Application of PRIM for understanding patterns in carbon dioxide model-observation differences

Gerken, Tobias; Davis, Kenneth J.; Keller, Klaus; Feng, Sha

doi:https://doi.org/10.5194/acp-25-13327-2025

Articles | Volume 25, issue 20

https://doi.org/10.5194/acp-25-13327-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/acp-25-13327-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 25, issue 20

Research article

|

22 Oct 2025

Research article |

| 22 Oct 2025

Application of PRIM for understanding patterns in carbon dioxide model-observation differences

Tobias Gerken, Kenneth J. Davis, Klaus Keller, and Sha Feng

Supplement

https://doi.org/10.5194/acp-25-13327-2025-supplement

Data sets

ACT-America: L3 Merged In Situ Atmospheric Trace Gases and Flask Data, Eastern USA (Version 1.2) Kenneth J. Davis et al. https://doi.org/10.3334/ORNLDAAC/1593

Full WRF-Chem Output in Support of the NASA Atmospheric Carbon and Transport (ACT)-America Project (7/1/2016 – 7/31/2019) Sha Feng et al. https://doi.org/10.26208/RQF5-Q142

Model code and software

TobGerken/PRIM_Analysis_Gerken2025: v0.5 Tobias Gerken https://doi.org/10.5281/zenodo.14727931

Download

Article (2746 KB)
Full-text XML

Short summary

We apply the Patient Rule Induction Method (PRIM) technique to airborne CO₂ and meteorological data to better understand atmospheric conditions and implications for carbon dioxide model-observation-mismatches. We found PRIM is capable of separating observations from different seasons and levels based on atmospheric conditions. Large magnitude carbon dioxide model-observation-differences were associated with non-typical atmospheric conditions, with implications for transport model evaluation.