Galactic Archaeology: Unveiling the History of the Milky Way

Lead Research Organisation: University of Central Lancashire
Department Name: Sch of Comput Engin and Physical Sci


An astonishing archaeological dig is underway - RAVE, the RAdial Velocity Experiment, is using the space motions of one hundred thousand stars (already several times the number previously accumulated throughout history) in the vicinity of the Sun to wind time's arrow back, the first step in a long-term community goal of making a painstaking star-by-star reconstruction of the assembly history of our Milky Way. Disentangling the underlying physics of galaxy formation, whether it be the distant 'far-field' or the local 'near-field' is one of the highest scientific priorities for both STFC and the UK astronomy community, and a fundamental motivation behind several billion-pound technological investments in which STFC has a vested interest. How did the swath of light which cuts across the night sky - the so-called Milky Way - originate? Comprised of hundreds of billions of stars, with ages spanning more than ten billion years, each moving on very different trajectories, with chemical abundance patterns varying by factors of more than a million from star-to-star, and colours and luminosities even more dramatically different, at first glance it would appear all but impossible to piece together the formation history of a disk (or spiral) galaxy like our own Milky Way. Even more daunting would apppear to be attempts to link this highly complex system today with distant 'blobs' of light seen at the edge of the visible Universe with facilities such as the Hubble Space Telescope and other STFC-supported observatories such as Gemini and the European Southern Observatory. Motivated by the extraordinary RAVE dataset, and the promise of proposed next-generation galactic archaeology experiments such as the European Space Agency's GAIA mission and the Gemini/Subaru WFMOS Project, we have developed sophisticated computational techniques capable of tracing the life cycle of galaxies such as the Milky Way, from distant 'blob' to the multi-billion star structure surrounding us today. We have developed a fast and flexible software package which includes virtually all of the important physical processes which drive the formation and evolution of galactic systems; while many codes are available for simulating the kinematics of purely stellar-like systems (so-called 'collisionless' systems), there are very few on the market today which include a parallel treatment of the relevant gas physics. Our software packages include the critical effects of not only gravity, but also gas dynamics, gas heating (from exploding stars and exotic supermassive black holes), gas cooling (from interstellar atomic processes), star formation, chemical enrichment from dying stars, and the luminosity and colour evolution of the stellar populations throughout the simulated galaxy. We are already presented with a number of mysteries regarding the origin of our Milky Way: why do its surrounding satellite 'building blocks' bear chemical fingerprints which look nothing like our Galaxy? why do these satellites appear in special orbits around the Galaxy, as opposed to random orientations? what is responsible for the flared warp in our otherwise flat Milky Way disk? what is the origin of the relatively young clusters observed in the outskirts of our Galaxy, clusters with chemical fingerprints which, again, look nothing like the stars around them? why is our mottled and patchy galactic halo otherwise so homogeneous chemically, save for just a few chemical elements? Datasets such as RAVE, WFMOS, and GAIA provide unprecedented spatial, kinematic, and chemical, data for enormous numbers of stars. With the support of STFC, our team will mine the fossil record of the Milky Way embedded in these new datasets (or guide their potential exploitation, as in the case of WFMOS and GAIA), in order to understand the detailed 3d shape of disk galaxies and their dark matter halos, and gain further insight into the role of accretion events in shaping these properties.


10 25 50
publication icon
Arrigoni M (2010) Galactic chemical evolution in hierarchical formation models - I. Early-type galaxies in the local Universe in Monthly Notices of the Royal Astronomical Society

publication icon

publication icon
Breddels M (2010) Distance determination for RAVE stars using stellar models in Astronomy and Astrophysics

publication icon
Brook C (2013) The Lowest Metallicity Stars in the LMC: Clues from MaGICC Simulations in Publications of the Astronomical Society of Australia

publication icon
Brook C (2012) MaGICC discs: matching observed galaxy relationships over a wide stellar mass range in Monthly Notices of the Royal Astronomical Society

publication icon
Brook C (2014) MaGICC baryon cycle: the enrichment history of simulated disc galaxies in Monthly Notices of the Royal Astronomical Society

publication icon
Brook C (2011) Hierarchical formation of bulgeless galaxies: why outflows have low angular momentum in Monthly Notices of the Royal Astronomical Society

publication icon
Brook C (2012) Thin disc, thick disc and halo in a simulated galaxy in Monthly Notices of the Royal Astronomical Society

Description Our group was the first to simulate a suite of galaxies comparable to our Milky Way, in a fully cosmological framework. The MaGICC (Making Galaxies in a Cosmological Context) was initiated and rose to fruition through the support of this grant. Our feedback prescription now underpins essentially all other such simulation efforts around the world. We have also identified entirely new means to assess the veracity of energy feedback prescription in simulations; specifically, the temporal evolution of metallicity gradients and the power spectrum of the interstellar medium. Feedback remains the primary uncertain physics within simulations, and these new ways in which to test feedback have been acknowledged and embraced by the community.
Exploitation Route Our work has been based upon a handful of extremely high resolution simulations; the critical next step must be the realisation of a statistical sample of galaxies spanning a range in mass and assembly history. These must be run at a comparable ~200pc resolution. Better linking these simulations with full galactic chemical evolution will be required, in order to exploit the next generation of observational experiments, including Gaia-ESO, GALAH, WEAVE, 4MOST, and Gaia.
Sectors Education
Description Our grant was an ambitious one, with 5 primary objectives outlined, and a number of ancillary sub-projects. Tangible outcomes for each have been realised, amongst the 57 refereed papers listed under 'Publications' (40 of which are theoretical developments in support of the work, and 17 relate to the RAVE science which underpins the constraints to which our theory papers are targeted). Each objective - (1) abundance gradients, (2) chemical tagging, (3) age-metallicity-kinematic relations, (4) galactic building blocks, (5) warps/thick discs - was underpinned by the need to develop leading-edge chemical and hydrodynamical tools; before the science objectives could be realised, we needed new simulations which ameliorated many of the problems facing the current generation. Our developments in this regard were two-fold : (i) making use of adaptive mesh refinement techniques (which capture shocks and instabilities in a superior manner to traditional particle-based approaches) to simulate the first Milky Way-like galaxies to redshift z=0 (done: Sanchez-Blazquez et al 2009; House et al 2011; Few et al 20011ab), and (ii) making significant revisions to the conventional manner by which star formation is handled, and energy is fed back into the ISM of simulated disc galaxies (done: Brook et al 2011abc; Pilkington et al 2011abc; Governato et al 2010). Through code development (i), we have been successful in meeting objective (5), providing new insights into the origin of the 'broken' light profiles seen in the most disc galaxies, through the link between gas warps and the density threshold for star formation in the outer discs of Milky Way-like analogs (Sanchez-Blazquez et al 2009), as well as the respective roles of secular and accretion-driven kinematic disc heating in forming thick stellar discs (House et al 2011; Sales et al 2010). For the first time, we have been able to examine the systematic differences in the predicted abundance gradients inferred from mesh- and particle-based disc galaxy simulations, in support of Objective (1); this fascinating study has highlighted the need for high-redshift empirical constraints, in order to direct future developments in the field (Pilkington et al 2011b). Through code development (ii), within our newly developed simulations, we have (a) incorporated a more physically-motivated star formation density threshold, (b) increased the efficiency of supernovae energy coupling to the surrounding ISM, (c) included the effects of thermalised energy deposition from massive star radiation pressure, (d) examined the importance of the initial mass function, and (e) significantly improved the link between ISM density and cooling 'suppression' of Type II supernova remnants (Brook et al 2011abc; Pilkington et al 2011ac; Stinson et al 2011; Maccio et al 2011). These developments have led to a suite of disc simulations which, for the first time, adhere to all scaling relations of which we are aware (Brook et al 2011c), including that between specific star formation and stellar mass, a relation that all extant simulations have failed to recover (Avila-Reese et al 2011). These particle-based simulations (and those we have developed in parallel with our GCD+ SPH code), have been employed in our abundance gradients work (Pilkington et al 2011b; Rahimi et al 2011). Disentangling the relationship between stellar age, kinematics, and metallicity, is fundamental to work by RAVE and our theory programme (Objective 3); our preliminary work in this area can be found in House et al (2011). Our detailed analysis of the new simulations have uncovered a fascinating problem for future exploration. Specifically, in massive galaxies, the metal distribution is overly skewed by accreted satellite stars (Objective 4: Calura et al 2011), while in dwarfs, the distribution is overly skewed by in-situ stars forming from an overly tight/coordinated age-metallicity relation (Objective 3: Pilkington et al 2011c).
First Year Of Impact 2008
Sector Education
Impact Types Societal
Description RAVET (Real and Virtual Eartly-Types) 
Organisation Instituto de Astrofisica de Canarias (IAC)
Country Spain, Kingdom of 
Sector Public 
PI Contribution Chemical and chemo-dynamical models provided in support of observational programmes led by the Spanish partners of RAVET.
Collaborator Contribution Provided empirical constraints to our computational and theoretical models.
Impact 6 refereed journal publications between UCLan and one of the Spanish partners; 3 additional papers submitted at the time of writing; PhD exchanges between UCLan, Valencia, and Madrid, provided unique study opportunities for students in both nations
Start Year 2008
Description Triple Science Celebration - Burnley College 
Form Of Engagement Activity Participation in an open day or visit at my research institution
Part Of Official Scheme? Yes
Geographic Reach Regional
Primary Audience Schools
Results and Impact Roughly 60 students across 5 schools from the northwest; material covered included practical application of Galactic Archaeology simulations and follow-up Q&A sessions with myself

Schools have followed up with requests for 1-2 hour long outreach sessions at our Alston Observatory.
Year(s) Of Engagement Activity 2010