Research results

Leverhulme Trust, STFC

Our group uses theoretical models of planetary climates, primarily the UK Met Office 3D model called the Unified Model, to understand and explore the climates of planets discovered around distant stars, termed exoplanets. Developing a range of models, and using them to interpret observations from telescopes such as the Hubble Space Telescope, as well as make predictions for future instruments such as the James Webb Space Telescope.

Linking with the Met Office, and across disciplines within and beyond the University of Exeter, has helped us create an exciting inter-disciplinary group including atmospheric chemistry experts, astrophysicsist and Earth climate scientists. Our group is at the forefront of exoplanet research, worldwide, developing models covering a range of physical complexities, which aids us in isolating the key physical processes dominating the target exoplanet climate, as well as covering a range of objects from extremely hot Jovian planets, to potentially habitable Earth-like planets. The Figure below shows both the simulated output, and a visualisation of the climate of a hot Jupiter: HD189733b.

Access to High Performance Computing (HPC), such as ISCA and its predecessor zen are absolutely vital as this sort of research requires calculations which can not be performed on desktop computers. It is absolutely vital that access to such facilities remains, and importantly, access remains for research projects without direct funding. Without this, many of our results would not have been possible, which have subsequently allowed us to leverage funding. In point our group would not exist and all of our publications would not have appeared without access to, initially free, HPC via the University. In addition, smaller scale, server machines have allowed rapid development to debug and evolve our models, before they are used on large-scale HPC facilities. This element is also vital, and one such machine, funded by our College has been invaluable to my group’s research. 

Our work has resulted in the identification of a potential mechanism leading to the significant expansion of a class of exoplanets called hot Jupiters, who often appear up to two times larger than our previous theories would predict. We have also demonstrated the need for the treatment of chemical kinetics, that is tracking the chemistry within an atmosphere, to interpret observations of exoplanets alongside the inclusion of clouds with, crucially, a treatment of their scattering effect. For gas giant exoplanets we have clearly demonstrated the three -dimensional mixing of material within these atmospheres which dominates the current observations of species within them. For terrestrial planets we have both determined a potentially habitable state for our nearest exoplanet, Proxima Centauri B, and shown potential signatures of water clouds in future observations, alongside highlighting the importance of treating the surface effects.

We have also partnered with a visual effects company to create scientifically driven visualisations of several exoplanet environments.

Mayne et al., (2014): http://adsabs.harvard.edu/abs/2014A%26A...561A...1M

Amundsen et al., (2014): http://adsabs.harvard.edu/abs/2014A%26A...564A..59A

Mayne et al., (2014b): http://adsabs.harvard.edu/abs/2014GMD.....7.3059M

Helling et al., (2016): http://adsabs.harvard.edu/abs/2016MNRAS.460..855H

Drummond et al., (2016): http://adsabs.harvard.edu/abs/2016A%26A...594A..69D

Amundsen et al., (2016): http://adsabs.harvard.edu/abs/2016A%26A...595A..36A

Amundsen et al., (2017): http://adsabs.harvard.edu/abs/2017A%26A...598A..97A

Boutle et al., (2017): http://adsabs.harvard.edu/abs/2017A%26A...601A.120B

Tremblin et al., (2017): http://adsabs.harvard.edu/abs/2017ApJ...841...30T

Mayne et al., (2017): http://adsabs.harvard.edu/abs/2017A%26A...604A..79M

Lewis et al., (2018): http://adsabs.harvard.edu/abs/2018ApJ...854..171L

Drummond et al., (2018): http://adsabs.harvard.edu/abs/2018ApJ...855L..31D

Goyal et al., (2018): http://adsabs.harvard.edu/abs/2018MNRAS.474.5158G

Drummond et al., (2018b): http://adsabs.harvard.edu/abs/2018A%26A...612A.105D

Lines et al., (2018): http://adsabs.harvard.edu/abs/2018A%26A...615A..97L

Drummond et al., (2018c): http://adsabs.harvard.edu/abs/2018arXiv181009724D

Lines et al., (2018b): http://adsabs.harvard.edu/abs/2018MNRAS.481..194L

http://exoclimatology.com/

ERC Starting Grant “ImagePlanetFormDiscs” (2015-2020)

The aim of the project is to study the processes in the inner few astronomical units of protoplanetary discs and to search for evidence for ongoing planet formation. For this purpose, we have commissioned the new MIRC-X beam combination instrument at the CHARA telescope array in California. One objective is to derive the physical conditions near the dust sublimation region and to constrain the global structure of these discs using radiative transfer modelling. For this purpose, we use the TORUS code that has been developed by colleagues around Prof. Tim Harries.

The HPC facility at Exeter enables us to develop and use the TORUS Monte Carlo radiative transfer code to model the infrared emission from protoplanetary disks. The radiative transfer modelling is essential to fully probe the physical nature of these disks and the HPC suite enables us to much more rapidly explore the parameter space.

In Davies et al. (2018), we find that the dust sublimation rim is best described by a curved geometry and composed mainly of large, micrometre-sized dust grains, which is consistent with grain growth in the inner disc regions.

Davies et al. (2018), ApJ in press (http://adsabs.harvard.edu/abs/2018arXiv180810762D)

Labdon et al., in preparation

http://imageplanetformdiscs.skraus.eu/

The heart of the TOFU project was the development of multi-D time implicit fully compressible hydrodynamic code MUSIC (Multi-dimensional stellar implicit code). No tool like MUSIC currently exists worldwide. It is uniquely suited to study complex hydrodynamics processes in star and planet interiors. The heart of the COBOM project is to perform the most comprehensive study ever performed of mixing processes in stars, one of the major uncertainty in stellar evolution theory, using a fundamentally new approach based on our code MUSIC.

Thanks to its flexibility and rapid access, the local HPC facility at Exeter has played a fundamental role for the development of the MUSIC code and final production phase for the TOFU project. The availability of a local facility is essential for efficient numerical tool developments. The possibility of having a performant local facility, complementary (and more flexible/available) than national/international supercomputer facilities, has been one of the selling points for the COBOM ERC project which has been recently awarded. 

This project aims to understand and characterize the hydrodynamic effects of debris blockage through a combination of laboratory experiments in flumes and computational fluid dynamics (CFD) modelling. The purpose is to enable bridge engineers to better protect bridge structures from the scour and hydrodynamic forces caused by debris accumulation in front of bridge piers and abutments. The key output will be a risk-based approach for assessing the scour, and uplift and lateral forces at individual bridges due to debris blockage during flood conditions. The approach will be incorporated within existing guidance for the assessment of bridges under hydraulic action.

A key part of the research is the development of CFD models for scour, their validation using laboratory results and the subsequent usage of the models for simulating a number of real-life scenarios. The vast majority of these simulations have been run on the ISCA supercomputing facilities. The time and productivity gains from moving to ISCA from our in-house computational cluster has been immense and is a critical factor in the progress we have made to date.

https://ore.exeter.ac.uk/repository/handle/10871/32599

https://ore.exeter.ac.uk/repository/handle/10871/28314

https://ore.exeter.ac.uk/repository/handle/10871/24897

In this project we aim to identify the key physical parameters of doped semiconductor nanocomposites (e.g. system dimensionality, sample length, unit cell size, density and quality of interfaces between the constituents of a composite, and carrier doping level) which can significantly reduce their thermal transport capability below the alloy and amorphous limits and help enhance their thermoelectric figure of merit ZT.

I have a long-established high-performance computer cluster, named ceres. With the help of the Leverhulme Trust grant I have added four more nodes to this cluster. All my research computing is carried out using this cluster. I have so far not used the University of Exeter HPC facilities for my research.

Thomas and Srivastava, J Phys CM 29 (2017) 505703;

J Appl Phys 123 (2018) 135703;

Phys Rev B 98 (2018) 094201;

Srivastava and Thomas, Phys Rev B 98 (2018) 035430.

ERC Consolidator Grant

Accretion, Winds, and Evolution of Spins and Magnetism of Stars

The AWESoMeStars project studies Sun-like and low-mass stars. The overarching goal is to develop a comprehensive, physical description of the origin and evolution of stellar rotation, magnetic activity, mass loss, and accretion.

We use university facilities for our MHD simulations of stellar winds (and soon of star-disk interactions and stellar evolution calculations). It has been crucial to the project to have the easy access and quick response facilities for running our numerical simulations. Also, having the capability to purchase time on a pay-as-you-go basis has allowed critical flexibility and eliminated wasted resources, for example, while the project was in the ramp-up phase.

So far, our main discoveries (that have been published to date) are in studies of stellar winds.
Generally speaking, we have now determined what are the most important properties of a star
that are predictors of angular momentum loss. Highlights include: We have used simulations to
determine how the acceleration of a stellar wind affects the angular momentum loss, finding that
our uncertainty in how winds accelerate will not contribute the largest uncertainty in being able
to predict the angular momentum loss. (Pantolmos & Matt 2017, and Pantolmos, Matt, & Zanni
2019 in prep.). We have used simulations to determine how combined magnetic geometries
affect stellar wind angular momentum loss and determined a formulation that can predict the
angular momentum loss for any magnetic geometry, without the need for costly numerical
simulations (Finley & Matt 2017; 2018a). We have used that formulation to study the variability in
angular momentum loss in the solar wind, which we find to vary by a factor of several over the
solar magnetic cycle (Finley, Matt, & See 2018b).

http://empslocal.ex.ac.uk/AWESoMeStars/articles/publications/

http://empslocal.ex.ac.uk/AWESoMeStars/