Binarity and multiperiodicity in high-amplitude delta Scuti stars
A. Derekas, L.L. Kiss, T.R. Bedding, M.C.B. Ashley, B. Csak, A. Danos, J.M. Fernandez, G. Furesz, Sz. Meszaros, Gy.M. Szabo, R. Szakats, P. Szekely, K. Szatmary
aa r X i v : . [ a s t r o - ph ] D ec Mon. Not. R. Astron. Soc. , 000–000 (0000) Printed 28 October 2018 (MN L A TEX style file v2.2)
Binarity and multiperiodicity in high-amplitude δ Scuti stars
A. Derekas ⋆ , L. L. Kiss , T. R. Bedding , M. C. B. Ashley , B. Cs´ak , , A. Danos ,J. M. Fernandez , G. F ˝ur´esz , Sz. M´esz´aros , Gy. M. Szab ´o , R. Szak´ats , P. Sz´ekely ,K. Szatm´ary Sydney Institute For Astronomy, School of Physics, University of Sydney, NSW 2006, Australia School of Physics, Department of Astrophysics and Optics, University of New South Wales, Sydney, NSW 2052, Australia Department of Optics and Quantum Electronics, University of Szeged, Hungary Max Planck Institute for Astronomy, K¨onigstuhl 17, D-69117 Heidelberg, Germany Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138, USA Department of Experimental Physics and Astronomical Observatory, University of Szeged, Szeged, D´om t´er 9, 6720 Hungary
Accepted ... Received ..; in original form ..
ABSTRACT
We have carried out a photometric and spectroscopic survey of bright high-amplitude δ Scuti (HADS) stars. The aim was to detect binarity and multiperiodicity (or both) in orderto explore the possibility of combining binary star astrophysics with stellar oscillations. Herewe present the first results for ten, predominantly southern, HADS variables. We detected theorbital motion of RS Gru with a semi-amplitude of ∼ − and 11.5 days period. Thecompanion is inferred to be a low-mass dwarf star in a close orbit around RS Gru. We foundmultiperiodicity in RY Lep both from photometric and radial velocity data and detected orbitalmotion in the radial velocities with hints of a possible period of 500–700 days. The data alsorevealed that the amplitude of the secondary frequency is variable on the time-scale of a fewyears, whereas the dominant mode is stable. Radial velocities of AD CMi revealed cycle-to-cycle variations which might be due to non-radial pulsations. We confirmed the multiperiodicnature of BQ Ind, while we obtained the first radial velocity curves of ZZ Mic and BE Lyn.The radial velocity curve and the O–C diagram of CY Aqr are consistent with the long-periodbinary hypothesis. We took new time series photometry on XX Cyg, DY Her and DY Peg,with which we updated their O–C diagrams. Key words: stars: variables: δ Scuti – binaries: general – binaries: spectroscopic – techniques:photometric – techniques: radial velocities – methods: observational δ Scuti stars are short-period pulsating variables of A-F spectraltypes, located at the intersection of the main sequence and the clas-sical instability strip in the Hertzsprung-Russell diagram. Typicalperiods are in the order of a few hours with amplitudes less than1 mag. A prominent group within the family comprises the high-amplitude δ Scuti stars (HADS), which have V-band amplitudeslarger than 0.3 mag. Population II members of the group are alsoknown as SX Phoenicis stars, often found in globular clusters (fora review of δ Scuti stars see e.g. Rodr´ıguez & Breger (2001)).HADS’s are the short-period counterparts of the classicalCepheids, excited by the κ -mechanism and pulsating in one or tworadial modes, usually in the fundamental and first-overtone modes(McNamara 2000). The data in Rodr´ıguez & Breger (2001) show ⋆ E-mail: [email protected] that there is no rapidly rotating HADS ( v sin i km s − ),suggesting an intimate relationship between the rotational stateand the excitation of pulsations. In recent years, the number ofHADS known with multimode oscillations has rapidly grown, hint-ing a new potential for asteroseismic studies of these objects (e.g.Poretti 2003; Poretti et al. 2005). Several investigations suggestedthat some of the stars may have non-radial pulsation modes present(McNamara 2000; Poretti et al. 2005). However, an important pa-rameter in modeling stellar oscillations is the mass of the star,which is usually constrained from evolutionary models. This willinevitably lead to great uncertainties in any kind of modeling at-tempts, so that independent mass estimates could be of paramountimportance. Thus, binary δ Scutis may play a key role in under-standing oscillations of these stars.There has been a great interest recently in δ Scutistars that reside in eclipsing binary systems (e.g. Kim et al.2003; Mkrtichian et al. 2006; Soydugan et al. 2006; c (cid:13) Derekas et al.
Table 1.
The list of programme stars. The asterisks mark multiperi-odic stars, for which this Table contains the dominant period only.References for the parameters are the following: (a) Rodr´ıguez et al.(1996); (b) Rodr´ıguez & Breger (2001); (c) Perryman et al. (1997b); (d)Kholopov et al. (1985-1988); (e) Szak´ats, Szab´o, & Szatm´ary (2008); (f)Blake et al. (2003); (g) Rodr´ıguez et al. (1995b); (h) Fu & Sterken (2003);(i) Derekas et al. (2003).Star Pop. a V max V min P (days) Obs.RY Lep* I 8 . m
20 9 . m
10 0.22514410 c I/sp.
AD CMi* I 9 . m
21 9 . m
51 0.12297443 d sp. BE Lyn I 8 . m
60 9 . m
00 0.09586952 e sp. DY Her I 10 . m
15 10 . m
66 0.14863135 d V XX Cyg II 11 . m
28 12 . m
13 0.13486511 f V BQ Ind* II b . m
78 10 . m
05 0.08200015 c I ZZ Mic* I 9 . m
27 9 . m
69 0.06718350 d BV/sp.
RS Gru I 7 . m
92 8 . m
51 0.14701131 g BV I/sp.
CY Aqr II 10 . m
42 11 . m
16 0.06103833 h BV I/sp.
DY Peg II 9 . m
95 10 . m
62 0.07292630 i V Pigulski & Michalska 2007; Christiansen et al. 2007). Currently,we know about 40 such δ Scutis, of which only one belongs to theHADS group (Christiansen et al. 2007). There are also a handful ofnon-eclipsing binary HADS (Rodr´ıguez & Breger 2001), usuallydeduced from the apparent cyclic period changes but with a fewexceptions, like the single-lined spectroscopic binary SZ Lyn. Thelow number of binary δ Scutis suggests a strong observational bias,because one needs accurate observations with a long time span todetect multiplicity unambiguously (Rodr´ıguez & Breger 2001).In this paper we present the results of our investigationsinto binarity and multiple periodicity in bright HADS variables.The sample was initially selected from the variable star cata-logue of the Hipparcos satellite, which contains 21 “SX Phe”type stars (Perryman et al. 1997b). Of these, here we discuss 9stars (and RY Lep in addition), i.e. our present sample containsalmost half of all known bright HADS. Using a wide range oftelescopes and instruments, we have been monitoring the targetstars over the last five years, extending the earlier studies byour group (Kiss & Szatm´ary 1995; Kiss et al. 2002; Derekas et al.2003; Szak´ats, Szab´o, & Szatm´ary 2008).The paper is organized as follows. The observations and thedata analysis are described in Sect. 2. The main discussion is inSect. 3, in which the results for individual stars are presented. Abrief summary is given in Sect. 4.
The observed stars and their main observational properties arelisted in Table 1. The full log of observations is given in Table A1in the Appendix. All data presented in this paper are available fordownload from the CDS, Strasbourg.Observations were carried out using seven different instru-ments at five observatories in Australia, Hungary and the USA ona total of 65 nights between 2003 October and 2008 July. In thefollowing we briefly describe the telescopes and the detectors usedin this project, in an order of increasing mirror size:–
Szeged Observatory, 0.4 m (Sz40) V -band CCD photometry of XX Cyg was carried out with the0.4 m Newtonian telescope at Szeged Observatory. The telescope Table 2.
New times of maximum (HJD-2400000).Star
HJD max
Filter Star
HJD max
FilterRS Gru 52920.0196 V CY Aqr 53334.9453 I RS Gru 52921.9311 V CY Aqr 53336.9592 I RS Gru 52922.0772 V CY Aqr 53337.9357 I RS Gru 52923.9905 V CY Aqr 54307.5293 V RS Gru 52925.0188 V CY Aqr 54308.5073 V CY Aqr 52920.9223 V XX Cyg 54307.4294 V CY Aqr 52920.9827 V XX Cyg 54309.4515 V CY Aqr 52921.0439 V XX Cyg 54677.3633 V CY Aqr 52923.0587 V DY Her 54304.4772 V CY Aqr 52926.9643 V DY Peg 54305.4731 V CY Aqr 52927.0258 V has an 11 ′ × ′ field of view and the detector was an SBIG ST-7CCD camera (765 ×
510 pixels at 9 µ m). We took observations forXX Cyg with exposure time of 45 s.– Siding Spring Observatory, 0.5 m (APT50)
Time-series CCD photometry was obtained for RY Lep, BQ Indand CY Aqr with the Automated Patrol Telescope (APT) at SidingSpring Observatory, which is owned and operated by the Univer-sity of New South Wales (UNSW). The telescope was originallya Baker-Nunn design converted into CCD imaging (Carter et al.1992) and has a three-element correcting lens and an f/1 spheri-cal primary mirror. The camera has an EEV CCD05-20 chip with770 × µ m) pixel to image a 2 × field of view. Weobtained I -band images with exposure times between 3 s and 60 s,depending on the star and the weather conditions.– Siding Spring Observatory, 24 ′′ /0.6 m (SSO60) Photoelectric photometry of ZZ Mic, RS Gru and CY Aqrwas obtained with the 0.6 m f/18 Cassegrain-reflector, on whicha single-channel photometer was mounted. It had a computer-controlled 8-hole filter wheel, for which the dwell times on eachfilter can be varied. The detector was a thermoelectrically cooledHamamatsu R647-4 photomultiplier tube with a 9 mm diameter bi-alkali (blue-sensitive) photocathode (Handler et al. 2000). We used B , V and I filters and the exposure time varied between 15 s and30 s, depending on the brightness of the observed star.– Piszk´estet˝o Station, 0.6 m (P60) V -band CCD photometry was obtained on CY Aqr, XX Cyg,DY Her and DY Peg with the 60/90/180 cm Schmidt-telescopemounted at the Piszk´estet˝o Station of the Konkoly Observatory.The detector was a Photometrics AT200 CCD camera (1536 × µ m pixels, FOV=28 ′ × ′ ).– Fred Lawrence Whipple Observatory, 1.5 m (MH150)
High resolution spectra were obtained on BE Lyn with the Till-inghast Reflection Echelle Spectrograph (TRES) and the 1.5 m tele-scope at the Fred Lawrence Whipple Observatory on Mt. Hopkins,Arizona. TRES is a high-throughput fiber-fed echelle. It is cross-dispersed, yielding a passband of 380-920 nm over the 51 spec-tral orders. It accommodates 3 optical fiber pairs (science+sky) ofdifferent diameters, to offer a match for seeing conditions. Simul-taneous ThAr calibration is also available via a separate fiber. Theavaliable resolutions are 64K, 35K and 31K, depending on the fibersize selected (1.5, 2.3 or 3.2 arcsec, respectively). The observationswere taken as part of the instrument commissioning, using the smallfiber and 300 s integration times. The 4.6k ×
2k detector was binned2 × c (cid:13) , 000–000 inarity and multiperiodicity in HADS variables – Siding Spring Observatory, 2.3 m (SSO230)
Spectroscopic observations were carried out for RY Lep,AD CMi, ZZ Mic, RS Gru and CY Aqr with the 2.3 m ANU tele-scope at the Siding Spring Observatory, Australia. All spectra weretaken with the Double Beam Spectrograph using the 1200 mm − gratings in both arms of the spectrograph. The projected slit widthwas 2 ′′ on the sky, which was about the median seeing during ourobservations. The spectra covered the wavelength ranges 4200–5200 ˚A in the blue arm and 5700–6700 ˚A in the red arm. The dis-persion was 0.55 ˚A px − , leading to a nominal resolution of about1 ˚A. The exposure time varied between 50 s and 180 s dependingon the observed star and the weather conditions.– Anglo-Australian Observatory, 3.9 m (AAT)
We used the 3.9 m Anglo-Australian Telescope equipped withthe UCLES spectrograph for 4.2 hours of high-resolution spec-troscopy of RS Gru. The observations were taken during servicetime. Our echelle spectra include 56 orders with a central wave-length of 6183 ˚A and a resolving power λ/ ∆ λ ≈
40 000. The ex-posure time was 300 s.All data were reduced with standard tools and procedures.Photoelectric photometry taken with the SSO60 instrument weretransformed to the standard system using the coefficients fromBerdnikov & Turner (2004). Uncertainties in the Fourier ampli-tudes were calculated following the considerations of Kjeldsen(2003). The BV standard magnitudes of comparison stars weretaken from Kharchenko (2001), while I standard magnitudes weretaken from the DENIS catalogue (The DENIS consortium 2005).The CCD observations were reduced in IRAF , including bias re-moval and flat-field correction utilizing sky-flat images taken dur-ing the evening or morning twilight. Differential magnitudes werecalculated with aperture photometry using two comparison stars ofsimilar brightnesses.Thanks to the dense sampling of the light curves, new times ofmaximum light for monoperiodic stars were easy to determine fromthe individual cycles. This was done by fitting fifth-order polynomi-als to the light curves around the maxima. We estimate the typicaluncertainty to be about ± S/N of each frequency was calculated following Breger et al.(1993).All spectra were reduced with standard tasks in IRAF. Re-duction consisted of bias and flat field corrections, aperture ex-traction, wavelength calibration, and continuum normalization. Wechecked the consistency of wavelength calibrations via the constantpositions of strong telluric features, which verified the stability ofthe system. Radial velocities were determined with the task fxcor ,applying the cross-correlation method using a well-matching the-oretical template spectrum from the extensive spectral library ofMunari et al. (2005). For consistency, the velocities presented inthis paper were all determined from a 50 ˚A region centered on theH α line. The high-resolution spectra for BE Lyn and RS Gru al-lowed us to compare hydrogen and metallic line velocities, which IRAF is distributed by the National Optical Astronomy Observatories,which are operated by the Association of Universities for Research inAstronomy, Inc., under cooperative agreement with the National ScienceFoundation. m a gn it ud e B-0.6VI+0.60.20.30.4 B - V ( m a g ) V -I ( m a g ) Figure 1.
Standard light and colour variations of RS Gru(E =2452920.0196; P=0.14701131 d). will be discussed in a subsequent paper. We made barycentric cor-rections to every radial velocity value. Depending on the signal-to-noise of the spectra, the estimated uncertainty of the radial veloci-ties ranged from 1–5 km s − . Radial velocities from the AAT andMH150 Echelle spectra have much better accuracy,
100 m s − . RS Gru (HD 206379; HIP 107231) is a monoperiodic HADS witha pulsation period of 0.147 d and a mean magnitude of ∼ − with a velocityamplitude of ∼
45 km s − . McNamara & Feltz (1976) obtained uvby β photometry and spectrographic data and determinedphysical parameters. van Citters (1976) acquired photoelectricradial velocity curves on two nights, while further photometricobservations were taken by Dean et al. (1977). New radial velocitymeasurements taken by Balona & Martin (1978a) showed unam-biguously the variation of the center-of-mass velocity, indicatingthe binary nature of RS Gru but the orbital period was notdetermined for nearly three decades. Further investigations weredone by Breger (1980); Andreasen (1983); McNamara (1985);Antonello et al. (1986); Garrido, Garcia-Lobo, & Rodr´ıguez(1990); Claret, Rodr´ıguez, & Garcia (1990); Rodr´ıguez et al.(1990). Period decrease was found by Rodr´ıguez et al. (1995a,b)and physical parameters were also calculated by Rodr´ıguez et al.(1995b). Joner & Laney (2004) took high-quality spectroscopicmeasurements and determined the radius and the absolute magni-tude for RS Gru. They again showed unambiguously that RS Gru c (cid:13)000
45 km s − . McNamara & Feltz (1976) obtained uvby β photometry and spectrographic data and determinedphysical parameters. van Citters (1976) acquired photoelectricradial velocity curves on two nights, while further photometricobservations were taken by Dean et al. (1977). New radial velocitymeasurements taken by Balona & Martin (1978a) showed unam-biguously the variation of the center-of-mass velocity, indicatingthe binary nature of RS Gru but the orbital period was notdetermined for nearly three decades. Further investigations weredone by Breger (1980); Andreasen (1983); McNamara (1985);Antonello et al. (1986); Garrido, Garcia-Lobo, & Rodr´ıguez(1990); Claret, Rodr´ıguez, & Garcia (1990); Rodr´ıguez et al.(1990). Period decrease was found by Rodr´ıguez et al. (1995a,b)and physical parameters were also calculated by Rodr´ıguez et al.(1995b). Joner & Laney (2004) took high-quality spectroscopicmeasurements and determined the radius and the absolute magni-tude for RS Gru. They again showed unambiguously that RS Gru c (cid:13)000 , 000–000 Derekas et al. v r a d ( k m / s ) Figure 2.
Radial velocities of RS Gru, phased with the pulsation period. is a spectroscopic binary with an orbital period of approximatelytwo weeks but no exact period was given.We obtained standard
BV I photoelectric photometry usingSSO60 on 4 nights in 2003 and 1 night in 2004. For differen-tial magnitudes we used two comparison stars: comp=HD 207193( V = 6 . mag, B − V = 0 . mag) and check=HD 207615( V = 8 . mag, B − V = 0 . mag). The full log of observationsis given in Table A1 and the light and colour variation are plottedin Fig. 1.To measure the orbital period of the system, we obtainedmedium-resolution spectroscopy on 16 nights between 2003 and2005 using the SSO230 instrument. In addition, we observed thestar with the AAT on 1 night in 2006. The whole phased datasetis shown in Fig. 2, where the continuously changing shift in thesystemic velocity is evident.We performed a period analysis of the radial velocities, whichrevealed the main pulsation frequency at f =6.802 c/d, its integerharmonics ( f , f ), and a low frequency component at about ∼ γ -velocities for each night by fitting the zero-point only. This waywe could determine the center-of-mass velocity on 15 nights (listedin Table 3). The Fourier spectrum of the data (Fig. 3) shows a broadhump of peaks around 0.1 c/d with the highest peak at 0.087 c/d,which we identify as the most likely orbital frequency. The phased γ -velocities (Fig. 4) show a reasonably convincing sine-wave, forwhich the best-fit curve (solid line in Fig. 4) indicates a velocityamplitude of K=6.5 km s − .From these, we can estimate the mass of the companion fromthe mass function (Hilditch 2001): f ( M ) = (1 . × − )(1 − e ) / K P = M sin i/ ( M + M ) = 3 . × − M ⊙ . Assum-ing that the δ Scuti mass is about 1.5–2.5 M ⊙ , the calculated min-imum mass for the companion at different inclinations is shown inTable 4. The derived masses show that the companion of the RS Gruis most likely a low-mass star. We can also estimate the semimajoraxis of the system, which is about ∼ a = 0 .
11 AU at M = 0 . M ⊙ and a = 0 .
13 AU at M = 0 . M ⊙ ).There is an interesting possibility to determine the pulsation frequency (c/d) A m p lit ud e ( k m / s ) f orb Figure 3.
Fourier spectrum of the γ -velocities of RS Gru. Inset shows thespectral window. v g a mm a ( k m / s ) Figure 4. γ -velocity variation of RS Gru phased with P orb = 11 . . Table 3.
Center of mass velocities of RS Gru.HJD-2 400 000 v γ HJD -2 400 000 v γ (d) (km s − ) (d) (km s − )52922.0214 73.5 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± Table 4.
Estimated mass ( M ) for the companion of RS Gru. M = 1 . M ⊙ M = 2 M ⊙ M = 2 . M ⊙ Inclination M M M ( ◦ ) ( M ⊙ ) ( M ⊙ ) ( M ⊙ )
90 0.09 0.11 0.1370 0.10 0.12 0.1450 0.12 0.15 0.1730 0.20 0.24 0.2710 0.66 0.78 0.89c (cid:13) , 000–000 inarity and multiperiodicity in HADS variables constant, which is most useful for pulsating stars in eclipsing bi-naries, where the radius of the stars can be determined accurately(Jørgensen & Grønbech 1978). Combining Kepler’s third law andthe pulsation constant formula: a P = G π ( M + M ) and Q = P pul (cid:18) M R (cid:19) (1)results in (using the same units): Q = 0 . P pul P orb (cid:16) R a (cid:17) − (cid:16) M M (cid:17) − (2)Adopting R = 2 . ± . R ⊙ (Balona & Martin 1978a), P pul =0 . d, P orb = 11 . d, a = 0 . ± . AU, M = 2 . ± . M ⊙ (Rodr´ıguez et al. 1995b), M = 0 . ± . M ⊙ , the resulting Q = 0 . ± . d is consistent with pulsations in the radial fun-damental mode. The dominant sources of error are the radius, theunknown inclination and the orbital semi-major axis, which pose asignificant limitation at this stage. It is nevertheless reassuring thatthe given orbital period and size yield a consistent picture of RSGru being a fundamental mode pulsator. The light variation of RY Lep (HD 38882; HIP 27400; V =8.2 mag; I =8.3 mag) was discovered by Strohmeier (1964) and the star wasthought to be an eclipsing binary with an unknown period for morethan two decades. The SIMBAD database still lists it as an eclips-ing binary. Diethelm (1985) obtained five nights of observationswhich revealed the real HADS nature of this star. He determinedthe pulsation period as 0.2254 d and also noted small cycle-to-cycle variations, but the data were not sufficient to draw a firmconclusion. Rodr´ıguez et al. (1995b) determined physical parame-ters based on one night of uvbyβ observations, covering one pulsa-tion cycle. Laney, Joner, & Schwendiman (2002) found aperiodicor possibly multiperiodic variations and detected binary motion inthe radial velocities with a period more than 500 days. Finally,Rodr´ıguez et al. (2004) has shown unambiguously the multiperi-odic nature of RY Lep (f =4.4416 c/d, f =6.60 c/d) and its binaritywas also suggested from an analysis of the O–C diagram (detailshave not yet been published).To study RY Lep photometrically, we obtained I -band CCDimages on 20 nights between October 2004 and January 2005 usingthe APT50 instrument. We obtained more than 5000 data pointswith 10 s exposures. The full log of observations is given in TableA1. For the aperture photometry we used two comparison stars:comp=GSC 05926-01037 ( V = 9 . mag, I = 9 . mag B − V = 1 . mag), check=HD 39036 ( V = 8 . mag, I = 8 . mag B − V = 1 . mag).We performed standard Fourier analysis of the data. Fig. 6shows the results of the frequency search. We identified the twomain pulsational frequencies at f =4.4415 c/d and f =6.5987 c/d.A further 14 statistically significant peaks were found in the data,which are mainly the various linear combinations of the two pulsa-tion modes, and thereupon validating the pulsational nature of f .The results of the period analysis are summarized in Table 5. Thefinal light curve fit is shown in Fig. 5.The resulting frequencies ( f , f ) are in very good agree-ment with those by Rodr´ıguez et al. (2004). We have two low-amplitude peaks that seem to be significant and may be relatedto pulsations ( f and f ), similarly to V743 Cen, AI Vel andVW Ari, where 3, 4 and 7 frequencies were detected, respectively Figure 6.
Amplitude spectra of 20 nights of I -band data for RY Lep. Theinsert shows the window function. From top to bottom, every panel showsan amplitude spectrum prewhitened with all frequencies marked in the pan-els above. 16 frequencies can be identified with S/N larger than 4.c (cid:13) , 000–000
Derekas et al. ∆ I (mag) HJD (2450000+)
Figure 5.
Individual light curves of RY Lep (small dots) with the light curve fit (continuous line).
Table 5.
The result of the period analysis for RY Lep. f and f are the twopulsation modes, the remaining peaks are predominantly to be the harmon-ics or linear combinations of these two frequencies.No. Frequency Amplitude S/N Frequency( d − ) (mmag) identification ± f f f f + f f f f f − f + 1 . f f f f f + f f f + 2 f f f f − f f f f f f f f (McAlary & Wehlau 1979; Walraven, Walraven, & Balona 1992;Liu et al. 1996). Three low-frequency components that are presum-ably artifacts ( f , f and f ). The frequency ratio of f and f is 0.6731 which is not compatible with the usual scenario of fundamental and first overtone radial modes ( FU / ≈ . ).Rodr´ıguez et al. (2004) identified f with the fundamental modeand f with a non-radial p mode. The frequency ratio could alsoindicate first and third overtone radial modes, for which theoreti-cal models predict / ≈ . (Santolamazza et al. 2001), butthe physical parameters of the star, such as temperature, luminos-ity and evolutionary mass, are not compatible with that possibility(Rodr´ıguez et al. 1995b, 2004).Looking at the pulsational amplitudes of RY Lep, one can no-tice some interesting features. In our data the amplitude ratio of the f and f frequencies is about 0.5. Contrary to this, observations byRodr´ıguez et al. (2004) implied a significantly lower amplitude ra-tio of about 0.1. If we use a trasformation factor F ≈ I -band and the V -band amplitudes (see fig. 2 of Balona & Evers(1999)), numbers in Table 5 imply ∆ V ( f ) = 164 . mmag and ∆ V ( f ) = 79 . mmag. In comparison, the data in Rodr´ıguez et al.(2004), obtained between 1998 and 2002, revealed ∆ V ( f ) =164 . mmag and ∆ V ( f ) = 11 . mmag (Rodr´ıguez, 2008, per-sonal communication). We conclude that f seems to be very stablein amplitude, whereas f shows strong amplitude variations, withrecent data implying an 8-times larger amplitude.The first spectroscopic measurement of RY Lep was presentedby Popper (1966), where the spectral type was determined as F0.Recently, Laney, Joner, & Schwendiman (2002) found clear evi-dence for binary motion using radial velocity measurements butthe data did not allow them to determine the orbital period whichappeared to be longer than 500 days.We obtained radial velocity measurements on four nights in2004 and two in 2005. The RV curves of the seasonal datasets c (cid:13) , 000–000 inarity and multiperiodicity in HADS variables v r a d ( k m / s ) Figure 7.
The RV curve of RY Lep phased with P=0.225 d. The data of 6nights in 2004 and 2005 clearly show the ∼ km s − of γ -velocity shiftbetween the two sets of observations. Note that the October 2004 dataset isbinned. (February 2004, October 2004 and December 2005) phased withthe main period ( f ) are presented in Fig. 7 with three differentsymbols.The pulsation amplitude of RY Lep is about 30 km s − andits multiperiodic nature causes cycle-to-cycle variations in the RVcurves. The February 2004 and December 2005 datasets have basi-cally the same γ -velocity values but the October 2004 dataset has ahigher value by about 25-30 km s − . This leads to the conclusionthat the orbital motion is clearly detected in the almost 700 day longdataset, which is in agreement with Laney, Joner, & Schwendiman(2002) but still does not allow us to determine the orbital period.To estimate the approximate nature of the companion, we assumedthat the orbital period is about 730 d (Laney, Joner, & Rodr´ıguez2003) and took the full range of 25 km s − in v γ as an estimateof the 2K velocity amplitude. Repeating the same calculations asfor RS Gru, the companion’s mass is about 1.1 ± ⊙ and theorbital semi-major axis is 2.3 AU, i.e. the companion is comparableto RY Lep in mass. The lack of noticeable spectral lines from thesecondary may suggest a white dwarf but a firm conclusion wouldrequire spectra with broader coverage. One of the best studied HADS is AD CMi (HD 64191; HIP 38473; V =9.38 mag), whose light variation was discovered by Hoffmeister(1934) and classified as an eclipsing binary by Zessewitch (1950).The first detailed study of the star was done by Abhyankar(1959) who took photometric and spectroscopic observationsbut the data were not sufficient to determine the radius us-ing the Wesselink method. Further observations were obtainedby Anderson & McNamara (1960); Epstein & Abraham de Epstein(1973); Dean et al. (1977); Balona & Stobie (1983). Breger (1975)used uvby β photometry to determine radius, mass and variations ofphysical parameters during the pulsation cycle, while McNamara(1985) found that the rotational velocity is smaller than 20 km s − .Fourier decomposition of the light curve (Antonello et al. 1986)showed a surprisingly high φ value, suggestive of overtone pul-sation. However, the star seems to pulsate in fundamental modeas other monoperiodic HADS stars do (Kilambi & Rahman 1993).There has been no explanation for this phenomenon. Kim (1990)and Kim & Joner (1994) determined the radius of AD CMi, using v r a d ( k m / s ) Figure 8.
RV curve of AD CMi phased with the pulsation period(E =2449401.1320 d; P=0.12297443 d). the visual surface brightness method and found a very good agree-ment with angular diameters from theoretical and empirical rela-tionships. Kilambi & Rahman (1993) analyzed 8 nights of UBV R photometry and calculated physical parameters that agree well withBreger (1975) and suggested the star is lying on the cool-edge ofthe instability strip of the Population I stars. Jiang (1987) reporteda continuous period increase at the star, which was confirmed byRodr´ıguez, Rolland, & Lopez de Coca (1988, 1990). The stabilityof light curve was studied by Rodr´ıguez (1999) who found no sig-nificant long-term changes in amplitude. The first suggestion forbinarity for AD CMi was presented by Fu & Jiang (1996), whofound a possible orbital period of 30 years from the O–C diagram.Most recently, Hurta et al. (2007) and Khokhuntod et al. (2007)have studied the period variations of AD CMi using published andnew data. They deduced the presence of light-time effect due to bi-narity and a slow period increase due to evolutionary effect. In ad-dition, Khokhuntod et al. (2007) detected an extra low-frequencycomponent in the photometric data, which provides a possible ex-planation for the large scatter of the O − C diagram.We performed spectroscopic measurements on 3 nights inFebruary 2004 (see Table A1). The phased RV data (Fig. 8) havea mean amplitude of ∼
25 km s − , while showing significantcycle-to-cycle variation. The mean velocity is about 40 km s − .Two radial velocity measurements are available in the literature(Abhyankar 1959; Balona & Stobie 1983). They determined γ -velocities at 34.5 km s − and 38.8 km s − , respectively. Hurta et al.(2007) interpreted the current O − C diagram as a combinationof the a continuous period increase and light-time effect. The am-plitude of the orbital motion is expected to be about 1.1 km s − (Hurta et al. 2007). Considering this orbital amplitude, our data arein good agreement with Balona & Stobie (1983). The Abhyankar(1959) data, six points in total, are of lower quality and have poorphase coverage, so that the larger difference is still compatible withour result.Petersen & Høg (1998) were the first to notice that AD CMimay be peculiar in terms of luminosity because the Hipparcos par-allax indicated that the star, among five others, was situated ap-proximately 3 mag below the standard P–L relation. They evensuggested the possible existence of an “AD CMi group”. We havechecked the new reduction of the Hipparcos data (van Leeuwen2007). The updated parallax π = 6 . ± . mas differs by about1- σ from the original value at π old = 8 . ± . mas. While c (cid:13)000
25 km s − , while showing significantcycle-to-cycle variation. The mean velocity is about 40 km s − .Two radial velocity measurements are available in the literature(Abhyankar 1959; Balona & Stobie 1983). They determined γ -velocities at 34.5 km s − and 38.8 km s − , respectively. Hurta et al.(2007) interpreted the current O − C diagram as a combinationof the a continuous period increase and light-time effect. The am-plitude of the orbital motion is expected to be about 1.1 km s − (Hurta et al. 2007). Considering this orbital amplitude, our data arein good agreement with Balona & Stobie (1983). The Abhyankar(1959) data, six points in total, are of lower quality and have poorphase coverage, so that the larger difference is still compatible withour result.Petersen & Høg (1998) were the first to notice that AD CMimay be peculiar in terms of luminosity because the Hipparcos par-allax indicated that the star, among five others, was situated ap-proximately 3 mag below the standard P–L relation. They evensuggested the possible existence of an “AD CMi group”. We havechecked the new reduction of the Hipparcos data (van Leeuwen2007). The updated parallax π = 6 . ± . mas differs by about1- σ from the original value at π old = 8 . ± . mas. While c (cid:13)000 , 000–000 Derekas et al. the new value pushes the absolute magnitude of AD CMi about1 mag brighter, there is still a significant shift left unexplained.Studies of HADS/SX Phe variables in clusters and nearby galaxies(e.g. Poretti et al. 2006) do not indicate this large spread in absolutemagnitude, so that we suspect that there might be a yet-to-identifysource of systematic error in some of the Hipparcos HADSs.Our data also shows cycle-to-cycle variations in the shape ofRV curve that are within a range of a 2-3 km s − as shown in thephase diagram in Fig. 8. This might be due to the presence of anadditional pulsation mode but our data are not extensive enoughto resolve multiple modes. If this secondary mode is the same onereported by Khokhuntod et al. (2007), its amplitude must changein time, because the very low amplitude in the Khokhuntod et al.(2007) data is hardly compatible with the − km s − cycle-to-cycle RV change we find in the spectroscopic measurements. It isinteresting to add that the Abhyankar (1959) data showed a largerpeak-to-peak amplitude of about 35 km s − , which may also bedue to cycle-to-cycle variations caused by a second excited mode. BQ Ind (HD 198830; HIP 103290) was discovered to be avariable by the Hipparcos satellite and has a mean magni-tude V =9.8 mag, I =9.7 mag and a period of 0.0819877 d(Perryman et al. 1997a). The multiperiodic nature of the star firstdiscovered by Sterken, Fu, & Brogt (2003), who determined twofrequencies ( f =12.1951 c/d, f =15.7686 c/d), corresponding tothe fundamental and first overtone modes. Since then, no furtherobservations have been reported in the literature.We performed CCD photometry on BQ Ind on six consecu-tive nights in 2004 with APT50. More than 700 data points wereobtained with 30-40 s exposure time in I -band; a log of observa-tions is given in Table A1. For the aperture photometry we usedtwo comparison stars: comp=GSC 08800-00069 ( V = 10 . mag, I = 9 . mag, B − V = 1 . mag), check=PPM 774605( V = 10 . mag, I = 9 . mag, B − V = 1 . mag).The amplitude spectrum is shown in Fig. 10. The primary peakwas found at f = 12 . d − and the next prewhitening stepyielded the secondary frequency at f = 15 . d − . After sub-tracting the two main frequencies we ended up at their linear com-bination and then the integer harmonics ( f , f ) of the primaryfrequency. Their parameters are summarized in Table 6.The resulting two frequencies ( f , f ) confirm the double-mode nature of BQ Ind, and are in very good agreement (within1%) with the frequencies determined by Sterken, Fu, & Brogt(2003), with no further frequencies in the residuals. The period ra-tio is f /f = 0 . , which suggests fundamental ( f ) and first-overtone ( f ) mode pulsation. The Fourier-fit of the individual lightcurves is presented in Fig. 9. The short-period variability of ZZ Mic (HD 199757; HIP 103684)was discovered by Churms & Evans (1961). Its average V mag-nitude is 9.43 mag and the pulsation period is 0.0654 d. Thefirst detailed analysis of this star was done by Leung (1968),who found cycle-to-cycle variation in ultraviolet light and alsodetected a period decrease. Later the photoelectric observationsand data analysis (Chambliss 1971; Rodr´ıguez 1999) did not con-firm any change in the light curve shape. Percy (1976) reana-lyzed Leung’s observations and deduced two periods: 0.0654 d frequency (c/d) A m p lit ud e ( m a g ) (a)(b)(c) f f f + f Figure 10.
Fourier analysis of BQ Ind. Panel a: Amplitude spectrum of thecomplete dataset. The insert shows the window function. Panel b: After re-moval of the main period and its harmonics, the secondary period is clearlyseen. Panel c: After removal of the secondary period, the next peak is thelinear combination of the two frequencies.
Table 6.
The result of the period analysis for BQ Ind.No. Frequency Amplitude S/N Mode( d − ) (mmag) combination ± f f f f + f f f f f and 0.0513 d, suggesting fundamental and first overtone pulsa-tions (Balona & Martin 1978b). Previously, Bessell (1969) ana-lyzed spectrophotometric and spectroscopic observations and de-termined the pulsation constant, masses and absolute magnitudes,concluding the first-overtone pulsating nature of ZZ Mic. The firstradius determination of the star was carried out by Balona & Martin(1978b). The last analysis of the star was done by Rodr´ıguez(1999), who studied the stability of the light curve and did not findany significant long-term amplitude change.We took three nights of photoelectric observations in 2004using B , V filters on SSO60. For the calculations of differ-ential magnitudes we used the following two comparison stars:comp=HD 199639 ( V = 7 . mag, B − V = 0 . mag) andcheck=HD 200320 ( V = 8 . mag, B − V = 0 . mag). Thephase diagrams are shown in Fig. 11.Since the discovery of ZZ Mic, it has been controversial interms of changing light curve shape and being multiperiodic. In or-der to study the question, we performed a period analysis of ouradmittedly meagre V-band data. The pre-whitening steps are plot- c (cid:13) , 000–000 inarity and multiperiodicity in HADS variables HJD (2450000+) ∆ I (mag)
Figure 9.
Individual light curves of BQ Ind (small dots) with the five-component fit. m a gn it ud e B-0.4V0.10.20.3 B - V ( m a g ) v r a d ( k m / s ) Figure 11.
Standard light, colour and radial velocity variations of ZZ Mic(E =2453305.9819 d; P=0.0671835 d). ted in Fig. 12, while the resulting parameters are listed in Table 7.The Fourier spectrum is dominated by the main pulsational period( f = =
14 mmag compared to A = f = f = f = S/N ratio of this frequency is 8, which is quite low compared to f and 2 f but still above the significance limit.Because of the limited data we have, we tried to detect thesecondary frequency in other publicly available data. We ana-lyzed the data from the All Sky Automated Survey (ASAS) project(Pojma´nski 2002). The Fourier spectrum of this data (Fig. 13)clearly shows the main pulsational period at f = ∼
23 mmag) is too high compared A m p lit ud e ( m a g ) frequency (c/d) (a)(b)(c) f f Figure 12.
Fourier spectra of ZZ Mic with the pre-whitening steps. Theinsert shows the spectral window. to the amplitude of the secondary frequency ( ∼
14 mmag), whichprevents any detection in the ASAS data.If we accept f as an independent mode, the ratio of the twomodes is f /f = 0 . . This suggests f is the fundamental modeand f is the first overtone mode. The fundamental mode identifi-cation for f is strongly supported by the ubvyβ photometry ofRodr´ıguez, Lopez-Gonz´alez, & Lopez de Coca (2000). Moreover, f must be a radial mode which was suggested from the phase shiftsin BV photometry by Rodr´ıguez et al. (1996).Moreover, a period ratio of 0.778 seems to be toolarge for a normal Pop. I HADS (Poretti et al. 2005;Petersen & Christensen-Dalsgaard 1996) which suggests thatZZ Mic is a Pop. II star. However, the value of f is not tooreliable, so the period ratio might be slightly different. Studies on c (cid:13)000
14 mmag), whichprevents any detection in the ASAS data.If we accept f as an independent mode, the ratio of the twomodes is f /f = 0 . . This suggests f is the fundamental modeand f is the first overtone mode. The fundamental mode identifi-cation for f is strongly supported by the ubvyβ photometry ofRodr´ıguez, Lopez-Gonz´alez, & Lopez de Coca (2000). Moreover, f must be a radial mode which was suggested from the phase shiftsin BV photometry by Rodr´ıguez et al. (1996).Moreover, a period ratio of 0.778 seems to be toolarge for a normal Pop. I HADS (Poretti et al. 2005;Petersen & Christensen-Dalsgaard 1996) which suggests thatZZ Mic is a Pop. II star. However, the value of f is not tooreliable, so the period ratio might be slightly different. Studies on c (cid:13)000 , 000–000 Derekas et al.
Table 7.
The result of the period analysis for the ZZ Mic.No. Frequency Amplitude S/N Mode( d − ) (mmag) combination ± f f f f Figure 13.
The Fourier spectrum of ZZ Mic using the ASAS data. Theinsert shows the spectral window. metal abundances and space motions (Breger 1980) suggest thatZZ Mic is a normal Pop. I HADS, which is also supported byRodr´ıguez, Lopez-Gonz´alez, & Lopez de Coca (2000).We obtained spectra simultaneously with BV light curves onone night using SSO230. The resulted RV curve is shown in thebottom panel of Fig. 11, which is the first radial velocity curve ob-tained of ZZ Mic. The full amplitude of the RV curve is 22 km s − . CY Aqr (HIP 111719; V =10.7 mag, I =10.3 mag) is one ofthe shortest period HADS in the galactic field, with a pulsa-tion period of 0.061038d, and has been subject to many in-vestigations. It was discovered by Hoffmeister (1935). A num-ber of early studies on the star are listed by Hardie & Tolbert(1961), who estimated physical parameters and found that theshape of the light curve varies. The period stability was stud-ied by Ashbrook (1954), who found no change in period but no-ticed a phase jump that seemed to be attributed to the star. Fur-ther studies were made by Zissell (1968); Nather & Warner (1972);Bohusz & Udalski (1980). Changes in light curve shape and thepossibility of another period were also investigated in several pa-pers, e.g. Elst (1972); Fitch (1973); Figer (1978) but both phe-nomena were discounted later by Geyer & Hoffmann (1975); Percy(1975); Purgathofer & Schnell (1984); Hintz & Joner (1997). Fi-nally, Coates et al. (1994) set a definite upper limit of 1.5 mmagin V for the amplitude of any long-lived secondary period.K¨amper (1985) made a thorough period study and his re-sults indicated the presence of random fluctuations in pulsationfrequency that cannot be explained by considering only evolu-tion. Other period change studies were performed by Rolland et al.(1986); Mahdy et al. (1988); Powell, Joner & McNamara (1995).McNamara, Powell, & Joner (1996) determined physical proper-ties. Fu, Jiang, & Liu (1994) suggested that the period changes dueto the presence of an unseen companion with an orbital periodof around 50 years. Zhou, Fu, & Jiang (1999) and Fu & Sterken(2003) studied the O–C diagram to characterize long-term periodevolution. They found a long-term cyclic component, and both m a gn it ud e B-0.7VI+0.30.20.4 B - V ( m a g ) -0.20 V -I ( m a g ) v r a d ( k m / s ) Figure 14.
Standard light, colour and radial velocity variations of CY Aqr(E =2452920.9223 d; P=0.061038328 d). suggested possible binarity for CY Aqr with an orbital period of ∼ ∼ BV I photoelectric, 5 nights I -band and 5 nights CCD V -band photometry with SSO60, APT50and P60 between 2003 and 2007. The integration time was 15swith SSO60 and 50s with the APT50. The full log of observa-tions is given in Table A1. Differential magnitudes were calcu-lated using the following comparison stars: comp=GSC 00567-02242 ( V = 9 . mag, I = 8 . mag, B − V = 1 . mag)and check=GSC 00567-01242 ( V = 10 . mag, I = 9 . mag, B − V = 1 . mag). The resulting light and colour curves are plot-ted in the top three panels of Fig. 14. We determined new times ofmaximum that are listed in Table 2. The O–C diagram (not shown)is in a very good agreement with light-time solution determined byFu & Sterken (2003).We obtained spectroscopic observations on two nights in 2003and 2004. The mean radial velocity is –38 km s − . This valueis in a good agreement with previous data by Struve (1949) andFernley et al. (1987), who measured –32 km s − and –40 km s − ,respectively. The predicted amplitude of mean velocity change dueto the binarity is ∼ − (Zhou, Fu, & Jiang 1999) which iscomparable to the accuracy of our observation. Furthermore, dueto the high eccentricity and long period of the binary system, themean velocity changed only slightly in the last 10 years, which isfar beyond our detection limit. In conclusion, our radial velocityobservations do not contradict the current understanding of the na-ture of CY Aqr. c (cid:13) , 000–000 inarity and multiperiodicity in HADS variables v r a d ( k m / s ) Figure 15.
Phased RV curve of BE Lyn (HJD=2449749.4651 d,P=0.09586952 d).
We obtained high-resolution spectroscopy of BE Lyn on threenights with MH150 in order to detect possible binarity or additionalpulsational frequencies. The period change of this star inspired a se-ries of studies by our group (Kiss & Szatm´ary 1995; Derekas et al.2003; Szak´ats, Szab´o, & Szatm´ary 2008) and, while the initial or-bital elements of the suspected binary system were ruled out andhence leaving the binarity unconfirmed, the lack of spectroscopicdata in the literature has kept this star in our focus.To our knowledge, our radial velocity measurements are thefirst obtained for BE Lyn. The phased RV curve is shown in Fig. 15,where we see characteristic shape and velocity amplitude for fun-damental mode pulsation. The center-of-mass velocity is measuredat 3.4 km s − , while the amplitude of the variation is ∼
34 km s − .There is no sign of gamma velocity change during the three nightsof observation and we also could not detect any non-radial modepulsation in the RV curve. Further study of the data in terms ofvelocity gradient within the stellar atmosphere is in progress. We performed times-series photometry on XX Cyg, DY Peg andDY Her. Previous studies of these stars are listed in Derekas et al.(2003). Since then, only DY Peg was studied by Hintz et al. (2004),who explained the period change of the star with two period breaksrather than continuously decreasing rate, as was previously thought.We obtained 3 nights of V -band CCD photometry on XXCyg with P60 and Sz40 and 1-1 night on DY Her and DY Peg withP60 during 2007 and 2008. The journal of observations is given inTable A1. We determined new times of maximum that are listedin Table 2. The updated O–C diagrams contain these and recentlypublished data collected from the literature (Agerer & H¨ubscher2003; H¨ubscher 2005; H¨ubscher, Paschke & Walter 2005;B´ır´o et al. 2006; H¨ubscher, Paschke & Walter 2006;Klingenberg, Dvorak, & Robertson 2006; H¨ubscher 2007;H¨ubscher & Walter 2007).We calculated the O–C diagram of XX Cyg using the follow-ing ephemeris: HJD max = 2451757 . . × E (Derekas et al. 2003), and the resulting diagram is shown in the toppanel of Fig. 16. The parabolic fit form of the O–C diagram is: HJD max = 0 . . × − E + 2 . × − E with an -0.00500.0050.010.015 O - C ( d ) XX Cyg-250000 -200000 -150000 -100000 -50000 0cycle number-0.006-0.00300.0030.006 O - C r e s ( d ) -0.0100.010.020.030.04 O - C ( d ) DY Her 0 50000 100000 150000cycle number-0.00500.005 O - C r e s ( d ) -0.008-0.006-0.004-0.00200.002 O - C ( d ) DY Peg 0 100000 200000 300000cycle number-0.00500.005 O - C r e s ( d ) Figure 16.
O–C diagram of DY Peg, DY Her, XX Cyg. rms of 0.00152 d. The second-order coefficient corresponds to arelative rate of period change P dPdt = 1 . × − yr − , which isonly slightly different from the value of . × − yr − given byBlake et al. (2003). Therefore, we can conclude that the period ofXX Cyg has been increasing at a same rate of the last few years.The residuals have no signs of any other change. c (cid:13) , 000–000 Derekas et al.
The O–C diagram of DY Her was calculated with the fol-lowing ephemeris:
HJD max = 2433439 . . × E (Derekas et al. 2003). The diagram is shown in the middle panelof Fig. 16. The O–C diagram was fitted with a parabolic form of HJD max = − . . × − E − . × − E withan rms of 0.0013 d, which gives P dPdt = − . × − yr − . Thisis 6% difference in the rate of period change previously given byDerekas et al. (2003). The residuals do not show any other periodchange, so we can conclude that the present results in period changeis very well agreed with the previous studies of DY Her and the staris showing continuous slow period decrease.Finally, we also updated the O–C diagram of DY Peg using thefollowing ephemeris: HJD max = 2432751 . . × E (Mahdy 1987) and shown the final O–C diagram in the bottompanel of Fig. 16. We performed a parabolic fit of the diagram thatresulted in the following form: HJD max = − . . × − E − . × − E with an rms of 0.0008 d. From this, wederived P dPdt = − . × − yr − which is in a good agree-ment with Mahdy (1987), Pe˜na, Gonz´alez & Hobart (1987) andDerekas et al. (2003). The residuals of the O–C diagram show somesigns of cyclic change over 100 000 cycles but the present data areinsufficient to draw a firm conclusion. We have carried out multicolor photometry and medium- and high-resolution spectroscopy of ten bright high-amplitude δ Scuti starsover 5 years. Our aim was to detect binarity and/or multiperiodicityin HADS variables in order to deepen our knowledge of interactionbetween oscillations and binarity.To put our binary targets in a broader context, we have com-piled a complete list of binary HADS variables, presented in Table8. How do RS Gru and RY Lep, the two newly confirmed spec-troscopic binaries, compare with other known systems? Lookingat the 8 stars in Table 8, we can see three distinct groups withmarkedly different orbital periods. RS Gru is one of the shortest-period binary and it is interesting to note that neither UNSW-V-500 nor RS Gru show evidence of multimode pulsations. RY Lepis similar to SZ Lyn both in the orbital period and the reasonablylarge mass of the companion. These intermediate-period systemsare also promising for detecting spectral features of the companionin the ultraviolet or infrared region, thus allowing a full dynami-cal mass determination. To be able to detect spectroscopically thebinary nature of the long-period systems, will require very high-precision spectroscopy, since the expected v γ change is in the rangeof 1 km s − .To summarize, the main results of this paper are the follows:(i) We monitored RS Gru spectroscopically on 17 nights in or-der to measure the orbital period. We derived the orbital period as11.5 days.(ii) We confirmed the multimode pulsation of RY Lep fromCCD photometry, detecting and refining the frequencies oftwo independent modes. Spectroscopic measurements also showthe multimode pulsation. We detected the orbital motion inthe radial velocity curve, confirming the preliminary results ofLaney, Joner, & Schwendiman (2002) on the binary nature ofRY Lep. Our 700 day-long dataset is in good qualitative agreementwith Laney, Joner, & Schwendiman (2002). The limits on the or-bital period and RV amplitude suggest a binary companion of about1 M ⊙ , possibly a white dwarf star. Table 8.
Summary of the estimated masses of the companions in theknown HADS binary systems. Sources for P orb and masses are: (1)Christiansen et al. (2007), (2) Moffett et al. (1988), (3) Fu et. al (2008)(4) Fu & Jiang (1999), (5) Hurta et al. (2007), (6) Fu, Sterken & Barrera(2004).Star P orb m comp ( M ⊙ ) Refs.UNSW-V-500 5.35 d ∼ ∼ (iii) The radial velocity curve of AD CMi shows cycle-to-cyclevariations that support the presence of a low frequency mode pul-sation reported by Khokhuntod et al. (2007). The center-of-massvelocity is in good agreement with the previous measurement byBalona & Stobie (1983) and does not contradict the binary hypoth-esis of the star, since the predicted γ -velocity change is around1 km s − .(iv) We obtained the first spectroscopic measurements forBE Lyn. The RV curve has an amplitude of ∼
34 km s − and thecenter-of-mass velocity is 3.4 km s − .(v) We confirmed the double-mode nature of BQ Ind, corre-sponding to the fundamental and first overtone modes.(vi) We detected a low-amplitude secondary period in the pho-tometry of ZZ Mic but further observations are needed to confirmits validity. The RV curve has a full amplitude of 22 km s − .(vii) We updated the O–C diagram for CY Aqr, corroborat-ing binarity found by Zhou, Fu, & Jiang (1999) and Fu & Sterken(2003). Our radial velocity data are in a good agreement with pre-vious observations by Struve (1949) and Fernley et al. (1987) buthave better accuracy.(viii) We obtained new time series photometry on XX Cyg, DYHer and DY Peg and updated their O–C diagrams with new times ofmaximum. DY Her and DY Peg show continuous period decrease,while XX Cyg has continuous period increase.Further analysis of the photometric and spectroscopic observations(e.g. determination of physical parameters) will be presented in asubsequent paper. ACKNOWLEDGMENTS
We are grateful to the referee for the comments that improved thepaper. AD is supported by an University of Sydney PostgraduateAward. GyMSz is supported by the Bolyai J´anos Research Fel-lowship of the Hungarian Academy of Sciences. The research wassupported by the Hungarian OTKA Grant T042509. This work hasbeen supported by the Australian Research Council. The authorsare grateful for Prof. Chris Tinney for taking the high-quality ser-vice observations with the AAT. We thank the CfA TAC for theirsupport by providing telescope time for this project. The NASAADS Abstract Service was used to access data and references. Thisresearch has made use of the SIMBAD database, operated at CDS-Strasbourg, France. c (cid:13) , 000–000 inarity and multiperiodicity in HADS variables REFERENCES
Abhyankar, K. D., 1959, ApJ, 130, 834Agerer, F., & H¨ubscher, J., 2003, IBVS, No. 5485Anderson, L. R., & McNamara, D. H., 1960, PASP, 72, 506Andreasen, G. K., 1983, A&A, 121, 250Antonello, E., Broglia, P., Conconi, P., & Mantegazza, L., 1986,A&A, 169, 122Ashbrook, J., 1954, AJ, 59, 6Balona, L. A., & Martin, W. L., 1978a, MNRAS, 184, 1Balona, L. A., & Martin, W. L., 1978b, MNRAS, 184, 11Balona, L. A., & Stobie, R. S., 1983, SAAOC, 7, 19Balona, L. A., & Evers, E. A., 1999, MNRAS, 302, 349Berdnikov, L. N., & Turner, D. G., 2004, Astron. and Astrop.Transactions, 23, 253Bessell, M. S., 1969, ApJS, 18, 195B´ır´o et al., 2006, IBVS, No. 5684Blake, R. M., Delaney, P., Khosravani, H., Tome, J., & Lightman,M., 2003, PASP, 115, 212Bohusz, E., & Udalski, A, 1980, AcA, 30, 359Breger, M., 1975, ApJ, 201, 653Breger, M., 1980, ApJ, 235, 153Breger, M., et al., 1993, A&A, 271, 482Carter, B. D., Ashley, M. C. B., Sun, Y.-S., & Storey, J. W. V.,1992, PASA, 10, 74Chambliss, C. R., 1971, ApJ, 165, 365Churms, J., & Evans, D. S., 1961, Observatory, 81, 25Christiansen, J.L., Derekas, A., Ashley, M. C. B., Webb, J. K.,Hidas, M. G., Hamacher, D. W., & Kiss, L. L., 2007, MNRAS,382, 239van Citters, G. W. Jr., 1976, AJ, 81, 766Claret, A., Rodr´ıguez, E., & Garcia, J. M., 1990, RMxAA, 21, 389Coates, D. W., Fernley, J. A., Sekiguchi, K., Barnes, T. G., &Frueh, M. L., 1994, MNRAS, 266, 1Dean, J. F., Cousins, A. W. J., Bywater, R. A., & Warren, P. R.,1977, MmRAS, 83, 69The DENIS consortium, 2005, Third release of DENIS data(September 2005), Vizier Catalogues, B/denisDerekas, A., et al., 2003, A&A, 402, 733Diethelm, R., 1985, A&A, 149, 465Eggen, O. J., 1956, PASP, 68, 142Elst, E. W., 1972, A&A, 17, 148Epstein, I, & Abraham de Epstein, A. E., 1973, AJ, 78, 83Figer, A., 1980, IBVS, No. 1388Fitch, W. S., 1973, A&A, 27, 161Fernley, J. A., Jameson, R. F., Sherrington, M. R., & Skillen, I. ,1987, MNRAS, 225, 451Fu, J. N., & Jiang, S. Y., 1996, IBVS, No. 4325Fu, J. N., & Jiang, S. Y., 1999, Delta Scuti Star Newsletter, 13, 9Fu, J. N., & Sterken, C., 2003, A&A, 405, 685Fu, J. N., Jiang, S. Y., & Liu, Y. Y., 1994, IBVS, No. 3970Fu, J. N., Sterken, C., & Barrera, L., 2004, ASPC, 318, 346Fu, J. N., et al., 2008, AJ, 135, 1958Garrido, R., Garcia-Lobo, E., & Rodr´ıguez, E., 1990, A&A, 234,262Geyer, E. H., & Hoffmann, M., 1975, A&AS, 21, 177Handler, G., et al., 2000, JAD, 6, 4Hardie, R. H., & Tolbert, C. R., 1961, ApJ, 134, 581Hilditch, R. W., 2001, An Introduction to Close Binary Stars,Cambridge University PressHintz, E. G., & Joner, M. D., 1997, PASP, 109, 639Hintz, E. G., Joner, M. D., Ivanushkina, M., & Pilachowski, C. A., 2004, PASP, 116, 543Hoffmeister, C., 1934, Astr. Nachr., 253, 195Hoffmeister, C., 1935, Beobachtungszirkular, 16, 45Hoffmeister, C., 1956, Ver¨off. Sonneberg Sternw., 3, 1H¨ubscher, J., 2005, IBVS, No. 5643H¨ubscher, J., 2007, IBVS, No. 5802H¨ubscher, J., & Walter, F., 2007, IBVS, No. 5761H¨ubscher, J., Paschke, A., & Walter, F., 2005, IBVS, No. 5657H¨ubscher, J., Paschke, A., & Walter, F., 2006, IBVS, No. 5731Hurta, Zs., P´ocs, M. D., & Szeidl, B., 2007, IBVS, No. 5774Jiang, S. Y., 1987, ChA&A, 11, 343Joner, M. D., & Laney, C. D., 2004, AAS, 205, 5414Jørgensen, H. E., & Grønbech, B., 1978, A&A, 66, 377K¨amper, B.-C., 1985, IBVS, No. 2802Kharchenko, N. V., 2001, KFNT, 17, 409, Vizier CataloguesI/280AKhokhuntod, P., Fu, J. N., Boonyarak, C., Marak, K., Chen, L., &Jiang, S. Y., 2007, ChJAA, 7, 421Kholopov et al. 1985-1988, General Catalogue of Variable Stars,Vol. III, Nauka, MoscowKilambi, G. C., & Rahman, A., 1993, BASI, 21, 47Kim, C., 1990, Ap&SS, 168, 153Kim, C., & Joner, M. D., 1994, Ap&SS, 218, 137Kim, S.-L., Lee, J W., Kwon, S.-G., Youn, J.-H., Mkrtichian, D.E., & Kim, C., 2003, A&A, 405, 231Kinman, T. D., 1961, Royal Obs. Bull., 37, 151Kiss, L. L., & Szatm´ary, K., 1995, IBVS, No. 4166Kiss, L. L., Derekas, A., M´esz´aros, Sz., & Sz´ekely, P., 2002,A&A, 394, 943Kjeldsen, H., 2003, Ap&SS, 284, 1Klingenberg, G., Dvorak, S. W., & Robertson, C. W., 2006, IBVS,No. 5701Laney, C. D., Joner, M., & Schwendiman, L., 2002, ASPC, 256,173Laney, C. D., Joner, M., & Rodr´ıguez, E., 2003, ASPC, 292, 203van Leeuwen, F., 2007, Astrophys. Space Sci. Lib., 350Lenz, P., & Breger, M., 2005, Comm. Asteroseis., 146, 53Leung, K.-C., 1968, AJ, 73, 6Liu, Y. Y., et al., 1996, A&AS, 120, 179Mahdy, M. A., 1987, IBVS, No. 3055Mahdy, H. A., Soliman, M. A., & Hamdy, M. A., 1988, IBVS, No.3276McAlary, C. W., & Wehlau, W. H., 1979, AJ, 84, 1211McNamara, D. H., 1985, PASP, 97, 715McNamara, D. H., 2000, ASPC, 210, 373McNamara, D. H., & Feltz, K. A. Jr., 1976, PASP, 88, 510McNamara, D. H., Powell, J. M., & Joner, M. D., 1996, PASP,108, 1098Mkrtichian, D. E., et al., 2006, Ap&SS, 304, 169Moffett, T. J., Barnes, T. G. III, Fekel, F. C., Jefferys, W. H., &Achtermann, J. M., 1988, AJ, 95, 153Munari, U., Sordo, R., Castelli, F., & Zwitter, T. 2005, A&A, 442,1127Nather, B., & Warner, R. E., 1972, MNRAS, 156, 315Oosterhoff, P. Th., & Walraven, Th., 1966, BAN, 18, 387Pe˜na, J. H., Gonz´alez, S. F., & Hobart, M. A., 1987, A&AS, 138,11Percy, J. R., 1975, A&A, 69, 251Percy, J. R., 1976, Proc. Solar and Stellar Puls. Conf., 60Perryman, M. A. C., et al., 1997a, A&A, 323, 49Perryman, M. A. C., et al., 1997b, The
Hipparcos and Tycho
Cat-alogues (ESA SP-1200; Noordwjik: ESA) c (cid:13) , 000–000 Derekas et al.
Petersen, J. O., & Christensen-Dalsgaard, J., 1996, A&A, 312,463Petersen, J. O., & Høg, E., 1998, A&A, 331, 989Pigulski, A., & Michalska, G., 2007, AcA, 57, 61Pojma´nski, G., 2002, AcA, 52, 397Popper, D. M., 1966, AJ, 71, 175Poretti, E., 2003, A&A, 409, 1031Poretti, E., et al., 2005, A&A, 440, 1097Poretti, E., et al., 2006, Mem. S. A. It., 77, 219Powell, J. M., Joner, M. D., & McNamara, D. H., 1995, PASP,107, 225Purgathofer, A., & Schnell, A., 1984, IBVS, No. 2500Rodr´ıguez, E., 1999, PASP, 111, 709Rodr´ıguez, E., & Breger, M., 2001, A&A, 366, 178Rodr´ıguez, E., Rolland, A., & Lopez de Coca, P., 1988, RMxAA,16, 7Rodr´ıguez, E., Rolland, A., & Lopez de Coca, P., 1990, IBVS, No.3427Rodr´ıguez, E., Lopez-Gonz´alez, M. J., & Lopez de Coca, P., 2000,A&AS, 144, 469Rodr´ıguez, E., Rolland, A., Lopez de Coca, P., Garrido, R., &Garcia-Lobo, E., 1990, RMxAA, 21, 386Rodr´ıguez, E., L´opez de Coca, P., Costa, V., & Mart´ın, S., 1995a,A&A, 299, 108Rodr´ıguez, E., Rolland, A., Costa, V., & Mart´ın, S., 1995b, MN-RAS, 277, 965Rodr´ıguez, E., Rolland, A., Lopez de Coca, P., & Mart´ın, S., 1996,A&A, 307, 539Rodr´ıguez, E., et al., 2004, Comm. in Asteroseis., 145, 48Rolland, A., Pe˜na, J. H., Lopez de Coca, P., Peniche, R., & Gon-zalez, S. F., 1986, A&A, 168, 125Santolamazza, P., Marconi, M., Bono, G., Caputo, F., Cassisi, S.,& Gilliland, R. L., 2001, ApJ, 554, 1124Soydugan, E., Soydugan, F., Demircan, O., & ˙Ibanoˇglu, C., 2006,MNRAS, 370, 2013Sterken, C., Fu, J.-N., & Brogt, E., 2003, ASPC, 292, 121Strohmeier, W., 1964, IBVS, No. 51Struve, O., 1949, AJ, 54, 137Szak´ats, R., Szab´o, Gy. M., & Szatm´ary, K., 2008, IBVS, No.5816Walraven, Th., Walraven, J., & Balona, L. A., 1992, MNRAS,254, 59Zessewitch, B. P., 1950, Astron. Circ. USSR, 100, 18Zhou, A. Y., Fu, J. N., & Jiang, S. Y., 1999, Ap&SS, 268, 397Zissel, R., 1968, AJ, 73, 696
APPENDIX A: FULL LOG OF OBSERVATIONS. c (cid:13) , 000–000 inarity and multiperiodicity in HADS variables Table A1.
Journal of observations.Date Filter Instrument Data points Obs. length Date Filter Instrument Data points Obs. length
RS Gru AD CMi
B, V, I
SSO60 93 3.6 h 2004–02–04 spec.
SSO230 63 4.9 h2003–10–09
B.V, I
SSO60 145 4.4 h 2004–02–07 spec.
SSO230 69 4.3 h2003–10–11
B, V, I
SSO60 125 2.9 h 2004–02–08 spec.
SSO230 47 2.6 h2003–10–12
B, V, I
SSO60 121 2.9 h
BQ Ind
B, V, I
SSO60 141 3.6 h 2004–10–25 I APT60 153 3.5 h2003–10–09 spec.
SSO230 151 3.9 h 2004–10–26 I APT60 46 1.1 h2004–09–25 spec.
SSO230 146 3.3 h 2004–10–27 I APT60 156 3.6 h2004–09–26 spec.
SSO230 170 3.4 h 2004–10–28 I APT60 133 3.0 h2004–09–27 spec.
SSO230 99 2.3 h 2004–10–29 I APT60 120 2.8 h2004–09–28 spec.
SSO230 14 0.3 h 2004–10–30 I APT60 98 2.2 h2005–05–28 spec.
SSO230 14 0.7 h
ZZ Mic spec.
SSO230 179 3.8 h 2004–10–27
B, V
SSO60 148 2.7 h2005–05–30 spec.
SSO230 203 4.1 h 2004–10–29
B, V
SSO60 105 1.7 h2005–05–31 spec.
SSO230 161 4.2 h 2004–10–30
B, V
SSO60 124 2.1 h2005–06–01 spec.
SSO230 86 3.4 h 2004–10–27 spec.
SSO230 41 2.4 h2005–06–02 spec.
SSO230 57 1.4 h
CY Aqr spec.
SSO230 116 2.9 h 2003–10–08
V, I
SSO60 193 4.0 h2005–08–18 spec.
SSO230 125 3.2 h 2003–10–10
V, I
SSO60 85 1.7 h2005–08–21 spec.
SSO230 8 1.1 h 2003–10–13
B, V
SSO60 175 2.9 h2005–08–22 spec.
SSO230 37 3.5 h 2004–11–24 I APT50 43 1.3 h2005–08–23 spec.
SSO230 98 3.4 h 2004–11–25 I APT50 47 1.3 h2006–07–21 spec.
AAT 40 4.2 h 2004–11–26 I APT50 33 0.9 h
RY Lep I APT50 45 1.3 h2004–10–25 I APT50 190 3.3 h 2004–11–28 I APT50 49 1.3 h2004–10–28 I APT50 246 3.7 h 2007–07–25 V P60 101 1.2 h2004–10–29 I APT50 335 4.9 h 2007–07–26 V P60 103 1.8 h2004–11–24 I APT50 105 1.9 h 2003–10–08 spec.
SSO230 87 3.6 h2004–11–25 I APT50 366 5.8 h 2004–07–04 spec.
SSO230 22 1.6 h2004–11–26 I APT50 345 6.1 h
BE Lyn I APT50 385 6.0 h 2007–10–25 spec
MH150 10 1.0 h2004–11–28 I APT50 380 6.2 h 2007–10–26 spec
MH150 22 2.1 h2004–12–24 I APT50 98 1.4 h 2007–10–27 spec
MH150 20 2.0 h2004–12–25 I APT50 60 1.2 h
XX Cyg I APT50 324 4.7 h 2007–07–25 V P60 117 3.2 h2004–12–28 I APT50 464 6.7 h 2007–07–27 V P60 201 3.6 h2004–12–29 I APT50 471 6.8 h 2008–07–29 V Sz40 85 3.3 h2004–12–30 I APT50 419 6.6 h
DY Her I APT50 144 2.1 h 2007–07–22 V P60 131 2.2 h2005–01–07 I APT50 140 2.0 h
DY Peg I APT50 137 1.9 h 2007–07–23 V P60 145 2.7 h2005–01–09 I APT50 135 1.9 h2005–01–10 I APT50 130 1.9 h2005–01–11 I APT50 128 1.9 h2004–02–05 spec.
SSO230 178 4.5 h2004–02–06 spec.
SSO230 134 3.9 h2004–10–26 spec.
SSO230 28 1.3 h2004–10–27 spec.
SSO230 75 3.9 h2005–12–19 spec.
SSO230 233 5.5 h2005–12–20 spec.
SSO230 203 5.5 hc (cid:13)000