Variables In Globular Clusters Research Articles

Are most Cataclysmic Variables in Globular Clusters dynamically formed?

AbstractWe have been investigating populations of cataclysmic variables (CVs) in a set of more than 300 globular cluster (GC) models evolved with themoccacode.[-120pt] One of the main questions we have intended to answer is whether most CVs in GCs are dynamically formed or not. Contrary to what has been argued for a long time, we found that dynamical destruction of primordial CV progenitors is much stronger in GCs than dynamical formation of CVs. In particular, we found that, on average, the detectable CV population is predominantly composed of CVs formed via a typical common envelope phase (≳70 per cent). However, core-collapsed models tend to have higher fractions of bright CVs than non-core-collapsed ones, which suggests then that the formation of CVs is indeed slightly favoured through strong dynamical interactions in core-collapsed GCs, due to the high stellar densities in their cores.

The 82nd Name-List of Variable Stars. Part I - RA 0hr to 18hr, Novae and Globular-cluster Variables

The 82nd Name-List of Variable Stars. Part I - RA 0hr to 18hr, Novae and Globular-cluster Variables

Metal-poor Type II Cepheids with Periods Less Than Three Days

We have analyzed 10 high-resolution spectra of Type II Cepheids with periods less than 3 days. We find that they clearly separate into two groups: those with near or slightly below solar metallicities, and those with [Fe/H] between −1.5 and −2.0. While the former are usually called BL Her stars, we suggest that the latter be called UY Eri stars. The UY Eri subclass appears to be similar to the short period variables in globular clusters of the Galactic Halo. Globular clusters with [Fe/H] > −1.0 almost never have Type II Cepheids.

MOCCA-SURVEY database I. Accreting white dwarf binary systems in globular clusters – I. Cataclysmic variables – present-day population

In this paper, which is the first in a series of papers associated with cataclysmic variables and related objects, we introduce the CATUABA code, a numerical machinery written for analysis of the MOCCA simulations, and show some first results by investigating the present-day population of cataclysmic variables in globular clusters. Emphasis was given on their properties and the observational selection effects when observing and detecting them. In this work we analysed in this work six models, including three with Kroupa distributions of the initial binaries. We found that for models with Kroupa initial distributions, considering the standard value of the efficiency of the common envelope phase adopted in BSE, no single cataclysmic variable was formed only via binary stellar evolution, i. e., in order to form them, strong dynamical interactions have to take place. We show and explain why this is inconsistent with observational and theoretical results. Our results indicate that the population of cataclysmic variables in globular clusters is, mainly, in the last stage of their evolution and observational selection effects can drastically change the expected number of observed cataclysmic variables. We show that the probability of observing them during the outbursts is extremely small and conclude that the best way of looking for cataclysmic variables in globular clusters is by searching for variabilities during quiescence, instead of during outbursts. For that, one would need a very deep observation which could reach magnitudes $\gtrsim$ 27 mag. Finally, we argue that cataclysmic variables in globular clusters are not necessarily magnetic.

A census of variability in globular cluster M 68 (NGC 4590)

We analyse 20 nights of CCD observations in the V and I bands of the globular cluster M68 (NGC 4590), using these to detect variable objects. We also obtained electron-multiplying CCD (EMCCD) observations for this cluster in order to explore its core with unprecedented spatial resolution from the ground. We reduced our data using difference image analysis, in order to achieve the best possible photometry in the crowded field of the cluster. In doing so, we showed that when dealing with identical networked telescopes, a reference image from any telescope may be used to reduce data from any other telescope, which facilitates the analysis significantly. We then used our light curves to estimate the properties of the RR Lyrae (RRL) stars in M68 through Fourier decomposition and empirical relations. The variable star properties then allowed us to derive the cluster's metallicity and distance. We determine new periods for the variable stars, and search for new variables, especially in the core of the cluster where our method performs particularly well. We detect an additional 4 SX Phe stars, and confirm the variability of another star, bringing the total number of confirmed variable stars in this cluster to 50. We also used archival data stretching back to 1951 in order to derive period changes for some of the single-mode RRL stars, and analyse the significant number of double-mode RRL stars in M68. Furthermore, we find evidence for double-mode pulsation in one of the SX Phe stars in this cluster. Using the different types of variables, we derived an estimate of the metallicity, [Fe/H]=$-2.07 \pm 0.06$ on the ZW scale, and 4 independent estimates of the distance modulus ($\mu_0 \sim$ 15.00 mag) for this cluster. Thanks to the first use of difference image analysis on time-series observations of M68, we are now confident that we have a complete census of the RRL stars in this cluster.

A Study of Double- and Multi-mode RR Lyrae Variables

AbstractWe present the results of our new study of known RR Lyrae variable stars. All observations available for these stars in the Catalina Surveys were analyzed, and double-mode variations were identified. We studied the Petersen diagram and the period distribution for the double-mode RR Lyrae variables in the Galactic field, pulsating in the first-overtone and fundamental modes. The double-peaked character of the period distribution was detected for Galactic RR(B) stars, corresponding to Oosterhoff’s classes of globular clusters, which indicates that the age and evolution stage of RR(B) stars in the field and RR Lyrae variables in globular clusters are probably the same. Besides, we discovered five RRC stars with two simultaneously excited non-radial pulsations (equidistant triplets).

A stellar population synthesis approach to the Oosterhoff dichotomy

We use color-magnitude diagram synthesis together with theoretical relations from non-linear pulsation models to approach the long-standing problem of the Oosterhoff dichotomy related to the distribution of the mean periods of fundamental RR Lyrae variables in globular clusters. By adopting the chemical composition determined from spectroscopic observations and a criterion to account for the hysteresis mechanism, we tuned age and mass-loss to simultaneously reproduce the morphology of both the turn-off and the Horizontal Branch of a sample of 17 globular clusters of the Milky Way and of nearby dwarf galaxies in the crucial metallicity range (-1.9<[Fe/H]<-1.4) where the Oostheroff transition is apparent. We find that the Oosterhoff dichotomy among Galactic globular clusters is naturally reproduced by models. The analysis of the relative impact of the various involved parameters indicates that the main responsibles of the dichotomy are the peculiar distribution of clusters in the age-metallicity plane and the hysteresis. In particular, there is a clear connection between the two main branches of the age-metallicity relation for Galactic globular clusters and the Oosterhoff groups. The properties of clusters' RR Lyrae belonging to other Oostheroff groups (OoInt and OoIII) are instead not well reproduced. While for OoIII clusters a larger helium abundance for a fraction of the cluster's stars can reconcile the model prediction with observations, some other parameter affecting both the Horizontal Branch morphology and the RR Lyrae periods is required to reproduce the behavior of OoInt clusters.

CV1 in the globular cluster M 22: confirming its nature through X-ray observations and optical spectroscopy

Context. Observations of cataclysmic variables in globular clusters appear to show a dearth of outbursts compared to those observed in the field. A number of explanations have been proposed, including low mass-transfer rates and/or moderate magnetic fields implying higher mass white dwarfs than the average observed in the field. Alternatively this apparent dearth may be simply a selection bias. Aims. We examine multi-wavelength data of a new cataclysmic variable, CV1, in the globular cluster M 22 to try to constrain its period and magnetic nature, with an aim at understanding whether globular cluster cataclysmic variables are intrinsically different from those observed in the field. Methods. We use the sub-arcsecond resolution of the Chandra ACIS-S to identify the X-ray counterpart to CV1 and analyse the X-ray spectrum to determine the spectral model that best describes this source. We also examine the low resolution optical spectrum for emission lines typical of cataclysmic variables. Cross correlating the Halpha line in each individual spectrum also allows us to search for orbital motion. Results. The X-ray spectrum reveals a source best-fitted with a high-temperature bremsstrahlung model and an X-ray unabsorbed luminosity of 1.8e32 erg/s (0.3-8.0 keV), which are typical of cataclysmic variables. Optical spectra reveal Balmer emission lines, which are indicative of an accretion disc. Potential radial velocity in the Halpha emission line is detected and a period for CV1 is proposed. Conclusions. These observations support the CV identification. The radial velocity measurements suggest that CV1 may have an orbital period of ~7 hours, but further higher resolution optical spectroscopy of CV1 is needed to unequivocally establish the nature of this CV and its orbital period.

Identifying Contaminated K-Band Globular Cluster RR Lyrae Photometry

ABSTRACTAcquiring near-infrared K-band (2.2 μm) photometry for RR Lyrae variables in globular clusters and nearby galaxies is advantageous, since the resulting distances are less impacted by reddening and metallicity. However, K-band photometry for RR Lyrae variables in M5, Reticulum, M92, ω Cen, and M15 display clustercentric trends. The Hubble Space Telescope (HST) Advanced Camera for Surveys (ACS) data imply that multiple stars in close proximity to RR Lyrae variables located near the cluster core, where the stellar density increases markedly, are generally unresolved in ground-based images. RR Lyrae variables near the cluster core appear to suffer from photometric contamination, thereby yielding underestimated cluster distances and biased ages. The impact is particularly pernicious, since the contamination propagates a systematic uncertainty into the distance scale, and hinders the quest for precision cosmology. The clustercentric trends are probably unassociated with variations in chemical composition, since an empirical K-band period-magnitude relation inferred from Araucaria/VLT (Very Large Telescope) data for RR Lyrae variables in the Sculptor dSph exhibits a negligible metallicity dependence: (0.059 ± 0.095) × [Fe/H]ZW, a finding that supports prior observational results. A future multiepoch, high-resolution near-infrared survey, analogous to the optical HST ACS Galactic Globular Cluster Survey, may be employed to establish K-band photometry for the contaminating stars discussed here.

Searching for Long-Period Variables in Globular Clusters: A Demonstration on NGC 1851 Using PROMPT

ABSTRACTWe demonstrate how a small, robotically controlled telescope can be used to monitor bright, long-period variable stars in dense stellar systems like Galactic globular clusters. Observations of NGC 1851 gathered with the No. 5 PROMPT 0.4 m telescope in BVRI yielded quality color-magnitude diagrams to well below the horizontal branch at V = 16.1 mag. We recovered many of the known RR Lyrae variables, clarified the nature of the three known bright variables in the cluster, detected two new long-period variables, and flagged seven more suspected variables. We describe methods that should yield good results in variable star searches and monitoring using this and other small telescopes.

XMM-Newton and Chandra Observations of Neutron Stars and Cataclysmic Variables in the Globular Cluster NGC 2808

AbstractWe report on XMM-Newton and Chandra observations of the globular cluster NGC 2808. We detect one quiescent low mass X-ray binary of the 3±1 expected, if these systems are formed through encounters, and we show evidence for the presence of 20±10 bright cataclysmic variables in the core with a luminosity above 4×1031 erg s−1. We also review the specific nature of cataclysmic variables in globular clusters with reference to recent VLT/FORS1 observations of a cataclysmic variable in M 22.

RR Lyrae variables in globular clusters and nearby galaxies

AbstractI point out that the Oosterhoff dichotomy for globular cluster and field RR Lyrae stars may place the strongest constraints so far on the number of dwarf spheroidal-like protogalactic fragments that may have contributed to the formation of the Galactic halo. The first calibration of the RR Lyrae period-luminosity relation in I, J, H, K taking evolutionary effects into account is provided. Problems in the interpretation of RR Lyrae light curves and evolutionary properties are briefly reviewed.

AGB variables in globular clusters

AbstractWith a large number of newly detected red variables in globular clusters we construct a log P – K diagram of these stars and compare it with findings from the LMC. We show the existence of a relation between the location in log P – K diagram and the velocity amplitude in the near infrared.

Metallicities for Double-Mode RR Lyrae Stars in the Large Magellanic Cloud

Metallicities for six double-mode RR Lyrae stars (RRd's) in the Large Magellanic Cloud have been estimated using the ΔS method. The derived [Fe/H] values are in the range [Fe/H] = -1.09 to -1.78 (or -0.95 to -1.58, adopting a different calibration of [Fe/H] versus ΔS). Two stars in our sample are at the very metal-rich limit of all RRd's for which metal abundance has been estimated, either by direct measure (for field objects) or on the basis of the hosting system (for objects in globular clusters or external galaxies). These metal abundances, coupled with mass determinations from pulsational models and the Petersen diagram, are used to compare the mass-metallicity distribution of field and cluster RR Lyrae variables. We find that field and cluster RRd's seem to follow the same mass-metallicity distribution, within the observational errors, strengthening the case for uniformity of properties between field and cluster variables. At odds to what is usually assumed, we find no significant difference in mass for RR Lyrae variables in globular clusters of different metallicity and Oosterhoff types, or there may even be a difference contrary to the commonly accepted one, depending on the metallicity scale adopted to derive the masses. This unusual result for the mass-metallicity relation is probably due, at least in part, to the inclusion of updated opacity tables in the computation of metal-dependent pulsation models.

Oscillation Patterns of SX Phœnicis Variables in Globular Clusters

AbstractBased on new analyses of light curves of SX Phœnicis variables in globular clusters and on results taken from the literature, an overview of oscillation patterns in this subgroup of δ Scuti variables is given. Belonging to the Blue Straggler stars, they are expected to have a more complicated life history than “normal” δ Scuti stars in the field or in open clusters. The overall picture is that the patterns found in this SX Phe group seem to be similar to the patterns that are well established for standard field δ Set stars.

Abundance Analyses of Field RV Tauri Stars. IV. AD Aquilae, DS Aquarii, V360 Cygni, AC Herculis, and V453 Ophiuchi

Abundance analyses are presented and discussed for five RV Tauri variables. Three stars—DS Aqr, V360 Cyg, and V453 Oph—are RV C stars by spectroscopic classification, i.e., metal lines are weak. They are shown to be metal poor with [Fe/H] from -1.0 to -2.2 with normal relative abundances of other elements. By contrast, AD Aql and AC Her are RV B stars with an odd abundance pattern: elements that condense into grains at a high temperatures are underabundant (i.e., [Fe/H] = -2.1 for AD Aql) but elements with a low condensation temperatures are much less underabundant (i.e., [S/H] = 0.0 and [Zn/H] = -0.1 for AD Aql). This abundance pattern is ascribed to a separation of dust and gas in the upper atmosphere of the star. The present analyses with previously published results are used to investigate the systematics of the dust-gas separation in RV Tauri variables. The process is apparently inoperative in stars with an initial metallicity of about [Fe/H] -1.0; RV C stars and similar variables in globular clusters are immune to the dust-gas separation. The process achieves more severe effects in RV B than in RV A stars. The strength of the abundance anomalies attributed to dust-gas separation is not correlated with reported infrared excesses. After correction for the effects of the dust-gas separation, there is no strong evidence from the abundances that evolution along the AGB and experience of the third dredge-up preceded the formation of the majority of the RV Tauri variables.

RV Tau Variables in Globular Clusters: Clues on Their Nature?

AbstractWe report on the detailed elemental abundances of RV Tau variables that reside in globular clusters, and how they compare with those of RV Tau variables in the field. Surprisingly, they seem to show definite differences, and at this stage, it is not clear why this is so.

The pulsation, temperatures and metallicities of Mira and semiregular variables in different stellar systems

Data on angular diameters and infrared photometry for late-type stars are assembled. It is shown that a consistent Teff scale can be established, combining results for Mira and non-Mira M-type stars. The log Teff versus (J- K) relation is much steeper than previously adopted, but is consistent with predictions from model stellar atmospheres. Comparison of the linear diameters of Miras measured in the red spectral region with those measured in the infrared shows that models successfully predict the extension observed in the red, and the combined data provide strong evidence that Miras are pulsating in their first overtone. Data on Miras and semiregular (SR) variables in globular clusters are compared with predictions from stellar evolution and pulsation theory. These data also support a steep log Teff versus (J- K) relation at low temperatures. The Miras and SR variables in 47 Tue conform to theoretical expectation if they are undergoing an average mass loss of ∼ 3 ×, 10−7 M⊙ yr−1. SR variables in both metal-rich and metal-poor globular clusters are probably pulsating, like the Miras in their first overtone. The general agreement between observation and theory now found suggests that infrared colour-period relations can be used to investigate overall metallicity differences between Miras in different stellar systems, at least at the shorter periods where circumstellar extinction is probably negligible. A comparison of Miras in Galactic globular clusters of known metallicity with those in the LMC and in the SgrI window of the Galactic Bulge indicates that Miras of periods 100 to 300 d in the LMC have a mean metallicity log z ∼ −0.6, whilst those in SgrI have log z ∼ −0.2, close to that of K giants in the NGC 6522 Bulge window. No evidence has yet been found for a dependence of the Mira period-luminosity relation on metallicity, and it is pointed out that theory does not at present give a definitive prediction of such an effect. Some stars of special interest are discussed in an appendix.

GCVS Problems with Flare Stars

The astrometric aspect of cataloguing variable stars becomes more and more important. During the past thirty years there has been approximately an three-fold increase of the number of known variables; now the rate of new discoveries becomes extremely high, and we may expect a new twofold increase in a couple of years due to discoveries made with space-borne instruments. To avoid confusion, we should continue our policy of naming new variables, and this is impossible without checking identifications with known variable stars, requiring good coordinates and/or finding charts. Operation of automatic telescopes also needs excellent coordinates just for pointing. For all these reasons, we are now working on improving considerably the accuracy standard of variable star coordinates, providing at least reasonably good coordinates for extragalactic variables (GCVS, 4th Edition, Vol. V, now in press, gives coordinates mainly to a second of arc) and variables in globular clusters, identifying variables, including those with rough published positions, with positional catalogues, etc.Flare stars are very important variables; potentially they are one of the most abundant types of variable stars. Their identification in minimum light may be rather complicated, especially for dense regions of star clusters and associations. Most regretfully, there are many cases of papers on flare stars introducing additional confusion connected with wrong designations, wrong coordinates, missing finding charts. Thus, there appear problems preventing us from incorporating some flare stars in the GCVS. Many definite variables remain in the status of “suspects”, being positionally close to known variables, and no finding chart being available either for the earlier known variable or for the new one. Sometimes we do include such a “pretender” into the GCVS, and years after a finding chart or good coordinates appear, and it turns out that we have two entries for the same star (see, for instance, numerous cases of duplicated names for flare stars in Taurus in the GCVS Vol. III).

Predictions of a population of cataclysmic variables in globular clusters

We have studied the number of cataclysmic variables (CVs) that should be active in globular clusters during the present epoch as a result of binary formation via two-body tidal capture. We predict the orbital period and luminosity distributions of CVs in globular clusters. The results arebased on Monte Carlo simulations combined with evolution calculations appropriate to each system formed during the lifetime of two specific globular clusters, omega Cen and 47 Tuc. From our study of these two clusters, which represent the range of core densities and states of mass segregation that are likely to be interesting, we extrapolate our results to the Galactic globlular cluster system. Although there is at present little direct observational evidence of CVs in globular clusters, we find that there should be a large number of active systems. We predict that there should be more than approximately 100 CVs in both 47 Tuc and omega Cen and several thousand in the Galactic globular cluster system. These numbers are based on two-body processes alone and represent a lower bound on the number of systems that may have been formed as a result of stellar interaction within globular clusters. The relation between these calculations and the paucity of optically detected CVs in globular clusters is discussed. Should future observations fail to find convincing evidence of a substantial population of cluster CVs, then the two-body tidal capture scenario is likely to be seriously constrained. Of the CVs we espect in 47 Tuc and omega Cen, approximately 45 and 20, respectively, should have accretion luminosities above 10(exp 33) ergs/s. If one utilizes a relation for converting accretion luminosity to hard X-ray luminosity that is based on observations of Galactic plane CVs, even these sources will not exhibit X-ray luminosities above 10(exp 33) ergs/s. While we cannot account directly for the most luminous subset of the low-luminosity globular cluster X-ray sources without assuming an evolutionary pattern that is different from that of the majority of CVs in the disk, we are able to account for all of the observed lower luminosity subset of these sources, many of which have been recently discovered through ROSAT observations. In order for our predicted integrated cluster X-ray luminosities to be consistent with observational upper limits, the relation between accretion and X-ray luminosities should be something like that inferred from the Galactic plane population of CVs. Our calculations predict a large number of systems with L(sub acc) is less than 10(exp 32) ergs/s. Although our calculations imply that globular clusters should have an enhancement of CVs relative to the number thought to be present in the Galactic disk, this enhancement is at most roughly an order of magnitude, not comparable to the factor of approximately 100 for low-mass X-ray binaries (LMXBs).