基于统计与光程分布的二氧化碳反演方法
发布时间：2018-09-09 19:34
【摘要】：为了研究气候变化,需要实现遥感卫星对二氧化碳(CO_2)的高精度测量。气溶胶和透射率较高的薄卷云的散射是影响大气中CO_2反演精度的主要环境因素。结合主成分分析(PCA)的统计方法和光程概率分布的密度函数(PPDF)方法,利用PCA方法得到大气CO_2反演的先验值,避免了因偏差过大而导致的运算结果无法接近真值;基于3层PPDF模型,解决了薄卷云和气溶胶散射引起的光子路径变化而导致的吸收谱线变化的问题。结果表明,PCA方法和PPDF方法联合反演的反演精度得到明显提高;对2013年塔克拉玛干沙漠GOSAT数据的反演结果进行分析,采用单一的PPDF方法得到的反演结果的方差为3.5,两种方法相结合得到的反演结果的方差为1.4,优于日本国立环境研究所(NIES)提供的反演方差(1.6)。
[Abstract]:In order to study climate change, high precision measurement of carbon dioxide (CO_2) by remote sensing satellites is needed. The scattering of aerosols and thin cirrus with high transmittance is the main environmental factor affecting the accuracy of CO_2 inversion in the atmosphere. Combining the statistical method of principal component analysis (PCA) and the density function (PPDF) method of optical path probability distribution, a priori value of atmospheric CO_2 inversion is obtained by using PCA method, which avoids that the calculation result caused by the deviation is too large to approach the true value, and based on the three-layer PPDF model, The problem of absorption line variation caused by photon path change caused by thin cirrus and aerosol scattering is solved. The results show that the inversion accuracy of the combined PPDF method and the GOSAT data of the Taklimakan Desert in 2013 has been improved obviously, and the inversion results of the GOSAT data in the Taklimakan Desert in 2013 have been analyzed. The variance of the inversion result obtained by using a single PPDF method is 3.5, and the variance of the inversion result obtained by the combination of the two methods is 1.4, which is better than the inversion variance provided by (NIES) of the National Institute of Environment of Japan (1.6).
【作者单位】： 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;中国科学技术大学;
【基金】：国家自然科学基金(41175037);国家自然科学基金青年科学基金(41601393) 高分辨对地观测系统重大专项(民用部分)(32-Y20A17-9001-15/17)
【分类号】：X87
本文编号：2233364
[Abstract]:In order to study climate change, high precision measurement of carbon dioxide (CO_2) by remote sensing satellites is needed. The scattering of aerosols and thin cirrus with high transmittance is the main environmental factor affecting the accuracy of CO_2 inversion in the atmosphere. Combining the statistical method of principal component analysis (PCA) and the density function (PPDF) method of optical path probability distribution, a priori value of atmospheric CO_2 inversion is obtained by using PCA method, which avoids that the calculation result caused by the deviation is too large to approach the true value, and based on the three-layer PPDF model, The problem of absorption line variation caused by photon path change caused by thin cirrus and aerosol scattering is solved. The results show that the inversion accuracy of the combined PPDF method and the GOSAT data of the Taklimakan Desert in 2013 has been improved obviously, and the inversion results of the GOSAT data in the Taklimakan Desert in 2013 have been analyzed. The variance of the inversion result obtained by using a single PPDF method is 3.5, and the variance of the inversion result obtained by the combination of the two methods is 1.4, which is better than the inversion variance provided by (NIES) of the National Institute of Environment of Japan (1.6).
【作者单位】： 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;中国科学技术大学;
【基金】：国家自然科学基金(41175037);国家自然科学基金青年科学基金(41601393) 高分辨对地观测系统重大专项(民用部分)(32-Y20A17-9001-15/17)
【分类号】：X87
【相似文献】
中国期刊全文数据库 前2条
1 刘毅;杨东旭;蔡兆男;;中国碳卫星大气CO_2反演方法:GOSAT数据初步应用[J];科学通报;2013年11期
2 ;[J];;年期
，本文编号：2233364
本文链接：https://www.wllwen.com/shengtaihuanjingbaohulunwen/2233364.html
