The Future of Extreme European Winter Weather

Every time there is an extreme weather event, there is much speculation about the role played by climate change. Questions that are particularly relevant for Europeans include how is climate change affecting the occurrence of flooding brought about by severe storminess, like that seen in the 2013/14 winter, or storms Emma and Friederike in 2017/18? How is it affecting cold air outbreaks like the "Beast from the East" in 2018? Until now, our capacity for studying the impact of climate change on weather extremes has been limited. This is because we could not run atmospheric models that can successfully simulate weather events like these for long enough to be able to study climate change's effect - this requires thousands of years of simulation time. But I have recently led development of a system that can achieve this, using distributed computing and an atmospheric model with a fine enough grid spacing to be able to simulate extreme storms and anticyclones realistically. I will exploit this system to show how climate change is affecting extreme weather systems, with a particular focus on winter storms in Europe. This will done by seeing how the model simulations respond to specified changes in greenhouse gas levels and other atmospheric constituents and changes in the oceans and sea ice associated with climate change. This allows an exquisite level of control over the climate change scenario being simulated, and makes it possible to simulate a wide range of futures.

However, it is not sufficient to run a single climate model and just report its best estimate of climate change's effect. It also needs to be shown what is the range of possible outcomes. This is necessary to ensure that decision-makers can prepare for the worst possible outcomes whilst not wasting resources on adapting to scenarios that are very unlikely. It is also very important to understand the physical mechanisms behind climate change's impact, so that we can make better judgements about how much to trust the model simulations.

I will investigate the main sources of uncertainty in how climate change will affect weather extremes in a given future scenario:
1. Uncertainty about how to best model the atmosphere.
2. Uncertainty about how atmospheric dynamics will change, for example the shift in the mean latitude of the jet stream, or changes in the frequency of blocking events (when stable, high-pressure systems form and divert storms to the north and south).
3. Uncertainty in the future factors that will influence the atmosphere, namely ocean temperatures, ice cover and the atmospheric chemical composition (including man-made greenhouse gases and aerosols).

To understand the mechanisms behind how climate change is affecting Europe's weather, I will first separate the changes in man-made greenhouse gas and aerosol levels in the atmosphere from the oceanic changes, then further divide the latter into changes in different regions (such as the North Atlantic, the Arctic, the tropics etc.). Experiments will be done to see the effect of each change separately. This will help to answer questions such as how are North Atlantic storms affected by warming of the ocean waters there, which may invigorate storms by evaporating more water, providing more fuel for their growth? Are changes more due to remote influences of the tropics and Arctic? There has been much speculation in the media about the melting of sea ice in the Arctic being a driver of European extreme weather, for instance, but is still very scientifically controversial, and its impact on extreme weather events has not been studied in this way before.

Overall, this research will give us much greater confidence and understanding about how climate change will be felt through extreme European winter weather, informing governments and industries about the difference that reducing greenhouse gas emissions will make, and helping decision-makers to plan for the future.

Groups of people that will benefit from this research include:

Climate change adaptation planners: key results of this project will include showing how climate change will affect the statistics of simulated winter extremes (such as severe storms, heavy rainfall and cold waves brought about by atmospheric blocking events) based on projections for the 2025-2034 and 2050-2059 periods. This will build on the work of projects like the UK Climate Projections 2018 and European Climate Prediction System projects. My results will show more clearly the range of possible changes in future extreme weather due to climate change, with a particular potential application being increasing the skill of projections of future flood risk through my partnership with Prof C. Huntingford at the Centre for Ecology and Hydrology (CEH). This is relevant for planning future resilience to flooding in cities.

Policy makers: another key result will be to show how extreme events are likely to change if the 1.5C or 2C global warming targets specified in the Paris Agreement are met. This will show how beneficial it is to meet these mitigation targets, informing national greenhouse gas emissions reductions policies and international negotiations. The results of this work will be highly relevant to the UK government's 5-yearly Climate Change Risk assessments.

Climate prediction centres (e.g. the Met Office): the proposed work will show what are the key sources of uncertainty in projections of how extreme weather will change. Methods to understand and investigate these sources of uncertainty will also be presented. This will help climate prediction centres to produce more reliable projections for the future, that forecast users can have more confidence in.

Water managers: the work will show how extremes in precipitation will vary in future climate change scenarios. The model data will be suitable for use by hydrological modellers to calculate flood risk, as has previously been done using data from a lower-quality climate model [Schaller et al., 2016, Nat. Clim. Change, 6, 627-634]. My collaboration with Prof C. Huntingford at CEH will ensure that this knowledge transfer occurs.

The insurance industry: insurers and reinsurers must make sure they can withstand extreme weather events. The large dataset of climate model simulations that will be produced in this work will be very valuable for providing information about how climate change will affect these risks, which cannot be estimated accurately with observations or small model datasets. The results can be applied to modify the inputs to the sector's "catastrophe models" that simulate the risks of damage to life and property given the risks of the occurrence of severe weather. My collaboration with Dr C. Rye at reinsurer SCOR will ensure that there is knowledge transfer to this sector.

The public: there is intense demand for information about how climate change affects extreme weather, as evidenced by the regular articles on the subject in the media. It is important that the information presented to the public is based on all of the possible sources of data and that uncertainty in the results is appropriately quantified, so the results are not likely to appear to change dramatically as new scientific work is done. This project will uniquely add information about how climate change affects extreme weather according to a physical atmospheric model that can simulate the relevant weather phenomena well. It will also show clearly the range of uncertainty on these estimates, helping to make them robust. The results and methods developed in the project will also be relevant for the attribution of observed extreme events to climate change, which is another highly impactful topic. Operational services for event attribution are being piloted by climate change predictions centres such as the Met Office, so this will also benefit those organisations.