遥感叶面积指数产品提取自然植被物候期对比  

Comparison of Natural Vegetation Phenology Metrics from Remote Sensing LAI Products

作　　者：李铮[1,2,3] 柏延臣[1] 何亚倩 

机构地区：[1]北京师范大学遥感科学国家重点实验室遥感与地理信息系统研究中心地理学与遥感科学学院,北京100875 [2]北京师范大学地表过程与资源生态国家重点实验室,北京100875 [3]北京师范大学减灾与应急管理研究院,北京100875 [4]Department of Geology and Geography,West Virginia University

出　　处：《遥感技术与应用》2015年第6期1103-1112,共10页Remote Sensing Technology and Application

基　　金：国家863计划项目(2012AA12A305);国家973计划项目(2013CB733403);国家自然科学基金项目(41271347)

摘　　要：物候是指示气候变化的关键因子,遥感技术的快速发展为物候监测提供了新的途径。遥感叶面积指数(LAI)产品包含了主要的物候信息,并广泛应用于植被物候的监测。了解不同数据产品在提取植被物候信息上的差异是评价遥感产品对物候期监测适用性的重要方面。以东北三省为研究区域,使用非对称性高斯函数拟合法进行数据平滑,利用动态阈值法提取MODIS、CYCLOPES和GLASS叶面积指数(LAI)产品的生长季开始时间(SGS)、生长季结束时间(EGS)和生长季长度(LGS)。研究表明:MODIS和GLASS产品提取的SGS、EGS和LGS比较接近,整体上一致性较好;CYCLOPES产品提取的SGS多数情况下晚于MODIS和GLASS产品而EGS早于MODIS和GLASS产品。通过可利用的实地物候观测数据验证表明:MODIS和GLASS产品提取林地的SGS与物候观测值比较接近,EGS略晚于物候观测值,CYCLOPES产品提取的林地的SGS和EGS更加可靠。Phenology is the key indicator of climate change,and the rapid development of remote sensing provides a new method to monitor phenology.It is really significant of acknowledging the differences in extracting vegetation phenology to assess the availability of remote sensing product in monitoring phenology.Taking three northeast provinces of China as study area,our study exploits asymmetric Gaussian function fitting method to smooth the data,and uses dynamic threshold method to extract the start of the growing season（SGS）,the end of the growing season（EGS）,and the length of the growing season（LGS）of MODIS,CYCLOPES and GLASS leaf area index data product.The results show that MODIS and GLASS data product have similar results in extracting SGS,EGS and LGS,and the consistency between MODIS and GLASS data product is better on the whole;The SGS extracted from CYCLOPES is later than MODIS and GLASS data product in the overwhelming majority of cases,but the EGS is earlier than MODIS and GLASS data product,thus it causes the LGS shorter than the other two data products.The validation of phonological phase by taking advantages of the phonological observation data show that the SGS from MODIS and GLASS data product is similar to the phonological observations,and the EGS is a litter later than the phonological observations;The SGS and EGS from CYCLOPES data product are similar to the phonological observations and it indicates that the SGS and EGS from CYCLOPES data product in woodland are much more reliable.

关 键 词：物候 叶面积指数(LAI) 生长季开始时间(SGS) 生长季结束时间(EGS) 生长季长度(LGS) 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]