THE HAWAII INFRARED PARALLAX PROGRAM. I. ULTRACOOL BINARIES AND THE L/T TRANSITION^*,†

2.1.1. Position Measurements

2.1.2. Cross-identifying Detections

2.1.3. Registering Dithers

2.1.4. Accounting for Distortion and Linear Terms

In optimizing the registration of dithers, we allowed for optical distortion as a third-order polynomial function in x and y, which was applied before the tangent projection. These distortion terms were derived from several data sets of the densest target field that lies within the Sloan footprint (2MASS 0850+1057) by fitting our measured (x, y) positions to SDSS-DR7 reference catalog coordinates. SDSS provides the best combination of source density on the sky and positional accuracy (≈40 mas as judged from the rms of our fits) among astrometric reference catalogs currently available. The residuals of our fits using first-, second-, and third-order terms are shown in Figure 2. There was no discernable improvement by including fourth-order terms, so we adopted the best-fit terms up to third order for our distortion solution, shown in Figure 3 with coefficients given in Table 2. We note that we also tried fitting for the distortion from our data alone, since dithered images can in principle constrain any nonlinear terms (e.g., Anderson & King 2003). However, the largest observed offset of any given star between two of our 1' dithers is only ≈2–3 pixels, even though the largest absolute offsets due to distortion are ≈10–20 pixels. Thus, we found that we have more leverage for determining the distortion by using a comparison to an absolute reference catalog. The scatter in the best-fit distortion terms determined from different data sets of 2MASS 0850+1057 reflects this fact as it is much lower for the catalog-matching approach compared with using the internal position residuals alone. We also tested the stability of the distortion pattern by both fixing and fitting for it in dense fields observed throughout our program. The astrometric residuals of star positions did not change significantly, validating our approach of using a single distortion solution for all images.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Table 2. Distortion Coefficients for WIRCam Northeast Array

Term	aij	bij
x2	$\hphantom{-} 1.173\times 10^{-6}$	−6.409 × 10−7
xy	−1.303 × 10−6	$\hphantom{-} 1.117\times 10^{-6}$
y2	$\hphantom{-} 5.105\times 10^{-7}$	−1.191 × 10−6
x3	−5.287 × 10−10	−1.466 × 10−10
x2y	−4.130 × 10−10	−4.589 × 10−10
xy2	−5.338 × 10−10	−3.884 × 10−10
y3	−1.353 × 10−10	−5.872 × 10−10

Notes. To apply this distortion correction, the origin must first be redefined as the optical axis:$x^{\prime } = x -2122.6900\\y^{\prime } = y + 81.6789,$where x and y are the pixel positions measured by SExtractor. Distortion-free positions may then be computed:$x^{\prime \prime } = x^{\prime } + a_{20}x^{\prime 2} + a_{11}x^{\prime }y^{\prime } + a_{02}y^{\prime 2} + a_{30}x^{\prime 3} + \ldots \\y^{\prime \prime } = y^{\prime } + b_{20}x^{\prime 2} + b_{11}x^{\prime }y^{\prime } + b_{02}y^{\prime 2} + b_{30}x^{\prime 3} + \ldots$This distortion correction only applies for the northeast array in the WIRCam mosaic.

Download table as:  ASCIITypeset image

We also accounted for differential aberration and refraction offsets in the process of registering dithered images. Both effects are essentially a linear transformation of star positions, since stars on one side of our 10' field experience slightly different positional offsets due to annual stellar aberration and atmospheric refraction than the opposite side of the field. Differential refraction can cause up to a few × 10⁻⁴ expansion of the scale along the elevation axis, and differential aberration can cause up to a ±2 × 10⁻⁴ seasonal change in the scale. Thus, it is important to account for these effects in order to monitor the stability of WIRCam's linear terms over time and between targets. We computed the appropriate offsets from equations in Kovalevsky & Seidelmann (2004, pp. 121–141) and applied the differential values (i.e., with the median offset subtracted) to the celestial coordinates in our minimization routine.

2.1.5. Resulting Positional Errors

The end product of combining measurements from each dithered data set was a catalog of median positions in celestial coordinates⁷ and the rms for each source as determined from ⩾10 dithered measurements. These rms values correspond to the often quoted astrometric quality metric of the "mean error for a single observation of unit weight" (m.e.1). Monet et al. (1992) quote m.e.1 values of 3–5 mas for the highest S/N stars in the USNO CCD program, Vrba et al. (2004) quote 8–10 mas for the brightest reference stars in the USNO infrared astrometry program, and Tinney et al. (2003) report a median rms of 12 mas for the NTT infrared astrometry program. The ultimate astrometric precision at each epoch may be expected to scale as $1/\sqrt{N_{\rm frames}}$, and the USNO CCD, USNO IR, and NTT programs obtained 1–2, 3, and 8 frames per epoch, respectively. Therefore, their precisions per epoch are 2–4 mas, 5–6 mas, and 4 mas, respectively. For our program, the rms of the position measurements for our targets were typically 6–18 mas (13 mas median; Figure 4). Because we obtained 20–30 frames for each data set, our astrometric precision per epoch is 1.5–3.0 mas (2.8 mas median). Thus, the quality of our astrometry is comparable to or better than previous ground-based parallax programs targeting ultracool dwarfs in the optical or infrared.

Zoom In Zoom Out Reset image size

In order to obtain multi-epoch astrometry for all objects in each of our target fields, we next associated the sources measured in different dithered data sets. We excluded the noisiest measurements from this analysis, typically applying an rms threshold of 30–60 mas (0.1–0.2 pixels). The positional shifts between epochs were estimated using a two-dimensional histogram approach as follows: all n₁ objects from the first image were each paired with all n₂ objects from a second image; the α and δ offsets between all n₁ × n₂ possible pairings were computed and binned in a two-dimensional histogram; the peak bin in (Δα, Δδ) space, which contains the min(n₁, n₂) true pairings, was found; the shift was computed by taking the median of the offsets contained in the peak bin. The bin size used was initially set to be arbitrarily large and then iteratively decreased until the number of pairs in the peak bin was <2 times the expected number (i.e., until true pairs dominated the peak bin). The crossmatching of positions was then performed in a similar fashion as for the individual dithered images: a match radius of 20 was employed to associate objects detected at different epochs. Such a large match radius is needed if the proper motion is large (≳ 1'' yr⁻¹), as is the case for some targets. We excluded sources from the multi-epoch astrometry catalog if they were detected at fewer than half of the epochs. This excludes faint sources that were only well detected in the best conditions, bright sources that were only below the saturation limit in poor conditions, and any other transient sources or long-lived artifacts that may be in the data set.

Because the initial association of object positions was based only on rough estimates of the position offsets between epochs, we optimized this registration by fitting for the offsets as well as allowing for relative changes in linear terms across different data sets. Thus, we replaced the initial guesses of the linear terms generated by the reference catalog matching (except for the first epoch, which we solve for later). We parameterized the linear terms as a rotation, x-axis pixel scale, ratio of y/x-axis pixel scales, and a shear term (Δy∝x). We used MPFIT to perform an unweighted least-squares minimization in a similar fashion as described for the dither matching in the previous section. After the first optimization, we fit every object in the field for proper motion and parallax and temporarily excluded objects that displayed significant parallax (>3σ) or proper motion (>30 mas yr⁻¹). This automated procedure typically excluded no more than 5% of the reference stars, and it always excluded the science target. We then determined the optimal registration solution a second and final time after excluding these objects.

We have performed as much of our analysis as possible using relative astrometry in order to preserve the fidelity of our position measurements. However, we must ultimately tie our astrometry to an absolute reference frame in order to determine, e.g., the actual pixel scale and orientation of our images. The most suitable catalogs for this purpose are 2MASS, which provides positions for infrared sources over the entire sky, and SDSS, which has a higher sky density of sources and higher astrometric precision but more limited sky coverage. In our shallowest images taken with the K_H2-band filter, we found that shallower reference catalogs were usually more appropriate (USNO-B1, Monet et al. 2003; and UCAC-3, Zacharias et al. 2010). For each field, we constructed a reference frame from the catalog that had the most sources in common with our images. We required reference catalog sources to have absolute position errors ⩽150 mas (e.g., for 2MASS: ERRMAJ < 015). We found the rough offset between our own astrometric catalog and the reference sources by using our aforementioned two-dimensional histogram approach. We then matched reference sources to our own using a match radius of 20. We excluded from this analysis any sources in our astrometric catalog that displayed significant proper motion (>30 mas yr⁻¹), as these would have introduced substantial scatter (≳03) to our comparison with reference catalog position measurements from typically ≈5–10 years ago.

Using the sources in common between our science images and the reference catalog (typically ≳ 30 stars; see Table 1), we determined the absolute astrometric frame for our CFHT images. We registered our positions to the reference catalog, allowing for an offset (i.e., to determine the absolute coordinates of our astrometry) and the linear terms. This solution allows us to compute the pixel scale and orientation in an absolute sense, completely replacing the temporary guess from the initial catalog crossmatching. In the final best-fit registration, the rms of all stars about their catalog positions was typically 60–80 mas for 2MASS and 30–50 mas for SDSS. This scatter is dominated by the reference catalog positional errors. (Thus, the actual relative astrometric uncertainties of 2MASS positions over our 10' field are a factor of ∼2 smaller than the nominal catalog errors of 100–150 mas.) After this final absolute calibration we found that our input guess for the absolute coordinates from image headers was accurate to within ≲ 1''.

2.3.1. Astrometric Stability of WIRCam

The best-fit parameters from the registration of multi-epoch data sets to an astrometric reference catalog enable us to assess the long-term astrometric stability of WIRCam. The level of precision with which we are able to monitor the changes in linear terms such as scale and rotation is fundamentally limited in two ways: (1) positional errors both in our data and reference catalogs introduce random and systematic errors in the derived terms and (2) the uncertainty in the higher order distortion terms is a source of systematic error in the derived linear terms. We have assessed the level of uncertainty in the scale introduced by both of these effects through Monte Carlo simulations. To test the contribution of random errors alone (i.e., case 1), we simulated many star fields with random positions distributed uniformly over a 10' × 10' field and found the best-fit scale to match them to a reference catalog that had normally distributed noise added to it. For a reference catalog accurate to 80 mas (i.e., akin to 2MASS), ≈30 reference stars were needed to achieve a fractional precision in the scale of 1 × 10⁻⁴ (Figure 5). This situation is typical of about half of our targets. For a higher fidelity reference catalog accurate to 40 mas (e.g., like SDSS), 30 reference stars give a much better scale precision of 5 × 10⁻⁵, and the very best case among our targets of 190 SDSS reference stars would give a precision of 2 × 10⁻⁵.

Zoom In Zoom Out Reset image size

The second source of error present in our determinations of linear terms is the uncertainty in the distortion solution. This is because the linear and higher order terms are partially degenerate when fitting polynomials for the distortion. In the reduction procedure described above, we used data sets containing ≈200 SDSS reference stars to determine the WIRCam distortion, and the catalog errors were estimated to be 40 mas from the rms of the fit residuals. Thus, we simulated many random star fields each containing 200 stars with normally distributed noise of 40 mas and found that fitting freely for both linear and distortion terms resulted in a scale uncertainty of 3 × 10⁻⁴ (Figure 5). This result is effectively independent of the assumed centroiding error in the star positions even up to our worst errors of 0.1 pixel because the reference catalog scatter dominates. This source of error is a few times larger than the uncertainty due simply to random errors in the reference catalog, and thus it is the limiting factor in our ability to measure the scale of WIRCam. From these simulations, the limiting systematic uncertainties in shear and rotation are 3 × 10⁻⁴ and 002 (i.e., 3 × 10⁻⁴ radians), respectively.

With these results in mind, we can now assess the stability of WIRCam from our astrometric monitoring data (Figure 6). (1) We are most sensitive to changes in the orientation of WIRCam and found a highly significant scatter of ±014 among data sets taken over our program. This scatter is clearly not Gaussian but rather is highly correlated with the observation date; the orientation of data sets taken on the same WIRCam observing run was nearly identical. This is consistent with the fact that the instrument is taken off of the telescope between observing runs. (2) We found the x pixel scale to be 030614 ± 000008 pixel⁻¹ (i.e., a fractional error of 3 × 10⁻⁴). Given the errors estimated above, this scatter is consistent with the pixel scale being constant over the duration of our program. This stability is impressive given that WIRCam is taken on and off the telescope for ∼8 observing runs per year. (3) We found the ratio between y and x pixel scales to be consistent with unity (0.9997 ± 0.0003), and the scatter in this value is consistent with the uncertainty given by our Monte Carlo simulations. (4) Finally, we found a significant shear term (which we have defined as Δy∝x) of −0.0013 ± 0.0004. If the angle between WIRCam's x and y axes were different from the 90° angle between north and east only by a rotation, this term would be zero. Instead, this shear term implies that the angle between WIRCam's x and y axes is actually 8993 ± 002 when projected onto the sky.

Zoom In Zoom Out Reset image size

Using our final astrometric catalog of WIRCam position measurements calibrated against an absolute reference frame, we fit for the proper motion and parallax of all sources in each target field. For each source, we used MPFIT to perform a least-squares minimization weighted by the standard errors of the position measurements. We fitted three parameters to the combined (α, δ) data: proper motion in right ascension (μ_α), proper motion in declination (μ_δ), and parallax (π). This is notably different from one standard approach taken in the literature of fitting two separate values of the parallax in α and δ. The parallax offsets were computed as follows:

$$\begin{equation} \Delta {\alpha } = \pi (X\sin {\alpha } - Y\cos {\alpha })/\cos {\delta } \end{equation} \tag{ 1 }$$

$$\begin{equation} \Delta {\delta } = \pi (X\cos {\alpha }\sin {\delta } + Y\sin {\alpha }\sin {\delta } - Z\cos {\delta }), \end{equation} \tag{ 2 }$$

where X, Y, and Z are the coordinates of the Earth relative to the barycenter of the solar system as given by the JPL ephemeris DE405. MPFIT minimized the residuals in (α, δ) after subtracting the relative parallax and proper motion offsets (three parameters) and the mean (α, δ) position (effectively removing 2 additional degrees of freedom). Thus, each fit to 2 × N_epoch measurements had 2 × N_epoch − 5 degrees of freedom (dof).

For each target, we then performed a Markov Chain Monte Carlo (MCMC) analysis on the astrometry in order to accurately determine the posterior distributions of all parameters. We adopted the formalism described by Ford (2005), which uses a Metropolis–Hastings jump acceptance criterion with Gibbs sampling that chooses only one parameter (at random) to be altered at each step in the chain. Before running our science chains, we first ran a test chain to determine the optimal step size (β) for each of our parameters in order to ensure efficient convergence. This initial chain was run according to the procedure outlined by Ford (2006) in which each value of β is periodically adjusted until the acceptance rate for that parameter comes within some tolerance (we chose 5%) of the target rate (we chose 0.25). We then ran 30 chains of 10⁴ steps, each one starting at different points in parameter space drawn at random by adding Gaussian noise, with σ equal to the step size, to the best-fit parameters from the MPFIT results. We computed the Gelman–Rubin statistic for our set of 30 chains, which Ford (2005) suggests should be <1.2 to ensure that the results are converged and well mixed. The Gelman–Rubin statistic was always <1.03 for all parameters, and typically <1.01. Finally, we discarded the first 10% of each chain as the "burn in" portion, using only the latter 90% for deriving the probability distributions of parameters.

At this stage we investigated the impact of DCR on our resulting parallaxes for targets observed in the J band. We assumed an effective wavelength of 1.2462 μm for the background star reference frame, based on the typical values for GKM stars as discussed earlier, and computed individual DCR offsets for the measured positions at each epoch using the method described in the introduction to this section (also see Figure 1). We added these offsets to the measured astrometry and performed our MCMC analysis a second time. We found that the change in the resulting parallax was almost always ⩽0.15σ. As a source of systematic error this is completely negligible as it would boost the final error by ⩽1% when added in quadrature. In a few special cases that are most sensitive to DCR shifts (i.e., three T7–T8 dwarfs with fewer than 10 epochs) the change in parallax was as large as 0.2σ–0.4σ. This would give a slightly larger boost of 2%–7% to their errors, but this is still negligible. In examining the ensemble of the 33 J-band targets for which we computed DCR parallax offsets, we found a mean±rms offset of −0.10 ± 0.19 mas (−0.06 ± 0.10 mas when excluding objects with parallax errors >2 mas), indicating that there is also no systematic offset in our parallaxes due to DCR.

We also performed tests on our data to determine when to consider a parallax measurement "done." Even though our MCMC analysis fully captures any uncertainty due to the degeneracy between proper motion and parallax over data sets spanning modest time baselines (≲ 2 years), we wanted to confirm that the parallaxes we present here will not change substantially with the addition of future data. To check this, for each object we determined the best-fit parallax using subsets of the data starting with the first three epochs (the minimum needed to constrain the five-parameter fit) and then adding one data point at a time for each successive epoch. As expected, the most important criterion for reaching a stable parallax solution was the time baseline. For all of our targets we found that a time baseline of ≈1.2 years was sufficient to reach a best-fit parallax value that remained stable with the addition of new data up to the last observation epoch (our longest time baseline to date is 4.3 years). Therefore, all of the parallaxes presented here are expected to have reached a stable, final value (median baseline of 2.4 years, minimum 1.8 years). We note that this minimum needed time baseline of 1.2 years will necessarily be longer for cases where the astrometric errors are significantly larger than ours or when the target parallax is smaller.

The results from our MCMC analysis are given in Table 3, and the astrometric data are shown in Figures 7–8. The minimum χ² value for each chain is commensurate with the degrees of freedom, which verifies that our adopted positional errors are accurate. There are three exceptions, the known binaries 2MASS J0518−2828AB (Burgasser et al. 2006c) and 2MASS J1404−3159AB (Looper et al. 2008) and the candidate unresolved binary SDSS J0805+4812 (Burgasser 2007b). Their large χ²/dof values can be attributed to the large perturbations present in the residuals after fitting for parallax and proper motion due to orbital motion.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

2.4.1. Correction for Orbital Motion

For the binaries in our sample which have orbit determinations, we apply a correction to the best-fit parallax and proper motion parameters to account for photocenter shifts due to orbital motion. All binaries in our sample only have relative astrometric orbit determinations, so the center of mass is not known. Thus, we use the relative orbital offsets and an assumed mass ratio (derived from evolutionary models) to compute the motion of the center of mass. This motion is then modified by a factor that depends on the binary's flux ratio in the bandpass used in our CFHT imaging in order to determine the actual motion of the photocenter. The coefficient by which to multiply the relative orbital motion is thus (ξ − γ), where

$$\begin{equation} \xi \equiv l / (1+l), \end{equation} \tag{ 3 }$$

$$\begin{equation} \gamma \equiv q / (1+q), \end{equation} \tag{ 4 }$$

l is the flux ratio (≡ f₂/f₁ = 10^−0.4Δm), and q is the mass ratio (≡ M₂/M₁). The coefficient (ξ − γ) is typically negative because flux is a steeper-than-linear function of mass, and thus the photocenter motion is opposite in sign from that of the orbit of the secondary relative to the primary.

We compute the astrometric offsets in a Monte Carlo fashion, using the Markov chain from the published orbit determination to draw the binary's orbital elements (e.g., see Liu et al. 2008). We also draw random: (1) flux ratios corresponding to the error in the measured J-band resolved photometry of the target binaries (or K-band photometry for targets using the CFHT K_H2 filter); and (2) mass ratios from evolutionary models are derived using the method described in Section 5.4 of Dupuy et al. (2009b). Our Monte Carlo approach enables us to appropriately track the correlation between the different parameters (e.g., the derived mass ratio depends on the K-band flux ratio and also the system mass via the orbital elements). For each Monte Carlo trial we subtracted the orbital motion offsets from our CFHT astrometry and recomputed the best-fit parallax and proper motion. This enabled us to derive systematic offsets and corresponding errors due to the uncertainties in the various input parameters.

Our corrections to the parallax and proper motion are given in Table 4 along with the predicted semimajor axis of the photocenter motion for each binary (a_phot). We quote a_phot as positive for photocenter motion has the same sign as the primary motion. For each target we added the randomly drawn orbital motion offsets to our MCMC chains in a Monte Carlo fashion. We note that the minimum χ² of the parallax fit either improved or did not change significantly for all binaries. The latter cases correspond to orbital motion that is nearly linear (e.g., for very long period binaries) and thus easily compensated for by a slightly different proper motion than the precorrected fit. The parallax offsets are quite small (always <0.5σ_π, median 0.08σ_π), which is not surprising since the orbital periods of these binaries are very long (≈10–20 years). The proper motion offsets are much more significant, since the orbital motion over 2–4 years of monitoring can largely be expressed as a linear term. (We note that even though these proper motions are corrected for orbital motion, they are still not "absolute" since the bulk proper motion of the background stars that define the astrometric reference frame is not known.)

Table 4. Orbital Motion Corrections to Parallax and Proper Motion

Target	aphot	q	Δm	Δμαcos δ	Δμδ	Δμ	ΔP.A.	Δπ	Δχ2	Orbit
	(mas)	(M2/M1)	(mag)	('' yr−1)	('' yr−1)	('' yr−1)	(deg)	('')		Ref.
LP 349-25AB	5.0 ± 1.7	0.86 ± 0.04	0.307 ± 0.008	0.0018(6)	0.0030(10)	0.0005(2)	−0.46(16)	−0.00040(13)	0.0	3
LP 415-20AB	10.0 ± 1.1	0.80 ± 0.03	0.728 ± 0.023	0.0026(3)	0.0000(1)	0.0025(3)	−0.32(5)	−0.00006(7)	−1.3	5
LHS 1901AB	7.4 ± 1.0	1.00 ± 0.00	0.113 ± 0.016	−0.0006(1)	−0.0032(5)	0.0024(4)	0.19(3)	0.00016(5)	0.0	3
2MASS J0746 + 2000AB	14.1 ± 1.9	0.92 ± 0.02	0.356 ± 0.024	0.0025(3)	−0.0018(2)	−0.0022(3)	−0.33(4)	−0.00008(2)	0.0	6
2MASS J0920 + 3517AB	3.3 ± 1.1	0.98 ± 0.02	0.25 ± 0.07	0.0017(6)	0.0007(3)	−0.0017(6)	−0.15(5)	−0.00044(18)	−11.8	4
LHS 2397aAB	78 ± 7	0.77 ± 0.08	2.80 ± 0.03	0.0270(24)	−0.0104(12)	−0.0257(22)	−1.51(16)	−0.00081(16)	−19.4	2
2MASS J1534−2952AB	6.3 ± 2.0	0.95 ± 0.03	0.162 ± 0.014	−0.0003(2)	−0.0014(6)	0.0012(5)	0.15(7)	−0.00001(3)	−0.2	7
2MASS J2132 + 1341AB	11.8 ± 2.0	0.81 ± 0.07	0.85 ± 0.04	−0.0067(12)	−0.0008(6)	−0.0005(5)	3.11(57)	0.00016(9)	−2.2	4
2MASS J2206−2047AB	2.6 ± 1.6	0.99 ± 0.03	0.067 ± 0.010	−0.0004(3)	−0.0000(1)	−0.0001(1)	0.61(54)	0.00000(14)	0.0	1
DENIS-P J2252−1730AB	9 ± 4	0.55 ± 0.04	0.94 ± 0.07	0.0001(8)	0.0010(37)	0.0005(16)	−0.10(47)	0.00001(15)	−4.1	4

Notes. Offsets to parallax and proper motion parameters due to orbital motion during our astrometric monitoring program. This is computed from the relative orbit parameters (reference given in the last column), our evolutionary model-derived mass ratio estimate (q), and the flux ratio in the observed bandpass (Δm). The semimajor axis of the resulting photocenter motion (a_phot) is shown for each binary. The difference in χ² between the original best-fit astrometric solution and orbit-corrected solution is also given (Δχ²). These offsets and their errors have already been accounted for in values given in Table 3. References. (1) Dupuy et al. 2009a; (2) Dupuy et al. 2009c; (3) Dupuy et al. 2010; (4) Dupuy 2010; (5) Dupuy & Liu 2011; (6) Konopacky et al. 2010; (7) Liu et al. 2008.

Download table as:  ASCIITypeset image

The three binaries showing large perturbations due to orbital motion (2MASS J0518−2828AB, SDSS J0805+4812AB, and 2MASS J1404−3159AB) unfortunately do not have orbit determinations, and thus we are unable to correct their astrometry as described in this section. Additional astrometric monitoring is needed before the orbits of these binaries can be determined from CFHT data alone.

2.4.2. Correction from Relative to Absolute Parallax

While our primary goal is to measure new parallaxes with CFHT, we also monitored several objects with published parallaxes to validate our methods. In this subsection we provide a detailed comparison of our parallaxes to published values in order to determine if there are any unaccounted for sources of systematic error in our data. As shown below, we determine that our parallaxes generally agree very well with published values, with a few exceptions, and thus readers primarily interested in science results may wish to skip this subsection.

Our "control" sample of ultracool dwarfs included (1) single objects (e.g., 2MASS J2224−0158), (2) wide but unresolved binaries that will have negligible orbital motion over our observations (e.g., 2MASS J1146+2230AB), and (3) binaries that are in our Keck AO dynamical mass sample for which we can independently check and/or improve the published parallax measurements (e.g., SDSS J0423−0414AB). In Figure 12 and Table 7 we show our absolute parallaxes compared to published measurements. (Note that we do not compare our proper motion measurements to published values because all such measurements are relative, not absolute, and we have no way to ascertain the absolute proper motion of the reference frame for published results that generally will be many times larger than the relative proper motion uncertainty.) Our parallaxes are consistent within <1.8σ in 23 of 27 cases (i.e., 85% of the time) and this subset of comparisons has a reasonable χ² (19.9 for 23 dof). The published values largely come from the USNO CCD (eight objects; Monet et al. 1992; Dahn et al. 2002) and IR (eight objects; Vrba et al. 2004) programs. The parallax values for this subsample range from 25.7 ± 0.9 mas (van Leeuwen 2007) to 174.3 ± 2.8 mas (Vrba et al. 2004). In Table 7, we also show how well previously published parallaxes have agreed with each other. We note that there are several instances of published values that are discrepant with each other at the ⩾2σ level (9 of the 31 cases listed), whereas only 1–2 would be expected for Gaussian errors. This implies that some of the parallax errors for the published sample are underestimated. We now consider the four objects for which our parallax is discrepant with the published value at >1.8σ.

• 1.  
  SDSS J0423−0414AB disagrees by 3.1σ with the parallax of Vrba et al. (2004). These authors emphasize the preliminary nature of all their parallaxes (see their Section 6) and present evidence that their errors may be somewhat underestimated. There are seven objects in common between their IR program and the USNO CCD program. The two sets of measurements are only consistent to within 0.5σ–2.7σ, with an unreasonably large χ² of 18.8 (7 dof). To achieve the median expected value of χ² = 6.3 would require multiplying their errors by a factor of 2.0. (Alternatively, the parallax errors from both programs may be underestimated by a factor of 1.72.) If we multiply the published parallax error of SDSS J0423−0414AB by 2.0, the discrepancy between our two measurements is much more modest (1.7σ).
• 2.  
  2MASS J0700+3157AB and 2MASS J1534−2952AB are 2.0σ and 6.3σ discrepant with the measurements of Thorstensen & Kirkpatrick (2003) and Tinney et al. (2003), respectively. We find that both published errors may be underestimated based on Monte Carlo simulations of the published data using an appropriate astrometric precision per epoch. Using the actual measurement epochs and precision per epoch of the published 2MASS J0700+3157AB data (J. Thorstenen 2010, private communication), we find an uncertainty in the parallax of 3.8 mas that is ≈2 times larger than the published error. Adopting this error would result in better agreement with our measurement (1.2σ difference). In the case of 2MASS J1534−2952AB, we retrieved the epochs of the observations from the ESO archive and assumed a range of astrometric precision based on the values given in Tinney et al. (2003), namely (7 mas to 20 mas)/$\sqrt{8}$ added in quadrature to the DCR offset error of 2–6 mas. This resulted in a parallax uncertainty of 2.7–3.7 mas, which is 2.3–3.1 times larger than the published error. At this level, the discrepancy with our parallax measurement is significantly decreased, though it still disagrees at the 2.9σ level. We also checked if orbital motion was significant and found that the correction offset for the parallax was negligible for the Tinney et al. (2003) epochs as it is for ours. We note that Tinney et al. (2003) used only eight reference stars (cf. our 475) and data spanning six distinct epochs over 2.0 years (cf. our 16 over 2.4 years), so our solution should be more robust.
• 3.  
  For LP 349-25AB, our parallax is 3.4σ different from the published value (75.8 ± 1.6 mas; Gatewood & Coban 2009). This is the one case that we cannot readily explain with information at hand. Gatewood & Coban (2009) do not discuss their astrometric precision per epoch, so we cannot assess their quoted error with Monte Carlo simulations. One source of systematic error could be their relatively small number of reference stars (12 versus our 33). Another effect could be DCR as Gatewood & Coban (2009) seem to have observed in a broadband optical filter (not described in their paper). For this object we used the narrowband K_H2 filter, so DCR will be completely negligible. We note that our value for the correction from relative to absolute parallax (1.7 ± 0.3 mas) agrees very well with theirs (1.6 ± 0.8 mas), so this cannot be the source of the discrepancy. Our orbit correction is very small (−0.40 ± 0.13 mas) and Gatewood & Coban (2009) see no significant perturbations due to orbital motion, so this is also unlikely to explain the discrepancy.
• 4.  
  There is one other published parallax that is discrepant with our results, 2MASS J0850+1057AB (2.5σ different than Dahn et al. 2002). This object has already been discussed by Faherty et al. (2011). They found that the Dahn et al. (2002) parallax was likely biased by a background star that was blended with the target at the time of those observations but which is now clearly separated from the science target at ≈4'' in both our data and those of Faherty et al. (2011). Our parallax of 30.1 ± 0.8 mas for 2MASS J0850+1057AB is in good agreement with both the values of 35 ± 8 mas and 26.2 ± 4.2 mas determined by Faherty et al. (2011) and Vrba et al. (2004), respectively, but with 5–10 times smaller error bars.

Zoom In Zoom Out Reset image size

Finally, we note that on average our absolute parallax measurements are not different from published results in any systematic fashion. The mean and rms of the differences between published parallax values and our own is 1.5 ± 3.4 mas (excluding the four discrepant cases discussed above and published values with >20 mas parallax errors). In fact, by using a much larger number of reference stars than previous parallax programs, we should be less sensitive to systematic errors in our correction to absolute parallax. Previous surveys typically used ≈5–10 reference stars, whereas on average we have >100 reference stars per field (N_ref = 20–475) and so are much less biased by outliers. This is further supported by the fact that the one target with a Hipparcos parallax for its stellar companion (HD 225118; 25.7 ± 0.9 mas) is in excellent agreement with our CFHT value for the M7.5 dwarf (2MASS J0003−2822; 25.0 ± 1.9 mas).

The vast majority of our ultracool binary targets do no have any published resolved spectroscopy providing component spectral type determinations. Some binaries have spectral types in the literature determined by a spectral decomposition, i.e., using integrated-light spectra and resolved photometry to estimate the deblended component spectra. However, methods used in the literature have varied substantially (e.g., Liu et al. 2006; Reid et al. 2006b; Siegler et al. 2007) and often rely on an input assumption for the relationship between spectral type absolute magnitude (e.g., Burgasser 2007b; Burgasser et al. 2010). This is problematic because the binary components cannot then truly be used to assess such empirical relations, which is a major goal of our work. Therefore, we have determined component types for all binaries with parallaxes in a uniform fashion that is completely independent of any assumptions about how magnitude should depend on spectral type.

Our approach is modeled somewhat after the spectral template matching method of Burgasser et al. (2010), but we have removed assumption for the relationship between spectral type absolute magnitude. A library of template spectra for single⁹ objects is compiled, and all possible pairs of these spectra are added together. Each pairing is allowed to have an arbitrary component flux ratio. The flux ratio and overall scale factor are adjusted to minimize the χ² of the difference with the target's measured integrated-light spectrum. For each pairing we then compute synthetic photometry in the bands for which we have measured flux ratios. We reject pairings that disagree significantly with the measured resolved photometry (p-value <0.05 in a χ² test). Thus, our final set of modeled binary spectra is purely selected on how well they match the measured integrated-light spectrum and resolved photometry. We then ranked this ensemble based on the χ² of the match to the integrated-light spectrum and computed weighted averages and errors of the component types and synthesized flux ratios using the method outlined in Section 4.3 of Burgasser et al. (2010). When assessing component types, we take these quantities and their nominal errors into consideration but do not treat them as absolute truth.

The input library of template spectra we used necessarily varied with the component spectral types. For binaries composed wholly of ⩾L3 dwarfs, we used the same library of 178 IRTF/SpeX prism spectra as Burgasser et al. (2010). Although this library is somewhat less numerous than the full set of spectra in the SpeX Prism Library, it has the significant advantage that Burgasser et al. (2010) report infrared spectral types on a consistent scheme for all templates. This is in contrast to types available in the literature, particularly for L dwarfs, which are based on a variety of infrared flux indices and sometimes only have optical types. Because this library only has a handful of early-L dwarf templates and no late-M dwarfs we had to use a different subset of spectra for earlier type binaries. For binaries with at least one <L3 component we simply used the full SpeX Prism Library with whatever spectral types were available in the literature (i.e., a mix of optical and infrared types). For uniformity, we resampled all spectra to a wavelength grid with 0.004 μm steps ranging from 0.78 to 2.40 μm. To reduce systematic errors due to inaccurate correction of telluric absorption, we excluded two wavelength regions (1.34 μm < λ < 1.41 μm and 1.81 μm < λ < 1.94 μm) when performing the spectral matching. In some cases, we had to use measured integrated-light spectra obtained with SpeX in SXD mode (R = 1200–2000), which we degraded to the standard SpeX prism resolution of 120 for accurate comparison to library templates. For such SXD data we exclude the K-band portion of the spectrum since that order does not overlap with the JH orders and thus its relative normalization would need to account for the uncertainty in the measured and synthesized integrated-light photometry in the K band.

Throughout our analysis, we conservatively assume that infrared types of late-M and -L dwarfs are uncertain by at least 1 subtype, with some templates having larger uncertainties of 1.5–2 subtypes, and that T dwarf types are uncertain by at least 0.5 subtype. This is based on the analysis of infrared types done by Burgasser et al. (2010) who compared their types to published values for 189 spectra of 178 sources. These authors found an intrinsic rms scatter of 1.1 and 0.5 subtypes in the ensemble of L and T dwarfs, respectively.

We assigned component types and uncertainties on a case by case basis, taking into account various factors such as larger than average spectral type uncertainties in the best-match templates; the full range of properties implied when there were multiple matches giving equally good fits; and constraints imposed by requiring consistency with the integrated-light type. When flux ratios were available from multiple sources (e.g., our MKO Keck photometry and HST/NICMOS medium-band data), we checked for consistency. We sometimes noted discrepancies with photometry from the literature when published errors were rather small. In these cases we excluded the published values as their errors are likely underestimated, and it did not change the derived spectral types significantly within the errors.

Our derived component spectral types and their corresponding uncertainties are listed in Table 8, and the single best template pairing for each binary is shown in Figures 13–15. In Table 8, we give references for the literature photometry used and also a list of the bandpasses utilized in constraining each fit. We also list separately those binaries for which we do not use component types from our spectral template matching because their types have been determined directly from resolved spectroscopy (e.g., LHS 1070BC; Leinert et al. 2000) or other analysis (e.g., CFBDSIR J1458+1013AB; Liu et al. 2011b).

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Table 8. Component Spectral Types of Ultracool Binaries with Parallaxes

Object	Primary	Secondary	Broadband	HST/NICMOS	Phot.
	Type	Type	Data	Data	Ref.
Derived from Our Template-matching Method
GJ 1001BC	L5 ± 0.5	L5 ± 0.5	JHK	110W, 170M	1
LP 349-25AB	M6.5 ± 1	M8 ± 1	JHK	...	5
SDSSp J0423−0414AB	L6.5 ± 1	T2 ± 0.5	K	110W, 170M	1, 2
2MASS J0518−2828AB	L6 ± 1	T4 ± 0.5	...	110W, 170M	1
2MASS J0700+3157AB	L3 ± 1	L6.5 ± 1.5	JHK	...	1
LHS 1901AB	M7 ± 1	M7 ± 1	JHK	...	5
SDSS J0805+4812AB	L4 ± 1	T5 ± 0.5	...	...	...
2MASS J0850+1057AB	L6.5 ± 1	L8.5 ± 1	JHK	110W, 145M, 170M	1, 3
Gl 337CD	L8.5 ± 1	L7.5 ± 2	JHK	...	1
2MASS J0920+3517AB	L5.5 ± 1	L9 ± 1.5	JHK	...	1
SDSS J0926+5847AB	T3.5 ± 1	T5 ± 1	...	110W, 170M	1
2MASS J1017+1308AB	L1.5 ± 1	L3 ± 1	K	...	1
SDSS J1021−0304AB	T0 ± 1	T5 ± 0.5	JHK	110W, 170M	1, 2
Gl 417BC	L4.5 ± 1	L6 ± 1	K	...	1
2MASS J1146+2230AB	L3 ± 1	L3 ± 1	...	...	...
2MASS J1209−1004AB	T2.5 ± 0.5	T6.5 ± 1	JHKCH4s	...	10
2MASS J1225−2739AB	T5.5 ± 0.5	T8 ± 0.5	JHKCH4s	...	1
DENIS-P J1228−1547AB	L5.5 ± 1	L5.5 ± 1	K	...	1
Kelu-1AB	L2 ± 1	L4 ± 1	JHK	...	7
2MASS J1404−3159AB	L9 ± 1	T5 ± 0.5	JHK	...	1
SDSS J1534+1615AB	T0 ± 1	T5.5 ± 0.5	JHK	...	8
2MASS J1534−2952AB	T4.5 ± 0.5	T5 ± 0.5	JHKCH4s	...	9
2MASS J1553+1532AB	T6.5 ± 0.5	T7.5 ± 0.5	JHKCH4s	110W, 170M	1, 2
2MASS J1728+3948AB	L5 ± 1	L7 ± 1	JHK	110W, 145M, 170M	1, 3
LSPM J1735+2634AB	M7.5 ± 0.5	L0 ± 1	JHK	...	1
2MASS J1750+4424AB	M6.5 ± 1	M8.5 ± 1	JHK	...	6
2MASS J1847+5522AB	M6 ± 0.5	M7 ± 0.5	JHK	...	6
SDSS J2052−1609AB	L8.5 ± 1.5	T1.5 ± 0.5	JHK	110W, 170M	1, 11
2MASS J2101+1756AB	L7 ± 1	L8 ± 1	K	...	6
2MASS J2132+1341AB	L4.5 ± 1.5	L8.5 ± 1.5	JHK	...	1
2MASS J2140+1625AB	M8 ± 0.5	M9.5 ± 0.5	JHK	...	6
2MASS J2206−2047AB	M8 ± 0.5	M8 ± 0.5	JHK	...	4
DENIS-P J2252−1730AB	L4.5 ± 1.5	T3.5 ± 1	K	110W, 170M	1
Published Values from Resolved Spectroscopy or Indices
LHS 1070BC	M8.5 ± 0.5	M9.5 ± 0.5	...	...	A
2MASS J0746+2000AB	L0 ± 0.5	L1.5 ± 0.5	...	...	B
HD 130948BC	L4 ± 1	L4 ± 1	...	...	C
Gl 569Bab	M8.5 ± 0.5	M9.0 ± 0.5	...	...	D
CFBDS J1458+1013AB	T9 ± 0.5	>T10	...	...	E
SCR J1845−6357AB	M8.5 ± 0.5	T6 ± 0.5	...	...	F
Ind Bab	T1 ± 0.5	T6 ± 0.5	...	...	G
2MASS J2234+4041AB	M6 ± 1	M6 ± 1	...	...	H

Notes. We list component spectral types derived using the template-matching method described in Section 5.2 supplemented by spectral type determinations from the literature based on resolved spectroscopy or resolved photometric indices. The fourth and fifth columns list the MKO broadband and HST/NICMOS medium-band flux ratios used in the template matching, respectively. The last column gives references for the photometry used or for the source of the resolved spectroscopy. References. (1) This work (Tables 5 and 6); (2) Burgasser et al. 2006c; (3) Burgasser et al. 2010; (4) Dupuy et al. 2009a; (5) Dupuy et al. 2010; (6) Konopacky et al. 2010; (7) Liu & Leggett 2005; (8) Liu et al. 2006; (9) Liu et al. 2008; (10) Liu et al. 2010; (11) Stumpf et al. 2011. (A) Leinert et al. 2000; (B) Bouy et al. 2004; (C) Goto et al. 2002; (D) Lane et al. 2001; (E) Liu et al. 2011b; (F) Kasper et al. 2007; (G) King et al. 2010; (H) Allers et al. 2009.

Download table as:  ASCIITypeset image

Finally, we note that there are several binaries with parallaxes for which we cannot derive component spectral types using this method, either because they do not have the needed spectral or photometric data or the data available do not sufficiently constrain the component types. Binaries with parallaxes for which we do not have spectra are 2MASS J0025+4759AB, HD 65216BC, LSPM J1314+1320AB, DENIS-P J1441−0945AB, and 2MASS J2331−0406AB. 2MASS J0856+2235AB, 2MASS J0952−1924AB, 2MASS J1239+5515AB, and LSR J1610−0040AB have no near-IR photometry. LHS 2397aAB has a spectrum and near-IR photometry, but the late-L companion is too faint to enable accurate spectral decomposition.

Our spectral decomposition analysis produces estimates of the flux ratios for bandpasses without resolved photometry. We synthesize flux ratios for every template pairing that agrees with the available resolved photometry (p value  <  0.05). To determine the flux ratio and corresponding uncertainty in a given bandpass, we then use the weighted average and error given by the aforementioned Burgasser et al. (2010) weighting scheme. There are several cases that benefit from these flux ratios, in order of most to least reliable: (1) binaries with resolved K_S photometry that we convert to MKO K (and vice versa, e.g., for ΔJ_MKO to ΔJ_2MASS), (2) binaries with HST/NICMOS 0.9–1.8 μm photometry that we convert to JH photometry, (3) binaries with, e.g., only a K-band ratio for which we determine J and H flux ratios; and (4) one binary without any flux ratios (SDSS 0805+4812AB). In the following analysis we account for these differing levels of reliability when reporting absolute magnitudes and examining the location of binary components on CMDs.

We combine our parallaxes with integrated-light photometry (and flux ratios in the case of binaries) to compute the absolute magnitudes for our sample. We have also compiled such measurements for all ultracool objects with published parallaxes. Table 9 presents a list of all ultracool dwarf parallax measurements, including objects that have parallax determinations by virtue of their companionship to stellar primaries. We compiled photometry from the literature for this entire sample, and for objects that have published photometry in only one system, (MKO or 2MASS) we use near-IR spectra, when available, to synthesize photometric conversions. For such objects we also synthesize Y − J colors to provide Y-band photometry when it is not measured directly. Using objects in the SpeX Prism Library with both 2MASS and MKO photometry we can test the quality of our synthetic photometric system offsets. The χ² of our computed 2MASS/MKO offsets compared to measured values was 52.0 (98 dof), 96.6 (95 dof), and 51.4 (93 dof) for J, H, and K bands, respectively. Since χ² is reasonable in all cases, we find that any systematic error in our computed offsets must be negligible compared to the uncertainty in the measured photometry (≲0.02 mag). However, when computing colors across different bandpasses we found an additional error of 0.05 mag was needed to explain the scatter in observed minus computed values. Thus, we treat synthesized photometric system offsets (e.g., J_MKO − J_2MASS) has having zero error, while we add 0.05 mag in quadrature to all synthesized Y − J photometry.

Table 9. All Ultracool Dwarfs with Parallaxes

Object	αJ2000	δJ2000	Epoch	πabs	μαcos δ	μδ	μ	P.A.	Vtan	Ref.	Note
	(deg)	(deg)	(MJD)	('')	('' yr−1)	('' yr−1)	('' yr−1)	(deg)	(km s−1)
SDSS J000013.54+255418.6	000.0564	+25.9055	54301.63	0.0708(19)	−0.0191(14)	0.1267(13)	0.1281(13)	351.4 ± 0.6	8.58 ± 0.24	1
LSR J0011+5908	002.8826	+59.1445	51492.22	0.1083(14)	−0.8997	−1.1654	1.4723	218	64	18
BRI 0021−0214	006.1027	−01.9723	51071.24	0.0866(40)	−0.0804(38)	0.1330(60)	0.1550(70)	328.8 ± 0.7	8.5 ± 0.5	26
LHS 1070A	006.1841	−27.1401	51542.05	0.1295(25)	−0.1330(50)	0.6401(31)	0.6537(30)	348.3 ± 0.4	23.9 ± 0.5	6
PC 0025+0447	006.9249	+05.0616	51768.41	0.0138(16)	0.0105(4)	−0.0008(3)	0.0105(4)	94.6 ± 1.8	3.6 ± 0.5	8
LP 349-25AB	006.9842	+22.3255	54687.57	0.0696(9)	0.4039(10)	−0.1654(15)	0.4365(9)	112.27 ± 0.21	29.7 ± 0.4	1	Young?
2MASSW J0030300−145033	007.6256	−14.8426	51840.16	0.0374(45)	0.2450(35)	−0.0282(18)	0.2466(36)	96.6 ± 0.4	31 ± 4	31
SDSSp J003259.36+141036.6	008.2474	+14.1770	51878.20	0.0300(50)	0.2730(70)	0.0391(35)	0.2760(70)	81.8 ± 0.7	43 ± 8	31
ULAS J003402.77−005206.7	008.5116	−00.8687	55051.60	0.0687(14)	−0.0167(10)	−0.3588(8)	0.3592(8)	182.66 ± 0.16	24.8 ± 0.5	1
2MASSW J0036159+182110	009.0674	+18.3529	51872.12	0.1142(8)	0.8991(6)	0.1200(16)	0.9071(6)	82.40 ± 0.10	37.66 ± 0.27	8
2MASS J00501994−3322402	012.5873	−33.3749	55050.57	0.0946(24)	1.1506(21)	0.9391(21)	1.4851(21)	50.78 ± 0.08	74.4 ± 1.9	1
RG 0050−2722	013.2279	−27.0999	51128.04	0.0460(100)	0.0562(48)	0.0901(45)	0.1063(45)	32.0 ± 2.6	10.9 ± 2.5	27
CFBDS J005910.90−011401.3	014.7961	−01.2336	55068.57	0.1032(21)	0.8847(11)	0.0440(12)	0.8858(11)	87.15 ± 0.08	40.7 ± 0.8	1
SDSSp J010752.33+004156.1	016.9684	+00.6990	51789.23	0.0641(45)	0.6280(70)	0.0914(36)	0.6350(70)	81.7 ± 0.3	47 ± 3	31
CTI 012657.5+280202	021.9132	+28.0982	50753.20	0.0305(5)	−0.1334(3)	−0.1348(3)	0.1896(2)	224.70 ± 0.10	29.5 ± 0.5	8
L 726-8AB	024.7550	−17.9507	51026.33	0.3750(40)	3.2771(6)	0.5908(6)	3.3299(6)	79.78 ± 0.01	42.1 ± 0.4	12
2MASS J01490895+2956131	027.2873	+29.9370	50755.30	0.0444(7)	0.1757(8)	−0.4021(7)	0.4388(7)	156.40 ± 0.10	46.9 ± 0.7	8
SDSS J015141.69+124429.6	027.9232	+12.7417	50704.37	0.0467(34)	0.7418(42)	−0.0368(21)	0.7427(42)	92.84 ± 0.16	76 ± 5	31
DENIS-P J020529.0−115925AB	031.3725	−11.9916	51869.20	0.0506(15)	0.4344(8)	0.0549(8)	0.4378(8)	82.80 ± 0.10	41.0 ± 1.2	8	Triple
SDSS J020742.48+000056.2	031.9285	+00.0157	51774.32	0.0293(40)	0.1588(31)	−0.0143(39)	0.1595(30)	95.1 ± 1.4	26 ± 4	20
2MASSI J0243137−245329	040.8072	−24.8916	51129.17	0.0936(36)	−0.2878(35)	−0.2076(29)	0.3548(41)	234.2 ± 0.3	18.0 ± 0.7	31
BRI B0246−1703	042.1708	−16.8560	51026.38	0.0620(50)	0.0210(90)	−0.2730(120)	0.2740(120)	175.7 ± 1.9	21.1 ± 2.1	27
TVLM 831-154910	042.5486	−01.8582	51116.18	0.0302(45)	0.0660(50)	−0.0559(44)	0.0870(60)	130.2 ± 1.2	13.6 ± 2.3	26
TVLM 831-161058	042.8053	+00.7934	51788.42	0.0177(22)	0.2340(70)	0.0399(43)	0.2380(70)	80.3 ± 1.0	64 ± 8	26
TVLM 831-165166	042.9277	−01.0350	51084.21	0.0195(39)	0.4010(110)	0.1970(70)	0.4470(110)	63.8 ± 0.7	109 ± 23	26
TVLM 832-10443	043.1095	+00.9395	51789.29	0.0360(4)	−0.1752(2)	−0.1032(3)	0.2033(1)	239.50 ± 0.10	26.8 ± 0.3	8
Teegarden's star	043.2535	+16.8815	51486.22	0.2592(9)	3.4228	−3.8081	5.1203	138	94	11
PSO J043.5395+02.3995	043.5401	+02.3997	55584.29	0.1710(450)	2.5490(110)	0.2310(110)	2.5590(110)	84.82 ± 0.25	71 ± 20	19
DENIS-P J0255.0−4700	043.7649	−47.0142	51153.08	0.2014(39)	0.9996(27)	−0.5655(37)	1.1485(22)	119.50 ± 0.20	27.0 ± 0.5	7
TVLM 832-42500	045.6455	−01.2737	51084.34	0.0363(40)	0.6720(190)	0.3630(140)	0.7640(210)	61.6 ± 0.9	100 ± 12	26
LP 412-31	050.2486	+18.9065	50747.41	0.0689(6)	0.3493(5)	−0.2557(6)	0.4329(3)	126.20 ± 0.10	29.78 ± 0.26	8
2MASSW J0326137+295015	051.5570	+29.8376	50771.36	0.0310(15)	−0.0188(8)	0.0668(8)	0.0694(8)	344.3 ± 0.7	10.6 ± 0.5	8
2MASSI J0328426+230205	052.1777	+23.0348	50748.33	0.0331(42)	0.0126(26)	−0.0597(49)	0.0610(49)	168.1 ± 2.3	8.7 ± 1.4	31
LSPM J0330+5413	052.7038	+54.2320	51172.21	0.1038(14)	−0.1510	−0.0050	0.1511	268	7	18
LP 944-20	054.8967	−35.4289	51154.22	0.2014(42)	0.3240(80)	0.2960(70)	0.4390(80)	47.6 ± 0.9	10.32 ± 0.29	27
2MASP J0345432+254023	056.4299	+25.6732	51530.21	0.0371(5)	−0.0960(3)	−0.0357(4)	0.1024(3)	249.60 ± 0.20	13.08 ± 0.18	8
LHS 1604	057.7502	−00.8792	51828.37	0.0681(19)	0.0086(11)	−0.4724(10)	0.4725(10)	178.96 ± 0.13	32.9 ± 0.9	21	Overlum.
2MASSI J0415195−093506	063.8381	−09.5835	55070.64	0.1752(17)	2.2143(12)	0.5361(12)	2.2782(12)	76.39 ± 0.03	61.6 ± 0.6	1
SDSSp J042348.57−041403.5AB	065.9517	−04.2339	54341.64	0.0721(11)	−0.3276(5)	0.0912(5)	0.3401(5)	285.56 ± 0.09	22.4 ± 0.3	1
LHS 191	066.5830	+03.6100	51569.05	0.0584(18)	−0.1182(16)	−1.0154(17)	1.0223(17)	186.64 ± 0.09	83.0 ± 2.5	21
LHS 197	071.5771	+48.7477	51545.16	0.0523(10)	1.0258(7)	−0.6345(7)	1.2062(7)	121.74 ± 0.03	109.3 ± 2.1	21
LSR J0510+2713	077.5838	+27.2342	50786.31	0.1007(16)	−0.2149	−0.6339	0.6694	199	32	18
LHS 1742a	077.6623	+19.4022	50755.39	0.0134(10)	0.7727(3)	−0.3075(3)	0.8316(3)	111.70 ± 0.02	294 ± 22	21	Subdwarf
LSR J0515+5911	078.8789	+59.1885	51197.13	0.0657(13)	0.1127	−1.0093	1.0156	174	73	18
2MASS J05185995−2828372AB	079.7498	−28.4773	54366.66	0.0437(8)	−0.0700(5)	−0.2756(5)	0.2844(5)	194.25 ± 0.10	30.9 ± 0.6	1
2MASS J05325346+8246465	083.2228	+82.7796	51238.10	0.0423(18)	2.0441(15)	−1.6654(15)	2.6367(16)	129.17 ± 0.03	296 ± 12	23	Subdwarf
LHS 207	084.5527	+79.5219	51822.51	0.0451(14)	0.8412(6)	−0.8581(6)	1.2016(6)	135.57 ± 0.03	126 ± 4	21
SDSSp J053951.99−005902.0	084.9667	−00.9839	51116.34	0.0761(22)	0.1643(22)	0.3159(32)	0.3561(35)	27.49 ± 0.28	22.2 ± 0.7	31
2MASSI J0559191−140448	089.8314	−14.0809	54519.25	0.0966(10)	0.5718(16)	−0.3330(16)	0.6617(16)	120.21 ± 0.14	32.5 ± 0.3	1	Overlum.
2MASS J06411840−4322329	100.3267	−43.3758	51271.04	0.0560(60)	0.2160(90)	0.6130(90)	0.6490(90)	19.4 ± 0.8	55 ± 6	2
2MASS J07003664+3157266AB	105.1533	+31.9562	54513.30	0.0867(12)	0.1424(7)	−0.5546(7)	0.5726(7)	165.60 ± 0.07	31.3 ± 0.4	1
ESO 207-61	106.9720	−49.0140	51600.09	0.0541(45)	−0.0100(60)	0.3910(70)	0.3910(70)	358.6 ± 0.8	34.3 ± 2.9	27
LHS 1901AB	107.7987	+43.4984	54513.31	0.0742(10)	0.3544(9)	−0.5662(9)	0.6680(9)	147.96 ± 0.08	42.7 ± 0.6	1
2MASS J07193188−5051410	109.8828	−50.8614	51615.10	0.0326(24)	0.1981(33)	−0.0614(38)	0.2074(33)	107.2 ± 1.1	30.2 ± 2.3	2
UGPS J072227.51−054031.2	110.6137	−05.6750	55257.50	0.2428(24)	−0.9037(17)	0.3518(14)	0.9698(17)	291.27 ± 0.08	18.94 ± 0.19	17
2MASSI J0727182+171001	111.8297	+17.1646	55125.63	0.1125(9)	1.0471(9)	−0.7641(9)	1.2962(9)	126.12 ± 0.04	54.6 ± 0.4	1
LHS 234	115.0801	−17.4125	50894.11	0.1070(16)	1.1440(19)	−0.5530(27)	1.2706(16)	115.80 ± 0.13	56.3 ± 0.8	6
2MASSI J0746425+200032AB	116.6764	+20.0089	54517.34	0.0811(9)	−0.3659(7)	−0.0527(5)	0.3697(7)	261.81 ± 0.08	21.61 ± 0.24	1
LP 423-31	118.0996	+16.2044	50752.50	0.0544(10)	0.1813	−0.3562	0.3997	153	35	11
SDSS J080531.84+481233.0AB	121.3814	+48.2094	54428.60	0.0431(10)	−0.4583(7)	0.0498(7)	0.4610(7)	276.20 ± 0.09	50.7 ± 1.2	1
DENIS J081730.0−615520	124.3750	−61.9211	51545.24	0.2030(130)	−0.3300(500)	1.0970(430)	1.1470(420)	343.0 ± 2.7	26.8 ± 2.0	3
2MASSI J0825196+211552	126.3320	+21.2645	50822.40	0.0938(10)	−0.5097(16)	−0.2884(19)	0.5856(14)	240.50 ± 0.20	29.6 ± 0.3	8
ULAS J082707.67−020408.2	126.7820	−02.0689	53736.50	0.0260(31)	0.0267(27)	−0.1089(24)	0.1122(23)	166.2 ± 1.4	20.5 ± 2.5	20
LHS 248	127.4562	+26.7763	50846.25	0.2758(30)	−1.1390	−0.6056	1.2900	242	22	29
SDSSp J083008.12+482847.4	127.5344	+48.4801	51192.31	0.0764(34)	−1.0060(60)	−0.7698(48)	1.2670(70)	232.58 ± 0.15	79 ± 4	31
LHS 2021	127.6357	+09.7876	51610.13	0.0598(45)	−0.5002(34)	−0.4487(36)	0.6720(22)	228.1 ± 0.4	53 ± 4	7
LHS 2026	128.1270	−01.5772	51144.29	0.0508(6)	0.1648(3)	−0.4691(3)	0.4972(3)	160.64 ± 0.03	46.4 ± 0.5	21
2MASS J08354256−0819237	128.9274	−08.3233	51201.24	0.1170(110)	−0.5200(90)	0.2850(100)	0.5930(80)	298.8 ± 1.0	23.9 ± 2.3	2
SDSSp J083717.22−000018.3	129.3217	−00.0050	51259.68	0.0340(130)	−0.0150(80)	−0.1720(170)	0.1730(170)	185.1 ± 2.8	24 ± 12	31
LHS 2034	130.1240	+18.4026	51105.50	0.0713(11)	−0.8132(6)	−0.4480(6)	0.9284(6)	241.15 ± 0.04	61.7 ± 1.0	21
LHS 2065	133.4008	−03.4923	51173.27	0.1173(15)	−0.5096(9)	−0.2004(10)	0.5476(9)	248.53 ± 0.10	22.13 ± 0.29	21
2MASSI J0856479+223518AB	134.1992	+22.5884	54428.62	0.0324(10)	−0.1869(10)	−0.0132(8)	0.1874(10)	265.95 ± 0.24	27.4 ± 0.9	1
LP 368-128	135.0983	+21.8348	51123.47	0.1569(27)	−0.5097(37)	−0.5823(34)	0.7739(22)	221.2 ± 0.3	23.4 ± 0.4	14
ULAS J090116.23−030635.0	135.3176	−03.1097	53736.50	0.0626(26)	−0.0386(23)	−0.2612(28)	0.2640(28)	188.4 ± 0.5	20.0 ± 0.9	20
DENIS-P J0909.9−0658	137.4896	−06.9718	51185.36	0.0425(42)	−0.1839(26)	0.0207(30)	0.1851(25)	276.4 ± 0.9	20.6 ± 2.1	2
2MASSW J0920122+351742AB	140.0506	+35.2949	54427.66	0.0344(8)	−0.1889(7)	−0.1985(7)	0.2740(8)	223.59 ± 0.13	37.8 ± 0.9	1
SDSS J092615.38+584720.9AB	141.5642	+58.7889	54513.41	0.0437(11)	0.0102(5)	−0.2163(5)	0.2165(5)	177.30 ± 0.12	23.5 ± 0.6	1
2MASSI J0937347+293142	144.3953	+29.5281	51636.28	0.1634(18)	0.9411(12)	−1.3155(12)	1.6174(12)	144.42 ± 0.04	46.9 ± 0.5	23	Low-Z?
2MASS J09393548−2448279	144.8979	−24.8078	51584.13	0.1873(46)	0.5734(23)	−1.0447(25)	1.1917(25)	151.24 ± 0.11	30.2 ± 0.7	5	Low-Z?
TVLM 262-111511	145.5939	+42.7599	50926.24	0.0340(60)	0.1680(80)	−0.2000(90)	0.2610(120)	140.1 ± 0.5	36 ± 7	26
ULAS J094806.06+064805.0	147.0252	+06.8014	53736.50	0.0272(42)	0.1990(70)	−0.2740(70)	0.3390(70)	143.9 ± 1.1	59 ± 9	20
TVLM 262-70502	147.9487	+42.5641	50931.23	0.0256(40)	0.0969(47)	−0.1760(70)	0.2010(80)	151.2 ± 0.9	37 ± 6	26
2MASS J09522188−1924319AB	148.0912	−19.4089	50931.11	0.0338(30)	−0.0611(35)	−0.1016(28)	0.1186(21)	211.0 ± 1.9	16.6 ± 1.5	7
2MASS J10043929−3335189	151.1637	−33.5886	51300.99	0.0550(60)	0.2435(37)	−0.2533(37)	0.3514(37)	136.1 ± 0.6	30 ± 3	2
TVLM 263-71765	152.7510	+42.7510	51230.33	0.0319(29)	−0.1330(60)	−0.1490(60)	0.2000(70)	221.8 ± 1.4	29.8 ± 2.9	26
SSSPM J1013−1356	153.2806	−13.9390	51214.24	0.0203(20)	0.0691(13)	−1.0249(13)	1.0272(13)	176.14 ± 0.07	240 ± 23	23	Subdwarf
2MASSI J1017075+130839AB	154.2818	+13.1442	54514.44	0.0302(14)	0.0479(5)	−0.1178(5)	0.1272(5)	157.86 ± 0.24	20.0 ± 0.9	1
ULAS J101821.78+072547.1	154.5908	+07.4298	53736.50	0.0250(20)	−0.1837(26)	−0.0151(31)	0.1843(26)	265.3 ± 1.0	34.9 ± 2.9	20
2MASS J10185879−2909535	154.7450	−29.1649	51243.10	0.0353(32)	−0.3401(20)	−0.0939(26)	0.3529(19)	254.6 ± 0.4	47 ± 4	2
SDSS J102109.69−030420.1AB	155.2902	−03.0723	54514.45	0.0299(13)	−0.1626(6)	−0.0745(7)	0.1789(6)	245.38 ± 0.21	28.3 ± 1.3	1
TVLM 213-2005	155.3643	+50.9179	51175.45	0.0301(4)	−0.3856(1)	0.0514(7)	0.3890(1)	277.60 ± 0.10	61.3 ± 0.8	8
2MASSI J1047538+212423	161.9744	+21.4065	50842.42	0.0947(38)	−1.6620(70)	−0.4741(42)	1.7280(80)	254.08 ± 0.13	87 ± 4	31
LHS 292	162.0524	−11.3356	51203.31	0.2203(36)	0.6162	−1.5252	1.6450	158	35	29
LHS 2314	162.2641	+05.0396	51586.21	0.0411(23)	−0.3651(15)	−0.4623(15)	0.5891(14)	218.30 ± 0.15	68 ± 4	21
Wolf 359	164.1203	+07.0147	51604.20	0.4191(21)	−3.8467	−2.6935	4.6960	235	53	29
DENIS-P J1058.7−1548	164.6995	−15.8048	50897.23	0.0577(10)	−0.2529(5)	0.0414(4)	0.2563(5)	279.30 ± 0.10	21.1 ± 0.4	8
SSSPM J1102−3431	165.5410	−34.5099	51264.16	0.0181(5)	−0.0671(6)	−0.0140(6)	0.0686(6)	258.2 ± 0.5	18.0 ± 0.5	25	TWA
LHS 2351	166.5791	+04.4758	51589.22	0.0481(31)	−0.3290(90)	0.3720(90)	0.4970(120)	318.5 ± 0.6	49 ± 3	27
SDSS J111010.01+011613.1	167.5412	+01.2700	54514.50	0.0521(12)	−0.2171(7)	−0.2809(7)	0.3550(7)	217.71 ± 0.11	32.3 ± 0.7	1
2MASS J11145133−2618235	168.7033	−26.3075	55280.39	0.1792(14)	−3.0189(11)	−0.3840(16)	3.0432(11)	262.75 ± 0.03	80.5 ± 0.6	1
LHS 2397aAB	170.4539	−13.2190	54520.49	0.0730(21)	−0.4869(23)	−0.0614(18)	0.4908(23)	262.81 ± 0.21	31.9 ± 0.9	1
2MASSW J1146345+223053AB	176.6441	+22.5152	54514.51	0.0349(10)	0.0256(7)	0.0894(8)	0.0930(8)	16.0 ± 0.4	12.6 ± 0.4	1
ULAS J115038.79+094942.9	177.6616	+09.8286	53736.50	0.0170(80)	−0.1070(160)	−0.0320(80)	0.1120(160)	253 ± 4	32 ± 19	20
LHS 2471	178.4695	+06.9989	51607.32	0.0703(27)	0.2572(19)	−0.8536(18)	0.8915(18)	163.23 ± 0.12	60.2 ± 2.3	21
2MASSW J1207334−393254b	181.8894	−39.5483	51300.18	0.0191(4)	−0.0642(4)	−0.0226(4)	0.0681(4)	250.6 ± 0.3	16.9 ± 0.4	10	TWA, planet
2MASSW J1207334−393254	181.8894	−39.5483	51300.18	0.0191(4)	−0.0642(4)	−0.0226(4)	0.0681(4)	250.6 ± 0.3	16.9 ± 0.4	10	TWA
2MASS J12095613−1004008AB	182.4851	−10.0679	54513.52	0.0458(10)	0.2661(5)	−0.3554(6)	0.4440(6)	143.18 ± 0.06	46.0 ± 1.0	1
2MASSI J1217110−031113	184.2963	−03.1870	51208.26	0.0908(22)	−1.0544(17)	0.0756(18)	1.0571(17)	274.10 ± 0.10	55.2 ± 1.3	28
BRI B1222−1222	186.2176	−12.6431	50903.27	0.0586(38)	−0.2610(110)	−0.1870(110)	0.3220(110)	234.4 ± 1.9	26.0 ± 1.9	27
2MASS J12255432−2739466AB	186.4763	−27.6630	50998.97	0.0751(25)	0.3849(19)	−0.6282(26)	0.7368(29)	148.50 ± 0.10	46.5 ± 1.6	28
DENIS-P J1228.2−1547AB	187.0639	−15.7935	54514.54	0.0448(18)	0.1344(9)	−0.1853(9)	0.2289(9)	144.04 ± 0.22	24.2 ± 1.0	1
2MASS J12373919+6526148	189.4133	+65.4374	51250.47	0.0961(48)	−1.0020(80)	−0.5250(60)	1.1310(90)	242.33 ± 0.23	55.9 ± 2.9	31
2MASSW J1239272+551537AB	189.8645	+55.2605	54513.53	0.0424(21)	0.1252(11)	−0.0004(10)	0.1252(11)	90.2 ± 0.5	14.0 ± 0.7	1
SDSSp J125453.90−012247.4	193.7247	−01.3799	51202.38	0.0849(19)	−0.4787(20)	0.1301(34)	0.4961(18)	285.2 ± 0.4	27.7 ± 0.6	8
SSSPM J1256−1408	194.0586	−14.1443	51238.21	0.0188(19)	−0.7399(13)	−1.0006(14)	1.2445(14)	216.48 ± 0.06	314 ± 31	23	Subdwarf
SDSS J125637.13−022452.4	194.1549	−02.4145	51220.27	0.0111(29)	−0.5121(19)	−0.2977(19)	0.5923(19)	239.83 ± 0.18	254 ± 70	23	Subdwarf
Kelu-1AB	196.4168	−25.6848	54514.56	0.0497(24)	−0.2992(12)	−0.0041(15)	0.2992(12)	269.21 ± 0.28	28.5 ± 1.4	1
LSPM J1314+1320AB	198.5850	+13.3337	51573.43	0.0610(28)	−0.2424	−0.1856	0.3053	233	24	18
ULAS J131508.42+082627.4	198.7851	+08.4409	53736.50	0.0430(80)	−0.0600(90)	−0.0960(90)	0.1130(100)	212 ± 4	12.6 ± 2.6	20
SDSSp J132629.82−003831.5	201.6242	−00.6421	51212.34	0.0500(60)	−0.2260(80)	−0.1070(60)	0.2510(90)	244.6 ± 1.0	24 ± 3	31
2MASSW J1328550+211449	202.2293	+21.2468	51321.15	0.0310(38)	0.2192(17)	−0.4282(18)	0.4811(18)	152.90 ± 0.20	74 ± 9	8
ULAS J133553.45+113005.2	203.9728	+11.5014	55287.48	0.0999(16)	−0.1908(14)	−0.2024(14)	0.2782(12)	223.3 ± 0.3	13.20 ± 0.22	1
SDSSp J134646.45−003150.4	206.6931	−00.5306	51943.33	0.0683(23)	−0.5032(32)	−0.1143(19)	0.5160(33)	257.20 ± 0.20	35.8 ± 1.2	28
2MASS J14044948−3159330AB	211.2071	−31.9924	54515.60	0.0421(11)	0.3448(10)	−0.0108(14)	0.3450(10)	91.79 ± 0.23	38.8 ± 1.0	1
SDSS J141624.08+134826.7	214.1009	+13.8080	55307.42	0.1097(13)	0.0952(14)	0.1329(14)	0.1635(14)	35.6 ± 0.5	7.07 ± 0.10	1	Low-Z?
LSR J1425+7102	216.2713	+71.0360	51318.21	0.0134(5)	−0.6050(3)	−0.1599(10)	0.6258(2)	255.20 ± 0.10	222 ± 9	9	Subdwarf
LHS 2919	217.0175	+13.9372	51227.42	0.0828(41)	−0.3584	−0.4742	0.5944	217	34	18
LHS 2924	217.1801	+33.1775	51612.34	0.0908(13)	−0.3457(4)	−0.7079(4)	0.7878(4)	206.03 ± 0.03	41.1 ± 0.6	21
LHS 2930	217.6578	+59.7236	51251.47	0.1038(14)	−0.8025(7)	0.1668(6)	0.8197(7)	281.74 ± 0.04	37.4 ± 0.5	21
SDSS J143517.20−004612.9	218.8217	−00.7703	51285.21	0.0100(50)	0.0220(80)	0.0100(60)	0.0250(90)	65 ± 11	12 ± 10	31
SDSS J143535.72−004347.0	218.8989	−00.7298	51285.21	0.0160(60)	0.0218(46)	−0.1050(90)	0.1080(90)	168.3 ± 2.3	32 ± 13	31
LHS 377	219.7513	+18.6607	51614.38	0.0284(8)	−0.0115(2)	−1.2158(3)	1.2159(3)	180.54 ± 0.01	203 ± 6	21	Subdwarf
2MASSW J1439284+192915	219.8682	+19.4875	50609.26	0.0696(5)	−1.2298(7)	0.4067(22)	1.2953(2)	288.30 ± 0.10	88.2 ± 0.6	8
SSSPM J1444−2019	221.0861	−20.3229	50941.14	0.0617(21)	−2.8939(25)	−1.9549(25)	3.4924(25)	235.96 ± 0.04	268 ± 9	23	Subdwarf
SDSSp J144600.60+002452.0	221.5026	+00.4144	51641.24	0.0455(33)	0.1800(70)	−0.0655(41)	0.1910(70)	110.1 ± 1.0	20.0 ± 1.6	31
LHS 3003	224.1596	−28.1632	50990.95	0.1590(50)	−0.4700(100)	−0.8440(120)	0.9660(130)	209.1 ± 0.5	28.7 ± 1.0	27
CFBDS J145829+10134AB	224.6225	+10.2284	55283.56	0.0313(25)	0.1740(21)	−0.3818(25)	0.4196(26)	155.50 ± 0.28	63 ± 5	1
TVLM 513-46546	225.2841	+22.8339	51613.40	0.0944(6)	−0.0246(3)	−0.0579(4)	0.0629(4)	203.0 ± 0.3	3.16 ± 0.03	8
TVLM 513-42404	225.5888	+25.4319	51320.26	0.0350(100)	−0.1210(90)	−0.0380(70)	0.1270(90)	253 ± 3	17 ± 5	26
2MASSW J1503196+252519	225.8321	+25.4237	54575.47	0.1572(22)	0.0901(16)	0.5618(16)	0.5690(16)	9.11 ± 0.16	17.16 ± 0.25	1
SDSS J150411.63+102718.3	226.0493	+10.4546	55050.24	0.0461(15)	0.3737(19)	−0.3692(20)	0.5253(19)	134.66 ± 0.22	54.0 ± 1.7	1	Overlum.
2MASSW J1507476−162738	226.9487	−16.4607	50936.22	0.1364(6)	−0.1615(15)	−0.8885(6)	0.9031(5)	190.30 ± 0.10	31.39 ± 0.14	8
TVLM 868-110639	227.5702	−02.6855	51257.27	0.0612(47)	−0.4050(120)	0.0240(60)	0.4060(120)	273.4 ± 0.8	31.5 ± 2.6	26
TVLM 513-8328	228.5856	+23.6848	50613.29	0.0241(45)	−0.1130(80)	−0.0480(70)	0.1230(90)	247 ± 3	24 ± 5	26
SDSS J153417.05+161546.1AB	233.5711	+16.2630	54515.65	0.0249(11)	−0.0799(7)	−0.0361(8)	0.0877(7)	245.7 ± 0.5	16.7 ± 0.8	1
2MASSI J1534498−295227AB	233.7083	−29.8747	54515.66	0.0624(13)	0.0934(10)	−0.2600(13)	0.2763(13)	160.24 ± 0.20	21.0 ± 0.4	1
DENIS-P J153941.9−052042	234.9246	−05.3452	51256.33	0.0645(34)	0.6013(27)	0.1047(34)	0.6104(26)	80.1 ± 0.3	44.9 ± 2.4	2
WISEPA J154151.66−225025.2	235.4649	−22.8405	55665.00	0.3500(1100)	−0.7800(2300)	−0.2100(2600)	0.8500(2300)	254 ± 18	11 ± 5	16
2MASS J15462718−3325111	236.6133	−33.4198	51003.16	0.0880(19)	0.1211(23)	0.1901(23)	0.2254(22)	32.5 ± 0.6	12.15 ± 0.29	28
2MASSW J1553022+153236AB	238.2585	+15.5442	54576.51	0.0751(9)	−0.3858(7)	0.1662(8)	0.4201(7)	293.30 ± 0.12	26.5 ± 0.3	1
LSR J1610−0040AB	242.6208	−00.6814	51243.36	0.0310(3)	−0.7942(21)	−1.2090(14)	1.4465(2)	213.30 ± 0.10	221.0 ± 1.8	9	Subdwarf
SDSSp J162414.37+002915.6	246.0599	+00.4877	51290.23	0.0909(12)	−0.3728(16)	−0.0091(19)	0.3730(16)	268.6 ± 0.3	19.45 ± 0.27	28
2MASS J16262034+3925190	246.5848	+39.4220	50932.43	0.0299(11)	−1.3815(10)	0.2394(10)	1.4021(10)	279.83 ± 0.04	223 ± 8	23	Subdwarf
SDSS J162838.77+230821.1	247.1624	+23.1388	54576.52	0.0751(9)	0.4123(8)	−0.4430(8)	0.6052(8)	137.06 ± 0.07	38.2 ± 0.5	1
2MASSW J1632291+190441	248.1213	+19.0780	50607.36	0.0656(21)	0.2931(9)	−0.0538(10)	0.2980(9)	100.40 ± 0.20	21.5 ± 0.7	8
LHS 3241	251.6315	+34.5821	51316.27	0.0843(8)	−0.3690	−0.3945	0.5402	223	30	11
WISE J164715.57+563208.3	251.8159	+56.5349	51308.41	0.1160(290)	−0.1660(90)	0.2420(80)	0.2940(80)	325.6 ± 1.7	12 ± 3	16	Very red
2MASSW J1658037+702701	254.5159	+70.4504	51626.38	0.0539(7)	−0.1468(7)	−0.3133(9)	0.3460(9)	205.10 ± 0.10	30.4 ± 0.4	8
DENIS-P J170548.3−051645	256.4514	−05.2795	51268.32	0.0450(120)	0.1110(100)	−0.1150(100)	0.1600(100)	136 ± 4	17 ± 5	2
2MASSI J1711457+223204	257.9406	+22.5346	50614.33	0.0331(48)	0.0310(70)	−0.0042(39)	0.0310(80)	98 ± 7	4.5 ± 1.3	31
2MASSW J1728114+394859AB	262.0481	+39.8164	54576.59	0.0387(7)	0.0358(5)	−0.0183(5)	0.0402(5)	117.2 ± 0.8	4.92 ± 0.11	1
LSPM J1735+2634AB	263.8045	+26.5793	54576.60	0.0667(14)	0.1496(7)	−0.3192(8)	0.3525(8)	154.88 ± 0.12	25.1 ± 0.5	1
WISEP J174124.27+255319.6	265.3526	+25.8929	51645.39	0.1820(380)	−0.4880(160)	−1.4760(160)	1.5550(160)	198.3 ± 0.6	41 ± 9	16
2MASSW J1750129+442404AB	267.5533	+44.4019	54576.60	0.0303(10)	−0.0151(8)	0.1433(9)	0.1441(9)	354.0 ± 0.3	22.6 ± 0.7	1
2MASS J17502484−0016151	267.6035	−00.2709	51257.41	0.1085(26)	−0.3992(32)	0.1957(33)	0.4445(32)	296.1 ± 0.4	19.4 ± 0.5	2
SDSSp J175032.96+175903.9	267.6372	+17.9845	51260.48	0.0362(45)	0.1780(70)	0.1000(50)	0.2040(80)	60.8 ± 1.1	27 ± 4	31
LP 44-162	269.3142	+70.7003	51316.33	0.0524(11)	0.0130	0.3292	0.3295	2	30	18
2MASSI J1835379+325954	278.9079	+32.9985	50923.48	0.1765(5)	−0.0807(13)	−0.7547(11)	0.7590(11)	186.10 ± 0.10	20.39 ± 0.06	22
LP 335-12	279.8879	+29.8712	51638.45	0.0793(20)	0.0798	−0.2183	0.2324	160	14	18
LP 44-334	280.0099	+72.6817	51319.46	0.0593(22)	−0.0330	0.1891	0.1920	350	15	18
2MASSW J1841086+311727	280.2859	+31.2911	50924.44	0.0236(19)	0.0594(32)	0.0416(26)	0.0726(37)	55.0 ± 1.5	14.6 ± 1.4	31
CE 507	280.8016	−33.3754	51620.40	0.0655(25)	−0.1557(42)	−0.3615(28)	0.3936(24)	203.3 ± 0.7	28.5 ± 1.1	7
LHS 3406	280.8422	+40.6725	51288.50	0.0707(8)	−0.1187(4)	0.5940(4)	0.6057(4)	348.70 ± 0.04	40.6 ± 0.5	21	Overlum.
SCR J1845−6357AB	281.2726	−63.9632	51693.19	0.2595(11)	2.5919(18)	0.6175(27)	2.6644(17)	76.60 ± 0.06	48.69 ± 0.21	14
2MASSI J1847034+552243AB	281.7648	+55.3788	54314.36	0.0298(11)	0.1244(10)	−0.0621(11)	0.1391(10)	116.5 ± 0.5	22.1 ± 0.8	1
LSR J2036+5059	309.0902	+51.0014	51702.31	0.0216(13)	0.7552(13)	1.2578(13)	1.4671(13)	30.98 ± 0.05	322 ± 19	23	Subdwarf
SDSS J205235.31−160929.8AB	313.1477	−16.1580	54314.45	0.0339(8)	0.3997(6)	0.1527(7)	0.4279(6)	69.09 ± 0.09	59.8 ± 1.4	1
2MASS J21011544+1756586AB	315.3143	+17.9496	51671.49	0.0301(34)	0.1440(29)	−0.1509(30)	0.2085(37)	136.3 ± 0.5	33 ± 4	31
LP 397-10	319.0262	+22.6462	51004.42	0.0484(11)	0.0680	0.1789	0.1914	21	19	11
[HB88] M18	319.6323	−45.0979	51437.04	0.0470(80)	0.4090(70)	−0.4700(80)	0.6230(90)	139.0 ± 0.5	63 ± 11	27
LSPM J2124+4003	321.1348	+40.0667	51689.39	0.0667(13)	0.5441	0.4455	0.7032	51	50	11
HB 2124−4228	321.8589	−42.2551	51410.09	0.0290(60)	0.1280(70)	−0.1140(60)	0.1720(90)	131.8 ± 0.7	28 ± 6	27
[HB88] M20	322.5359	−44.7744	51410.11	0.0370(160)	0.0630(60)	−0.4370(80)	0.4410(80)	171.8 ± 0.7	56 ± 29	27
2MASSI J2132114+134158AB	323.0480	+13.6995	54314.50	0.0360(7)	0.0194(14)	−0.1225(7)	0.1240(7)	171.0 ± 0.6	16.3 ± 0.3	1
2MASSW J2140293+162518AB	325.1220	+16.4217	54314.49	0.0325(11)	−0.0686(8)	−0.0828(8)	0.1075(8)	219.6 ± 0.4	15.7 ± 0.5	1
LSPM J2158+6117	329.6441	+61.2850	51448.32	0.0592(22)	0.7923	0.1061	0.7993	82	64	11
2MASSW J2206228−204705AB	331.5952	−20.7847	54635.61	0.0357(12)	0.0130(9)	−0.0319(11)	0.0344(11)	157.8 ± 1.5	4.57 ± 0.21	1
GRH 2208−20	332.7083	−19.8736	50998.35	0.0247(5)	−0.2068(11)	−0.6600(4)	0.6917(2)	197.40 ± 0.10	132.7 ± 2.7	8
TVLM 890-60235	335.7731	+00.5030	51742.31	0.0194(22)	−0.0706(16)	−0.0165(28)	0.0725(15)	256.9 ± 2.2	17.7 ± 2.1	26
2MASSW J2224438−015852	336.1839	−01.9830	54316.47	0.0862(11)	0.4686(5)	−0.8648(6)	0.9836(6)	151.55 ± 0.03	54.1 ± 0.7	1	Very red
LHS 523	337.2267	−13.4216	50989.35	0.0888(49)	−0.3166	−1.0357	1.0830	197	58	29
2MASS J22344161+4041387AB	338.6734	+40.6941	51096.09	0.0031(6)	−0.0017(5)	−0.0031(5)	0.0035(5)	209 ± 8	5.4 ± 1.2	30	LkHα 233
LP 460-44	338.9544	+18.6750	50725.17	0.0435(36)	0.3234	0.0423	0.3262	83	36	11
ULAS J223955.76+003252.6	339.9823	+00.5479	53736.50	0.0100(50)	0.1250(50)	−0.1080(50)	0.1660(50)	130.9 ± 1.8	75 ± 50	20
DENIS-P J225210.73−173013.4AB	343.0458	−17.5031	54318.51	0.0632(16)	0.3972(23)	0.1443(36)	0.4226(20)	70.0 ± 0.5	31.7 ± 0.8	1
SDSSp J225529.09−003433.4	343.8711	−00.5760	51393.27	0.0162(26)	−0.0362(14)	−0.1763(25)	0.1799(26)	191.6 ± 0.4	53 ± 9	31
2MASS J23062928−0502285	346.6220	−05.0413	51075.18	0.0826(26)	0.9221(22)	−0.4719(32)	1.0358(18)	117.10 ± 0.19	59.4 ± 1.9	7
APMPM J2330−4737	352.5672	−47.6128	51834.07	0.0727(33)	−0.5648(46)	−0.9741(34)	1.1261(26)	210.10 ± 0.26	73 ± 3	7
APMPM J2331−2750	352.8406	−27.8306	51341.43	0.0691(21)	0.0772(20)	0.7598(13)	0.7637(13)	5.80 ± 0.15	52.4 ± 1.6	7
APMPM J2344−2906	355.8833	−29.1076	51126.01	0.0323(46)	0.3412(39)	−0.2233(48)	0.4077(29)	123.2 ± 0.8	60 ± 9	7
2MASSI J2356547−155310	359.2282	−15.8864	51011.33	0.0690(34)	−0.4434(21)	−0.6002(25)	0.7462(29)	216.46 ± 0.11	51.2 ± 2.5	31
APMPM J2359−6246	359.6786	−62.7618	51526.09	0.0480(22)	0.5728(25)	0.0836(38)	0.5789(25)	81.7 ± 0.4	57.3 ± 2.7	7
Ultracool Companions
2MASSI J0003422−282241	000.9277	−28.3783	55050.53	0.0257(9)	0.2808(9)	−0.1415(8)	0.3145(10)	116.75 ± 0.13	58.0 ± 2.1	30	Overlum.
GJ 1001BC	001.1452	−40.7350	51392.29	0.0769(40)	0.6436(32)	−1.4943(21)	1.6270(18)	156.70 ± 0.12	100 ± 5	14
HD 1160B	003.9887	+04.2511	51872.27	0.0097(5)	0.0211(5)	−0.0142(4)	0.0255(5)	123.9 ± 0.9	12.5 ± 0.6	30	Young
LHS 1070BC	006.1841	−27.1401	51542.05	0.1295(25)	−0.1325(49)	0.6401(31)	0.6537(30)	348.3 ± 0.4	23.9 ± 0.5	6
2MASS J00250365+4759191AB	006.2652	+47.9887	51123.26	0.0228(9)	0.2750(7)	0.0117(8)	0.2752(7)	87.57 ± 0.16	57.2 ± 2.2	30
HD 3651B	009.8291	+21.2548	50726.24	0.0904(3)	−0.4607(3)	−0.3695(3)	0.5905(3)	231.27 ± 0.03	30.96 ± 0.11	30
GJ 1048B	038.9997	−23.5224	51128.19	0.0470(9)	0.0836(10)	0.0136(8)	0.0848(10)	80.8 ± 0.5	8.55 ± 0.20	30
β Pic b	086.8212	−51.0665	51518.17	0.0514(1)	0.0046(1)	0.0831(2)	0.0832(2)	3.20 ± 0.08	7.67 ± 0.02	30	β Pic, planet
CD-35 2722 B	092.3301	−35.8253	51470.29	0.0470(30)	−0.0085(47)	−0.0615(41)	0.0622(41)	188 ± 4	6.3 ± 0.6	32	AB Dor
Gl 229B	092.6443	−21.8645	51191.20	0.1738(10)	−0.1371(5)	−0.7142(8)	0.7272(8)	190.87 ± 0.04	19.83 ± 0.12	30
AB Pic b	094.8038	−58.0543	51506.26	0.0217(7)	0.0144(8)	0.0446(8)	0.0469(8)	17.8 ± 0.9	10.2 ± 0.4	30	Tuc-Hor
HD 46588B	101.6148	+79.5846	51236.15	0.0560(3)	−0.0991(2)	−0.6037(3)	0.6118(3)	189.33 ± 0.02	51.83 ± 0.25	30
HD 49197B	102.3389	+43.7591	51122.39	0.0223(6)	−0.0359(7)	−0.0496(6)	0.0612(6)	215.9 ± 0.6	13.0 ± 0.4	30
HD 65216BC	118.4222	−63.6473	51550.16	0.0281(6)	−0.1229(7)	0.1455(7)	0.1905(7)	319.81 ± 0.22	32.1 ± 0.7	30
HIP 38939B	119.5055	−25.6497	51218.22	0.0540(11)	0.3620(6)	−0.2457(7)	0.4375(6)	124.16 ± 0.09	38.4 ± 0.8	30
WD 0806−661B	121.8111	−66.3135	51538.20	0.0522(17)	0.3403(29)	−0.2896(33)	0.4468(18)	130.4 ± 0.5	40.6 ± 1.3	24
2MASSs J0850359+105716AB	132.6495	+10.9544	54428.61	0.0301(8)	−0.1442(6)	−0.0126(6)	0.1447(6)	265.01 ± 0.24	22.8 ± 0.6	1
Gl 337CD	138.0612	+14.9943	50770.48	0.0491(5)	−0.5246(5)	0.2457(4)	0.5793(6)	295.10 ± 0.04	55.9 ± 0.6	30
LP 261-75B	147.7729	+35.9673	50897.30	0.0160(70)	−0.1000(70)	−0.1610(90)	0.1890(110)	211.8 ± 1.6	56 ± 32	31	Young
HD 89744B	155.5620	+41.2407	50908.18	0.0254(3)	−0.1199(3)	−0.1389(3)	0.1835(3)	220.81 ± 0.09	34.3 ± 0.4	30
Gl 417BC	168.1070	+35.8037	50943.25	0.0456(4)	−0.2490(4)	−0.1510(4)	0.2912(4)	238.76 ± 0.07	30.3 ± 0.3	30	Young
2MASSW J1200329+204851	180.1372	+20.8143	51325.16	0.0340(110)	−0.1264	0.2710	0.2990	335	42	29
LHS 330	187.3095	+53.5517	51219.49	0.0396(11)	−1.2007(3)	0.1357(4)	1.2083(3)	276.45 ± 0.02	145 ± 4	21
Ross 458C	195.1739	+12.3541	54572.50	0.0855(15)	−0.6163(15)	−0.0136(10)	0.6164(15)	268.74 ± 0.09	34.2 ± 0.6	30	Young?
HD 114762B	198.0826	+17.5171	50837.48	0.0259(8)	−0.5796(5)	−0.0022(4)	0.5796(5)	269.79 ± 0.04	106 ± 3	30	Subdwarf
2MASS J13204159+0957506	200.1733	+09.9641	51620.35	0.0259(16)	−0.2511(13)	−0.1431(12)	0.2890(14)	240.31 ± 0.21	53 ± 3	30
2MASS J13204427+0409045	200.1845	+04.1513	51603.35	0.0323(9)	−0.5089(8)	0.2028(8)	0.5478(8)	291.73 ± 0.08	80.4 ± 2.1	30
ULAS J141623.94+134836.3	214.0998	+13.8101	54598.50	0.1097(13)	0.0952(13)	0.1329(14)	0.1635(14)	35.6 ± 0.5	7.07 ± 0.10	1	Low-Z?
SDSS J141659.78+500626.4	214.2495	+50.1072	51604.38	0.0219(6)	−0.2993(5)	0.1854(5)	0.3521(5)	301.77 ± 0.09	76.2 ± 2.2	30
BD +01 2920B	215.8369	+01.2773	55220.02	0.0582(5)	0.2240(4)	−0.4777(4)	0.5276(4)	154.88 ± 0.04	43.0 ± 0.4	30	Low-Z?
GD 165B	216.1629	+09.2862	51638.29	0.0317(25)	−0.2182	−0.1260	0.2520	240	38	29
Proxima Cen	217.4288	−62.6796	51615.33	0.7699(5)	−3.7738(4)	0.7705(20)	3.8517(1)	281.54 ± 0.03	23.72 ± 0.02	4
DENIS-P J144137.3−094559AB	220.4049	−09.7664	51251.40	0.0364(36)	−0.1981(29)	−0.0156(44)	0.1987(29)	265.5 ± 1.3	25.9 ± 2.6	7
G 239-25B	220.5902	+66.0558	51298.32	0.0932(13)	−0.3017(13)	−0.0361(16)	0.3038(13)	263.2 ± 0.3	15.46 ± 0.23	30
HD 130948BC	222.5659	+23.9118	51305.23	0.0550(3)	0.1439(4)	0.0327(3)	0.1476(4)	77.20 ± 0.13	12.72 ± 0.09	30	Young?
Gl 569Bab	223.6218	+16.1011	51336.17	0.1036(17)	0.2760(18)	−0.1217(16)	0.3016(18)	113.8 ± 0.3	13.81 ± 0.24	30
Gl 570D	224.3123	−21.3633	50949.25	0.1712(9)	1.0305(10)	−1.7151(9)	2.0009(8)	149.00 ± 0.03	55.4 ± 0.3	30
TVLM 513-42404B	225.5888	+25.4344	51320.26	0.0350(100)	−0.1210(90)	−0.0380(70)	0.1270(90)	253 ± 3	17 ± 5	26
ULAS J150457.65+053800.8	226.2411	+05.6342	52757.75	0.0538(28)	−0.6087(32)	−0.5026(33)	0.7895(31)	230.45 ± 0.24	70 ± 4	30
Gl 584C	230.8443	+30.2489	51597.42	0.0560(8)	0.1171(4)	−0.1717(4)	0.2078(5)	145.72 ± 0.12	17.60 ± 0.25	30
HR 6037B	244.2728	−67.9411	48347.50	0.0192(4)	−0.0460(3)	−0.0840(3)	0.0958(3)	208.70 ± 0.18	23.7 ± 0.5	30	Young
GJ 618.1B	245.1089	−04.2754	51694.10	0.0299(27)	−0.4153(19)	−0.0219(17)	0.4159(19)	266.98 ± 0.24	66 ± 6	30
vB 8	253.8971	−08.3945	51279.32	0.1545(7)	−0.8147(6)	−0.8691(6)	1.1912(6)	223.15 ± 0.03	36.55 ± 0.17	21
GJ 660.1B	258.2134	−05.1236	51268.41	0.0501(36)	0.1809(47)	−0.6938(35)	0.7169(34)	165.4 ± 0.4	68 ± 5	30
SDSS J175805.46+463311.9	269.5227	+46.5528	50973.38	0.0710(19)	−0.0166(23)	0.5801(17)	0.5803(17)	358.36 ± 0.22	38.7 ± 1.0	30
PZ Tel B	283.2745	−50.1805	51836.02	0.0194(10)	0.0176(11)	−0.0836(8)	0.0855(8)	168.1 ± 0.8	20.9 ± 1.1	30	β Pic
vB 10	289.2401	+05.1506	51390.22	0.1701(8)	−0.5888(8)	−1.3691(8)	1.4903(8)	203.27 ± 0.03	41.53 ± 0.20	21
HR 7329B	290.7134	−54.4240	51823.01	0.0207(2)	0.0256(2)	−0.0827(2)	0.0866(2)	162.82 ± 0.14	19.79 ± 0.20	30	β Pic
Gl 758B	290.8918	+33.2220	51658.43	0.0635(4)	0.0833(3)	0.1622(3)	0.1824(3)	27.19 ± 0.10	13.63 ± 0.08	30
GJ 1245B	298.4795	+44.4153	50977.46	0.2202(15)	0.4023(4)	−0.4661(4)	0.6157(2)	139.20 ± 0.05	13.26 ± 0.09	13
HR 7672B	300.9799	+17.0845	51683.40	0.0563(4)	−0.3927(3)	−0.4059(3)	0.5648(3)	224.05 ± 0.03	47.6 ± 0.3	30
Gl 802B	310.8300	+55.3478	51351.40	0.0635(13)	0.8779(10)	1.7222(10)	1.9330(10)	27.01 ± 0.03	144.3 ± 2.9	15
HD 203030B	319.7425	+26.2306	50748.13	0.0245(7)	0.1326(8)	0.0084(6)	0.1328(8)	86.36 ± 0.28	25.7 ± 0.8	30	Young
HN Peg B	326.1186	+14.7688	51081.26	0.0559(5)	0.2296(5)	−0.1133(4)	0.2561(5)	116.26 ± 0.08	21.72 ± 0.18	30	Young
Wolf 940B	326.6618	−00.1774	54385.50	0.0835(39)	0.7619	−0.5139	0.9190	124	52	29
Ind Bab	331.0438	−56.7827	51490.11	0.2761(3)	3.9609(2)	−2.5392(2)	4.7050(2)	122.66 ± 0.00	80.79 ± 0.08	30
G 216-7B	339.3857	+39.3777	51096.12	0.0512(16)	0.0187(17)	−0.3423(13)	0.3429(13)	176.88 ± 0.29	31.8 ± 1.0	30
HR 8799d	346.8694	+21.1344	48347.50	0.0254(7)	0.1079(6)	−0.0496(5)	0.1188(6)	114.69 ± 0.24	22.2 ± 0.6	30	Planet, young
HR 8799b	346.8694	+21.1344	48347.50	0.0254(7)	0.1079(6)	−0.0496(5)	0.1188(6)	114.69 ± 0.24	22.2 ± 0.6	30	Planet, young
HR 8799c	346.8694	+21.1344	48347.50	0.0254(7)	0.1079(6)	−0.0496(5)	0.1188(6)	114.69 ± 0.24	22.2 ± 0.6	30	Planet, young
HR 8799e	346.8694	+21.1344	48347.50	0.0254(7)	0.1079(6)	−0.0496(5)	0.1188(6)	114.69 ± 0.24	22.2 ± 0.6	30	Planet, young
2MASS J23310161−0406193AB	352.7567	−04.1054	51114.07	0.0383(5)	0.1792(6)	−0.1914(6)	0.2622(6)	136.88 ± 0.13	32.5 ± 0.5	30
APMPM J2354−3316C	358.5387	−33.2741	51386.34	0.0442(18)	−0.3173(25)	−0.4062(22)	0.5154(15)	218.0 ± 0.3	55.2 ± 2.3	24

Notes. This table gives astrometric parameters for all ultracool dwarfs with parallax determinations. To be included in this list an object must have a spectral type ⩾M6 or K-band absolute magnitude  >  8.5 mag. For parameters in units of arcseconds, errors are given in parentheses in units of 10⁻⁴ arcsec. (α, δ, MJD): coordinates that correspond to the epoch listed. (π_abs, μ_αcos δ, μ_δ, μ, P.A.): absolute parallax and proper motion parameters listed both as (α, δ) and as total amplitude (μ) and position angle. V_tan: tangential velocity computed from the proper motion and parallax. Ultracool companions to stars (or other ultracool dwarfs) are listed separately, even if the source of the parallax determination is not for the primary (e.g., vB 8 has an independent parallax measurement). Since some literature sources do not provide uncertainties in the proper motion we cannot compute some V_tan errors. "Note" column indicates special characteristics of some objects: subdwarfs, planetary-mass objects, members of specific young moving groups or otherwise young objects (≲ 300 Myr). References. (1) This work; (2) Andrei et al. 2011; (3) Artigau et al. 2010; (4) Benedict et al. 1999; (5) Burgasser et al. 2008b; (6) Costa et al. 2005; (7) Costa et al. 2006; (8) Dahn et al. 2002; (9) Dahn et al. 2008; (10) Ducourant et al. 2008; (11) Gatewood & Coban 2009; (12) Geyer et al. 1988; (13) Harrington et al. 1993; (14) Henry et al. 2006; (15) Ireland et al. 2008; (16) Kirkpatrick et al. 2011; (17) Leggett et al. 2012; (18) Lépine et al. 2009; (19) Liu et al. 2011a; (20) Marocco et al. 2010; (21) Monet et al. 1992; (22) Reid et al. 2003b; (23) Schilbach et al. 2009; (24) Subasavage et al. 2009; (25) Teixeira et al. 2008; (26) Tinney et al. 1995; (27) Tinney 1996; (28) Tinney et al. 2003; (29) van Altena et al. 1995; (30) van Leeuwen 2007; (31) Vrba et al. 2004; (32) Wahhaj et al. 2011.

Download table as:  ASCIITypeset images: 1 2 3 4 5 6

We also include mid-IR photometry from Spitzer/IRAC (e.g., Patten et al. 2006; Leggett et al. 2007, 2010) and the WISE All-Sky Source Catalog¹⁰ (Wright et al. 2010). We checked for WISE quality flags indicating possibly spurious or contaminated detections for all objects after noting that Kirkpatrick et al. (2011) include sources with nonzero flags in their Table 1. We visually inspected the WISE image atlas in the worst cases, i.e., "H" and "D" flags indicating possible spurious detections, and found that the sources are in fact likely to be real. A published example of one such object is HD 46588B shown in Figure 1 of Loutrel et al. (2011), which is flagged in the WISE catalog as potentially spurious even though it is real. After vetting sources against the WISE image atlas, we do not find the need to exclude any flagged WISE photometry from the following analysis.

Tables 10 and 11 present the resulting collections of apparent magnitudes in the near- and mid-IR, respectively. In total, there are 314 objects in 255 systems that have parallax measurements. In Tables 12 and 13 we list absolute magnitudes sorted by spectral type, along with references for any high angular resolution imaging available. This encompasses numerous HST imaging programs with WFPC2, NICMOS, ACS, and STIS; AO surveys at Keck, VLT, Gemini, Subaru, Palomar, CFHT, and Lick; as well as speckle and lucky imaging surveys. We note that there are unpublished archival data for many of the objects compiled in our table, but we only count observations for which analysis has been published. The only exception is for the subset of objects observed by our Keck AO binary survey that we have determined to be unresolved in FWHM = 005–010 imaging (M. C. Liu et al., in preparation). The intention of this compilation is to identify a clean sample of likely single objects (i.e., with no companions outside ≈01) from objects that have not been surveyed for binarity. Thus, we assign null entries for the handful of unresolved spectroscopic and astrometric binaries that have been imaged at high angular resolution in order to remove them from the subset of likely single objects.