Work

Analysis of the Current Situation of CO2 Satellite Observation

Year: 2025

Type: article

Abstract: Accurate quantification of carbon dioxide (CO2) sources and sinks is becoming a key aspect in recent carbon flux research; yet our understanding of satellite performance on regional scales remains insufficient. In this work, the column-averaged dry-air mole fraction of CO2 retrieved from OCO-2 v11.1r and GOSAT v03.05 was evaluated against CarbonTracker (CT) using data from March 2022 to August 2023. Also, the satellite data were validated against those from the Total Carbon Column Observing Network (TCCON) for March 2022 to February 2024. Comparison with CT revealed that both satellites had a general negative bias over land and the best performance in spring. In Southern Hemisphere land regions, the satellites captured monthly variability reliably, with OCO-2 obtaining the most accurate monthly concentrations. In Northern Hemisphere land regions, CT demonstrated the best performance, although both satellites accurately quantified monthly variations in some regions. In tropical land regions, none of the satellites showed superior performance. OCO-2 data showed bias features in sub-regional areas such as East and South Asia. For ocean regions, the bias was the largest in spring. Phase offset, slight underestimation of concentrations, and seasonal biases were found over several ocean regions in OCO-2 time series, whereas GOSAT was unable to provide reasonable results. When comparing TCCON with OCO-2 and GOSAT data, we found systematic errors of −0.12 and −0.56 ppm and root mean square errors of 1.08 and 1.70 ppm, respectively, mainly contributed by topographic variation and aerosol load. The errors were the smallest in spring and larger in summer and winter. Both CT- and TCCON-based analyses indicated that current satellite products may have better performance in desert surfaces. Clouds, aerosols, and surface pressure still challenged OCO-2 retrieval, while the bias-correction process can be emphasized for GOSAT.

Source: Remote Sensing

DOI: 10.3390/rs17213635

Authors: Yuanbo Li, Kun Wu, Yuk Ling Yung, Xiaomeng Wang, Jixun Han

Institutions: Nanjing University of Information Science and Technology, California Institute of Technology

Language: English

Topic: Atmospheric and Environmental Gas Dynamics

Subfield: Global and Planetary Change

Field: Environmental Science

Domain: Physical Sciences

SDG: Climate action

Open Access status: gold

Estimated APC paid: $3,258