I extract reliable signals from noisy, incomplete data, and measure how observational limits bias what we think we know.
My research studies turbulence in astrophysical plasmas, from galaxy clusters to the solar wind. The methods (spectral analysis, Bayesian inference, stochastic modelling and simulation) transfer to any field where data are sparse, noisy or indirectly observed.
Galaxy cluster mass estimates underpin precision cosmology, but they depend on non-thermal pressure from gas motions driven by mergers, feedback and galaxy motions, usually modelled as turbulence.
I measure turbulent cascade rates, energies and characteristic scales from surface brightness fluctuations, and quantify how observational effects bias those measurements.
How magnetohydrodynamic turbulence evolves as the solar wind flows outward matters for space weather, cosmic rays and the evolution of heliospheric fields.
I test how modelling assumptions change the predicted evolution of energies and length scales, and constrain model parameters against spacecraft observations using Bayesian methods.
Detecting the inflationary B-mode signal would constrain the tensor properties of the Universe, but it is heavily obscured by Galactic foregrounds: synchrotron dominates at low frequencies and thermal dust at high frequencies.
I work on developing stochastic models to provide constrained realizations of the un-observed small-scale sky. This work necessarily sits at the very beginning of future CMB missions. Providing the tools necessary to evaluate how instrument architecture influences our ability to detect primordial B-modes.
Spectral density estimation with Fourier and real-space methods, spherical harmonics, and spatial and time-series analysis in noisy, data-limited regimes.
Parameter estimation and model selection using MCMC and nested sampling.
Magnetohydrodynamic turbulence, turbulence transport modelling, stochastic and fractal models, and numerical differential equation solvers.
Python (NumPy, SciPy, Matplotlib), Git and GitHub, LaTeX.
A Python package that simplifies common tools for astrophysical plasma turbulence analysis, and lets you test observational processing pipelines on synthetic data with known ground truth.
A Python package that provides stochastic models of full-sky polarized Galactic synchrotron emission with free-parameters to capture non-Gaussian statistical and phase coherent behaviour.
Contributor to a software-defined computer network controller for production networks.
Selected work. Full list on ORCID.