Downscaled XCO₂ Estimation Using Data Fusion and AI-Based Spatio-Temporal Models

One of the well-known greenhouse gases (GHGs) produced by anthropogenic human activity is carbon dioxide (CO₂). Understanding the carbon cycle and how negatively it affects the ecosystem requires analysis of the rise in CO₂ concentration. This work aims to map CO₂ concentration for the entire surface, making it useful for regional carbon cycle analysis. Here, column-averaged CO₂ dry mole fraction, called XCO₂, measured by the orbiting carbon observatory-2 (OCO-2) satellite, is used. Because of spectral interference by the clouds and aerosols, there are many missing footprints in the Level-2 swath of OCO-2, making it disruptive to understand any assessment related to the carbon cycle. The objective of this work is to predict 1 km2 XCO₂ using data resampling and machine learning models. This work achieves a minimum mean absolute error (MAE) and root mean square error (RMSE) of 0.3990 and 0.8090 ppm, using the monthly models.