Machine learning emulators of dynamic global vegetation models for fast gridded carbon budget predictions
physics.ao-ph
Submitted: 2026-11-03
Updated: 2026-11-03
Comments: 6 figures; Methods and Supplementary Information (10 figures, 6 tables) included
Code: https://github.com/agiamason/NRT_LAND
Project page: https://mdosullivan.github.io/GCB
License: http://creativecommons.org/licenses/by/4.0/
The gist: The land carbon sink explains most of the variability of the atmospheric CO2 growth rate, but the Global Carbon Budget estimates it with dynamic global vegetation models (DGVMs) run once a year.
Terminology
Abstract
The land carbon sink explains most of the variability of the atmospheric CO2 growth rate, but the Global Carbon Budget estimates it with dynamic global vegetation models (DGVMs) run once a year. Emulators could provide it sooner if they reproduce each model outside its training period. Here we show that convolutional neural network emulators of 19 of these models, driven by ERA5 climate and their own past predictions, reproduce the global land sink of these models and its spread over two years not used in training. Global monthly net biome production (NBP) anomalies were reproduced with a median R squared of 0.81. The error of the multi-model mean annual NBP was at most 0.23 GtC yr-1. At the grid-cell level, the median R squared of monthly anomalies was 0.30, partly because of errors that accumulate in the emulated NBP history.
Sources
Related papers
- Anthropogenic Heat in Urban Climate Systems: Forcing, Sensitivity, and Feedback
- NORi: An ML-Augmented Ocean Boundary Layer Parameterization
- A Checklist to assess the energy and carbon impacts of ML/AI applications in Earth System Modeling
- Skillful Thunderstorm Forecasts Learned Directly from Observations
- Wind vector retrieval from miniaturized wave-enabled sea-surface drifters
- Stochastic coupling of climate variables and ice volume over the Late Pleistocene glacial cycles