地理科学进展 ›› 2016, Vol. 35 ›› Issue (1): 25-34.doi: 10.18306/dlkxjz.2016.01.004

• • 上一篇    下一篇

基于高光谱分析的草地叶绿素含量估算研究进展

马文勇(), 王训明*()   

  1. 中国科学院地理科学与资源研究所,北京 100101
  • 出版日期:2016-01-31 发布日期:2016-01-31
  • 通讯作者: 王训明 E-mail:mawenyong2162@126.com;xunming@igsnrr.ac.cn
  • 作者简介:

    作者简介:马文勇(1988-),男,山东新泰人,博士生,主要从事沙漠化与遥感监测研究,E-mail: mawenyong2162@126.com

  • 基金资助:
    国家杰出青年科学基金项目(41225001)

Progress on grassland chlorophyll content estimation by hyperspectral analysis

Wenyong MA(), Xunming WANG*()   

  1. Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China
  • Online:2016-01-31 Published:2016-01-31
  • Contact: Xunming WANG E-mail:mawenyong2162@126.com;xunming@igsnrr.ac.cn
  • Supported by:
    National Science Fund for Distinguished Young Scholars of China, No.41225001

摘要:

叶绿素是草地进行光合作用最重要的色素,与氮素、蛋白质、水分等其他植被生化参数均有着密切关系,是草地光合能力及生理状况的良好指示剂。利用高光谱数据建模分析是实现大面积草地叶绿素含量估算的一种重要手段。本文将基于高光谱分析估算草地叶绿素含量的方法总结为:基于红边位置及光谱指数的经验模型和辐射传输模型两类。经验模型通过建立叶绿素含量与红边位置、光谱指数之间统计关系来估算叶绿素含量,参数简单,实用性较强;但光谱指数构造形式多样且与草地叶绿素含量关系复杂,在一定程度上影响了叶绿素的估算精度。辐射传输模型以叶绿素含量与辐射能量的作用过程作为其理论基础,模型中参数较多且对估算尺度敏感,有待于进一步完善。目前草地叶绿素估算的研究相对薄弱,专门用于估算的模型较少。未来的工作一方面应致力于发展和改进适宜于草地的光谱指数,同时确定合适的辐射传输模型参数以改进模型对草地的监测效果;另一方面,如何由叶片尺度拓展到冠层尺度进而拓展到像元尺度,从而更好地实现大面积草地叶绿素含量估算,是一项既具有重要意义又有挑战性的工作。

关键词: 高光谱分析, 草地叶绿素, 红边位置, 光谱指数, 辐射传输模型

Abstract:

As the key pigment for photosynthesis, chlorophyll is closely related to nitrogen, protein, moisture, and other biochemical parameters of vegetation. Chlorophyll is a good indicator for photosynthesis activity and physiological state of grasslands. Hyperspectral analysis is a powerful means for accurately retrieving grassland chlorophyll contents in large areas. Based on the existing methods for monitoring grassland chlorophyll contents, this article summarizes and groups such methods into two categories: empirical models of red edge position (REP) / spectral indices, and radiative transfer models. With statistical analysis of the relationships between chlorophyll contents and REP/ spectral indices, empirical models were extensively employed for estimating grassland chlorophyll contents. These models use simple parameters and the REP that is closely related to grassland chlorophyll contents can be conveniently calculated. Due to the diverse structural forms of the spectral indices and the complicated relationships between spectral indices and chlorophyll contents, accuracy of chlorophyll content estimation is still limited, although novel algorithms of REP calculation and, to a lesser extent, new spectral indices, have significantly improved the estimation accuracy of grassland chlorophyll contents. Based on the relationships of chlorophyll contents and radiation energy, radiative transfer models were founded. Multiple parameters are required in the models, which are highly sensitive to the scale of estimation, and these models need to be further improved. Currently there exist few studies on grassland chlorophyll content estimation and only very few models have been developed for grassland chlorophyll content estimation. Future research should focus on developing new indices or improving existing indices and determining appropriate parameters of radiative transfer models to achieve better results for grassland application. In addition, current studies are limited to leaf and canopy scale analyses, and therefore expanding grassland chlorophyll content estimation to the landscape scale will be an important and challenging task.

Key words: hyperspectral analysis, grassland chlorophyll, red edge position, spectral index, radiative transfer model