生物多样性 ›› 2018, Vol. 26 ›› Issue (8): 862-877.DOI: 10.17520/biods.2018143

• • 上一篇    下一篇

遥感技术在鸟类生态学研究中的应用

雷倩1,2, 李金亚1, 马克明1,*()   

  1. 1 中国科学院生态环境研究中心城市与区域生态国家重点实验室, 北京 100085
    2 中国科学院大学, 北京 100049
  • 收稿日期:2018-05-15 接受日期:2018-08-14 出版日期:2018-08-20 发布日期:2018-09-27
  • 通讯作者: 马克明
  • 作者简介:# 共同第一作者
  • 基金资助:
    国家重点研发计划项目(2017YFC0505800, 2016YFC0500406)和国家自然科学基金(41601439)

Applications of remote sensing technology in avian ecology

Qian Lei1,2, Jinya Li1, Keming Ma1,*()   

  1. 1 State Key Laboratory of Urban and Regional Ecology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 100085
    2 University of Chinese Academy of Sciences, Beijing 100049
  • Received:2018-05-15 Accepted:2018-08-14 Online:2018-08-20 Published:2018-09-27
  • Contact: Ma Keming
  • About author:# Co-first authors

摘要:

获取鸟类活动及生境信息是鸟类生态学研究的基础, 而遥感技术弥补了传统野外调查方法的缺陷, 提供了获取多种信息的新途径。应用遥感技术的鸟类生态学研究热点从最初的种群行为观察, 到栖息地选择, 再到生境适宜性、破碎化及人为干扰探究等, 随着技术的不断发展也在扩展和变化。不同波段或组合下的遥感技术各有所长。光学遥感应用广泛, 尤其是信息量较大的红外波段图像和作为野外鸟巢及物种活动监测常用工具的红外相机; 多光谱图像常用于栖息地制图以及地物识别, 高空间分辨率的数据甚至可对鸟类种群进行直接计数; 高光谱数据则可对光谱特征相似的地物进行更为精确的区分和反演; 激光雷达遥感主要用于栖息地植被结构的三维探测, 为了解鸟类栖息地选择提供更好的依据。微波遥感在飞鸟探测上应用颇多, 近年来多极化数据在复杂栖息地精确制图上也具有优势, 但成本较高、解译复杂且推广度较低。在实际应用中, 遥感数据时空尺度的选择会影响研究结果, 部分遥感反演参数也缺乏生态学意义。多源遥感数据的结合应用能够提升制图分类的精度, 实现数据的时空分辨率互补, 优化鸟类生态研究所需参数。未来的遥感技术在鸟类生态学中的应用应致力于提供更加明确的光谱信息、相对简便的解译方法, 以及更为合理的多源数据组合方式等。

关键词: 微波遥感, 红外, LiDAR, 多光谱, 高光谱, 生境反演, 地物识别

Abstract:

Avian ecological studies tend to center on birds and their habitats. According to the literature, studies in avian ecology have shifted from focusing on behavior and habitat selection to focusing on human disturbance, habitat suitability and habitat structure, which has been made possible partially due to remote sensing (RS) technology. Characteristics and applications of RS data are varied. Here, we assessed various RS methods, considering the current state of avian ecology. Light remote sensing is most commonly used. Infrared trigger cameras and video complement field work to monitor brooding, defensive and other behaviors, while the infrared images contain massive amounts of data. Multi-spectral images are used most frequently for mapping habitat and can directly track species when captured at a high spatial resolution. Hyperspectral data has great potential in classifying objects with similar spectral characteristics. LiDAR data mainly contributes to studies of habitat structure. Researchers have used Radar to monitor flying birds over extended periods of time, where the microwave images with multi-polarization may promote the precision of mapping complex habitats. In practice, we recognize that data scale may affect study results and that some RS inversion model parameters lack ecological significance. Multi-source data could enhance mapping accuracy and provide context for the intersection of spatial and temporal resolutions of images. In the future, RS technology development should pay more attention to provide specific spectral information, more convenient interpretation methods, and more rational multi-source data combinaions, for a better use of them.

Key words: microwave remote sensing, infrared, LiDAR, multi-spectral, hyperspectral, habitat parameter inversion, object identification