Exoplanets and stellar physics

Note that other projects targeting exoplanet science are offered by the Atmospheric, Oceanic and Planetary Physics sub-department, and listed on their website.

Supervisor: Suzanne Aigrain

The PLATO space mission, due for launch in March 2027, is the European Space Agency’s next M-class mission. It will search for planetary transits, including those caused by planets with similar sizes and orbits to the Earth, and simultaneously characterise the host stars using asteroseismology. Measuring the masses of the planets using follow-up radial velocity observations from the ground is also a core part of the mission. Starting 6 months into the mission, just as the first science data are beginning to flow, the PhD student get involved in the verification and validation activities for the Exoplanet Analysis System, the pipeline which searches for transits in the light curves and models them. They will also contribute to early science projects designed to demonstrate the capabilities of PLATO before its data and pipelines have been fully validated (for example characterisation of previously know planetary systems that PLATO will observe). As the mission matures, they will have the opportunity to get involved in the discovery and characterisation of new planetary systems. 

Supervisors: Matthias Tecza, Ryan Griffiths

The upcoming Planetary Camera and Spectrograph instrument (PCS) aims to exploit the high angular resolution delivered by the 40 m Extremely Large Telescope to detect and characterise rocky, potentially habitable exoplanets around nearby M and K-dwarf stars [1]. To achieve this, PCS must be sensitive to companions with planet-to-star flux ratios of 10⁻⁸–10⁻⁹ – 10³ times fainter than the limits of current instruments. Residual stellar speckles, which are formed by quasi-static imperfections in the imaging system, are the primary limitation for reaching this sensitivity as they obscure light from faint companions and cause false detections [2]. These stellar speckles can, however, be suppressed through manipulation of the phase and amplitude of the light propagating through the instrument - typically using one or two deformable mirrors. 

Through deducing the relationship between speckles formed in science images and the applied deformable mirror shapes, we can remove speckles without modifications to the instrument design. Deep reinforcement learning (DRL) offers a promising approach to learning this complex, non-linear relationship and recent studies have demonstrated improvements in image contrast in simulations of broadband imaging [3,4].

This project will explore applying deep reinforcement learning to speckle suppression for spectroscopic imaging. This will involve designing suitable reward functions and policy networks that incorporate spectral information and calibration data. There is scope to explore the effectiveness of different algorithms (e.g. on-/off-policy and model-based/model-free) for improving contrast. The project will also investigate how these methods can be integrated into observing routines and how effectively they track and suppress slowly evolving speckle patterns.

While these studies can be carried out in simulation, this project aims ultimately to validate a DRL-based speckle suppression technique on a high-contrast imaging experiment located in the lab at Oxford [5]. The test bench contains a full adaptive optics and coronagraph simulator coupled to an image-slicer integral field spectrograph. This experimental validation will serve as a vital step before moving to on-sky testing with existing high-contrast imaging instruments. We aim to apply the resulting DRL control method on existing on-sky instruments to make new detections of giant exoplanets by improving their contrast at small angular separations.

Reading List

1.  https://doi.eso.org/10.18727/0722-6691/5221
2. https://doi.org/10.1086/520913
3. https://doi.org/10.1051/0004-6361/202558504 
4. https://arxiv.org/pdf/2407.18733 
5. https://doi.org/10.1117/12.3104913

Supervisors: David DeBoer, Kris Zarbadami, Neil Bowles

Note: students interested in this project should apply to the Atmospheric Oceanic and Planetary Physics programme (not the Astrophysics one). Their website, which also lists other projects in planetary physics can be found here.

We uniquely rely on radio telescopes to understand the bulk of baryonic matter in the early Universe and the evolution of the Universe across early Cosmic times.  There are two unavoidable limitations to our exploration of the Cosmos with terrestrial-based radio telescopes:

1. the prevalence of interfering radio transmissions from human activity, and
2. the Earth’s ionosphere and atmosphere, which are opaque or disruptive to many important radio frequency ranges.

These limitations are overcome by reaching above the atmosphere and beyond Earth, a capability that, to date, has generally been the exclusive purview of expensive, government-funded space-based “Great Observatories”.

However, we are now on the cusp of a new era, with cost-effective and opportunity-laden access to space (including the lunar surface) becoming a reality. Such unfettered access will enable disparate groups to send instruments there, bringing about a transition from Earth-based observatories to space-based and cislunar telescopes (Fig. 1). Although this shift provides exciting opportunities for new science, it will also introduce sources of radio frequency interference (RFI) to the quietest remaining location to which we have access: the lunar Farside. Measurements from a generally RFI-free environment must be taken before the lunar skies have a multitude of orbiters.

One mission in the planning stage is the Lunar Farside Technosignature and Transients Telescope (LFT3). Being led out of Oxford, this mission would seek to arrive on the lunar farside surface before 2030, to take unique-in-history measurements from <1 - 2700 MHz.. The science case is wide and spans technosignatures, transients, cosmology and lunar studies. This project would explore instrumentation for scientific outcomes for LFT3 and any potential “ride-along” instruments. This could include all mission aspects as well as foundational and preparatory science and instrumentation. The overall goal is science from the Moon, which also includes regolith studies and operations from the Moon. Specifically, this project would work with the LFT3 instrument design and strive to maximize the scientific impact of the data to be collected from the Moon at this unique time.