Ok, so this one is more work than a personal project, but I wanted to give an overview of my work, in a more informal way. The below is a summary of my PhD, that I completed at the Royal Meteological Institute of Belgium, and the Université catholique de Louvain.

The title of my PhD is: Dynamics of Atmospheric Multi-Scale Systems: A Low-Order Model Perspective.

Background and Context

The day to day weather in the midlatitudes, the area bounded between the Tropic of Cancer/Capricorn and the Arctic/Antarctic circles, is governed by large eddies, called synoptic scale systems [Holton, 2004]. These eddies consist of relatively high or low pressure systems that typically bring settled and unsettled weather, respectively. They are usually over three thousand kilometres in diameter, and they tend to move from west to east due to the jet stream, which is a fast moving band of air located at an altitude of approximately 10km [Stendel 2021]. In North Western Europe the flow of the jet brings, on average, warm moist air from the Atlantic, leading to a relatively mild and moist climate. The below image shows the location of the midlatitudes.

The eddies in the midlatitudes are generated by instabilities that form in the jet stream. This is called baroclinic instability[Barry, 2003], and is caused by the difference in the amount of solar heating that the equatorial region receives, compared with the polar regions. In other words, a temperature gradient exists between the equator and the poles, and this drives the eddies. This temperature gradient is larger in winter and weaker in summer, due to where the sun is hitting the planet. This leads to an increase in baroclinic instability in the winter, which leads to more powerful eddies [Charney, 1947]. Examples of this are Storm Éowyn that hit Ireland and Britain in early 2025, leading to wind speed records being broken, as well as a strong high pressure systems that brought sustained cold temperatures over the winter of 2017-2018 (Storm Emma, a.k.a ``the beast from the east”)[Moore, 2018; Met Office, 2025].

Eddies can remain over one location for extended periods of time. An example of this is called atmospheric blocking, where high pressure systems remain relatively stationary over one geographical location, and the usual westerly (zonal) flow is diverted [Tibaldi, 2018]. This leads to heat waves in summer, or cold snaps in winter. An example of atmospheric blocking, that occurred during April 2025, is shown in the two synoptic charts below. The variation in possible weather events, that are seen in the midlatitudes, is referred to as Low Frequency Variability (LFV). This covers a large category of possible weather events that can occur over time spans of weeks to months, longer than the life cycle of an individual eddy [Holton, 2004]. This can include blocking events that disrupt the regular movements of eddies which last a week or so, all the way up to interannual variability that can oscillate over the course of years, such as the El Niño-Southern Oscillation [Ghil, 2020].

Climate change is altering the global radiation balance, resulting in the atmosphere capturing increased amount of solar radiation. As stated earlier, the weather events of the midlatitudes are driven by a temperature gradients, so a change to the global distribution of heating will impact the weather and possible variability in the midlatitudes [Vallis 2015]. Currently, heat waves in Europe are responsible for tens of thousands of deaths per year [Bhatnagar, 2024]. As climate change increases the average temperature of the planet, extreme heat events are expected to increase the number of mortalities in the future [Ballester, 2023]. In Western Europe it has been found that climate models are underestimating the temperature and frequencies of extreme temperature events due to changes in the atmospheric circulation [Vautard 2023]. The concern is that the current state of the art models do not accurately capture all of the circulation responses to external forcings, or that they systematically underestimate the strengths of possible eddies. In addition, there is a growing concern that tipping elements in the climate system could alter the climate abruptly and irreversibly [Armstrong Mckay, 2022]. These tipping elements could have a major impact on atmospheric circulation. For example, the jet stream over Western Europe could be impacted greatly by the weakening of the Atlantic Meridional Ocean Circulation [Van Westen, 2024].

Our current understanding of how LFV in the atmosphere forms is still not complete. Numerical Weather Prediction (NWP), or weather forecasting, struggles with accurately forecasting certain aspects of this variability. For example, when atmospheric blocking events will occur or decay [Davini, 2020]. Conversely, this inherent variability offers an opportunity to extend the usual limit of weather forecasting (approximately two weeks), by understanding the mechanisms behind the repeating patterns [Hannachi, 2017].

Thesis outline

In the thesis I analysed the behaviour of an simplified model of the midlatitude atmoshpere. This was done by using a 2-layer models of different resolutions and couplings, to study low frequency variability.

Chapter 1 Low order models of the atmosphere display low frequency variability, meaning that there can be repeating patterns in the behaviour of the atmosphere over long time spans. Such behaviour could be useful for providing longer range forecasts than traditional weather forecasts are capable of. Unstable periodic orbits have been used before to categorise parts of the state space, and provide some level of predictability on the future path of trajectories. Here we wanted to identify whether these orbits can help us understand transitions between parts of the state space. This was studied in the context of variability in the ground temperature of a coupled model, and how this impacts blocking locations in the model. We found that the orbits can identify regimes in the attractor, and that they can provide an indicator of transitions between these regimes.

Chapter 2 There is still great uncertainty in the impact of climate change on the circulation in the midlatitudes. This is exacerbated by multiple changing processes in the atmosphere producing the opposite responses. This can lead to a small over all change in complex models, and thus overshadow the underlying individual responses. It is crucial to have a solid understanding of each of the potential processes that could be impacted by climate change so we can ensure global climate models are capturing each of these processes, and hopefully as a result the combined response, accurately. To study complex behaviour, it is common to employ a hierarchy of models to investigate the key underlying behaviours from a particular change in forcings. For this reason in this chapter we employ a reduced order atmospheric model to investigate the impacts at the `bottom rung’ of the hierarchy.

Chapter 3 To model the weather and climate systems it is necessary to make certain assumptions to reduce the complexity of the system to allow us to model these in a feasible amount of time, with limited resources. There are almost always tradeoffs between accuracy and efficiency, however, when making these approximations. The resulting climate states in models are largely impacted by the radiation scheme used for transferring energy in the model, for example.

Usually, to reduce the complexity of the numerical integration, the nonlinear term controlling outgoing longwave readiation has been linearised. However, linearising nonlinear processes can also remove the possibility for multistable solutions. We found that by removing this linearisation the model produced multiple stable climatologies.

Chapter 4 Turbulent energy exchanges in the atmosphere are important to model to understand how energy is transferred between scales as this can have a large impact on the resulting weather. At large scales, the atmosphere acts similar to that of a two dimensional flow. As a result, it also displays some behaviours that are found in two dimensional turbulence. This is seen primarily in an inverse energy flow, or cascades. However, observational data does not always match well with the theory of geostrophic turbulence.

The behaviour of geostrophic turbulence is difficult to model due to the scales involved. To help understand these issues mentioned we built and investigated the properties of a simplified model of geostrophic turbulence. This model was based on the two layer quasigeostrophic model to provide a link with the previous chapters in this thesis, with the hope that such models could provide a useful tool for modelling energy transfers in future. We found that the shell model can capture the same energy flows and energy spectra that theory of two layer quasigeostrophic models predicts.

My PhD project was part of a wider project called CriticalEarth, where myself and 14 other PhD students were studying abrupt transitions and tipping points in the climate system.

See more about the wider project at https://www.criticalearth.eu

References

Holton, James R. 2004. An Introduction to Dynamic Meteorology. 4th ed. Vol. 88. International Geophysics Series. Elsevier Academic Press.

Stendel, Martin, Jennifer Francis, Rachel White, Paul D. Williams, and Tim Woollings. 2021. “The Jet Stream and Climate Change.” In Climate Change. Elsevier. https://doi.org/10.1016/B978-0-12-821575-3.00015-3.

Barry, Roger G., and Richard J Chorley. 2003. Atmosphere, Weather and Climate. 8th ed. Routledge. https://doi.org/10.4324/9780203871027.

Charney, J. G. 1947. “The Dynamics of Long Waves in a Baroclinic Westerly Current.” Journal of Meteorology 4 (5): 136–62. https://doi.org/10.1175/1520-0469(1947)004<0136:TDOLWI>2.0.CO;2.

Tibaldi, Stefano, and Franco Molteni. 2018. Atmospheric Blocking in Observation and Models. Vol. 1. Oxford University Press. https://doi.org/10.1093/acrefore/9780190228620.013.611.

Bhatnagar, Bhanu. 2024. “Statement – Heat Claims More than 175 000 Lives Annually in the WHO European Region, with Numbers Set to Soar.” https://www.who.int/europe/news/item/01-08-2024-statement--heat-claims-more-than-175-000-lives-annually-in-the-who-european-region--with-numbers-set-to-soar.

Ballester, Joan, Marcos Quijal-Zamorano, Raúl Fernando Méndez Turrubiates, et al. 2023. “Heat-Related Mortality in Europe during the Summer of 2022.” Nature Medicine 29 (7): 1857–66. https://doi.org/10.1038/s41591-023-02419-z.

Vallis, Geoffrey K., Pablo Zurita‐Gotor, Cameron Cairns, and Joseph Kidston. 2015. “Response of the Large‐scale Structure of the Atmosphere to Global Warming.” Quarterly Journal of the Royal Meteorological Society 141 (690): 1479–501. https://doi.org/10.1002/qj.2456.

Vautard, Robert, Julien Cattiaux, Tamara Happé, et al. 2023. “Heat Extremes in Western Europe Increasing Faster than Simulated Due to Atmospheric Circulation Trends.” Nature Communications 14 (1): 6803. https://doi.org/10.1038/s41467-023-42143-3.

Armstrong McKay, David I., Arie Staal, Jesse F. Abrams, et al. 2022. “Exceeding 1.5°C Global Warming Could Trigger Multiple Climate Tipping Points.” Science 377 (6611): eabn7950. https://doi.org/10.1126/science.abn7950.

Van Westen, René M., Michael Kliphuis, and Henk A. Dijkstra. 2024. “Physics-Based Early Warning Signal Shows That AMOC Is on Tipping Course.” Science Advances 10 (6): eadk1189. https://doi.org/10.1126/sciadv.adk1189.

Ghil, Michael, and Valerio Lucarini. 2020. “The Physics of Climate Variability and Climate Change.” Reviews of Modern Physics 92 (3): 035002. https://doi.org/10.1103/RevModPhys.92.035002.

Davini, Paolo, and Fabio D’Andrea. 2020. “From CMIP3 to CMIP6: Northern Hemisphere Atmospheric Blocking Simulation in Present and Future Climate.” Journal of Climate 33 (23): 10021–38. https://doi.org/10.1175/JCLI-D-19-0862.1.

Hannachi, Abdel., David M. Straus, Christian L. E. Franzke, Susanna Corti, and Tim Woollings. 2017. “Low-Frequency Nonlinearity and Regime Behavior in the Northern Hemisphere Extratropical Atmosphere.” Reviews of Geophysics 55 (1): 199–234. https://doi.org/10.1002/2015RG000509.