文献类型：期刊文献

中文题名：RS在森林病虫害监测研究中的应用

英文题名：Utilization of remote sensing for detecting forest damage caused by insect infestations or diseases

作者：郭志华[1] 肖文发[1] 张真[1] 陈昌洁[1] 赵宪文[2]

第一作者：郭志华

机构：[1]中国林科院森林生态环境与保护研究所;[2]中国林科院资源信息研究所

年份：2003

卷号：12

期号：4

起止页码：73-81

中文期刊名：自然灾害学报

外文期刊名：Journal of Natural Disasters

收录：CSTPCD;;Scopus;北大核心:【北大核心2000】；CSCD:【CSCD2011_2012】；

基金：科技部社会公益研究项目(2001DIA10004);国家林业局重点实验室基金

语种：中文

中文关键词：森林病虫害;失叶;遥感;变化监测;GIS技术

外文关键词：forest insect pests and disease; defoliation; remote sensing; change detection

分类号：S763;TP79

摘要：森林病虫害是影响森林健康的主要因素之一。森林病虫害的快速准确监测、检测及其损害评估无论对森林经营者还是生态学家都具有重要意义。在大的空间尺度上,遥感数据早已被广泛有效地用于检测森林病虫害所引起的失叶现象。目前,在森林病虫害的遥感监测方面所利用的遥感数据主要有LandsatTM,MSS和SPOTHRV等卫星遥感数据和多种航空遥感数据;此外,各种地面辅助地理数据和野外调查数据可以提高对森林病虫害的监测精度。不同类型的森林往往感染不同的森林病虫害,而不同的森林病虫害所引起的森林受损症状不同,有的病虫害导致森林反射光谱的显著变化,有的则导致森林大量失叶。因此,对不同病虫害的遥感监测方法不同。随着技术的进步,传统的图像增强、图像分类和影像差等病虫害监测技术已显粗糙。近年来,更多、更复杂的图像处理方法被广泛用于检测病虫害引起的森林变化,各种数学方法和GIS技术也为病虫害的遥感监测提供了新的活力。虽然森林病虫害的遥感监测面临挑战,但随着计算机软硬件技术和遥感技术的进步,森林病虫害的遥感检测依然具有广阔的发展空间。
Forest insect pest and disease is one of the main factors influencing forest health. Detecting and monitoring its outbreaks, and estimating the area affected, are therefore important for both forest managers and forest ecologists. On a large spacial scale, remote sensing has widely been used to detect and monitor forest damage caused by insect pests. At present, the remote sensing data that could be used to detect forest damage mainly includ Landsat TM, MSS, SPOT HRV, and a lot of aerial remote sensing data. Besides, the ground auxiliary data such as digital elevation model (DEM) and digital stand data, and other field investigation data such as insect population and etc., could be used to improve monitoring precision. Usually, different kinds of insect or disease infest different types of forest, and the different insect or disease could result in different damage symptoms. Some insect infestations or diseases resulted in a marked spectral change ; on the other hand, some insect infestations or diseases caused a general loss of foliage. Therefore, the different change detection methods are required for mapping damage of forests such as insect defoliation. Relative robust change detection methods were image differencing and image classification, but they did not adequately address differences in sun elevation angles, atmospheric conditions, or phonological changes between images recorded at different dates. So, more sophisticated change detection methods perform transformations of the image space to detect forest changes. On the other hand, many mathematical methods such as wavelet analysis, neural network analysis and cellular automata network and etc., and geographical information system had been developed and applied to forest pest detection.Although there were many challenges in detecting forest insect defoliation and disease damage, with the progress of hardware and software of computers and progress with remote sensing technologies, remote sensing detection will still be a very important approach to identify changes in forest health over large space and long time. 