resources

🔬 Resources & Tools

A curated collection of datasets, benchmarks, and publications from the AI4Carbon Initiative and the broader research community working on machine learning for carbon cycle science.

📊 Datasets & Benchmarks

Explore our collection of open-source datasets and benchmarks designed to accelerate machine learning research in atmospheric transport and carbon cycle modeling.

CarbonBench

CarbonBench

Benchmark dataset for Eulerian atmospheric transport models

Dataset
Learn More

📚 Publications

Recent and seminal publications on machine learning for carbon cycle science, atmospheric transport modeling, and inverse modeling techniques. Publications are tagged by topic:

* indicates joint first authorship

2025

  1. Global Daily Column Average CO2 at 0.1^∘ \texttimes 0.1^∘ Spatial Resolution Integrating OCO-3, GOSAT, CAMS with EOF and Deep Learning
    Franz Pablo Antezana Lopez , Guanhua Zhou , Guifei Jing , and 4 more authors
    Scientific Data, Feb 2025
    Machine learning Monitoring Methods
    PDF
  2. Atmospheric Transport Modeling of CO2 With Neural Networks
    Vitus Benson , Ana Bastos , Christian Reimers , and 3 more authors
    Journal of Advances in Modeling Earth Systems, Feb 2025
    Machine learning Transport Benchmark Methods
    PDF
  3. High-Resolution Greenhouse Gas Flux Inversions Using a Machine Learning Surrogate Model for Atmospheric Transport
    Nikhil Dadheech , Tai-Long He , and Alexander J. Turner
    Atmospheric Chemistry and Physics, May 2025
    Machine learning Transport Inversion
    PDF
  4. Simulating Out-of-Sample Atmospheric Transport to Enable Flux Inversions
    Nikhil Dadheech , and Alexander J. Turner
    EGUsphere, Jul 2025
    Transport Inversion Methods
    PDF
  5. FootNet v1.0: Development of a Machine Learning Emulator of Atmospheric Transport
    Tai-Long He , Nikhil Dadheech , Tammy M. Thompson , and 1 more author
    Geoscientific Model Development, Mar 2025
    Machine learning Transport Benchmark
    PDF
  6. FootNet v1.0: Development of a Machine Learning Emulator of Atmospheric Transport
    Tai-Long He , Nikhil Dadheech , Tammy M. Thompson , and 1 more author
    Geoscientific Model Development, Mar 2025
    Machine learning Transport Benchmark
    PDF
  7. Conditional Diffusion-Based Retrieval of Atmospheric CO2 from Earth Observing Spectroscopy
    William R. Keely , Otto LamminpÀÀ , Steffen Mauceri , and 3 more authors
    Apr 2025
    Machine learning Monitoring Methods
    PDF
  8. Constraining a Data-Driven CO\textsubscript2 Flux Model by Ecosystem and Atmospheric Observations Using Atmospheric Transport
    Samuel Upton , Markus Reichstein , Wouter Peters , and 7 more authors
    EGUsphere, May 2025
    Ecosystem Transport Inversion Monitoring
    PDF

2023

  1. A Machine Learning Emulator for Lagrangian Particle Dispersion Model Footprints: A Case Study Using NAME
    Elena Fillola , Raul Santos-Rodriguez , Alistair Manning , and 2 more authors
    Geoscientific Model Development, Apr 2023
    Machine learning Transport Methods
    PDF

2022

  1. On the Potential of a Neural-Network-Based Approach for Estimating XCO\textsubscript2 from OCO-2 Measurements
    François-Marie Bréon , Leslie David , Pierre Chatelanaz , and 1 more author
    Atmospheric Measurement Techniques, Sep 2022
    Machine learning Monitoring Methods
    PDF

2020

  1. Spatiotemporal Reconstructions of Global CO2-fluxes Using Gaussian Markov Random Fields
    Unn Dahlén , Johan Lindström , and Marko Scholze
    Environmetrics, Sep 2020
    Inversion Methods
    PDF

🔗 Related Resources

📖 Global Greenhouse Gas Watch (G3W)

WMO initiative to establish a comprehensive, integrated and user-focused global system for monitoring greenhouse gases and supporting climate action. Learn more →

🌐 Virtual Institute for Carbon and Climate (VICC)

The Schmidt Sciences Virtual Institute for Carbon and Climate supports interdisciplinary research and collaboration to advance understanding of the carbon-climate system. Learn more →

đŸ’» Join the Community

Subscribe to our mailing list to stay updated on new datasets, publications, and community initiatives.