Can air-sea coupling solve the signal-to-noise paradox in climate predictions?

Lead Research Organisation: University of Leeds
Department Name: School of Earth and Environment


This project tackles a major conundrum in climate science: the so-called "signal-to-noise paradox". It has recently become apparent that many state-of-the-art climate models underestimate the response of the atmospheric circulation to a broad range of external drivers1. This creates the perplexing situation where model predictions of the real world are more skilful than when the model is asked to predict itself (so-called perfect model test). This issue is particularly acute in the North Atlantic and has profound implications for climate science, since the models used to make projections of future climate contain biases that affect how they can be interpreted and used for climate services. Despite the wide-ranging implications of the signal-to-noise paradox, its origins remain elusive and there are currently no solutions for how to fix it. This project will be the first observation-based study of one potential source of the signal-to-noise error related to air-sea interactions over the storm tracks. The student will use EO data, machine learning techniques and state-of-the-art climate simulations to shed new light on the causes of the signal-to-noise paradox.
The storm track is a key feature of North Atlantic weather and climate and recent evidence suggests it may be central to explaining the signal-to-noise paradox2,3. Developing extratropical cyclones (ETCs) play a major role in transporting heat and momentum in the atmosphere. Fluxes of moisture and heat from the ocean provide an important source of energy to developing ETCs.
This project will address the hypothesis that the origin of the signal-to-noise error lies in the strength of air-sea coupling being underestimated in models, leading to errors in ETC development that ultimately lead to a too weak atmospheric circulation response. If this hypothesis is supported by the results, we will aim to develop an observational emergent constraint to reduce uncertainty in future projections of North Atlantic climate. The project outcomes are expected to lead to a step change in understanding the origins and potential solutions to the signal-to-noise paradox.
Aims and objectives The aim is to test the hypothesis that the signal-to-noise paradox is a consequence of simulated errors in North Atlantic air-sea interactions. To achieve this aim, the following objectives are set:
- Use EO data to identify ETCs and characterise the air-sea fluxes at different stages of ETC lifecycle (genesis, intensification, lysis), focusing on frontal development in the Gulf Stream.
- Compare the observed air-sea energy exchanges within ETCs to those simulated in stateof-the-art high resolution climate models
- Establish whether there is a relationship between biases in air-sea coupling and the magnitude of the signal-to-noise error in models and use EO observed quantities to constrain modelled spread (emergent constraint).


10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
NE/T00939X/1 30/09/2020 29/09/2027
2885250 Studentship NE/T00939X/1 30/09/2023 29/06/2027 Yvonne Anderson