JSTOR

1. INTRODUCTION

Var Her 04 (EQ J183926+260409; , mas [J2000.0]; GCVS designation not yet assigned) is a newly discovered cataclysmic variable in Hercules. It was discovered at photographic (Tri‐X) magnitude 11.5 by Yuji Nakamura (Kameyama, Mie‐ken, Japan) on 2004 June 13.632 UT (Green et al. 2004). Announcement of the discovery was made on 2004 June 22 by the American Association of Variable Star Observers (AAVSO) after consultation with the IAU Central Bureau of Astronomical Telegrams (CBAT).1 The first known observation was found in the All Sky Automated Survey 3 database (ASAS‐3; Pojmanski 2002) on 2004 June 15.2462, when it was measured at . The first AAVSO‐coordinated observation was made on 2004 June 23.677.

Var Her 04 is the northwest component of a double star separated by (Fig. 1). Astrometry was measured using the 1.55 m USNO Flagstaff Station (NOFS) telescope. The southeast component matches the position of 2MASS J18392619+2604087 ( , mas [J2000.0]) within its margin of error. The southeast component has and likely contaminates all observations made without filters or in and by a system that cannot separate the two. This includes all observations in this paper, unless otherwise noted. Var Her 04’s location falls within the error margins of the location of 1RXS J183927.1+260409, a bright X‐ray source in the ROSAT catalog (average hardness ratio ; Voges et al. 1999).

(35 KB)

Fig. 1.— Var Her 04 (northwest component) and companion in at quiescence. Image taken on 2004 August 15, with the University of Hawaii 2.2 m telescope (0 2 pixel⁻¹; north is up, east is right; image ). Photometry of Ver Her 04 in quiescence is contaminated by the companion.

Open New Window

2. OBSERVATIONS

AAVSO observers made 7551 CCD measurements of the object over 16 days following the announcement of the discovery. Observers reduced their own data using comparison stars published on an AAVSO chart. They were also instructed to synchronize their imaging computer clocks with the USNO Network Time Protocol (NTP) server. Initial photometric calibration of the field was done by B. Gary from Hereford, Arizona, over multiple nights with a 0.36 m telescope, ST‐8XE CCD, and filters at prime focus (Table 1). Comparison stars were chosen based on brightness and to avoid variability associated with M‐class stars. Later, precision photometry was obtained over multiple nights using the USNO‐FS 1.0 m telescope, along with a large set of Landolt standards (Landolt 1992) covering a range of color and air mass. The field was observed over multiple nights with the University of Hawaii 2.2 m telescope on Mauna Kea. These images were calibrated with the comparison stars from the photometry of A. Henden. Using point‐spread function fitting, Henden also obtained precise astrometry for both the variable and companion, along with photometry of the latter.

TABLE 1 Comparison Star Photometry Used by AAVSO Observers

Open New Window

The variable was observed at the 1.8 m Vatican Advanced Technology Telescope on 2004 June 24 (UT). CCD data were taken in the Johnson V filter continuously for 5 hr. The majority of the images were 10 s exposures (plus a 30 s readout). The data were bias‐corrected and flat‐fielded using twilight sky images.

The optical light curve (Fig. 2) is characterized by a steady decline of 0.17 mag day⁻¹ until 2004 June 24.5000 (JD 2,453,181.00), when an inflection in the light curve gave rise to a gentle hump that spanned 0.1 mag over 2 days. On 2004 June 26.5 (JD 2,453,182.00), the light curve resumed a slow decline until 2004 June 30.0358 (JD 2,453,186.55) when a rapid decline began. The decline resulted in a drop of 1.1 mag. The last observation of the decline was made 0.5062 days after it began. The star remained steady around through the end of observations in early September. Afterwards, photometry was obtained during quiescence, providing a measurement of (Table 2). Precision photometry of the companion revealed some brightening during the time period that the cataclysmic variable (CV) was declining, and then it too went into a state of quiescence (Table 3).

(42 KB)

Fig. 2.— Var Her 04’s light curve. Data averaged from AAVSO sources. Combining faint unfiltered observations leads to the larger errors at late time.

Open New Window

TABLE 2 Precision Photometry of the Cataclysmic Variable in Quiescence

Open New Window

TABLE 3 Photometry of the Companion

Open New Window

A blue spectrum was obtained on 2004 June 23.32 UT and again on 2004 June 25.45 with the FAST spectrograph (Fabricant et al. 1998) at the Cassegrain focus of the 1.5 m Tillinghast telescope at the F. L. Whipple Observatory on Mount Hopkins. The FAST spectrograph uses a Loral CCD with a spatial scale of 1 1 pixel⁻¹ in the binning mode used for these observations. The exposure time for the observation was 360 s. The slit was oriented with a position angle of 70°; this was not the optimal parallactic angle (Filippenko 1982), but the low air mass (1.01) and wide slit (3 ) imply that differential slit losses were insignificant. Standard CCD processing and optimal spectral extraction were done with IRAF.2 Custom routines were used to calibrate the flux data using the sdO comparison star BD +28°4211 (Stone 1977) in the region of 3400–4500 Å, and the sdG comparison star BD +26°2606 (Oke & Gunn 1983) in the 4500–7500 Å region. Telluric absorption features were removed using the intrinsically featureless spectrum of BD +26°2606 (Wade & Horne 1988; Matheson et al. 2000).

The spectrum (Fig. 3) shows a blue continuum with double‐peaked Balmer lines. The FWHM of the Hα line is 2000 km s⁻¹, and it shows some change in shape between the two nights (Fig. 4). The Balmer lines become dominated by absorption at the blue end of the spectrum. Weak, double‐peaked emission lines of He i are detected, in addition to a Na i absorption feature. Very broad bumps identified as He ii and high ionization states of carbon and nitrogen, generally associated with Wolf‐Rayet features, are also seen. The spectrum is similar to that of WZ Sge during outburst (Brosch et al. 1980).

(65 KB)

Fig. 3.— Blue (370–750 nm) spectra from Mount Hopkins on 2004 June 23.32, revealing the typical spectrum of a cataclysmic variable.

Open New Window

(80 KB)

Fig. 4.— Comparison of Hα and Hβ features in the Mount Hopkins spectra taken on 2004 June 23.32 and 2004 June 25.00.

Open New Window

3. SUPERHUMPS

Superhumps were observed from the beginning of AAVSO observations on 2004 June 23.3677 (JD 2,453,179.8677) to 2004 June 30.5592 (JD 2,453,187.0592; Fig. 5). Before combining data for statistical analysis, each observer’s data set was individually transformed to a uniform zero point by subtracting a linear fit from each night’s observations. This was done so that we could remove the overall trend of the outburst and combine all observations into a single data set. We used a date‐compensated discrete Fourier transform (Ferraz‐Mello 1981), combined with the CLEANest algorithm (Foster 1995) for period analysis. Superhumps appeared in the earliest time series data set beginning on 2004 June 23.3677. Analysis of the observations until 2004 June 30.5592 (JD 2,453,187.0592) give a superhump period of days (Fig. 6). Then a period of days emerged in quiescence (Fig. 7). The power spectrum in quiescence is also more complicated. This is probably because of noise from the faint photometry and influence of the red companion. Further time series CCD observations spread out over the next 10 days revealed no significant change in the power spectrum.

(101 KB)

Fig. 5.— Superhumps during outburst.

Open New Window

(31 KB)

Fig. 6.— Power spectrum from 2004 June 23.3677 to 2004 June 30.5592, revealing a strong superhump period of days ( cycles day⁻¹) .

Open New Window

(43 KB)

Fig. 7.— Quiescence power spectrum from 2004 July 1.5 to July 9.2. The orbital period of days ( cycles day⁻¹) emerges among noise caused by the faintness of the object and the influence of the companion.

Open New Window

4. ARCHIVAL DATA

The blue RH‐series patrol plates at the Harvard College Observatory Plate Stacks were inspected for previous outbursts on plates dating from 1929 to 1950. Four outbursts were detected, and one more possible outburst was detected at the plate limit (Table 4). The ASAS‐3 does not detect the star during almost nightly observations from 2003 March 30.4 to 2004 June 6.3. A search through the orphan files in the Northern Sky Variability Survey database (NSVS; Woźniak et al. 2004) revealed no detected outbursts from 1999 April 5.5 to 2000 March 23.5.

TABLE 4 Previous Outbursts on Harvard Patrol Plates

Open New Window

5. ANALYSIS

The orbital period, outburst amplitude, and lack of a recently detected outburst3 suggest classification as a ugwz dwarf nova. However, Var Her 04 displayed more frequent and consistent outburst behavior from 1929 to 1950, and there is an absence of the expected ugwz echo outbursts in the 2004 outburst.

If days is the superhump period and days is the orbital period, then we derive a superhump‐period excess and an approximate mass ratio of , as derived from (Patterson 1998)

There are other objects that display similar erratic outburst behavior and have orbital periods and mass ratios similar to Var Her 04: EG Cnc, WZ Sge, AL Com, and DU UMa (Patterson 1998). This small group of stars are believed to be hydrogen‐burning “period bouncers,” which are dwarf novae that have periods that are lengthening after evolving through the period minimum. These are theoretically the most common type of CV, but also the most elusive, as a result of their inherently faint nature and infrequent outbursts. The most obvious common trait of these stars is very low ϵ, which is dependent on mass ratio. The low mass ratio implies a very small secondary, which provides a hint toward the relative age of the system, with older systems more likely to have survived a period bounce.

The faint red companion is close to the variable, making it difficult to obtain precision photometry and astrometry of both stars during quiescence. Photometry of the companion suggests a hint of brightening that is coincident with the outburst of the variable and beyond the expected photometric error. The online POSS I/II averaged coordinates were checked for proper motion, but they consist of the averaged position of the red and blue plates. The former will be contaminated by the companion, so proper motion could not be estimated. Additionally, direct parallax measurement would be quite difficult because of the companion, but may be possible in 3 years with a large‐aperture telescope. The change in position angle should be detectable in a few years if the stars are part of the same system. If so, then the companion may have been reflecting outburst light from the variable.

6. CONCLUSIONS

AAVSO observations and study of recent and past Var Her 04 superoutbursts reveal an enigmatic system with an outburst light curve and supercycle unlike any other known object. Var Her 04 is most similar to the small group of now five possible ugsu hydrogen‐burning “period bouncers.” Searches for past outbursts in other archives, and careful monitoring of current behavior, is needed to establish supercycle behavior evolution. AAVSO observers will carefully monitor the star over the next few years for a new outburst. Precision photometry, X‐ray observations, and quality spectra in quiescence would be useful in firmly establishing the mass ratio of the system. In addition, future proper motion studies of the companion using large aperture telescopes is needed to determine any relationship with the faint red companion.