高分辨率太阳吸收光谱二氧化碳反演中参数的敏感性分析
发布时间：2019-05-16 11:42
【摘要】：基于高分辨率傅里叶变换红外太阳吸收光谱可以准确测量大气中二氧化碳的柱总量。基于光谱反演算法中的前向模型,分析了改变前向模型中不同参数和不同天顶角情况下对反演结果的影响及其原因。选取具有代表性的两天的测量光谱,改变前向模型中连续体倾斜量值、内部视场角、零偏置和多普勒效应四个模型参数,观察不同先验模型参数的扰动对二氧化碳测量结果的影响。结果表明,不同模型参数的扰动引起反演的二氧化碳柱平均干空气混合比(XCO_2)相对偏差并不一样;不同的测量时间,相同参数扰动引起的二氧化碳含量相对偏差也不同。其中连续体倾斜量值的变化对反演结果的影响最大,其变化引起的相对偏差波动范围在0.1%~0.2%之间;内部视场角、零偏置和多普勒效应的变化对反演结果影响较小,引起的二氧化碳的相对偏差分别在-0.045%~0.02%,-0.045%~0.015%和-0.03%~0.04%之间。最后用二氧化碳反演平均核解释了反演误差来源。研究结果对光谱反演算法中模型参数的设定和提高测量的准确度提供了理论依据。
[Abstract]:The total amount of carbon dioxide in the atmosphere can be accurately measured by high resolution Fourier transform infrared solar absorption spectroscopy. Based on the forward model in spectral inversion algorithm, the influence of different parameters and zenith angles in the forward model on the inversion results and its causes are analyzed. The representative two-day measurement spectrum is selected to change the model parameters of continuum inclination, internal field of view angle, zero bias and Doppler effect in the forward model. The effects of disturbance of different prior model parameters on the measurement results of carbon dioxide were observed. The results show that the relative deviation of the average dry air mixing ratio (XCO_2) of carbon dioxide column caused by the disturbance of different model parameters is not the same, and the relative deviation of carbon dioxide content caused by the disturbance of the same parameters is also different at different measurement time. Among them, the variation of continuum inclination value has the greatest influence on the inversion results, and the fluctuation range of relative deviation caused by the change is between 0.1% and 0.2%. The variation of internal field of view angle, zero bias and Doppler effect has little effect on the inversion results, and the relative deviation of carbon dioxide is between-0.045% and 0.015%, 0.045% and 0.015%, respectively. Finally, the source of inversion error is explained by the average kernel of carbon dioxide inversion. The results provide a theoretical basis for the setting of model parameters and the improvement of measurement accuracy in spectral inversion algorithm.
【作者单位】： 中国科学技术大学环境科学与光电技术学院;中国科学技术大学地球与空间科学学院;中国科学院安徽光学精密机械研究所环境光学与技术重点实验室;
【基金】：国家自然科学基金项目(41405134,41575021,91544212,41605018) 安徽省自然科学基金项目(1608085MD79) 重点研发计划青年基金项目(2016YFC0200800)资助
【分类号】：O433
[Abstract]:The total amount of carbon dioxide in the atmosphere can be accurately measured by high resolution Fourier transform infrared solar absorption spectroscopy. Based on the forward model in spectral inversion algorithm, the influence of different parameters and zenith angles in the forward model on the inversion results and its causes are analyzed. The representative two-day measurement spectrum is selected to change the model parameters of continuum inclination, internal field of view angle, zero bias and Doppler effect in the forward model. The effects of disturbance of different prior model parameters on the measurement results of carbon dioxide were observed. The results show that the relative deviation of the average dry air mixing ratio (XCO_2) of carbon dioxide column caused by the disturbance of different model parameters is not the same, and the relative deviation of carbon dioxide content caused by the disturbance of the same parameters is also different at different measurement time. Among them, the variation of continuum inclination value has the greatest influence on the inversion results, and the fluctuation range of relative deviation caused by the change is between 0.1% and 0.2%. The variation of internal field of view angle, zero bias and Doppler effect has little effect on the inversion results, and the relative deviation of carbon dioxide is between-0.045% and 0.015%, 0.045% and 0.015%, respectively. Finally, the source of inversion error is explained by the average kernel of carbon dioxide inversion. The results provide a theoretical basis for the setting of model parameters and the improvement of measurement accuracy in spectral inversion algorithm.
【作者单位】： 中国科学技术大学环境科学与光电技术学院;中国科学技术大学地球与空间科学学院;中国科学院安徽光学精密机械研究所环境光学与技术重点实验室;
【基金】：国家自然科学基金项目(41405134,41575021,91544212,41605018) 安徽省自然科学基金项目(1608085MD79) 重点研发计划青年基金项目(2016YFC0200800)资助
【分类号】：O433
【相似文献】
相关期刊论文 前10条
1 黄梦桥,罗汉;回归模型参数的距离估计方法[J];数学理论与应用;2001年02期
2 李抗美;预测油田产量的伽马模型求解方法——与《求解Γ模型参数的新方法》作者商榷[J];石油勘探与开发;2002年04期
3 朱悦新;我国人口模型参数辨识[J];西安交通大学学报;1982年02期
4 何s铉，
本文编号：2478256
本文链接：https://www.wllwen.com/kejilunwen/wulilw/2478256.html
