| 15 Oct 2025 | Subsequently updated
Understanding biases and changes in European heavy precipitation using dynamical flow precursors
Abstract. We address the problem of understanding precipitation in climate models. Using a novel decomposition applied to two large ensemble simulations, we disaggregate biases and forced changes in European heavy precipitation occurrence according to different weather conditions and isolate synoptic-scale dynamical contributions from the local-scale conversion of synoptic forcing into precipitation. We categorise weather conditions using multivariate, regionally-specific heavy precipitation precursors that target precipitation-causing flow patterns, revealing a larger role for dynamics in explaining model biases and projected changes than suggested by previous work. We demonstrate that biases in heavy precipitation across models and regions can emerge from errors on very different scales, with compensating biases between scales being common. This has important implications for model selection, for example for downscaling or storyline applications. In terms of forced changes in heavy precipitation, we show that apparent model agreement can arise from markedly different future scenarios with different levels of implied risk.
Our results demonstrate the utility of flow-dependent diagnostics for exposing the origins of climate model biases, which can distort a model’s precipitation response in future projections. With an eye to informing researchers in model development and validation, we demonstrate which combinations of dynamical versus conversion biases lead to specific types of distortion, and emphasise that these cannot be corrected for without a flow-dependent perspective. This framework allows us to introduce an intuitive heuristic for guiding model selection and interpretation, and to extract usable climate information from imperfect models.
Competing interests: Camille Li is editor of the EGU journal Weather and Climate Dynamics.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Review of “Understanding biases and changes in European heavy precipitation using dynamical flow precursors” by Oldham-Dorrington et al.
This manuscript introduces a flow-dependent decomposition framework for analyzing heavy precipitation biases and forced changes in two major large-ensemble climate simulations (CESM2 LENS2 and MPI-GE). The paper classifies synoptic states using region-specific multivariate precursor patterns (Z500, U850, V850), enabling a novel partition of precipitation errors into dynamical (synoptic forcing occurrence) and conversion (local-scale processes converting forcing to precipitation). The authors apply this to 38 regions across Europe and all seasons.
Overall, this paper is impressively comprehensive, and the results reveal new insights into compensating biases, dynamical controls, and the physical mechanisms behind future changes in heavy precipitation frequency. The paper is clearly written, well structured, and methodologically rigorous. It will be of high interest to the climate dynamics, hydroclimate, and impacts communities. The identification of widespread compensating biases and distortions in forced changes is especially valuable for model evaluation, downscaling, and storyline applications.
I find the manuscript to be a strong and valuable contribution suitable for publication after minor revisions. My comments below aim to enhance clarity, interpretation, and broader applicability.
In short, this is a well-designed and insightful manuscript that advances our understanding of flow-dependent heavy-precipitation frequency biases and changes. With clarifications on terminology, broader discussion of intensity considerations, and guidance on ensemble-size requirements, the paper will be even more impactful and accessible to a wide interdisciplinary audience.