Star formation and Stellar Evolution


Galactic Plane ecology: HII regions and their exciting stars

Janet Drew, Tim Gledhill, Mercedes Ramos-Lerate (Madrid, ESAC)

Astronomy has in common with biology the need to understand how environments work as interactive systems (the business of ecology). Here, the type of environment in question is the star-forming region in which massive stars have formed -- an HII region. It is widely accepted HII regions are important environments, significantly shaping the evolution of galaxies (via output radiation, supernova explosions and powerful winds). This project is timed to overlap the following 3 opportunities: the  2018 release of Gaia astrometry, providing much improved constraints on Galactic stellar distances and space motions; the completion of the IPHAS survey (led by Drew, see www.iphas.org); the commissioning of the WEAVE massive-multiplex spectrograph on the William Herschel Telescope from early 2019. The PhD project would combine two strands: (i) data-mining of IPHAS H-alpha imagery of the northern Galactic Plane, aimed at building intelligent selections of HII-region pointings for the WEAVE survey, (ii)  analysis of OB stars across the same sky area, using Gaia and ground-based survey data (including IPHAS), to build a higher accuracy 3-D picture of where the OB stars are and their true luminosities. These investigations will come together via WEAVE follow-up spectroscopy tracing the kinematics of both HII-regions and the OB stars in the second half of the PhD programme.

The student will be exposed to methods in advanced data science as well as contributing at the frontier of spatially-resolved interstellar medium astrophysics. The project is a collaboration with a member of staff at the Gaia ground station and data centre in Madrid. The WEAVE survey is itself a large collaborative programme involving large numbers of European scientists.


Galactic and Stellar Archaeology in the Gaia Era

Chiaki Kobayashi, Sean Ryan

Elements heavier than helium are formed in stars, and produced from different astronomical sources (core-collapse supernovae, Type Ia supernovae, asymptotic giant branch, gamma-ray bursts, and neutron-star mergers). Our hydrodynamical simulation code self-consistently includes these enrichment sources as well as other relevant physics such as  star formation and feedback (chemodynamical simulations). The student will study the origin of elements by comparing our simulation results of the Local Group (our Milky Way Galaxy and dwarf satellite galaxies) to the observational data from galactic archaeology surveys such as with the GAIA satellite mission. The student can also explore more detailed physics by modifying the existing code if he/she wishes.


Eruptive Variability in Protostars

Phil Lucas, Jan Forbrich, Janet Drew

This project is based on data from the VISTA VVV survey, the first large scale infrared exploration of the Milky Way in the time domain (co-led by Dr Lucas).

Stars and planets are formed via the accretion of matter from a circumstellar disc, which in turn is fed by a surrounding envelope of gas and dust. This is an unsteady process and huge eruptions are sometimes observed when the accretion rate jumps by a factor of 1000, causing a dramatic increase in the star's luminosity. This is often accompanied by ejection of some matter from the system in a fast jet. It is thought that this may be common behaviour for normal stars during formation but this has been hard to prove because there were no large scale infrared variability surveys before VVV.

The VVV survey has recently discovered a large population of eruptive variables, most of them protostars that are hidden from view in visible light. These protostars have diverse outburst durations and spectra. At the same time, recent ALMA maps showing unexpected rings in nearby protostellar discs has shown that we have much to learn about their basic properties. The aim of this project is to further explore this population and try to understand it better. By taking an empirical approach we can guide future theoretical models of star assembly and planet formation. This will involve working with members of the international VVV collaboration and using telescopes in Chile for spectroscopy and adaptive optics imaging, as well as archival multi-waveband photometry. It is likely that we will also make many unexpected discoveries in this first exploration of the infrared variable sky. Mid-IR time series data on YSOs from the Spitzer YSOVAR project may also be considered to broaden the analysis. The understanding developed in this PhD will prepare the student for future work in the growing area of time domain astrophysics with LSST, GAIA and Pan-STARRS.


Exploring the Infrared Variable Sky

Phil Lucas, Jan Forbrich, Hugh Jones

This project is based mainly on data from the VISTA VVV survey, the first large scale infrared exploration of the Milky Way in the time domain (co-led by Dr Lucas).

The VVV survey observed about 500 million stars in a large part of the Milky Way over a 5 year period, detecting several million variable stars in the near infrared. It also measured the proper motions for all the stars (many of which are undetected by GAIA) and this helps us to determine their distances via comparison with a Galactic rotation model. The aim of this project is to develop a broad understanding of the infrared variable populations in the Milky Way. This will involve using a variety of time domain methods (and quite likely developing new ones) in order to empirically determine the main groups of sources that exist by considering light curve properties, colours, luminosities and Galactic locations. It is quite likely that new types of variable star will be discovered, given that the dataset has already yielded many unclassifiable variable stars. This project can be quite flexible given the range of different approaches and questions that can be explored. The student would also undertake observational follow-up of newly discovered variables using telescopes in Chile. We will be able to draw on expertise from across the VVV collaboration and from academics in other departments, such as Computer Science. We expect that the methods developed can also be applied to other datasets, e.g. the YSOVAR mid-infrared project (of which Dr Forbrich is a member).


Self-enrichment in massive star clusters with super-massive stars

Martin Krause, Jim Dale, Donna Rodgers-Lee, Chiaki Kobayashi

Almost all of the elements in the Universe have been synthesised in stars, ejected by stellar winds and then recycled via the interstellar medium, self-gravitating gas clouds into the next generations of stars and their planets. How this exactly works is a matter of ongoing research, which we investigate with observations and computer simulations. Specifically for massive star clusters, it is known that elements produced in massive stars during the first million years of the clusters’ life is dumped locally into low-mass stars, in which these metals may be found even more than 10 Gyr after formation of the cluster. The elemental pattern of the detected elements has led recently to the suggestion that hypothetical short-lived stars of more than 1,000 solar masses (so-called supermassive stars) might have contributed to the enrichment process.

In this project, the metals ejected in the winds of massive and super-massive stars and their mixing into star-forming gas will be simulated with the 3D hydrodynamics code PLUTO. We have an active, experienced simulation group with several members using PLUTO. The output of the simulations should then be used to assess the question which kind of stars could have contributed to the enrichment in massive star clusters.


Modelling stellar feedback in simulations of star formation

Jim Dale and Martin Krause

Star formation is a complex problem involving at minimum hydrodynamics and self-gravity. However, for much of their lives, the structure and dynamics of star-forming molecular clouds are dominated by stellar feedback, and star formation cannot be understood without also modelling feedback processes. In particular, feedback is probably responsible for regulating and terminating star formation in individual clouds.

Most of the theoretical work in star formation is now done via computer simulations, and a lot of work has been devoted to simulating feedback. The effects of accretion heating and jets from low-mass protostars on low-mass molecular clouds has been examined (e.g. Offner & Arce 2014), as has the influence of feedback from massive (proto)stars, such as photoionisation and stellar winds (e.g Dale, Ngoumou, Ercolano & Bonnell 2014), on high-mass clouds.

However, before a massive molecular cloud can form massive protostars, it must first form low-mass protostars, which are much more common. At the earliest stages of star formation, these objects are in fact the only sources of feedback on the cloud, and they control the environment into which the massive stars are eventually born. It follows, therefore, that the effects of massive stars cannot be properly understood without modelling the preceding influence of low-mass protostars.

The purpose of this project is to investigate, using a series of numerical experiments, what is the effect of allowing *both* low- and high-mass protostellar feedback to operate self-consistently in the same cloud. The simulations will make use of the Smoothed Particle Hydrodynamics code GANDALF (Hubber, Rosotti & Booth 2017).

GANDALF is already able to model photoionisation from massive stars. The first task for this project will be to implement new algorithms to model the accretion heating and jets from low-mass protostars. Once this has been done and validated, a series of simulations will be done in which all types of feedback operate self-consistently and in the correct time-sequence, so that their influence on one another and on the evolution of the clouds can be understood.


A comprehensive study of collisions between giant molecular clouds

Jim Dale and Martin Krause

Simulations of spiral galaxies (e.g. Dobbs, Pringle & Duarte-Cabral 2015) suggest that collisions between molecular clouds should be common events, and in fact most clouds experience at least one collision during their lives. Additionally, there is a substantial body of observations claiming that particular star clusters or star-forming regions in the Milky Way are the results of such collisions (e.g. NGC 3603, Fukui et al. 2014, or M20, Torii et al. 2017).

Galactic-scale simulations are unable to resolve star formation, and interest in modelling collisions at the molecular cloud scale has recently revived (e.g. Haworth et al. 2015). However, these papers typically examine only a very small corner of the possible parameter space of such encounters. A comprehensive suite of models is sorely needed.

This project involves running a large number of simulations of collisions between pairs of turbulent model clouds using the Smoothed Particle Hydrodynamics code GANDALF (Hubber, Rosotti & Booth 2017). The most important parameters that the simulation suite will explore are the relative velocity of the clouds, the impact parameter, the mass ratio and the initial virial state. In all cases, the ranges of these parameters are already known from observations and/or galactic-scale simulations. This will enable a detailed study of the influence of cloud-cloud encounters on the star formation process within the clouds.


The Orion Radio All-Stars

Jan Forbrich, Mark Thompson, Jim Dale

The Orion Nebula Cluster (ONC) is the most prominent nearby star formation region. Even very early stages of protostellar evolution produce intense X-ray and nonthermal radio emission, irradiating protoplanetary disks and young planets. The ONC is an ideal target to study these physical phenomena, and now is a great time to do this, following the advent of major new and recently upgraded radio astronomy observatories. These studies are in turn laying important groundwork for the upcoming Square Kilometre Array (SKA).

We are conducting a large international radio astronomical survey of the ONC, involving continuing observations with three of the best radio interferometers (VLA, VLBA, and ALMA), as well as simultaneous Chandra X-ray observations and infrared surveys. Our study has already produced a sevenfold increase in the known radio population over previous studies, and it has begun to reveal both complex YSO variability patterns and intricate radio detail in the proplyds - photoevaporating disks around young stars impacted by massive stars next door. It will now finally be possible to study both thermal and nonthermal radio emission in great detail. This project will provide unprecedented constraints on the magnetospheric high-energy activity of young stellar objects, mass accretion, new insights into the impact of massive stars on their environment, and new astrometric constraints on the dynamics of embedded YSOs in the ONC, crucially complementing Gaia with VLBI precision astrometry. We will use and contribute to novel imaging algorithms for interferometric wideband continuum data, which will become more and more common in radio astronomy.

Depending on your interests, you will focus on a few of these aspects, exploring the radio population in the ONC in a full multi-wavelength context. This PhD will give you in-depth experience of working with the most cutting edge radio interferometers in a multi-wavelength star formation context, ideally positioning you to take advantage of the SKA when it is commissioned. Extended work visits to the National Radio Observatory in Socorro, New Mexico, home of both the VLA and the VLBA, could be part of this project.


Investigating our Galaxy with runaway stars

Ralf Napiwotzki

Runaway stars are young stars found faraway from their birth place in the Galactic disc. They were usually identified as blue objects at high Galactic latitude. They are the result of ejection due to an close encounter in a stellar cluster or the supernova explosion of their companion in a binary system. Although very interesting in their own right, they are also ideal tools to investigate our Galaxy. Runaway stars can shed light on two aspects of the structure and evolution of our Galaxy.

Present day chemical abundance gradient. Stars nearer to the centre of the Galaxy are born with a higher metallicity than stars further away. Measuring an accurate value of this gradient is a very valuable ingredient to understand the evolution of our Galaxy.

Tracing spiral arms. We know that we are part of a spiral galaxy, but our knowledge of shape and location of spiral arms is quite limited. Very little is known of the spiral structure behind the Galactic centre.

O and B stars are good targets to address both questions. They are young and intrinsically bright. However, they are born in the middle of the Galactic plane and observations are made difficult by reddening and confusion. Early-type runaway stars allow us to overcome this problem. They are found kpc above and below the Galactic plane, well out of the extinction layer.

Until now, identifying runaway stars was a tricky, time consuming issue, because of confusion with hot evolved stars (white dwarfs and their precursors). This will be overcome by the forthcoming catalogue of parallaxes measured by Gaia making the distinction between runaways and lookalikes easy. Spectroscopic follow up will be carried out to measure radial velocities and determine chemical abundances. Once this is done all the information needed to trace back the stars' trajectory to its birth place in the disc is available. Doing this we will know where the star was born.