Beyond Bias: Radon-Based Technique for Reducing Uncertainty in Greenhouse Gas Verification Frameworks

The accuracy of greenhouse gas (GHG) emission estimates is significantly limited by uncertainties in atmospheric transport models (ATMs). These uncertainties largely arise from difficulties in accurately representing sub-grid turbulence and mixing processes. Furthermore, the use of modelled meteorological data to filter observations before inversion frameworks results in the exclusion of 40–75% of continuous GHG measurements, thereby reducing the reliability of emission estimates.

To overcome these challenges, we propose the use of radon measurements - a naturally occurring radioactive noble gas with well-characterised sources and sinks. Radon will be used as a metric to define atmospheric mixing classes, providing a novel approach to validate ATM performance and address its inherent uncertainties. These mixing classes, which reflect varying atmospheric stability conditions, offer a valuable benchmark for evaluating model parameterisations and meteorological inputs.

Our study utilises radon measurements from the Weybourne Atmospheric Observatory (UK) and Cabauw Tower (Netherlands) to assess the reliability of meteorological inputs and parameterisation in ATMs. Preliminary results demonstrate that radon-derived mixing classes can reduce biases in data filtering while improving the representation of atmospheric transport dynamics. This innovative method helps to bridge gaps in current inversion frameworks, enabling more accurate GHG emission estimates and supporting the development of evidence-based climate policies.