12 September 2023 in Geo-spatial Information Science
15 December 2025 in IEEE Transactions on Geoscience and Remote Sensing
Carbon dioxide (CO2), a major greenhouse gas, has a profound impact on global climate change. Satellite remote sensing is a crucial approach for monitoring the column-averaged dry-air mole fraction of CO2 (XCO2). Currently, XCO2 retrieval primarily relies on fully physical algorithms, where the radiative transfer model (RTM) imposes a significant computational burden during the forward simulation process. This inefficiency limits the capability for near-real-time monitoring, especially for large-scale retrievals expected from next-generation wide-swath carbon satellites. To address this challenge, this study proposes a deep learning-based surrogate modeling approach to accelerate RTM simulations and enable fast and accurate spectral radiance prediction, thereby facilitating efficient XCO2 retrieval. Specifically, we design a two-stage neural network surrogate model for the SCIATRAN RTM, combining an AutoEncoder and a deep neural network. The former effectively reduces the dimensionality of hyperspectral data while preserving spectral fidelity, and the latter captures complex nonlinear mapping between atmospheric/surface states and satellite-observed radiance under diverse spatiotemporal scenarios. Results demonstrate that the surrogate model significantly improves computational efficiency while maintaining high accuracy, achieving millisecond-level prediction speeds. The model performs robustly across test datasets, finer-step extrapolation scenarios, and real-conditions validation using OCO-2 observations, yielding R² values close to 1.00 and a minimum RMSE of 2.87×10⁻⁴ W/(m²∙sr∙nm). Meanwhile, we extend the surrogate model to the calculation of Jacobian matrices of atmospheric and surface parameters, demonstrating high fidelity and stability. Preliminary results demonstrate the potential of the surrogate model for XCO2 retrieval when embedded within the ACOS retrieval algorithm. This study provides a novel and efficient solution for high-speed XCO2 retrieval, supporting real-time CO2 monitoring and carbon flux estimation.
27 February 2026 in IEEE Transactions on Geoscience and Remote Sensing
CO2 emissions from power plants are considered the primary source of anthropogenic emissions. Estimating CO2 emissions from power plants is essential for tracking global carbon emissions and understanding the global carbon cycle. Satellite remote sensing offers an independent approach to monitoring global CO2 emissions. The first space-borne IPDA LiDAR, Aerosol and Carbon Detection LiDAR (ACDL) instrument, onboard the DQ-1, was successfully launched in April 2022. It enables global measurements to be taken during the day and at night. In addition, the ACDL instrument can capture finer plume patterns, attributed to its small footprint of 70 m. However, there are still uncertainties in the CO2 emissions quantified from satellite observations. To reduce the impact of the uncertainty of the wind field on the CO2 emission estimation, we propose a parameter, plume skewness, to evaluate the wind direction based on the ACDL observations. The effective DAOD enhancements are identified, and wind direction inversion is conducted on 6 power plants using the ACDL observations from June to December 2022. Then, the assessment of power plant emissions is performed with the satellite-derived wind direction, and the CO2 emission results are validated against the EPA inventory. Results indicate that the CO2 emissions with satellite-derived wind direction consist well with the reported emissions, with a bias of less than 20%. Furthermore, the CO2 emission estimates with satellite-derived wind direction are compared with the results with ERA5 wind direction. The mean bias reduced from 11% to 4%, indicating the reliability of satellite-derived wind direction.
13 April 2026 in Geoscience Frontiers
2 February 2026 in ISPRS Journal of Photogrammetry and Remote Sensing