EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 666 (October 2022) A&A, 666 (2022) A136 Full HTML
Open Access
Issue
A&A
Volume 666, October 2022
Article Number A136
Number of page(s) 15
Section Stellar structure and evolution
DOI https://doi.org/10.1051/0004-6361/202244006
Published online 17 October 2022

© P. Harmanec et al. 2022

Licence Creative CommonsOpen Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article is published in open access under the Subscribe-to-Open model. Subscribe to A&A to support open access publication.

1. Introduction

The B0.5IV star V1294 Aql (also known as HD 184279, BD+03°4065, HIP 96196, SAO 124788, MWC 319; , δ2000.0 = +03°45′40779) is one of a few Be stars for which photometric variations over a few decades are rather well documented. This is so thanks to the curious fact that the star was recommended as a secondary standard for UBV photometry by Johnson & Harris (1954) and Johnson (1955) and therefore was relatively often observed by various observers. When Dahn & Guetter (1973) and Tempesti & Patriarca (1976) reported pronounced light variations of this ‘secondary standard’, other research teams reanalysed their photometric observations to document the variations further. This early history has been summarised in detail by Horn et al. (1982), who confirmed an earlier suspicion by Ballereau & Hubert-Delplace (1982) that a correlation exists between light and spectral changes. They collected the records about the presence of Balmer emission and compared them to the light and colour variations based on historical records and based also on their new UBV photometry from Hvar Observatory. They noted that the observed cyclic long-term variability contradicts what is usually observed for the long-term V/R changes of the Balmer emission lines and expressed doubts about their interpretation by the model of a revolving elongated disk. Comparing the light curve of V1294 Aql to similar light curves of BU Tau and V744 Her, they noted that the development of a new shell phase in all three objects was accompanied by a pronounced light decrease. This type of variability was later classified as an inverse correlation between the brightness and emission strength by Harmanec (1983) and was interpreted as an essentially geometrical effect for situations in which the circumstellar envelope in question is seen more or less equator-on. A study of the disk models, which grow in size and/or density with time, by Sigut & Patel (2013) basically confirmed this conjecture. The correlation was documented by Horn et al. (1983) throughout the entire interval of 50 years covered by spectral records with the help of a long series of photographic magnitudes.

The first detailed spectroscopic study of V1294 Aql was published in two papers by Ballereau & Chauville (1987) and Ballereau & Chauville (1989). They demonstrated the presence of cyclic long-term V/R variations of the double Balmer emission lines with a cycle length of some 6 yr and concluded that they are caused by a slow revolution of an elongated envelope (the model of McLaughlin 1961a,b). They also investigated the possibility whether the star might be a spectroscopic binary, but found no evidence for it. Moujtahid et al. (1998) studied the correlation of the second Balmer jump with visual brightness and long-term radial-velocity (RV) changes for a number of Be stars. For V1294 Aql, they concluded that the Balmer jump parameter D1 correlated with both these quantities, attaining maximum in 1979, when the long-term cyclic RV variations were also at maximum.

Mennickent et al. (1997) discussed possible observational tests of the models of long-term cyclic variations of the Be star disks. Investigating spectral, light, and colour variations of V1294 Aql over the time interval from JD 2 440 000 to 2 450 000, they noted that the brightness extrema corresponded to the phase transitions of the V/R variations from V/R >  1 to V/R <  1 and vice versa, but admitted that this behaviour was not seen during the third V/R transition over the investigated interval of time. They tentatively concluded, however, that the disk of V1294 Aql is seen more or less equator-on (in agreement with Horn et al. 1982) and that the revolution of its elongated structure (“one-armed global oscillation”) projected against the star during the phase transitions causes the observed light decreases through attenuation of the stellar flux.

Chini et al. (2012) carried out a spectroscopic survey of 249 O-type and 581 B-type stars in a search for the duplicity. Based on variable RV, they reported that V1294 Aql is a single-line spectroscopic binary. In a search for hot subdwarf companions to Be stars in the spectra obtained by the International Ultraviolet Explorer (IUE), Wang et al. (2018) reported a null detection for such a companion to V1294 Aql. Brandt (2021) cross-corelated the HIPPARCOS and Gaia catalogues in an effort to identify astrometrically accelerating objects. V1294 Aql was identified as a candidate for a system with a faint companion.

2. Observations and reductions

2.1. Spectroscopy

New electronic spectra were obtained at four observatories: with the Ondřejov 2 m reflector (OND) and a coudé spectrograph, with the Dominion Astrophysical Observatory 1.22 m reflector (DAO) and a coudé spectrograph, with the Cerro Armazones 1.5 m Hexapod Telescope (HPT), the Bochum Echelle Spectroscopic Observer (BESO) spectrograph, which is similar to FEROS (Fuhrmann et al. 2011), and with the Cerro Tololo Inter-American Observatory (CTIO) 1.5 m reflector with the CHIRON echelle spectrograph (Tokovinin et al. 2013). We also used one archival ESO FEROS echelle spectrum (Kaufer et al. 1997; Kaufer & Pasquini 1998), the medium-resolution CCD spectra obtained and published by Christian Buil2, and a selection of amateur CCD spectra with resolutions better than 10 000 from the BeSS spectroscopic database (Neiner et al. 2011). Table 1 lists a journal of all spectral observations used.

Table 1.

Journal of electronic spectra.

The initial reduction of all Ondřejov and DAO spectra (bias subtraction, flat-fielding, creation of 1D spectra, and wavelength calibration) was carried out in IRAF. Initial reduction of the HTP and CTIO spectra was carried out at the respective observatories. Rectification, removal of residual cosmics and flaws, and RV measurements of all spectra were carried out with the Pascal program SPEFO (Horn et al. 1996; Škoda 1996), namely the latest version 2.63 developed by J. Krpata (Krpata 2008). SPEFO displays direct and flipped traces of the line profiles superimposed on the computer screen that the user can slide to achieve a precise overlapping of the parts of the profile for which the RV is to be measured. All RVs measurements were carried out independently by PH and also by AH, who studied the spectra as a part of his student’s research project. In addition to the wings of the Hα and He I 6678 emissions, we also measured the bottom of the (sometimes asymmetric) cores of the Hα, He I 6678, Si II, and Fe II shell absorptions. Both sets of measurements were intercompared, the larger deviations were carefully checked, and the mean of the measurements for each line was then used here. Only after these RV measurements were completed was a new program for spectral reduction reSPEFO, a modern replacement of SPEFO, written in JAVA and running on different platforms (Linux, Windows) developed by A. Harmanec3. It can, among other things, import the spectra that were originally reduced in SPEFO and treat spectra stored as FITS files. We used this new program to measure line intensities, equivalent widths, and the V/R ratio of the double Hα emission.

2.2. Photometry

We attempted to collect all available observations with known dates of observations. Basic information about all data sets can be found in Table 2, and the details of the photometric reductions and standardisation are described in Appendix A.

Table 2.

Journal of available photometry with known dates of observations.

For the convenience of other investigators, we also publish all our individual observations together with their HJDs. All measured RVs are listed in Table 3, spectrophotometric quantities are provided in Table 4, and photometric observations are collected in Table 5. These three tables are only available in electronic form at the CDS.

3. Long-term spectral, light, and colour changes of V1294 Aql

As mentioned above, we attempted to collect and homogenise all available photometric observations and measurements of RVs and line strengths from the records with known dates of observations. Long-term behaviour of these quantities and their mutual correlations are discussed in the following sub-sections.

3.1. Value of visual estimates of brightness

The visual estimates of brightness by skilful amateur observers are commonly used to determine the times of minima (or maxima) of periodic variables and to monitor light changes of variables with a large amplitude (over several magnitudes). The scatter band of visual estimates is typically about 01 to 015, but it can be pushed down by talented individuals who, moreover, follow a few principles: (a) to perform only one visual estimate per night (without recalling the previous one), and (b) to reduce the estimates to Johnson V magnitudes of the comparison stars (known to one thousandth of a magnitude), not to the Harvard scale of magnitudes, which is only accurate to 01. One of us (SO) contacted the first author of this study back in 2003 to inform him that he had observed another light decrease of V1294 Aql for about 03. We then agreed to test his ability to obtain accurate visual estimates via parallel photoelectric observations at Hvar, San Pedro Mártir (SPM), Tubitak National Observatory (TUG), and Çanakkale. Figure 1 shows the comparison of visual estimates and Johnson V photoelectric photometry from several stations over the time interval covered by visual estimates. The visual estimates agree very well with the general trend of variations that were recorded via photoelectric photometry, but the deep light minimum appears broader than that recorded by photoelectric photometry. One possible reason is that the minimum was observed close to the end of visibility of the star in the sky and the visual estimates were not corrected for the differential extinction.

thumbnail Fig. 1.

Time plot of the Otero visual brightness estimates (empty red circles) along with Johnson V photometry from Hvar (blue dots), SPM (green), Çanakkale (cyan), and TUG (magenta).

3.2. Correlation between the long-term light and spectral changes in time

Figure 2 is a time plot of all available observations secured in or transformed into the Johnson UBV magnitudes. The top panel shows that the usual brightness level in V is occasionally disturbed by rapid light decreases of different durations. Moreover, we note that there is also a secular, steady slow light decrease of the undisturbed brightness of the star outside the more rapid light decreases until about HJD 2455000, when it suddenly changed to a steeper secular light increase. We return to this new phenomenon in a separate section below. The second panel of Fig. 2 shows that the B−V index followed the brightness changes, but with a small amplitude, while the U−B index showed a similar pattern of changes, but with a larger amplitude and with a more or less steady secular reddening.

thumbnail Fig. 2.

Time plot of available observations over the whole time interval of about 25 000 d covered by the data. Top panel: yellow brightness observations, which could be transformed into Johnson V magnitude. Second and third panel: available B−V and U−B colour index observations. Bottom panel: V/R changes in the peaks of the double Hα emission. In the three panels with photometry, the differential observations are shown as blue dots, and all-sky observations are shown as black dots. Bottom panel: blue circles denote the DAO spectra, red circles show the OND spectra, green circles represent BESO spectra, magenta circles show BeSS spectra, black circles show Castanet Tolosan spectra, and the black crosses plot the data from the literature.

An enlarged Fig. 3 covers only the more recent time interval, when electronic spectra became available. It shows that in the time intervals that are sufficiently densely covered by the data, the sharp light decreases are accompanied by similarly sharp strengthening in the Hα emission.

thumbnail Fig. 3.

Time plot showing the correlation between the secular brightness and colour variations in the V-band magnitudes and B−V and U−B colour indices (black shows all-sky and blue shows differential observations), and the V/R changes, EW, and strength of the Hα emission for the more recent electronic spectra. The colour symbols for spectra from different sources are the same as in Fig. 2.

Figure 4 shows the V/R changes of Si II 6347 Å line. It reveals a pattern similar to that observed for Hα, but the time coverage is less dense because these variations could not be measured in time intervals when the emission was faint.

thumbnail Fig. 4.

Time plot of the V/R changes recorded for the electronic spectra in the Si II 6347 Å line. The colour symbols for spectra from different sources are the same as in Fig. 2.

Figure 5 shows the variation of the shell absorption RVs (characterised by He I RVs, for which data are also published by Ballereau & Chauville 1989) and emission-line RVs measured on the wings of the Hα line in electronic spectra. We note that while the shell RVs shows large cyclic changes that have also been observed for a number of other Be stars, the RVs measured on the wings of the Hα emission are secularly stable and show only mild changes on a shorter timescale.

thumbnail Fig. 5.

Time plot of radial velocities. Top panel: RV of the He I shell lines. Bottom panel: RV measured on the steep wings of the Hα emission. Data from individual instruments are shown by different symbols: the circles are the same as in Fig. 2, and black crosses plot data from Ballereau & Chauville (1989). The ranges on the two axes in the two plots are different.

3.3. Unusual colour variations

Many systematically studied Be stars are known to exhibit always the same and a rather clear type of either a positive or an inverse correlation between the long-term brightness variations, a characteristic type of behaviour in the colour-colour diagram, and the Balmer emission-line strength as defined by Harmanec (1983, 2000). He identified these two types of correlation as an aspect effect. For Be stars with an inverse type of correlation, light decreases are followed by the rise of the Balmer emission-line strength and by a shift along the main sequence towards later spectral subclasses in the U−B versus B−V diagram. For a positive type of correlation, the brightenings are followed by the rise of the emission strength and a shift from the main sequence towards the supergiant sequence in the U−B versus B−V diagram. The inverse correlation is observed for stars that are seen more or less equator-on (a growing gaseous envelope is attenuating the light of the central object), while the positive correlation is observed for stars that are seen more pole-on (inner optically thick parts of the growing envelope mimic an apparent increase in the stellar radius). Several examples of both types of correlation in the colour-colour diagram can be found, for instance, in Fig. 2 of Božić et al. (2013).

The situation is dramatically different for V1294 Aql. The UBV observations accumulated over several decades cover a large part of the whole colour-colour diagram, without a single clear pattern.

To understand better what is going on, we investigated the colour-colour diagrams for different segments of the long-term changes. Figure 6 shows the colour changes separately for the old data secured before JD 2450000, for more recent data from the secular brightness increase (observations after JD 2457000), and for observations covering two episodes of a rapid increase and decrease in the Hα emission associated with sharp light decreases. The pattern is remarkably similar for both these episodes. Formally, it looks like a positive correlation. However, the phases of minimum brightness and maximum strength of the emission correspond to data that are close to the main sequence, even below it when the reddening is considered. The older data are all clustered above the supergiant sequence, while the recent data lie along the supergiant sequence for late-B and early-A spectral classes. All this indicates that we observe a combination of several different types of long-term changes.

thumbnail Fig. 6.

U−B vs. B−V diagram for several distinct data subsets. Top panels: older data until JD 2450000 (left). All-sky observations are shown as black circles, and data from stations defined in Table 2 are denoted as follows: 01 (blue), 04 (green), 12 (red), and 26 (magenta). More recent data from the interval of secular light brightening, all from station 01 (blue) (right). Bottom panels: show UBV observations from the two sharp increases in the emission-line strength accompanied by light decreases. Time interval JD 2452741 – 2453309, which covers the first sharp light decrease (left, cf. Figs. 1 and 3.) Data from the time interval JD 2455357 – 2456094 corresponding to the second sharp increase in the emission-line strength (right). Data from stations 1, 30, 66, and 89 of Table 2 are shown by blue, red, green, and magenta dots, respectively. The main sequence and the supergiant sequence based on data from Golay (1974) (pp. 79–80) are shown, as is the reddening line.

4. Duplicity of V1294 Aql

The idea that duplicity can be an important factor for the very existence of the Be phenomenon is not new. Plavec & Horn (1969), Kříž (1969) and Plavec (1970) have argued that at least some Be stars could be binaries that are observed in the later phases of mass exchange between the binary components. Kříž & Harmanec (1975) and Harmanec & Kříž (1976) formulated the general hypothesis that Be stars are mass-accreting components of binaries and showed that this idea can also explain several types of time variations observed for Be stars. Additional arguments were provided by Plavec (1976) and Peters (1976). However, as pointed out already by Plavec (1976), if all Be stars have Roche-lobe filling secondaries, more eclipsing binaries should be observed among them. Later investigations also led to the finding that the presence of Roche-lobe filling secondaries can be excluded for some Be stars that were found to be spectroscopic binaries, such as V744 Her = 88 Her (Doazan et al. 1982a,b) or V439 Her = 4 Her (Heard et al. 1975; Harmanec et al. 1976). This led Pols et al. (1991) to suggest that many Be stars might be objects created by large-scale mass transfer that were observed in phases after the mass transfer ceased. The expected secondaries of such objects would be hot compact stars, white dwarfs in some cases. These are the most easily detectable sources in far-UV spectra. Evidence for a hot secondary to the well-known Be binary φ Per was found first from the antiphase variation in the He II 4686 Å emission seen in the photographic spectra (Poeckert 1981) and later from He I 6678 Å emission in the electronic spectra obtained by Gies et al. (1993). Its ultimate direct detection as an O VI subdwarf came from the study of the far-UV spectra from the Hubble Space Telescope by Gies et al. (1998). The secondary was then resolved with optical spectro-interferometry by Mourard et al. (2015). Detections for several other systems followed. Wang et al. (2018) carried out a systematic search for the presence of hot secondaries and summarised our knowledge of already known cases. Wang et al. (2021) detected nine new Be+sdO binaries from analyses of the Hubble Space Observatory spectra, and Klement et al. (2022) reported the first interferometric detection and signatures of the orbital motion for three known Be+sdO systems. On the other hand, Bodensteiner et al. (2020) carried out a systematic search for Be stars with main-sequence secondaries, with a completely null result. This constitutes indirect evidence that the mass exchange is or was behind the formation of binaries with Be primaries. Hastings et al. (2021) carried out evolutionary calculations of mass exchange in binaries in an effort to set some limits on the fraction of Be stars produced by binary interaction. They found that under certain conditions, this fraction can be quite high.

One always has to be cautious when analysing binaries with clear signatures of the presence of circumstellar matter in the system. The experience from our previous studies of individual Be stars (Božić et al. 1995; Koubský et al. 1997; Harmanec et al. 2000, 2002a; Linnell et al. 2006; Ruždjak et al. 2009) shows that the binary nature of particular Be stars is most easily detected via periodic RV variations of the steep emission wings of the Hα line and often also via the periodic changes in the V/R ratio of the double Balmer emission lines.

Period analyses of all Hα emission-line RVs of V1294 Aql, using both the Deeming (1975) and Stellingwerf (1978) methods, revealed that the RV of the Hα emission wings varies with a period of 193 d and a semi-amplitude of ∼5 km s−1. The same periodicity is also detected in the RV of the Hα absorption core and in the absorption RVs of Si II doublet at 6347 and 6371 Å and Fe II 6456 Å after long-term changes are removed.

Using the program FOTEL (Hadrava 1990, 2004), we derived the circular-orbit orbital elements for all Hα emission-wing RVs and for those from the high-resolution spectra alone. The mutual agreement of the two solutions is very satisfactory. They are presented in Table 6, and the corresponding RV curve is plotted in Fig. 7. In the rest of this study, we adopt the following linear ephemeris:

(1)

based on the solution for all spectra.

thumbnail Fig. 7.

Radial-velocity curve corresponding to the orbital solution, based on RVs measured on the steep wings of the Hα emission, plotted for phases from ephemeris (top; eq. (1)). Orbital curve based on RV of the He I 6678 Å shell line prewhitened for the long-term changes, plotted for the same ephemeris (bottom). Data from individual instruments are shown by different symbols. The circles are the same as in Fig. 2, and black triangle shows CTIO.

Table 6.

Orbital solutions based on the Hα emission RVs.

To be fair, we note that some deviations from the mean RV curve in the upper panel of Fig. 7 are rather large. This is, for instance, the case of two Ondřejov spectra taken on HJD 2457128.6 and 2457137.6, when a very steep rise of the emission strength had occurred. We remeasured these spectra several times, but the result was the same. Their RVs are almost in anti-phase to the orbital RV curve near phase 0.3. We show the corresponding line profiles in Fig. 8 together with another profile, taken about two weeks later, which already gives a RV in accord with the orbital motion. The two peculiar RVs were given zero weight in the orbital solution.

thumbnail Fig. 8.

Three Ondřejov Hα profiles. They are mutually shifted in ordinate by 1.0 of the continuum level for better clarity. Profiles from HJD 2457128.5976 and 2457137.5762 have anomalously positive RVs of the emission wings, which stem from the episode of a large strengthening of the emission. The next profile, from HJD 2457154.5435, has a normal orbital RV.

We tentatively adopted the mass of the Be primary after Zorec et al. (2016) and estimated the basic properties of the system for several possible orbital inclinations (see Table 7). Because no lines of the secondary were detected in the optical spectra and because no companions to Be stars were ever found among main-sequence objects (Bodensteiner et al. 2020), we conclude that the secondary is not a Roche-lobe filling object, but most probably a hot subdwarf star or white dwarf. It should be looked for in the far-UV spectral region. For the Gaia DR2 parallax of 00007059, the projected angular separation of the binary components is 00048, which might be resolved with present-day optical interferometers such as the currently tested interferometer SPICA (Pannetier et al. 2020).

Table 7.

Possible properties of the binary system.

5. Correlations between orbital and long-term changes

Inspection of the light, colour, and emission-line strength variations seems to indicate that the rapid episodes of large changes such as those near JD 2452900 or JD 2457300 occurred during one binary orbital period. To investigate the problem, we plot phase diagrams of the variability of the V magnitude for several more recent shorter time intervals in Fig. 9. The two large light decreases that are accompanied by strong increases in the Hα emission-line strength apparently occurred around phases of elongations, with the Be primary receding from us. At the same time, the plot shows that in another observing season, brightenings were observed around similar orbital phases. The same is also confirmed by a phase plot of the Hα emission-line strength; see Fig. 10.

thumbnail Fig. 9.

Phase plots of V magnitude for several seasons of Hvar observations for binary ephemeris (Eq. (1)). From top to bottom: data from JD 2457568 − 57657, JD 2457931 − 58079, JD 2458318 − 58392, and JD 2458664 − 58740.

thumbnail Fig. 10.

Phase plots of the strength of the Hα emission for binary ephemeris (Eq. (1)). Data from individual sources are denoted as follows: Circles in blue show DAO spectra, red circles show OND spectra, green circles represent BESO spectra, magenta circles show BeSS, and black circles show Castanet spectra. The black crosses show data from the literature.

We also investigated the time behaviour of the V/R ratio of the double Hα emission. In this case as well, V1294 Aql appears to be quite unusual. As Figs. 2 and 3 show, the V/R variations are different in different time intervals and do not recall either the long-term cyclic changes known for Be stars with one-armed global oscillations or phase-locked changes. In several panels of Fig. 11 we show enlarged plots of the V/R changes. Instants of expected phase-locked V/R maxima predicted by the orbital ephemeris (1) are shown by vertical lines.

thumbnail Fig. 11.

Enlarged subsets of time variability of the HαV/R ratio with the instants of the expected phase-locked maxima predicted by the orbital ephemeris (eq. (1)).

6. Rapid changes

Although we collected a large number of photometric observations of V1294 Aql, their time distribution is not suitable for a search for rapid periodic changes. Perhaps the only observations suitable for such a search are the early V observations by Lynds (1959). He himself concluded that his observations definitively indicate variations in brightness, which appear to be somewhat erratic, however, and no period could be found. The observations were secured within one month during a time interval that was not affected by secular variations. Our period analysis revealed sinusoidal variations with a semi-amplitude of 00139(29). A least-squares fit led to ephemeris (Eq. (2)), the rms of one observation being 00067. The corresponding phase plot is shown in Fig. 12.

(2)

thumbnail Fig. 12.

Possibly periodic rapid light changes based on Lynds (1959)V magnitude photometry and plotted for ephemeris (Eq. (2)).

This indicates that a scatter of at least 003 is to be expected in individual observations on longer timescales.

Lefèvre et al. (2009) carried out an automatic period search in the Hipparcos Hp photometry to find new periodic variables among OB stars. They identified V1294 Aql as a possible slowly pulsating B star (SPB) with a period of 7752. We cannot confirm their result. They apparently did not take the secular light change in Hp photometry into account; see the upper panel of Fig. 2 here.

Zorec et al. (2016) estimated the following physical properties of the Be component:

Teff = (30120 ± 2540) K, log g = (4.08 ± 0.40) [cgs], mass M = (16.9 ± 2.7) M⊙, v sin i = (207 ± 18) km s−1, critical rotational velocity v = (517 ± 64) km s−1, and the inclination of the rotational axis i = (37° ±9°). For these values, the period 0648 appears as a reasonable rotation period of the Be component. In passing we note that at the time of writing, the star has not been observed by the TESS satellite.

7. Fourth timescale

Harmanec (1998a) has called attention to the fact that the brightness of the Be star ω CMa outside of the episodes of brightenings accompanied by the growth of emission-line strength (typical of the positive correlation discussed above) has been decreasing secularly. His observation was later confirmed with more recent photometry (Ghoreyshi et al. 2018, 2021). These authors and also Marr et al. (2021), who studied another Be star, V2048 Oph = 66 Oph, modelled the secular variability and the episodes of brightening and increases in the Balmer emission-line strength with some success, estimating the required viscosity values for individual episodes, and also discussing some limitations of their effort. The yellow light curve of V2048 Oph is shown in the upper panel of Fig. 1 of Marr et al. (2021). It shows a secular light decrease between 1980 and 2000, occasionally interrupted by brightenings reminiscent of a positive correlation. However, the strength of the Hα emission is near its maximum over the same time interval of about 20 years, and only then does it gradually decrease. No emission has been observed since about 2010. However, as Marr et al. (2021) pointed out, the outer parts of the disk are still seen in the radio wavelengths.

We collected and homogenised the V photometry of V2048 Oph from the archive of UBV photometry provided by J.R. Percy, from Hvar, SPM, Johnson et al. (1966), Haupt & Schroll (1974) and Kozok (1985) and transformed the HIPPARCOSHp (Perryman 1997) photometry into Johnson V and the observations of Hill et al. (1976) secured in the DAO photometric system into UBV using the transformation formulæ provided by Harmanec & Božić (2001). The V light curve of V2048 Oph based on the above-mentioned data sets is shown in the upper panel of Fig. 13.

thumbnail Fig. 13.

Secular photometric changes of several well-observed Be stars.

A similar secular light decrease has also been reported for V744 Her = 88 Her, a Be star with an inverse type of correlation (Harmanec & Božić 2013). We show its light curve complemented by more recent observations, adapted from Božić et al. (in prep.) in Fig. 13 as well. In the same figure, we also show the V and B light curves of the Be star EW Lac = HD 217050 from another study in preparation. In this case, a secular increase in brightness is observed. We note that the large scatter around the mean trend is related to the known rapid light variability of EW Lac on a timescale shorter than one day. The fourth panel of Fig. 13 shows the plot of Hvar V photometry of φ And, another Be star with a positive type of correlation. A mild light decrease over several decades of observations is visible.

Searching the literature, we found a few more examples. A secular light increase has also been observed for the Be star with a positive correlation γ Cas (Harmanec 2002, Fig. 5) over nearly 30000 days.

Finally, as we have shown here, the secular light decrease of V1294 Aql has recently changed to a secular light increase. All this shows the large variety of different possible evolutions of the circumstellar disk, its replenishment, and a gradual dissipation.

8. Discussion

In spite of the effort of several generations of stellar astronomers, the engine leading to the occasional formation of circumstellar disks around Be stars has not been firmly identified so far. One possible explanation is based on the idea that Be stars are rapidly rotating non-radial pulsators (NRP) and that the additional force needed to facilitate the outflow of gas and angular momentum transfer from the stellar equator arises from a constructive interference of two or more NRP modes (Rivinius et al. 1998a,b; Baade et al. 2017a,b; Baade & Rivinius 2020; Borre et al. 2020; Labadie-Bartz et al. 2021). Especially the systematic photometries from space observatories were used and analysed to support this conjecture. Confirmation of this scenario would require the creation of new, self-consistent models, however, which would show that Be stars are indeed pulsationally unstable over the whole area that they occupy in the Hertzsprung-Russell (HR) diagram. It should also be mentioned that Baade et al. (2017b) warned that evidence for constructive interference of pulsational modes for a larger number of Be stars is lacking and pointed out additional problems such as the rotational splitting of modes and the presence of rapid changes in the circumstellar envelopes during active phases. An alternative view was suggested by Harmanec (1998a), who argued that the dominant period of rapid changes undergoes small cyclic changes. Modelling such a situation, he found that a standard period analysis of a corresponding series of observations returns a multiperiodicity with several close periods. Yet another possibility was suggested by Harmanec et al. (2002b), who argued that the presence of a secondary can facilitate outflow from the equatorial parts of the gaseous disk of the Be primary even in systems that are not filling their Roche lobes. This idea is problematic, however, in that the effect is rather small in the majority of cases.

For a long time, various other suggestions have been made that the Be phenomenon and observed variability patterns of Be stars might be causally related to their binary nature (e.g. Plavec & Horn 1969; Kříž & Harmanec 1975; Harmanec & Kříž 1976; Harmanec 1987; Pols et al. 1991; Panoglou et al. 2018; Bodensteiner et al. 2018, 2020; Langer et al. 2020; Bodensteiner et al. 2021, among others). The approach of Klement et al. (2019) is worth mentioning. These authors studied the spectral energy distribution of several Be stars and provided arguments that their disks had to be truncated by the Roche lobes. This constitutes an indirect argument for the presence of companions to these objects.

The phase-locked V/R changes observed for several Be binaries represent another interesting phenomenon. As already noted above, they are usually observed roughly in phase with the orbital RV changes of the Be stars in question (Harmanec et al. 1976, 2002a; Doazan et al. 1982b; Juza et al. 1991; Štefl et al. 2007). A phase-locked emission-line variation with a single maximum and minimum per orbital period was also found by Borre et al. (2020) for γ Cas. These authors used an interesting detection technique. They analysed local pixel-per-pixel line fluxes across the Hα profile in a series of higher-resolution BeSS spectra. Panoglou et al. (2018) modelled the phase-locked V/R changes as global oscillations in the circumstellar disks with two spiral patterns and concluded that the phase-locked V/R changes should exhibit two maxima and minima during one orbital period. This clearly disagrees with the available observations mentioned above. The only case for which a double-wave V/R curve was detected is V696 Mon = HR 2142 (Peters 1972, 1976). There is a natural explanation of roughly sinusoidal phase-locked V/R changes in our view, as discussed in Appendix C of Wolf et al. (2021). The circumstellar disk probably occupies almost the whole volume of the Roche lobe near the orbital plane. This causes there to be more gaseous material in the part of the disk facing the secondary than on the opposite side. Because the disk rotates, more emission power is available on the side facing the secondary, and this naturally leads to phase-locked V/R changes that are in phase with the RV changes of the Be component.

To show how confusing the interpretation of V/R changes can be, we compare the V/R changes of Hα and He I 6678 Å for two shorter time intervals in Fig. 14. In the first interval, the variations for both lines are in phase, while in the second interval, antiphase variation is observed. We note that this is a consequence of the fact that the He shell RV becomes very negative (a temporarily elongated envelope?) and apparently weakens the V peak of a relatively faint He emission. This is best illustrated in Fig. 15, where the Hα and He I 6678 profiles for two dates are shown.

thumbnail Fig. 14.

Apparent V/R changes observed for the Hα and He I 6678 emission lines intercompared for two time segments. The variation in the shell He RV is also shown. An apparently anti-phase behaviour is observed in the second time interval, when the shell RV becomes quite negative and the He shell line blends with the V peak of the faint He emission. The same colours as in the previous time plots are used to distinguish spectra from individual observatories.

thumbnail Fig. 15.

Apparent V/R changes observed for the Hα and He I 6678 emission lines intercompared for two dates. An apparently anti-phase behaviour is observed for the later date, when the He I 6678 shell RV becomes quite negative and the He shell line blends with the V peak of the faint He emission.

This study of V1294 Aql demonstrates very clearly how hard it is to identify and understand mutually coexisting and overlapping variability patterns governing the observed spectral, light, and colour changes. Attempts at modelling them quantitatively, planned for a continuation of this study, are expected to shed more light on the mysterious Be phenomenon. We also suggest that further monitoring of the object with systematic photometry, high-resolution spectroscopy, and especially with the optical interferometry could help to reveal the secrets of this intriguing Be binary, or possibly a multiple system, as indicated by the analysis of the astrometric data (Brandt 2021).

1

For normal stars, the Balmer jump parameter D, which is defined as log of the flux longward 3700 Å divided by the flux shortward 3700 Å, is a good measure of the stellar effective temperature. The presence of circumstellar matter can cause the second Balmer jump, which is occasionally seen in emission.

2

For the description of his instrumentation and data reduction, see http://www.astrosurf.com/buil/us/bestar.htm.

4

The whole program suite with a detailed manual, examples of data, auxiliary data files, and results is available at http://astro.troja.mff.cuni.cz/ftp/hec/PHOT.

Acknowledgments

We gratefully acknowledge the use of the latest publicly available version of the program FOTEL written by P. Hadrava. We thank A. Aret, A. Budovičová, P. Chadima, M. Dovčiak, J. Fuchs, P. Hadrava, J. Juryšek, E. Kiran, L. Kotková, R. Křiček, J. Libich, J. Nemravová, P. Rutsch, S. Saad, P. Škoda, S. Štefl, and V. Votruba, who obtained some of the Ondřejov spectra used in this study. J.R. Percy kindly puts the archive of his systematic UBV observations of bright Be stars in three observatories at our disposal. We also acknowledge the constructive suggestions of an anonymous referee to the first version of this study. Over the years, this long-term project was supported by the grants 205/06/0304, 205/08/H005, P209/10/0715, and GA15-02112S of the Czech Science Foundation, by the grants 678212 and 250015 of the Grant Agency of the Charles University in Prague, from the research project AV0Z10030501 of the Academy of Sciences of the Czech Republic, and from the Research Program MSM0021620860 Physical study of objects and processes in the solar system and in astrophysics of the Ministry of Education of the Czech Republic. The research of PK was supported by the ESA PECS grant 98058. HB, DR, and DS acknowledge financial support from the Croatian Science Foundation under the project 6212 “Solar and Stellar Variability”. This work has made use of data from the European Space Agency (ESA) mission Gaia (https://www.cosmos.esa.int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC; https://www.cosmos.esa.int/web/gaia/dpac/consortium). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement. We also used some spectra of the BeSS database, operated at LESIA, Observatoire de Meudon, France: http://basebe.obspm.fr. Finally, we acknowledge the use of the electronic database from the CDS, Strasbourg, and the electronic bibliography maintained by the NASA/ADS system.

References

  1. Alvarez, M., & Schuster, W. J. 1982, Rev. Mex. Astron. Astrofis., 5, 173 [Google Scholar]
  2. Baade, D., & Rivinius, T. 2020, in Stars and their Variability Observed fromSpace, eds. C. Neiner, W. W. Weiss, D. Baade, et al., 35 [Google Scholar]
  3. Baade, D., Rivinius, T., Pigulski, A., Carciofi, A., & BEST Collaboration 2017a, in The B[e] Phenomenon: Forty Years of Studies, eds. A. Miroshnichenko, S. Zharikov, D. Korčáková, & M. Wolf, ASP Conf. Ser., 508, 93 [Google Scholar]
  4. Baade, D., Rivinius, T., Pigulski, A., & Carciofi, A., & BRITE Executive Science Team 2017b, in The Lives and Death-Throes of Massive Stars, eds. J. J. Eldridge, J. C. Bray, & L. Xiao, 329, 384 [NASA ADS] [Google Scholar]
  5. Ballereau, D., & Chauville, J. 1987, A&AS, 70, 229 [NASA ADS] [Google Scholar]
  6. Ballereau, D., & Chauville, J. 1989, A&A, 214, 285 [NASA ADS] [Google Scholar]
  7. Ballereau, D., & Hubert-Delplace, A. M. 1982, in Be Stars, ed. M. Jaschek, & H.-G. Groth, IAU Symp., 98, 171 [NASA ADS] [CrossRef] [Google Scholar]
  8. Bodensteiner, J., Baade, D., Greiner, J., & Langer, N. 2018, A&A, 618, A110 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  9. Bodensteiner, J., Shenar, T., & Sana, H. 2020, A&A, 641, A42 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  10. Bodensteiner, J., Shenar, T., Mahy, L., et al. 2021, in MOBSTER-1 Virtual Conference: Stellar Variability as a Probe of Magnetic Fields in Massive Stars, 24 [Google Scholar]
  11. Borre, C. C., Baade, D., Pigulski, A., et al. 2020, A&A, 635, A140 [EDP Sciences] [Google Scholar]
  12. Božić, H., Harmanec, P., Horn, J., et al. 1995, A&A, 304, 235 [Google Scholar]
  13. Božić, H., Harmanec, P., & Koubský, P. 2013, Cent. Eur. Astrophys. Bull., 37,\pg{\break} 9 [NASA ADS] [Google Scholar]
  14. Brandt, T. D. 2021, ApJS, 254, 42 [NASA ADS] [CrossRef] [Google Scholar]
  15. Chini, R., Hoffmeister, V. H., Nasseri, A., Stahl, O., & Zinnecker, H. 2012, MNRAS, 424, 1925 [Google Scholar]
  16. Cousins, A. W. J. 1973, Mon. Notes Astron. Soc. South Afr., 32, 11 [Google Scholar]
  17. Cousins, A. W. J., & Dean, J. F. 1978, Mon. Notes Astron. Soc. South Afr., 37, 114 [Google Scholar]
  18. Cousins, A. W. J., & Stoy, R. H. 1963, Photoelectric Magnitudes and Colours of Southern Stars [Google Scholar]
  19. Dahn, C. C., & Guetter, H. H. 1973, ApJ, 179, 551 [NASA ADS] [CrossRef] [Google Scholar]
  20. Deeming, T. J. 1975, Ap&SS, 36, 137 [Google Scholar]
  21. Doazan, V., Harmanec, P., Koubský, P., Krpata, J., & \v{Z}d’{\’{a}}rsk\’{y}, F. 1982a, A&AS, 50, 481 [NASA ADS] [Google Scholar]
  22. Doazan, V., Harmanec, P., Koubský, P., Krpata, J., & \v{Z}d’\’{a}rsk\’{y}, F. 1982b, A&A, 115, 138 [NASA ADS] [Google Scholar]
  23. Fernie, J. D. 1983, ApJS, 52, 7 [NASA ADS] [CrossRef] [Google Scholar]
  24. Fuhrmann, K., Chini, R., Hoffmeister, V. H., et al. 2011, MNRAS, 411, 2311 [NASA ADS] [CrossRef] [Google Scholar]
  25. Ghoreyshi, M. R., Carciofi, A. C., Rímulo, L. R., et al. 2018, MNRAS, 479,\pg{\break} 2214 [Google Scholar]
  26. Ghoreyshi, M. R., Carciofi, A. C., Jones, C. E., et al. 2021, ApJ, 909, 149 [NASA ADS] [CrossRef] [Google Scholar]
  27. Gies, D. R., Willis, C. Y., Penny, L. R., & McDavid, D. 1993, PASP, 105, 281 [NASA ADS] [CrossRef] [Google Scholar]
  28. Gies, D. R., Bagnuolo, W. G., Jr, Ferrara, E. C., et al. 1998, ApJ, 493, 440 [NASA ADS] [CrossRef] [Google Scholar]
  29. Golay, M. 1974, in Introduction to Astronomical Photometry, (Dordrecht: Reidel), Astrophys. Space Sci. Libr., 41 [CrossRef] [Google Scholar]
  30. Hadrava, P. 1990, Contr. Astron. Obs. Skalnaté Pleso, 20, 23 [Google Scholar]
  31. Hadrava, P. 2004, Publ. Astron. Inst. Acad. Sci. Czech Rep., 92, 1 [Google Scholar]
  32. Harmanec, P. 1983, Hvar Obs. Bull., 7, 55 [NASA ADS] [Google Scholar]
  33. Harmanec, P. 1987, in IAU Colloq. 92: Physics of Be Stars, eds. A. Slettebak, & T. P. Snow, 339 [Google Scholar]
  34. Harmanec, P. 1998a, A&A, 334, 558 [NASA ADS] [Google Scholar]
  35. Harmanec, P. 1998b, A&A, 335, 173 [NASA ADS] [Google Scholar]
  36. Harmanec, P. 2000, in IAU Colloq. 175: The Be Phenomenon in Early-Type Stars, eds. M. A. Smith, H. F. Henrichs, & J. Fabregat, ASP Conf. Ser., 214, 13 [NASA ADS] [Google Scholar]
  37. Harmanec, P. 2002, ASP Conf. Ser., 279, 221 [NASA ADS] [Google Scholar]
  38. Harmanec, P., & Božić, H. 2001, A&A, 369, 1140 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  39. Harmanec, P., & Božić, H. 2013, Cent. Eur. Astrophys. Bull., 37, 3 [Google Scholar]
  40. Harmanec, P., & Horn, J. 1998, J. Astron. Data, 4, 5 [Google Scholar]
  41. Harmanec, P., & K{\v{r}}{\’{\i}}\v{z}, S. 1976, in IAU Symp. 70: Be and Shell Stars, ed. A. Slettebak, 385 [CrossRef] [Google Scholar]
  42. Harmanec, P., Koubský, P., Krpata, J., & Ždárský, F. 1976, Bull. Astron. Inst. Czechoslovakia, 27, 47 [NASA ADS] [Google Scholar]
  43. Harmanec, P., Horn, J., & Juza, K. 1994, A&AS, 104, 121 [NASA ADS] [Google Scholar]
  44. Harmanec, P., Habuda, P., Štefl, S., et al. 2000, A&A, 364, L85 [NASA ADS] [Google Scholar]
  45. Harmanec, P., Božić, H., Percy, J. R., et al. 2002a, A&A, 387, 580 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  46. Harmanec, P., Bisikalo, D. V., Boyarchuk, A. A., & Kuznetsov, O. A. 2002b, A&A, 396, 937 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  47. Hastings, B., Langer, N., Wang, C., Schootemeijer, A., & Milone, A. P. 2021, A&A, 653, A144 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  48. Haupt, H. F., & Schroll, A. 1974, A&AS, 15, 311 [NASA ADS] [Google Scholar]
  49. Heard, J. F., Hurkens, R., Harmanec, P., Koubský, P., & Krpata, J. 1975, A&A, 42, 47 [NASA ADS] [Google Scholar]
  50. Hill, G., Hilditch, R. W., & Pfannenschmidt, E. L. 1976, Publ. Dominion Astrophys. Obs. Victoria, 15, 1 [NASA ADS] [Google Scholar]
  51. Horn, J., Harmanec, P., Koubský, P., et al. 1982, Bull. Astron. Inst. Czechoslovakia, 33, 308 [NASA ADS] [Google Scholar]
  52. Horn, J., Hudec, R., & Koubský, P. 1983, Inf. Bull. Variable Stars, 2304, 1 [NASA ADS] [Google Scholar]
  53. Horn, J., Kubát, J., Harmanec, P., et al. 1996, A&A, 309, 521 [NASA ADS] [Google Scholar]
  54. Johnson, H. L. 1955, Ann. Astrophys., 18, 292 [NASA ADS] [Google Scholar]
  55. Johnson, H. L., & Harris, D. L. 1954, ApJ, 120, 196 [NASA ADS] [CrossRef] [Google Scholar]
  56. Johnson, H. L., Iriarte, B., Mitchell, R. I., & Wisniewski, W. Z. 1966, Commun. Lunar Planet. Lab., 4, 99 [NASA ADS] [Google Scholar]
  57. Juza, K., Harmanec, P., Hill, G. M., Tarasov, A. E., & Matthews, J. M. 1991, Bull. Astron. Inst. Czechoslovakia, 42, 39 [NASA ADS] [Google Scholar]
  58. Kaufer, A., & Pasquini, L. 1998, in Optical Astronomical Instrumentation, ed. S. D’Odorico, SPIE COnf. Ser., 3355, 844 [NASA ADS] [CrossRef] [Google Scholar]
  59. Kaufer, A., Wolf, B., Andersen, J., & Pasquini, L. 1997, The Messenger, 89, 1 [NASA ADS] [Google Scholar]
  60. Klement, R., Carciofi, A. C., Rivinius, T., et al. 2019, ApJ, 885, 147 [Google Scholar]
  61. Klement, R., Schaefer, G. H., Gies, D. R., et al. 2022, ApJ, 926, 213 [NASA ADS] [CrossRef] [Google Scholar]
  62. Koubský, P., Harmanec, P., Kubát, J., et al. 1997, A&A, 328, 551 [NASA ADS] [Google Scholar]
  63. Kozok, J. R. 1985, A&AS, 61, 387 [NASA ADS] [Google Scholar]
  64. Krpata, J. 2008, http://astro.troja.mff.cuni.cz/ftp/hec/SPEFO [Google Scholar]
  65. Kuschnig, R., Paunzen, E., & Weiss, W. W. 1994, Inf. Bull. Variable Stars, 4070, 1 [NASA ADS] [Google Scholar]
  66. Kříž, S. 1969, in Mass Loss from Stars, ed. M. Hack, Astrophys. Space Sci. Library, 13, 257 [CrossRef] [Google Scholar]
  67. Kříž, S., & Harmanec, P. 1975, Bull. Astron. Inst. Czechoslovakia, 26, 65 [Google Scholar]
  68. Labadie-Bartz, J., Baade, D., Carciofi, A. C., et al. 2021, MNRAS, 502, 242 [NASA ADS] [CrossRef] [Google Scholar]
  69. Langer, N., Baade, D., Bodensteiner, J., et al. 2020, A&A, 633, A40 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  70. Lefèvre, L., Marchenko, S. V., Moffat, A. F. J., & Acker, A. 2009, A&A, 507, 1141 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  71. Linnell, A. P., Harmanec, P., Koubský, P., et al. 2006, A&A, 455, 1037 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  72. Lynds, C. R. 1959, ApJ, 130, 577 [Google Scholar]
  73. Lyubimkov, L. S., Rachkovskaya, T. M., Rostopchin, S. I., & Tarasov, A. E. 1998, Astron. Rep., 42, 312 [NASA ADS] [Google Scholar]
  74. Manfroid, J., Sterken, C., Bruch, A., et al. 1991, ESO Sci. Rep., 8, 1 [NASA ADS] [Google Scholar]
  75. Manfroid, J., Sterken, C., Cunow, B., et al. 1994, Eur. South. Obs. Sci. Rep., 14, 1 [Google Scholar]
  76. Marr, K. C., Jones, C. E., Carciofi, A. C., et al. 2021, ApJ, 912, 76 [NASA ADS] [CrossRef] [Google Scholar]
  77. McLaughlin, D. B. 1961a, JRASC, 55, 73 [NASA ADS] [Google Scholar]
  78. McLaughlin, D. B. 1961b, JRASC, 55, 13 [NASA ADS] [Google Scholar]
  79. Mennickent, R. E., Sterken, C., & Vogt, N. 1997, A&A, 326, 1167 [NASA ADS] [Google Scholar]
  80. Mermilliod, J.-C., Mermilliod, M., & Hauck, B. 1997, A&AS, 124, 349 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  81. Moreno, H. 1971, A&A, 12, 442 [NASA ADS] [Google Scholar]
  82. Moujtahid, A., Zorec, J., Hubert, A. M., Garcia, A., & Burki, G. 1998, A&AS, 129, 289 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  83. Mourard, D., Monnier, J. D., Meilland, A., et al. 2015, A&A, 577, A51 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  84. Neiner, C., de Batz, B., Cochard, F., et al. 2011, AJ, 142, 149 [Google Scholar]
  85. Pannetier, C., Mourard, D., Berio, P., et al. 2020, SPIE Conf. Ser., 11446, 114460T [NASA ADS] [Google Scholar]
  86. Panoglou, D., Faes, D. M., Carciofi, A. C., et al. 2018, MNRAS, 473, 3039 [NASA ADS] [CrossRef] [Google Scholar]
  87. Perryman, M. A. C. ESA 1997, in The Hipparcos and Tycho Catalogues, Astrometric and Photometric Star Catalogues Derived from the ESA Hipparcos Space Astrometry Mission, (Noordwijk, Netherlands: ESA Publications Division), ESA SP Ser., 1200 [Google Scholar]
  88. Peters, G. J. 1972, PASP, 84, 334 [NASA ADS] [CrossRef] [Google Scholar]
  89. Peters, G. J. 1976, in Be and Shell Stars, A. Slettebak, 70, 417 [NASA ADS] [CrossRef] [Google Scholar]
  90. Plavec, M. 1970, PASP, 82, 957 [NASA ADS] [CrossRef] [Google Scholar]
  91. Plavec, M. 1976, in Be and Shell Stars, A. Slettebak, 70, 439 [NASA ADS] [CrossRef] [Google Scholar]
  92. Plavec, M., & Horn, J. 1969, in Mass Loss from Stars, ed. M. Hack, Astrophys. Space Sci. Libr., 13, 242\pg{\newpage} [NASA ADS] [CrossRef] [Google Scholar]
  93. Poeckert, R. 1981, PASP, 93, 297 [Google Scholar]
  94. Pojmanski, G. 2002, Acta Astron., 52, 397 [NASA ADS] [Google Scholar]
  95. Pols, O. R., Cote, J., Waters, L. B. F. M., & Heise, J. 1991, A&A, 241, 419 [NASA ADS] [Google Scholar]
  96. Prša, A., Harmanec, P., Torres, G., et al. 2016, AJ, 152, 41 [Google Scholar]
  97. Rivinius, T., Baade, D., Štefl, S., et al. 1998a, Be Star Newsletter, 33, 15 [NASA ADS] [Google Scholar]
  98. Rivinius, T., Baade, D., Štefl, S., et al. 1998b, in Cyclical Variability in Stellar Winds, eds. L. Kaper,&A. W. Fullerton, 207 [CrossRef] [Google Scholar]
  99. Ruždjak, D., Božić, H., Harmanec, P., et al. 2009, A&A, 506, 1319 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  100. Schuster, W. J., & Guichard, J. 1984, Rev. Mex. Astron. Astrofis., 9, 141 [Google Scholar]
  101. Sigut, T. A. A., & Patel, P. 2013, ApJ, 765, 41 [Google Scholar]
  102. Škoda, P. 1996, in Astronomical Data Analysis Software and Systems V, ASP Conf. Ser., 101, 187 [Google Scholar]
  103. Stellingwerf, R. F. 1978, ApJ, 224, 953 [Google Scholar]
  104. Sterken, C., Manfroid, J., Anton, K., et al. 1993, Eur. Southern Obs. Sci. Rep., 12, 1 [NASA ADS] [Google Scholar]
  105. Sterken, C., Manfroid, J., Beele, D., et al. 1995, Eur. Southern Obs. Sci. Rep., 16, 1 [NASA ADS] [Google Scholar]
  106. Tempesti, P., & Patriarca, R. 1976, Inf. Bull. Variable Stars, 1164, 1 [Google Scholar]
  107. Tokovinin, A., Fischer, D. A., Bonati, M., et al. 2013, PASP, 125, 1336 [NASA ADS] [CrossRef] [Google Scholar]
  108. Štefl, S., Okazaki, A. T., Rivinius, T., & Baade, D. 2007, in Active OB-Stars: Laboratories for Stellar and Circumstellar Physics, eds. A. T. Okazaki, S. P. Owocki, & S. Štefl, ASP Conf. Ser., 361, 274 [Google Scholar]
  109. van der Wal, P. B., Nagel, C., Voordes, H. R., & de Boer, K. S. 1972, A&AS, 6, 131 [NASA ADS] [Google Scholar]
  110. Wang, L., Gies, D. R., & Peters, G. J. 2018, ApJ, 853, 156 [Google Scholar]
  111. Wang, L., Gies, D. R., Peters, G. J., et al. 2021, AJ, 161, 248 [Google Scholar]
  112. Wolf, M., Harmanec, P., Božić, H., et al. 2021, A&A, 647, A97 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  113. Zorec, J., Frémat, Y., Domiciano de Souza, A., et al. 2016, A&A, 595, A132 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]

Appendix A: Details of photometric observations and their reduction and homogenisation

Differential observations from Hvar, Ondřejov, San Pedro Martír, Tubitak, and Çanakkale were all reduced and carefully transformed into the standard Johnson UBV system with the help of the reduction program HEC22; see Harmanec et al. (1994) and Harmanec & Horn (1998) for the observational strategy and details of the data reduction. 4 The latest HEC22 rel.18 version was used. It allows the use of non-linear transformation formulæ, includes the colour extinction coefficient among the seasonal transformation coefficients, and allows the time variation of linear extinction coefficients to be modelled in the course of the observing nights. Our standard strategy is to observe the check star as frequently as the variable to determine the real data quality. This also permitted replacing the originally selected comparison by the check star when the former was found to be microvariable after some time. This was indeed the case of observations of V1294 Aql. Our original comparison 35 Aql = HD 183324 = HR 7400 was reported to be a λ Boo microvariable with a period of 00211 and a full amplitude of 002 by Kuschnig et al. (1994) and received the variable star name V1431 Aql. We now use our original check star HR 7397 = HD 183227 as the comparison, but we note that because our observing sequences usually consisted of three to five cycles of individual observations, they are typically some 005 – 006 long and the variability of 35 Aql is in a sense smeared out. We therefore retained some series of observations relative to 35 Aql, where the change to the new comparison HR 7397 would decrease the quality of differential observations.

Below, we provide some details of the individual data sets defined in Table 2 and their reductions.

  • Station 01 – Hvar: These differential observations have been secured by a number of observers over many years, with HR 7397 and HR 7438 as the comparison and check star, respectively, within the framework of the long-term program of monitoring the light and colour variations of bright Be stars. The all-sky magnitudes from Hvar were carefully homogenised and serve as our primary source of UBV and UBVR magnitudes. The Hvar mean values, which were always added to magnitude differences from all stations, are listed in Table A.1. Since the summer of 2013, the R filter was installed to the Hvar photometer, and UBVR observations were collected. We note that the transparency curve of the R filter closely corresponds to that of the standard Cousins R filter. However, because we were not able to find enough northern bright standard stars with the Cousins V​ − ​R values, we derived robust mean values of Johnson V​ − ​R indices from Johnson et al. (1966) and reduced our observations to the Johnson system.

  • Station 02 – Brno: These BVR observations were secured by P. Svoboda in his private observatory with a Sonnar 0.034 m refractor and a CCD camera. They were reduced to the form of magnitude differences by the author. The Hvar all-sky values of the comparison star were then added to them.

  • Station 04 – Ondřejov:UBV observations secured with the 0.65 m Ondřejov reflector and a photometer with an EMI tube and transformed into the standard Johnson system.

  • Station 12 – La Silla ESO: There are two independent observation sets. The early observations are all-sky uvby observations published by Kozok (1985). The rest are differential uvby observations obtained with the Danish 0.50 m reflector (later called Strömgren Automatic Telescope (SAT) (see Manfroid et al. 1991; Sterken et al. 1993; Manfroid et al. 1994; Sterken et al. 1995, for the published data and details of the observations and reduction) relative to 35 Aql, HD 7397 being used as the check star. We then used the transformation from the Strömgren into the Johnson system derived by Harmanec & Božić (2001) to obtain values that can be directly compared to other UBV datasets.

  • Station 20 – Toronto Differential BV observations secured during the international campaign on bright Be stars by John Percy, who kindly placed them at our disposal. They were transformed into the standard Johnson system via linear transformations by the author, and the Hvar all-sky values of the comparison star were added to the magnitude differences.

  • Station 26 – Haute Provence Two UBV observations were secured by Haupt & Schroll (1974). Individual observations were kindly provided by H. Haupt upon our request and corrected to our adopted values for the comparison star.

  • Station 30: SPM 1.50 m and 0.84 m reflectors The early all-sky observations (before JD 2446000) were secured at the 1.50 m reflector in the 13C system and reduced by their authors (Alvarez & Schuster 1982; Schuster & Guichard 1984). They were then transformed into the standard Johnson UBV system with the help of transformation formulæ derived by Harmanec & Božić (2001). More recent differential UBV observations were secured by MW with the 0.84 m reflector and Cuenta-pulsos photon-counting photometer. Variable extinction during the nights was monitored, and the data were reduced to the standard UBV system.

  • Station 37 – Jungfraujoch These all-sky seven-colour (7-C) observations were secured in the Geneva photometric system at the Jungfraujoch Sphinx mountain station 0.40 m reflector and kindly placed at our disposal by G. Burki. They were transformed into the standard Johnson UBV system with the help of transformations derived by Harmanec & Božić (2001).

  • Station 44: Mt Palomar 0.51 m reflectorLynds (1959) carried out numerous differential observations of early-type stars classified as giants, including V1294 Aql. He observed in yellow light, with a Corning 3384 filter and an EMI 6094 tube. Most of his observations are only published in the form of time plots of the data, variable minus comparison. In Table A.2 we compare the mean magnitude differences between the respective stars for all his microvariables with the mean differences in the standard Johnson V filter. It is obvious that for all practical purposes, the yellow filter used by Lynds measures magnitudes that are very close to the Johnson V filter. The largest △(dV−dy) difference was found for V373 Cas, which is now known to be a Be star and a spectroscopic binary (Lyubimkov et al. 1998). The Hipparcos Hp magnitude has a 01 range of variations (Perryman 1997). We therefore digitised the yellow observations of V1294 Aql from the enlarged Fig. 13 of Lynds (1959) and derived HJDs and magnitude differences that were then added to the V magnitude of the comparison star HR 7438 from Table A.1. We estimate that the HJDs are accurate to ±00003, which is very satisfactory.

  • Station 61 – Hipparcos: These all-sky observations were reduced to the standard V magnitude via the transformation formulæ derived by Harmanec (1998b). To derive this transformation, the correct values of the Johnson B−V and U−B indices are required. We checked the overlapping Hvar UBV photometry to find that the B−V index remained basically constant over the time interval covered by the Hp photometry at a value of 010. The U−B index varied from −077 to −064, but we verified that these extremes cause an error of only about 001 in the transformation into Johnson V. We therefore used B−V=−010 and U−B=−070 to transform Hp into V.

  • Station 66 – Tubitak: These differential UBV observations were secured with the 0.40 m reflector of the Turkish national observatory and an SSP-5A solid-state photometer and were transformed into the standard system.

  • Station 89 – Çanakkale: These differential UBV observations were secured at Çanakkale mountain station with a 0.40 m reflector and an SSP-5 solid-state photometer and were transformed into the standard system.

  • Station 93 – ASAS3 V photometry: We extracted these all-sky observations from the ASAS3 public archive (Pojmanski 2002), using the data for diaphragm 1, which has the lowest rms errors on average. We omitted all observations of grade D and observations with rms errors larger than 004. We also omitted a strongly deviating observation at HJD 2452662.6863.

Table A.1.

Standard UBVR magnitudes of the stars used by different observers as comparison stars in their differential observations of V1294 Aql.

Table A.2.

Lynds (1959) versus standard Johnson magnitude differences for all suitable microvariables observed by Lynds. The mean UBV values of all considered stars were taken (with one exception) from the General Catalogue of Photometric Data (see Mermilliod et al. 1997). The yellow magnitude differences ‘variable minus comparison’ of Lynds are denoted by dy. The UBV magnitudes of the corresponding comparison are always listed below each program star.

All Tables

Table 1.

Journal of electronic spectra.

In the text
Table 2.

Journal of available photometry with known dates of observations.

In the text
Table 6.

Orbital solutions based on the Hα emission RVs.

In the text
Table 7.

Possible properties of the binary system.

In the text
Table A.1.

Standard UBVR magnitudes of the stars used by different observers as comparison stars in their differential observations of V1294 Aql.

In the text
Table A.2.

Lynds (1959) versus standard Johnson magnitude differences for all suitable microvariables observed by Lynds. The mean UBV values of all considered stars were taken (with one exception) from the General Catalogue of Photometric Data (see Mermilliod et al. 1997). The yellow magnitude differences ‘variable minus comparison’ of Lynds are denoted by dy. The UBV magnitudes of the corresponding comparison are always listed below each program star.

In the text

All Figures

thumbnail Fig. 1.

Time plot of the Otero visual brightness estimates (empty red circles) along with Johnson V photometry from Hvar (blue dots), SPM (green), Çanakkale (cyan), and TUG (magenta).
In the text
thumbnail Fig. 2.

Time plot of available observations over the whole time interval of about 25 000 d covered by the data. Top panel: yellow brightness observations, which could be transformed into Johnson V magnitude. Second and third panel: available B−V and U−B colour index observations. Bottom panel: V/R changes in the peaks of the double Hα emission. In the three panels with photometry, the differential observations are shown as blue dots, and all-sky observations are shown as black dots. Bottom panel: blue circles denote the DAO spectra, red circles show the OND spectra, green circles represent BESO spectra, magenta circles show BeSS spectra, black circles show Castanet Tolosan spectra, and the black crosses plot the data from the literature.
In the text
thumbnail Fig. 3.

Time plot showing the correlation between the secular brightness and colour variations in the V-band magnitudes and B−V and U−B colour indices (black shows all-sky and blue shows differential observations), and the V/R changes, EW, and strength of the Hα emission for the more recent electronic spectra. The colour symbols for spectra from different sources are the same as in Fig. 2.
In the text
thumbnail Fig. 4.

Time plot of the V/R changes recorded for the electronic spectra in the Si II 6347 Å line. The colour symbols for spectra from different sources are the same as in Fig. 2.
In the text
thumbnail Fig. 5.

Time plot of radial velocities. Top panel: RV of the He I shell lines. Bottom panel: RV measured on the steep wings of the Hα emission. Data from individual instruments are shown by different symbols: the circles are the same as in Fig. 2, and black crosses plot data from Ballereau & Chauville (1989). The ranges on the two axes in the two plots are different.
In the text
thumbnail Fig. 6.

U−B vs. B−V diagram for several distinct data subsets. Top panels: older data until JD 2450000 (left). All-sky observations are shown as black circles, and data from stations defined in Table 2 are denoted as follows: 01 (blue), 04 (green), 12 (red), and 26 (magenta). More recent data from the interval of secular light brightening, all from station 01 (blue) (right). Bottom panels: show UBV observations from the two sharp increases in the emission-line strength accompanied by light decreases. Time interval JD 2452741 – 2453309, which covers the first sharp light decrease (left, cf. Figs. 1 and 3.) Data from the time interval JD 2455357 – 2456094 corresponding to the second sharp increase in the emission-line strength (right). Data from stations 1, 30, 66, and 89 of Table 2 are shown by blue, red, green, and magenta dots, respectively. The main sequence and the supergiant sequence based on data from Golay (1974) (pp. 79–80) are shown, as is the reddening line.
In the text
thumbnail Fig. 7.

Radial-velocity curve corresponding to the orbital solution, based on RVs measured on the steep wings of the Hα emission, plotted for phases from ephemeris (top; eq. (1)). Orbital curve based on RV of the He I 6678 Å shell line prewhitened for the long-term changes, plotted for the same ephemeris (bottom). Data from individual instruments are shown by different symbols. The circles are the same as in Fig. 2, and black triangle shows CTIO.
In the text
thumbnail Fig. 8.

Three Ondřejov Hα profiles. They are mutually shifted in ordinate by 1.0 of the continuum level for better clarity. Profiles from HJD 2457128.5976 and 2457137.5762 have anomalously positive RVs of the emission wings, which stem from the episode of a large strengthening of the emission. The next profile, from HJD 2457154.5435, has a normal orbital RV.
In the text
thumbnail Fig. 9.

Phase plots of V magnitude for several seasons of Hvar observations for binary ephemeris (Eq. (1)). From top to bottom: data from JD 2457568 − 57657, JD 2457931 − 58079, JD 2458318 − 58392, and JD 2458664 − 58740.
In the text
thumbnail Fig. 10.

Phase plots of the strength of the Hα emission for binary ephemeris (Eq. (1)). Data from individual sources are denoted as follows: Circles in blue show DAO spectra, red circles show OND spectra, green circles represent BESO spectra, magenta circles show BeSS, and black circles show Castanet spectra. The black crosses show data from the literature.
In the text
thumbnail Fig. 11.

Enlarged subsets of time variability of the HαV/R ratio with the instants of the expected phase-locked maxima predicted by the orbital ephemeris (eq. (1)).
In the text
thumbnail Fig. 12.

Possibly periodic rapid light changes based on Lynds (1959)V magnitude photometry and plotted for ephemeris (Eq. (2)).
In the text
thumbnail Fig. 13.

Secular photometric changes of several well-observed Be stars.
In the text
thumbnail Fig. 14.

Apparent V/R changes observed for the Hα and He I 6678 emission lines intercompared for two time segments. The variation in the shell He RV is also shown. An apparently anti-phase behaviour is observed in the second time interval, when the shell RV becomes quite negative and the He shell line blends with the V peak of the faint He emission. The same colours as in the previous time plots are used to distinguish spectra from individual observatories.
In the text
thumbnail Fig. 15.

Apparent V/R changes observed for the Hα and He I 6678 emission lines intercompared for two dates. An apparently anti-phase behaviour is observed for the later date, when the He I 6678 shell RV becomes quite negative and the He shell line blends with the V peak of the faint He emission.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.