Probabilistic storyline attribution using machine learning
Abstract
A fundamental goal in climate attribution is to estimate how forced climate change contributes to observed extreme weather events. The storyline attribution method compares an observed weather event, conditional on its atmospheric dynamic state (i.e., atmospheric circulation), in the current, 'factual' climate to an event with very similar circulation conditions in a hypothetical, 'counterfactual' climate. However, physical climate models cannot directly transfer these storyline counterfactuals across different climate forcing states. Statistical and machine learning techniques may overcome this limitation; yet, emulating circulation-conditional extreme events under different climate states is challenging. Here, we demonstrate distributional autoencoders (DAEs) as a versatile method for generating climate counterfactuals. They model the full distribution of spatially resolved European temperature fields conditional on the atmospheric circulation state and the mean global warming level. These distributions allow for deriving meaningful conditional probability ratios, which is a particular advantage of the DAE-based storyline approach. We train DAEs on fully coupled climate model simulations and we evaluate the modelled distributions across different factual and storyline-based counterfactual climate model simulations. In an illustrative case study, we revisit the 2003 European heatwave and we generate counterfactuals for a hypothetical `2003-like European heatwave' using ERA5 circulation, which we hypothesize to occur a quarter century (2028) and a half century (2053) after 2003. The conditional intensity would increase from 29.3 °C in 2003, to 30.3 °C and 32.1 °C in 2028 and 2053, respectively and conditional probability ratios would be 2.1 and 3.2 when compared to 2003.
Versions
Related
- The October 2024 extreme precipitation event over Valencia: storyline attribution of the synoptic-scale thermodynamic drivers
- Advancing Near-Term Water Quality Forecasting with Explainable Machine Learning and Probabilistic Uncertainty Quantification
- Long-term ozone formation sensitivity in China: spatiotemporal evolution and machine learning attribution
- Regional climate risk assessment from climate models using probabilistic machine learning
- Machine-learning-based approach for solar radiation model uncertainty identification, attribution and bias correction
BibTeX
@misc{loer2026probabilistic,
title = {{Probabilistic storyline attribution using machine learning}},
author = {Frieder Loer and Maybritt Schillinger and Sebastian Sippel},
howpublished = {arXiv},
year = {2026},
month = jun,
doi = {10.48550/arxiv.2606.02550},
eprint = {2606.02550},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2606.02550},
note = {CC BY-SA 4.0},
}
RIS
TY - UNPB
TI - Probabilistic storyline attribution using machine learning
AU - Frieder Loer
AU - Maybritt Schillinger
AU - Sebastian Sippel
PY - 2026
DA - 2026/06/01
PB - arXiv
DO - 10.48550/arxiv.2606.02550
UR - https://arxiv.org/abs/2606.02550
AB - A fundamental goal in climate attribution is to estimate how forced climate change contributes to observed extreme weather events. The storyline attribution method compares an observed weather event, conditional on its atmospheric dynamic state (i.e., atmospheric circulation), in the current, 'factual' climate to an event with very similar circulation conditions in a hypothetical, 'counterfactual' climate. However, physical climate models cannot directly transfer these storyline counterfactuals across different climate forcing states. Statistical and machine learning techniques may overcome this limitation; yet, emulating circulation-conditional extreme events under different climate states is challenging. Here, we demonstrate distributional autoencoders (DAEs) as a versatile method for generating climate counterfactuals. They model the full distribution of spatially resolved European temperature fields conditional on the atmospheric circulation state and the mean global warming level. These distributions allow for deriving meaningful conditional probability ratios, which is a particular advantage of the DAE-based storyline approach. We train DAEs on fully coupled climate model simulations and we evaluate the modelled distributions across different factual and storyline-based counterfactual climate model simulations. In an illustrative case study, we revisit the 2003 European heatwave and we generate counterfactuals for a hypothetical `2003-like European heatwave' using ERA5 circulation, which we hypothesize to occur a quarter century (2028) and a half century (2053) after 2003. The conditional intensity would increase from 29.3 °C in 2003, to 30.3 °C and 32.1 °C in 2028 and 2053, respectively and conditional probability ratios would be 2.1 and 3.2 when compared to 2003.
ER -