Large Variety of New Pulsating Stars in the OGLE-III Galactic Disk Fields
P. Pietrukowicz, W.A. Dziembowski, P. Mroz, I. Soszynski, A. Udalski, R. Poleski, M.K. Szymanski, M. Kubiak, G. Pietrzynski, L. Wyrzykowski, K. Ulaczyk, S. Kozlowski, J. Skowron
aa r X i v : . [ a s t r o - ph . GA ] J a n Large Variety of New Pulsating Starsin the OGLE-III Galactic Disk Fields ∗ P. P i e t r u k o w i c z , W. A. D z i e m b o w s k i , P. M r ´o z ,I. S o s z y ´n s k i , A. U d a l s k i , R. P o l e s k i , ,M. K. S z y m a ´n s k i , M. K u b i a k , G. P i e t r z y ´n s k i , , L. W y r z y k o w s k i , , K. U l a c z y k ,S. K o z l o w s k i and J. S k o w r o n Warsaw University Observatory, Al. Ujazdowskie 4, 00-478 Warszawa, Polande-mail: (pietruk,wd,pmroz,soszynsk,udalski,rpoleski,msz,mk,pietrzyn,wyrzykow,kulaczyk,simkoz,jskowron)@astrouw.edu.pl Department of Astronomy, Ohio State University, 140 W. 18th Ave.,Columbus, OH 43210, USA Universidad de Concepci´on, Departamento de Astronom´ıa,Casilla 160-C, Concepci´on, Chile Institute of Astronomy, University of Cambridge, Madingley Road,Cambridge CB3 0HA, UK
ABSTRACTWe present the results of a search for pulsating stars in the 7.12 deg OGLE-III Galacticdisk area in the direction tangent to the Centaurus Arm. We report the identification of20 Classical Cepheids, 45 RR Lyr type stars, 31 Long-Period Variables, such as Miras andSemi-Regular Variables, one pulsating white dwarf, and 58 very likely δ Sct type stars. Basedon asteroseismic models constructed for one quadruple-mode and six triple-mode δ Sct typepulsators, we estimated masses, metallicities, ages, and distance moduli to these objects. Themodeled stars have masses in the range 0.9–2.5 M ⊙ and are located at distances between2.5 kpc and 6.2 kpc. Two triple-mode and one double-mode pulsators seem to be PopulationII stars of the SX Phe type, probably from the Galactic halo. Our sample also includescandidates for Type II Cepheids and unclassified short-period ( P < .
23 d) multi-mode starswhich could be either δ Sct or β Cep type stars. One of the detected variables is a very likely δ Sct star with an exceptionally high peak-to-peak I -band amplitude of 0.35 mag at the veryshort period of 0.0196 d. All reported pulsating variables but one object are new discoveries.They are included in the OGLE-III Catalog of Variable Stars.Finally, we introduce the on-going OGLE-IV Galactic Disk Survey, which covers morethan half of the Galactic plane. For the purposes of future works on the spiral structure andstar formation history of the Milky Way, we have already compiled a list of known GalacticClassical Cepheids. Galaxy: disk – Stars: variables: Cepheids – Stars: variables: delta Scuti –Stars: variables: RR Lyrae – variables: general
Pulsations are present in many phases of stellar evolution and are reflected ina variety of observed types of pulsating variable stars. Pulsating stars occupydifferent regions on the Hertzsprung–Russell diagram. They are seen amongmain-sequence (MS) and post-MS stars ( e.g. , β Cep, γ Dor, δ Sct type), giants( e.g. , Classical Cepheids, Miras, RR Lyr type), and white dwarfs ( e.g. , ZZ Cet,V777 Her type).Pulsations provide information on the structure and evolution of stars. Theobserved properties of Classical Cepheids ( δ Cep type stars), such as periods, ∗ Based on observations obtained with the 1.3-m Warsaw telescope at the Las CampanasObservatory of the Carnegie Institution for Science. amplitudes, and colors allow for testing of hydrodynamical models of stars( e.g. , Kanbur, Ngeow and Buchler 2004, Kanbur et al. e.g. , Pietrukowicz2003, Turner, Abdel-Sabour Abdel-Latif and Berdnikov 2006, Poleski 2008). Re-cent studies of period changes show that some Cepheids, including the nearestPolaris, are very likely undergoing enhanced mass loss (Neilson et al. et al. e.g. , Popova 2006, Majaess, Turner and Lane 2009, Bobylev and Bajkova2012). Recent works indicate a steady decrease in metallicity in Cepheids withthe increasing distance from the Galactic center in the thin disk and no corre-lation between the metallicity and the vertical distance from the Galactic plane(Genovali et al. et al. e.g. ,Soszy´nski et al. e.g. , Storm et al. e.g. , Vilardell, Jordi and Ribas 2007, Riess, Fliri and and Valls-Gabaud 2012),M33 ( e.g. , Gieren et al. e.g. , Gieren et al. e.g. , Tammann, Reindl and Sandage 2011), and more distant, like M81 (Freed-man et al. e.g. , Soszy´nski et al. e.g. , Szczygie l, Pojma´nski and Pilecki 2009, Sesar et al. et al. et al. e.g. , Kinemuchi et al. et al. et al. et al. e.g. , Pietrzy´nski et al. et al. et al. et al. e.g. , Whitelock et al. et al. e.g. , in Cen A, Rejkuba 2004).Pulsating variables of δ Sct, γ Dor, β Cep type, and Slowly Pulsating B(SPB) stars showing multiple modes are of particular interest for testing seismicmodels of MS and post-MS stars ( e.g. , Daszy´nska-Daszkiewicz, Dziembowskiand Pamyatnykh 2005, Balona and Dziembowski 2011, Balona et al. e.g. , Li, Xu and Li2010, Breger et al. et al. et al. (2010)proposed to use rotational splitting asymmetries to probe the internal rotationprofile of β Cep type stars.For more than two decades the number of detected pulsating stars has in-creased rapidly. For example, the Optical Gravitational Lensing Experiment(OGLE) – a long-term large-scale sky survey focused on stellar variability (Udal-ski et al. e.g. , Soszy´nski et al.
The observations presented in this paper were collected with the 1.3-m Warsawtelescope at Las Campanas Observatory, Chile, during the third phase of theOptical Gravitational Lensing Experiment (OGLE-III) in years 2001–2009. Theobservatory is operated by the Carnegie Institution for Science. The OGLE-IIICCD camera consisted of eight chips with the total field of view of 35 ′ × ′ andthe scale of 0.26 arcsec/pixel. More details on the instrumentation setup can befound in Udalski (2003).The OGLE-III Galactic disk fields covered an area of 7.12 deg near theGalactic plane between longitudes +288 ◦ and +308 ◦ . Coordinates of the cen-ters, the time coverage, and the number of data points of the monitored fieldsare given in Table 1 and shown in Figs. 1–2 in Pietrukowicz et al. (2013). Asignificant majority of frames (815–2698 per field) were collected in the standard I -band filter with an exposure time of 120 or 180 s. Additional observations,consisting of only 3–8 measurements, were taken in the V -band filter with anexposure time of 240 s. A total number of approximately 8 . × stars withbrightness between I = 12 . I = 21 . etal. et al. − with the help of the F NPEAKS code (written by Z. Ko laczkowski, private commu-nication) for stars containing at least 30 measurements in the I -band. Around345 500 detections were visually inspected for any kind of variability. This wasdone in two stages. In the first stage, we inspected about 100 500 stars hav-ing a variability signal-to-noise ratio S/N ≥
14, and in the second stage about245 000 objects with 10 ≤ S/N <
14. We also performed an additional search forhigh frequencies between 24 d − and 100 d − for stars with I < . V − I < . e.g. , Mr´oz et al. et al. P < .
25 d).We manually removed obvious outliers from the light curves of our stars andimproved their periods using the TATRY code (Schwarzenberg-Czerny 1996).In the next step, we searched the Cepheids and short-period stars for additionalmodes. Based on the periods, period ratios, and amplitudes of the short-periodvariables we selected 57 good candidates for δ Sct type stars. Another candidatefor a δ Sct variable together with a very likely pulsating white dwarf were foundas a result of the high-frequency search.Each variable was calibrated from the instrumental to the standard magni-tudes using transformation relations given in Udalski et al. (2008). Since thenumber of V -band measurements is very small, we rely on average instrumentalbrightnesses h v i and h i i , and color h v i − h i i . For high-amplitude stars, we fitthe instrumental i -band light curve to the instrumental v -band data by adjust-ing the amplitude and phase shift. The average brightness in both passbandswas derived from the fits. In the case of low-amplitude multi-mode pulsatingstars the h v i and h i i values were determined as the arithmetic mean from allmeasurements after converting magnitudes into fluxes. For stars without anyavailable V -band measurement, except LPVs, we assumed an average V − I color of neighboring stars within a radius of 1. ′ et al. (2010). For LPVs without points in V , we assumedan average V − I color determined from the remaining LPVs. For all stars with V − I > . I -band photometry had to be additionally correctedaccording to formula presented in Szyma´nski et al. (2011). The catalog of pulsating stars in the Galactic disk fields, containing tables withbasic parameters, time-series I - and V -band photometry, and finding charts, isavailable to the astronomical community from the OGLE Internet Archive: http://ogle.astrouw.edu.pl and ftp://ftp.astrouw.edu.pl/ogle/ogle3/OIII-CVS/gd/ where one can find the following catalogs on pulsating stars: cep , dsct , lpv , rrlyr , t2cep , wd , and unclassified . The stars are arranged according to increas-ing right ascension and named as OGLE-GD-CEP-NNNN, OGLE-GD-DSCT-NNNN, OGLE-GD-LPV-NNNN, OGLE-GD-RRLYR-NNNN, OGLE-GD-T2CEP-NNNN, and OGLE-GD-WD-NNNN, where NNNN are four-digit consecutivenumbers for each of the type. In the case of short-period unclassified variableswe leave the name as the following catenation: FIELD.CHIP.ID. In the datatables, we provide coordinates of the variables, their pulsation periods, perioduncertainties, information on brightness, and classification to a subtype, if ex-ists. Our search for variable objects brings identification of 20 Classical Cepheids. AllCepheids but one object are new discoveries. Variable OGLE-GD-CEP-0019was detected by the EROS-II collaboration during the Galactic Spiral Arms(GSA) observation program and named EROS2 GSA J133319-640707 (Derue etal. et al.
Fig. 1. Phased light curves of 20 Classical Cepheids detected in the OGLE-III disk area.Objects are sorted according to increasing period.
All Cepheids have the average apparent brightness between I = 13 . I = 16 . V due tounfortunate position just outside the reference image in this band. The observed V − I color of the remaining 19 Classical Cepheids is in a range between 1.1 magand 4.2 mag. Such a wide magnitude range is the result of the non-uniformlydistributed interstellar extinction in the Galactic disk. It is worth to note thatthe monitored area contains other eleven known Classical Cepheids, but all ofthem are saturated in the OGLE data. The observed amplitudes (0.2–0.4 mag) and shape of the light curves of sixstars (Fig. 2) resemble those of Type II Cepheids ( e.g. , Soszynski et al. V − I between 1 . . I vs. V − I diagrams (Fig. 3) is in agreement with the proposedtypes of old population Cepheids. However, our confidence is not full due to thefact that all objects but OGLE-GD-T2CEP-0001 were detected at an angulardistance < ◦ Fig. 2. Phased light curves of six candidates for Type II Cepheids.
Within the twenty-one disk fields we found 45 RR Lyr stars of which 36 are of thetype “ab” (fundamental mode) and nine of the type “c” (first overtone). Lightcurves of the variables are presented in Fig. 4. Object OGLE-GD-RRLYR-0005was detected independently in two overlapping fields CAR109 and CAR115.The increased time span allowed a better determination of the pulsation periodfor this star. Twelve of our RRab variables ( i.e. , one-third of the sample) exhibitthe Blazhko effect (Blaˇzko 1907), which manifests as long-period modulations inthe light curve shape (Smith 2004). The observed Blazhko periods are between21 d and 90 d.We used an empirical relation presented in Smolec (2005) to assess the metal-licity of the RRab stars on the Jurcsik (1995) scale (J95):[Fe / H] J95 = − . − . P + 0 . φ , σ sys = 0 . , where P is the pulsation period and φ = φ − φ is a Fourier phase combinationderived from the I -band light curve. In Fig. 5, we present the obtained metallic-ity distributions for all RRab stars as well as those without the Blazhko effect.The stars form essentially two groups: a metal-rich one with [Fe/H] J95 & − . J95 . − .
8. This division likely reflects the
Fig. 3. Color-magnitude diagrams with marked positions (red points) of candidates for TypeII Cepheids. Field name is given in each panel. Red arrows indicate the direction of reddeningwith an assumed ratio of total-to-selective exinction R I = A I /E ( V − I ) = 1 . presence of stars from two Galactic populations: the thick disk and halo, re-spectively. Such explanation is in agreement with properties of RR Lyr starsobserved in the Galaxy ( e.g. , Kinemuchi et al. et al. We identified 31 new LPVs in the OGLE-III Galactic disk area. In Fig. 6,we present their I -band light curves phased with our best-measured pulsationperiods. Eighteen of the LPVs are Mira stars, while the other thirteen objectsare Semi-Regular Variable (SRV) stars. For ten LPVs the observations areinsufficient to estimate their peak to peak amplitudes in I . Due to the lack of V -band data we were not able to assess the V − I color for four Mira stars. MirasOGLE-GD-LPV-0023 and 0027 have the largest mean observed color among alldetected variables within the OGLE-III disk fields, V − I ≈ . δ Sct type Stars
We classified 58 stars as very likely δ Sct type variables. Fifty-seven of themwere found during the search for frequencies lower than 24 d − and one objectduring the high-frequency search (24–100 d − ). The alternative within therealm of short-period pulsating stars is the β Cep type. The period ranges forthese two types largely overlap, although β Cep stars have much higher massesand effective temperatures. Unfortunately, our data do not allow us to assessthese parameters, mostly because of the heavy and unknown reddening in theobserved regions. Thus, we cannot exclude that some of the stars, which we
Fig. 4. Phased light curves of 45 RR Lyr variables found within the OGLE-III Galactic diskfields. The type of the variables is given in the lower left corner of each panel. classified as δ Sct, may be in fact β Cep stars. However, this seems unlikelybecause amplitudes as high as in our stars ( > . β Cep stars (seeStankov & Handler 2005). Furthermore, the period range in our multi-periodicobjects is wider than typically found in β Cep stars.Twenty-eight stars in our sample are fundamental-mode pulsators (see lightcurves in Fig. 7). Of course we cannot exclude that some other modes areexcited in these objects, but their amplitudes must be much smaller. Stars withone or two dominant radial modes are known as High Amplitude Delta Scutistars (HADS). They are rare. According to Lee et al. (2008) only about 0.3%of the total population of Galactic δ Sct stars belongs to this subtype. Whatcauses that a star chooses this high-amplitude form of pulsation is unknown.
Fig. 5. Distributions of photometrically derived metallicities for all 36 newly detected RRabtype stars (solid line) and a subgroup of 24 variables without the Blazhko effect (dotted line).The metallicity is given in the Jurcsik (1995) scale.Fig. 6. Light curves of 31 LPVs detected in the OGLE-III disk area. Fig. 7. Phased light curves of 28 detected fundamental-mode δ Sct type variables.
In the remaining 30 stars of our sample we detected at least two modes.Their light curves phased with the strongest period are plotted in Fig. 8. Theamplitudes of light variations are large by δ Sct star standards. However, itis not clear if the dominant peaks are due to radial modes. In this work, weassume this to be the case. Only with this optimistic assumption an inferenceon stellar parameters based on periods alone is possible.Twenty-two δ Sct stars exhibit two radial modes (in eighteen cases F and1O), six stars exhibit three radial modes (mostly F, 1O, and 2O), and one star,OGLE-GD-DSCT-0025, seems to have four radial modes (F, 1O, 2O, and 3O, seeperiodograms in Fig. 9). In some of the stars, we detected additional non-radialmodes. Objects with more than two bona fide radial modes are particularlyvaluable in this context. Table 1 lists data for such objects.The identification of the radial orders of excited modes in multi-mode pul-sators was done with the help of the Petersen diagram (shorter-to-longer periodratio vs. the logarithm of the longer period). The diagram, presented in Fig. 10,1
Fig. 8. Light curves of 30 detected multi-mode δ Sct type stars. includes the multi-mode δ Sct as well as double-mode δ Cep type stars. The rela-tions for the former type looks like continuations of the relations for multi-modeCepheids. This has been already noted for numerous variables detected in theLarge Magellanic Cloud (see Fig. 3 in Poleski et al. δ Sct stars are just a low-mass version Cepheids crossing theinstability strip for the first time. What matters is that this continuity supportsthe radial mode hypothesis for modes in δ Sct stars.One of the disk Classical Cepheids, OGLE-GD-CEP-0001, is a first-overtonepulsator with an additional secondary period near 0 . P . About 140 suchCepheids from the Small Magellanic Cloud (SMC) and 30 from the Large Mag-ellanic Cloud (LMC) were reported in Soszy´nski et al. (2008) and Soszy´nski etal. (2010), respectively. The origin of this secondary period remains unknown.An interesting application of the Petersen diagrams is probing of metal abun-2 Fig. 9. Periodograms with marked frequencies after subsequent pre-whitenings (numbered inthe upper left corners) for the quadruple-mode star OGLE-GD-DSCT-0025. Radial modesand their combinations are labeled.
Table 1:
Observed periods for seven δ Sct type stars with at least three radial modesVariable P F P P P OGLE-GD--DSCT- [d] [d] [d] [d]-0018 0.13152629(81) - 0.08140070(20) 0.06896097(53)-0021 0.09766834(39) 0.07692128(44) 0.06186007(56) --0025 0.1740594(19) 0.13310790(70) 0.10646403(51) 0.0896032(29)-0033 0.06330743(12) 0.04894428(14) 0.03949024(70) --0034 0.06369610(25) 0.04993199(9) 0.04106100(26) --0048 0.10740182(5) 0.08293446(1) 0.06655117(3) --0049 0.36141456(19) 0.27314926(19) 0.21699323(26) - dance in stars (see Popielski, Dziembowski and Cassisi 2000, Buchler and Szab´o2007, Buchler 2008, and Soszy´nski et al. δ Sct stars of our sample, which revealpairs of radial modes of consecutive orders. The data are compared with model3
Fig. 10. Petersen diagram for multi-mode pulsators in the OGLE-III disk area. Open symbolscorrespond to Classical Cepheids, filled symbols to δ Sct type stars. Circles refer to double-mode, triangles – triple-mode, and squares – quadruple-mode pulsators. values in the separate diagrams employing the (F,1O) and (1O,2O) pairs shownin Fig. 11. The lines present the values calculated for stellar models along evo-lutionary tracks at selected metallicity parameters, Z , and masses, M , whichare depicted in Fig. 11. The tracks extend over the MS (in some cases) andpost-MS part of the pulsation instability range. The relative hydrogen abun-dance X =0 .
72 and the mixing length parameter α MLT =1 . Z -values is required to explain star positionsin both diagrams. For triple-mode stars OGLE-GD-DSCT-0021 and 0034 inTable 1 we need Z ≈ × − , which corresponds to Population II objects,called SX Phe stars. Similar is required for double-mode object OGLE-GD-DSCT-0032 at log P F ≈ − .
08 and P /P F ≈ . P F ≈ − .
78 and P /P F ≈ . β Cep stars, which are muchmore massive and metal rich may populate this part of the diagram. To explainpositions of stars OGLE-GD-DSCT-0025, 0033, 0048, and 0049 and a number4
Figure 11: Location in the Petersen diagrams of objects with more than one bona fide radial mode detected compared with the model predictions. The leftand right panels refer to the (F,1O) and (1O,2O) pairs, respectively. Circles,triangles, and squares refer to stars with two, three, and four radial modesdetected, respectively. The same evolutionary models were used in both panelsof each row. The numbers shown in the right panels give the metallicity Z andmass in M ⊙ .of double-mode objects, a close to solar metal abundance is needed.With data on three radial mode periods (Table 1), adopting the same condi-tions as for calculations made for the models used in Fig. 11., we may try to de-termine exactly all stellar parameters. Proceeding this way, Moskalik & Dziem-bowski (2005) calculated their seismic models of two triple-mode Cepheids. Wetried the same method and found that it had worked well for some of ourstars but not all of them. The problem may be seen in the upper panels ofFig. 11. Consider for instance star OGLE-GD-DSCT-0021 (log P F ≈ − .
01 and P /P F ≈ . M and Z , thoughwe are close to models with lowest values (solid lines). We found the same fordifferent choices of X , α MLT , and α OV parameters and at moderate rotation.The limited precision of our period calculations in some cases prohibited the useof this method. Thus, as the best models of all seven objects in Table 1 werechosen those minimizing D ≡ qP k (log P obs k − log P mod k ) . Adding the 3O period5for the quadruple-mode star OGLE-GD-DSCT-0025 to that fit has a negligibleeffect on inferred parameters, resulting in the worse fit and the value of D goingup from 0.002 to 0.005.In Table 2, along with model parameters, we list the reddening, E ( V − I ),and distance modulus m − M . These quantities were determined from the meanmagnitudes measured in the two bands and the corresponding absolute magni-tudes interpolated from the tables provided by Castelli and Kurucz (2004). Theobtained distances range from 2.5 to 6.2 kpc, which places the seven variablesin the foreground of stars from the Centaurus spiral arm.Table 2: Parameters of our best models for seven δ Sct type stars with at least threeradial modesVariable
M Z
Age log T eff E ( V − I ) m − M RemarksOGLE-GD--DSCT- [M ⊙ ] [Gyr] [K] [mag] [mag]-0018 1.74 0.00848 1.16 3.862 0.37 13.00 post-MS-0021 0.94 0.00046 5.68 3.825 0.65 13.66 SX Phe-0025 1.92 0.01234 1.02 3.835 0.36 12.00 post-MS-0033 1.45 0.01253 1.51 3.835 0.43 12.94 MS-0034 1.01 0.00044 4.36 3.887 0.90 13.54 SX Phe-0048 1.46 0.00713 1.83 3.842 0.53 13.95 post-MS-0049 2.46 0.01618 0.58 3.789 0.58 13.85 post-MS The main source of uncertainty in the numbers given in Table 2 is the lack ofdata on the relative hydrogen abundance and rotation rate as well as insufficientknowledge about efficiency of convective transport and overshooting. In orderto assess the range of uncertainties, we carried out calculations with varingvalues of X , α MLT , α OV , and v rot . In particular for star OGLE-GD-DSCT-0049, we obtained the following expressions for inferred parameters with anexplicit dependence on these quantities: M = 2 . − . X − .
72) + 0 . α MLT − . − . α OV − . α rot ,Z = 0 . − . X − .
72) + 0 . α MLT − . − . α OV − . α rot , Age = 0 . − . X − .
72) + 0 . α MLT − . − . α OV − . α rot ,E ( V − I ) = 0 . − . X − .
72) + 0 . α MLT − . − . α OV − . α rot ,m − M = 13 . − . X − .
72) + 0 . α MLT − .
5) + 0 . α OV − . α rot , where α rot = (cid:18) v rot
90 km / s (cid:19) . Searching for frequencies between 24 d − and 100 d − led us to the detectionof a star with the dominant period of 0.01962154(1) d=28 . I -band amplitude of 0.35 mag. Its light curveis presented in the last panel of Fig. 8. The object shows two additional peaksequally distant from the dominant mode in the frequency space: ± . − from 50.9644 d − (Fig. 12). The high amplitude at the very short pulsationperiod causes a problem with proper classification of this object. It seems tobe an extreme case of δ Sct type stars. In the revised catalog of 636 δ Sct starsprepared by Rodriguez et al. (2000), nineteen objects have periods below or6around 0.03 d, and only three objects below or around 0.02 d. The correspond-ing maximum V -band amplitudes are 0.06 mag and 0.02 mag, respectively. Asimilar triple-peak structure of the oscillation spectrum was found in the δ Sctstar 1 Mon (V474 Mon): 7 . ± . − . Balona and Stobie (1980) inter-preted that structure as a dipolar ( ℓ = 1) triplet, where the dominant peak isdue to a radial mode and the two side oscillations are likely dipole modes splitby stellar rotation. If the three peaks in our star are due to a dipolar tripletthen ν rot ≈ . − , which implies v rot ∼ R/R ⊙ ≈
150 km/s, well acceptablevalue for a δ Sct type object. Therefore we named this intriguing object asOGLE-GD-DSCT-0058.
Fig. 12. Power spectra of the short-period pulsator OGLE-GD-DSCT-0058 with labelledfrequencies.
In the observed Galactic fields, we detected 60 stars with periods < .
23 d, I -band variability range > .
05 mag, and shapes characteristic for pulsating stars.Fifty-two of the stars show more than one periodicity, but the period ratioscannot be explained by the presence of pure radial modes in δ Sct. In some ofthe stars we observe close, very likely non-radial modes (see Fig. 13). In Fig. 14,we present the I vs. V − I diagram for one of the OGLE fields with markedlocations of the short-period stars. Most of the objects are probably δ Sct typestars, but we cannot exclude that some of them are more distant β Cep typestars. Information on spectral type would help in the final classification of theseobjects.
The high-frequency search (24–100 d − ) led us to the initial detection of nearlyone hundred objects with a signal-to-noise ratio S/N >
10. All but one objectshave V − I > . V − I ≈− .
23 mag has the period of0.01273972(1) d=1100 . I -band amplitude of 0.010 mag. Theobserved color, period, and amplitude are typical for pulsating white dwarfs, forinstance of ZZ Cet type ( e.g. , Eggen 1985, Van Grootel et al. The completeness of the search for pulsating stars depends on their type. All ourClassical Cepheids were detected during the inspection of stars with a signal-to-noise ratio S/N ≥
14. We did not find any new Cepheid among objects with7
Fig. 13. Example light curves of short-period pulsators showing two or more close modes.These stars could be either δ Sct or β Cep type variables, but due to the lack of informationon intrinsic color or surface temperature it is almost impossible to distinguish between thetwo types. ≤ S/N <
14. The discovered Cepheids are bright and have the average I -bandmagnitudes < .
5. Given the brightness and the corresponding rms variabil-ity σ I < .
03 mag (Szyma´nski et al. > . <
14. Sinusoidal-like RRc type objects are difficult to distinguish fromhundreds of regularly variable candidate spotted stars and some pulsators ofthis type could be missed. Their completeness could be as low as 50%.For LPVs the completeness seems to depend on the time coverage of thefields. We did not detect these variables in the fields with only one well-coveredseason, i.e. , CAR106, CEN106, and CEN107. The new LPVs were found in thedisk fields which contain about 83% of all monitored stars. This percentageis a rough estimate of the completeness for high-amplitude regularly pulsatingLPVs, in particular the Mira variables.Proper classification of short-period pulsators based on the
V I photometryis particularly difficult for stars in the Galactic disk fields. However, the resultsfor the fundamental-mode HADS variables seem to be very complete, since 24objects out of the 28 detected fundamental-mode δ Sct stars were found duringthe inspection of stars with the variability signal-to-noise ratio S/N ≥ Fig. 14. Color-magnitude diagram of the field CAR113.4 located about 0. ◦ R I = A I /E ( V − I ) = 1 . Twenty-one fields in the direction tangent to the Centaurus Arm of the Galacticdisk were observed by the OGLE survey during the third phase of the projectin years 2001–2009. The analysis of the data has brought an identification of20 bona fide
Classical Cepheids, 45 RR Lyr type stars, 31 LPVs, one pulsat-ing white dwarf, and 58 very likely δ Sct type stars. Only one of the diskvariables, OGLE-GD-CEP-0019, had been identified before. Ten Cepheids arefundamental-mode, two are first-overtone, and eight are double-mode pulsators,mainly F/1O and 1O/2O. One Cepheid, OGLE-GD-CEP-0001, pulsates in thefirst overtone and a second mysterious mode. All detected Classical Cepheids9are brighter than I = 16 . δ Sct type stars pulsate in the fundamentalmode. In 22 stars of this type we found two radial modes, in six stars threeradial modes, and in one object four radial modes. Many stars exhibit additionalnon-radial modes. The presence of more than two radial modes in seven starsallowed the determination of their distance and physical parameters based onpulsation models. We show that two triple-mode and one double-mode stars arevery likely Population II pulsators of SX Phe type. We note that star OGLE-GD-DSCT-0012 cannot be reproduced by our models. It could be either a δ Scttype star pulsating in some non-radial modes or, less probable, a β Cep typestar. Information on spectral type would help in the definitive classification ofthis and another 60 short-period pulsating variables found in the OGLE diskarea. Among the detected stars there is also a candidate δ Sct star with anexceptionally high full I -band amplitude of 0.35 mag at the very short periodof 0.0196 d.The detection completeness of Classical Cepheids, RRab type stars, andfundametal-mode δ Sct type stars with amplitudes > . I <
18 mag is assessed at the level of 100%. In the case of the RRc variablesseveral stars could be missed. The search for LPVs was estimated to be completein about 83%.The OGLE photometry in the disk fields allowed a confident classification ofClassical Cepheids, RR Lyr, most of LPVs and some δ Sct type variables. How-ever, many other types of pulsating variables, besides β Cep type stars, are verylikely present in the data. Among them could be low-amplitude Cepheids, SPBstars, γ Dor type stars, hybrid β Cep/SPB (Pamyatnykh 1999), δ Sct/ γ Dorstars (Grigahc`ene et al.
The detection of new Classical Cepheids is one of the aims of a new Galacticsurvey conducted in the framework of the OGLE-IV project at the 1.3-m War-saw telescope. The survey covers more than half of the Galactic plane withlongitudes from − ◦ to +60 ◦ and latitudes − ◦ . b . +3 ◦ . By taking 30 sexposures we plan to find variables in a brightness range 10 < I <
18 mag. InFig. 16, we present a color image of almost entire fourth quarter of the Galac-tic disk with the overlaid location of the OGLE-III and OGLE-IV fields. Thetotal OGLE-IV disk area is about 200 times larger that the area analyzed inthis paper. With this coverage we plan to achieve a nearly complete census ofhigh-amplitude Classical Cepheids.For the purpose of the new survey we compiled data on known GalacticClassical Cepheids from the literature. The list of confirmed Classical Cepheidsfrom the General Catalogue of Variable Stars (GCVS, Samus et al. et al. † . We used photometric data from the Northern Sky Variability Survey(NSVS, Wo´zniak et al. P = 68 .
464 d. The compiled list of 841 GalacticClassical Cepheids is also available in the on-line OGLE archive (see links inSection 3).Fig. 17 shows the distribution of the Galactic Classical Cepheids in the ( l , b )coordinates. Approximately 85% of the stars are located within a − ◦ < b < +5 ◦ stripe and ≈
64% within a − ◦ < b < +3 ◦ stripe along the Galactic plane. Fromthe distribution we can infer that obscured Milky Way plane regions in the firstand fourth quarters ( − ◦ ≤ l < +90 ◦ ) may hide many (probably at least severalhundreds) not yet detected Cepheids.In Fig.18, we present the histogram of the pulsation periods of known funda-mental-mode Galactic Classical Cepheids and compare it with period distribu-tions for stars in the SMC (Soszy´nski et al. et al. et al. ≈ . ≈ . ≈ . Acknowledgments.
The authors would like to thank Richard I. Ander-son for discussions on the presented project, Grzegorz Pojma´nski for his help inclose inspection of selected ASAS objects, and Thomas Fruth for providing theBEST-II survey photometric data on candidates for Galactic Cepheids. Thiswork has been supported by the Polish Ministry of Sciences and Higher Ed-ucation grants No. IP2012 005672 under the Iuventus Plus program and No.IdP2012 000162 under the Ideas Plus program. W.D. is supported by the Pol-ish National Science Centre through grant No. DEC-2012/05/B/ST9/03932.The OGLE project has received funding from the European Research Councilunder the European Community ′ s Seventh Framework Programme (FP7/2007-2013)/ERC grant agreement No. 246678 to A.U. REFERENCES
Akhter, S., Da Costa, G. S., Keller, S. C., and Schmidt, B. P. 2012,
Astrophys. J. , , 23.Alard, C., and Lupton, R. H. 1998, Astrophys. J. , , 325.Balona, L. A., and Stobie, R. S. 1980, MNRAS , , 931.Balona, L. A., and Dziembowski, W. A. 2011, MNRAS , , 591.Balona, L. A., et al. MNRAS , , 3028.Bernhard, K., and Wils, P. 2009, Peremennye Zvezdy Prilozhenie , , 5.Blaˇzko, S. 1907, Astron. Nachr. , , 327.Bobylev, V. V., and Bajkova, A. T. 2012, Astronomy Letters , , 638.Bono, G., Marconi, M., Cassisi, S., Caputo, F., Gieren, W., and Pietrzy´nski, G 2005, Astrophys. J. , , 966.Breger, M., et al. MNRAS , , 1721.Buchler, J. R., and Szab´o, R. 2007, Astrophys. J. , , 723. † Buchler, J. R. 2008,
Astrophys. J. , , 1412.Castelli, F., and Kurucz R. L. 2004, , , astro-ph/0405087.Daszy´nska-Daszkiewicz, J., Dziembowski, W. A., and Pamyatnykh, A. A. 2005, Astron.Astrophys. , , 641.Derue, F., et al. Astron. Astrophys. , , 149.Eggen, O. J. 1985, P.A.S.P. , , 1029.Freedman, W. L. et al. Astrophys. J. , , 628.Genovali, K., et al. Astron. Astrophys. , , A132.Gieren, W., Pietrzy´nski, G., Walker, A., Bresolin, F., Minniti, D., Kudritzki, R.-P., Udalski,A., Soszy´nski, I., Fouqu´e, P., Storm, J, and Bono, G. 2004, Astron. J. , , 1167.Gieren, W., et al. Astrophys. J. , , 69.Greco, C., Clementini, G., Catelan, M., Held, E. V., Poretti, E., Gullieuszik, M., Maio, M.,Rest, A., DeLee, N., Smith, H. A., and Pritzl, B. J. 2009, Astrophys. J. , , 1323.Grigahc`ene, A. et al. Astrophys. J. , , L192.Jurcsik, J. 1995, Acta Astron. , , 653.Kanbur, S. M., Ngeow, C.-C., and Buchler, J. R. 2004, MNRAS , , 212.Kanbur, S. M., Marconi, M., Ngeow, C., Musella, I., Turner, M., James, A., Magin, S., andHalsey, J. 2010, MNRAS , , 695.Kinemuchi, K., Smith, H. A., Wo´zniak, P. R., and McKay, T. A. 2006, Astron. J. , , 1202.Kinman, T. D., Morrison, H. L., and Brown, W. R. 2009, Astron. J. , , 3198.Lee, Y.-H., Kim, S. S., Shin, J., Lee, J., and Jin, H. 2008, PASJ , , 551.Lemasle, B. et al. . 2013, Astron. Astrophys. , , A31.Li, Z. P., Xu, Y., and Li, Q. S. 2010, P.A.S.P. , , 1432.Luck, R. E., Andrievsky, S. M., Kovtyukh, V. V., Gieren, W., and Graczyk, D. 2011, Astron.J. , , 51.Majaess, D. J., Turner, D. G., and Lane, D. J. 2009, MNRAS , , 263.Mateu, C., Vivas, A. K., Downes, J. J., Brice˜no, C., Zinn, R., and Cruz-Diaz, G. 2012, MNRAS , , 3374.Menzies, J. W., Whitelock, P. A., Feast, M. W., and Matsunaga, N. 2010, MNRAS , , 86.Miceli, A., Rest, A., Stubbs, C. W., Hawley, S. L., Cook, K. H., Magnier, E. A., Krisciunas,K., Bowell, E., and Koehn, B. 2008, Astrophys. J. , , 865.Moskalik, P., and Dziembowski, W.A. 2005, Astron. Astrophys. , , 1077.Mr´oz, P, Pietrukowicz, P., Soszy´nski, I., Udalski, A., Poleski, R., Szyma´nski, M. K., Kubiak,M., Pietrzy´nski, G., Wyrzykowski, L., Ulaczyk, K., Koz lowski, S., and Skowron, J2013, Acta Astron. , , 135.Neilson, H. R., Langer, N., Engle, S, G., Guinan, E., and Izzard, R. 2012, Astrophys. J. , ,L18.Pamyatnykh, A. A. 1999, Acta Astron. , , 119.Papar´o, M., et al. Astron. Astrophys. , , A27.Pietrukowicz, P. 2003, Acta Astron. , , 63.Pietrukowicz, P., Udalski, A., Soszy´nski, I., Nataf, D. M., Wyrzykowski, L., Poleski, R.,Koz lowski, S., Szyma´nski, M. K., Kubiak, M., Pietrzy´nski, G., and Ulaczyk, K. 2012, Astrophys. J. , , 169.Pietrukowicz, P., Mr´oz, P., Soszy´nski, I., Udalski, A., Poleski, R., Szyma´nski, M. K., Kubiak,M.,Pietrzy´nski, G., Wyrzykowski, L., Ulaczyk, K., Koz lowski, S., and Skowron, J.2013, Acta Astron. , , 115.Pietrzy´nski, G., Gieren, W., Szewczyk, O., Walker, A., Rizzi, L., Bresolin, F., Kudritzki, R.-P.,Nalewajko, K., Storm, J., Dall’Ora, M., and Ivanov, V. 2008, Astron. J. , , 1993.Pojma´nski, G. 1998, Acta Astron. , , 35.Pojma´nski, G. 2002, Acta Astron. , , 397.Pojma´nski, G. 2003, Acta Astron. , , 341.Pojma´nski, G. and Maciejewski, G. 2004, Acta Astron. , , 153.Pojma´nski, G. and Maciejewski, G. 2005, Acta Astron. , , 97.Pojma´nski, G., Pilecki, B., and Szczygie l, D. M. 2005, Acta Astron. , , 275.Poleski, R. 2008, Acta Astron. , , 313.Poleski, R., Soszy´ski, I., Udalski, A., Szyma´nski, M. K., Kubiak, M., Pietrzy´nski, G.,Wyrzykowski, L, Szewczyk, O., and Ulaczyk, K. 2010, Acta Astron. , , 1.Popielski, B. L., Dziembowski, W. A., and Cassisi, S. 2000, Acta Astron. , , 491.Popova, M. ´E. 2006, Astronomy Letters , , 244.Rejkuba, M. 2004, Astron. Astrophys. , , 903.Riess, A. G., Fliri, J., and Valls-Gabaud, D. 2012, Astrophys. J. , , 156.Rodr´ıguez, E., L´opez-Gonz´alez, M. J., and L´opez de Coca, P. 2000, Astron. Astrophys. Suppl. Ser. , , 469.Rodr´ıguez-L´opez, C., MacDonald, J., and Moya, A. 2012, MNRAS , , L44.Samus, N. N., et al. Sarajedini, A., Mancone, C. L., Lauer, T. R., Dressler, A., Freedman, W., Trager, S. C.,Grillmair, C., and Mighell, K. J. 2009,
Astron. J. , , 184.Schwarzenberg-Czerny, A. 1996, Astrophys. J. , , L107.Sesar, B., et al. Astrophys. J. , , 717.Smith, H. A. 2004, Camb. Astrophys. Ser. , , “RR Lyrae Stars”,Cambridge University Press.Smolec, R. 2005, Acta Astron. , , 59.Soszy´nski, I., Poleski, R., Udalski, A., Szyma´nski, M. K., Kubiak, M., Pietrzy´nski, G.,Wyrzykowski, L., Szewczyk, O., and Ulaczyk, K. 2008, Acta Astron. , , 163.Soszy´nski, I., Poleski, R., Udalski, A., Szyma´nski, M. K., Kubiak, M., Pietrzy´nski, G.,Wyrzykowski, L., Szewczyk, O., and Ulaczyk, K. 2010, Acta Astron. , , 17.Soszy´nski, I., Dziembowski, W. A., Udalski, A., Poleski, R., Szyma´nski, M. K., Kubiak, M.,Pietrzy´nski, G., Wyrzykowski, L., Ulaczyk, K., Koz lowski, S., and Pietrukowicz, P.2011a, Acta Astron. , , 1.Soszy´nski, I., Udalski, A., Pietrukowicz, P., Szyma´nski, M. K., Kubiak, M., Pietrzy´nski, G.,Wyrzykowski, L, Ulaczyk, K., Poleski, R., and Koz lowski, S. 2011b, Acta Astron. , ,285.Soszy´nski, I., Udalski, A., Szyma´nski, M. K., Kubiak, M., Pietrzy´nski, G., Wyrzykowski, L,Ulaczyk, K., Poleski, R., Koz lowski, S., Pietrukowicz, P., and Skowron, J. 2013, ActaAstron. , , 21.Stankov, A., and Handler, G. 2005, Astrophys. J. Suppl. Ser. , , 193.Storm, J., Gieren, W., Fouqu´e, P., Barnes, T. G., Soszy´nski, I., Pietrzy´nski, G., Nardetto, N.,and Queloz, D. 2011, Astron. Astrophys. , , 95.Su´arez, J. C., Andrade, L., Goupil, M. J., and Janot-Pacheco, E. 2010, Astron. Nachr. , ,1073.S¨uveges, M. et al. MNRAS , , 2528.Szczygie l, D. M., Pojma´nski, G., and Pilecki, B. 2009, Acta Astron. , , 137.Szyma´nski, M. K., Udalski, A., Soszy´nski, I., Kubiak, M., Pietrzy´nski, G., Poleski, R.,Wyrzykowski, L., and Ulaczyk, K. 2010, Acta Astron. , , 295.Szyma´nski, M. K., Udalski, A., Soszy´nski, I., Kubiak, M., Pietrzy´nski, G., Poleski, R.,Wyrzykowski, L., and Ulaczyk, K. 2011, Acta Astron. , , 83.Tammann, G. A., Reindl, B., and Sandage, A. 2011, Astron. Astrophys. , , A134.Turner, D. G., Abdel-Sabour Abdel-Latif, M, and Berdnikov, L. N. 2006, P.A.S.P. , , 410.Udalski, A., Szyma´nski, M., Kaluzny, J., Kubiak, M., and Mateo, M. 1992, Acta Astron. , ,253.Udalski, A. 2003, Acta Astron. , , 291.Udalski, A., Szyma´nski, M. K., Soszy´nski, I, and Poleski, R. 2008, Acta Astron. , , 69.Ulaczyk, K., Szyma´nski, M. K., Udalski, A., Kubiak, M., Pietrzy´nski, G., Soszy´nski, I.;Wyrzykowski, L., Poleski, R., Gieren, W., Walker, A. R., and Garc´a-Varela, A. 2013, Acta Astron. , , 159.Van Grootel, V., Dupret, M.-A., Fontaine, G., Brassard, P., Grigahc`ene, A., and Quirion,P.-O. 2012, Astron. Astrophys. , , A87.Vilardell, F., Jordi, C., and Ribas, I. 2007, Astron. Astrophys. , , 847.Whitelock, P. A., Menzies, J. W., Feast, M. W., Matsunaga, N., Tanab´e, T., and Ita, Y. 2009, MNRAS , , 795.Wo´zniak, P. R. 2000, Acta Astron. , , 421.Wo´zniak, P. R., et al. Astron. J. , , 2436.Yang, S.-C., Sarajedini, A., Holtzman, J. A., and Garnett, D. R. 2010, Astrophys. J. , ,799. Fig. 16. Map of the OGLE-IV Galactic Disk Survey fields with longitudes between − ◦ and0 ◦ . Dashed white lines mark the equatorial and Galactic coordinate grids. The whole OGLE-IV disk survey covers the Galactic plane between longitudes − ◦ . l . +60 ◦ and latitudes − ◦ . b . +3 ◦ . Green filled squares denote the location of the OGLE-III fields analyzed inthis paper.4