State Key Laboratory of Resources and Environment Information System, Institute of Geographical Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China; College of Resources and Environment, University of Chinese Academy of Sciences, Beijing 100049, China; Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences, Beijing 100101, China.
State Key Laboratory of Resources and Environment Information System, Institute of Geographical Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China; College of Resources and Environment, University of Chinese Academy of Sciences, Beijing 100049, China.
Sci Total Environ. 2017 Dec 1;601-602:1575-1590. doi: 10.1016/j.scitotenv.2017.06.018. Epub 2017 Jun 9.
Ground observations can capture CO concentrations accurately but the number of available TCCON (Total Carbon Column Observing Network) sites is too small to support a comprehensive analysis (i.e. validation) of satellite observations. Atmospheric transport models can provide continuous atmospheric CO concentrations in space and time, but some information is difficult to generate with model simulations. The HASM platform can model continuous column-averaged CO dry air mole fraction (XCO) surface taking TCCON observations as its optimum control constraints and an atmospheric transport model as its driving field. This article presents a comparison of the satellite observations with a HASM XCO surface obtained by fusing TCCON measurements with GEOS-Chem model results. We first verified the accuracy of the HASM XCO surface using six years (2010-2015) of TCCON observations and the GEOS-Chem model XCO results. The validation results show that the largest MAE of bias between the HASM results and observations was 0.85ppm and the smallest MAE was only 0.39ppm. Next, we modeled the HASM XCO surface by fusing the TCCON measurements and GEOS-Chem XCO model results for the period 9/1/14 to 8/31/15. Finally, we compared the GOSAT and OCO-2 observations with the HASM XCO surface and found that the global OCO-2 XCO estimates more closely resembled the HASM XCO surface than the GOSAT XCO estimates.
地面观测可以准确捕捉 CO 浓度，但现有的 TCCON（总碳柱观测网络）站点数量太少，无法支持卫星观测的全面分析（即验证）。大气传输模型可以提供连续的大气 CO 浓度在空间和时间上，但有些信息很难通过模型模拟生成。HASM 平台可以模拟连续柱平均 CO 干空气摩尔分数（XCO）表面，以 TCCON 观测为最佳控制约束，以大气传输模型为驱动场。本文比较了卫星观测与 HASM XCO 表面，该表面通过融合 TCCON 测量值和 GEOS-Chem 模型结果获得。我们首先使用六年（2010-2015 年）的 TCCON 观测和 GEOS-Chem XCO 模型结果验证了 HASM XCO 表面的准确性。验证结果表明，HASM 结果与观测值之间的最大平均绝对误差（MAE）偏差为 0.85ppm，最小 MAE 仅为 0.39ppm。接下来，我们通过融合 TCCON 测量值和 GEOS-Chem XCO 模型结果来模拟 HASM XCO 表面，用于 9/1/14 至 8/31/15 期间。最后，我们将 GOSAT 和 OCO-2 观测与 HASM XCO 表面进行了比较，发现全球 OCO-2 XCO 估计值更接近 HASM XCO 表面，而不是 GOSAT XCO 估计值。
