Mnras0411 2770

Mon. Not. R. Astron. Soc. 411, 2770–2791 (2011) doi:10.1111/j.1365-2966.2010.17890.
Catalogues of hot white dwarfs in the Milky Way from GALEX’s

ultraviolet sky surveys: constraining stellar evolution
Luciana Bianchi,1 Boryana Efremova,1 James Herald,1 Léo Girardi,2

Alexandre Zabot,3 Paola Marigo4 and Christopher Martin5
Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

1 Department of Physics and Astronomy, Johns Hopkins University, 3400 N. Charles Street, Baltimore, MD 21218, USA
2 Astronomical Observatory of Padova, INAF, Vicolo dell’Osservatorio 5, I-35122 Padova, Italy
3 Universidade Federal da Fronteira Sul, Campus de Laranjeiras do Sul, Brazil
4 Department of Astronomy, University of Padua, Vicolo dell’Osservatorio 3, I-35122 Padova, Italy
5 California Institute of Technology, Pasadena, CA 91125, USA
Accepted 2010 October 16. Received 2010 August 3; in original form 2010 March 9
ABSTRACT
We present comprehensive catalogues of hot star candidates in the Milky Way (MW), selected
from Galaxy Evolution Explorer (GALEX) far-UV (FUV; 1344–1786 Å) and near-UV (NUV;
1771–2831 Å) imaging. The FUV and NUV photometry allows us to extract the hottest stellar
objects, in particular hot white dwarfs (WD), which are elusive at other wavelengths because of
their high temperatures and faint optical luminosities. We generated catalogues of UV sources
from two GALEX’s surveys: All-Sky Imaging Survey (AIS; depth AB magnitude ∼19.9/
20.8 in FUV/NUV) and Medium-depth Imaging Survey (MIS; depth ∼22.6/22.7 mag). The two
catalogues (from GALEX fifth data release) contain 65.3/12.6 million (AIS/MIS) unique UV
sources with errorNUV 0.5 mag, over 21 435/1579 deg2 . We also constructed subcatalogues
of the UV sources with matched optical photometry from Sloan Digital Sky Survey (SDSS;
seventh data release): these contain 0.6/0.9 million (AIS/MIS) sources with errors 0.3 mag
in both FUV and NUV, excluding sources with multiple optical counterparts, over an area
of 7325/1103 deg2 . All catalogues are available online. We then selected 28 319 (AIS)/9028
(MIS) matched sources with FUV − NUV < −0.13; this colour cut corresponds to stellar
T eff hotter than ∼18 000 K (the exact value varying with gravity). An additional colour cut
of NUV−r > 0.1 isolates binaries with largely differing T eff s, and some intruding quasi-
stellar objects (QSOs; more numerous at faint magnitudes). Available spectroscopy for a
subsample indicates that hot-star candidates with NUV−r < 0.1 (mostly ‘single’ hot stars)
have negligible contamination by non-stellar objects. We discuss the distribution of sources
in the catalogues, and the effects of error and colour cuts on the samples. The density of
hot-star candidates increases from high to low Galactic latitudes, but drops on the MW plane
due to dust extinction. Our hot-star counts at all latitudes are better matched by MW models
computed with an initial–final mass relation (IFMR) that favours lower final masses. The
model analysis indicates that the brightest sample is likely composed of WDs located in the
thin disc, at typical distances between 0.15 and 1 kpc, while the fainter sample comprises also
a fraction of thick disc and halo stars. Proper motion distributions, available only for the bright
sample (NUV < 18 mag), are consistent with the kinematics of a thin-disc population.
Key words: catalogues – surveys – stars: evolution – white dwarfs – Galaxy: stellar content
– ultraviolet: stars.
branch (AGB) and planetary nebula (PN) phases, in which they shed
1 I N T RO D U C T I O N
much of their mass. The ejected material enriches the interstellar
The vast majority of stars (initial mass 8 M ) end their lives medium (ISM) with newly synthesized nuclear products (mainly
as white dwarfs (WD), after passing through the asymptotic giant He, C, N and possibly O) to different extents, depending on the
initial stellar mass and exact evolutionary path (e.g. Marigo 2001;
Karakas 2010). Intermediate-mass stars are the main providers of
E-mail: [email protected] carbon and nitrogen, whereas low-mass stars are the most relevant
C 2010 The Authors
Monthly Notices of the Royal Astronomical Society

C 2010 RAS
Hot white dwarfs: UV catalogues 2771
component for the mass budget of stellar remnants in galaxies. et al. (2007), Heller, Homeier & Dreizler (2009, WD-M star bina-
Most of the stellar mass is shed in the AGB and PN phases, but ries) from spectroscopy; other authors used optical-IR photometry
the evolution through these phases is still subject to considerable to search for binaries among the known WDs (e.g. Watcher et al.
uncertainties, in particular regarding mass loss and the efficiency of 2003; Tremblay & Bergeron 2007). Special classes such as CVs are
the third dredge-up. Stars within an initial mass range of ∼0.8 to addressed by other works (e.g. Szkody et al. 2009 and references
8 M end as WDs with a narrow mass range, mostly below 0.8 M . therein; Gänsicke et al. 2009), while others studied in detail the
It is important to understand how the mass of their precursors relates very local population (e.g. Holberg, Oswalt & Sion 2002 discuss
to the final WD mass, in order to understand the relative contribution 122 objects within 20 pc from the Sun, which they estimate to be an
of different stars to the chemical enrichment of elements such as 80 per cent complete sample within this distance, and which include

He, C, N and O. some double-degenerate systems).
While the evolution of the WD progenitors in the main-sequence Finding the hottest, smallest stars, however, remained a challenge
phase is fairly well understood and observationally constrained, the prior to the GALEX UV sky surveys, which provide deep sensitivity
hot-WD population is hitherto quite elusive, owing to their small and large area coverage. For example, 105 were found in the original
radius, hence low optical luminosity, and extremely hot temper- EUVE whole-sky survey, with small subsequent additions obtained
atures, to which optical colours are insensitive (see e.g. Bianchi by combining EUVE and ROSAT [see e.g. Dupuis (2002) for a
2007; Bianchi et al. 2007a,b) as well as to their very short lifetimes review and discussion]. The ROSAT whole-sky survey produced
on the constant-luminosity post-AGB phase. To make matters more 175 WD in X-rays, mostly DAs (Fleming et al. 1996, from ROSAT
difficult, the post-AGB luminosity at a given stellar temperature PSPC).
varies significantly according to the stellar mass, making it impos- Such catalogues of confirmed or candidate WDs enable the study
sible to infer absolute luminosity from other physical parameters. of these objects as astrophysical probes of stellar evolution, of MW
The evolutionary time spent on the constant-luminosity post-AGB structure, of the local neighbourhood, etc. Most importantly, com-
phase and on the cooling track is also a strong function of the mass prehensive catalogues enable the selection of targets for follow-up
(Vassiliadis & Wood 1994). Therefore, the exact relation between spectroscopy, which then provides the detailed physics of these
progenitor’s initial mass and WD mass [initial–final mass relation objects, especially when extended to the UV and far-UV (FUV)
(IFMR)] remains to date a crucial missing link in our understanding wavelengths, as proven by the score of results enabled by IUE,
of stellar evolution and chemical enrichment of the Galaxy. FUSE and HST spectrographs.
A characterization of the population of hot WDs in the Milky Way Bianchi (2007, 2009) and Bianchi et al. (2007a, 2009a, 2005)
(MW) can reduce these uncertainties, and lead to a better under- have demonstrated the power of FUV and near-UV (NUV) mea-
standing of processes that drive the chemical evolution of galaxies surements, afforded for the first time over large areas of the sky by
like the MW. UV photometry combined with optical measurements GALEX, to unambiguously detect and characterize extremely hot
significantly increases the sensitivity to the hottest temperatures. stellar sources. Not only the UV wavelengths are more sensitive to
For example, the colour difference between a T eff = 50 000 K and the temperatures of hottest stars, but the combination of UV and
20 000 K star is about 1.5 mag in FUV-g, but <0.4 mag in U − optical colours also allows a better separation of different classes of
B, and <0.15 mag in g − r which are comparable to photometric astrophysical objects (e.g. Bianchi 2009 and references therein).
errors when large surveys are considered. The sensitivity gained by Our present work provides a selection of hot-star candidates from
extending the measurements to UV wavelengths is more critical for UV photometry; about 40 000 hot-star candidates with FUV, NUV
discerning the hottest stars (see e.g. figs 5–7 of Bianchi et al. 2007a; photometric errors 0.3 mag (about 74 000 with photometric errors
Bianchi 2009). 0.5 mag), have also SDSS optical photometry. The majority are
The census and characterization of cool compact objects has likely hot WD with log(g) between 7 and 9. The catalogue covers
significantly improved in recent years thanks to optical and IR different latitudes and enables a first analysis of this stellar pop-
surveys. The Eisenstein et al. (2006) catalogue from the Sloan Dig- ulation with MW models. Larger catalogues of UV sources with
ital Sky Survey (SDSS) fourth data release (DR4) contains 9316 and without optical SDSS match are also constructed, and made
spectroscopically confirmed WDs and 928 subdwarfs over an area available as online products.
of 4783 deg2 ; about one fourth (2741) have T eff > 18 000 K as In Section 2, we describe the method used to construct ‘clean’
estimated by the SDSS pipeline automated spectral analysis. An catalogues of unique UV sources from two GALEX surveys with
additional ∼5–6000 WD are expected from seventh data release different depths and coverage, and subsets of these catalogues with
(DR7; Kleinman, Nitta & Koester 2009). Seven ultracool WD were matched optical photometry, and we present the catalogues’ char-
added to the census by Harris et al. (2008). Gontcharov et al. (2010) acteristics. In Section 3, we extract samples of hot-star candidates,
extracted from Two Micron All Sky Survey (2MASS), Tycho-2, and analyse them with MW models in Section 4. Discussion and
XPM and UCAC3 catalogues combined 34 WDs, 1996 evolved conclusions are given in Section 5.
(11 000 < T eff < 60 000 K) and 7769 unevolved (T eff < 7000 K)
subdwarfs, using multi-colour photometry and proper motions of
2 T H E U V S K Y S U RV E Y S A N D T H E S O U R C E
stars with 6 < Ks < 14 mag. The current version (2008) of the
C ATA L O G U E S
McCook & Sion (1999) catalogue of spectroscopically confirmed
WDs includes 10 132 entries (all types), a factor of 5 increase over
2.1 The sky survey data
a decade, with respect to the original (1999) version listing 2249
WDs. Vennes et al. (2002) give a catalogue of 201 DA WDs, spectro- The Galaxy Evolution Explorer (GALEX) is imaging the sky in FUV
scopically confirmed from the ∼1000 H-rich DA WDs discovered (1344–1786 Å, λeff = 1538.6 Å) and NUV (1771–2831 Å, λeff =
in the 2dF QSO redshift survey; Croom et al. (2004) 2dF cata- 2315.7 Å) simultaneously, with a field-of-view of 1.◦ 2 diameter and
logue includes 2071 WDs over 2000 deg2 . WDs in binaries from a resolution of 4.2/5.3 arcsec (FUV/NUV) (Morrissey et al. 2007).
the SDSS were catalogued and studied by several authors, see e.g. The images are sampled with 1.5 arcsec pixels. Nested surveys
Rebassa-Mansergas et al. (2010) and references therein, Silvestri with different depth and coverage are in progress. The widest sky

C 2010 The Authors, MNRAS 411, 2770–2791

C 2010 RAS
2772 L. Bianchi et al.

Figure 1. Top: sky coverage (in Galactic coordinates) of the GALEX data release GR5 showing the major surveys: AIS (orange), MIS (green) and DIS (blue).
Bottom: the portion of the GALEX GR5 sky coverage for AIS and MIS overlapping with SDSS DR7.
coverage is provided by the All-Sky Imaging Survey (AIS) and the coverage of the overlap is 7325 (AIS) and 1103 (MIS) deg2 . Details
Medium [depth] Imaging Survey (MIS), that reach typical depths of the area calculation are provided in Section 2.4. The sky coverage
of 19.9/20.8 mag (FUV/NUV) and 22.6/22.7 mag (FUV/NUV), re- of AIS and MIS in the GALEX data release GR5 and its overlap
spectively, in the AB magnitude system. The Nearby Galaxy Survey with SDSS DR7 are shown in Fig. 1.
(NGS; Bianchi et al. 2003; Bianchi 2009; Gil de Paz et al. 2007),
with over 300 fields at MIS depth, targeted nearby, hence fairly
extended, galaxies, therefore it has been excluded in the present 2.2 The catalogue of unique UV sources
work, lest some sources from galaxies ‘shredded’ by the pipeline Here, we describe the procedure used to construct the catalogue of
intrude on our catalogue. See also Bianchi (2009, 2010); Bianchi unique GALEX sources (i.e. eliminating repeated measurements).
et al. (2007a); Bianchi et al. (2010) for a general discussion of the All catalogues described in this paper are made publicly avail-
content of the UV sky surveys. able from our website http://dolomiti.pha.jhu.edu/uvsky, and from
In this paper, we use data from the GALEX fifth data release MAST at http://galex.stsci.edu, and as High-Level Science Products
(GR5) AIS and MIS surveys, which include a total of 28 269 and (http://archive.stsci.edu/hlsp/, in the ‘Catalogues’ section). There-
2161 fields, respectively. The data are taken from the Multimission fore, we provide here information on how they were constructed that
Archive at the Space Telescope Science Institute (MAST) archive. will be relevant for potential users, as well as to others interested in
We restrict the catalogues to sources within the central 1◦ diameter constructing future versions of similar samples.
of the field (for good photometry and astrometry, and to exclude We extracted catalogues of GALEX sources from STScI
edge artefacts). With such restriction, and eliminating overlaps, MAST (www.mast.stsci.edu), at the CASJobs SQL interface
these surveys cover a total unique area of 21 434.8 (AIS) and 1578.6 (www.galex.stsci.edu). Sources were extracted from the table ‘pho-
(MIS) deg2 (Section 2.4). Section 2.2 describes the construction toobjall’, from the MIS and AIS surveys separately, with the criteria
of the catalogue of unique GALEX sources. In order to separate that the source distance from the field centre had to be 0.◦ 5 and
the UV sources by astrophysical classes, we examine in this work the photometric error less than 0.5 mag in NUV. In other words, we
the portions of the GALEX GR5 AIS and MIS surveys that are initially included in our general GALEX source catalogue all NUV
also included in the footprint of the SDSS DR7, which provides reliable detections, regardless of whether they have also an FUV
five optical magnitudes: u g r i z in addition to the GALEX FUV, detection. For our selection of hot-star candidates, we will eventu-
NUV magnitudes. The overlap between GALEX GR5 and SDSS ally impose the additional criterion of good FUV photometry. The
DR7 includes 10 316/1655 (AIS/MIS) GALEX fields, and the area additional restriction of errFUV < 0.5 mag significantly reduces the

C 2010 RAS
400 10000
MIS 8000
Number of fields
300 AIS
6000
200
4000
100
2000

0 0
0 1000 2000 3000 4000 0 100 200 300 400
FUV Exposure time (sec) FUV Exposure time (sec)
1.2•104
MIS AIS
1.0•104
Number of UV sources in each field
8.0•103
6.0•103
4.0•103
2.0•103
0
-50 0 50
Galactic latitude
Figure 2. Top: distribution of FUV exposure times for AIS (10 s bins) and MIS surveys (50 s bins); a few fields have longer exposure times, off the scale of the
plots. Bottom: number of UV sources (all UV detections, not just our selected hot stars) in each GALEX field (1◦ diameter), before merging the catalogues and
removing overlaps. Three shades of blue (light/medium/dark) for AIS indicate exposure times of <120/120–220/>220 s, respectively, and lighter/darker green
for MIS fields indicate FUV exposures less/more than 1200 s. Generally, the number of UV source detections in a field increases with depth of exposure, as
expected, and as seen more distinctly for the AIS at the high latitudes, uncomplicated by dust extinction. A few fields, including some with short exposures, have
overdensities. While the number of MW stars increases towards the MW disc, as shown by the AIS sources, at MIS depth most UV sources are extragalactic,
hence show no correlation with Galactic latitude except for the foreground extinction. The sharp drop in the centre, due to dust extinction, defines the MW dust
disc.
number of sources (by a factor of up to 10; see Table 2), with respect The typical exposure time for MIS is 1500 s, which is met or
to the total number of NUV detections, and of course introduces exceeded by the majority of fields. The AIS survey aims at exposure
a bias in the source catalogue, the hottest and bluest sources (the times of the order of ∼100 s. We retained also fields with exposures
subject of this paper) being not affected but the redder sources being shorter than typical. Therefore, while the typical depth of the two
progressively eliminated, as discussed in Sections 2.5 and 3.3. For surveys (AIS and MIS) differ by ∼2 mag, the exposure level is not
more discussions about statistical properties, and biases inherent to strictly homogeneous across each catalogue.
sample selections, see Bianchi et al. (2010). Fig. 2 (bottom panel) shows the number of GALEX sources in
A few observations planned as part of the MIS or AIS surveys each GALEX field. We colour-coded the fields by three ranges of ex-
actually partly failed, and resulted in one of the detectors (most often posure time for the AIS (two for MIS), since the number of sources
FUV) not being exposed (such observations are typically repeated detected above a given error-cut generally increases with exposure
later). The observations with one of the two bands having zero time. This plot is useful to check for fields with overdensities, since
exposure time were eliminated from our catalogue, otherwise they the surveys, the broad AIS in particular, include also some stellar
will bias the statistics of FUV-detection over the total number of clusters. In general, the spread in number of sources per field is just
NUV detections, and the corresponding fields were not counted in about a factor of 2 for MIS. A few AIS fields around latitude −30◦
the area calculation. have an overdensity of almost 1 dex; however the total number of
Fig. 2 shows the distribution of exposure times (FUV is shown, AIS fields in each 10◦ -wide latitude bin (used in our analysis; see
but NUV is generally equal or larger), for the AIS and MIS fields. the next section) is very large, and a few overdensities do not affect


C 2010 RAS
our analysis of stellar counts with MW models over wide areas. As with the GALEX searches, the SDSS searches were done in an
Fields with high density of sources, and in particular the stellar automated way on 1◦ -wide Galactic latitude strips using the casjobs
clusters included in the surveys, will be separately analysed else- command-line tool. We found that there is currently a bug in the
where. They are included in our catalogue for completeness and for SDSS version of this tool (v0.03) that causes the last character of a
possible use by others, although customized photometry is desirable file to be dropped when extracting query results. We compensated
in very crowded fields (e.g. de Martino et al. 2008). for this by writing our query to pad the end of the file with an
Nearby galaxies are generally observed as part of the ‘NGS’ additional character.
survey which is excluded from our catalogue, but a few large galax-
ies are also in the footprint of AIS and MIS and bright knots of
2.3.1 Multiple matches

galaxies ‘shredded’ by the pipeline may enter the catalogue as sep-
arate sources (see Bianchi et al. 2007a). A few such sources may A given GALEX source may have multiple SDSS matches within
have FUV−NUV <−0.13 and appear point like in the SDSS cata- the search radius, given the higher spatial resolution of the SDSS
logue, and therefore may enter our catalogue of hot-star candidates (∼1.4 arcsec). In such cases the CASjob search returns multiple
(Section 3), but they would not affect the statistical results. For lines with the same GALEX source and the various SDSS matches.
other, more specific uses of our catalogue, they may be removed We ‘ranked’ the multiple SDSS matched sources based on the dis-
by checking against a list of nearby galaxies, as we did in Bianchi tance, with the closest SDSS source being retained as the ‘match’
et al. (2007a). Using the ‘child’ flags from the SDSS pipeline, that for the GALEX source and the additional SDSS sources being
track deblended sources, proved not to be useful in identifying and noted. When using UV-optical colours for source classification,
weeding out sources from ‘shredded galaxies’. UV sources with multiple optical matches must be excluded, be-
The searches and download of GALEX sources with the above cause, even though the closest optical source may be the actual
criteria were done using the java CASJobs command-line tool counterpart, the UV photometry at the GALEX spatial resolution
(casjobs.jar). One problem we often encountered when using this of ∼5 arcsec may be a composite measure of two optical sources,
tool to extract and download query results from the data base is and therefore the colour would not be meaningful. The fraction
that sometimes the download pipe gets broken prematurely, and the of UV sources with multiple optical matches, eliminated from the
results are truncated without any warning being given. In order to analysis sample, is then taken into account when estimating den-
verify that the output file contained all the results, a separate count sity of astrophysical sources (# per deg2 ). This fraction is given in
query was run on the SQL server and compared to the number of Table 2. Sources with multiple optical matches are listed separately
output sources in the downloaded file each time. It was necessary in our hot-star catalogues, and are included with ‘rank’>0 in the
to subdivide the search in small latitude strips due to various limits total matched-source catalogue. The relative number of UV sources
set by the CASjobs interface (e.g. query length, output file size), as with multiple optical matches, shown in Fig. 3, increases towards
well as to avoid the frequent problem of the long searches being low Galactic latitudes, as expected. The figure refers to our final cat-
interrupted. alogue with match radius of 3 arcsec, and point-like sources only:
The GALEX archive contains multiple observations of the same approximately 10 per cent of the UV sources have more than one
source, when some fields overlap or are repeated. Having all mea- optical counterpart, at intermediate and high Galactic latitudes, in
surements of each source is useful for variability studies (which
will be addressed in a forthcoming work) and for choosing the best
measurement when several are available. For our present purpose
we constructed from the total output a unique-source catalogue,
in the following manner. GALEX sources were considered possi-
ble duplicates if they lied within 2.5 arcsec of each other. If two
GALEX sources were within this distance, but had the same ‘pho-
toextractid’ (i.e., they are both from the same observation), they
were both considered unique. Otherwise, they were assumed to be
multiple observations of the same source. We choose, to represent
the unique source, the measurement with the longest NUV exposure
time. In the case of equal exposure times, the observation where the
source was closer to the field centre was chosen (i.e. the source
with the smallest ‘fov_radius’ value from the ‘photoobjall’ table),
as photometric quality is usually better in the central part of the
field.
2.3 The matched UV-optical source catalogue

A portion of the GALEX survey areas is in the footprint of the SDSS
Figure 3. Fraction of UV sources having multiple optical matches within
DR7 (see Fig. 1, bottom panel), and for the UV sources in this area
a 3 arcsec match radius, as a function of Galactic latitude. Green dots show
we constructed a catalogue of matched optical sources. We uploaded MIS data, blue dots show AIS data, divided in strips of 1◦ (small dots), as
the coordinates of our GALEX unique sources into the SDSS SQL well as averaged over 5◦ (blue circles). The increase towards the Galactic
interface (version 3_5_16 rev 1.70, at casjobs.sdss.org) and queried plane reflects the higher density of optical sources at low MW latitudes. The
for SDSS matches against the ‘photoprimary’ table (SDSS source plot also shows a slight North–South asymmetry. The fractions shown are
catalogue that includes only unique sources) using an initial search for point-like sources, with error cuts of 0.3 mag. The fraction increases if
radius of 4.2 arcsec. A match radius of 3 arcsec was eventually used a larger match radius is used. The red dots are the incidence of spurious
in the final catalogue, as we shall see. matches (see text).

C 2010 RAS
agreement with our previous work on earlier data releases which 2.3.4 Comparison with other catalogues
covered mainly high Galactic latitudes (Bianchi et al. 2005, 2007a).
We point out, in the interest of users of our catalogues, that there
The fraction increases to slightly over 20 per cent for latitudes |b|
are several differences from other basic matched source catalogues
≈ 10◦ –20◦ , reflecting the higher density of stars in the MW disc,
posted on MAST. Most notably, we include only the central 1◦ di-
and is still very uncertain at latitudes |b| < 10◦ , where we have little
ameter portion of the field, therefore we conservatively eliminate all
area coverage. For a larger match radius (4.2 arcsec), the fraction of
edge sources (mostly defects or bad measurements but also some
multiple matches increases significantly, relative to the total number
good sources), but we retain all good portions. Secondly, we con-
of matches, and so does the incidence of spurious matches (Sec-
sider all existing measurements for each source, from which the
tion 2.3.3), therefore we adopted a final match radius of 3 arcsec in
best is chosen. For example, the matched catalogue described by

our catalogue. The fraction is also higher if extended sources are
Budavari et al. (2009) instead eliminates overlap by reducing each
included.
GALEX field to a hexagon (selection of primary). In this proce-
dure, the vertices of the hexagons include sources near the edge
(i.e. possibly also sources with poor photometry, and artefacts).
2.3.2 Other caveats
Moreover, since the hexagons are fixed at nominal positions, but
In the interest of others who may want to apply the same proce- the actual pointings may differ, unnecessary gaps between fields
dures, we also mention that, due to the large current area coverage (which may not exist in the actual observations) are introduced,
and consequent large number of sources, when it is necessary to as the authors point out. Our partition, described in Section 2.4,
perform the searches on small contiguous portions of the sky (e.g. and following Bianchi et al. (2007b), avoids such problems and
1◦ strips), it may happen that the coordinates of a GALEX UV enables better and more homogeneous photometry quality, as well
source fall in one strip (close to the latitude limit), and the SDSS as easy and precise calculations of the unique-area coverage for our
match falls in the next latitude range, due to a small difference in GALEX surveys, and of overlap area with any other survey, consis-
coordinates. Such cases are included in the final catalogues in the tent with the actual source catalogue rather than with nominal field
latitude range which is appropriate according to the position of the centres. This is preferable for our analysis. Such differences may
GALEX source. Care should also be taken when using the Galactic cause samples selected from different matched catalogues, to differ
coordinates as returned by the GALEX data base, as those coordi- slightly.
nates differ by 0.5 arcsec from those derived by converting the
GALEX RA, Dec. coordinates to Galactic coordinates using stan-
dard astronomical packages (e.g. WCSTools from the Smithsonian
Astrophysical Observatory). These differences probably arise be-
cause the GALEX data base stores Galactic coordinates using real 2.4 Area coverage
data types, while using double precision data types to store RA and
In order to derive the density of sources extracted from our cata-
Dec. values. Therefore, sorting GALEX search results based on the
logues, we computed the total areas of unique coverage, taking into
Galactic coordinates returned by the data base may give different
account overlap between fields, in the GALEX GR5, and then the
results than using WCSTools to derive those coordinates from the
area overlapping with SDSS DR7, following the method of Bianchi
GALEX RA and Dec.
et al. (2007b). Since we restricted the catalogues to sources within
Another contingency that must be tested for, in the above proce-
the central 1◦ of the GALEX field, for each field we considered an
dures, is whether a given SDSS source matches multiple GALEX
effective radius of 0.◦ 5. Our code scans the entire sky and calculates
sources. This can occur if an SDSS source lies somewhere in be-
the unique area covered by the GALEX fields, and the portion of this
tween two GALEX sources which were deemed unique. In this case,
area covered also by the SDSS. First, the whole sky is divided in
the match query returns a match of each of the GALEX sources with
small, approximately square, tassels, along Galactic longitude and
the same SDSS object. We retain the match with the GALEX source
latitude. We used steps in l,b of 0.05/0.◦ 1 for MIS/AIS, respectively.
which is the closest to the SDSS source, and eliminate the other.
Then, we find the distance between the centre of each tassel in our
This occurs rarely.
whole-sky grid and the centre of each GALEX field, and sum the
areas of the tassels that are within half degree of a field centre,
avoiding to count the same tassel twice, which eliminates overlap
2.3.3 Spurious matches
between fields.
The probability of spurious matches was estimated as follows. We The error in the estimated area depends on the step used to
randomly selected 30 per cent of the coordinates from our matched calculate the grid of tassels (tassel size), and the number of tassels.
GALEX-SDSS catalogues (in separate 5◦ bins of Galactic latitude), The area of each tassel decreases at high latitudes (the step in l,b is
and searched the SDSS data base against coordinates offset by kept constant over the whole sky), hence the statistical error due to
0.5 arcmin from those of the real sources. The spurious match the tassels along the field’s edge (i.e. tassels that fall partly inside
rate, shown in Fig. 3 (red dots), is the number of incidences and partly outside the 1◦ circle of a GALEX field) varies with the
where one or more matches were found (within a search radius location of the fields, both with latitude and with the relative location
of 3 arcsec) divided by the number of uploaded coordinates. As of the field centre to a grid step. Such errors cancel out statistically
for the multiple matches, it is higher towards the Galactic plane, for a large number of fields. For a very small area however, i.e.
where the stellar density increases and therefore also the proba- for a few fields, it is desirable to use smaller tassels. We estimate
bility of random matches within a given radius. The rate is of the the uncertainty by computing the areas several times, each time
order of a few per cent at high and intermediate latitudes. How- offsetting the positions of the GALEX field centres by about half
ever, matches that are positional coincidences are likely to have the size of a tassel, in both latitude and longitude directions. The
odd colours and the fraction may be lower in selected analysis resulting uncertainty is 1 deg2 , for our total area and the areas of
samples. each latitude bin.


C 2010 RAS
In order to calculate the area of overlap between the GALEX MIS Table 1. Sky coverage for GR5 MIS and AIS, and GR5×DR7.
and AIS surveys and SDSS DR7, we initially matched the centres
of all tassels included in our GALEX coverage (MIS and AIS) to the Latitude Area (deg2 ) E(B − V)a
SDSS DR7 footprint: the areas of tassels deemed within the SDSS range GALEX GR5 GR5+DR7 Mean 1σ
footprint were summed to obtain the area coverage of our matched
MIS
source catalogues. However, we discovered an issue in the SDSS −90 to −80 29.4 0.0 n.a. n.a.
data base DR7 footprint that prevented its use for calculating the −80 to −70 59.3 19.3 0.04 0.01
overlap areas. The output returned from the SDSS data base foot- −70 to −60 257.4 99.2 0.03 0.01
print query is currently somewhat incorrect, perhaps due to some −60 to −50 242.5 86.2 0.05 0.02
−50 to −40

SDSS fields which appear included in the DR7 footprint in the data 150.5 120.4 0.06 0.02
base (therefore counting towards area coverage in any estimate), −40 to −30 109.0 68.04 0.07 0.03
however do not have sources in the ‘photoprimary’ table or cor- −30 to −20 27.4 25.5 0.08 0.02
responding images because the observation failed or was marked −20 to −10 0.0 0.0 n.a. n.a.
−10 to 0 0.0 0.0 n.a. n.a.
as bad for some reason. We have consulted several experts of the
0 to 10 0.0 0.0 n.a. n.a.
SDSS data base, but we found no way so far to identify such fields
10 to 20 5.8 5.0 0.04 0.01
specifically. When we plotted on the sky the SDSS matched sources 20 to 30 89.1 85.5 0.04 0.01
(obtained from the SDSS ‘photoprimary catalogue’), and the sky- 30 to 40 210.3 203.2 0.04 0.02
tassels deemed by the footprint query to be within the footprint, we 40 to 50 161.3 157.1 0.03 0.02
noted some not irrelevant areas of mismatch (most notably, in the 50 to 60 175.7 173.0 0.04 0.01
latitude strip 50◦ –60◦ South we found the largest ‘false positive’ 60 to 70 58.1 58.1 0.03 0.01
area, about 17.5 deg2 in total; the discrepancies are much smaller 70 to 80 2.8 2.8 0.02 0.00
at other latitudes). We attempted to perform the footprint query on 80 to 90 0.0 0.0 n.a. n.a.
the SDSS data base using different syntaxes. Specifically, a basic Total 1578.6 1103.0 0.04 0.02
all-inclusive footprint SQL search returns several ‘false-positive’ AIS
answers (i.e. several contiguous areas that the data base considers −90 to −80 249.3 0.0 0.02 0.00
within the footprint, have no SDSS sources), hence the area based −80 to −70 714.0 99.6 0.03 0.01
on such outputs would be overestimated with respect to the source −70 to −60 1137.4 128.4 0.03 0.01
catalogue; a more detailed footprint query, confining the search to −60 to −50 1490.2 223.4 0.05 0.02
‘PRIMARY footprint area, returns both false-positives and false- −50 to −40 1752.0 306.8 0.07 0.03
negatives. We finally could not use the SDSS footprint query, and −40 to −30 1910.2 268.8 0.09 0.06
wrote a separate code to independently search, for our entire GALEX −30 to −20 1616.3 168.7 0.10 0.06
−20 to −10 1015.9 100.9 0.17 0.10
sky coverage, sub-areas with and without SDSS sources; we consid-
−10 to 0 211.3 37.6 0.41 0.24
ered the areas devoided of any SDSS sources as outside the SDSS
0 to 10 349.5 27.4 0.45 0.31
footprint.1 This procedure was complicated because (1) in order to 10 to 20 1487.0 156.3 0.12 0.14
decide whether a tassel contained SDSS sources, we had to con- 20 to 30 2002.1 454.5 0.06 0.05
sider SDSS sources of any colour, not just our hot-star catalogue 30 to 40 2115.2 917.9 0.05 0.02
(which includes the rarest stars in the sky, hence some tassels may 40 to 50 1866.2 1174.7 0.03 0.02
actually not have any hot star even though they are located within 50 to 60 1480.0 1221.2 0.03 0.01
the footprint), and (2) we had to consider much larger sky-tassels 60 to 70 1067.3 1067.3 0.02 0.01
than our original grid, to avoid false-negatives, and then iterate 70 to 80 738.0 738.0 0.02 0.01
within the ‘positive’ tassels with a progressively finer grid (down 80 to 90 233.2 233.2 0.02 0.01
Total 21434.8 7324.5 0.05 0.07
to 0.05 deg2 , about 1/16th of a GALEX field), to confine the uncer-
tainty of the area estimate. It is important for our analysis, which aE
B−V values are computed from the Schlegel et al. (1998) maps for the
compares density of sources (#/deg2 ) with model predictions, to esti- centres of the GALEX fields, then averaged. They do not reflect an average
mate areas consistent with the source catalogue used, in this case the Galactic trend, since they follow the specific distribution of fields shown in
PRIMARY catalogue from the SDSS. We used our area estimates in Fig. 1, which is rather non-uniform at low latitudes.
the following analysis. We stress, for future reference, that there was
no way to detect the inconsistency between catalogue and data base
2.5 Catalogue statistics and effect of error cuts on the sample
footprint other than by plotting the distribution of sources and foot-
print tassels on sky coordinates in various ways. This is a desirable Fig. 4 shows the UV magnitude distribution of our GALEX unique-
test on any such work. source catalogue, and the effect of progressive error cuts in NUV,
The GALEX GR5 AIS and MIS sky coverage, and the GALEX- and in both NUV and FUV. While a progressively stringent error cut
SDSS overlap area in the matched GR5×DR7 releases, restricting in one band (NUV) simply truncates the sample to a brighter mag-
the GALEX field to 1◦ diameter and eliminating overlaps between nitude limit, an additional error cut in FUV, which effectively raises
GALEX fields, are given in Table 1, for the whole coverage and for the faint limit of the sample in this band, causes a relatively higher
separate ranges of Galactic latitude. decrease of redder sources, and the histogram of sources distribu-
tion becomes double-peaked. The same effects are seen (right-hand
panel) for the matched GALEX-SDSS sources catalogue. This cata-
logue can also be separated in point like and extended sources using
1 These regions may be ’in’ the footprint for SDSS data base purposes, but the SDSS spatial information (∼1.4 arcsec resolution). Such dis-
they contribute no sources to the catalogue, hence counting them towards tinction shows that most of the faint-magnitude peak of the source
the area coverage would be inconsistent. distribution is due to extended sources, which are likely galaxies,

C 2010 RAS

Figure 4. Distribution of UV sources in NUV magnitudes, and effects of error cuts and colour cuts. Left: the whole catalogue of GALEX unique sources
(MIS), described in Section 2.2, with progressive error cuts in NUV, and in both FUV and NUV. A more stringent NUV error cut simply causes a brighter
cut-off at the faint end, as expected, while the requirement of significant detection (or good photometry) also in FUV modifies the overall distribution (‘redder’
sources being eliminated), which becomes double-peaked. The distribution of the ‘bluest’ sources (FUV−NUV <−0.13), which include hot stars (this paper)
and some QSOs (Bianchi et al. 2009a), is also shown. Right: NUV magnitude distribution of the matched UV-optical sources. Here, we also use the spatial
information from the optical data, to separate point-sources (eliminating most galaxies, but not QSOs with a high contrast between the central source and the
underlying galaxy). The point-like sources with FUV−NUV <−0.13 and FUV, NUV error 0.3 mag (solid orange line) are the hot-star candidate sample
analysed in this paper, and the green histogram are the ‘single’ hot stars. The figure, with the analysis in Section 4, shows that a potential halo component
(NUV ∼ 24–25 mag) is eliminated by our error cut at the current survey’s depth; on the other hand, including objects with larger errors would cause our colour
selection to include low-z QSOs.
as can be expected (e.g. Bianchi 2009). The most restricted sample sen to largely exclude all extragalactic objects, based on colours de-
in the right-panel histograms, the point-like matched sources with rived from canonical templates of QSOs and galaxies (e.g. Bianchi
error 0.3 mag in both FUV and NUV, and FUV−NUV <−0.13, 2009 and references therein; see also Fig. 5).
are the subject of this paper; it contains mostly hot-star candi- As shown by Fig. 5 in this work, and by fig. 7 (lower pan-
dates (Teff > approximately 18 000 K), with some contamination els) of Bianchi et al. (2007a), at AIS depth (∼21th mag) sources
by QSOs at faint magnitudes and red optical colours, discussed in with FUV−NUV <−0.13 have very little contamination by ex-
Section 3.2 (see also Bianchi et al. 2009a). tragalactic objects, but at fainter magnitudes a number of QSOs
The analysis of the hot-star sample with MW models (see later) and galaxies with extremely ‘blue’ FUV-NUV colours is found.
shows that the magnitude limit introduced by our error cut of Bianchi et al. (2009a) have examined the nature of such uncommon
0.3 mag eliminates mostly halo and thick-disc MW stars from the QSOs, and their figs 1 and 2 show the location of these objects
sample (between the green solid and dashed lines on the right-hand in the colour–colour diagrams. These QSOs have extremely ‘blue’
panel). On the other hand, including sources with larger photomet- FUV−NUV (more negative than typical QSOs at any redshift), but
ric errors would introduce significant contamination of the sample optical colours typical of QSOs and galaxies. Therefore, they over-
by non-stellar objects (see Fig. 5). lap, in UV–optical colour–colour diagrams, with the locus of hot
More statistical analysis of the catalogues, and discussion of stars with a cool companion, i.e. hot stars having very blue (i.e.
potential biases in flux-limited sample selection, is given by Bianchi negative) FUV−NUV colour but optical colours redder than what
et al. (2010). their FUV−NUV would imply for a single hot source. We discuss
the QSO contamination in Section 3.2.
We restricted the catalogue to sources with photometric errors
3 S E L E C T I O N O F H OT S TA R S
0.3 mag in both FUV and NUV. Fig. 4, previously discussed, and
The matched GALEX-SDSS sources in our catalogues are shown in Table 2 in this work show the effect of the error cuts on the sample’s
colour–colour diagrams in Fig. 5 as density plots. Model colours for statistics. We additionally use one optical band (r) to separate the
major classes of astrophysical objects are also shown to guide the hot sources in ‘single’ star candidates, having colour NUV−r <
eye in interpreting the distribution of sources. We refer to Bianchi 0.1, and ‘binary’ candidates having NUV−r > 0.1. In more detail,
et al. (2009a) and references therein for other similar figures and the NUV−r boundary between single and binary hot stars has a
description of the model colours. complex dependence on stellar T eff and radii of the binary pair,
We selected hot stellar candidates by extracting the matched which we will address in a different work. For example, a hot
point-like sources with FUV−NUV <−0.13 (in the AB magni- WD with an A-type companion, and pairs with small mass ratios,
tude system), which corresponds to stars hotter than ≈18 000 K, the
precise T eff depending on the stellar gravity, and on the model atmo-
spheres adopted (discussed later).2 This FUV−NUV limit was cho- FUV-NUV colour is basically reddening free. In fact, the higher selective
extinction Aλ /EB−V in the FUV range is approximately equalled by the
effect of the broad 2175 Å feature in the NUV range (see Fig. 2 and Bianchi
2 The measured colours also depend on extinction, however for a MW- 2009). The absorption in each UV band, however, is much higher than in
type dust, with RV = 3.1, and moderate reddening amounts, the GALEX optical bands.


C 2010 RAS

Figure 5. Colour–colour diagrams for the GALEX UV sources matched to SDSS sources, at AIS and MIS depth. Blue/black densities are point-like/extended
sources, respectively. Model colours for stars (T eff sequence, for different gravities), QSO (redshift sequence, cyan) and galaxies (age sequences, green)
are shown. The two purple stellar sequences (label ‘WD’) in the right-side panel are for log g = 9.0 and 7.0. Although gravity cannot be separated well
photometrically at high T eff s, the vast majority of hot stellar sources are clustered in between these two model sequences, and much fewer are seen along
the log g = 5 and 3 (red and yellow) sequences at hot temperatures, as expected. In the ≈2.5 mag shallower AIS survey the relative number of stars and
extragalactic objects is comparable, while at MIS depth the latter dominate. From such diagrams the hot-star samples are selected (see text), with colour cuts
shown by the green lines in the left-hand plot.
will fall in the colour selection of our ‘single’ hot-star sample. binary companion or QSOs, which have redder UV-optical colours,
The adopted value is a useful cut to eliminate most extragalactic will be detectable also in the SDSS imaging. This explains why the
sources contaminating our FUV−NUV <−0.13 sample, as they density of MIS ‘single’ hot matched sources (light-green) drops at
have NUV−r > 0.1, however it is obviously an oversimplification faint magnitudes earlier than the dashed-green histogram, the dif-
for a detailed study of binaries. In sum, our so-called (for simplicity) ference between the two being larger than what can be ascribed to
‘binary’ sample includes stellar pairs with a hot WD and a cooler QSO contamination (Section 3.2).
star, and some QSOs; the ‘single’ sample includes all single stars In order to estimate the incompleteness of faint hot-WD counts
and some binaries. We impose no error cut on the r magnitudes, in in our matched sample, we searched for GALEX MIS sources with
order to not limit the sample of the hottest WDs, which are faint FUV−NUV <−0.13 that are within the SDSS DR7 footprint, but
at optical wavelengths. If the restriction of error 0.3 mag were do not have SDSS counterpart. We found ≈1500 such sources, hav-
imposed to the rband, the density of objects in the MIS sample ing no optical match in our catalogue with a match radius of 3 arcsec
would be reduced by about 30 per cent (GALEX is ‘deeper’ than (Section 2.3). We performed a number of tests to verify if these are
SDSS for very hot stars of low luminosity, see Section 3.1); the loss real sources. We matched them against the SDSS catalogue, extend-
would be much smaller for the brighter AIS sample. ing the match radius to 6 arcsec, and found 399 additional matches,
The density of hot-star candidates is shown in Fig. 6 and will 138 of them are classified as ‘point-like’ sources at the SDSS res-
be discussed later. The magnitude distribution of the total hot-star olution. These mostly appear to be actual sources in the images,
candidate sample (FUV−NUV <−0.13) is shown with shadowed although we note that for faint sources, the SDSS classification
histograms in Fig. 8 and beyond, and the ‘single’ hot-star candi- of ‘extended’ and ‘point like’ is not always reliable, as shown by
dates (FUV−NUV <−0.13 and NUV−r < 0.1) with solid-colour Bianchi et al. (2009a). It is expected that a small number of sources
histograms. may have optical coordinates differing by more than 3 arcsec from
the UV position; sometimes this is due to a nearby source not fully
resolved. For the remaining 1100 UV sources with no SDSS match
3.1 Hot stars not detected by SDSS out to 6 arcsec, visual inspection of random subsamples revealed
While the SDSS depth of 22.3/23.3/23.1/22.3/20.8 AB magnitude in some to be part of extended cirrus emission, many seem likely
u,g,r,i,z provides a fairly complete match to the AIS UV source cata- sources, a few cases are parts of a shredded galaxy. The ‘kron ra-
logue, the hottest, smallest stars detected in the MIS may fall below dius’ from the GALEX pipeline gives an indication of the spatial
the SDSS limit. This can be guessed by comparing in Figs 8, 10, 12 extent of the source: 189 sources have kron radius larger than 3.5,
the green-dashed histogram (matched GALEX-SDSS sources with hence are probably not stellar sources, but visual inspection sug-
FUV−NUV <−0.13) with the light-green filled-colour histogram gests about 75 per cent of them to be real sources. Even visual
(matched sources with FUV−NUV<−0.13 and NUV−r < 0.1, i.e. inspection, however, is not always conclusive, for faint sources or
‘single’ hot stars). The hottest single WDs will be faintest at optical complicated fields, and better resolution or deeper exposures would
wavelengths (e.g. a star with log g = 9 and T eff = 50 000/100 000 K be needed for a final sample. In sum, a very large fraction of the
would have FUV−r = −2.33/−2.61 in AB magnitude, according 1500 objects unmatched within 3 arcsec are actual sources, and an
to our TLUSTY model grids). Therefore, hot WDs still detectable undetermined fraction may be actual hot stars.
in UV at the depth of our MIS sample (see Fig. 4) may be below In spite of the large uncertainties, we added all these objects to
the SDSS detection limit, while hot WDs with an optically brighter the ‘single’ hot matched sources catalogue, and show the total as

C 2010 RAS

Table 2. Statistical characteristics of the Catalogues and effects of error cuts.
No. of GALEX
sources
Latitude No. of GALEX FUV−NUV No. of matched sources No. of point-like matched sources
range sources <−0.13 errNUV 0.5 errNUV,FUV errNUV,FUV errNUV 0.5 errNUV,FUV errNUV,FUV with FUV−NUV <−0.13
errNUV errFUV,NUV errFUV,NUV Rank0 Rank1 0.5 0.3 0.5 0.3 errNUV,FUV NUV−r <0.1
0.5 0.5 0.5 0.3 Rank0 Rank1 Rank0 Rank1 Rank0 Rank1 Rank0 Rank1 Rank0 Rank1 0.5 0.3 0.3
MIS
−90 to −85 66 929 16 822 2444 622 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
−85 to −80 177 553 49 043 6304 1540 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
−80 to −75 112 394 32 092 4021 971 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
394 332 109 647 12 984 3070 87 456 2708 33 147 1300 15 510 794 17 782 638 2584 211 1158 127 321 91 42

−75 to −70
−70 to −65 798 670 223 254 25 020 6151 156 597 3994 63 005 2099 34 584 1450 33 311 1021 5554 345 2759 224 671 201 78
−65 to −60 1409 943 400 165 44 106 11 391 360 403 10 991 149 854 5826 87 609 4079 71 801 2433 12 151 866 6395 565 1386 426 155

−60 to −55 1244 707 328 909 40 804 10 773 228 705 7106 86 721 3448 49 148 2388 49 844 1659 7401 501 4061 334 931 296 125
−55 to −50 632 803 143 121 21 932 6162 164 257 5294 49 621 2093 25 136 1291 41 518 1388 4477 308 2374 204 711 257 112
−50 to −45 609 861 137 541 21 637 5756 304 497 12 522 92 372 5102 46 024 3217 77 719 3427 8067 812 4342 506 1230 472 244
C 2010 RAS
−45 to −40 428 767 89 077 13 169 3620 207 041 10 578 57 421 3948 29 654 2414 61 074 3127 5178 651 2869 413 745 311 157
−40 to −35 538 838 103 347 15 242 4283 211 133 9835 50 373 3174 27 961 1978 83 015 3652 6030 580 3634 349 843 396 228
−35 to −30 323 258 59 584 8370 2715 124 660 8736 27 139 2794 16 141 1929 58 244 3735 3658 607 2374 419 558 303 174
−30 to −25 214 275 31 837 5398 1749 133 209 8714 23 127 2116 11 900 1213 75 788 4697 4076 529 2404 301 656 365 240
−25 to −20 48 433 6190 1035 357 34 035 2638 4825 494 2441 272 22 202 1652 1109 149 646 94 182 93 65
15 to 20 43 887 8626 997 304 26 939 1383 6195 433 3488 279 14 452 716 1297 120 812 81 132 77 51
20 to 25 226 867 47 323 6099 1672 144 508 7255 35 983 2484 18 325 1450 64 767 3040 5810 575 3372 358 635 308 218
25 to 30 403 868 91 523 10 958 2782 281 365 12 461 79 869 4695 39 329 2745 101 054 4591 10 017 974 5594 565 1075 520 346
30 to 35 711 824 152 925 21 310 5682 474 446 18 387 132 915 6920 66 709 4116 155 923 6218 14 995 1263 8182 727 2079 862 488
35 to 40 937 945 203 696 27 963 7589 621 803 24 188 173 805 9065 91 210 5604 200 400 8114 18 511 1602 10 080 971 2720 1094 601
40 to 45 641 846 152 205 21 044 5661 430 026 15 963 133 123 6575 68 076 4018 119 976 4803 12 413 1116 6306 654 1892 744 413
45 to 50 707 063 162 317 21 478 5659 460 636 15 730 139 985 6496 74 669 4238 127 343 4598 12 844 1041 6839 658 1885 639 325
50 to 55 731 762 170 967 20 334 5143 492 531 19 798 149 059 7947 80 742 4944 135 058 5360 13 351 1257 7468 769 1700 641 371
55 to 60 690 921 170 036 20 115 4864 461 420 18 063 146 692 7491 76 642 4663 119 973 4788 12 470 1150 6521 704 1657 528 293
60 to 65 377 644 97 916 11 366 2899 255 057 9001 86 150 4034 45 134 2581 62 828 2388 7169 649 3695 415 912 309 142
65 to 70 99 569 25 651 3073 716 68 009 2407 22 683 1058 10 337 609 13 951 546 1857 151 832 83 238 75 44
70 to 75 23 953 6405 678 136 16 778 537 5757 238 2588 137 3126 137 446 46 198 26 62 20 12
−90 to 90 12 597 912 3020 219 387 881 102 267 5745 511 228 289 1749 821 89 830 923 357 56 409 1711 149 72 728 171 465 15 503 92 915 9547 23 221 9028 4924
AIS
−85 to −90 186 959 29 800 6088 1187 2118 66 488 22 137 9 686 24 60 2 21 1 5 0 0
−80 to −85 639 403 107 385 20 863 4235 15 996 448 3458 144 1075 65 4635 152 399 33 177 17 55 28 24
−75 to −80 946 695 157 960 31 268 6157 42 060 1425 9492 431 2998 179 11 306 416 922 79 448 32 134 60 41
−70 to −75 1379 695 220 313 43 504 8513 102 750 4001 23 114 1254 7740 600 28 113 1238 2273 223 1058 116 309 138 97
Hot white dwarfs: UV catalogues
2779

Table 2 – continued
2780
No. of GALEX
sources
Latitude No. of GALEX FUV−NUV No. of matched sources No. of point-like matched sources
range sources <−0.13 errNUV 0.5 errNUV,FUV errNUV,FUV errNUV 0.5 errNUV,FUV errNUV,FUV with FUV−NUV <−0.13
errNUV errFUV,NUV errFUV,NUV Rank0 Rank1 0.5 0.3 0.5 0.3 errNUV,FUV NUV−r <0.1
0.5 0.5 0.5 0.3 Rank0 Rank1 Rank0 Rank1 Rank0 Rank1 Rank0 Rank1 Rank0 Rank1 0.5 0.3 0.3
−65 to −70 1779 025 267 234 53 565 10 329 104 406 3670 24 373 1188 8059 514 28 907 1114 2464 221 1070 102 350 141 99
−60 to −65 1823 298 273 723 54 203 10 650 151 592 6068 34 985 1981 11 556 944 44 240 1714 3466 342 1584 181 489 233 176
L. Bianchi et al.
−55 to −60 2202 425 309 450 62 380 12 610 149 424 6616 29 732 1895 9752 819 50 822 2148 3580 355 1700 168 537 294 214
−50 to −55 2221 574 260 072 55 813 11 764 137 391 6620 21 591 1640 7251 732 56 334 2497 3270 348 1589 154 553 331 234
−45 to −50 2526 007 271 544 59 711 12 805 195 821 10 880 29 085 2665 10 396 1253 81 131 4309 4433 562 2221 277 801 473 338
−40 to −45 2644 017 259 939 59 142 14 759 157 644 12 248 20 715 3038 7274 1400 79 249 5390 3708 680 1808 341 758 441 302
−35 to −40 2824 496 243 813 55 332 13 542 243 682 15 446 28 209 3371 9899 1562 133 778 7605 5717 831 2900 417 1208 725 501
−30 to −35 3022 905 251 126 53 008 14 535 171 132 14 681 14 919 2977 5579 1395 112 514 7969 4148 857 2175 454 938 570 394
−25 to −30 2725 467 171 367 39 306 12 449 219 113 20 491 15 759 3593 6299 1665 159 944 12 263 5423 1154 2978 571 1330 829 611
−20 to −25 2565 319 121 358 29 997 11 039 158 390 19 012 7832 2255 3563 1116 133 521 13 447 4060 922 2233 504 1037 650 476
−15 to −20 2294 738 90 915 21 775 9748 104 999 13 902 4288 1362 2245 739 94 135 10 745 2685 665 1536 384 660 406 291
−10 to −15 1474 126 54 091 11 682 6026 107 441 14 988 4086 1142 2582 762 102 078 12 821 3235 736 2048 511 607 350 231
−5 to −10 468 735 20 256 3196 1719 15 893 2690 834 224 547 147 15 334 2358 707 158 441 102 77 39 22
0 to −5 81 792 5544 338 126 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
5 to 0 112 359 9822 673 324 4730 816 364 140 263 120 4577 774 340 133 241 113 37 18 2
10 to 5 896 414 41 563 5055 2532 67 009 10 668 4191 1170 2914 871 64 784 9577 3683 921 2520 682 285 169 86
15 to 10 2209 741 93 883 17 027 7825 133 415 20 698 6792 1979 4160 1267 124 034 17 107 5085 1207 3230 806 758 458 301
20 to 15 2785 094 137 819 28 011 10 512 248 149 22 946 22 199 3385 9760 1708 192 683 16 088 8372 1376 4610 747 1398 892 688
25 to 20 3078 141 180 485 38 516 11 924 482 621 34 646 50 125 5965 19 504 2777 328 234 21 367 14 284 2013 7510 1020 2539 1604 1230
30 to 25 3074 765 199 400 45 162 11 949 565 885 34 039 59 037 6033 20 866 2622 336 893 18 763 14 374 1854 7217 885 2777 1683 1308
35 to 30 3068 226 221 555 50 831 11 949 770 294 40 366 91 294 7455 30 887 3131 420 587 20 138 18 413 1879 9082 840 3615 2135 1714
40 to 35 2827 184 215 451 50 937 10 501 911 569 43 682 109 497 8095 35 498 3333 459 207 20 279 19 905 1846 9697 822 3899 2278 1774
45 to 40 2867 019 241 576 57 082 10 742 1149 694 53 336 144 002 10 172 45 305 4158 490 863 21 484 21 555 2157 10 284 980 4171 2345 1858
50 to 45 2589 087 237 387 55 351 9692 1187 271 53 682 158 334 10 737 47 012 4333 455 468 19 957 21 918 2243 10 131 1013 4017 2139 1679
55 to 50 2183 650 219 294 49 511 8631 1135 384 51 155 165 181 11 031 48 979 4336 414 632 18 158 21 966 2136 9994 973 3845 1979 1556
60 to 55 2031 169 227 445 49 608 8489 1207 517 55 161 187 254 12 923 56 759 5317 390 611 17 457 21 534 2264 9850 1071 3576 1838 1410

65 to 60 1617 230 193 927 41 370 7220 1058 375 50 788 173 459 12 127 52 661 5039 332 166 15 743 19 066 2203 8630 1048 3063 1531 1197
70 to 65 1419 335 168 945 36 062 6064 927 973 42 934 151 376 10 316 44 148 4136 278 120 12 937 16 547 1808 7447 822 2578 1325 1022
75 to 70 1204 662 135 513 29 867 4605 792 486 37 015 121 558 8221 34 147 3190 227 147 10 912 13 190 1494 5749 675 2127 1033 792
80 to 75 860 091 88 480 19 796 2865 563 957 27 906 78 969 5847 21 026 2238 155 680 8158 8642 1069 3668 475 1370 645 512
85 to 80 471 126 59 667 12 490 1978 310 436 15 381 53 200 3836 14 757 1510 87 185 4318 5673 638 2445 290 842 401 307
90 to 85 164 322 20 454 4760 795 107 786 5417 18 269 1330 5094 523 27 164 1384 1682 212 754 102 267 138 119
−90 to 90 65266 291 5808 556 1253 280 290 790 13 704 403 753 888 1868 061 149 944 590 692 64 510 5926 762 340 811 286 779 35 621 13 9046 17 696 51 012 28 319 21 606
Note: The columns ‘rank0’ give the number of UV sources with one SDSS counterpart (within the match radius of 3 arcsec), and ‘rank1’ of those with multiple optical matches. UV sources with multiple matches
may have composite UV colours, therefore are excluded from the analysis sample; their exclusion is accounted for by correcting the density of sources by (rank1/(rank0+rank1)). The fraction from Columns 16–17
(from which our analysis sample is extracted) is plotted in Fig. 3. Numbers in Columns 18, 19 and 20 include only ‘rank0’ sources.

C 2010 RAS
and two as galaxies. Only one is classified as QSO by the SDSS
spectroscopic pipeline, but it is actually a hot WD, as shown by
Bianchi et al. (2009a). This implies a purity of almost 100 per cent
for the ‘single’ hot-star candidates, down to the magnitude limit
of the SDSS spectroscopy at least. Again, we stress, however, that
the spectroscopic subset is serendipitous for our purpose, but not
necessarily unbiased.
In the AIS sample, out of 28 319 total hot-star candidates (21 606
of which with NUV−r < 0.1), 4448 (3737) have spectra, classified

as 4075 (3721) stars, 309 (nine) QSOs, and 59 (seven) galaxies;
corresponding to 91.6 per cent (99.6 per cent for ‘single’ hot-star
candidates) purity. The higher content of stellar sources in the AIS
sample, even for the ‘binaries’, is due to the brighter magnitude
limit.
A crude, more direct indication that there is contamination by
extragalactic sources in the ‘binary’ hot-star locus is that the fraction
of ‘binaries’ (NUV−r > 0.1) among the hot sources is about 25 per
cent for the AIS and 45 per cent for the MIS. We note that the AIS
value of 25 per cent is not reflecting the fraction of actual binaries,
because the pairs whose components have similar T eff are included
in the ‘single’ colour cut, and on the other hand, AIS ‘binaries’ may
Figure 6. Density of ‘single’ hot stars at different Galactic latitudes (all also contain QSOs.
magnitudes combined, green for MIS, blue for AIS). The stellar counts are In view of the contamination by QSOs in the ‘binary’ hot-star
not corrected for MW extinction, which is more severe at low latitudes, and
sample, in the following analysis with MW models we will consider
causes the sharp drop seen on the MW plane. The bright-green histogram is
the counts of ‘single’ hot-star candidates, and assume a canonical
the GALEX-SDSS matched MIS single hot-star candidates, the dark-green
histogram includes also hot UV sources with no optical match, within the binary fraction of 30 per cent in the models.
SDSS footprint (see text). The ranges 10◦ –20◦ N and 70◦ –80◦ N have only 6
and 3 deg2 MIS coverage in the matched footprint, therefore the statistical 3.3 Characteristics of the Hot-Star Candidate Catalogue
significance is much less than in other bins.
The number of hot sources selected (point like, with error 0.3 mag
filled-dark-green histograms in Figs 6, 8, 10, 12. They must be in FUV and NUV, and FUV−NUV <−0.13; see Section3), and of
considered a very generous upper limit, and only a reminder that the subset with NUV−r < 0.1 (‘single’), is given in the last two
the light-green histograms suffer from incompleteness for hot stars columns of Table 2. Stellar counts as a function of magnitude will
with NUV fainter than ∼22 mag. be analysed in the following sections. Fig. 6 shows the density of
‘single’ hot-star candidates (all magnitudes combined) as a function
of Galactic latitude. The density increases by over a factor of 7, from
3.2 Purity and completeness of the sample
the poles towards the Galactic plane. Owing to the MW extinction,
In order to estimate the probability that the photometrically selected the counts are a lower limit at all latitudes, but especially near the
hot-star candidates actually are hot stars, and the possible contam- Galactic disc. Therefore, the variation with latitude shown by this
ination of the sample by other types of objects, we examined the figure is less than the actual one. For the MIS, the density of hot stars
subsamples of our catalogues for which SDSS spectra exist. Out in the matched sample (light-green histogram) is incomplete due to
of 9028 MIS-matched sources with FUV−NUV <−0.13 and FUV, the SDSS limit (Section 3.1). The dark-green histogram shows the
NUV error 0.3 mag, 810 have SDSS spectra, 104 of which are MIS ‘single’ hot-star candidates including the hot GALEX sources
sources with NUV−r > 0.1, i.e. in the ‘binaries’ locus. Of these without optical match. Sources with NUV−r > 0.1 are not shown
‘binary’ candidates with existing spectra, 58 are spectroscopically due to the QSO contamination (Section 4.3); adding the stellar
classified by the SDSS pipeline as stars, four as galaxies, 42 as binaries in this colour range would increase the number density but
QSOs. Therefore, almost half of the MIS sources with FUV−NUV not change the trend.
<−0.13 and NUV−r > 0.1 could be QSOs (as found also by In the next section, we analyse the density of hot stellar candidates
Bianchi et al. 2007a), and half could be hot stars with a cool com- as a function of magnitude; the samples are divided in strips of 10◦
panion. In more detail, QSO numbers increase at fainter magni- Galactic latitude, in order to examine the structure of this MW
tudes, and have FUV-NUV colours closer to our limit (−0.13) than stellar component.
the hot WD (Bianchi et al. 2009a). Because stars and QSO counts
vary with magnitude in different ways, and especially because this
4 A N A LY S I S , C O M PA R I S O N W I T H M I L K Y
statistic can be highly biased by the SDSS selection of spectroscopic
WAY M O D E L S
targets, which is of course unrelated to our present selection, and
because the spectroscopic survey does not reach the depth of the From the point of view of stellar evolution, the sample presented in
MIS photometric survey, we refrain from assuming a correction for this paper includes essentially two kinds of evolved stellar objects:
the fraction of extragalactic objects in the binaries sample based on the post-AGB stars that have just expelled their envelopes and are
current data. crossing the HR diagram towards higher T eff s at constant luminosity,
A much higher purity is found for the ‘single’ hot-star candi- on their way to become PNe nuclei, and the hot WDs which are
dates (FUV−NUV <−0.13 and NUV−r<0.1). Out of 4924 in the fading both in T eff and in luminosity. Post-AGB stars are very
MIS sample, 706 have spectra, 703 of which are classified as stars, elusive because they evolve very fast, on time-scales of the order


C 2010 RAS
of 103 –105 yr (Vassiliadis & Wood 1994). Hot WDs evolve with
longer time scales, but at significantly fainter luminosities. Both are
elusive at all wavelengths except the UV. Our hot-star census based
on the UV sky surveys provides the first opportunity to examine a
comprehensive, unbiased sample of such objects. Our hot post-AGB
and WD candidate sample extends over a significant sky coverage,
and clearly presents a disc-like distribution, concentrated towards
the Galactic disc (Fig. 6). Such sample will be analysed in the
context of a model for the MW geometry below.

4.1 Constraining Milky Way models
In this section we analyse the number counts and sky distribution
of our hot-star candidates, using the TRILEGAL stellar population
synthesis code (Girardi et al. 2005). TRILEGAL creates mock cat-
alogues of stars belonging to the MW, and then distributes them
along the line-of-sight. It extracts the simulated stars from extended
libraries of evolutionary tracks and synthetic stellar spectra, assum-
ing reasonable prescriptions for the distributions of ages, masses and
metallicities of the MW stellar components: thin and thick discs,
halo and bulge. We note here that the bulge component does not
contribute almost any source to our hot-star sample, and is therefore
not shown in the plots of our model results. Figure 7. Comparison of model magnitudes (NUV) and colour (FUV-NUV)
The thin-disc density, ρ thin , decreases exponentially with the of stars with log(g) = 7.0 and 9.0, of varying T eff . Model magnitudes are
galactocentric radius, R, and as a squared hyperbolic secant function constructed from our grid of pure-H TLUSTY NLTE model spectra and from
in the direction perpendicular to the plane, z: Koester (2008) LTE models. The latter include convection, which TLUSTY
(ver198) does not. Our colour cut of FUV−NUV <−0.13 corresponds to
ρdisc = C exp(−R/hR ) sech2 (2 z/hz ) . (1) T eff ≈ 18 800/17 200 K (log (g) = 9.0/7.0) in the TLUSTY models, and
to T eff = 18 000 K in solar Kurucz-model colours with log (g) = 5.0 (not
The scalelength is set to hR = 2800 pc, whereas the scaleheight
shown). While TLUSTY and Koester model colours with log (g) = 7.0 agree
increases with the population age t as quite well for T eff up to 26 000 K, the two grids are discrepant by ∼0.15 mag
hz (t) = 95 (1 + t/(4.4 Gyr))1.67 . (2) for log (g) = 9.0, where the Lyα wings are broad enough to enter the GALEX
FUV band when convection is taken into account. The situation reverses at
Such an increase is necessary to describe the observed increase of hotter T eff s. A colour cut at FUV−NUV <−0.5 benefits from less scatter
the velocity dispersion W with age (see e.g. Holmberg, Nordström among model grids, which we could take as a measure of less uncertainty,
& Andersen 2009). however the resulting sample (the very hottest stars) would be drastically
The thick disc follows the same functional form but with both reduced, and so the statistical significance.
scale parameters fixed as hR = 2800 pc and hz = 800 pc. The con-
stants C, for the thin and thick discs, are adjusted so that the surface modifications will be detailed in a subsequent paper (Zabot et al.,
thin-disc density is equal to 59 M pc−2 , whereas the local thick in preparation) . Post-AGB stars, PNe nuclei and WDs of types
disc density is 0.0015 M pc−3 . The halo is modelled as an oblate DA and DB have been included using the evolutionary tracks from
spheroid following a deprojected r1/4 law, with a local density of Vassiliadis & Wood (1994) and Althaus, Serenelli & Benvenuto
0.00015 M pc−3 . Finally, the Sun is located at R = 8700 pc, z = (2001) and Althaus et al. (2005), together with the Koester (2008)
24.2 pc. synthetic spectra. In the present version we have considered only
The default parameters for the geometry of the MW components the DAs, since they are the dominant type among hot WDs (Hansen
in TRILEGAL are calibrated to reproduce the star counts in a lo- & Liebert 2003). The fraction of DBs is known to increase sig-
cal sample extracted from the Hipparcos catalogue, and in several nificantly for WDs cooler than 12 000 K, thanks to convection in
multi-band catalogues including the shallow, all-sky 2MASS, and a the He layers (Bergeron, Ruiz & Leggett 1997; Hansen & Liebert
few deep surveys such as EIS-deep and DMS (Girardi et al. 2005). 2003). Such cool WDs, however, are not included in our selection
For regions out of the Galactic plane, i.e. for |b| > 10◦ , errors in the of sources. The TRILEGAL code uses the Koester (2008) grids to
star counts predicted by TRILEGAL are typically of about 10 to 20 assign magnitudes to the theoretical stars. In Fig. 7 we show a com-
per cent down to K ∼ 14 (Girardi et al., in preparation) . parison with magnitudes calculated from TLUSTY models, and in
The reddening is taken into account in the MW models as follows. Fig. 8 we compare stellar counts obtained by selecting model stars
Along a given line-of-sight, the value of E(B − V) at infinity is by T eff and by synthetic colours, in order to illustrate the sometimes
taken from the Schlegel, Finkbeiner & Davis (1998) maps. The neglected effect of the model-atmosphere uncertainties.
total extinction is then distributed along the line-of-sight as if it Following our preliminary findings of discrepancies between pre-
were generated by an exponential dust layer with scaleheight hdust z ; dictions and hot-star counts based on earlier data releases (Bianchi
its default value is hdust
z = 110 pc. In this way, the closest simulated et al. 2009b), the TRILEGAL code has been modified to also allow
stars are unreddened, whereas those at distances of a few hundred the choice of IMFR independent from the prescription adopted for
parsec are reddened by the full amount predicted by the Schlegel the previous evolutionary phases. In this way, the mass distribution
et al. (1998) maps. of the WDs does not need to follow the constraints imposed by
We use a new version of TRILEGAL, which has been modified the previous TP-AGB tracks, which come from Marigo & Girardi
for the purposes of this analysis in several ways; most of these (2007, hereafter MG07).

C 2010 RAS

Figure 8. Modelling WD counts: the effect of the IFMR. The density of hot-star candidates at two sample latitudes (10◦ -wide strips) are shown, with model
predictions (halo, thin- and thick-disc components, and total) The models in the top-left panel are computed with the TRILEGAL default IFMR (MG07) and
the others with W2000 IFMR. The thick disc and halo counts become significant at magnitudes fainter than ∼19, and ∼20, for these two IFMR, respectively.
Models with W2000 IFMR match better the data down to magnitudes ∼20–21, below which the AIS becomes incomplete. In the top and bottom-right
panels the model counts are obtained selecting stars with T eff > 18 000 K in the model calculations, and in the bottom-left panel selecting stars by colour cut
(FUV−NUV < −0.13) with TRILEGAL’s transformation from T eff to magnitudes, from the same calculations. The comparison illustrates the uncertainties
introduced by the transformation of the isochrones into magnitudes via model atmospheres. The filled-green histograms are GALEX MIS hot ‘single’ sources
with SDSS match (light green), and including sources with no SDSS detection (dark green).
Model predictions for sample Galactic latitudes are shown in Therefore, the mismatch between model predictions computed
Fig. 8, computed with default TRILEGAL parameters for the with TRILEGAL’s default parameters and observed stellar counts
MW geometry and the default IFMR from MG07, as well as the (Fig. 8, top-left panel), also previously noted by Bianchi et al.
Weidemann (2000) (hereafter W2000) IFMR. We show the thin- (2009b), prompted us to explore different IFMRs in TRILEGAL.
disc, thick-disc, and halo stellar components, as well as the total Although other factors, such as the WD birthrate, or the assumed
predicted counts. The IFMR from MG07 largely overpredicts faint MW extinction model, or a different geometry may also affect
star counts, while the W2000 IFMR produces an overall better match model predictions, the IFMR is a most critical aspect for modelling
with the observed counts, at all Galactic latitudes (see also Fig. 10), hot-WD counts, and was therefore explored in this first analysis.
and was therefore adopted in all our calculations that follow. Given The geometry is better constrained by low-mass stars (e.g. Girardi
the relevance of the IFMR in the context of stellar evolution, we ex- et al. 2005), which are more numerous, and their counts based on
plain in the next section how it affects the hot-star count predictions, optical-IR bands are less affected by extinction than a UV-based
which can be tested by our data. catalogue. The right-hand panels of Fig. 8 show that the W2000
IFMR produces a better match of the model calculations with the
observed stellar counts than the MG07 IFMR, and this trend is seen
at all Galactic latitudes.
4.2 The initial–final mass relation These two IFMRs are illustrated in Fig. 9, together with additional
While the MW geometry has been derived by previous studies of IFMRs recently derived by Ferrario et al. (2005) and Kalirai et al.
low-mass stars, the IMFR is one of the least constrained factors (2008) from extensive high signal-to-noise ratio spectroscopy of
in our understanding of stellar evolution, in spite of its importance WDs in Galactic open clusters. Despite the uncertainties related
for determining the yield of chemical elements, and ultimately the to the age-dating of the clusters, their field contamination, and the
Galaxy chemical evolution. The reasons for this uncertainty in- possible dependences on metallicity, they are somewhat similar to
clude insufficient statistics of post-AGB stars. The final WD mass the W2000 semi-empirical IFMR. The MG07 IFMR, instead, is
is confined to a very small range (∼0.5–1.0 M ) compared to the derived from theoretical evolutionary tracks and, as shown in the
∼0.8–8 M range of their main-sequence initial masses, and the figure, yields significantly higher final masses for all M ini < 2 M .
post-AGB evolutionary time-scale varies steeply within the small In order to understand why the model-predicted hot-WD counts
WD mass range. The luminosity remains constant over a large range depend on the IFMR, let us first consider the distribution of WD
of T eff (up to >100 000 K), and again depends on the remnant mass, masses in present-day surveys of the solar neighbourhood. Em-
making it much harder to establish their distance and absolute lu- pirical determinations have always concluded that the WD mass
minosity than for main-sequence stars. distribution presents a strong peak at low masses, with a maximum


C 2010 RAS

Figure 9. Left: some of the IFMRs implemented in TRILEGAL, and tested in this work. The MG07 IFMR depends on metallicity, so that different age–
metallicity relations (AMR) will sample it in different ways; the curve shown here is for the Rocha-Pinto et al. (2000) AMR, which is used to model the MW
thin disc in TRILEGAL. Other IFMRs are the semi-empirical one from W2000, and the purely empirical ones by Ferrario et al. (2005) and Kalirai et al. (2008).
Right: the distribution of initial and final WD masses (top and bottom panels, respectively) derived from TRILEGAL models using the MG07 (red lines) and
W2000 (black lines) IFMR. The different lines illustrate the contributions from the different MW components, for one sample latitude, down to NUV = 21.5.
close to 0.55 or 0.65 M and a well-defined tail of more mas- The WD lifetime τ WD18k , defined as the total time after the AGB
sive WDs, as well as a faint low-mass tail believed to be either in which the star will be hotter than 18 000 K, is longer for WDs
observational errors or the result of binary evolution (e.g. Berg- of higher mass, because, although they evolve much faster in the
eron, Saffer & Liebert 1992; Bragaglia, Renzini & Bergeron 1995; constant-luminosity phase, they reach higher T eff than lower-mass
Madej, Należyty & Althaus 2004; Liebert, Bergeron & Holberg WDs, and spend a longer time in their WD cooling track before
2005; Kepler et al. 2007; Hu, Wu & Wu 2007; Holberg et al. 2008). fading below the T eff = 18 000 K limit. Models using different
The origin of this peaked mass distribution can be readily derived IFMRs simply associate the final masses, hence also these τ WD18k
from basic population synthesis theory (see Marigo 2001; Ferrario lifetimes, to different initial masses, without affecting the WD pro-
et al. 2005). Given a volume-limited sample containing stellar pop- duction rate. The end result in terms of mass distribution N(M ini )
ulations of all ages between T = 0 and T = 10 Gyr, the distribution (see equation 4) is that smaller WD masses will weigh the dis-
of WD initial masses will be given by tribution towards shorter τ WD , hence decreasing the numbers of
N(Mini ) ∝ φM (Mini ) ψ[T − τH (Mini )] τWD (Mini ) , (3) observed WDs with T eff > 18 000 K. This effect explains the higher
number of hot WDs predicted by the MG07 IFMR wit respect to
where φ M is the IMF, ψ[T − τ H ] is the star formation rate at the W2000 IFMR (Fig. 8). In fact, as can be inferred from Fig. 9,
the time of stellar birth T − τ H , and τ H and τ WD are the main the W2000 IFMR gives significantly lower WD final masses (hence
sequence and WD lifetimes, respectively. The first two terms in the shorter τ WD18k ) for the interval of initial masses between 0.6 and
right-hand side of this equation represent the production rate (or, ∼2 M . Because of the IFM being skewed towards lower masses,
alternatively, the ‘death rate’) of evolved stars with different masses, these objects represent numerically the majority within the whole
in number of stars per unit time. Considering a Salpeter (1955) IMF, mass range.
−2.35
φM ∝ Mini , and a reasonably constant star formation rate ψ over Fig. 9 (right-hand panel) shows the distribution of initial and final
the galaxy history; the equation above indicates a marked peak of WD masses derived from TRILEGAL for a sample latitude, using
the production rate at the minimum initial mass for the formation either the MG07 or W2000 IFMR. It shows that most of the hot
of a WD, which is about 1 M for a galaxy age of T = 10 Gyr. For WDs observed by GALEX are expected to derive from low-mass
−2.35
higher initial masses, the production rate should fall as Mini . This stars, with initial masses below ∼2 M . Moreover, it indicates
behaviour will be shared by any galaxy model with nearly constant that the bulk of observed WDs have low masses, typically 0.55
star formation, independently of its IFMR. M WD 0.6 M in the case of the W2000 IFMR. There is no direct
Assuming a constant and monotonic IFMR, M WD (M ini ), the WD confirmation of this particular prediction, although it is in line with
mass distribution is given by WD mass determinations derived from spectroscopic surveys of

dMWD −1 more limited samples (e.g. Bergeron et al. 1992; Bragaglia et al.
N(MWD ) ∝ N(Mini ) , (4) 1995; Madej et al. 2004; Liebert et al. 2005; Kepler et al. 2007; Hu
dMini
therefore, for almost-linear and linear IFMRs like the W2000, Fer- et al. 2007; Holberg et al. 2008).
In the calculations with MG07 IFMR the distribution of predicted
rario et al. (2005), and Kalirai et al. (2008) ones, the WD produc-
WD masses is quite different from the W2000 case, with a broad
tion rate is still expected to behave like a power law peaked at the
smallest masses. This smallest mass corresponds to a WD mass of mass distribution extending from 0.55 to 0.65 M , followed by a
sort of gap, and a second peak of WD masses located slightly above
∼0.55 M (see Fig. 9).

C 2010 RAS
0.7 M . These features reflect the positive slope of this IFMR up still too sparse for a conclusive comparison, which will be instead
to M ini 2 M , and its plateau at M WD 0.7 M for 2 M ini attempted by matching the AIS data with GSC2, and with the in-
3.5 M (see Fig. 9, left). Another particularity of this IFMR is the creased MIS coverage expected in the GALEX ‘Extended Mission’
high masses (about 0.64 M ) predicted for the WDs belonging to phase.
the halo and thick disc. This prediction conflicts with recent mass Finally, because our UV-optical matched samples are complete
determinations of WDs in old star clusters, which indicate values to UV magnitudes brighter than ∼20–21, therefore mostly sensitive
close to 0.53 M (Kalirai et al. 2008, 2009). A similar trend is seen to the thin-disc population (Figs 8 and 10), and because the models
at different latitudes. with default MW geometry slightly overpredict the stellar counts at
Overall, an IFMR such as that of W2000 produces predicted low (especially Northern) Galactic latitudes, we explored different

counts closer to the observed numbers, at bright magnitudes where geometries of the thin disc with TRILEGAL. In particular, in Fig. 12
our sample is complete, and at the same time mass distributions in we show the effect of adopting more strongly peaked density distri-
agreement with those derived from spectroscopy of nearby WDs. butions at the Galactic plane: either the simple vertical exponential
These aspects underscore the difficulties of deriving the IFMR from law, ∝ ρ z/hz [which has been adopted in many MW models, starting
theoretical evolutionary tracks of AGB stars, even when these tracks from the classical Bahcall & Soneira (1980, 1984)’s], or the simple
are directly calibrated with observations of AGB stars in the Local hyperbolic secant law, sech(2z/hz ). The results confirm that the de-
Group, as is the case of MG07. This work, and future analyses fault geometry adopted by Girardi et al. (2005) and in this work, i.e.
planned on a wider sky area to explore concurrently the effects the squared hyperbolic secant law, provides the closest prediction to
of the MW dust geometry, suggests that a deeper investigation of the observed counts. The same distribution does also provide a good
AGB evolution is necessary, especially concerning the theoretical match to star counts of ‘normal stars’ in 2MASS and in the local
prescriptions for mass loss, to reconcile constraints provided by Hipparcos sample (Girardi et al. 2005). The slight overprediction
WDs in our Galaxy and those provided by AGB stars in nearby and underprediction of hot-star counts at certain latitudes may also
galaxies. be due to extinction by interstellar dust, which may be either patchy
We finally note that even models calculated with W2000 IFMR (as in fact it is) or have a slightly different distribution than assumed.
tend to predict more faint-star counts than our ‘single’ WD sample These remaining discrepancies and the effects of dust distribution
(Fig. 10). The mismatch is significant if we consider the red model may be addressed with a wider MIS coverage in the future.
lines (model stars selected by their synthetic FUV-NUV colour),
which should be the most consistent with the observed sample
4.4 The kinematic of Milky Way stars
selection, to the extent that the model atmospheres used by TRILE-
GAL are correct. Therefore, we explored also the case of a Kalirai The plots of stellar density versus Galactic latitude presented in the
et al. (2008) IFMR: the predicted counts are lower at faint mag- previous sections clearly indicate that we are looking at a population
nitudes (especially the halo component fainter than NUV ∼ 22), of objects for which the mean distance is larger than the mean
as expected from Fig. 9 (left), while still generally in line with the scaleheight on the thin disc. At typical distances of a few hundred
bright star counts. Because of the incompleteness of our sample parsec, there will be no parallax measurements for such stars. The
at magnitudes fainter than ∼21, and other factors that cannot be most direct distance information we can find for them is related to
conclusively constrained with the current sample, we defer more their proper motions.
quantitative conclusions which rests on a comparison of faint stars The table ‘USNO’ in the SDSS data base contains proper motions
to a future work, and we adopt in this work the W2000 IFMR, that for SDSS objects made after recalibrating USNO-B1.0 (Munn et al.
matches well the bright star counts, to explore other effects. 2004) against SDSS astrometry. The systematic errors are quoted
to less than 1 mas year−1 . The UCAC3 proper motion catalogue is
strongly incomplete for our sample, owing to its magnitude limit
4.3 Milky Way geometry
of f .mag ∼ 17 (which roughly corresponds to r ∼ 17). In fact, we
As explained above the default parameters in the TRILEGAL code, checked and found that just a handful of our hot-WD candidates are
in particular the MW geometry, were defined from previous anal- present in UCAC3.
yses of stellar counts from several surveys, performed with this Fig. 13 presents the histograms of proper motions, μ, for the hot
code. Particularly important in this context were the 2MASS star stars in the 20–30N strip of MIS, and for three different cuts in NUV
counts, which are sensitive to the low-mass stars (<0.8 M ). These magnitude. The observed distributions are compared with those
low-mass stars do not overlap with the mass range of the hot-WD predicted from the TRILEGAL models, computed for the same
progenitors (about 0.8–8 M ), however they are very numerous area and for the W2000 IFMR. The model distributions are derived
and long-lived, and are little affected by uncertainties in stellar evo- as follows. The observed velocity ellipsoids (σ U , σ V , σ W ) of each
lutionary models. Therefore, we should expect the geometry to be Galaxy component are taken form the literature. Each simulated star
well constrained by previous works, except for the limited extent of is then given a random space velocity that follows the Schwarzschild
the previous surveys. distribution for this ellipsoid. Space velocities are then corrected
The agreement between our data and model predictions (with for the solar motion (cf. Dehnen & Binney 1998) and projected on
W2000 IFMR) is generally good at bright magnitudes, where our the sky using the transformations derived from Hipparcos. The final
sample is complete. In more detail, we should also explore the de- results are simulated proper motions and radial velocities. We adopt
pendence of observed and predicted stellar counts with longitude. an age-dependent velocity ellipsoid for the thin disc, derived from
The effect is more prominent towards the Galactic plane, there- the Geneva-Copenhagen Survey of Solar neighbourhood (Holmberg
fore we examined latitudes between 20◦ and 30◦ North and South, et al. 2009). For the thick disc and halo, velocity ellipsoids are taken
and computed models with evenly spaced coverage along these from Layden et al. (1996). These prescriptions provide a first-order
strips. The models predict, as expected, different counts towards description of the general distributions of proper motions and radial
the Galaxy centre and anti-centre, shown in Fig. 11. The coverage velocities, but are not expected to describe the details of the local
of our present GALEX-SDSS matched catalogue at low latitudes is velocity streams, nor the effects caused by disc shear in the more


C 2010 RAS
Figure 10. Density of hot stars, separated in 10◦ ranges of Galactic latitude. The area coverage for AIS and MIS, indicated on each panel, varies significantly Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022
among latitudes, and the statistics are better where the coverage is large. TRILEGAL model counts, computed with W2000 IFMR, are shown with lines for
thin disc (dotted), thick disc (dash–dotted) and halo (dashed), as well as total (solid line). The black lines are model stars selected by T eff > 18 000 K (upper
line: total, lower line: single model stars); the red lines are stars selected from the same models by colour cut (FUV−NUV <−0.13) as explained in Fig. 8. The
match is good overall, at bright magnitudes, but better at Northern high latitudes and Southern low latitudes (except for |b| < 20◦ ). The dashed green-outlined
histograms include both single and binary MIS hot-star candidates (the latter including also QSOs at faint magnitudes). The solid-colour histograms are the
‘single’ hot-star candidates (NUV−r < 0.1). The models were computed covering the centres of the GALEX MIS fields in each latitude strip, except for the
lowest latitudes (|b| < 20) where the TRILEGAL models are computed following the distribution of the AIS fields, because MIS has little or no coverage.
The large difference between North and South 20◦ –30◦ latitudes, seen both in observations and predictions, is due to the MIS-SDSS overlap including more
directions towards the Galactic centre in the South, and away from it in the North; the longitude dependence is illustrated in Fig. 11.
distant stars. Moreover, we do not simulate measurement errors in NUV <18 mag. There is quite a good agreement both in the num-
the proper motions. More details will be given elsewhere. bers and distributions of proper motions. The agreement in the
The upper panel of Fig. 13 (right) compares the observed numbers derives from the fact that the photometry is quite com-
and model distributions for a bright WD sample, with a cut at plete, and just 4 per cent of this bright subsample does not have a

C 2010 RAS

Figure 11. Model-predicted counts at latitudes between 20◦ and 30◦ North,
and South, for four symmetric longitude quadrants. Model calculations were
performed in 1◦ fields evenly distributed in longitude, and combined in 90◦
sections, to illustrate the dependence of the model-predicted hot star counts
on longitude. The strongest differences are seen between centre and anti-
centre directions, as expected, and the North–South asymmetry is most
prominent in directions towards the Galactic centre. Lines show thin disc
(dotted), thick disc (dash–dotted), halo (dashed) and total counts (solid). Figure 12. Thin-disc description. In view of the fact that our present data
are most sensitive to the thin-disc population, we compare model results
proper motion listed in USNO-B. This means that both photometry computed with three different thin-disc geometries: in the top panel the
and proper motion information are very complete in this case, and TRILEGAL default geometry, in the middle panel an exponential (option
1) and in the bottom panel a z sech (option 2) description (see text). The
quite similar to the ones expected for a thin-disc population.
top panel, with the default geometry description given in equations (1) and
The middle and bottom panels of Fig. 13 (right) compare the (2), gives the closest results to the observed counts, confirming previous
observed and simulated μ distribution for fainter samples, with findings by Girardi et al. (2005). This option was adopted in all our model
cuts at NUV < 19.5 and NUV < 21.5 mag, respectively. It shows calculations shown. Colours and symbols as in Fig. 10.
that these samples become incomplete, first in the proper motions
at NUV = 19.5, and then in photometry at magnitudes NUV = work will address identification of thick disc and halo objects with
21.5. Note, for instance, the increasing fraction of stars without follow-up spectroscopy. The distance distribution is fairly similar
proper motion measurement in USNO-B, which goes from 13 to 54 at other latitudes.
per cent from NUV < 19.5 to NUV < 21.5. Finally, Fig. 14 presents the expected distributions of radial veloc-
From this comparison we conclude that our hot WDs do follow ities for the latitude 20–30 N sample, from the models, and the radial
the kinematics expected for the thin disc, even if the proper motion velocities derived from SDSS spectra, available for a subsample of
information is severely incomplete for the faintest WD candidates. our hot-star catalogue (Section 3.2). This is a biased subset because
Distance distributions predicted by the same models are shown the choice of SDSS spectroscopic targets is obviously unrelated
in Fig. 13 (left). This plot indicates that the brightest (and complete) to our selection of hot stellar objects. The worse effect, however,
MIS sample of NUV < 18 mag is likely made of WDs located in the making the comparison inconclusive, is the large errors in velocity
thin disc at typical distances between 0.15 and 1 kpc in the latitude determinations from the SDSS pipeline (see Fig. 14, right).
range 20◦ –30◦ North. The fainter sample could instead contain a
significant fraction of stars at distances larger than 2 kpc, comprising
5 C O N C L U S I O N S A N D S U M M A RY
also a small fraction of thick disc and halo WDs. Note, however,
that this distribution of distances is inferred from models, and we From GALEX’s data release GR5 we constructed catalogues of
presently have no data to verify the distances of our sample stars, unique UV sources (i.e. eliminating repeated observations), and
but for the proper motions already depicted in Fig. 13 (right); future subcatalogues with matched SDSS optical photometry. We


C 2010 RAS

Figure 13. Left: distance distributions predicted by our TRILEGAL models for the hot stars in the 20–30 N strip of MIS, and for three different cuts in NUV
magnitude. This plot indicates that the brightest (and complete) MIS sample of NUV < 18 mag is likely made of WDs located in the thin disc at typical
distances between 0.15 and 1 kpc (in this latitude range). The faintest samples could instead contain a significant fraction of stars at distances larger than 2 kpc,
comprising also a small fraction of thick disc and halo WDs. Right: distribution of proper motions (amplitudes): observed distribution (green solid histograms)
are compared to model predictions (continuous lines), computed for the same area. The numbers in each panel indicate the NUV magnitude cut, the number
of predicted hot stars, of observed ones in MIS and of those which have proper motion information from USNO-B.
Figure 14. Distribution of velocities (left) from the SDSS spectroscopic pipeline for the subsample of MIS hot stars with spectra. The comparison is
inconclusive because of the large errors (often larger than the values) in the pipeline velocity measurements, shown in the right-hand panel.
extracted catalogues of hot-star candidates (FUV−NUV <−0.13), area-limited surveys). We explored in this work different thin-disc
mostly comprising hot WDs. Over 38 000 such sources with photo- geometries and different IFMR. GALEX provides an unprecedented
metric error 0.3 mag have SDSS photometry (∼74 000 with UV census of the evolved descendants of 0.8–8 M stars, and the anal-
photometric error 0.5 mag), excluding UV sources with multiple ysis of our current hot-WD sample places some constraints on the
optical matches, that are between ∼8 to 30 per cent of the total. IFMR, one of the crucial ingredients to understand the chemical
The GALEX surveys AIS and MIS cover different latitudes at their enrichment of the ISM.
respective depths (∼21 and ∼23 AB magnitude), enabling a first The brightest (and complete) sample is likely made of hot WDs
quantitative analysis of the hot-WD stellar population with MW located in the thin disc at typical distances between 0.15 and 1 kpc.
models. Current descriptions of MW stellar components (halo, thin The faintest samples could instead contain a significant fraction
and thick disc) are mostly based on stellar counts of low-mass of stars at distances larger than 2 kpc, comprising also hot WDs
(<0.8 M ) stars (2MASS, shallow but all-sky, plus a few deep but in the thick disc and halo. This work is based on the analysis of

C 2010 RAS
matched GALEX-SDSS sources, and the SDSS magnitude limit Let us first consider, in the IFMRs in Fig. 9, the part for
prevents the analysis of thick-disc and halo components. Hot WDs Mini < Mmin dredge
. The chemical yields produced by these stars (mainly
13
are, however, detectable at such larger distances at MIS depth C, N and 4 He) are determined by the first dredge-up, and the
14
(see dark-green histograms in Figs 8 and 10, and Section 3.1), cool-bottom process (Boothroyd & Sackmann 1999) or thermo-
and will be investigated in a future work with deeper optical haline mixing (Charbonnel & Zahn 2007) operating during the
data. RGB, as no subsequent dredge-up events take place on the AGB
Model predictions of hot-star counts at different latitudes, com- according to the standard stellar evolution theory. For these low-
puted with currently accepted MW geometry, and assuming W2000 mass stars, the AGB yield, defined by equation (5), simplifies as
ej
IFMR, match the data quite well at intermediate latitudes and bright MyAGB (k) = [XkRGB − Xk0 ]
MAGB , where XkRGB is the surface abun-

ej
magnitudes, but are somewhat discrepant at low latitudes and faint dance left after the RGB phase, and
MAGB is the total amount of
magnitudes. There is a slight but systematic North–South asymme- mass ejected on the AGB. In turn, the expelled mass is related to
ej
try, and MW models with canonical geometry for stars and dust the final mass through
MAGB = MAGB,0 − Mfinal , where M AGB,0
match better the data at Southern low latitudes, and at Northern denotes the stellar mass at the onset of the AGB phase. The final
high latitudes, indicating that the dust distribution on the MW disc masses of these stars with Mini < Mmin dredge
can be reduced by simply
may also be better constrained by our data than was previously invoking more efficient mass loss along the AGB. Their chemical
possible. The dust distribution will be investigated with the future yields would increase correspondingly, since less envelope mass
more extended sky coverage from the ongoing GALEX mission. will be locked in the central core following the outward displace-
ej
However, the observed asymmetry may also be due to the longi- ment of the H-burning shell, so that
MAGB will be larger. For
tude dependence of the stellar populations, as our present matched- instance, for an AGB mass of M AGB,0 = 1.2 M , a final mass M final
sources sample covers unevenly North and South Galactic latitudes = 0.55 M versus M final = 0.60 M would imply an increase of the
(Fig. 1). Our MIS matched-sources sample is incomplete at magni- AGB chemical yield by (0.60 − 0.55)/(1.2 − 0.60) ≈ 10 per cent,
tudes fainter than about ∼21.5 due to the SDSS limits for the hottest as well as a larger total chemical yield, including the RGB wind
sources. At all latitudes, and at magnitudes where the samples are contribution.
complete, MW model predictions of hot WD counts computed with For Minitial ≥ Mmin dredge
, in addition to the first dredge-up (and
IFMRs favouring lower final masses match the GALEX hot WD possibly second dredge-up on the early-AGB), the chemical yields
counts significantly better than the IFMR currently postulated by include also the enrichment, mainly in 4 He and primary 12 C, pro-
MG07 from AGB stars data. This result contributes an important duced by the He-shell flashes and brought up to the surface by the
constraint to stellar evolution in the final phases which drive the third dredge-up, and the enrichment of 14 N produced by the hot-
yield of chemical elements. bottom burning process occurring in the most massive AGB stars
(Minitial 4.0–4.5 M ). Again, smaller final masses are reached
when higher mass-loss rates occur, and/or deeper dredge-up (i.e.
5.1 The implications for stellar evolution larger λ). These two factors may affect the chemical yields in op-
posite ways. On one hand, a high mass loss limits the number of
What are the implications of constraining the IFMR for the galaxy
dredge-up episodes, hence lowering the chemical yields; on the
chemical evolution? The IFMR contains the record of the nucle-
other hand, a deeper dredge-up conveys larger amounts of carbon
osynthesis history during the TP-AGB phase, as well as determines
and helium to the surface layer, which are then ejected. In addition,
the contribution of AGB stars in terms of dark remnants. The IFMR
mass loss and dredge-up efficiency are actually inter-related, a larger
and the final yield of chemical elements are significantly affected
surface C abundance generally favouring stronger dust-driven mass
by the occurrence (or not) of the so-called ‘third dredge-up’. At
loss in C stars (e.g. Mattsson et al. 2008), while a significant reduc-
each dredge-up episode the core mass is reduced by an amount
tion of the mass of the envelope may decrease its penetration (i.e.
M dredge = λ
M c, TP , where λ is the classical efficiency param-
lower λ) during the third dredge-up (see e.g. Karakas, Lattanzio
eter, and
M c, TP denotes the mass growth of the core during the
& Pols 2002). In sum, a higher mass expelled does not directly
preceeding interpulse period.
imply that such material is relatively more enriched of processed
In the extreme case where λ ∼ 1, no increment of the C–O core
elements.
is predicted. As a consequence, for low-mass stars not experiencing
An IFMR similar to Weidemann (2000) weighs the final masses
the third dredge-up (λ = 0, M < Mmin dredge
), the remnant mass scales
towards lower values than e.g. the IFRM of MG07 in the range
directly with the duration of the TP-AGB lifetime, while for stars
M initial 3.0 M , while the two relations essentially agree at higher
in which the third dredge-up takes place (λ > 0, M ≥ Mmin dredge
)
M initial . For Minitial Mmin dredge
, forcing the theoretical IFMR to con-
this simple proportionality is replaced by a more complex interplay
verge on a W2000-type IFMR by assuming a higher mass loss would
between core growth, dredge-up and mass loss. For such more mas-
produce a modest increase of the chemical yields from these ob-
sive stars (M ≥ Mmin dredge
, where Mmin
dredge
≈ 1.0–1.5M , depending
jects (because more mass is ejected), compared to the MG07 IMFR
on metallicity and model details) the final mass and the chemical
case. As for the initial mass range Mmin dredge
≤ Minitial ≤ 3.0 M , a
yields are related as we will briefly discuss below (see Marigo &
higher mass loss (necessary to reach a smaller final mass, as sug-
Girardi 2001 for more details).
gested by our WD counts) may reduce the number of dredge-up
For the generic element k, the corresponding AGB yield is defined
episodes on the TP-AGB and this may lead to a lower yield of pro-
as
cessed material. In other words, constraining the IFMR translates
into constraining the total mass lost in the AGB phase. The chem-
MyAGB (k) = [Xk (t) − Xk0 ]Ṁ(t) dt, (5)
τAGB ical composition of the ejecta from these stars, however, is also
critically sensitive to the depth of the third dredge-up, which can be
where Xi0 is the initial abundance at the epoch of star forma- calibrated by measurements of abundances in PNe (Marigo et al.
tion, Xi (t) is the value at time t and Ṁ(t) is the current mass-loss 2003) and by a better estimate of the duration of the AGB phase at
rate. different metallicity (Marigo et al., in preparation).


C 2010 RAS
5.2 Future work AC K N OW L E D G M E N T S
With the current MIS coverage (1103 deg2 overlapping with SDSS Data presented in this paper were obtained from the Multimission
photometry in GR5), we compared with MW models stellar counts Archive at the Space Telescope Science Institute (MAST). STScI is
summed in 10◦ latitude strips, combining all longitudes in order operated by the Association of Universities for Research in Astron-
to obtain acceptable statistics. In order to disentangle geometry of omy, Inc., under NASA contract NAS5-26555. Support for MAST
the stellar populations, dust extinction and stellar evolution in more for non-HST data is provided by the NASA Office of Space Science
detail, we need to also analyse the longitude dependence of stellar via grant NAG5-7584 and by other grants and contracts.
counts, as demonstrated by Figs 10 and 11. We will explore dust GALEX is a NASA Small Explorer, launched in 2003 April. We
effects in the MW disc by matching the AIS with GSC2, for a larger gratefully acknowledge NASA’s support for construction, opera-

sky coverage (see Bianchi et al. 2010). The density of hot stars (the tion, and science analysis of the GALEX mission, developed in
rarest in nature) is very low, and large area coverage is needed for cooperation with the Centre National d’Etudes Spatiales of France
good statistics. and the Korean Ministry of Science and Technology. LB and JH
The MW halo and thick disc components, however, become sig- acknowledge partial support from FUSE GI grant H901 (NASA
nificant at UV magnitudes fainter than about 21, therefore require NNX08AG97G). AZ acknowledges financial support from CNPq-
a wider MIS coverage (planned in the GALEX Extended Mission MCT/Brazil. LG acknowledges partial funding from contract ASI-
phase) to be usefully constrained (see Figs 1, 8 and 10), and deeper INAF I/016/07/0. We are very grateful to A. Thakar for discussions
optical surveys. of many issues regarding the SDSS data base, and to L. G. Althaus,
Finally, hot WDs in binaries, again elusive at optical wavelengths, A. Kanaan, S. O. Kepler, and D. Koester for providing their WD
are uniquely and unambiguously revealed and characterized by the models.
GALEX UV sky survey, matched to optical surveys. These binaries
sample different types of stellar pairs (different T eff ranges) than
those detected and characterized by optical surveys, thanks to the
UV photometry (e.g. fig. 6 of Bianchi et al. 2007b). Because our cur- REFERENCES
rent photometric selection includes contamination by extragalactic Althaus L. G., Serenelli A. M., Benvenuto O. G., 2001, MNRAS, 323,
objects in the ‘binary’ locus, and on the other hand the binaries 471
do not significantly change the distribution of stellar counts, which Althaus L. G., Garcı́a-Berro E., Isern J., Córsico A. H., 2005, A&A, 441,
was the subject of the present analysis, the binary candidates will 689
be investigated in a future work. Bahcall J. N., Soneira R. M., 1980, ApJS, 44, 73
Bahcall J. N., Soneira R. M., 1984, ApJS, 55, 67
Bergeron P., Saffer R. A., Liebert J., 1992, ApJ, 394, 228
Bergeron P., Ruiz M. T., Leggett S. K., 1997, ApJS, 108, 339
Bianchi L., 2007, in Gómez de Castro A. I., Barstow M. A., eds,
5.3 Final remarks: the online catalogues UV Astronomy: Stars from Birth to Death. UCM, Madrid, p. 65
The present catalogue of ‘single’ hot-star candidates has a high (http://www.ucm.es/info/nuva)
Bianchi L., 2009, Ap&SS, 320, 11
purity, as estimated from the serendipitous (but not unbiased) spec-
Bianchi L., 2010, in Shustov B., Gómez de Castro A. I., Sachkov M., eds,
troscopic SDSS coverage. While we expect most of the hottest
Ap&SS (special issue ‘The UV Universe 2010’), in press
sources to be high gravity objects (CSPN, subdwarfs or WD), we Bianchi L., Madore B., Thilker D., Gil de Paz A., Martin C., The GALEX
did not apply any cut in gravity, therefore main-sequence stars and Team 2003, in Livio M., Brown T., eds, The Local Group as an Astro-
supergiants hotter than ∼18 000 K are also included in our cata- physical Laboratory. STScI Publ., Baltimore, p. 10
logues. Their number is very small compared to the evolved stars, Bianchi L. et al., 2005, ApJ, 619, L27
however their inclusion in the catalogue makes it more generally Bianchi L. et al., 2007a, ApJS, 173, 659
useful. Bianchi L. et al., 2007b, in Gómez de Castro A. I., Barstow M. A.,
Finally, we remind future users of our catalogues that GALEX UV eds, UV Astronomy: Stars from Birth to Death. UCM, Madrid, p. 95
sources with more than one optical counterpart were excluded from (http://www.ucm.es/info/nuva)
Bianchi L., Hutchings J. B., Efremova B., Herald J., Bressan A., Martin C.,
the analysis of the hot-star sample (and should be excluded from any
2009a, AJ, 137, 3761
sample based on photometric colour selection), but their fraction can
Bianchi L., Efremova B., Herald J., Girardi L., 2009b, in Van Steenberg
be corrected statistically using Table 2. They are included (flagged M. E., Sonneborn G., Moos H. W., Blair W. P., eds, AIP Conf. Ser.
with ‘rank’ = 1) in our online catalogues. A few per cent of the Vol. 1135, Future Directions in Ultraviolet Spectroscopy. Am. Inst.
matched sources may be spurious matches (Fig. 3) but these would Phys., New York, p. 326
mostly have random UV-optical colours. The hot-star candidates Bianchi L. et al., 2010, in Shustov B., Gómez de Castro A. I., Sachkov M.,
with NUV−r > 0.1 includes also some extragalactic objects, the eds, Ap&SS (special issue ‘The UV Universe 2010’), in press
relative fraction is magnitude and latitude dependent. Boothroyd A. I., Sackmann I.-J., 1999, ApJ, 510, 232
The source catalogues (GALEX unique sources, GALEX-SDSS Bragaglia A., Renzini A., Bergeron P., 1995, ApJ, 443, 735
matched sources, hot-stars samples) are available in electronic form Budavari T. et al., 2009, ApJ, 694, 1281
Charbonnel C., Zahn J.-P., 2007, A&A, 467, L15
only, from the author’s web site (http://dolomiti.pha.jhu.edu/uvsky)
Croom S. M., Smith R. J., Boyle B. J., Shanks T.., Miller L., Outram P. J.,
where files description can be found, and will be also posted on
Loaring N. S., 2004, MNRAS, 349, 1397
MAST (http://galex.stsci.edu and http://archive.stsci.edu/hlsp/). A de Martino C., Bianchi L., Pagano I., Herald J., Thilker D., 2008, Mem. Soc.
statistical analysis and discussion of properties, useful to potential Astron. Ital., 79, 704
users of our (and similar) catalogues for understanding complete- Dehnen W., Binney J. J., 1998, MNRAS, 298, 387
ness and biases of any sample selection, is given by Bianchi et al. Dupuis J., 2002, in Howell S. B., Dupuis J., Golombek D., Walter F. M.,
(2010). Cullison J., eds, ASP Conf. Ser. Vol. 264, Continuing the Challenge of

C 2010 RAS
EUV Astronomy: Current Analysis and Prospects for the Future. Astron. Koester D., 2008, Mem. Soc. Astron. Ital., 75, 282
Soc. Pac., San Francisco, p. 15 Layden A. C., Hanson R. B., Hawley S. L., Klemola A. R., Hanley C. J.,
Eisenstein D. J. et al., 2006, ApJS, 167, 40 1996, AJ, 112, 2110
Ferrario L., Wickramasinghe D., Liebert J., Williams K. A., 2005, MNRAS, Liebert J., Bergeron P., Holberg J. B., 2005, ApJS, 156, 47
361, 1131 McCook G. P., Sion E. M., 1999, ApJS, 121, 1
Fleming T. A., Snowden S. L., Pfeffermann E., Briel U., Greiner J., 1996, Madej J., Należyty M., Althaus L. G., 2004, A&A, 419, L5
A&A, 316, 147 Marigo P., 2001, A&A, 370, 194
Gänsicke B. T. et al., 2009, MNRAS, 397, 2170 Marigo P., Girardi L., 2001, A&A, 377, 132
Gil de Paz A. et al., 2007, ApJS, 173, 185 Marigo P., Girardi L., 2007, A&A, 469, 239 (MG07)
Girardi L., Groenewegen M. A. T., Hatziminaoglou E., da Costa L., 2005, Marigo P., Bernard Salas J., Pottasch S. R., Tielens A. G. G. M., Wesselius

A&A, 436, 895 P. R., 2003, A&A, 409, 619
Gontcharov G. A., Bajkova A. T., Fedorov P. N., Akhmetov V. S., 2010, Mattsson L., Wahlin R., Höfner S., Eriksson K., 2008, A&A, 484, L5
preprint (arXiv:1003.3456) Morrissey P. et al., 2007, ApJS, 173, 682
Hansen B. M. S., Liebert J., 2003, ARA&A, 41, 465 Munn J. A. et al., 2004, AJ, 127, 3034
Harris H. C. et al., 2008, ApJ, 679, 697 Rebassa-Mansergas A., Gänsicke B. T., Schreiber M. R., Koester D.,
Heller R., Homeier D., Dreizler S., 2009, A&A, 496, 191 Rodrı́guez-Gil P., 2010, MNRAS, 402, 620
Holberg J. B., Oswalt T., Sion E. M., 2002, ApJ, 571, 512 Rocha-Pinto H. J., Maciel W. J., Scalo J., Flynn C., 2000, A&A, 358, 850
Holberg J. B., Sion E. M., Oswalt T., McCook G. P., Foran S., Subasavage Salpeter E. E., 1955, ApJ, 121, 161
J. P., 2008, AJ, 135, 1225 Schlegel D. J., Finkbeiner D. P., Davis M., 1998, ApJ, 500, 525
Holmberg J., Nordström B., Andersen J., 2009, A&A, 501, 941 Silvestri N. M. et al., 2007, AJ, 134, 741
Hu Q., Wu C., Wu X., 2007, A&A, 466, 627 Szkody P. et al., 2009, AJ, 137, 4011
Kalirai J. S., Hansen B. M. S., Kelson D. D., Reitzel D. B., Rich R. M., Tremblay P. E., Bergeron P., 2007, ApJ, 657, 1013
Richer H. B., 2008, ApJ, 676, 594 Vassiliadis E., Wood P. R., 1994, ApJS, 92, 125
Kalirai J. S., Saul Davis D., Richer H. B., Bergeron P., Catelan M., Hansen Vennes S., Smith R. J., Boyle B. J., Croom S. M., Kawka A., Shanks T.,
B. M. S., Rich R. M., 2009, ApJ, 705, 408 Miller L., Loaring N., 2002, MNRAS, 335, 673
Karakas A. I., 2010, MNRAS, 403, 1413 Wachter S., Hoard D. W., Hansen K. H., Wilcox R. E., Taylor H. M.,
Karakas A. I., Lattanzio J. C., Pols O. R., 2002, Pub. Astron. Soc. Australia, Finkelstein S. L., 2003, ApJ, 586, 1356
19, 515 Weidemann V., 2000, A&A, 363, 647 (W2000)
Kepler S. O., Kleinman S. J., Nitta A., Koester D., Castanheira B. G.,
Giovannini O., Costa A. F. M., Althaus L., 2007, MNRAS, 375, 1315
Kleinman S.J, Nitta A., Koester D., 2009, J. Phys.: Conf. Ser., 172, 2020 This paper has been typeset from a TEX/LATEX file prepared by the author.


C 2010 RAS

Mnras0411 2770

Uploaded by

Copyright:

Available Formats

Mnras0411 2770

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mnras0411 2770

Uploaded by

Copyright:

Available Formats

Mon. Not. R. Astron. Soc. 411, 2770–2791 (2011) doi:10.1111/j.1365-2966.2010.17890.

Catalogues of hot white dwarfs in the Milky Way from GALEX’s

Luciana Bianchi,1 Boryana Efremova,1 James Herald,1 Léo Girardi,2

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

C 2010 The Authors

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

2.3 The matched UV-optical source catalogue

C 2010 The Authors, MNRAS 411, 2770–2791

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

C 2010 The Authors, MNRAS 411, 2770–2791

Monthly Notices of the Royal Astronomical Society

C 2010 The Authors, MNRAS 411, 2770–2791

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

C 2010 The Authors, MNRAS 411, 2770–2791

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

C 2010 The Authors, MNRAS 411, 2770–2791

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

C 2010 The Authors, MNRAS 411, 2770–2791

Monthly Notices of the Royal Astronomical Society

Downloaded from https://academic.oup.com/mnras/article/411/4/2770/983368 by Indian Institute of Astrophysics user on 08 May 2022

Monthly Notices of the Royal Astronomical Society

You might also like