Data Quality Control and Rapid Retrieval Algorithm for Ground-Based CO2 Concentration Inversion
LI Shu1, 3, YANG Le-yi1, 3, CHU Xiao-xue2, 3*, YE Song1, 3, SHI Hai-liang4, GAN Yong-ying1, 3, WANG Xin-qiang1, 3, WANG Fang-yuan1, 3
1. School of Optoelectronic Engineering, Guilin University of Electronic Technology, Guilin 541004, China
2. School of Life and Environmental Sciences, Guilin University of Electronic Technology, Guilin 541004, China
3. Guangxi Key Laboratory of Optoelectronic Information Processing, Guilin 541004, China
4. Anhui Institute of Optics and Fine Mechanics, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031, China
Abstract：Accurate monitoring of atmospheric CO2 concentrations is crucial for climate change mitigation. Ground-based remote sensing provides high spatiotemporal resolution data, yet its accuracy is affected by multiple factors. This study utilizes the SCIATRAN radiative transfer model to simulate radiation transmission processes under various observational conditions, analyzing impacts of a priori profiles, spectral resolution, solar zenith angle (SZA), relative azimuth angle (RAA), temperature, aerosol optical depth (AOD), and boundary layer humidity on CO2 retrieval. The following data quality control standards are established: (1) During observations, implement angle tolerance limits based on predicted daily CO2 levels: SZA≤1.5° and RAA≤28° for high-concentration periods, with stricter thresholds of SZA≤1° and RAA≤27° for low-concentration periods. (2) Retrieval processes must incorporate real-time temperature and pressure profile data, such as the ERA5 reanalysis dataset (European Centre for Medium-Range Weather Forecasts) featuring dynamically updated parameters. (3) Urban-type aerosols are adopted, excluding data where AOD>0.3 or humidity>80%. A novel Global-Local Synergistic Optimization (GLSO) algorithm is proposed by integrating the strengths of the Genetic Algorithm (GA) and the Levenberg-Marquardt (L-M) method. Implementation of GLSO on EM27/SUN observation data demonstrates significant improvements: compared with the conventional L-M method, GLSO reduces iteration counts by 40%, decreases CO2 total column deviation from 0.85% to 0.80%, and achieves a 0.13% discrepancy with TCCON XCO2 data, outperforming the 0.27% deviation from L-M. Furthermore, the GLSO-derived CO2 concentrations show less than 1% deviation from official GOSAT CO2 products. This study establishes a robust framework for enhancing the precision and reliability of ground-based CO2 monitoring through the optimization of observational protocols and advanced retrieval algorithms.
Key words：Ground-based remote sensing; SCIATRAN; Sensitivity analysis; CO2 retrieval
李 树，杨乐怡，储小雪，叶 松，施海亮，甘永莹，王新强，王方原. 地基CO2浓度反演的数据质量控制及快速反演算法[J]. 光谱学与光谱分析, 2025, 45(11): 3226-3234.
LI Shu, YANG Le-yi, CHU Xiao-xue, YE Song, SHI Hai-liang, GAN Yong-ying, WANG Xin-qiang, WANG Fang-yuan. Data Quality Control and Rapid Retrieval Algorithm for Ground-Based CO2 Concentration Inversion. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(11): 3226-3234.
