The UK Infrared Telescope M 33 monitoring project. IV. Variable red giant stars across the galactic disc
We have conducted a near-infrared monitoring campaign at the UK InfraRed Telescope (UKIRT), of the Local Group spiral galaxy M 33 (Triangulum). The main aim was to identify stars in the very final stage of their evolution, and for which the luminosity is more directly related to the birth mass than the more numerous less-evolved giant stars that continue to increase in luminosity. In this fourth paper of the series, we present a search for variable red giant stars in an almost square degree region comprising most of the galaxy’s disc, carried out with the WFCAM instrument in the K band. These data, taken during the period 2005–2007, were complemented by J- and H-band images. Photometry was obtained for 403 734 stars in this region; of these, 4643 stars were found to be variable, most of which are Asymptotic Giant Branch (AGB) stars. The variable stars are concentrated towards the centre of M 33, more so than low-mass, less-evolved red giants. Our data were matched to optical catalogues of variable stars and carbon stars and to mid-infrared photometry from the Spitzer Space Telescope. Most dusty AGB stars had not been previously identified in optical variability surveys, and our survey is also more complete for these types of stars than the Spitzer survey. The photometric catalogue is made publicly available at the Centre de Données astronomiques de Strasbourg.
keywords:stars: evolution – stars: luminosity function, mass function – stars: mass-loss – stars: oscillations – galaxies: individual: M 33 – galaxies: stellar content
The Local Group galaxy Triangulum (Hodierna 1654) – hereafter referred to as M 33 (Messier 1771) – offers us a unique opportunity to study stellar populations, their history and their feedback across an entire spiral galaxy and in particular in its central regions, that in our own Milky Way are heavily obscured by the intervening dusty disc (van Loon et al. 2003; Benjamin et al. 2005). Our viewing angle with respect to the M 33 disc is more favourable (56– – Zaritsky, Elston & Hill 1989; Deul & van der Hulst 1987) than that of the larger M 31 (Andromeda), whilst the distance to M 33 is not much different from that to M 31 ( mag – Bonanos et al. 2006). For these reasons, numerous surveys have been conducted to study M 33 in different wavelength ranges including optical (Macri et al. 2001; Hartman et al. 2006; Massey et al. 2006); radio (Engargiola et al. 2003; Rosolowski et al. 2007); X-ray (Pietsch et al. 2004); and infrared (IR – Two Micron All Sky Survey [2MASS], Skrutskie et al. 2006; McQuinn et al. 2007). Large populations of Asymptotic Giant Branch (AGB) stars have been identified in M 33 (Cioni et al. 2008), as well as red supergiants (RSGs) up to progenitor masses in excess of 20 M (Drout, Massey & Meynet 2012). Many of them are dusty Long Period Variables (LPVs – McQuinn et al. 2007; Thompson et al. 2009), and these have been found also in the central parts of M 33 (Javadi, van Loon & Mirtorabi 2011a; Javadi et al. 2013).
In the final stage of stellar evolution, low–medium mass (0.8–8 M) stars enter the AGB phase (Marigo et al. 2008) and more massive stars ( M) enter the RSG phase (Levesque et al. 2005; Levesque 2010). These two phases of evolution trace stellar populations over a range in age from Myr to Gyr, and hence the evolution of their host galaxies over essential all cosmological time. Radial pulsations of the atmospheric layers in AGB stars and RSGs yield long-period variability of order 150–1500 days in the photometric light curves (e.g., Wood et al. 1992; Wood 1998; Pierce et al. 2000; Whitelock et al. 2003). Most evolved AGB stars pulsate in the fundamental mode, while less evolved AGB stars and RSGs pulsate in an over-tone; as a result the amplitude of variability expressed in magnitude of AGB stars are larger than that of RSGs and less evolved AGB stars. These LPVs are powerful tools to study the star formation history of galaxies, and to this aim various variability surveys have been conducted of M 33 over recent years (Macri et al. 2001; Mochejska et al. 2001a,b; Hartman et al. 2006; Sarajedini et al. 2006; McQuinn et al. 2007).
The coolest (–4000 K) and most luminous (–60 000 L) AGB stars create large amounts of dust in the starâs atmosphere. This dust is ejected into space and can cloak the star, especially in the optical light, and add luminosity at IR wavelengths. Likewise, RSGs stand out especially at IR wavelengths. Hence, among all surveys, those with the Spitzer Space Telescope (McQuinn et al. 2007) and the UK InfraRed Telescope (UKIRT, Javadi et al. 2011a) were more successful in detecting dusty LPVs. Since these evolved stars shed a large amount of mass into the interstellar medium (ISM), they are important factors in changing the chemical composition of galaxies and incrementing the rate of stellar birth.
The main objectives of our project are described in Javadi, van Loon & Mirtorabi (2011c): to construct the mass function of LPVs and derive from this the star formation history in M 33; to correlate spatial distributions of the LPVs of different mass with galactic structures (spheroid, disc and spiral arm components); to measure the rate at which dust is produced and fed into the ISM; to establish correlations between the dust production rate, luminosity, and amplitude of an LPV; and to compare the in situ dust replenishment with the amount of pre-existing dust. Paper I in the series presented the photometric catalogue of stars in the inner square kpc (Javadi et al. 2011a), with Paper II presenting the galactic structure and star formation history (Javadi, van Loon & Mirtorabi 2011b), and Paper III presenting the mass-loss mechanism and dust production rate (Javadi et al. 2013). This paper describes the extension of the survey to a nearly square degree area covering much of the M 33 optical disc. Subsequent papers in the series will cover the star formation history and mass return in this enlarged area.
In Section 2 we present the observational data, method of photometry on the images, and accuracy of these measurements. The methodology and completeness of our search for LPVs, and characterization of their amplitude of variability are discussed in Section 3. Section 4 describes the photometric catalogue of all detected stars, which is made available at the Centre de Données astronomiques de Strasbourg (CDS). In section 5 we describe the properties and distribution of the detected LPVs and also compare these with other (optical and IR) variability surveys and stellar catalogues. Section 6 summarizes and concludes the results.
Observations were made with three of UKIRT’s imagers: UIST, UFTI and WFCAM. UIST and UFTI cover the central part ( kpc) and the photometry, star formation history and mass return in this region was explained in Papers I, II and III. In this sequel we focus on the data from WFCAM, which cover a much larger part of M 33.
|Date (y m d)||Q||Filter||Epoch||(min)||Airmass|
|2005 09 18||3||K||1||20.3||1.035–1.058|
|2005 09 18||2||K||1||20.3||1.072–1.110|
|2005 09 18||4||K||1||20.3||1.248–1.338|
|2005 09 19||1||K||1||20.3||1.021–1.018|
|2005 10 18||3||K||2||20.3||1.019–1.021|
|2005 10 18||2||K||2||20.3||1.025–1.040|
|2005 10 18||4||K||2||20.3||1.053–1.083|
|2005 10 18||1||K||2||20.3||1.101–1.149|
|2005 11 04||1||K||3||20.3||1.018–1.023|
|2005 11 04||2||K||3||13.5||1.028–1.036|
|2005 12 23||2||K||4||27.0||1.019–1.022|
|2005 12 23||3||K||3||20.3||1.028–1.046|
|2006 07 21||1||K||4||20.3||1.425–1.325|
|2006 07 21||2||K||5||20.3||1.287–1.214|
|2006 07 21||3||K||4||20.3||1.183–1.132|
|2006 07 21||4||K||3||20.3||1.109–1.074|
|2006 10 28||1||K||5||27.0||1.294–1.126|
|2006 10 28||1||J||1||20.3||1.102–1.076|
|2006 10 29||1||K||6||20.3||1.445–1.347|
|2006 10 29||1||H||1||27.0||1.295–1.209|
|2006 10 29||1||J||2||27.0||1.115–1.062|
|2006 10 30||1||J||2||33.8||1.200–1.109|
|2006 10 30||4||K||4||33.8||1.085–1.044|
|2006 10 31||4||H||1||6.8||1.301–1.301|
|2006 12 05||2||K||6||20.3||1.019–1.025|
|2006 12 12||3||K||5||20.3||1.082–1.052|
|2006 12 12||3||H||1||20.3||1.040–1.027|
|2007 01 14||1||K||7||20.3||1.124–1.169|
|2007 01 14||1||J||3||20.3||1.217–1.284|
|2007 01 14||1||H||2||20.3||1.342–1.441|
|2007 01 15||2||K||7||20.3||1.119–1.163|
|2007 01 16||2||H||1||20.3||1.063–1.092|
|2007 01 17||3||J||1||20.3||1.031–1.047|
|2007 01 18||2||J||2||20.3||1.029–1.044|
|2007 01 25||3||K||6||20.3||1.072–1.104|
|2007 01 25||3||H||2||20.3||1.161–1.215|
|2007 09 14||1||K||8||20.3||1.122–1.086|
|2007 09 14||1||J||4||13.5||1.070–1.058|
|2007 09 14||1||H||3||13.5||1.046–1.038|
|2007 09 19||2||K||8||20.3||1.774–1.606|
|2007 10 04||2||J||3||13.5||1.208–1.181|
|2007 10 04||2||H||2||13.5||1.155–1.132|
|2007 10 13||3||K||7||20.3||1.108–1.076|
|2007 10 13||3||H||3||13.5||1.056–1.046|
|2007 10 24||4||K||5||20.3||1.135–1.097|
|2007 10 24||3||J||2||13.5||1.078–1.064|
|2007 10 24||4||J||1||13.5||1.050–1.041|
|2007 10 24||4||H||2||13.5||1.025–1.022|
The monitoring campaign comprises observations with the Wide Field CAMera (WFCAM) such that four separately pointed observations (“tiles”, viz. M 33-1, M 33-2, M 33-3 and M 33-4) may be combined to cover a filled square area of sky covering 0.89 square degree (13 kpc 13 kpc) at a pixel size of . The approximate centres of the camera pointings are respectively (, ), (, ), (, ), (, ). Observations were made in the K band (UKIRT filter K98) over the period September 2005–October 2007 (Table 1). On some occasions observations were made also in the J band and/or H band (UKIRT filters J98 and H98, respectively) to provide colour information. The total integration on one tile was achieved through four separate exposures. The number of epochs varies per tile between five and eight, but overlapping regions will have been observed more frequently.
The large pixel scale of WFCAM means that the point spread function (PSF) is not always adequately sampled for the crowded fields in M 33 under modal observing conditions; to remedy this we employed a microstepping scheme to improve the sampling of the PSF. At each position of the nine-point microstepping sequence a 5-sec exposure was taken, and a small offset (“jitter”, ) was applied between each of nine subsequent microstep sequences. This cycle was repeated three times. Thus, a typical observation accrued a total integration time of min; however, in practice fewer or more repeats were performed depending on conditions, time available, or for technical reasons (interruptions).
The images were processed using the WFCAM pipeline by the Cambridge Astronomy Survey Unit (CASU – http://casu.ast.cam.ac.uk/). The main steps include:
Dark current correction;
Flat field correction to remove pixel sensitivity differences and gain differences between data channels and between detectors;
Confidence map generation: a normalized inverse weight map denoting the confidence associated with the flux values in each pixel;
Defringing, provided the fringe spatial pattern was available;
Sky subtraction, either using the ditter sequences themselves (our case) or using observations of offset sky regions;
Image persistence and detector crosstalk correction, i.e. modelling and removing electronic effects;
Combine (“interleave”) the images from the microstepping sequence;
Shift and average individual images from the jittering sequence;
Catalogue generation. Objects are identified from the images and listed in FITS binary tables. This catalogue includes assorted aperture flux measures, intensity-weighted centroid estimates, and shape information such as intensity-weighted second moments to encode the equivalent elliptical Gaussian light distribution;
Astrometric calibration. All objects in the catalogue are matched to astrometric standards to define a World Coordinate System (WCS) for each image/catalogue;
Photometric zeropoint from comparison of instrumental magnitudes with 2MASS (K band is ).
Our programme IDs are U/05B/18, U/06B/40 and U/07B/17. We have complemented our data with WFCAM archival data taken for two projects, viz. U/05B/7 and U/05B/H47 which we briefly describe below.
|Date (y m d)||Q||Filter||Epoch||(min)||Airmass|
|2005 09 29||4–1||J||1||1.0||1.196–1.197|
|2005 09 29||4–1||H||1||4.5||1.206–1.227|
|2005 09 29||4–1||K||1||4.5||1.227–1.023|
|2005 09 30||2–1||J||1||1.0||1.023–1.023|
|2005 09 30||2–1||H||1||4.5||1.024–1.024|
|2005 09 30||2–1||K||1||4.5||1.028–1.028|
|2005 09 30||2–3||J||1||1.0||1.055–1.056|
|2005 09 30||2–3||H||1||4.5||1.060–1.060|
|2005 09 30||2–3||K||1||4.5||1.069–1.069|
|2005 10 24||1–1||J||1||1.0||1.335–1.333|
|2005 10 24||1–1||H||1||4.5||1.320–1.319|
|2005 10 24||1–1||K||1||4.5||1.293–1.292|
|2005 10 24||1–2||J||1||1.0||1.050–1.020|
|2005 10 24||1–2||H||1||4.5||1.340–1.210|
|2005 10 24||1–2||K||1||4.5||1.210–1.120|
|2005 10 24||1–3||J||1||1.0||1.204–1.202|
|2005 10 24||1–3||H||1||4.5||1.193–1.193|
|2005 10 24||1–3||K||1||4.5||1.175–1.174|
|2005 10 24||1–4||J||1||1.0||1.158–1.157|
|2005 10 24||1–4||H||1||4.5||1.145–1.144|
|2005 10 24||1–4||H||1||4.5||1.130–1.130|
|2005 11 05||3–1||K||1||4.5||1.073–1.073|
|2005 11 05||3–1||H||1||4.5||1.064–1.064|
|2005 11 05||3–1||J||1||1.0||1.055–1.055|
|2005 11 05||3–2||K||1||4.5||1.052–1.052|
|2005 11 05||3–2||H||1||4.5||1.046–1.046|
|2005 11 05||3–2||J||1||1.0||1.039–1.038|
|2005 11 27||1–1||J||2||1.0||1.292–1.295|
|2005 12 16||1–1||J||2||1.0||1.107–1.106|
|2005 12 16||1–1||H||2||4.5||1.101–1.101|
|2005 12 16||1–1||K||2||4.5||1.090–1.089|
|2005 12 16||1–2||J||2||1.0||1.079–1.078|
|2005 12 16||1–2||H||2||4.5||1.074–1.074|
|2005 12 16||1–2||K||2||4.5||1.065–1.064|
|2005 12 16||1–3||J||2||1.0||1.055–1.054|
|2005 12 16||1–3||H||2||4.5||1.051–1.051|
|2005 12 16||1–3||K||2||4.5||1.044–1.044|
|2005 12 16||1–4||J||2||1.0||1.038–1.038|
|2005 12 16||1–4||H||2||4.5||1.035–1.035|
|2005 12 16||1–4||K||2||4.5||1.031–1.031|
The data of this programme (Cioni et al. 2008) were taken to survey the luminous red giant stars of Local Group galaxies. M 33 was observed on four occasions: 29 and 30 September, 24 October, 5 November and 16 December 2005 (Table 2). photometry was obtained from a mosaic of four fields (instead of the one central field in our case), covering an area square degrees. The and data were acquired employing a three-point jitter pattern with microstepping and 10-sec exposures per position, giving a total integration time of 270 sec. The data were acquired using a five-point jitter pattern with three 10-sec exposures but no microstepping, resulting in a total integration time of 150 sec. The data were processed by the CASU.
|Date (y m d)||Q||Filter||Epoch||(min)||Airmass|
|2005 10 20||1||K||1||8.3||1.077–1.059|
|2005 10 20||2||K||1||8.3||1.052–1.041|
|2005 10 20||3||K||1||8.3||1.038–1.030|
|2005 10 20||4||K||1||8.3||1.028–1.023|
The M 33 data for this programme (PI: M. Irwin) were acquired on 20 October 2005 (Table 3). The covering area is nearly identical to ours, with the camera pointings called M 33-position 1, M 33-position 2, M 33-position 3 and M 33-position 4, respectively.
We present the combined, square-degree mosaic of M 33 in the K band in figure 1. The previously investigated, square-kpc area is delineated with a box. While the spiral structure is evident the images are much less affected by extinction by interstellar dust than images at optical wavelengths. The very bright Galactic foreground red giant HD 9687 (, ) has left a trail of ghost imprints to its East, but saturation is not normally a problem across the mosaic. Most stars above the tip of the red giant branch (RGB) are resolved also with WFCAM, except within the central few arcsec dominated by the nuclear star cluster.
2.3 Cross correlation of catalogues
The photometric catalogues of M 33 were retrieved from the public WFCAM Science Archive (WSA). Only the overlapping area of the other catalogues (and images) with those from our own programme will be considered here; they are meant to provide further epochs that increase the sensitivity and reliability of the variability detection.
The FITS catalogues contain only un-calibrated fluxes (, in counts) obtained in a series of apertures. By knowing the photometric zeropoint at unity airmass, , the extinction coefficient, (Krisciunas et al. 1987), the airmass at the start and end of observation, and , and exposure time, , the (telluric) extinction-corrected and flux-calibrated magnitudes, , are determined by:
In order to determine the mean magnitude and light curve of all stars over the epochs in which they were detected, unique IDs need to be assigned to all individual stars. To this aim, we cross-identified each star within every catalogue. The matches were obtained by performing search iterations using growing search radii, in steps of out to , on a first-encountered first-associated basis but after ordering the principal photometry in order of diminishing brightness (to avoid rare bright stars being erroneously associated with any of the much larger number of faint stars).
2.4 Photometric calibration
Given the high level of crowding especially towards the central parts of M 33, and the importance of accurate relative photometry between epochs in order to correctly separate variable from non-variable sources, it is essential to check the accuracy of the WSA catalogues. To this aim, we performed PSF photometry using the DAOPhot package within IRAF (Stetson 1987) on one of the frames (from camera 1) in the central part field M 33-3. An empirical constant PSF model with a 2D elliptical Gaussian function was used to construct the PSF from seven isolated stars. Then, the allstar task was used to estimate the instrumental magnitude for 32,627 stars identified with the daofind task. The transformation factor from instrumental magnitude to standard magnitude was obtained from the standard magnitudes of 30 stars in common from the UIST catalogue (Paper I), hence deriving for the WFCAM data.
The celestial coordinates of the stars were calculated using the IRAF ccmap–cctran tasks. In this frame, the WSA catalogue lists 36 301 stars. We cross-identified these with the DAOPhot catalogue within a search radius of ; hence, 24 140 stars were identified in common. Figure 2 shows the difference in magnitude between the WSA and DAOPhot photometry, against the WSA magnitude; the histogram of magnitude differences is shown in figure 3. The vast majority of stars have magnitudes that are consistent between the two methods of measurement within a few tenths of a magnitude; among the 3.5 per cent of stars with mag, 78 per cent are located near (within of) the centre of M 33 where crowding is more severe than elsewhere in the mosaic. This renders just 0.9 per cent of stars with suspect photometry. Upon visual inspection of the image, it was found that in most cases there is a faint star near a bright star, and the photometric difference is simply the result of erroneous cross-identification. Consequently, the accuracy of the WSA magnitudes is acceptable.
2.4.1 Relative calibration
The relative calibration between frames was obtained from the mean magnitudes of stars in common within the magnitude interval . While these will include some variable stars the vast majority will not vary by more than 10 per cent rendering their mean magnitude accurate to well within a per cent. Then the photometry from the different frames was brought in line with each other by applying corrections that equalised these mean magnitudes. The corrections are shown in figure 4, also for the UIST survey (they were not shown in Paper I). Note that the corrections are small – generally a few per cent and always less than 10 per cent.
3 Variability analysis
The search for variable stars was done by calculating the variability index for each star. This index was introduced by Welsh & Stetson (1993) and developed further by Stetson (1996). In this method, first the observations are paired on the basis of timespan between observations such that the observations of each pair have a timespan less than the shortest period expected for the kind of variable star of interest. In case more than two observations were performed within the timespan of shortest periodicity, those sets of observations would be paired in more than one pair. Hence, the index is calculated:
Here, observations and have been paired and for each pair a weight is assigned; the product of normalized residuals, 111Following Stetson (1996), if ., where , is the residual of measurement from the mean magnitude, normalized by the error of the measurement, ; and is the total number of observations. Note that and may refer to observations taken in different filters. The index for non-variable stars is approximately zero as the residuals arising from random noise are uncorrelated and their product will therefore tend to zero for large sets of measurements.
The effect of a small number of observations or corrupt data can be limited by means of a backup index, viz. the Kurtosis index:
The shape of the light variations define the value of the Kurtosis index; for example, for a Gaussian distribution which is concentrated towards the average brightness level (as would be random noise), and for data affected by a single outlier (when ).
Here, we use the variability index (Stetson 1996):
where is the total weight assigned to a given star and is the total weight a star would have if observed in every single observation.
Figure 5 shows how the variability index varies with K-band magnitude. For comparison with our previous UIST survey (Paper I), in figure 6 we show the distribution for the central square kpc of M 33, where we indicate the WFCAM detections of stars that had been identified as variable in the UIST survey. Both graphs reveal a noticeable “branch” of stars with larger than usual between –18 mag; these are likely AGB stars with Mira-type variability.
Several tests were performed to select the optimal variability index threshold. First we inspected the histograms of variability index within several magnitude bins in the range 16–19 mag for all detected stars including the M 33 disc and central regions (Fig. 7). To determine the variability index threshold, a Gaussian function was fitted to each of these histograms. The Gaussian function should be near-perfectly fitted to those data for low values of , while it departs from the histograms for larger values of . This appears to happen around but at smaller values for faint stars; we thus decided, in first instance, to set the threshold at for the detection of variability.
Next, we test our procedure by comparing the WFCAM selected variables, with , with the UIST catalogue of variables in the central square kpc (from Paper I). The percentage of WFCAM variables within the magnitude ranges shown in figure 7 is about twice that derived from the UIST survey. If anything, the opposite would be expected as the UIST survey had a superior cadence of observations. Indeed, when selecting only stars from this central region of M 33, the histograms of variability index broaden, and a more appropriate threshold would be (Fig. 8). On the other hand, a threshold of seems appropriate for the disc of M 33, i.e. excluding the central square kpc (Fig. 9).
To further assess the validity of the choice of variability index threshold, we examined the location of the putative variable stars in a colour–magnitude diagram (CMD), for different choices of threshold: , and (Fig. 10). While there is no clear difference in the selection of variable stars on the AGB and RSG portions of the CMD (roughly at –18.5 mag and mag), all but the thresholds result in many putative variable stars at fainter magnitudes and along the blue and bright vertical sequence (around mag) where no large-amplitude red giant variables are expected.
Finally, we inspected the radial distribution of putative variable stars for the case where we apply different choices for the threshold for the central square kpc and the (remainder of the) disc, as compared to applying a single threshold for all (Fig. 11). The former case results in a break in gradient around – i.e. well outside the central square kpc – where results in a seemingly disproportionally increased number of variable stars closer to the centre. This gradient is not sustained within the central – i.e. corresponding roughly to the central square kpc. Indeed, the distribution turns over and much fewer variable stars are identified once was applied. The same artificial behaviour is not seen when the same threshold is applied across the entire M 33 galaxy.
Based on the histograms, CMD and radial distributions we decided to apply a single variability threshold of .
3.1 Comparison between the WFCAM and UIST catalogues of variable stars within the central square kpc
One final assessment of the variability detection success is made by a more careful comparison of the common area of the WFCAM and UIST surveys. After sorting both catalogues in order of diminishing brightness in K band, matches were obtained through successive search iterations using increasing search radii, in steps of out to . Within the coverage of UIST (Paper I), using WFCAM we detected 11 114 stars; from 18 398 stars detected with UIST, using WFCAM we detected 10 095 stars. Applying , we find 667 variable stars located within the central square kpc covered by UIST (out of 4643 in total across the WFCAM coverage); applying , this becomes 2696 – i.e. more than three times as many as were found in the more capable UIST survey. Again applying , only 192 out of the 812 UIST LPVs were identified with WFCAM (a success rate of 24%). This also suggests that the UIST survey, too, is incomplete.
Figures 12 and 13 show CMDs in which are highlighted those variable stars which have and those which have not been detected with either WFCAM of UIST, when applying a WFCAM variability index threshold of and , respectively. These CMDs suggest that is a more sensible threshold than , for two reasons; firstly, it yields a somewhat smaller number of variables with WFCAM than with UIST, which is expected as WFCAM has more difficulty in isolating blended stars and also the WFCAM survey is based on fewer epochs. Secondly, the fraction of UIST variables that were found with WFCAM is smaller but the fraction of WFCAM variables that were found with UIST is much larger (31%) than when applying (14%).
Since the central part of M 33 comprises a considerable variation in stellar density, the same data displayed in figures 12 and 13 are shown in figure 14 as a function of radial distance from the centre of M 33. The rate at which variable stars are detected increases towards the centre, and as a consequence the number of variable stars found in one of the surveys that is not recovered within the other survey also increases towards the centre. However, when applying a variability index threshold of , the WFCAM survey yields many more variable stars than the UIST survey, whilst for this is about equal. This again favours the choice of the latter threshold. The success rate of the WFCAM survey to recover UIST variable stars displays only a shallow gradient with radial distance from the centre of M 33 (Fig. 15); it averages % for this central region (but is % for the square area encompassing this circular area).
The recovery rate of UIST variables within the WFCAM catalogue of variable stars, and vice versa, is plotted as a function of WFCAM variability index threshold in figure 16. At , a near-equal fraction of variable stars is recovered within each of the surveys (%) – the UIST survey is slightly more successful than the WFCAM survey, as expected (see above). As increases, the number of variable stars identified in the WFCAM survey decreases but these variable stars will be the ones with larger amplitudes. Hence, the UIST survey will be more complete in including those WFCAM variable stars, reaching % success rate for . On the other hand, the WFCAM survey will miss more of the UIST variable stars (generally the ones with smaller amplitudes), and its success rate drops to just one per cent for . Hence, the choice for is supported once again.
3.2 Amplitudes of variability
We estimate the amplitude of variability by assuming a sinusoidal lightcurve shape. The amplitude is then:
where is the standard deviation in our data and 0.701 is the standard deviation of a unit sine function. The standard deviation is uncertain when the number of measurements () is small; we showed in Paper I that is the minimum acceptable to reach % fidelity.
The estimated K-band amplitude is plotted vs. K-band magnitude in figure 17. As is well known by now (Wood et al. 1992; Wood 1998; Whitelock et al. 2003; Paper I), the amplitude of variability exhibits a clear tendency to diminish with increasing brightness, which is partly due to the definition of magnitude as a relative measure rather than less powerful pulsation (van Loon et al. 2008). The amplitude is generally about a magnitude or less, but a small fraction of variable stars (8%) seem to reach mag. Very dusty AGB stars – which are rare – are known to reach such large amplitudes (Wood et al. 1992; Wood 1998; Whitelock et al. 2003). Among these extreme variables, 311 stars have ; excluding these stars, only 20 stars remain with mag. The brightest variables, with mag, are foreground red giants; among the faintest stars, with mag, our survey becomes increasingly less sensitive to small-amplitude variables.
4 Description of the catalogue
|Part I: stellar mean properties (403,734 lines)|
|2||Right Ascension (J2000)|
|4||Mean J-band magnitude|
|6||Mean H-band magnitude|
|8||Mean K-band magnitude|
|10||Number of J-band measurements|
|11||Number of H-band measurements|
|12||Number of K-band measurements|
|16||Estimated K-band amplitude|
|Part II: multi-epoch data (3,623,332 lines)|
|3||Filter (J, H or K)|
|5||Error in magnitude|
The photometric catalogue including all variable and non-variable stars is made publicly available at the Centre de Données astronomiques de Strasbourg (CDS). The content is described in Table 4. It is composed of two parts, part I comprising the mean properties of the stars and part II tabulating all the photometry (for the benefit of generating lightcurves, for instance).
The astrometric accuracy of the catalogue is r.m.s., tied to the 2MASS system. This accuracy was found to be consistent with the results from our cross-correlations with three other optical and IR catalogues (cf. Section 5.3).
5.1 The near-IR variable star population
Figure 18 presents near-IR CMDs for the whole region of M 33 monitored with WFCAM. The Large-amplitude variable stars we identified are highlighted in green. These are mainly found between –18 mag, and are largely absent among fainter stars (below the tip of the RGB at mag). Some brighter variable RSGs are found (around mag), but the clump of variables stars with mag and redder colour in than are foreground stars and perhaps saturated. Variable stars dominate the redder stars to the right of the vertical sequence comprising the bulk of stars.
The distributions over brightness (Fig. 19) and colour (Fig. 20) provide another means of assessing the properties of the variable stars. The largest fraction of stars that are found to be variable occurs between –17 mag, albeit less than that of the UIST survey in the central region in the same magnitude interval. The variable star population reaches a peak around mag; it then drops at fainter magnitudes even though the total stellar population keeps increasing. This mainly arises from two factors: firstly, many stars in this magnitude interval have not yet reached the final phase of their evolution and they will still evolve to higher luminosities and lower temperatures before they develop large-amplitude variability; secondly, the birth-mass and K-band brightness relation flattens considerably for low-mas AGB stars (see Paper II). At mag or mag almost all stars ( mag) are variable; these are dusty, strongly pulsating and heavily mass-losing AGB stars.
The level of contamination by foreground stars can be assessed with the trilegal simulation tool (Girardi et al. 2005). We used default parameters for the structure of the Galaxy, simulating a 0.9 square degree field in the direction (, ). Only a relatively small number of foreground stars are expected, with fairly neutral colours or below our completeness limit (Fig. 21). The part of the CMD occupied by large-amplitude variable stars is relatively uncontaminated by foreground stars.
The stellar populations can be described using isochrones calculated by Marigo et al. (2008) (Fig. 22). The isochrones were calculated for solar metallicity () for all stellar populations except the oldest, least chemically evolved one with for which we adopted . To account for a (shallow) metallicity gradient across the disc of M 33, we show CMDs of each of the 16 tiles of the M 33 mosaic, overlain with isochrones for in the central region but further out in the disc (Fig. 23).
The isochrones from Marigo et al. (2008) are the most realistic models to be used for the purpose of this study, for the folowing reasons:
The star’s evolution is followed all the way through the thermal pulsing AGB until the post-AGB phase. Crucially, two important phases of stellar evolution are included, viz. the third dredge-up mixing of the stellar mantle as a result of the helium-burning phase, and the enhanced luminosity of massive AGB stars undergoing hot bottom burning (HBB; Iben & Renzini 1983);
The molecular opacities which are important for the cool atmospheres of evolved stars have been considered in the models of stellar structure. The transformation from oxygen-dominated (M-type) AGB stars to carbon stars in the birth mass range –4 M is accounted for (cf. Girardi & Marigo 2007);
The dust production in the winds of LPVs, and the associated reddening is included;
The radial pulsation mode is predicted;
Combination of their own models for intermediate-mass stars ( M), with Padova models for more massive stars ( M; Bertelli et al. 1994), gives a complete coverage in birth mass ( M);
Magnitudes are calculated on a wide range of common optical and IR photometric systems;
The isochrones are available via an internet-based form, in a user-friendly format.
5.2 Spatial distribution of LPVs
Maps of the surface density of the number of large-amplitude variable stars, AGB stars, RGB stars and massive stars are presented in figure 24. The same selection criteria as in Paper II are used to select these different populations: the demarcation between massive stars and less-massive giant stars is defined to run from mag to mag, such that massive stars have colours bluer than this (down to mag) or have mag, whilst AGB stars and RGB stars are redder than this and have mag or mag, respectively.
The variable stars, AGB stars and massive stars are concentrated towards the centre but the RGB stars do not show such strong central concentration. This is in good agreement with what we found for the central square kpc of M 33 in Paper II. The distribution of the variable stars mostly mimics that of the AGB stars, as expected, though the somewhat stronger central concentration in the former suggests that more massive stars (AGB stars as well as RSGs) make a larger contribution in the central part of M 33 than further out in the disc. Only hints can be seen of the spiral arm pattern in these maps.
5.3 Cross-identifications in other catalogues
We have cross-correlated our UKIRT/WFCAM variability search results with those from two intensive optical monitoring campaigns (CFHT, Hartman et al. 2006, in 5.3.1; DIRECT, Macri et al. 2001, in 5.3.2). We have also compared with the optical catalogue of Rowe et al. (2005) which includes narrow-band filters that they used to identify carbon stars (in 5.3.3); RSGs from Drout et al. (2012) in 5.3.4; and a small number of Luminous Blue Variable (LBV) stars from Humphreys et al. (2014) in 5.3.5. In addition, we compared our data with the mid-IR variability search performed with the Spitzer Space Telescope (McQuinn et al. 2007) in 5.3.6. The matches were obtained by search iterations using growing search radii, in steps of out to , on a first-encountered first-associated basis after ordering the principal photometry in order of diminishing brightness (K-band for the UKIRT catalogue, i-band/I-band for the optical catalogues, and 3.6-m band for the Spitzer catalogue). Finally, we examined our data on the recently identified 24-m variables (Montiel et al. 2014) which includes the giant H ii region NGC 604 (in 5.3.7).
5.3.1 CFHT optical variability survey
A variability survey of M 33 was carried out with the 3.6-m Canada–France–Hawai’i Telescope (CFHT) on 27 nights comprising 36 individual measurements, between August 2003 and January 2005 (Hartman et al. 2006). Out of two million point sources in a square-degree field, they identified more than 1300 candidate variable blue and red supergiants, more than 2000 Cepheids, and more than 19 000 AGB and RGB LPVs. Their catalogue comprises Sloan g-, r- and i-band photometry to a depth of mag.
Out of 36 000 variable stars detected with CFHT in our field of M 33, our UKIRT/WFCAM survey detected 29 600 stars (82%), of which 1818 were found by us to be variable (Fig. 25). Most of the CFHT variables that were missed in our survey are fainter than the RGB tip; these are not the type of variables that our survey aims to detect. The AGB and RSG variables that were not detected in our survey have modest amplitudes (especially at IR wavelengths). On the other hand, our survey detected some of the dustiest AGB variables that were missed by the CFHT survey.
5.3.2 DIRECT optical variability survey
The DIRECT variability survey (Macri et al. 2001) aimed to determine a distance estimate of M 33 (and M 31) using detached eclipsing binaries and Cepheids. It was carried out on 95 nights with the F.L. Whipple observatory’s 1.2-m telescope and on 36 nights with the Michigan–Dartmouth–MIT 1.3-m telescope, between September 1996 and October 1997. Their catalogue contains Johnson B- and V-, and Cousins I-band photometry for all stars with mag, and the V-band variability index (cf. Section 3).
The central region of M 33 observed in DIRECT is completely covered in our WFCAM survey. The DIRECT survey listed 57 581 stars among which 1383 have (which the DIRECT team assumed as the variability threshold). Within the coverage of DIRECT are located 116 606 stars from the WFCAM catalogue, including 1444 WFCAM variables. Of the 24 422 stars from DIRECT that were detected in our WFCAM survey, 412 were found by us to be variable. Hence, most of the WFCAM variables were missed by the DIRECT survey (Fig. 26). This is unsurprising since dusty AGB variables are faint at optical wavelengths, but the relatively short duration of the DIRECT survey (little more than a year) may have contributed to it missing also less dusty AGB variables.
5.3.3 Carbon star survey
Rowe et al. (2005) used the CFHT and a four-filter system in 1999 and 2000 to cover much of M 33. The filter system was designed to identify carbon stars on the basis of their cyanide (CN) absorption as opposed to other red giants that display titanium-oxide (TiO) absorption, using narrow-band filters centered on 8120 and 7777 Å, respectively. They added broad-band Mould V- and I-band filters to aid in selecting cool stars. Carbon stars in their scheme have [CN][TiO] and mag, and M-type stars have [CN][TiO] mag at the same V–I criterion (they only considered stars for this purpose that had errors on these colours of mag).
Within the area in common with our WFCAM survey, Rowe et al. detected 2 079 334 stars, from which 306 292 stars (3.5%) were identified with WFCAM (a further 85 098 stars from the WFCAM catalogue are outside of their coverage). Among these, 11 040 stars are M-type and 3134 are carbon stars. To improve the cross-correlation and to avoid co-incidences because of the high density of optical sources, a pre-selection was made on carbon stars and M-type stars that are likely to have been detected at near-IR wavelengths. The average offset between the two catalogues was found to be only in both RA and DEC.
The main giant branches – viz. the RSGs around mag and the AGB around mag – are traced by M-type stars, with carbon stars digressing towards redder colours at –17 mag (Fig. 27). The carbon star distribution is consistent with a typical Gyr () isochrone or a little younger, i.e. birth masses around 2–3 M. Some of the bright carbon stars show signs of reddening to mag presumably due to circumstellar dust; as these are among the brightest detected carbon stars, they must indicate the termination point in carbon star evolution. Carbon stars are found fainter than the RGB tip, down to mag; these may have formed through binary mass transfer. Such faint carbon stars are known in other, especially metal poor populations, e.g., in the Sagittarius dwarf irregular galaxy (Gullieuszik et al. 2007) or in the Galactic globular cluster Centauri (van Loon et al. 2007). Note that our WFCAM survey detected large-amplitude variability in carbon stars only brighter than mag, i.e. above the RGB tip and consistent with thermal pulsing AGB stars that have become carbon stars as a result of third dredge-up.
Figure 28 presents a near-IR colour–colour diagram for AGB and RGB stars only, as other populations tend to confuse the picture in such diagram. Stars with mag are assumed to be on the RGB, whilst stars with mag are considered to be on the AGB (note that this selection excludes some of either type but maximises the purity of these two samples). For both selections we only kept stars with photometric errors on the colour mag. The M-type and carbon stars classified by Rowe et al. are highlighted. Some of the M-type, but especially carbon stars follow part of a sequence towards red colours, but the reddest of these, at mag and mag have generally not been bright enough at optical wavelengths for spectral typing. These are dusty LPVs identified in our WFCAM survey. The realm of the RGB stars spreads out over colour but generally this happens in one but not both colours at once, suggesting blends and/or other photometric uncertainties are to be blamed.
5.3.4 Red Supergiant stars
Drout et al. (2012) identified RSGs (and yellow supergiants) in M 33 using the Hectospec multi-fiber spectrograph on the 6.5-m Multiple Mirror Telescope in two observing campaigns, in 2009 and 2010.
They divided their list in three different categories: those supergiants that were selected both photometrically and kinematically, assigned rank 1; those confirmed with just one method, assigned rank 2; and those which were not selected by either method and which are likely foreground dwarfs, assigned rank 3. They identified 189 rank-1 stars, 12 rank-2 stars and 207 rank-3 stars.
Our WFCAM survey detected 381 of the red stars in the survey by Drout et al. (93%), from which 186 rank-1 stars (98%) and 13 rank-2 stars. Of the rank-1 stars, 14 were found by us to be variable, as was one rank-2 star. Their RSGs clearly delineate a branch in the near-IR CMD (Fig. 29; see also Fig. 30). It is possible that some of the reddened sources are also RSGs but they will have been too faint for spectral typing and hence not been included in the work by Drout et al. (2012). Three of the variables in the top of the AGB branch may instead be massive AGB stars or super-AGB stars (Fig. 29).
5.3.5 The Luminous Blue Variable Var C
Humphreys et al. (2014) identified a small number of LBVs in M 33, one of which is called “Var C”. Since its discovery, a series of eruptions have been witnessed in this star: 1940–1953 (Hubble & Sandage 1953); 1964–1970 (Rosino & Bianchini 1973); and 1982–1993 (Humphreys et al. 1988; Szeifert et al. 1996). By 1998 it had returned to a minimum state (Burggraf et al. 2014), but soon another, shorter episode of maximum light was seen from 2001 until 2005 (Viotti et al. 2006; Clark et al. 2012). Currently it is in a hot quiescent stage (Humphreys et al. 2014).
Var C was detected in our WFCAM survey four, five and ten times in the J-, H- and K-band, respectively. The photometry is reliable since there is no neighbouring star bright enough to seriously affect the photometry. The mean magnitudes are mag, mag and mag; with mag it is one of the bluer confirmed variable stars in our list. Over the two years of our monitoring campaign with WFCAM, Var C has steadily diminished its brightness, though in 2007 it seems to have stabilised (Fig. 30).
5.3.6 Spitzer mid-IR variability survey
Five epochs of Spitzer Space Telescope imagery in the 3.6-, 4.5- and 8-m bands have been analysed by McQuinn et al. (2007), to identify variable stars using a smilar method to that we used ourselves.
The Spitzer images covered nearly a square degree, slightly larger than our WFCAM survey; out of 40 571 Spitzer sources, 2868 stars fall outside the WFCAM monitoring coverage. Among the stars that Spitzer detected within the region in common with our survey, 36 411 are also in our photometric catalogue, down to a little below the RGB tip (Fig. 31). Hence, the recovery rate is %. The recovery rate of the Spitzer survey for RSGs, bright AGB stars and dusty AGB star variables from our WFCAM survey is good. Blending is only problematic for stars within the central square kpc of M 33 (cf. Paper I), where the recovery rate decreases to %.
Among the 2923 variable stars identified with Spitzer, 985 were identified as variable stars also in our WFCAM survey. This corresponds to a recovery rate of 35% (excluding 113 stars that fall outside the WFCAM coverage), i.e. slightly higher than the recovery rate by the WFCAM survey of UIST variable stars in the central square kpc (cf. Section 3.1). On the other hand, 3661 of our WFCAM variable stars had not been identified as variable stars in the Spitzer survey. Both surveys do well in detecting variable dusty AGB stars; the WFCAM survey is also sensitive to fainter, less dusty variable red giants (Fig. 32) that were too faint for Spitzer.
Figure 33 shows a mid-IR CMD, with a well-populated sequence off from which a branch extends towards redder colours and fainter 3.6-m brightness starting around –16 mag and mag. This branch mostly comprises dust-enshrouded objects, with a high fraction of WFCAM variable stars. The WFCAM variables are plentiful also among the brighter 3.6-m sources in the diagram; these are massive AGB stars and RSGs. The M-type stars form a sequence of increasing 3.6-m brightness, with carbon stars located along part of this sequence. The near absence of M-type or carbon stars in the red branch of dust-enshrouded objects is due to the limited sensitivity of the optical spectral typing survey. The Padova isochrones seem to underestimate the 3.6-m brightness by a factor three or so, most notable in the missing of the 1-Gyr isochrone of the red branch.
5.3.7 Sources variable at 24 m
Montiel et al. (2014) have identified 24 sources in M 33 that are variable at a wavelength of 24 m, based on Spitzer data with the MIPS instrument. One of these is NGC 604 (see below); the other sources could be dusty evolved stars.
We find that VC6, 8, 9, 13, 16 21 and 23 (their nomenclature) are variable with high confidence and VC10, 17 and 20 probably (but with fewer near-IR epochs so less reliable). These are all red, with mag (Fig. 34). We do not detect variability in VC14 but its photometry – mag, mag – is consistent with being a RSG and thus confirms the identification by Montiel et al. with the M2-type 760-d semiregular variable star VHK 71 (van den Bergh, Herbst & Kowal 1975); VC17 is likely variable and its photometry – mag, mag – is consistent wiht it being a dusty RSG.
On the other hand, for VC7 we only find a moderately red star without detected variability; this is probably the counterpart of the optical variable star with a 143-d period with which Montiel et al. associated VC7, but not of the 24-m variable itself as the 10-m absorption in the latter appears incompatible with such rapid variability.
None of the 24-m variables with mag – which includes all of their far-IR detections – were identified by us as near-IR variables. If this is because they are in fact not dusty evolved stars then this would explain why they do not obey the evolved stars sequences in their [3.6]–[8.0] vs. [3.6]–[4.5] diagram; they may instead be young stellar objects.
VC1 in Montiel et al. (2014) corresponds to NGC 604, the second largest extra-galactic H ii region in the Local Group (after 30 Doradus in the LMC). It is a young star forming region with an age of 3–5 Myr (e.g., Wilson & Matthews 1995; Pellerin 2006). Located some 12 from the centre of M 33, it spans around 4.1 pc with a core–halo structure (Melnick 1980). NGC 604 has been studied throughout the electromagnetic spectrum, including radio (Churchwell & Grass 1999; Tosaki et al. 2007), infrared (Higdon et al. 2003), optical (Tenorio-Tagle et al. 2004), ultraviolet (Keel et al. 2004), and X-ray (Maíz-Apellániz et al. 2004). These studies were mostly concerned with the effect of the massive star formation on the surrounding ISM. Using WFCAM we have identified 12 near-IR variable stars within NGC 604, which are shown in figure 35.
In the near-IR CMD (Fig. 34), the sources within NGC 604 are neither particularly luminous nor red – in fact, most have mag. We suspect that this group of variables may include any or all of the following types of sources: (i) young early-type stars; (ii) stars whose photometry is affected by nearby stars and/or nebulosity; and (iii) a dusty carbon star not born in NGC 604 (at mag).
WFCAM on UKIRT was used to extend our near-IR monitoring survey of the Local Group spiral galaxy M 33, from the central square kpc to a square degree. K-band observations were complemented with occasional J- and H-band observations to provide colour information.
The photometric catalogue comprises 403 734 stars, among which 4643 stars display large-amplitude variability. Investigation of the lightcurves and location on the near-IR CMD with respect to theoretical models of stellar evolution indicate that most of these stars are AGB stars or RSGs. They are concentrated towards the centre of M 33, more so than RGB stars. The majority of very dusty stars which are heavily reddened even at IR wavelengths are variable.
Cross-matching with optical monitoring campaigns and mid-IR variability searches conducted with Spitzer, shows that the UKIRT/WFCAM catalogue of variable stars is vastly more complete for the dusty variables than the optical surveys, and more complete for less dusty variables than the Spitzer surveys. Our catalogue is made publicly available at the CDS.
This work forms the basis for the next two papers in this series, to derive the SFH (Paper V) and to measure the rate and location of dust production (Paper VI) across M 33.
We thank the staff at UKIRT for their excellent support of this programme, and the referee for her/his constructive report. JvL thanks the School of Astronomy at IPM, Tehran, for their hospitality during his visits. We are grateful for financial support by The Leverhulme Trust under grant No. RF/4/RFG/2007/0297, by the Royal Astronomical Society, and by the Royal Society under grant No. IE130487.
- [Benjamin et al.(2005)] Benjamin R. A., et al., 2005, ApJ, 630, L149
- [Bonanos et al.(2006)] Bonanos A. Z. et al., 2006, ApJ, 652, 313
- [Burggraf et al.(2014)] Burggraf B., et al., 2014, A&A, submitted
- [Clark et al.(2012)] Clark J. S., et al., 2012, A&A, 541, A146
- [Churchwell & Goss (1999)] Churchwell E., & Goss W. M., 1999, ApJ, 514, 188
- [Cioni et al.(2008)] Cioni M.-R. L., et al., 2008, A&A, 487, 131
- [Deul & van der Hulst (1987)] Deul E. R., van der Hulst J. M., 1987, A&AS, 67, 509
- [Drout, Massey & Meynet (2012)] Drout M., Massey P., Meynet G., 2012, ApJ, 750, 97
- [Engargiola et al.(2003)] Engargiola G., Plambeck R. L., Rosolowsky E., Blitz L., 2003, ApJ, 149, 343
- [Girardi et al.(2005)] Girardi L., Groenewegen M. A. T., Hatziminaoglou E., da Costa L., 2005, A&A, 436, 895
- [Gullieuszik et al.(2007)] Gullieuszik M., Rejkuba M., Cioni M. R., Habing H. J., Held E. V., 2007, A&A, 475, 467
- [Hartman et al.(2006)] Hartman J. D., Bersier D., Stanek K. Z., Beaulieu J.-P., Kałuny J., Marquette J.-B., Stetson P. B., Schwarzenberg-Czerny A., 2006, MNRAS, 371, 1405
- [Higdon et al.(2003)] Higdon S. J. U., Higdon J. L., van der Hulst J. M., Stacey G. J., 2003, ApJ, 592, 161
- [Hodierna (1654)] Hodierna G. B., 1654, De Systemate Orbis Cometici, Deque Admirandis Coeli Characteribus (About the systematics of the cometary orbit, and about the admirable objects of the sky), Palermo
- [Hubble & Sandage (1953)] Hubble E., Sandage A., 1953, ApJ, 118, 353
- [Humphyers et al.(1988)] Humphreys R. M., Leitherer C., Stahl O., Wolf B., Zickgraf F.-J., 1988, A&A, 203, 306
- [Iben & Renzini (1983)] Iben I. Jr., Renzini A., 1983, ARA&A, 21, 271
- [Javadi, van Loon & Mirtorabi (2011a)] Javadi A., van Loon J. Th., Mirtorabi M. T., 2011a, MNRAS, 411, 263 (Paper I)
- [Javadi, van Loon & Mirtorabi (2011b)] Javadi A., van Loon J. Th., Mirtorabi M. T., 2011b, MNRAS, 414, 3394 (Paper II)
- [Javadi, van Loon & Mirtorabi (2011c)] Javadi A., van Loon J. Th., Mirtorabi M. T., 2011c, in: Why Galaxies Care About AGB Stars II, eds. F. Kerschbaum, T. Lebzelter & R. F. Wing, ASPC, 445, 497
- [Javadi et al.(2013)] Javadi A., van Loon J. Th., Khosroshahi H., Mirtorabi M. T., 2013, MNRAS, 432, 2824 (Paper III)
- [Keel et al.(2004)] Keel W. C., Holberg J. B., Treuthardt P. M., 2004, AJ, 128, 211
- [Levesque (2010)] Levesque E. M., 2010, in: Hot and Cool – Bridging Gaps in Massive Star Evolution, eds. C. Leitherer, P. Bennett, P. Morris and J. Th. van Loon, ASPC (San Francisco: ASP), 425, p103
- [Levesque et al.(2005)] Levesque E. M., Massey P., Olsen K. A. G., Plez B., Josselin E., Maeder A., Meynet G., 2005, ApJ, 628, 973
- [Macri et al.(2001)] Macri L. M., Stanek K. Z., Sasselov D. D., Krockenberger M., Kałuny J., 2001, AJ, 121, 861
- [Maíz-Apellániz et al.(2004)] Maíz-Apellániz J., Pérez E., Mas-Hesse J. M., 2004, AJ, 128, 1196
- [Marigo et al.(2008)] Marigo P., Girardi L., Bressan A., Groenewegen M. A. T., Silva L., Granato G. L., 2008, A&A, 482, 883
- [Massey et al.(2006)] Massey P., Olsen K. A. G., Hodge, P. W., Strong S. B., Jacoby G. H., Schlingman W., Smith R. C., 2006, AJ, 131, 2478
- [Melnick (1980)] Melnick J., 1980, A&A, 86, 304
- [McQuinn et al.(2007)] McQuinn K. B. W., et al., 2007, ApJ, 664, 850
- [Messier (1771)] Messier C., 1771, Mem. Acad., p448
- [Mochejska et al.(2001a)] Mochejska B. J., Kałuny J., Stanek K. Z., Sasselov D. D., Szentgyorgyi A. H., 2001a, AJ, 121, 2032
- [Mochejska et al.(2001b)] Mochejska B. J., Kałuny J., Stanek K. Z., Sasselov D. D., Szentgyorgyi A. H., 2001b, AJ, 122, 2477
- [Montiel et al.(2014)] Montiel E. J., Srinivasan S., Clayton G. C., Engelbracht C. W., Johnson C. B., AJ, in press (arXiv:1411.6008)
- [Pellerin (2006)] Pellerin A., 2006, AJ, 131, 849
- [Pierce, Jurcević & Crabtree (2000)] Pierce M. J., Jurcević J. S., Crabtree D., 2000, MNRAS, 313, 271
- [Pietsch et al.(2004)] Pietsch W., Misanovic Z., Haberl F., Hatzidimitriou D., Ehle M., Trinchieri G., 2004, A&A, 426, 11
- [Rosino & Bianchini (1973)] Rosino L., Bianchini A., 1973, A&A, 22, 453
- [Rosolowski et al.(2003)] Rosolowski E., Engargiola G., Plambeck R., Blitz L., 2003, ApJ, 599, 258
- [Skrutskie et al.(2006)] Skrutskie M. F., et al., 2006, AJ, 131, 1163
- [Stetson (1987)] Stetson P. B., 1987, PASP, 99, 191
- [Stetson (1993)] Stetson P. B., 1993, in: Stellar Photometry – Current Techniques and Future Developments, eds. C. J. Butler and I. Elliott, IAU Coll. Ser. 136 (Cambridge: Cambridge University Press), p291
- [Stetson (1996)] Stetson P. B., 1996, PASP, 108, 851
- [Szeifert et al.(1996)] Szeifert T., Humphreys R. M., Davidson K., Jones T. J., Stahl O., Wolf B., Zickgraf F.-J., 1996, A&A, 314, 131
- [Rowe et al.(2005)] Rowe J. F., Richer H. B., Brewer J. P., Crabtree D. R., 2005, AJ, 129, 729
- [Tenorio-Tagle et al.(2000)] Tenorio-Tagle G., Muñoz-Tuñon C., Pérez E., Maíz-Apellániz J., Medina-Tanco G., 2000, ApJ, 541, 720
- [Thompson et al.(2009)] Thompson T. A., Prieto J. L., Stanek K. Z., Kistler M. D., Beacom J. F., Kochanek C. S., 2009, ApJ, 705, 1364
- [Tosaki et al.(2007)] Tosaki T., Miura R., Sawada T., Kuno N., Nakanishi K., Kohno K., Okumura S. K., Kawabe R., 2007, ApJ, 664, L27
- [van den Bergh, Herbst & Kowal (1975)] van den Bergh S., Herbst E., Kowal C. T., 1975, ApJS, 29, 303
- [van Loon (2010)] van Loon J. Th., 2010, in: Hot and Cool – Bridging Gaps in Massive Star Evolution, eds. C. Leitherer, P. Bennett, P. Morris and J. Th. van Loon, ASPC (San Francisco: ASP), 425, p279
- [van Loon et al.(1997)] van Loon J. Th., Zijlstra A. A., Whitelock P. A., Waters L. B. F. M., Loup C., Trams N. R., 1997, A&A, 325, 585
- [van Loon et al.(1999)] van Loon J. Th., et al., 1999, A&A, 351, 559
- [van Loon et al.(2005)] van Loon J. Th., Cioni M.-R. L., Zijlstra A. A., Loup C., 2005, A&A, 438, 273
- [van Loon et al.(2007)] van Loon J. Th., van Leeuwen F., Smalley B., Smith A. W., Lyons N. A., McDonald I., Boyer M. L., 2007, MNRAS, 382, 1353
- [van Loon et al.(2008)] van Loon J. Th., Cohen M., Oliveira J. M., Matsuura M., McDonald I., Sloan G. C., Wood P. R., Zijlstra A. A., 2008, A&A, 487, 1055
- [Votti et al.(2006)] Viotti R. F., Rossi C., Polcaro V. F., et al., 2006, A&A, 458, 225
- [Whitelock et al.(2003)] Whitelock P. A., Feast M. W., van Loon J. Th., Zijlstra A. A., 2003, MNRAS, 342, 86
- [Wilson & Matthews (1995)] Wilson C. D., Matthews B. C., 1995, ApJ, 455, 125
- [Wood (1998)] Wood P. R., 1998, A&A, 338, 592
- [Wood et al.(1992)] Wood P. R., Whiteoak J. B., Hughes S. M. G., Bessell M. S., Gardner F. F., Hyland A. R., 1992, ApJ, 397, 552
- [Zaritsky, Elston & Hill (1989)] Zaritsky D., Elston R., Hill J. M., 1989, AJ, 97, 97