The Omegawhite Survey for Short-Period Variable Stars I: Overview and First Results
Total Page:16
File Type:pdf, Size:1020Kb
Mon. Not. R. Astron. Soc. 000, 1{15 (2015) Printed 12 September 2018 (MN LaTEX style file v2.2) The OmegaWhite Survey for Short-Period Variable Stars I: Overview and First Results S.A Macfarlane1;2?, R. Toma3, G. Ramsay3, P.J Groot1, P.A Woudt2, J.E Drew4, G. Barentsen4, J. Eisl¨offel5, 1Department of Astrophysics/IMAPP, Radboud University, P.O. Box 9010, 6500 GL Nijmegen, The Netherlands 2Astrophysics,Cosmology and Gravity Centre, Department of Astronomy, University of Cape Town, Private Bag X3, Rondebosch 7701, South Africa 3Armagh Observatory, College Hill, Armagh, BT61 9DG, Northern Ireland 4School of Physics, Astronomy & Mathematics, University of Hertfordshire, College Lane, Hatfield, Hertfordshire, AL10 9AB, U.K. 5Th¨uringerLandessternwarte, Sternwarte 5, D-07778 Tautenburg, Germany 12 September 2018 ABSTRACT We present the goals, strategy and first results of the OmegaWhite survey: a wide- field high-cadence g-band synoptic survey which aims to unveil the Galactic population of short-period variable stars (with periods < 80 min), including ultracompact binary star systems and stellar pulsators. The ultimate goal of OmegaWhite is to cover 400 square degrees along the Galactic Plane reaching a depth of g = 21.5 mag (10σ), using OmegaCam on the VLT Survey Telescope (VST). The fields are selected to overlap with surveys such as the Galactic Bulge Survey (GBS) and the VST Photometric Hα Survey of the Southern Galactic Plane (VPHAS+) for multi-band colour information. Each field is observed using 38 exposures of 39 s each, with a median cadence of ∼2.7 min for a total duration of two hours. Within an initial 26 square degrees, we have extracted the light curves of 1.6 million stars, and have identified 613 variable candi- dates which satisfy our selection criteria. Furthermore, we present the light curves and statistical properties of 20 sources which have the highest-likelihood of being variable stars. One of these candidates exhibits the colours and light curve properties typically associated with ultracompact AM CVn binaries, although its spectrum exhibits weak Balmer absorption lines and is thus not likely to be such a binary system. We also present follow-up spectroscopy of five other variable candidates, which identifies them as likely low-amplitude δ Sct pulsating stars. Key words: surveys { binaries: close { Galaxy:bulge { methods: observational { methods: data analysis { techniques: photometric. arXiv:1508.06277v1 [astro-ph.SR] 25 Aug 2015 1 INTRODUCTION nova outbursts (e.g. supernova discoveries in the Palomar Transient Factory (PTF), Law et al. 2009). Time domain astrophysics has been transformed over the last decade. Whereas in the past, photometric data of indi- In the field of Galactic binary research, one key goal has vidual objects were painstakingly obtained using high speed been to discover individual systems with astrophysically in- photometry or dedicated long term projects, now a whole teresting properties. Of particular interest are ultracompact series of synoptic projects have been developed to observe binaries (UCBs) which have an orbital period (Porb) of 6 large areas of sky over short time-scales giving photometric 70 min, implying that the secondary star cannot be a main- data on thousands or millions of objects. The diversity of sequence star (Rappaport et al. 1982). Furthermore, these goals of these projects is considerable, ranging from detect- hydrogen-deficient objects are predicted to be the strongest ing transiting exo-planets (e.g. the main aim of the `Super- known sources of gravitational wave radiation (GWR) in the WASP' project, Pollacco et al. 2006) to discovering super- passband of the satellite observatory eLISA (Amaro-Seoane et al. 2013; Roelofs et al. 2007), and as such are important calibrators that provide verification of the existence and de- ? Email: [email protected] tectability of GWR. Moreover, the evolution of these binary c 2015 RAS 2 S. Macfarlane et al. Table 1. Overview of recent high-cadence synoptic surveys Survey Field Location Total Sky Coverage Depth Variability sensitivity (deg) (deg2)(V mag) (min) a OmegaWhite jbj 6 10 400 21.5 > 6 b RATS jbj 6 30 46 22.5 > 5 c RATS-Kepler 6 6 jbj 6 21 49 22.5 > 5 d FSVS jbj > 20 23 24.0 > 24 DLSe jbj < 10 21 25.5 > 15 f Kepler 15 6 jbj 6 25 116 20.0 > 1 or > 30 SuperWASPg all sky all sky 15.0 > 10 aThis paper, bRapid Temporal Survey (Ramsay & Hakala 2005), c(Ramsay et al. 2014), dFaint Sky Variability Survey (Groot et al. 2003), Deep Lens Survey e(Becker et al. 2004), f (Borucki et al. 2010), gWide Angle Search for Planets (Pollacco et al. 2006) systems is influenced by the emission of GWR in addition periods, rapidly oscillating Ap or Am star systems (roAp, to the mass transfer phase. Therefore, the study of UCBs roAm), and pulsating white dwarfs (ZZ Ceti variables) ex- will also help to answer key questions of late-stage binary hibit periodic variations on amplitudes ranging from a per- evolution. Earlier estimates of their intrinsic numbers sug- cent or less up to several tens of percent on timescales of a gested a relatively high foreground contribution from UCBs few to tens of minutes. to the gravitational wave signal from merging supermassive δ Sct stars have A{F spectral types and are known black holes within the eLISA band. However, in a series of to have short pulsation periods (typically 0.02 - 0.25 days, papers using SDSS and PTF data (Roelofs et al. 2009; Rau Chang et al. 2013), with absolute magnitudes 3 - 5 mag et al. 2010; Carter et al. 2013; Levitan et al. 2015), it is fainter than Cepheids. However, they are believed to be the now clear that the predicted number density of AM CVn bi- second most common variable stars in the Galaxy (Breger naries (semi-detached and mass transferring UCBs) in the 1979), and their short period makes them relatively easy to Solar neighbourhood is 5 ± 3 × 10−7pc−3, a factor 50 lower detect in high-cadence surveys (e.g. Ramsay & Hakala 2005). than previous estimates by Nelemans et al. (2001). Other δ Sct stars can be used as precise distance tracers if the surveys which have recently had success in discovering new fundamental radial pulsation mode can be identified (e.g. AM CVn systems include the Catalina Real-Time Transient Petersen & Christensen-Dalsgaard 1999; McNamara et al. Survey (CRTS Drake et al. 2009), and the All Sky Auto- 2007). For high-amplitude δ Sct stars (HADS), where the mated Survey for Supernovae (ASAS-SN, e.g. Wagner et al. pulsation amplitude is greater than 0.3 mag, it is expected 2014). that the dominant period will be due to the fundamental or The emission line method for finding AM CVn systems, first-overtone radial mode. The shortest period δ Sct stars on which the SDSS work is based, is most sensitive to sys- in the Galactic field have a dominant pulsation period of tems with Porb > 30 min. More recently, the PTF survey ∼26 min (Kim et al. 2010). has been identifying outbursting AM CVns which have or- One survey that set out to discover short-period systems bital periods shorter than this, but still greater than 22 min was the Rapid Temporal Survey (RATS, Ramsay & Hakala (Levitan et al. 2015). From the total number of 43 known 2005; Barclay et al. 2011). RATS was carried out using the AM CVns, only 6 have periods shorter than 20 minutes. It 2.5-m Isaac Newton Telescope (INT) on La Palma and the is those systems with the shortest orbital period (5 min < 2.2-m Max Planck Gesellschaft telescope (MPG/ESO) be- Porb < 20 min) which are predicted to be the strongest tween 2003{2010, and covered a total of 46 square degrees emitters of gravitational waves. Determining their popula- favouring fields near the Galactic plane. The fact that no tion size is important for both the development of eLISA AM CVn was discovered was (in retrospect) not a surprise and for deducing the relative importance of the three pos- since the latest space densities imply that only a few sys- tulated formation channels of these binaries (for a detailed tems (at best) are expected in a survey covering this area. review, see Solheim 2010). A way to identify those AM CVn However, the survey did discover many new and interesting stars with the shortest orbital periods is through their pho- short-period variables, including a possible hybrid sdBV pul- tometric behaviour as they show a periodic modulation on, sator exhibiting long-period g-modes (Ramsay et al. 2006) or close to, the orbital period (e.g. the first PTF AM CVn and a dwarf nova discovered through its high-amplitude discovered, Levitan et al. 2011). Variations are expected due quasi-periodic oscillations (QPOs) in quiescence (Ramsay to eclipses, ellipsoidal variations, irradiation, superhumps, et al. 2009). or anisotropic disc hotspot emission in the system. The de- The OmegaWhite Survey, which is now being conducted tection of these especially short-period systems is the main using OmegaCam on the VST, has a similar observing strat- motivation behind the OmegaWhite survey. egy to RATS (See Table 1 for a comparison of similar stel- However, short time-scale photometric variations can lar wide-field surveys). However, OmegaWhite aims to cover also originate from physical changes within the internal a much wider area (400 square degrees) at lower average structure or atmosphere of the source. Examples of such Galactic latitudes (jbj < 5 degrees). In this paper we present: sources include flare stars or fast pulsating stars such as δ the OmegaWhite observing strategy, the reduction pipeline, Scuti stars (δ Sct, Breger 2000) or SX Phoenicis variables our method for identifying objects of interest, an outline of (e.g.