation Index) values. Generally, a tree-ring chronology is established using dozens of tree cores with missing rings and false rings being corrected for each core using the classical cross-dating technique. Therefore,the length of a chronology can represent the stand age well at the sampling site,and the lengths of the chronologies found in the study region are used as the stand age data. The stand age data are from 22 published tree-ring sampling sites which have specific tree-ring width chronologies’lengths. Then, a simple linear regression model between the two proxies is established using R language. Finally,potential locations of old tree stands are indicated using the relationship between NDVI values and the lengths of the chronologies. The method proposed in this study could as an alternative to tree-age information,to help tree-ring researchers locate old tree stands quickly and accurately.

关 键 词：老龄树群 遥感数据 树轮年表长度 归一化植被指数 R语言 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]