Dynamics of atmospheric multi-scale systems : a low-order model perspective

Hamilton, Oisin
(2025)

Files

250903_Hamilton_Oisin.pdf
  • Open Access
  • Adobe PDF
  • 65.14 MB

Details

Authors
  • Hamilton, OisinUCLouvain
    author
Supervisors
Vannisetm, Stéphane
;
Crucifix, Michel
Abstract
The weather in the midlatitudes displays low frequency variability (LFV), or long term repetitions, over time spans much longer than the usual two week forecasting horizon. This LFV originates from couplings, or interactions, between the atmosphere and other components of the climate system, as well as internal variability within the atmosphere. The overall aim of this thesis is to study LFV to increase our understanding of how such variability originates, and how it could be impacted by the changing climate. This thesis goes about answering these questions using reduced order models, which aim to distil the core dynamics of the coupled atmosphere behaviour. The following specific areas are addressed. Atmospheric blocking is studied from a regime perspective, where persistent weather patterns are linked with specific structures in state space. We used unstable periodic orbits (UPOs) to understand the observed clustering behaviour of the model. We found that the onsets and decay of blockings can be linked with transitions between clusters of orbits. In this study we also extended a method of approximating trajectories with orbits, known as shadowing, by looking at the cumulative number of times a UPO shadows a model trajectory over a fixed time period. This concept was used to identify different life cycles of each cluster in state space by creating corresponding sets of UPOs. The above analysis was used to also investigate how climate change could impact LFV in the midlatitudes. This was done using the same model, but with non-constant parameter values across the domain, as well as altering the model resolution. The results align with the literature, to show that climate change has opposing responses in the atmosphere, leading to uncertainties in how the atmosphere could respond to climate change. We also confirm that reduced order spectral models can present different dynamics, depending on the resolution, or number of wave modes included. To overcome this limitation this study developed and tested models of intermediate resolution. Climate regimes have also been explored in the context of a reduced order ocean-atmosphere system. By incorporating the full nonlinear longwave radiation scheme, rather than the previous linearised version, it was found that multiple climate states, with distinct average global temperatures, coexist for the same parameter values. In addition to different flow structures and low frequency variability patterns in the atmosphere, this work showed that the distinct chaotic attractors present different levels of predictability and dimensionality. Finally, as classical reduced-order models only focus on the large scales of the flow, we investigated the impact, on the energy transfers, of including a large number of spatial scales. A simplified model to study turbulent energy transfers (a shell model) was developed to study how energy is transferred between scales in the atmosphere. The aim of this is to investigate the scales and wave triad interactions that are required to model a two layer quasigeostrophic atmosphere. This work demonstrated that extremely simplified models can capture the same turbulent cascade as predicted by theory. This provides a framework to study the accuracy of energy transfers in the spectral two layer quasigeostrophic model, that we used throughout this thesis. In short, in this thesis, we built a number of reduced order models to investigate the dynamics of the atmosphere, revealing the richness and diversity of the dynamical behaviours present. The findings of this thesis could guide future studies, using more accurate climate models, on the emergence of LFV, and investigations into how different components of the climate system interact.
Affiliations

Citations

Hamilton, O. (2025). Dynamics of atmospheric multi-scale systems : a low-order model perspective. https://hdl.handle.net/2078.5/260959