On combining climate models into weighted ensembles
- 1Alfred-Wegener-Institut Helmholtz-Zentrum für Polar- und Meeresforschung (Bremerhaven, Germany)
Abstract
Several methods have been proposed and used to refine estimates of future climate change based on combined output from comprehensive climate models. While previously the so-called model democracy approach was used to combine model outputs, where every model is given equal weight, it is now widely accepted that using model weights that account for model performance and model independence is necessary to obtain more reliable results. However, most existing approaches rely, implicitly or explicitly, on a similar statistical basis, while describing things in different ways. Here we distinguish between approaches that are based on the performance of individual models (individual performance weighting) and approaches that are based on the performance of the weighted ensemble as a whole (ensemble performance weighting). At the same time, we formulate both in probabilistic Bayesian terms to make their application and comparison straightforward. Using simple constructed examples, we demonstrate that the ensemble performance weighting approach implicitly accounts for co-dependencies among models, which arguably makes the computation of independence weights for the purpose of model weighting obsolete. We also show that a set of weighted models within the ensemble weighting approach will naturally tend to artificially reduce uncertainty and that this is strongly influenced by the choice of the prior distribution over weight vectors. The distinction between individual and ensemble performance weighting is both methodological and conceptual. Formulating both approaches in general probabilistic Bayesian terms as done here, can serve as a common basis for future developments with regard to ensemble model weighting in Earth system science. The discussed concepts are generally applicable across domains and are thus possibly relevant for other modeling communities as well.
Versions
Related
- Reduction of uncertainty in near-term climate forecast by combining observations and decadal predictions
- A multi-scale dataset combining 3D plant architecture, leaf gas exchange, and whole-plant fluxes in young oil palm under controlled climate scenarios
- GCMagicc v1: a fast generative emulator for multivariate climate-impact ensembles
- Combining Thermochronology and Geochronology to Constrains Orogenic Processes in the Alps and Tibet
- Investigating hydrological modeling uncertainties in the Mediterranean region by combining precipitation and soil moisture products
BibTeX
@article{grusdt2026combining,
title = {{On combining climate models into weighted ensembles}},
author = {Britta Grusdt and Mahé Perrette and Alexander Robinson},
journal = {Earth System Dynamics},
year = {2026},
month = oct,
doi = {10.5194/esd-17-1589-2026},
url = {https://esd.copernicus.org/articles/17/1589/2026/},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - On combining climate models into weighted ensembles
AU - Britta Grusdt
AU - Mahé Perrette
AU - Alexander Robinson
PY - 2026
DA - 2026/10/28
JO - Earth System Dynamics
DO - 10.5194/esd-17-1589-2026
UR - https://esd.copernicus.org/articles/17/1589/2026/
AB - Several methods have been proposed and used to refine estimates of future climate change based on combined output from comprehensive climate models. While previously the so-called model democracy approach was used to combine model outputs, where every model is given equal weight, it is now widely accepted that using model weights that account for model performance and model independence is necessary to obtain more reliable results. However, most existing approaches rely, implicitly or explicitly, on a similar statistical basis, while describing things in different ways. Here we distinguish between approaches that are based on the performance of individual models (individual performance weighting) and approaches that are based on the performance of the weighted ensemble as a whole (ensemble performance weighting). At the same time, we formulate both in probabilistic Bayesian terms to make their application and comparison straightforward. Using simple constructed examples, we demonstrate that the ensemble performance weighting approach implicitly accounts for co-dependencies among models, which arguably makes the computation of independence weights for the purpose of model weighting obsolete. We also show that a set of weighted models within the ensemble weighting approach will naturally tend to artificially reduce uncertainty and that this is strongly influenced by the choice of the prior distribution over weight vectors. The distinction between individual and ensemble performance weighting is both methodological and conceptual. Formulating both approaches in general probabilistic Bayesian terms as done here, can serve as a common basis for future developments with regard to ensemble model weighting in Earth system science. The discussed concepts are generally applicable across domains and are thus possibly relevant for other modeling communities as well.
SN - 2190-4987
SN - 2190-4979
ER -