大气与环境光学学报 ›› 2016, Vol. 11 ›› Issue (2): 125-133.

• 环境光学监测技术 • 上一篇    下一篇

基于航空偏振数据的地表BPDF模型优化研究

舒存铭1,2, 孙晓兵1, 王涵1,2, 龚冠源1,2   

  1. (1中国科学院安徽光学精密机械研究所中国科学院通用光学定标与表征技术重点实验室, 合肥 230031; 
    2中国科学院大学, 北京 100049)
  • 收稿日期:2015-03-17 修回日期:2015-03-27 出版日期:2016-03-28 发布日期:2016-03-18
  • 通讯作者: 舒存铭(1988-),男,湖北黄冈人,硕士研究生,主要从事偏振光学遥感方面的研究。 E-mail:shucunming2008@163.com
  • 作者简介:舒存铭(1988-),男,湖北黄冈人,硕士研究生,主要从事偏振光学遥感方面的研究。
  • 基金资助:

    中国科学院科技创新重点项目(KGFZD-125-13-006)、国家973计划项目(613225-02)资助

Model Optimization of tDidirectional Polarization Reflectance Distribution Function of Land Surface Based on Airborne Polarimetric Data

SHU Cunming1,2, SUN Xiaobing1, WANG Han1,2, GONG Guanyuan1,2   

  1. (1 Key Laboratory of Optical Calibration and Characterization, Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei 230031, China; 
    2 University of Chinese Academy of Siences, Beijing 100049, China)
  • Received:2015-03-17 Revised:2015-03-27 Published:2016-03-28 Online:2016-03-18

摘要:

针对航空偏振气溶胶反演涉及的地表双向偏振辐射分布函数(BPDF)特性问题,分析了地表双向偏振反射分布物理机理,结合经典地表BPDF模型参数特点,提出了基于阴影方程和归一化植被指数的优化地表BPDF模型。利用航空辐射偏振遥感数据,验证了新模型的精度,相对于实测结果,平均相对偏差小于9.9%。与三种经典模型模拟结果相比,新模型值随不同观测角度变化时,其一致性很好。根据新模型精度,还模拟了地表偏振反射率偏差与气溶胶光学厚度反演误差的关系,发现地表偏振反射率偏差为4×10-4时,带来反演气溶胶光学厚度误差小于0.03,满足气溶胶偏振反演应用要求。

关键词: 遥感, 地表, 偏振, 模型

Abstract:

For the characteristic problem of bidirectional polarization reflectance distribution function(BPDF) in airborne polarimetric aerosol optic depth retrieval, the physics mechanism of BPDF is analyzed, and a BPDF model is put forward combining the characteristic BPDF model with the four classical models. This model is based on the shadow function and normalized difference vegetation index factor. The accuracy of new model is validated by the data of atmosphere multi-angle polarization radiometer, and the relative error is less than 9.9%. Compared with the three classic models, the new model shows good consistency when the viewing angle changes. Based on the accuracy of the new model, the relationship is simulated between the bias of polarized reflectance of land surface and the error of the retrieved aerosol optic depth. The results indicate that only less than 0.03 of the absolute error of the retrieved aerosol optic depth is caused when the polarized reflectance of land surface changed 4×10-4, and the application requirement of polarized aerosol retrieve can be satisfied.

Key words: remote sensing, land surface, polarization, model

中图分类号: