Generating high-resolution CO2 maps by Machine Learning-based geodata fusion and atmospheric transport modelling (GeCO)

The spatiotemporal distribution of greenhouse gases and their sources on Earth has so far been studied primarily at relatively coarse resolutions. Reliable information on local emissions and their spatiotemporal variability is largely lacking. Such data are urgently needed to design local climate change mitigation strategies and to validate mitigation efforts.

The GeCO project aims to achieve significant progress by highly resolved measurement and modeling of selected greenhouse gas sources and their turbulent atmospheric dispersion. Both source distribution and gas dispersion are modeled using computational, data-intensive approaches through collaboration between experts in environmental physics and geoinformatics. Cutting-edge methods, including machine learning (ML) and geodata fusion, are applied. Despite the high resolution, computational demands are minimized through a so-called catalog approach, which allows greenhouse gas dispersion to be simulated over extended periods without compromising spatial detail.