Publications
Hong Kong Observatory internship (June–September 2026)
I developed a model-agnostic machine-learning post-processor for tropical cyclone intensity and central pressure forecasts from global AI weather models. A first-author manuscript has been submitted to Tropical Cyclone Research and Review, and I will link it here once the preprint is public.
Computational statistical physics (2026): the 3D Heisenberg model
My final project for Computational Statistical Physics at ETH Zürich: a Monte Carlo study of the three-dimensional classical Heisenberg model, measuring the critical temperature and five critical exponents with Metropolis and Wolff cluster simulations on lattices up to 256³ spins. All six results are within 1σ of the 3D O(3) literature values.
Semester project (2026): effects of spatial and temporal remapping on tropical cyclone potential intensity
A semester project in the Atmospheric Physics group at ETH Zürich, supervised by Prof. Ulrike Lohmann and Dr Bernhard Enz. Using the ERA5 ensemble reanalysis over the western tropical Atlantic, I looked at how the potential intensity of tropical cyclones depends on the spatial and temporal resolution of its inputs.
Summer research project (2025): synchronisation in 2D turbulence
I studied the synchronisation of small scales in two-dimensional turbulence driven by Kolmogorov forcing, using JAX-CFD, and checked my code against results in the literature, and began extending the analysis to include Coriolis effects on the beta-plane. The project was supervised by Dr Jacob Page, with assistance from Dr Dmitriy Zhigunov.
JAX-CFD and Dedalus on Kolmogorov flow
Summer research project (2024): the three-body problem
An eight-week project in the School of Physics and Astronomy, University of Edinburgh, supervised by Prof. Jenni Smillie.
The project had two aims: to reproduce and visualise important orbits of the three-body problem, and to understand two landmark theoretical results, Chenciner and Montgomery’s proof that the figure-eight orbit exists and the convergence of Sundman’s series. I wrote symplectic integrators up to fourth order, used them to study the Lagrange, figure-eight and Broucke–Hadjidemetriou–Hénon orbits and Euler’s problem of two fixed centres, and ended with a search for new well-behaved orbits.
Write-up, final report (PDF) and visualisation with Allison Lau. The report’s values of the parameter e for the two-centre orbits (0.986, 1.08 and 2.56) are wrong, and its orbit search (section 7) is rerun and reinterpreted in the log below.
Research log
I kept a public log while the project ran. It starts from first simulations and works up to the existence proof of the figure-eight orbit:
- Plotting and Visualising the Three-Body Problem 24 Jun 2024
- Symplectic Integrators 25 Jun 2024
- Ellipses and the Two-Body Problem 27 Jun 2024
- Problem of Two Fixed Centres (Part I) 28 Jun 2024
- Understanding and Visualising the Shape Sphere 4 Jul 2024
- Problem of Two Fixed Centres (Part II) 9 Jul 2024
- Sundman's Power Series Solution to the Three-Body Problem (Part I) 12 Jul 2024
- Sundman's Power Series Solution to the Three-Body Problem (Part II) 16 Jul 2024
- Proving the Figure-Eight Orbit Exists: Action Minimisation on the Shape Sphere 17 Jul 2024
- Classifying and Finding New Possible Orbits 22 Jul 2024
- Problem of Two Fixed Centres (Part III) 26 Jul 2024
Background: Overview of the two-body problem.
Research problem: liquid slosh with pendulums and waves
This began as a group research project at the Edinburgh Scientific Researchers Association (ESRA), with Patrick Truong, Gregory Barbanel and Abdullah Fouzuddeen. Starting from a published single-pendulum model of liquid slosh in packaging machines, we looked at two ways of capturing the shape of the liquid surface rather than just its tilt: several coupled pendulums, and a wave equation for the surface.