New Study Uses Machine Learning to Retrieve Carbon Monoxide Data from GIIRS Satellite

A recent study presents a radiative transfer model-driven machine learning technique for retrieving carbon monoxide from the world's first hyperspectral Geostationary Interferometric Infrared Sounder (GIIRS) onboard Fengyun-4B (FY-4B) satellite, providing complementary insights into air quality and pollutant transport over East Asia. The study, published in the Journal of Remote Sensing, takes carbon monoxide as an example to explore the reliability of retrieval using an efficient machine learning method compared to traditional physical method.

The core idea of this machine learning approach is to rapidly convert the CO spectral features extracted from GIIRS measurements into columns through a trained model and simultaneously estimate the uncertainty based on the error propagation theory. Comparisons with the retrieval results of traditional physical methods and ground-based observations reveal consistent spatial distribution and temporal variation across these different datasets. Dr. Dasa Gu, a leading researcher on the project, stated that machine learning methods have the potential to provide reliable CO products without the computationally intensive iterative process required by traditional retrieval methods.