THE SOLAR NEIGHBORHOOD. XXI. PARALLAX RESULTS FROM THE CTIOPI 0.9 m PROGRAM: 20 NEW MEMBERS OF THE 25 PARSEC WHITE DWARF SAMPLE

Photometric observations have been collected since the inception of CTIOPI during scheduled observing runs when sky conditions were photometric. Standard stars from Graham (1982) and Landolt (1992, 2007) were taken nightly through a range of airmasses to calibrate fluxes to the Johnson–Kron–Cousins system and to calculate extinction corrections. Bias subtraction and flat-fielding (using calibration frames taken nightly) were performed using standard IRAF packages. An aperture of 14'' diameter was used when possible (consistent with Landolt 1992) to determine stellar flux. Cosmic rays within this aperture were removed before flux extraction. Aperture corrections were applied when neighboring sources fell within the adopted aperture. In these cases, the largest aperture that did not include flux from the contaminating source was used and ranged from 4'' to 12'' in diameter. Total uncertainties (including internal night-to-night variations as well as external fits to the standard stars) in the optical photometry are ± 0.03 mag in each filter (Henry et al. 2004). In the cases of WD 0628−020 and WD 2351−335, the primaries (red dwarfs) were significant contaminants (especially in I) so that aperture photometry alone was not possible. Instead, a point-spread function (PSF) fit using interactive data language (IDL) package mpfit2dpeak was generated for the primaries and then subtracted from the images. Aperture photometry was then performed on the subtracted images. The same procedure was performed for extracting photometry for the primaries (i.e., the WD secondaries were PSF fit and removed).

Our photometric results are given in Table 3, which is divided into three samples based on the trigonometric parallaxes presented here: (1) new 25 pc WD members, (2) WDs beyond 25 pc, and (3) known 10 pc ASPENS targets. Companions to WDs (i.e., LP600−43, LHS 234 and LHS 4039) for which photometric analyses were performed are included in the table just below their WD companions.

Table 3. Photometric Results

WD Name (1)	Alternate Name (2)	VJ (3)	RKC (4)	IKC (5)	No. of Nights (6)	π Filter (7)	σ (mag) (8)	No. of Nights (9)	No. of Frames (10)	J (11)	σJ (12)	H (13)	σH (14)	KS (15)	$\sigma _{K_S}$ (16)	Notes (17)
New 25 pc White Dwarfs
0038−226	LHS 1126	14.50	14.08	13.66	3	R	0.006	21	90	13.34	0.03	13.48	0.03	13.74	0.04
0121−429	LHS 1243	14.83	14.52	14.19	4	R	0.005	11	60	13.86	0.02	13.63	0.04	13.53	0.04
0141−675	LHS 145	13.82	13.52	13.23	3	V	0.007	29	166	12.87	0.02	12.66	0.03	12.58	0.03
0419−487	GJ 2034	14.37	13.76	12.46	3	R	0.010	12	64	10.72	0.02	10.15	0.02	9.85	0.03	a
0628−020	LP 600−42	15.32	15.06	14.75	3	I	[0.040]	15	70	⋅⋅⋅	⋅⋅⋅	⋅⋅⋅	⋅⋅⋅	⋅⋅⋅	⋅⋅⋅	b,c
⋅⋅⋅	LP 600−43	15.50	14.13	12.41	3	I	0.007	15	74	10.73	0.03	10.14	0.03	9.86	0.02	c,d
0751−252	SCR 0753−2524	16.27	15.78	15.31	4	R	[0.029]	13	45	14.75	0.03	14.47	0.03	14.30	0.09	b
0806−661	L97−3	13.73	13.66	13.61	4	R	0.007	13	65	13.70	0.02	13.74	0.03	13.78	0.04
0821−669	SCR 0821−6703	15.34	14.82	14.32	3	R	0.007	17	86	13.79	0.03	13.57	0.03	13.34	0.04
1009−184	WT 1759	15.44	15.18	14.91	3	I	0.007	18	77	14.68	0.04	14.52	0.06	14.31	0.07
1036−204	LHS 2293	16.24	15.54	15.34	3	R	0.006	14	52	14.63	0.03	14.35	0.04	14.04	0.07
1202−232	LP 852−7	12.80	12.66	12.52	3	R	0.007	16	75	12.40	0.02	12.30	0.03	12.34	0.03
1223−659	GJ 2092	14.02	13.82	13.62	3	V	[0.028]	14	61	13.33	0.04	13.26	0.06	13.30	0.06	b
1315−781	L40−116	16.15	15.74	15.36	3	R	0.009	15	63	14.89	0.04	14.67	0.08	14.58	0.12
1436−781	LTT 5814	16.10	15.81	15.48	3	R	0.006	18	63	15.04	0.04	14.88	0.08	14.76	0.14
2008−600	SCR 2012−5956	15.84	15.40	14.99	4	V	0.008	21	84	14.93	0.05	15.23	0.11	15.41	Null
2008−799	SCR 2016−7945	16.35	15.96	15.57	4	R	0.008	12	55	15.11	0.04	15.03	0.08	14.64	0.09
2040−392	L495−82	13.75	13.76	13.69	3	R	0.019	16	67	13.78	0.02	13.82	0.03	13.81	0.05	a,e
2138−332	L570−26	14.48	14.31	14.16	4	V	0.007	16	67	14.17	0.03	14.08	0.04	13.95	0.06
2336−079	GD 1212	13.28	13.27	13.24	4	R	0.009	19	74	13.34	0.03	13.34	0.02	13.35	0.03
2351−335	LHS 4040	14.52	14.38	14.19	3	I	[0.043]	13	62	13.99	0.11	13.86	0.25	13.73	0.11	b,c
⋅⋅⋅	LHS 4039	13.46	12.33	10.86	3	I	0.008	13	62	9.48	0.02	8.91	0.02	8.61	0.02	c,f
Beyond 25 pc White Dwarfs
0928−713	L64−40	15.11	14.97	14.83	3	R	0.006	16	66	14.77	0.03	14.69	0.06	14.68	0.09
1647−327	LHS 3245	16.20	15.85	15.49	3	R	0.009	12	46	15.15	0.05	14.82	0.08	14.76	0.11
2007−219	GJ 781.3	14.40	14.33	14.25	3	V	0.007	32	146	14.19	0.02	14.20	0.04	14.26	0.08
Known 10 pc White Dwarfs (ASPENS Targets)
0552−041	LHS 32	14.47	13.99	13.51	3	R	0.006	23	156	13.05	0.03	12.86	0.03	12.78	0.03
0738−172	LHS 235	13.06	12.89	12.72	4	I	0.009	15	92	12.65	0.02	12.61	0.03	12.58	0.04
⋅⋅⋅	LHS 234	16.69	14.69	12.41	4	I	0.006	15	92	10.16	0.02	9.63	0.02	9.29	0.02	g
0752−676	LHS 34	13.96	13.58	13.20	3	R	0.006	12	70	12.73	0.02	12.48	0.03	12.36	0.02
0839−327	LHS 253	11.86	11.77	11.65	3	V	0.007	16	94	11.58	0.03	11.54	0.03	11.55	0.03
1142−645	LHS 43	11.50	11.34	11.20	3	V	0.007	28	173	11.18	0.01	11.13	0.04	11.10	0.03
2251−070	LHS 69	15.70	15.11	14.56	3	R	0.007	16	83	14.01	0.03	13.69	0.04	13.55	0.05
2359−434	LHS 1005	12.97	12.82	12.66	3	R	0.007	12	87	12.60	0.03	12.43	0.02	12.45	0.02

Notes. ^aLikely variable at the ∼1%–2% level (see Section 5.2). ^bVariability analysis contaminated by nearby source, hence the brackets in Column 8 indicating erroneous variability. ^cOptical photometry was extracted using PSF fitting rather than by an aperture. ^dCommon proper motion companion to WD 0628−020. ^eNew ZZ Ceti pulsating WD. ^fCommon proper motion companion to WD 2351−335. ^gCommon proper motion companion to WD 0738−172.

Download table as:  ASCIITypeset image

Multi-epoch optical VRI photometry (Columns 3–5) as well as near-infrared JHK_S photometry (Columns 11–16) extracted from the Two Micron All Sky Survey (2MASS) database are listed in Table 3. We performed a photometric variability analysis of the parallax target (hereafter referred to as the "PI star") relative to the reference stars using the parallax data taken in a single filter, as outlined in Honeycutt (1992). Columns 6–10 list the number of different nights the object was observed for photometry, the parallax filter, the standard deviation of the PI star's magnitude in that filter from the variability analysis, the number of nights parallax data were taken, and the total number of parallax frames used in the variability analysis. In general, objects with standard deviation values larger than 0.02 mag are considered variable, those with values between 0.01 and 0.02 mag are likely variable at a few percent level, and those less than 0.01 mag are "steady." There are four cases where the standard deviations are larger than 0.02 mag (WD 0628−020, WD 0751−252, WD 1223−659, and WD 2351−335) and in all four cases, there are contaminating sources within a few arcseconds. Thus, the large standard deviations are due to varying degrees of contamination within the photometric aperture depending on seeing conditions rather than intrinsic variability. Two objects show standard deviations between 0.01 and 0.02 mag (WD 0419−487 and WD 2040−392). WD 0419−487 has an unresolved red dwarf companion in a short period orbit (see Section 5.2) so that mild variability might be expected. WD 2040−392 is a new pulsating ZZ Ceti WD (see Section 5.2).

A complete discussion of parallax data acquisition and reduction techniques can be found in Jao et al. (2005). Briefly, once an object (or system) is selected to be observed for a trigonometric parallax determination, the object needs to be "set up." This process consists of selecting a telescope pointing as well as a filter bandpass (VRI) with which all subsequent astrometry observations will be taken. The telescope pointing is selected so that a fairly homogeneous distribution of reference stars encircle the PI star, do not reside near bad columns, and are as close as possible to the PI star rather than near the edges of the CCD. Also, every effort is made to place the PI star as close to the center of the CCD as possible. Of course, compromises are made in cases of sparse fields.

Once sufficient data have been collected (criteria that define a definitive parallax are outlined in Section 3.2.1), each frame is inspected and poor quality frames (e.g., bad seeing, telescope guiding problems) are discarded. Centroids for the reference field and PI star are extracted using SExtractor (see the discussion in Section 3.2.2; Bertin & Arnouts 1996). The centroids are corrected for differential color refraction (DCR) based on the color of each reference star and PI star. Typically, this correction shifts the stars' positions by no more than a few mas.

Automatic quality control constraints on ellipticity, elongation, and full width at half-maximum (FWHM) eliminate frames and individual stars (reference and PI) of poor quality not noticed during manual inspection. A fundamental trail plate is chosen as a reference plate and rotated based on a comparison with either the Guide Star Catalog 2.2 or 2MASS (used by default except when there are too few stars in the PI star field that are catalogued in the 2MASS database). All star centroids on other plates are recalibrated to account for different scaling in both the X and Y directions, as well as the different amounts of translation and rotation. A least-squares reduction via the Gaussfit program (Jefferys et al. 1988) is performed, assuming the reference star grid has Σπ_i= 0 and Σμ_i= 0, where π and μ are parallax and proper motion, respectively. After verification of a good reference field (e.g., no reference stars within 100 pc, no problematic reference stars because of unresolved multiplicity), the final reduction produces a relative trigonometric parallax for the PI star.

Even the reference stars trace out small parallax ellipses because these stars are not infinitely far away. Thus, a correction to absolute parallax must be performed. This is accomplished in one of (at least) three different ways: (1) using a model of the Galaxy for the disk and halo, (2) spectroscopic parallaxes for each of the reference stars, or (3) photometric parallaxes for each of the reference stars. Because accurate photometry is needed for DCR correction and is already available, we use photometric parallaxes for the reference stars to correct to absolute parallax using the CCD distance relations of Henry et al. (2004). These relations assume the reference stars are single main-sequence dwarfs and do not take into account contamination from evolved stars, unresolved double stars, or reddening. We identify and remove outlying points estimated to be within ∼100 pc due either to evolved stars or unresolved double stars whose distances are underestimated. Given that all of the WDs presented here are relatively nearby, PI star reddening is negligible. However, the distant reference stars can be reddened and appear to be closer than they truly are. This is the case for two WD fields (WD 1223−659 and WD 1647−327) for which we have adopted an average correction to absolute of 1.2 mas (based on all other WD fields presented here) with a conservative error of 0.3 mas (consistent with the largest absolute correction errors presented here).

3.2.1. Definitive Parallax Criteria

3.2.2. New SExtractor Centroiding Algorithm

3.2.3. Cracked V Filter

Atmospheric modeling procedures of the WDs are identical to those presented in Subasavage et al. (2008) and references therein, with the exception that the trigonometric parallax constrains the surface gravity (instead of assuming log g= 8 as was done in that publication). Briefly, optical/near-IR magnitudes are converted into fluxes using the calibration of Holberg et al. (2006) and compared to the SEDs predicted by the model atmosphere calculations. The observed flux, f^m_λ, is related to the model flux by the equation

$$\begin{equation} f_{\lambda }^m= 4\pi (R/D)^2 H_{\lambda }^m \end{equation} \tag{ 1 }$$

where R/D is the ratio of the radius of the star to its distance from Earth, H^m_λ is the Eddington flux (dependent on T_eff, log g, and atmospheric composition) properly averaged over the corresponding filter bandpass, and π in this context is the mathematical constant (elsewhere throughout this paper, π refers to the trigonometric parallax angle). Our fitting technique relies on the nonlinear least-squares method of Levenberg-Marquardt (Press et al. 1992), which is based on a steepest descent method. The value of χ² is taken as the sum over all bandpasses of the difference between both sides of Equation (1), weighted by the corresponding photometric uncertainties. Only T_eff and [π(R/D)²] are free parameters and the uncertainties of both parameters are obtained directly from the covariance matrix of the fit. The main atmospheric constituent (hydrogen or helium) is determined by the presence of Hα from spectra published in the literature (references listed in Table 5) or by comparing fits obtained with both compositions.

We start with log g = 8.0 and determine T_eff and [π(R/D)²], which combined with the distance D obtained from the trigonometric parallax measurement yields directly the radius of the star R. The radius is then converted into mass using evolutionary models similar to those described in Fontaine et al. (2001) but with C/O cores, q(He) ≡ log M_He/M_⋆ = 10⁻² and q(H) = 10⁻⁴ (representative of hydrogen-atmosphere WDs), and q(He) = 10⁻² and q(H) = 10⁻¹⁰ (representative of helium-atmosphere WDs).¹⁴ In general, the log g value obtained from the inferred mass and radius (g = GM/R²) will be different from our initial guess of log g = 8.0, and the fitting procedure is thus repeated until an internal consistency in log g is reached. The parameter uncertainties are obtained by propagating the error of the trigonometric parallax measurements into the fitting procedure.

Physical parameter determinations for the DQ and DZ WDs are identical to the procedures outlined in Dufour et al. (2005, 2007). Briefly, the photometric SED provides a first estimate of the atmospheric parameters with an assumed value of metal abundances using solar abundance ratios. The optical spectrum is fit to better constrain the metal abundances and to improve the atmospheric parameters from the photometric SED. This procedure is iterated until a self-consistent photometric and spectroscopic solution is reached.

Only two objects' spectra are modeled here for the first time, WD 0806−661 and WD 1009−184. For the remaining DQ (those with carbon) and DZ (those with calcium) stars, spectral modeling to obtain abundances was performed and presented in Dufour et al. (2005, 2007) and Subasavage et al. (2007, 2008). The atmospheric abundances will not change with the inclusion of the parallaxes; however, the surface gravities (hence masses) are sensitive to changes in distance and have been updated in Table 5. For WD 0806−661, the optical spectrum shows no carbon absorption, yet it is classified as a DQ based on ultraviolet (UV) spectra. Thus, the UV spectrum was used for fitting (see Section 5.2). For DZ stars, trace amounts of hydrogen not directly visible can be present in the atmosphere and affect the spectral profiles of the calcium absorption lines (Dufour et al. 2007). In the case of WD 1009−184, whose spectrum was obtained using the same telescope/instrument setup and reduction procedures as described in Subasavage et al. (2008), the spectral fit including a log (H/He) =−3 better reproduced the calcium H & K lines than if no hydrogen were present (see Section 5.2).

In order to best constrain the physical parameters for this sample, weighted mean parallaxes and errors are calculated for systems with previous parallax determinations as well as those that have common proper motion companions with previous/new parallax determinations. The parallax values that are used to model the physical parameters are listed in Table 5 (Column 2) as well as the number of individual parallaxes used in the weighted mean and corresponding references (Columns 3 and 4). Columns 5 and 6 list the effective temperatures and surface gravities as well as corresponding errors. Columns 7 and 8 list the composition(s) used in the atmospheric modeling (with any secondary constituents listed in parentheses) and the spectral-type reference. Columns 9–13 list the derived masses, absolute magnitudes, luminosities, WD ages (not including main-sequence lifetimes), and any notes for the systems. While the errors listed for mass, luminosity, and age are formal errors, they are remarkably well constrained when accurate trigonometric parallaxes are available.

WD 0038−226. This WD has mild absorption bands similar to the carbon Swan bands found in DQ stars but the bands are shifted blueward. Initially, it was thought that these bands were actually the Swan bands but were pressure-shifted because of increased pressures in cool He-rich WD atmospheres (Liebert & Dahn 1983). Another explanation was that the features were caused by the hydrocarbon C₂H (Schmidt et al. 1995). The most recent possible explanations revisit the idea of pressure-shifted Swan bands alone or in conjunction with Swan bands produced by highly rotationally excited C₂ (Hall & Maxwell 2008). Additional theoretical investigations are necessary to better understand the properties of C₂ in the high-pressure, high-temperature helium environment that a WD atmosphere would provide. We present a significantly better parallax (the previous parallax had an error = 10.3%), that confirms this object is within 10 pc and is the nearest known WD with the aforementioned spectral anomaly.

In addition, Bergeron et al. (1994) have shown that this object displays collision-induced absorption in the infrared and can be attributed to collisions of molecular hydrogen with helium. Bergeron et al. (1997) have modeled the SED using a mixed H/He composition and arrived at a satisfactory fit. The updated physical parameters found in Table 5 were derived by an identical analysis (including the use of their BVRIJHK photometry) except with the updated trigonometric parallax presented here. The helium abundance derived with the updated parallax [log (He/H) = 1.31] is less than that found by Bergeron et al. (1997) [log (He/H) = 1.86].

WD 0121−429. A recently discovered magnetic DA that is thought to be an unresolved double degenerate with one component being a magnetic DA and the other being a featureless DC (Subasavage et al. 2007). The physical parameters are not listed in Table 5 because any number of component masses and luminosities can reproduce the SED fit.

WD 0141−675. A DA WD that was one of only two new 25 pc WD members to be within 10 pc.

WD 0419−487. A WD + red dwarf pre-cataclysmic eclipsing binary also known as RR Caeli with an orbital period of 7.3 hr for which Maxted et al. (2007) derive masses and radii. Contamination from the main-sequence component is significant both spectroscopically and photometrically so that no atmospheric modeling of the WD was possible (it is the outlying point in the H–R Diagram in Figure 3).

WD 0628−020. A known WD that has a red dwarf companion (exact spectral type unknown) with a separation of 45 at 3179. Parallax data were taken in I to optimize observations for the red dwarf companion. Thus, the parallax for the WD is more uncertain because it is somewhat poorly exposed in our images. The weighted mean parallax listed in Table 5, which represents our two measurements, should be taken as the distance to the system. The proper motions of the components differ by several sigma of the formal errors while the position angles are consistent (though the errors are fairly large). By comparing astrometric results of several distant wide binaries on CTIOPI (as yet unpublished), we find that the proper motion and position angle values agree to within 1–2σ. Thus, the formal errors are likely not significantly understated. We make some basic assumptions about component masses and orientation (i.e., face-on orbit) of the system to evaluate the plausibility of orbital motion to account for the discrepancy. Indeed, it is probable that orbital motion during the nearly four years of observation of this closely separated binary at 20.48 ± 0.47 pc is likely the cause of the discrepancy in μ.

WD 0738−172. A known nearby WD that has a M6.0V companion with a separation of 206 at position angle 2610. Because the companion is fainter in both V and R as well as redder, the parallax observations were taken in I so that an independent parallax to the secondary red dwarf (of comparable brightness to the primary at I) could also be obtained. The parallax values for the pair are in excellent agreement (see Table 4) but the proper motion values differ by several sigma of the formal errors. Similar to WD 0628–020, we make basic assumptions about the system and again, orbital motion during four years of observation of this nearby (9.14 ± 0.05 pc) system is likely the cause of the discrepancies in μ and position angle.

WD 0751−252. A recently discovered WD that is a common proper motion companion to LTT 2976 (Subasavage et al. 2005b). The trigonometric parallax determined in this work for the WD (56.54 ± 0.95 mas) is in marginal agreement with the Hipparcos parallax for the primary (51.52 ± 1.46 mas; van Leeuwen 2007).

WD 0806−661. A DQ WD that shows carbon features only in the UV (in the optical, it appears as a featureless DC). Thus, a UV spectrum from the International Ultraviolet Explorer (IUE) archive was included in the analysis. First, a model that includes carbon with an abundance below the detectability limit in the optical (e.g., log (C/He) ∼ −4 to −6) was fit to the VRIJHK magnitudes. However, the effective temperature and surface gravity derived from this model produced a poor fit to the UV spectrum. Then, the UV spectrum was fit independent of the photometry (i.e., T_eff and log g were allowed to vary). This produced a much better fit with exception of two carbon lines at 1657 Å and 1930 Å (see Figure 6). The asymmetry in the observed spectrum is not reproduced in the model fit likely because of the failure of the impact approximation for the van der Waals broadening used in the model (Koester et al. 1982). With the T_eff and log g fixed from the spectroscopic fit, the photometry was again fit to arrive at the final physical parameters found in Table 5, including an abundance of log (C/He) =−5.55 ± 0.12. We note that the discrepancy in the temperatures derived from the UV spectrum and the optical/near-IR photometry is fairly small (∼10,250 K versus ∼11,300 K), though significant. It is possible that the models fail to address some component of the input physics, such as a missing opacity source that gives rise to this discrepancy between UV and optical effective temperature determinations.

Zoom In Zoom Out Reset image size

WD 0821−669. A cool DA WD that was uncovered during a trawl of the SuperCOSMOS Sky Survey (SSS) database (Subasavage et al. 2005a, 2007). It is one of the oldest and nearest WDs of the new 25 pc members – 6.22 ± 0.16 Gyr at a distance of 10.65 ± 0.12 pc. Thus, it is now the 23rd nearest WD system.

WD 1009−184. A DZ WD that is difficult to model accurately because its effective temperature is near the point at which additional pressure effects become important and are not included in the DZ models. Also, trace amounts of hydrogen may exist in the atmosphere but at levels too low to detect spectroscopically, yet affect the profile of the calcium absorption as discussed in Dufour et al. (2007). Our best fit using both photometry and spectroscopy produces a T_eff= 5940 ± 280 K, including trace amounts of hydrogen [log (H/He) =−3] and an abundance of log (Ca/He) =−10.37 ± 0.20 (see Figure 7). However, it is likely that additional pressure effects have an impact, so the physical parameters should be considered preliminary estimates.

Zoom In Zoom Out Reset image size

WD 1036−204. A peculiar DQ WD that has carbon Swan absorption bands; however, the absorption is significantly deeper in this object because of the presence of a large magnetic field (Bues 1999). Again, no atmospheric modeling was possible but this object will serve as another important test case for model revisions in the future.

WD 2008−600. A DC WD that was recently discovered to have a flux deficiency in the infrared because of collisions by molecular hydrogen with helium, similar to WD 0038−226. Subasavage et al. (2007) have modeled this object using a mixed H/He composition and a preliminary trigonometric parallax. The updated physical parameters listed in Table 5 replace the preliminary parallax with the more accurate one presented here, and we derive a helium abundance of log (He/H) = 2.60. Also, this object has the largest tangential velocity (112 km sec⁻¹) of the systems presented here.

WD 2040−392. A DA WD that is listed in the McCook-Sion White Dwarf Catalog (McCook & Sion 1999)¹⁵ but had no follow-up observations until Subasavage et al. (2007) obtained a photometric distance estimate of 23.1 ± 4.0 pc (in excellent agreement with the trigonometric parallax distance of 22.63 ± 0.51 pc presented in this work). In addition, as discussed in Section 3.1, this object is variable at the ∼2% level. The physical parameters listed in Table 5 (i.e., T_eff= 10,830 ± 310 K, log g= 8.03 ± 0.04, mass = 0.62 ± 0.02 M_☉, and absolute V= 11.98 ± 0.06) are entirely consistent with new pulsating ZZ Ceti WDs discovered by Gianninas et al. (2006).

For confirmation of ZZ Ceti-type pulsations, data were acquired at the CTIO 0.9 m using the same central quarter of the CCD as used for parallax data. Observations were taken in white light to maximize the signal for the target and eleven reference stars with a temporal resolution of ∼1 minute. Relative aperture photometry was performed on the target and reference stars using an aperture diameter of 12''. The data span 1 hr, during which three cycles of a regular pulsation are clearly evident (see Figure 8), thus confirming the target is a ZZ Ceti pulsator. The Fourier (amplitude) spectrum identifies the dominant period of ∼980 s with an amplitude of ∼3%.

Zoom In Zoom Out Reset image size

WD 2251−070. A cool DZ WD for which the atmospheric model appropriate for DZs fails to reproduce the observed spectrum (see Dufour et al. 2007) likely because of additional pressure effects not accounted for in the model. Also, this object's effective temperature is at the limit of the model grid. Physical parameters should be considered preliminary estimates.

WD 2336−079. A DA WD that Gianninas et al. (2006) recently discovered is a pulsating ZZ Ceti WD. T_eff and log g listed in that publication are uncertain. With the high quality parallax presented here, the parameters are now better constrained and in reasonable agreement with Gianninas et al. (2006). Also, with a minuscule proper motion of 33.6 ± 1.0 mas, this object is the slowest moving WD known in the 25 pc WD sample. In fact, it was first cataloged by Giclas et al. (1975) and labeled as a suspected WD simply based on its blue color. Indeed, this object has the largest effective temperature of all of the systems presented here. Thus, this object provides additional support for the possibility that a sample of nearby WDs with little to no proper motions may exist, especially given that the majority of WDs in a volume-limited sample are cooler such that their colors alone are unexceptional.

WD 2351−335. A DA WD that has both a M4V primary (LHS 4039) and a recently discovered active M8.5V companion (APMPM J2354-3316C; Scholz et al. 2004). The WD has a separation of 65 at position angle 1822 from the primary and the M8.5V has a separation of 1028 at position angle 913 from the primary. Parallax data were taken in I to optimize observations for the M4V primary. Thus, the secondary WD was poorly exposed in the images giving rise to the largest parallax error of those presented here. The weighted mean parallax listed in Table 5, which represents our two measurements, should be taken as the parallax of the system. The tertiary M8.5V was unknown at the time parallax observations began (mid-2003) so, while the object is visible in our frames, exposure times were not sufficient to obtain usable astrometry.