Foundation models improve 96-hour carbon intensity forecasts in nordic zones
Zero-shot time-series foundation models against the Khepri v1.5 baselines for 96-hour carbon-intensity forecasts in nine Nordic bidding zones: a preregistered comparison
Machine Learning
Summary
Forecasting carbon intensity helps manage energy use and reduce emissions. The authors compared powerful new time-series models against standard forecasting methods for predicting carbon intensity up to four days ahead in Nordic regions. They found the new models generally gave more accurate forecasts than traditional methods across multiple zones. The results were carefully checked and confirmed on past data, though a small more recent test could not definitively prove superiority. Overall, the new models show promise for better short-term carbon intensity predictions.
What this means in practice
- •For energy grid operators: Use foundation models to improve 96-hour forecasts of carbon intensity for better grid management in Nordic bidding zones.
- •For power market analysts: Enhance short-term carbon intensity predictions in Scandinavian power markets by integrating pretrained time-series models without retraining.
Authors
Eirik Botten Nicolaysen
Abstract
We compare three zero-shot time-series foundation models (Timestral-5M, Timestral-209M and Chronos-2, 8192-hour context, no covariates) with the Khepri v1.5 SARIMA and GBM baselines, with their code and hyperparameters unchanged, for 96-hour forecasts of hourly carbon intensity in the Nordic bidding zones NO1-NO5 and SE1-SE4. The design, the analysis code and a minimum detectable effect (MDE) were timestamped before any foundation-model forecast by an OpenTimestamps proof (Bitcoin-confirmed) and the GPG-signed tag prereg-v1; the proofs are archived in Zenodo (DOI 10.5281/zenodo.23881315). The preregistration was also submitted to OSF Registries 2026-11-02 07:49 UTC (osf.io/hncz5; embargo requested, pending OSF approval). On the 2025 test (361 daily origins, contamination not ruled out) all three models have a lower zone-averaged MAPE than both baselines, by 0.952 to 1.347 percentage points (Diebold-Mariano with the Harvey-Leybourne-Newbold correction, all p < 0.0001), and each beats SARIMA in 9 of 9 zones. On a post-release holdout of 14 origins from 2026-10-16 the point estimates have the same sign (0.729 to 1.326 percentage points lower), but all lie within the preregistered MDE of 2.989 points and are reported as not determined. For calibration (RQ4) the preregistration locked a Diebold-Mariano-HLN test on CRPS; the MDE applies to MAPE only. Holdout CRPS point estimates (n = 14) are 1.1114 (Timestral-5M), 1.0733 (Timestral-209M) and 1.1198 (Chronos-2) against 1.4165 gCO2-eq/kWh for SARIMA (each p < 0.0001). The SARIMA quantiles are constructed from the fitted model's Gaussian predictive distribution (deviation D-03), not taken from a native v1.5 forecast, so the difference may partly reflect the SARIMA intervals. The contamination indicator does not exceed its preregistered threshold for any comparison; with 14 holdout origins it has low power (HAC standard errors 0.124 to 0.360 pp).