the Pristine survey

The Pristine survey, led by Dr. Else Starkenburg (Kapteyn Astronomical Institute, University of Groningen, the Netherlands) and Dr. Nicolas Martin (CNRS, University of Strasbourg, France), is searching for the most pristine stars in and around our Galaxy. The survey is able to search for these rare stars with unprecedented efficiency: instead of taking spectra of millions of stars, we employ a special colour filter on the Canada-France-Hawaii Telescope to search for stars with relatively few metals in their atmospheres.

More detailed information about the Pristine survey can be found here.

Below are recent news items, updates from the team and new papers. 

Chemodynamics of bright very metal-poor stars with the INT

(May 2024) The recent Gaia data release 3 has resulted in an abundance of bright metal-poor candidates, for example using the Pristine photometric method (Martin, Starkenburg et al. 2023). We successfully followed up a large sample of very/extremely metal-poor stars with spectra using the Isaac Newton Telescope (INT). In this work, we study their dynamical properties - e.g. are they mostly prograde or retrograde? what substructures do they belong to? Have a look at the paper for all the details: 

Viswanathan et al. (2024, preprint): link

Figure caption: energy vs. angular momentum for the stars observed in this work. The star symbols are colour-coded by the spectroscopic metallicity we derived for them. Various halo substructures are highlighted. 

Ancient stars in the Sagittarius dwarf galaxy

(May 2024) We followed up some of the most metal-poor stars in Sagittarius (selected from PIGS) with high-resolution spectroscopy, to investigate what the early chemical evolution of this disrupting dwarf galaxy looks like. Overall the chemistry seems quite similar to other classical dwarf galaxies. Both massive stars and AGB stars contribute to the abundances of VMP ([Fe/H] < -2.0) stars, but not type Ia supernovae. More information here: 

Sestito, Vitali et al. (2024, preprint): link

Figure caption: metallicity versus barium abundance, for stars in the Galactic halo (small grey points), previous measurements in Sagittarius (black) and the stars observed in this work (blue). Only a couple of VMP stars had barium measured before, we add 10 new stars. 

Orbital properties of the oldest stars in the inner Milky Way

(December 2023) In the most recent instalment of the Pristine Inner Galaxy Survey (PIGS 🐷) we study how metal-poor stars in the inner few kiloparsec of the Milky Way move around. We find that they are part of a concentrated population, which is slowly rotating around the Galactic centre. We suggest these stars are a mix of stars born in the very early Milky Way, in large early building blocks and in small accreted dwarf galaxies. Read more here:

Ardern-Arentsen et al. (2023, preprint): link

Figure caption: as a function of metallicity, this figure shows the fraction of stars that is confined within a certain distance from the Galactic centre. The confined fraction drops with decreasing metallicity, but even very metal-poor stars (with [Fe/H] < -2.0) are still quite confined - roughly 60% of them never travel beyond 5 kpc. 

Discovery of an extremely lithium-rich very metal-poor star

(October 2023) In our most recent Pristine paper, we study a serendipitously discovered star in detail to investigate the origin for its extremely high lithium abundance. We are able to measure the 6Li/7Li isotope ratio, which suggests that the main proposed scenario for lithium-rich giants might not be able to explain this star. However, such conclusions are very sensitive to three-dimensional effects in the models - something for future study. For more information, see the paper: 

Sitnova et al. (2023): link

Figure caption: best fit of the two lithium lines used in this work, using non-LTE 1D models (blue lines, and in purple dashed lines separately for 6Li and 7Li). The LTE profiles are shown as green dotted lines and do not fit.  

The GHOST tale of a PIGSlet

(August 2023) Using the recently commissioned GHOST spectrograph at the Gemini South telescope, we studied the detailed chemistry of an intriguing PIGS star that is moving on an elliptical orbit close to the Galactic plane. Our analysis reveals it has been enriched by only a few (energetic) supernovae, and that it likely formed in a small dwarf galaxy that was brought into the Milky Way during early Galactic assembly. All the details can be found here: 

Sestito et al. (2023, preprint): link

Figure caption: Chemical abundance pattern of P180956 (blue and red symbols) and best-fit theoretical supernova yields (best fit = magenta, second-best fit = orange). This star was likely enriched by 3-5 SNe events, in each case from one high-mass star + several lower mass stars. 
Figure caption: for an independent very/extremely metal-poor giants sample we compare high-resolution spectroscopic metallicities (x-axis) to our photometric metallicities based on synthetic CaHK from the BP/RP spectra (y-axis). There is a very good correlation, extending down to the lowest metallicities.

Millions of photometric metallicities from Pristine DR1 & Gaia BP/RP spectro-photometry

(August 2023) Exciting news: our first data release is now available! We also extend the "Pristine-like photometric metallicities" to the full sky using synthetic CaHK magnitudes derived from the Gaia BP/RP spectra. These magnitudes have also been extremely useful to calibrate the Pristine photometry. Combined, both catalogues include more than 2 million metal-poor star candidates as well as more than 200,000 and ~8,000 very and extremely metal-poor candidates. Have a look and explore! The preprint can be found here: 

Martin & Starkenburg et al. (2023): link

A curious carbon-rich star in PIGS studied in detail

(May 2023) In this paper we study an inner Galaxy carbon-enhanced metal-poor (CEMP) in detail, and derive abundances for 32 different elements. We find that this star is in a binary system, and that it is enhanced in both slow and rapid neutron-capture process elements the first of its kind in the Galactic bulge. When we fit different model predictions, the best fit is for intermediate neutron-capture models. Read more here: 

Mashonkina et al. (2023): link

Figure caption: heavy element abundance pattern for the star studied in this work, compared to i-process predictions from two different sources. These fits are better than separate r+s fits. 
Figure caption: location of the selected stars in the Pristine+SDSS colour-colour diagram. All stars lie in the upper part of the diagram, where the most metal-poor stars are expected to be located.

Exploring the lowest metallicity regime in Pristine

(April 2023) In preparation for Pristine follow-up with WEAVE, we have explored the extreme regions of the Pristine colour-colour space by targeting stars with photometric [Fe/H] < 3.5 for high-resolution spectroscopic follow-up. We found that it is crucial to remove photometrically variable stars for a clean selection of true extremely metal-poor stars. For more details, have a look at our latest paper: 

Lombardo et al. (2023): link

The edge of the Hercules dwarf galaxy

(April 2023) In the latest Pristine Dwarf Galaxy Survey paper, we study the Hercules dwarf galaxy. We find three new members using our metallicity-sensitive photometry, one of which is located in the very outskirts of the dwarf galaxy. After careful removal of foreground contamination, we find that there is no sign of a velocity gradient in Hercules, and conclude that there is no clear evidence for tidal disruption of this dwarf galaxy. Read more:

Longeard et al. (2023, preprint): link

Figure caption: known and new members of the Hercules dwarf galaxy. Her_5 is at almost 10 half-light radii from the centre. 

Pristine face-to-face meeting on La Palma

(April 2023) This week we are meeting on La Palma, La Isla Bonita, to discuss the latest updates and plans for the Pristine survey. One of the highlights: seeing the WEAVE multi-object spectroscopic instrument on the William Herschel Telescope! We are excitedly looking forward to the upcoming WEAVE spectra, especially for (Pristine-selected) extremely metal-poor stars. 

Digging in Pristine + LAMOST to find the most metal-poor stars

(January 2023) In the latest Pristine paper we combined the strengths of photometric pre-selection of metal-poor candidates from Pristine with the large LAMOST spectroscopic database, with the goal of finding the undiscovered most metal-poor stars in LAMOST. We followed up a number of promising candidates using the GTC, and present our results here:

Arentsen et al. (2023, accepted for publication in MNRAS): link

Figure caption: spectral fits for the most metal-poor stars in our follow-up of Pristine+LAMOST selected stars. 

Figure caption: comparing the chemical abundances of one of the stars in the sample to yields from PISNe. This star likely contains ~50% enrichment from a PISN. 

Cu and Zn abundances in metal-poor giants

(November 2022) In this work we selected very bright (9.5 < G < 11) metal-poor candidates from Pristine photometry and observed them with high-resolution spectroscopy. We focused on the often-neglected abundances copper and zinc, which are sensitive probes of Pair Instability Supernovae (PISNe) events. Have a read of the paper for more details: 

Caffau, Lombardo et al. (2023): link

Probing interstellar absorption with PIGS spectra

(November 2022) Instead of studying stars like we usually do in Pristine, in this new paper we investigate the interstellar material between us and the targets in PIGS. The combination of the high extinction towards the bulge and the low metallicity of the stars makes several diffuse interstellar bands (DIBs) prominent in the PIGS spectra. We study the relative strength of two DIBs and find that it depends on the Galactic latitude and the reddening, suggesting they might have different carriers and different spatial distributions. Read more here:

Zhao et al. (2022, accepted for publication in MNRAS): link

Figure caption: relative strength of the two DIBs at 442.8nm and 862.1nm as a function of reddening (top) and Galactic latitude (bottom). If the two DIBs had a similar carrier and/or similar spatial distributions, their ratio should not show clear trends - but this is not what we find. 

High-resolution spectroscopic follow-up of PIGS with GRACES

(August 2022) We are happy to present the newest Pristine paper, describing the detailed spectroscopic follow-up of 17 very metal-poor stars from the Pristine Inner Galaxy Survey (PIGS) with the GRACES instrument. We derive abundances and orbital parameters, and investigate possible connections to pair instability supernovae, globular clusters and dwarf galaxies. For more details, have a look at the paper: 

Sestito et al. (2022, preprint): link

Figure caption: comparison of some of the GRACES abundances with literature stars in the field and in globular clusters. The top left corner of this space is mostly populated by second generation globular cluster stars, and some of the PIGS stars lie in this region as well.

A photometric metallicity analysis of the Sagittarius dwarf spheroidal galaxy

(April 2022) The most recent Pristine Inner Galaxy Survey (PIGS) paper does not study the Galactic bulge, but instead focuses on the nearby Sagittarius (Sgr) dwarf spheroidal galaxy. We use the photometric metallicity information from Pristine to study the different spatial distributions of metal-poor and metal-rich stars in Sgr, finding a clear metallicity-gradient throughout our entire footprint. Read more: 

Vitali et al. (2022, preprint): link 

Figure caption: The ratio of metal-poor (MP, [Fe/H] < -1.3) to metal-rich (MR, [Fe/H] > -1.0) stars in Sagittarius increases with increasing distance from the centre of the dwarf galaxy. The colouring of the rings in the top panel corresponds to the colours in the bottom panel. 
In grey, shown are the results of N-body simulations for a globular cluster without dark matter (top) and a system with dark matter (bottom), with the declination on the x-axis. The system with dark matter better reproduces the velocity dispersion of the C-19 stars (shown in orange). From Errani et al. (2022). 

What is the origin of the C-19 stream?

(March 2022) This week there are two new Pristine papers out on the fascinating and puzzling extremely metal-poor C-19 system (see previous post). Yuan et al. present UVES spectroscopy of seven member stars (among which two new members), and find that chemically C-19 looks like a globular cluster (GC) stream, but dynamically it is hotter than expected for a stream of GC origin. Errani et al. study the origin of C-19 using N-body simulations, and find that one way to reproduce the characteristics of C-19 would be if its GC progenitor was embedded in a dark matter halo something not typically expected for GCs. There might be a second part of the stream hiding in a different region of the sky, which could shed more light on the properties of the stream progenitor. Stay tuned for more on C-19 in the future! 

Yuan et al. (2022, preprint): link
Errani et al. (2022, preprint): link

The Milky Way's most metal-poor stellar structure

(January 2022) We are finally able to share the exciting discovery of the C-19 stream with the world! In our investigation of Streamfinder streams with Pristine metallicities, we found one particularly promising candidate with an extremely low metallicity. Subsequent spectroscopic follow-up showed that the stream has an average [Fe/H] of 3.38, the lowest ever measured for any stellar structure. The other chemical abundances point towards a globular cluster origin for C-19, raising questions about the "metallicity floor" of globular cluster formation. It looks like we still have a lot to learn about these ancient fossils!

Martin et al. (2022, Nature): link

More general information in the press release here

A map of the Milky Way globular clusters overlaid on top of the Milky Way map obtained by the Gaia satellite. The color of the symbols represent their metallicity. The stars of the C-19 stream are shown with star symbols and have a much lower metallicity than any cluster, 2500 times fewer heavy elements when compared to the Sun. Credits: N. Martin & Observatoire Astronomique de Strasbourg; Canada-France-Hawaii Telescope / Coelum; ESA/Gaia/DPAC
Metallicities vs. mean Galactocentric distance for the Pristine+Streamfinder streams, colour-coded by the mean angular momentum along the direction perpendicular to the MW plane. 

Pristine metallicities for 26 stellar streams

(January 2022) In this paper we use the powerful combination of the Pristine photometric metallicities with the stream discoveries from STREAMFINDER + Gaia EDR3. We estimate metallicities for 26 stellar streams - several of these streams are new candidates, and three of the streams appear to be extremely metal-poor! Read more: 

Martin et al. (2022, preprint): link

The largest high-resolution spectroscopy Pristine sample

(December 2021) In the newest Pristine paper, we present a "CFHT ESPaDOnS view on the Milky Way halo and disc populations" using high-resolution spectroscopy for 132 Pristine-selected stars. We discuss the abundances of alpha-elements, s-process elements and carbon and derive orbital properties for all the stars in the sample, which cover a large range of metallicities. Read more here: 

Lucchesi et al. (2022, accepted for publication in MNRAS): link

(Figure) Derived carbon abundances for the ESPaDOnS sample, with main-sequence/lower RGB stars in the top panel and upper RGB stars in the bottom panel (a comparison literature sample is shown in grey). The Pristine selection appears to lack carbon-rich giants. 
Supernova model yields compared to the NLTE abundance patterns of Pr 221 (top) and Pr 237 (bottom). The best fit models are plotted as a continuous black lines.

A chemical analysis of two ultra-metal-poor stars

(September 2021) One of the main purposes of the Pristine survey is to identify and study in detail the most metal-poor stars. In this paper, we present new data for two previously discovered ultra metal-poor stars, Pr 221 and Pr 237. We derive non-LTE corrected abundances, and find that both stars likely had low-energy supernova progenitors. Read more here: 

Lardo et al. (2021, preprint): link 

Probing the outskirts of the dwarf galaxy Boötes I

(July 2021) In the fourth instalment of the Pristine Dwarf-Galaxy Survey, we studied the outskirts of the ultrafaint dwarf galaxy Boötes I, using spectra from AAT with AAOmega+2dF. We detect a velocity gradient and a metallicity gradient, and we find a larger elongation than previously found. These observations suggest that Boötes I may have been affected by tidal interactions with the Milky Way. Read more: 

Longeard et al. (2021, preprint): link

Spatial distribution of Boötes I members (red squares), compared to that of a simulated stream under the hypothesis that Boötes I is disrupting (green diamonds). 
Rotational velocity as function of Pristine metallicity [Fe/H]. Many different structures are visible, among which the very and extremely metal-poor extensions of the disc.

Uncovering the very metal-poor tail of the thin disc

(June 2021) In this new Pristine paper, we combine our CaHK photometry with the excellent astrometry from Gaia EDR3 to study the rotational properties of stars towards the Galactic anti-centre. Intriguingly, we find a well populated extension of the kinematical thin disc down to [Fe/H] ~ −2, with a scarcer extension reaching the extremely metal-poor regime, down to [Fe/H] ∼ −3.5 dex. To read more:

Fernández-Alvar et al. (2021, preprint): link

PIGS III: carbon-enhanced metal-poor stars in the bulge

(May 2021) In the newest Pristine Inner Galaxy Survey (PIGS) paper, we study the occurrence of carbon-enhanced metal-poor (CEMP) stars in the bulge region. Previously only a handful of CEMP stars were known there, and with PIGS we increase this by 96 CEMP stars! However, the fraction of CEMP stars seems to be smaller than in the rest of the halo. Why could that be? Read more:

Arentsen et al. (2021): link

Metallicity vs. carbon abundance diagram. The new CEMP stars from PIGS are shown as black dots, on top of a comparison sample of halo CEMP stars in colour. 

Virtual Pristine meeting

(March 2021) This year we met online for our "face-to-face" collaboration meeting, which was very fruitful despite happening quite early and/or late across various timezones. We discussed ongoing data efforts, projects, papers, proposals and interesting related science - there are a lot of exciting things happening, so keep an eye out!

Abundances and orbital properties using Pristine GRACES spectra

(December 2020) In a new paper, we present high-resolution observations with the GRACES instrument for 30 Pristine-selected stars with [Fe/H] < 2.5. There are several stars with interesting chemical signatures. Additionally, an orbital analysis of the sample reveals five extremely metal-poor stars with orbits in the plane of the Milky Way. What this tells us about the formation of our Galaxy? Read more here: 

Kielty et al. (2020, preprint): link

Dynamical properties of the GRACES sample. The five planar extremely metal-poor stars are highlighted with star symbols. 
Very metal-poor (VMP) stars in several bulge surveys. With PIGS, the number of known VMP stars in the inner Galaxy is increased significantly.

PIGS: the largest set of metal-poor stars in the Milky Way bulge

(June 2020) One of the most challenging places to look for metal-poor stars is in the Galactic bulge. It is plagued by lots of dust, stellar crowding and huge numbers of metal-rich stars that need to be filtered out. In the recent second Pristine Inner Galaxy Survey (PIGS) paper, we present our successful efforts at finding these stars. 

Arentsen et al. (2020): link 

New paper: Revealing the nature of Sagittarius II

(May 2020) It is not trivial to determine the nature of the faintest satellites around the Milky Way – globular clusters and small dwarf galaxies can look very similar. One particularly challenging case is that of Sagittarius II, which was also the topic of a previous Pristine paper. In this new paper we present convincing evidence that this system is a globular cluster, although it is quite unique due to its relatively large size. 

Longeard et al. (2020): link

The observed velocity dispersions of most globular clusters (black) and confirmed dwarf galaxies (blue) of the MW as a function of their absolute magnitudes, and the location of Sagittarius II (red) 
The observed spectra for two extremely metal-poor stars ([Fe/H] < 3.0) from the FORS2 sample.

Pristine XI: metal-poor stars observed at the VLT

(March 2020) In the Pristine XI paper, we analyse a sample of low-resolution spectra of metal-poor stars, observed with FORS2 at the ESO VLT. We test the power of combining Gaia information with the Pristine CaHK photometry to pre-select metal-poor candidates. 

Two interesting results: we find some very metal-poor stars that are not enhanced in α-elements, and we find that this sample has a low fraction of carbon-enhanced metal-poor (CEMP) stars. A deeper investigation of both is necessary in the future. 

Caffau et al. (2020): link

Pristine team meeting 

(February 2020) From February 10-14, we are meeting at the ISSI in Bern, Switzerland for our Pristine collaboration meeting. Since the team is always growing, these meetings are useful to hear what's going on. Besides such updates, we have two days of discussion on preparation for WEAVE. We are excited about the next few years! 

Our ISSI website: https://www.issibern.ch/teams/pristine/

The tail of the halo metallicity distribution function from this work (final one in black) and from previous works based on spectroscopic samples or theory

New paper: the metal-poor tail of the Galactic halo

(January 2020) Both from the theoretical and the observational perspective, the number of extremely and ultra metal-poor stars in the Milky Way is poorly constrained. In this work, we use the unique capabilities of Pristine to produce photometric metallicities down to [Fe/H] = −4 for ten thousands of (turnoff) stars. We find that there may be more of the most metal-poor stars than expected previously!

Youakim et al. (2020): link 

Pristine X: very metal-poor stars in the Milky Way disk

(November 2019) Today marks a special day: the tenth Pristine paper is now available as a preprint! In this paper, we studied the orbital properties of our spectroscopic Pristine sample and a sample of LAMOST very metal-poor stars. We found that many of them have orbits that stay within the plane of the Milky Way, which is quite unexpected!

Sestito et al. (2020): link 

Pristine VII: the Galactic outer halo with BHB stars

(October 2019) It has been a productive time for us lately, with another Pristine paper being accepted for publication! In this work, we use the capabilities of the Pristine CaHK photometry combined with SDSS u-band photometry to cleanly select blue horizontal branch (BHB) stars. These are great for studying the Milky Way's halo out to far distances!

Starkenburg et al. (2019): link

(Figure) Photometric selection plane for BHB stars: the top row shows the previous selection using SDSS photometry only, whereas the bottom row includes the Pristine CaHK photometry. The separation between blue stragglers (red) and BHBs (blue) is much more evident.

The first Pristine bulge paper

(October 2019) We are happy to present the first paper in the Pristine Inner Galaxy Survey (PIGS) series! In this paper, we study the kinematics of metal-poor stars in the Galactic bulge region.

Arentsen et al. (2019): link

Toomre diagram for the stars studied in this paper, showing most of the very metal-poor stars have halo-like velocities, as expected.

Pristine IX: high-resolution spectra & orbits

(October 2019) We are skipping two papers into the future: the preprint for the Pristine IX paper is now available. In it we look at a sample of bright (V<15) very metal-poor stars with the CFHT ESPaDOnS spectrograph, studying a subset of abundances and the orbital properties of the stars. 

Venn et al. (preprint): link

3 years of Pristine follow-up: paper VI

(September 2019) In the past three years we have done a lot of low- and intermediate resolution spectroscopic follow-up of metal-poor Pristine candidates. Many, many observing nights of a large number of collaboration members at the INT, WHT and NTT have resulted in a beautiful sample that we are proud to present!

Aguado, Youakim et al. (2019): link

The current photometric footprint of the Pristine survey, covering ~5000 deg2. The stars that have been followed up are shown as cyan points, these were selected from the ~2500 deg2 which was available at that time.

New paper: Pristine V

(May 2019) The fifth Pristine survey paper (Bonifacio et al. 2019) is now accepted and online. In this paper, we determined atmospheric parameters, chemical abundances and orbital parameters for 40 bright Pristine candidates observed with SOPHIE@OHP.

We investigated a new method for estimating metallicity, Teff and log g combining Pristine photometry with Gaia photometry and parallaxes.

Links to the paper: MNRAS or open access

News: Pristine ISSI meeting

(January 2019) - From January 7-11, 2019, we met at the International Space Science Institute (ISSI) in Bern, Switzerland. Topics of discussion were our photometric metallicities, preparations for spectroscopic follow-up with WEAVE, the dedicated surveys on the dwarf galaxies and on the Galactic bulge, abundance measurements from intermediate/high-resolution spectra (including NLTE corrections), and the synergy between Pristine and Gaia. More information: http://www.issibern.ch/teams/pristine/

News: open positions

(November 2018) - The Pristine collaboration is advertising two postdoctoral research positions in France, at the Observatoire astronomique de Strasbourg, in Strasbourg, and at the Observatoire de la Côte d'Azur, in Nice. Application deadline January 7th, 2019. More information here

News: discovery

(October 2018) - Recently, the Pristine survey discovered of one of the most metal-poor stars known. A press release can be found here

The Pristine logo

(October 2018) - Our survey now has a logo. The CaH&K filter on CFHT makes us what we are! 

News: Pristine meeting

(September 2018) - Recently, we had our Pristine face-to-face meeting at the NRC Herzberg Institute in Victoria, Canada. We shared updates about the photometry, low- and high-resolution spectroscopic follow-up and different ongoing projects. We are looking forward to publishing many interesting papers in the (near) future!