VLMOs in USco using Gaia DR1

Very Low-Mass Stars and Brown Dwarfs in Upper Scorpius using Gaia DR1: Mass Function, Disks and Kinematics

[ Institut de Recherche sur les Exoplanètes, Université de Montréal, Montréal, QC, Canada, H3T 1J4 Faculty of Science, York University, 4700 Keele Street, Toronto, Canada, ON M3J 1P3 [ SUPA, School of Physics & Astronomy, University of St. Andrews, North Haugh, St. Andrews, KY16 9SS, UK [ Faculty of Science, York University, 4700 Keele Street, Toronto, Canada, ON M3J 1P3
September 11, 2017October 16, 2017October 17, 2017
September 11, 2017October 16, 2017October 17, 2017
September 11, 2017October 16, 2017October 17, 2017
Abstract

Our understanding of the brown dwarf population in star forming regions is dependent on knowing distances and proper motions, and therefore will be improved through the Gaia space mission. In this paper, we select new samples of very low mass objects (VLMOs) in Upper Scorpius using UKIDSS colors and optimised proper motions calculated using Gaia DR1. The scatter in proper motions from VLMOs in Upper Scorpius is now (for the first time) dominated by the kinematic spread of the region itself, not by the positional uncertainties. With age and mass estimates updated using Gaia parallaxes for early type stars in the same region, we determine masses for all VLMOs. Our final most complete sample includes 453 VLMOs of which 125 are expected to be brown dwarfs. The cleanest sample is comprised of 131 VLMOs, with 105 brown dwarfs. We also compile a joint sample from the literature which includes 415 VLMOs, out of which 152 are likely brown dwarfs. The disc fraction among low-mass brown dwarfs () is substantially higher than in more massive objects, indicating that discs around low-mass brown dwarfs survive longer than in low-mass stars overall. The mass function for is consistent with the Kroupa IMF. We investigate the possibility that some ‘proper motion outliers’ have undergone a dynamical ejection early in their evolution. Our analysis shows that the color-magnitude cuts used when selecting samples introduce strong bias into the population statistics due to varying level of contamination and completeness.

brown dwarfs — stars: lowmass — stars: mass function — open clusters and associations: individual: Upper Sco
journal: AJ\correspondingauthor

Neil J. Cook

0000-0003-4166-4121]Neil J. Cook

0000-0001-8993-5053]Aleks Scholz

0000-0001-5349-6853]Ray Jayawardhana

1 Introduction

Most newly formed stars have masses significantly lower than the Sun. The characteristic mass of star formation, the peak of the Initial Mass Function (IMF), is around 0.2, almost independent of environment (Bonnell et al., 2007). The mass distribution of objects formed in young clusters extends far below the sub-stellar limit at 0.08 and into the planetary (fusion-less) mass domain at (Luhman, 2012; Scholz et al., 2012). In this very low mass (VLM) domain, a variety of formation channels might play a role, including turbulent fragmentation of clouds, dynamical ejections from multiple systems, or disc fragmentation (see Whitworth et al., 2007).

Very low mass object (VLMOs) are also viable host stars for exoplanet systems, as evidenced by discoveries of Earth-sized or -massed planets around mid M dwarfs (Berta-Thompson et al., 2015; Anglada-Escudé et al., 2016; Dittmann et al., 2017; Gillon et al., 2017). The ubiquity of these systems poses an interesting challenge for core accretion theories, as discs around young objects in this mass domain usually do not seem to have sufficient material to form these type of systems (Scholz et al., 2006; Testi et al., 2015; Pascucci et al., 2016), implying very rapid formation. VLMOs are also possible hosts of ultracool dwarfs with L, T, or Y spectral types (Bardalez Gagliuffi et al., 2013; Cook et al., 2016, 2017).

Identifying and characterizing the VLM population in star forming regions provides the observational constraints on star formation scenarios as well as the samples for in-situ studies of planet formation. Traditionally, the selection of VLM cluster members is based on cuts in color-magnitude and proper motion space, followed by spectroscopy to confirm (Luhman et al., 2003; Wilking et al., 2004; Scholz et al., 2012). So far, proper motion cuts were limited to a few nearby regions with space motions significantly offset from the Galactic background. The output from the astrometry mission Gaia is about to change that (Gaia Collaboration et al., 2016a, b). It is anticipated that the final Gaia data releases will provide the first large, uniform sample of parallaxes for young brown dwarfs in addition to sub-milliarcsecond precision in proper motions.

Gaia published its first data release in 2017 (Gaia Collaboration et al., 2016a, b, henceforth Gaia DR1). While Gaia DR1 does not yet provide Gaia-internal parallaxes and has not yet reached optimum astrometric precision, it can already be used for improving current selection methods for young VLMOs and to refine the resulting samples, as we demonstrate in this paper for the nearest OB association Upper Scorpius. Using the combined parallaxes from the Tycho-Gaia Astrometric Solution (TGAS) for bright young stars, the estimates for distances, age and spatial depth for nearby star forming regions can be solidified. With the help of the Gaia DR1 astrometry, the scatter in proper motions from VLMOS in Upper Scorpius is now (for the first time) dominated by the kinematic spread of the region itself, not by the positional uncertainties. For this particular region later data releases are unlikely to significantly improve the member selection from proper motions. Upper Scorpius is a region mostly free from reddening, therefore follow-up spectroscopy is not as essential here as in other star forming regions.

In this paper, we estimate distance, age and spatial depth from higher-mass Upper Scorpius members (Section 2). We establish new samples of VLM members of Upper Scorpius using photometry from the United Kingdom Infrared Digital Sky Survey Galactic Clusters Survey (UKIDSS/GCS, Lawrence et al., 2007, see Section 3), and optimized astrometry obtained by combining Gaia DR1 with other catalogs (Section 4). Using these new samples we estimate masses (Section 5.1), test the the disc fraction as a function of mass (Section 5.2) and the mass function (Section 5.3). We also make a first attempt at tracking down kinematic outliers (Section 5.4), i.e.  objects with proper motions significantly different from their nearby siblings, which could be those who experienced an early dynamical ejection from a disc. The paper is intended to pave the way for future studies of other star forming regions based on the next Gaia data releases.

For the purposes of this paper, we use the term VLMO for all objects with masses below , including very low mass stars, brown dwarfs and planetary mass objects.

2 The Upper Scorpius Association with Gaia DR1

Figure 1: The 74 higher-mass members of Upper Scorpius used to determine distance, age and spatial depth for Upper Scorpius (in blue). The upper left panel shows their distribution in proper motion, the upper right panel shows their distance spread, the bottom left panel shows the distribution in space, and the lower right panel shows the color-magnitude diagram with 7, 10, and 15  isochrones (cyan, red, yellow) from Siess et al. (2000). Over-plotted in red are the rejected candidates.

When deriving stellar properties for members of star forming regions, the main source of uncertainty are distance and age. Gaia DR1 does not provide parallaxes for very low mass members of Upper Scorpius – this is expected to be included in later data releases – but the TGAS catalog does contain parallaxes for a substantial sample of early-type stellar members. TGAS111Documentation available from https://gaia.esac.esa.int/documentation/GDR1/Data_processing/chap_cu3tyc/, is a combination of the Tycho-2 with the Gaia catalog, listing astrometric data for 2.5 million stars (Michalik et al., 2014; Gaia Collaboration et al., 2016a, b). In this section we aim to use this dataset to re-determine distance and age for Upper Scorpius .

We start with the sample of high-mass members of Upper Scorpius from de Zeeuw et al. (1999), selected as a moving group using Hipparcos astrometry. The mean Hipparcos distance in this sample is 1451 pc. From the 120 members listed there, 85 have an entry in TGAS (with 1″matching radius). From the catalog of primarily K-type members by Pecaut et al. (2012), we add 36 stars without Hipparcos, but with TGAS entry. To clean the sample, we remove objects with parallax error (16 objects), without Tycho-2 photometry (4 objects) or magnitude error  mag (24 objects), and with implausible distances  pc (9 objects) leaving 74 objects (47 objects rejected in total, taking into account those rejected by multiple criteria).

Five of the rejected stars at distance  pc are classified as K-M giants in the Michigan Spectral Survey (Houk & Smith-Moore, 1988), among them HIP83542, also rejected by Pecaut et al. (2012). In Figure 1, the cleaned sample has clusters around   mas  , with standard deviation of 3 mas  .

The cleaned sample of 74 stars have a median TGAS distance of 146.1 pc, and a mean TGAS distance of 1452 pc. The median parallax error in this sample is 0.36 mas – this includes a systematic error of 0.14 mas, see Table 2 in Arenou et al. (2017). The parallax error translates to a median distance error of 7.8 pc and a scaled error of  pc for the distance of the association. Thus, this provisional distance estimate from TGAS is within the error bars of the Hipparcos result. The parallax distribution (see Figure 1, upper right panel) has a standard deviation of 15.2 pc. Subtracting the errors in quadrature, this suggests a depth of the region along the line of sight of about  pc. For comparison, in the plane of the sky, the sample is contained within an area of 20 deg diameter in both RA and DEC, corresponding to 50 pc at the distance of Upper Scorpius .

From the cleaned sample, we produce a color-magnitude diagram (see Figure 1, lower right panel). The absolute magnitudes have been calculated using the individual TGAS distances for each star, eliminating the error caused by the depth of the region. Uncertainties in and are comparable to the size of the symbols for most objects. Variability is expected to have a minor effect on the position of the stars in the diagram for a region of this age. Without widespread accretion or discs (Luhman, 2012), the only plausible source of variability is magnetic activity, which typically does not cause large amplitudes, even at this young age (Grankin et al., 2008). Therefore, the spread in the color-magnitude diagram likely corresponds to a real age spread in Upper Scorpius .

Over-plotted in Figure 1 are the 7, 10, and 15  isochrones222http://www.astro.ulb.ac.be/~siess/pmwiki/pmwiki.php/WWWTools/Isochrones (cyan, red, yellow) from Siess et al. (2000), using a metallicity of (solar metallicity, Siess et al. 2000) and the Kenyon & Hartmann (1995) conversion table. These three isochrones bracket most of the sample for . For later-type objects the spread in the data points exceeds the model dispersion, partly due to increased photometric errors. Thus 7-15  is a plausible minimum estimate for the age spread in this region. This is consistent with recent age studies for Upper Scorpius , see Pecaut & Mamajek (2016); Fang et al. (2017), but in conflict with earlier claims of 5 2  by Preibisch et al. (2002). Note that we checked isochrones with non-solar metallicity and the differences were negligible for our purposes.

3 Very low mass objects in Upper Scorpius

Many previous works have studied the population of VLMOs in Upper Scorpius (e.g.  Slesnick et al. 2008; Dawson et al. 2011; Lodieu et al. 2011; Luhman & Mamajek 2012; Dawson et al. 2013; Lodieu 2013; Lodieu et al. 2013; Dawson et al. 2014). Most use various color cuts to select potential VLMOs, and use proper motions (either calculated or obtained from large catalogs) to identify candidates moving in a similar way to known Upper Scorpius members. One of the main uncertainties of membership (when distance is unknown) is the uncertainties associated with the calculated proper motions. Thus more precise proper motions tend to lead to identifying a better sample of VLMOs in Upper Scorpius .

In this section we compile two samples, the first is a sample directly taken from the literature (and thus based on various colors cuts and slightly different selection criteria), henceforth the ‘L-sample’, the second is a large uniform sample, based on the initial color selection from Lodieu (2013), henceforth the ‘C-sample’.

Figure 2: Color cuts applied to the samples obtained from the WSA. The gray contours show the data for the ‘C-ZYJHK DR10’ sample before the cuts were applied. Colored symbols show objects from the literature sample. The stars represent sources confirmed spectroscopically (Lodieu et al. 2011 or Dawson et al. 2014).

3.1 The L-sample

Although there are many surveys that study Upper Scorpius , we decided to choose those surveys that identify VLMO members using UKIDSS GCS (Lawrence et al., 2007) or similar (i.e.  VISTA) photometry so that we had sub-samples that had Z, Y, J, H, K photometry (’L-ZYJHK sample’); or had H and K photometry (‘L-HK only sample’). We combined data from Dawson et al. (2011), Lodieu et al. (2011), Dawson et al. (2013), Lodieu (2013), Lodieu et al. (2013), and Dawson et al. (2014) to obtain a sample of 789 unique objects, of which 493 were in the L-ZYJHK sample and 295 were in the L-HK only sample using photometry from both UKIDSS GCS DR10 (henceforth DR10) and the GCS Science verification release (henceforth SV; Dye et al., 2006). Tables 7 and 8 give the full detail on how many objects were in each source catalog.

3.2 The C-sample

The data for the C-sample were obtained using the WFCAM Science Archive (WSA Hambly et al., 2008)333WSA available on-line at http://surveys.roe.ac.uk/wsa. using SQL queries (see Appendix B). We followed the initial sampling used by Lodieu (2013), using identical bright saturation limits, limiting merged passband selection to 1″and retaining only point-like, non-duplicated sources. We decided to also follow the sub-sample selection of Lodieu (2013), defining samples that had Z, Y, J, H, K photometry (’C-ZYJHK sample’); or as having H and K photometry (‘C-HK only sample’). We obtained 2,653,897 sources for the C-ZYJHK sample from DR10, 157,325 sources for the C-ZYJHK sample from SV, and 7,473,530 for the ‘HK sample’ of which 4,814,722 do not have Z, Y and J photometry (i.e.  the C-HK only sample).

Following Lodieu (2013) we split the C-ZYJHK sample into a sub-sample affected by reddening and a sub-sample not affected by reddening (henceforth denoted as  for the reddened sample), and remove those objects with ‘HK extinction’ in the C-HK only sample (see table 1 from Lodieu, 2013). The C-ZYJHK DR10 sample had 1,722,423 sources flagged as not affected by reddening and 931,474 flagged as being affected by reddening. The C-ZYJHK SV sample had no sources flagged as affected by reddening. The C-HK only sample had 3,652,715 that were not removed due to reddening.

To select VLMOs from the full samples we used the literature sample and the color cuts identified by Lodieu (2013). Our final color cuts are nearly identical to Lodieu (2013) except that we add an additional cut to , this was in order to make sure our samples were not affected by the tail of the giant branch (See the against plot in Figure 2). The cuts are listed below.

We decided to keep two different combinations of these cuts to see the effect they had on the final population selected, the first used all the above cuts (denoted by a  HK ) and the second used all the color cuts except the ‘HHK cut’.

Thus we have samples with ‘C-ZYJHK DR10’, ‘C-ZYJHK DR10  ’, ‘C-ZYJHK DR10 HK ’, ‘C-ZYJHK DR10   HK ’, ‘C-ZYJHK SV’, ‘C-ZYJHK SV HK ’, and ‘C-HK only’ (where the ‘C’ distinguishes the sub-samples from the L-samples described in Section 3.1). The number of objects left after the color cuts are shown in Table 1 and a full break down of numbers is presented in tables 7 and 8.

Sample Total before cuts ZZJ HHK JJK ZZK YYJ JKZK Total after cuts
C-ZYJHK DR10 1,722,423 5,654 - 23,134 9,977 12,122 245,810 1,305
C-ZYJHK DR10   931,474 1,538 - 9,940 2,569 5,824 400,451 811
C-ZYJHK SV 157,325 135 - 887 143 206 29,723 86
C-ZYJHK DR10  HK 1,722,423 5,654 2,359 23,134 9,977 12,122 245,810 66
C-ZYJHK DR10    HK 931,474 1,538 318 9,940 2,569 5,824 400,451 77
C-ZYJHK SV  HK 157,325 135 71 887 143 206 29,723 33
C-HK only 3,652,715 - 1,526 - - - - 1,526
Table 1: The results for the C-samples after the color cuts are applied.

3.3 Discussion of sub-samples

In summary we have two samples, the L-sample, constructed directly from the literature and the C-sample, constructed from UKIDSS DR10 and SV SQL Queries and color cuts. The L-sample is split into a ‘ZYJHK’ sample (comprising of objects with Z, Y, J, H and K photometry) and the ‘HK only’ (comprising of those object with only H and K photometry), these are named the ‘L-ZYJHK’ and ‘L-HK only’ samples. The C-sample is also split into a ‘ZYJHK’ and the ‘HK only’ sample (with the same definition), split by the data origin (i.e.  either UKIDSS GCS DR10 or UKIDSS GCS SV) and by whether we use the ‘HHK cut’ (‘HKcut’) and whether the area the objects resides is flagged as having reddening (  ). We do this so we can analyze the affect different cuts have on our results. Table 2 describes these sample subsets and their differences.

Sample Sub-sample From DR10 From SV Has Z, Y, J, H, and K photometry Flagged as affected by reddening  HK Cut used
l-sample L-ZYJHK - - - -
L-HK only - - - -
c-sample C-ZYJHK DR10 HK
C-ZYJHK DR10   HK
C-ZYJHK SV HK
C-ZYJHK DR10
C-ZYJHK DR10 
C-ZYJHK SV
C-HK only
Table 2: The definition of the samples

Each of our sub-samples has specific properties due to the imposed selection criteria. The C-HK only sample is going to be the least well defined sample due to the lack of Z, Y and J photometry and therefore lack of all the color cuts except the  HK cut. The   samples (i.e.  C-ZYJHK DR10   and C-ZYJHK DR10   HK ) are expected to be more contaminated due to the increased reddening those source experience (i.e.  reddened objects will be scattered across the color cuts). The SV samples (i.e.  C-ZYJHK SV and C-ZYJHK SV HK ) rely on science verification photometry and will be less complete than the DR10 samples. The SV samples also cover a slightly different spatial location than the DR10 samples and thus may have slight differences in age (see the age gradient from figure 9 of Pecaut & Mamajek 2016). The L-sample consists of some higher mass objects (due to the lack of, for example, brightness cuts), however due to some of these objects being spectroscopically confirmed we do not further reduce the L-sample and use it for comparative purposes only.

Our C-samples are much improved over the previously existing L-sample (due to our use of the best available color cuts and the better proper motion cuts (see Section 4). The addition of the  HK cut to the C-ZYJHK samples add a brightness cut which essentially acts as a mass cut ( see effects in Section 5.3 and 5.2). This cut has the effect of avoiding contamination from red giants at the bright end of the magnitude distribution. Therefore, the C-ZYJHK DR10 HK should be the least contaminated sample among our sub-samples, but also the most restrictive. On the other hand, the C-ZYJHK DR10 should be the most complete. Throughout this paper we do all analysis on all samples to check how much of an effect selection has on any results we obtain.

4 Proper motion analysis

Figure 3: Proper motion vector diagram showing the one and two sigma ellipses used to define Upper Scorpius membership for the L-ZYJHK sample (for the best precision proper motions). The numbers of objects found were compared to a two-dimensional Gaussian distribution of equal center and covariance. Median uncertainties are shown in the upper right corner.

As with many previous moving group membership surveys (e.g.  Dawson et al. 2011, 2013; Lodieu 2013; Lodieu et al. 2013) we define Upper Scorpius membership as having a proper motion in both the Right Ascension and Declination directions consistent with that of Upper Scorpius . Since the release of the Gaia DR1 there have been new catalogs generated using the Gaia DR1 positions for objects without any Gaia proper motions (i.e.  the Gaia DR1 secondary catalog of 1.1 billion sources, Gaia Collaboration et al., 2016b). Three of the largest proper motion catalogs currently using Gaia DR1 positions (and overlapping on-sky with Upper Scorpius ) are the Hot Stuff for One Year catalog, (HSOY; containing 583 million stars Altmann et al., 2017), the Gaia-PS1-SDSS proper motion catalog (GPS1; containing 350 million stars, Tian et al., 2017), and the US Naval Observatory CCD astrograph catalog 5 (UCAC5; containing 107 million stars, Zacharias et al., 2017). We cross-matched (selecting the closest source within a 3″matching radius) all sub-samples (belonging to both the L-sample and the C-sample) with HSOY, GPS1 (also giving us access to the Pan-STARRS1 proper motions; PS1 Chambers, 2011), UCAC5, as well as the PPMXL catalog (Roeser et al., 2010, containing 900 million sources), and the proper motions associated with each source from UKIDSS GCS DR10. From these proper motions the most precise total proper motion was selected for each object (where total proper motion and associated uncertainty are defined in Equation 1).

(1)

where is the proper motion component in the Right Ascension direction () and is the proper motion component in the Declination direction.

We decided to exclude PPMXL proper motions as no sources with only PPMXL proper motions had uncertainties better than 10  mas  . The stars that had a suitable proper motion measurement 716 out of 789 for the L-sample, all 2259 C-ZYJHK DR10 objects, 68 out of 86 C-ZYJHK SV objects, all 17 C-ZYJHK SV HK , 1,519 out of 1,526 C-HK only sample (see tables 7 and 8 for a full break down of numbers).

Sample Total before cuts Total after pm cuts
L-ZYJHk 453 415
L-HK 241 175

C-ZYJHK DR10
1,305 171
C-ZYJHK DR10   881 224
C-ZYJHK SV 68 58
C-ZYJHK DR10  HK 66 49
C-ZYJHK DR10    HK 77 68
C-ZYJHK SV  HK 17 14
C-HK only 1,519 346
Table 3: The result of the Upper Scorpius membership selection.

Our uncertainties in proper motion are sufficiently small that we do not need to select members based on a proper motion uncertainty circle (this is in contrast to previous studies where large uncertainties dominate the velocity dispersion of Upper Scorpius ). However, since the proper motion of Upper Scorpius is very small, we decided to use the L-sample to define a two-sigma membership ellipse for Upper Scorpius (such that we avoid an overlap with  mas  ). We compared the median and standard deviations for the L-ZYJHK, L-HK only and the combined sample. The L-ZYJHK sample was found to have a center of =(-9.80, -19.94)  mas  , with standard deviations of =(7.51, 6.96)  mas  . The L-HK only sample was found to have a center of =(-7.78, -18.39)  mas  , with standard deviations of =(7.82, 9.14)  mas  . The combined L-sample was found to have a center of =(-9.04, -19.46)  mas  , with standard deviations of =(7.88, 8.08)  mas  .

We thus chose to define candidate members of Upper Scorpius as those within an ellipse of center =(-9.80, -19.94) and x and y radii of =(15.10, 13.95)  mas  (defined from the L-ZYJHK distribution). This was then used to select members from the L-sample and C-sample. We keep 415/453 and 175/241 of those objects from the L-ZYJHK and L-HK only samples respectively. For the C-ZYJHK DR10  sample we kept in 224/881 candidates, and 68/77 C-ZYJHK DR10   HK candidates. For the C-ZYJHK  sample we identified 171/1,305 candidates, and for the C-ZYJHK HK sample we kept 49/66 candidates. For the C-ZYJHK SV sample we identify 58/68 as Upper Scorpius candidates, and 14/17 for the C-ZYJHK SV HK sample. The C-HK only sample resulted in 346/1,519 candidates being identified. These numbers are summarized in Table 3 and tables 7 and 8 have a full break down of numbers. Figure 3 shows the L-ZYJHK sample used to selected candidates from both the L-samples and the C-samples.

The addition of a proper motion cut does confirm the characterization of the samples given in Section 3.3. Specifically, the C-HK sample without ZYJ photometry is heavily contaminated, as expected. The same applies to the C-ZYJHK DR10 and C-ZYJHK DR10   samples. We expect most of the contamination in these samples to be at the bright magnitude end, i.e.  at high masses in the VLM domain, because in this regime the population of young Upper Scorpius members is not well separated from the background population in color-magnitude diagrams. As mentioned above, the cleanest sample in our list are the ones with the  HK cut, which removes objects at the bright end.

5 Properties of VLMOs in Upper Scorpius

With an estimated age of 10  (with a spread between 7 and 15  ) and assuming a distance of 145 pc (with a spread of 13 pc, Section 2) it is possible to estimate mass and luminosity by fitting the photometry to theoretical isochrones. We use the 8, 10 and 15 Baraffe et al. (2015) isochrones (BHAC15) to give a lower, median and upper bound to each of our objects with UKIDSS Z, Y, J, H and K photometry (i.e.  we only fit sources which have all five photometric magnitudes), we choose 8 as the lower bound as the 7 isochrone is not computed for BHAC15. In this section we describe the fitting process and use these, with Wide-Field Infrared Survey Explorer (WISE, Wright et al., 2010) data to infer a disc fraction, analyze the mass distributions and explore the proper motion distribution of our candidates.

Figure 4: Example isochronal fit for L-sample object UGCS J161625.98-211222.9. Fit gives a mass of 0.04 .
Figure 5: Log mass histogram for the C-sample using the  HK cut (left) as compared to cases without (right). The  HK cut effectively constrains the mass of the objects to 0.2  (black vertical line) whereas without the  HK cut the masses extend to higher mass objects. Over-plotted, in both sub-plots for reference, is the L-sample (with no  HK cut applied).
Figure 6: Absolute Z magnitude against (Z-J) color (Hertzsprung-Russell diagram) for the L-sample. Objects with discs are marked with an orange star. Median uncertainties are shown with the green cross. The distribution is consistent with an average age of 10  and a spread from 8 to 15  .

5.1 Isochronal fitting

Using the 8, 10 and 15  BHAC15 isochrones for UKIDSS we used chi-squared minimization (using the apparent UKIDSS magnitudes converted to absolute magnitudes using a distance of 146 pc, see Section 2) to select the best fit model for a lower, median and upper bounding model. The nature of the BHAC15 isochrones means for a given set of photometry (ı.e. Z,Y,J,H and K) and age we get a mass estimate for each object (with an associated luminosity, effective temperature,  , radius or surface gravity, for each mass estimate).

The mass estimate attached to each age (8, 10 and 15  ) were then combined, giving an expected value, an upper and a lower uncertainty (described in Equation 2).

(2)

where is the mass estimate associated with the best fit (lower, median and upper bounding) model. This gave us appropriate uncertainties for the estimated mass based on the spread in ages found for Upper Scorpius (Section 2). We do not interpolate between the models and these estimated masses assume that all objects have an age between 8 and 15  with a median of 10  . An example fit can be seen in Figure 4 for L-sample object UGCS J161625.98-211222.9.

The estimated mass distributions for the C-sample using the  HK cut as compared to cases without can be seen in Figure 5 (with the L-sample over-plotted in both cases for comparison). From Figure 5 the differences between the sub-samples becomes clear. The L-samples contain a significant number of higher-mass stars. The application of the  HK cut (left panel compared to right panel) shows that this cut effectively removes objects of mass and hence avoids contamination. The Hertzsprung-Russell diagram for the L-sample is shown in Figure 6. The distribution of colors seems consistent with a typical age of 10  .

Defining brown dwarfs to have a mass less than 0.075  we calculated the number of our objects that are likely brown dwarfs (with the uncertainty coming from those that overlap in mass due to their mass estimate uncertainty). The L-ZYJHK sample was found to have 15238 out of the 415 objects as likely brown dwarfs, the C-ZYJHK DR10 HK , C-ZYJHK DR10    HK and C-ZYJHK SV HK samples were found to have 4211, 5310 and 106 respectively, and the C-ZYJHK DR10, C-ZYJHK DR10   and C-ZYJHK SV samples were found to have 4813, 6713 and 106 respectively. These numbers are presented in Table 4. We note that the number of brown dwarfs are quite similar in the samples with and without the  HK cut; in this mass domain the samples should avoid most contamination.

Sample Total objects in sample Likely brown dwarfs
L-ZYJHK 415 15238
C-ZYJHK DR10 HK 49 4211
C-ZYJHK DR10   HK 68 5310
C-ZYJHK SV HK 14 106
C-ZYJHK DR10 171 4813
C-ZYJHK DR10  224 6713
C-ZYJHK SV 58 106
Table 4: Numbers of possible brown dwarfs (Mass less than 0.075  ).

We were concerned that any objects with discs (see Section 5.2) may, due to their young age, have H and K photometry that is not well represented by one of the BHAC15 isochrones (i.e.  there should be an additional component added to the flux due to the presence of a disc). For this reason we also fitted the masses using only ZYJ and ZYJH and flagged any objects which were identified as having possible discs (from Section 5.2). This led to having three mass estimates for each object and thus we were able to see any differences due to ‘bad’ H and K photometry. After comparing objects with possible discs to those without discs no discernible difference was seen, thus the mass estimates were not affected by those objects having discs. Most objects had a maximum variation (due to using different sets of photometry) of 0.01 thus we decided to retain our mass estimates using all five bands.

Our chi-squared minimization does not take into account the uncertainties in magnitude (shown on Figure 4) and thus it does not take into account any uncertainty due to distance or spread in the association. Note that the spread in the association is significantly larger than the distance uncertainty on Upper Scorpius (1452 pc), i.e.  13 pc from Section 2, thus an additional uncertainty on the mass estimates will be introduced (this will be solved with later Gaia data releases giving distances to individual objects).

As a further sanity check for our mass estimates we compared our results to the Lodieu et al. (2011) and Dawson et al. (2014) samples which both have spectroscopically confirmed members of Upper Scorpius . We compare the spectral types from the literature to the mass estimates from the isochrones and compare the mass estimates from the literature to the mass estimates from the isochrone fits (see Figure 7). The figures show the expected trends and broad agreement, but there are also clear discrepancies. In the left panel of Figure 7 some objects (with very low masses) have surprisingly early spectral types, Dawson et al. (2014) concluded that some of these objects might be further away than the rest of the objects identified as being part of Upper Scorpius . For these objects our mass estimate will be underestimated. In the right panel of Figure 7 we find that our mass estimates are systematically higher (by ) compared to Lodieu et al. (2011). That study derives masses by comparing bolometric magnitudes (derived from J-band) with NextGen/DUST models (Baraffe et al. 1998 and Chabrier et al. 2000 respectively), using a distance consistent with ours, but assume an age of 5 Myr (priv. comm. Lodieu 2017). Between 5 and 10 Myr, VLMOs drop in luminosity, i.e. assuming a younger age leads to lower mass estimates.

Figure 7: The comparison between our mass estimates from the isochrones to those objects in Upper Scorpius with spectral type and masses from the literature (members with spectra) from Lodieu et al. (2011) and Dawson et al. (2014).
Figure 8: Disc fraction as a function of estimated mass (from isochrones). All samples were chosen to have 50 bins ranging from 0.0 to 0.6  (bin sizes were chosen to represent, approximately, the median uncertainty in mass estimates) and used 10,000 samples in our Monte-Carlo analysis (see Section 5.2) for those objects with mass estimates and with WISE photometry. Those flagged with discs were separated and fractions of object per mass bin were calculated. For those samples with the  HK cut (left panel) we focus on the region from 0 – 0.25  level as no objects in the C-ZYJHK samples had larger masses.
Figure 9: Mass functions (from isochrones). All samples were chosen to have 50 bins ranging from 0.0 to 0.6  (bin sizes were chosen to represent, approximately, the median uncertainty in mass estimates) and used 10,000 samples in our Monte-Carlo analysis (see Section 5.2) for those objects with mass estimates. The dashed lines show the best fits to a mass function ( for  , for 0.08  ), where is allowed to vary and the solid lines show the Kroupa mass function ( for  , for 0.08  ). All fits are scaled arbitrarily.

5.2 Disc fraction as a function of mass

Using the  color excess cut and W3 excess cut from Dawson et al. (2013, shown in Equation 3) we were able to identify possible discs in our candidate members.

(3)

For the L-ZYJHK sample we found 47/244 of the Upper Scorpius members with mass estimates had discs (19.32.5%). For the C-ZYJHK DR10 HK , C-ZYJHK DR10   HK and C-ZYJHK SV HK samples we found 14/46, 11/64 and 4/14 respectively (30.46.8%, 17.24.7% and 28.612.1%). For the C-ZYJHK DR10, C-ZYJHK DR10  and C-ZYJHK SV samples we found 21/167, 29/217 and 17/57 respectively (12.62.6%, 13.42.3% and 29.86.1%), all fractions are presented in Table 5. Combined these give a weighted mean disc fraction of 16.81.7%. All uncertainties are calculated as one-sigma uncertainties assuming binomial statistics (see tables 7 and 8 for a full break down of numbers). This value is broadly consistent with previously published disc fractions for VLM stars and brown dwarfs in this region (Jayawardhana et al., 2003; Scholz et al., 2007).

Sample Number of objects with discs Number of objects in sample Disc fraction
L-ZYJHK 47 244 19.32.5%
C-ZYJHK DR10 HK 14 46 30.46.8%
C-ZYJHK DR10   HK 11 64 17.24.7%
C-ZYJHK SV HK 4 14 28.612.1%
C-ZYJHK DR10 21 167 12.62.6%
C-ZYJHK DR10  29 217 13.42.3%
C-ZYJHK SV 17 57 29.86.1%
Table 5: The disc fractions found for the Upper Scorpius objects with WISE photometry.

We then used the mass estimates to plot disc fraction as a function of mass, where we account for uncertainties in the mass using a Monte-Carlo approach to draw samples from a Gaussian distribution for each object and then bin up the total samples for candidates, see Figure 8. For each object our Monte-Carlo approach draws a large number of samples from a Gaussian mass distribution (using the full-width-half-maximum as the larger of the lower and upper uncertainty band). Thus our binning process takes into account the uncertainties in mass estimates and shows the distribution as if there were a larger number of objects (note all disc fractions are carefully extrapolated in this process).

Figure 10: Proper motion excess diagrams for the C-ZYJHK DR10 HK sample. Excess is defined in Equation 4.

The disc fractions derived in that manner give two important insights. One is that the choice of the sample has a non-negligible effect on the outcome. This is particularly apparent from the right panel in Figure 8, which shows the sample without the  HK cut. Based on our assessment of contamination of these samples (see Section 3.3 and Section 4 the disc fractions derived from these samples have to be treated with caution. Contamination by background red giants might increase the fraction of objects with infrared excess in these samples, while contamination by background dwarf stars would reduce it. The strong fluctuations of disc fraction as a function of mass seen in these samples will be caused primarily by the varying influence of these contaminating samples, rather than actual changes in the disc fraction of VLMOs in Upper Scorpius. Discrepancies on the brown dwarf disc fraction presented in the literature may to a large extent be caused by differences in sample selection.

Second, the clean samples with the  HK cut do show that disc fractions within the sub-stellar domain increase with decreasing mass. This is seen in all three samples in the left panel of Figure 8. This trend was already stated in previous work, most notably by Luhman & Mamajek (2012). In our samples, the disc fraction for 0.05  is about 2-3 times larger than at  . This is solid evidence for disc lifetimes significantly exceeding 10 for low-mass brown dwarfs, confirming previous claims based on smaller samples (Riaz et al., 2009).

5.3 Mass function

Using the same MCMC process as in Section 5.2 we worked out the total number of objects in each mass bin. The mass function was then calculated using the total number of objects in each mass bin divided by the size of the mass bin. Uncertainties in number are assumed to be , and the uncertainty on is assumed to be the size of the mass bin. This means the uncertainty on is dominated by the uncertainty on the mass estimates. In Figure 8 we plot the mass function with the median uncertainties shown in yellow and the mass functions from Kroupa (2001), for  , for  , scaled arbitrarily to match our number of objects. Note that , and thus , are 10,000 times higher than our samples due to the Monte-Carlo samples used.

From Figure 8 we can see that both the C-ZYJHK DR10 samples are a good match to the Kroupa IMF (between and  , with best fits values of 0.45 and 0.38 for C-ZYJHK HK and C-ZYJHK   HK respectively). For the higher mass objects our brightness cuts impose mass cuts that are seen as the steep decrease in the number of objects with masses above 0.1  . At low masses (especially in the case of the UKIDSS science verification data) we see a gradual decrease in numbers of objects due to fainter objects being missed by UKIDSS (more so in the science verification data due to its preliminary shallow nature). Overall for the small area not affected by the brightness cuts or the faintness limit our data is consistent with the Kroupa IMF. A value of is also in line with previous determinations of the IMF in this region, e.g.  , Lodieu et al. (2013) find .

Sample Total objects in sample Non-outliers Outliers Expected Outliers from Gaussian
L-ZYJHK 415 347 68 57
C-ZYJHK DR10 HK 49 43 6 7
C-ZYJHK DR10   HK 68 59 9 10
C-ZYJHK SV HK 14 10 4 2
C-ZYJHK DR10 171 134 37 22
C-ZYJHK DR10  224 187 37 30
C-ZYJHK SV 58 44 14 8
Table 6: The number of outliers (those with excess greater than 2 sigma) compared to the number of expected outliers from a Gaussian distribution (again beyond 2 sigma).

The right panel in Figure 8 which shows the samples without the  HK cut again illustrates the effects of contamination on the mass function. In particular, there is a consistent ‘bump’ in the mass function just above 0.1  , which is most likely introduced by background objects. Echoing our previous comments, we would like to caution using this selection method for candidate members to derive population statistics for  , unless comprehensive spectroscopic characterisation is carried out to confirm youth and membership.

5.4 Proper motion outliers

In this subsection we explore the possibility that some of the VLMOs in Upper Scorpius have a dynamic history that deviates from the bulk of the population, for example, because they experienced an ejection in the early stages of their evolution, which might have stopped in-fall and constrained the mass. Ejections like that are part of a number of proposed formation scenarios for sub-stellar objects (e.g.   Whitworth et al., 2007).

To test for this possibility, we compare the proper motion for each target compared to its ten nearest neighbors (nearest in Right Ascension and Declination). An excess in proper motion was calculated for each object and is defined in Equation 4.

(4)

where is the median of the ten nearest neighbor objects proper motion in the Right Ascension/Declination direction, and is the standard deviation of the ten nearest neighbor objects in the Right Ascension/Declination direction. Thus excess is in units of sigma, where a value of zero would equate to an object having a proper motion component exactly equivalent to that of it’s neighbors.

This allowed us to flag any candidates with excesses beyond 2 sigma (defined by a circle in proper motion space). For the L-ZYJHK and L-HK only sample we found 68/415 and 31/175 outlier in our Upper Scorpius candidates. For the C-ZYJHK DR10 HK , C-ZYJHK DR10   HK and C-ZYJHK SV HK samples we found 6/49, 9/68 and 4/14 outliers in our Upper Scorpius candidates. For the C-ZYJHK DR10, C-ZYJHK DR10  and C-ZYJHK SVsamples we found 37/171, 37/224, 14/58 outliers in our Upper Scorpius candidates. Figure 10 shows the excess distribution in proper motion for two samples, with the 2 sigma circle drawn and outliers highlighted in red. The numbers are reported in Table 6.

Comparing the number of outliers with the outliers expected from a Gaussian distribution (last column in Table 6), there is currently no strong evidence for the presence of a population of brown dwarfs with kinematical properties distinct from the bulk population in the same area. For samples without the  HK cut, the number of outliers is again expected to be affected by contamination, as discussed throughout this paper. We note that a proper motion of 2 mas, currently the median uncertainty of our optimized proper motions, translates into a velocity in the plane of the sky of 1.3 kms, which is comparable with the typical velocity dispersion found in radial velocity surveys of young brown dwarfs (Joergens, 2006). Ejection velocities may in some cases be significantly beyond that level (Stamatellos & Whitworth 2009, Li et al. 2015), i.e.  the lack of proper motion outliers already provides useful limits for formation scenarios. This topic is an area where we expect future data releases from Gaia to provide improved constraints.

6 Concluding Remarks

The Gaia mission is a powerful new tool in understanding star forming regions due to its new and future astrometry. For the first time regions such as Upper Scorpius are no longer limited by the uncertainty on proper motion (and with future Gaia data releases unknown distances). As such we show that both age and mass estimates can be vastly improved, giving insight into the initial mass function and through published photometry from WISE, the disc fractions of these populations. However, we caution that future inferences on populations (such as for example, but in no way limited to, mass functions and disc fraction) will be very dependent on the selection criteria used. It may be that rigorous selection (i.e.  via full spectroscopic analysis) is required to really identify bona fide VLMOs and brown dwarfs from background objects and other contamination. This is especially valid for regions where reddening is important, but also applies, as shown in this paper, to regions free from extinction. Despite this we find that our mass functions (at least between ) are consistent with the Kroupa IMF, that the disc fraction among low-mass brown dwarfs () is substantially higher than in more massive objects. We note that proper motions from Gaia will give us an opportunity to detect objects that were dynamically ejected early on. If done correctly, with future Gaia data and full spectroscopic follow-up the potential for nearby star forming regions to advance our knowledge of low-mass, very-low mass and planetary mass objects is extremely exciting.

Acknowledgments

We would like to thank the anonymous referee whose careful reading of this paper and excellent comments were very welcome. We would like to thank Ogyen Verhagen, former undergraduate student at the University in St Andrews, whose final year project results motivated parts of this work. This work was supported in part by NSERC grants to RJ. This work has made use of data from the European Space Agency (ESA) mission Gaia (https://www.cosmos.esa.int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC, https://www.cosmos.esa.int/web/gaia/dpac/consortium). Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement. Specifically we use Gaia DR1 data (Gaia Collaboration et al. 2016a, Gaia Collaboration et al. 2016b) with more details available from Arenou et al. (2017), Lindegren et al. (2016), van Leeuwen et al. (2017), Carrasco et al. (2016) and Evans et al. (2017). We make use of data products from WISE (Wright et al., 2010), which is a joint project of the UCLA, and the JPLCIT, funded by NASA. The UKIDSS project is defined in Lawrence et al. (2007). UKIDSS uses the UKIRT Wide Field Camera (WFCAM; Casali et al. 2007). The photometric system is described in Hewett et al. (2006), and the calibration is described in Hodgkin et al. (2009). The pipeline processing and science archive are described in Hambly et al. (2008). We have used data from the DR10 data release and the science verification data, which are described in detail at http://wsa.roe.ac.uk. This work is based in part on services provided by the GAVO Data Center. This research has made use of the VizieR catalog access tool, CDS, Strasbourg, France (Ochsenbein et al., 2000). We acknowledge the use of data products from the HSOY catalog of Altmann et al. (2017), the GPS1 catalog of Tian et al. (2017), the UCAC5 catalog of Zacharias et al. (2017), the Pan-STARRS1 catalog of Chambers (2011) and the PPMXL database of Roeser et al. (2010). This research has made use of NASA’s Astrophysics Data System.

\software

Astropy (Astropy Collaboration et al., 2013), Ipython (Pérez & Granger, 2007), Matplotlib (Barrett et al., 2005; Hunter, 2007), Numpy (Jones et al., 2001; Oliphant, 2007), Scipy (Jones et al., 2001; Oliphant, 2007), Stilts (Taylor, 2006), Topcat (Taylor, 2005), tqdm (da Costa-Luis et al., 2017)

Appendix A SQL Queries

a.1 The ZYJHK Sample

ZYJHK sample query (Private communications with N. Lodieu see Lodieu, 2013). This query returned 2,653,897 sources from UKIDSS DR10 and 157,325 sources from the UKIDSS SV.

/* Start */
SELECT
    sourceID, ra, dec, zAperMag3, zAperMag3Err, yAperMag3, yAperMag3Err, jAperMag3, jAperMag3Err, hAperMag3, hAperMag3Err, k_1AperMag3, k_1AperMag3Err, muRa, muDec, sigMuRa, sigMuDec
FROM
    gcsSource
WHERE
       ra BETWEEN 232.0 AND 255.0
       AND dec BETWEEN -30.0 AND -15.0
       /* Bright saturation cut-offs */
       AND zaperMag3 > 11.3
       AND yaperMag3 > 11.5
       AND japerMag3 > 11.0
       AND haperMag3 > 11.30
       AND k_1aperMag3 > 9.90
       /* Limit merged passband selection to 1 arcsec */
       AND zXi BETWEEN -1.0 AND +1.0
       AND yXi BETWEEN -1.0 AND +1.0
       AND jXi BETWEEN -1.0 AND +1.0
       AND hXi BETWEEN -1.0 AND +1.0
       AND k_1Xi BETWEEN -1.0 AND +1.0
       AND zEta BETWEEN -1.0 AND +1.0
       AND yEta BETWEEN -1.0 AND +1.0
       AND jEta BETWEEN -1.0 AND +1.0
       AND hEta BETWEEN -1.0 AND +1.0
       AND k_1Eta BETWEEN -1.0 AND +1.0
       AND (jppErrBits < 131072)
       AND (hppErrBits < 131072)
       AND (k_1ppErrBits < 131072)
       /* Retain only point-like sources */
       AND (
        (
         ((zClass BETWEEN -2 AND -1) OR (zClassStat BETWEEN -3.0 AND +3.0))
        AND
         ((yClass BETWEEN -2 AND -1) OR (yClassStat BETWEEN -3.0 AND +3.0))
        AND
         ((jClass BETWEEN -2 AND -1) OR (jClassStat BETWEEN -3.0 AND +3.0))
        AND
         ((hClass BETWEEN -2 AND -1) OR (hClassStat BETWEEN -3.0 AND +3.0))
        AND
         ((k_1Class BETWEEN -2 AND -1) OR (k_1ClassStat BETWEEN -3.0 AND +3.0))
        )
        OR mergedClass BETWEEN -2 AND -1 OR mergedClassStat BETWEEN -3.0 AND +3.0
       )
       /* Retain only the best record when duplicated in an overlap region */
       AND (priOrSec = 0 OR priOrSec = frameSetID)
/* End */

a.2 The HK-only Sample

HK-only sample query (Private communications with N. Lodieu see Lodieu, 2013). This query returned 7,473,530 sources from UKIDSS DR10.

/* Start */
SELECT
    sourceID, ra, dec, zAperMag3, zAperMag3Err, yAperMag3, yAperMag3Err, jAperMag3, jAperMag3Err, hAperMag3, hAperMag3Err, k_1AperMag3, k_1AperMag3Err, muRa, muDec, sigMuRa, sigMuDec
FROM
    gcsSource
WHERE
       ra BETWEEN 232.0 AND 255.0
       AND dec BETWEEN -30.0 AND -15.0
       /* Bright saturation cut-offs */
       AND (zaperMag3 < -0.9e9 OR zaperMag3 > 11.3)
       AND (yaperMag3 < -0.9e9 OR yaperMag3 > 11.5)
       AND (japerMag3 < -0.9e9 OR japerMag3 > 11.0)
       AND haperMag3 > 11.30
       AND k_1aperMag3 > 9.90
       /* Limit merged passband selection to 1 arcsec */
       AND (zXi BETWEEN -1.0 AND +1.0 OR zXi < -0.9e9)
       AND (yXi BETWEEN -1.0 AND +1.0 OR yXi < -0.9e9)
       AND (jXi BETWEEN -1.0 AND +1.0 OR jXi < -0.9e9)
       AND hXi BETWEEN -1.0 AND +1.0
       AND k_1Xi BETWEEN -1.0 AND +1.0
       AND (zEta BETWEEN -1.0 AND +1.0 OR zEta < -0.9e9)
       AND (yEta BETWEEN -1.0 AND +1.0 OR yEta < -0.9e9)
       AND (jEta BETWEEN -1.0 AND +1.0 OR jEta < -0.9e9)
       AND hEta BETWEEN -1.0 AND +1.0
       AND k_1Eta BETWEEN -1.0 AND +1.0
       AND (hppErrBits < 131072)
       AND (k_1ppErrBits < 131072)
       /* Retain only point-like sources */
       AND (
        (
         ((zClass BETWEEN -2 AND -1) OR (zClassStat BETWEEN -3.0 AND +3.0) OR (zClass = -9999))
        AND
         ((yClass BETWEEN -2 AND -1) OR (yClassStat BETWEEN -3.0 AND +3.0) OR (yClass = -9999))
        AND
         ((jClass BETWEEN -2 AND -1) OR (jClassStat BETWEEN -3.0 AND +3.0) OR (jClass = -9999))
        AND
         ((hClass BETWEEN -2 AND -1) OR (hClassStat BETWEEN -3.0 AND +3.0))
        AND
         ((k_1Class BETWEEN -2 AND -1) OR (k_1ClassStat BETWEEN -3.0 AND +3.0))
        )
        OR mergedClass BETWEEN -2 AND -1 OR mergedClassStat BETWEEN -3.0 AND +3.0
       )
       /* Retain only the best record when duplicated in an overlap region */
       AND (priOrSec = 0 OR priOrSec = frameSetID)
/* End */

Appendix B The samples by number

In tables 7 and 8 we show the source counts (i.e.  the number of objects used from the original tables), the number of objects with proper motions in the HSOY, GPS1, UCAC, GCS, and PPMXL catalogs. We show the number of objects with a ‘most precise’ proper motion (in the ‘best pm’ column, i.e.  the smallest uncertainty as chosen from the HSOY, GPS1, UCAC and GCS catalogs, see Section 4). We show the source counts for those object that have WISE photometry, have a disc as indicated by the Dawson cuts (‘W1W2 disc’, ‘W3 disc’ and the combination of the two ‘disc’ column, see Section 5.2). The ‘mass est.’ column describes whether we were able to find a isochronal model that fit the data (see Section 5.1). The ‘USco’ column gives the number in each which conforms to our selection criteria for an Upper Scorpius member (see Section 4) and the numbers of objects in Upper Scorpius with discs, with a isochronal mass estimate and with both a disc and a isochronal mass estimate are shown in the last three columns. Tables 9 and 10 shows the column descriptions for the samples (tables available online in machine readable format).

{longrotatetable}
Flag Total HSOY GPS1 UCAC GCS PPMXL best pm WISE W1W2 Disc W3 Disc Disc mass est. USco USco+ disc USco+ Wise+ mass est. USco disc+ mass est.
In D11 28 12 21 0 28 23 28 12 1 0 1 28 20 1 8 1
In L11 91 63 77 36 91 83 91 62 6 14 17 91 88 16 59 16
In D13 108 53 80 1 108 86 108 52 11 5 11 108 104 11 49 11
In L13a ZYJHK 201 124 154 36 201 159 201 124 13 12 17 184 195 17 107 15
In L13a ZYJHK  120 73 84 23 120 96 120 72 6 4 8 116 108 7 59 7
In L13a HK 250 180 191 29 250 225 250 176 32 18 32 9 184 30 2 0
In L13a ZYJHK SV 70 38 55 5 70 61 70 38 6 4 8 66 64 7 33 7
In L13b 25 9 7 0 25 13 25 9 4 2 4 25 22 3 7 3
In D14 30 14 24 0 30 25 30 14 4 3 4 30 23 4 9 4
In GCS SV 122 79 103 38 122 110 122 77 10 17 21 117 112 20 70 20
In GCS DR10 716 469 549 126 716 609 716 462 66 55 83 453 610 79 244 47
L-ZYJHK 453 276 345 93 453 372 453 273 33 36 49 453 415 47 244 47
L-HK only 241 178 187 29 241 220 241 174 32 18 32 0 175 30 0 0
Total 716 469 549 126 716 609 716 462 66 55 83 453 610 79 244 47
Table 7: The source counts for the L-sample. For the L-sample numbers are also presented by original literature source catalog and how many sources are present in UKIDSS GCS DR10 and the UKIDSS GCS Science verification data. Tables available online in machine readable format.

Notes:

  • Source catalogs are as follows: D11 Dawson et al. (2011); L11 Lodieu et al. (2011); D13 Dawson et al. (2013); L13a Lodieu (2013); L13b Lodieu et al. (2013); D14 Dawson et al. (2014); and GCS Lawrence et al. (2007) - the UKIDSS GCS catalog.

  • ZYJHK, R, and HK here is as in Lodieu (2013) not as in our definitions of ZYJHK,     or  HK .

  • SV denotes UKIDSS GCS science verification data, DR10 denotes data release 10 data.

{longrotatetable}
The C-sample with the HK cut applied.
Flag Total HSOY GPS1 UCAC GCS PPMXL best pm WISE W1W2 Disc W3 Disc Disc mass est. USco USco+ disc USco+ Wise+ mass est. USco disc+ mass est.
C-ZYJHK DR10 HK 66 21 24 0 66 36 66 63 14 3 15 66 49 14 46 14
C-ZYJHK DR10   HK 77 24 35 0 77 42 77 73 12 1 12 77 68 11 64 11
C-ZYJHK SV  HK 17 12 14 0 17 15 17 17 5 0 5 17 14 4 14 4
C-ZYJHK Total HK 160 57 73 0 160 93 160 153 31 4 32 160 131 29 124 29
C-HK only 1526 870 807 225 1526 1032 1519 1395 180 98 214 1526 346 88 331 88
The C-sample without the HK cut applied.
Flag Total HSOY GPS1 UCAC GCS PPMXL best pm WISE W1W2 Disc W3 Disc Disc mass est. USco USco+ disc USco+ Wise+ mass est. USco disc+ mass est.
C-ZYJHK DR10 1305 1212 1177 1063 1305 1262 1305 1281 21 25 42 1305 171 21 167 21
C-ZYJHK DR10  811 688 728 546 811 752 811 797 24 19 33 811 224 29 217 29
C-ZYJHK SV 86 43 64 8 68 65 68 67 14 11 18 68 58 17 57 17
C-ZYJHK Total 2202 1943 1969 1617 2184 2079 2184 2145 59 55 93 2184 453 67 441 67
Table 8: The source counts for the C-sample. Tables available online in machine readable format.
\startlongtable
Column Name Description Unit UCD
UID Unique identifier (1) meta.id;meta.main
RAdeg Right Ascension in decimal degrees (J2000) deg pos.eq.ra;meta.main
DEdeg Declination in decimal degrees (J2000) deg pos.eq.dec;meta.main
Zmag UKIDSS Z magnitude mag phot.mag;em.IR
e_Zmag uncertainty in Zmag mag stat.error;em.IR
Ymag UKIDSS Y magnitude mag phot.mag;em.IR
e_Ymag uncertainty in Ymag mag stat.error;em.IR
Jmag UKIDSS J magnitude mag phot.mag;em.IR.J
e_Jmag uncertainty in Jmag mag stat.error;em.IR.J
Hmag UKIDSS H magnitude mag phot.mag;em.IR.H
e_Hmag uncertainty in Hmag mag stat.error;em.IR.H
Kmag K magnitude mag phot.mag;em.IR.K
e_Kmag uncertainty in Kmag mag stat.error;em.IR.K
pm Most precise proper motion mas/yr pos.pm
e_pm Uncertainty in pm mas/yr stat.error;pos.pm
r_pm Reference for pm (2) meta.bib
pmRA Most precise proper motion in RA mas/yr pos.pm;pos.eq.ra
e_pmRA Uncertainty in pmRA mas/yr stat.error;pos.pm;pos.eq.ra
pmDE Most precise proper motion in DE mas/yr pos.pm;pos.eq.dec
e_pmDE Uncertainty in pmDE mas/yr stat.error;pos.pm;pos.eq.dec
WISE Has WISE photometry (3) code.meta
Disk Flagged as having a disc (3) (4) code.meta
MassFit Best fit for mass solMass phys.mass
bMassFit Lower uncertainty bound in MassFit solMass stat.error;phys.mass
BMassFit Upper uncertainty bound in MassFit solMass stat.error;phys.mass
TeffFit Best fit for effective temperature K phys.temperature.effective
bTeffFit Lower uncertainty bound in TeffFit K stat.error;phys.temperature.effective
BTeffFit Upper uncertainty bound in TeffFit K stat.error;phys.temperature.effective
LumFit Best fit for luminosity solLum phys.luminosity
bLumFit Lower uncertainty bound in LumFit solLum stat.error;phys.luminosity
BLumFit Upper uncertainty bound in LumFit solLum stat.error;phys.luminosity
log(g)Fit Best fit for surface gravity [cm/s2] phys.gravity
blog(g)Fit Lower uncertainty bound in log(g)Fit [cm/s2] stat.error;phys.gravity
Blog(g)Fit Upper uncertainty bound in log(g)Fit [cm/s2] stat.error;phys.gravity
RadFit Best fit for radius solRad phys.size.radius
bRadFit Lower uncertainty bound in RadFit solRad stat.error;phys.size.radius
BRadFit Upper uncertainty bound in RadFit solRad stat.error;phys.size.radius
chilow Low fit stat.fit.chi2
chimid Mid fit stat.fit.chi2
chihi High fit stat.fit.chi2
WiseAndMass Has WISE photometry and a mass estimate meta.code
sigExpmRA Sigma excess in pmRA stat.value;pos.pm;pos.eq.ra
sigExpmDE Sigma excess in pmDE stat.value;pos.pm;pos.eq.dec
fnpmout Not an outlier in pm; 2 (3) meta.code.member
fpmout Is an outlier in pm; 2 (3) meta.code.member
11footnotetext: Of this merged detection as assigned by merge algorithm. ID is unique over entire WSA via program ID prefix.22footnotetext: hsoy = Hot Stuff for One Year (HSOY) catalog (Altmann et al. 2017); gps1 = Gaia-PS1-SDSS (GPS1) catalog (Tian et al. 2017); gcs = The UKIRT Infrared Deep Sky Survey (UKIDSS) catalog (Lawrence et al. 2007); and ucac = UCAC5: New Proper Motions Using Gaia DR1 catalog (Zacharias et al. 2017).33footnotetext: 1 = True, 2 = False44footnotetext: Dawson et al. (2013) WISE W1-W2 or WISE W3 disc.
Table 9: Column descriptions for C-sample tables for C-ZYJHK DR10, C-ZYJHK DR10  , C-ZYJHK SV, C-ZYJHK DR10 HK , C-ZYJHK DR10   HK , and C-ZYJHK SV HK and tables are available online in machine readable format.
\startlongtable
Column Name Description Unit UCD
UID Unique ID from catalog creation meta.id;meta.main
RAdeg Right Ascension in decimal degrees (J2000) deg pos.eq.ra;meta.main
DEdeg Declination in decimal degrees (J2000) deg pos.eq.dec;meta.main
Coord Source of main coordinates meta.bib
Cat Catalog contained in (1) meta.bib
Zmag Most precise Z magnitude mag phot.mag;em.opt.Z
r_Zmag Reference for Zmag (1) meta.bib
e_Zmag uncertainty in Zmag mag stat.error;em.opt.Z
Ymag Most precise Y magnitude mag phot.mag;em.opt.Y
r_Ymag Reference for Ymag (1) meta.bib
e_Ymag uncertainty in Ymag mag stat.error;em.opt.Y
Jmag Most precise J magnitude mag phot.mag;em.IR.J
r_Jmag Reference for Jmag (1) meta.bib
e_Jmag uncertainty in Jmag mag stat.error;phot.mag;em.IR.J
Hmag Most precise H magnitude mag phot.mag;em.IR.H
r_Hmag Reference for Hmag (1) meta.bib
e_Hmag uncertainty in Hmag mag stat.error;phot.mag;em.IR.H
Kmag Most precise K magnitude mag phot.mag;em.IR.K
r_Kmag Reference for Kmag (1) meta.bib
e_Kmag uncertainty in Kmag mag stat.error;phot.mag;em.IR.K
SpTCat spt from source catalog src.spType
r_SpTCat Reference for SpTcat (1) meta.bib
MassCat Mass from source catalog solMass phys.mass
r_MassCat Reference for MassCat (1) meta.bib
fZYJHK Flag whether source ahs valid ZYJHK photometry (2) meta.code
fHKonly Flag that source only has HK photometry (2) meta.code
pm Most precise proper motion mas/yr pos.pm
e_pm Uncertainty in pm mas/yr stat.error;pos.pm
r_pm Reference for pm (3) meta.bib
pmRA Most precise proper motion in RA mas/yr pos.pm;pos.eq.ra
e_pmRA Uncertainty in pmRA mas/yr stat.error;pos.pm;pos.eq.ra
pmDE Most precise proper motion in DE mas/yr pos.pm;pos.eq.dec
e_pmDE Uncertainty in pmDE mas/yr stat.error;pos.pm;pos.eq.dec
MassFit Best fit for mass solMass phys.mass
bMassFit Lower uncertainty bound in MassFit solMass stat.error;phys.mass
BMassFit Upper uncertainty bound in MassFit solMass stat.error;phys.mass
TeffFit Best fit for effective temperature K phys.temperature.effective
bTeffFit Lower uncertainty bound in TeffFit K stat.error;phys.temperature.effective
BTeffFit Upper uncertainty bound in TeffFit K stat.error;phys.temperature.effective
LumFit Best fit for luminosity solLum phys.luminosity
bLumFit Lower uncertainty bound in LumFit solLum stat.error;phys.luminosity
BLumFit Upper uncertainty bound in LumFit solLum stat.error;phys.luminosity
log(g)Fit Best fit for surface gravity [cm/s2] phys.gravity
blog(g)Fit Lower uncertainty bound in log(g)Fit [cm/s2] stat.error;phys.gravity
Blog(g)Fit Upper uncertainty bound in log(g)Fit [cm/s2] stat.error;phys.gravity
RadFit Best fit for radius solRad phys.size.radius
eRadFitL Lower uncertainty bound in RadFit solRad stat.error;phys.size.radius
eRadFitU Upper uncertainty bound in RadFit solRad stat.error;phys.size.radius
chilow Low fit stat.fit.chi2
chimid Mid fit stat.fit.chi2
chihi High fit stat.fit.chi2
Disk Flagged as having a disc (2) (4) meta.code
WiseAndMass Has WISE photometry and a mass estimate (2) meta.code
sigExpmRA Sigma excess in pmRA stat.value;pos.pm;pos.eq.ra
sigExpmDE Sigma excess in pmDE stat.value;pos.pm;pos.eq.dec
fnotPMout Not an outlier in pm; 2 (2) meta.code.member
fPMout Is an outlier in pm; 2 (2) meta.code.member
11footnotetext: GCSSV = from UKIDSS Science verification (Lawrence et al. 2007); GCSDR10 = from UKIDSS DR10 (Lawrence et al. 2007); D11 = from Dawson et al. (2011)]; L11 = From Lodieu et al. (2011); D13 = from Dawson et al. (2013); L13a = from Lodieu (2013): L13a_HK from HK sample, L13a_ZYJHK_SV from the ZYJHK UKIDSS Science verification sample,L13a_ZYJHK_red from the ZYJHK UKIDSS DR10 affected by reddening sample, L13a_ZYJHK_nored from the ZYJHK UKIDSS DR10 devoid of reddening sample; L13b = from Lodieu et al. (2013); D14 = from Dawson et al. (2014).22footnotetext: 1 = True, 2 = False33footnotetext: hsoy = Hot Stuff for One Year (HSOY) catalog (Altmann et al. 2017); gps1 = Gaia-PS1-SDSS (GPS1) catalog (Tian et al. 2017); gcs = The UKIRT Infrared Deep Sky Survey (UKIDSS) catalog (Lawrence et al. 2007); and ucac = UCAC5: New Proper Motions Using Gaia DR1 catalog (Zacharias et al. 2017).44footnotetext: From isochrones (i.e have ZYJHKW1W2 photometry).
Table 10: Column descriptions for the L-sample tables (L-ZJYHK and L-HKonly). Tables available online in machine readable format.

References

  • Altmann et al. (2017) Altmann, M., Roeser, S., Demleitner, M., Bastian, U., & Schilbach, E. 2017, A&A, 600, L4
  • Anglada-Escudé et al. (2016) Anglada-Escudé, G., Amado, P. J., Barnes, J., et al. 2016, Nature, 536, 437
  • Arenou et al. (2017) Arenou, F., Luri, X. fand Babusiaux, C., Fabricius, C., et al. 2017, A&A, 599, A50
  • Astropy Collaboration et al. (2013) Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558, A33
  • Baraffe et al. (1998) Baraffe, I., Chabrier, G., Allard, F., & Hauschildt, P. H. 1998, A&A, 337, 403
  • Baraffe et al. (2015) Baraffe, I., Homeier, D., Allard, F., & Chabrier, G. 2015, A&A, 577, A42
  • Bardalez Gagliuffi et al. (2013) Bardalez Gagliuffi, D. C., Burgasser, A. J., & Gelino, C. R. 2013, Mem. Soc. Astron. Italiana, 84, 1041
  • Barrett et al. (2005) Barrett, P., Hunter, J., Miller, J. T., Hsu, J.-C., & Greenfield, P. 2005, in Astronomical Society of the Pacific Conference Series, Vol. 347, Astronomical Data Analysis Software and Systems XIV, ed. P. Shopbell, M. Britton, & R. Ebert, 91
  • Berta-Thompson et al. (2015) Berta-Thompson, Z. K., Irwin, J., Charbonneau, D., et al. 2015, Nature, 527, 204
  • Bonnell et al. (2007) Bonnell, I. A., Larson, R. B., & Zinnecker, H. 2007, Protostars and Planets V, 149
  • Carrasco et al. (2016) Carrasco, J. M., Evans, D. W., Montegriffo, P., et al. 2016, A&A, 595, A7
  • Casali et al. (2007) Casali, M., Adamson, A., Alves de Oliveira, C., et al. 2007, A&A, 467, 777
  • Chabrier et al. (2000) Chabrier, G., Baraffe, I., Allard, F., & Hauschildt, P. 2000, ApJ, 542, 464
  • Chambers (2011) Chambers, K. C. 2011, in Bulletin of the American Astronomical Society, Vol. 43, American Astronomical Society Meeting Abstracts #218, 113.01
  • Cook et al. (2016) Cook, N. J., Pinfield, D. J., Marocco, F., et al. 2016, MNRAS, 457, 2192
  • Cook et al. (2017) —. 2017, MNRAS, 467, 5001
  • da Costa-Luis et al. (2017) da Costa-Luis, C., L., S., Mary, H., et al. 2017, tqdm/tqdm: tqdm v4.19.1 stable, , , doi:10.5281/zenodo.1001783. https://doi.org/10.5281/zenodo.1001783
  • Dawson et al. (2011) Dawson, P., Scholz, A., & Ray, T. P. 2011, MNRAS, 418, 1231
  • Dawson et al. (2013) Dawson, P., Scholz, A., Ray, T. P., et al. 2013, MNRAS, 429, 903
  • Dawson et al. (2014) —. 2014, MNRAS, 442, 1586
  • de Zeeuw et al. (1999) de Zeeuw, P. T., Hoogerwerf, R., de Bruijne, J. H. J., Brown, A. G. A., & Blaauw, A. 1999, AJ, 117, 354
  • Dittmann et al. (2017) Dittmann, J. A., Irwin, J. M., Charbonneau, D., et al. 2017, Nature, 544, 333
  • Dye et al. (2006) Dye, S., Warren, S. J., Hambly, N. C., et al. 2006, MNRAS, 372, 1227
  • Evans et al. (2017) Evans, D. W., Riello, M., De Angeli, F., et al. 2017, A&A, 600, A51
  • Fang et al. (2017) Fang, Q., Herczeg, G. J., & Rizzuto, A. 2017, ApJ, 842, 123
  • Gaia Collaboration et al. (2016a) Gaia Collaboration, Prusti, T., de Bruijne, J. H. J., et al. 2016a, A&A, 595, A1
  • Gaia Collaboration et al. (2016b) Gaia Collaboration, Brown, A. G. A., Vallenari, A., et al. 2016b, A&A, 595, A2
  • Gillon et al. (2017) Gillon, M., Triaud, A. H. M. J., Demory, B.-O., et al. 2017, Nature, 542, 456
  • Grankin et al. (2008) Grankin, K. N., Bouvier, J., Herbst, W., & Melnikov, S. Y. 2008, A&A, 479, 827
  • Hambly et al. (2008) Hambly, N. C., Collins, R. S., Cross, N. J. G., et al. 2008, MNRAS, 384, 637
  • Hewett et al. (2006) Hewett, P. C., Warren, S. J., Leggett, S. K., & Hodgkin, S. T. 2006, MNRAS, 367, 454
  • Hodgkin et al. (2009) Hodgkin, S. T., Irwin, M. J., Hewett, P. C., & Warren, S. J. 2009, MNRAS, 394, 675
  • Houk & Smith-Moore (1988) Houk, N., & Smith-Moore, M. 1988, Michigan Catalogue of Two-dimensional Spectral Types for the HD Stars. Volume 4, Declinations -26.0 deg to -12.0 deg
  • Hunter (2007) Hunter, J. D. 2007, Computing in Science & Engineering, 9, 90. http://aip.scitation.org/doi/abs/10.1109/MCSE.2007.55
  • Jayawardhana et al. (2003) Jayawardhana, R., Ardila, D. R., Stelzer, B., & Haisch, Jr., K. E. 2003, AJ, 126, 1515
  • Joergens (2006) Joergens, V. 2006, A&A, 446, 1165
  • Jones et al. (2001) Jones, E., Oliphant, T., Peterson, P., & Others. 2001, SciPy: Open source scientific tools for Python, , . http://www.scipy.org/
  • Kenyon & Hartmann (1995) Kenyon, S. J., & Hartmann, L. 1995, ApJS, 101, 117
  • Kroupa (2001) Kroupa, P. 2001, MNRAS, 322, 231
  • Lawrence et al. (2007) Lawrence, A., Warren, S. J., Almaini, O., et al. 2007, MNRAS, 379, 1599
  • Li et al. (2015) Li, Y., Kouwenhoven, M. B. N., Stamatellos, D., & Goodwin, S. P. 2015, ApJ, 805, 116
  • Lindegren et al. (2016) Lindegren, L., Lammers, U., Bastian, U., et al. 2016, A&A, 595, A4
  • Lodieu (2013) Lodieu, N. 2013, MNRAS, 431, 3222
  • Lodieu et al. (2013) Lodieu, N., Dobbie, P. D., Cross, N. J. G., et al. 2013, MNRAS, 435, 2474
  • Lodieu et al. (2011) Lodieu, N., Dobbie, P. D., & Hambly, N. C. 2011, A&A, 527, A24
  • Luhman (2012) Luhman, K. L. 2012, ARA&A, 50, 65
  • Luhman et al. (2003) Luhman, K. L., Briceño, C., Stauffer, J. R., et al. 2003, ApJ, 590, 348
  • Luhman & Mamajek (2012) Luhman, K. L., & Mamajek, E. E. 2012, ApJ, 758, 31
  • Michalik et al. (2014) Michalik, D., Lindegren, L., Hobbs, D., & Lammers, U. 2014, A&A, 571, A85
  • Ochsenbein et al. (2000) Ochsenbein, F., Bauer, P., & Marcout, J. 2000, A&AS, 143, 23
  • Oliphant (2007) Oliphant, T. E. 2007, Computing in Science & Engineering, 9, 10. http://aip.scitation.org/doi/abs/10.1109/MCSE.2007.58
  • Pascucci et al. (2016) Pascucci, I., Testi, L., Herczeg, G. J., et al. 2016, ApJ, 831, 125
  • Pecaut & Mamajek (2016) Pecaut, M. J., & Mamajek, E. E. 2016, MNRAS, 461, 794
  • Pecaut et al. (2012) Pecaut, M. J., Mamajek, E. E., & Bubar, E. J. 2012, ApJ, 746, 154
  • Pérez & Granger (2007) Pérez, F., & Granger, B. E. 2007, Computing in Science and Engineering, 9, 21. http://ipython.org
  • Preibisch et al. (2002) Preibisch, T., Brown, A. G. A., Bridges, T., Guenther, E., & Zinnecker, H. 2002, AJ, 124, 404
  • Riaz et al. (2009) Riaz, B., Lodieu, N., & Gizis, J. E. 2009, ApJ, 705, 1173
  • Roeser et al. (2010) Roeser, S., Demleitner, M., & Schilbach, E. 2010, AJ, 139, 2440
  • Scholz et al. (2012) Scholz, A., Jayawardhana, R., Muzic, K., et al. 2012, ApJ, 756, 24
  • Scholz et al. (2006) Scholz, A., Jayawardhana, R., & Wood, K. 2006, ApJ, 645, 1498
  • Scholz et al. (2007) Scholz, A., Jayawardhana, R., Wood, K., et al. 2007, ApJ, 660, 1517
  • Siess et al. (2000) Siess, L., Dufour, E., & Forestini, M. 2000, A&A, 358, 593
  • Slesnick et al. (2008) Slesnick, C. L., Hillenbrand, L. A., & Carpenter, J. M. 2008, ApJ, 688, 377
  • Stamatellos & Whitworth (2009) Stamatellos, D., & Whitworth, A. P. 2009, MNRAS, 392, 413
  • Taylor (2005) Taylor, M. B. 2005, in Astronomical Society of the Pacific Conference Series, Vol. 347, Astronomical Data Analysis Software and Systems XIV, ed. P. Shopbell, M. Britton, & R. Ebert, 29
  • Taylor (2006) Taylor, M. B. 2006, in Astronomical Society of the Pacific Conference Series, Vol. 351, Astronomical Data Analysis Software and Systems XV, ed. C. Gabriel, C. Arviset, D. Ponz, & S. Enrique, 666
  • Testi et al. (2015) Testi, L., Skemer, A., Henning, T., et al. 2015, ApJ, 812, L38
  • Tian et al. (2017) Tian, H.-J., Gupta, P., Sesar, B., et al. 2017, ArXiv e-prints, arXiv:1703.06278
  • van Leeuwen et al. (2017) van Leeuwen, F., Evans, D. W., De Angeli, F., et al. 2017, A&A, 599, A32
  • Whitworth et al. (2007) Whitworth, A., Bate, M. R., Nordlund, Å., Reipurth, B., & Zinnecker, H. 2007, Protostars and Planets V, 459
  • Wilking et al. (2004) Wilking, B. A., Meyer, M. R., Greene, T. P., Mikhail, A., & Carlson, G. 2004, AJ, 127, 1131
  • Wright et al. (2010) Wright, E. L., Eisenhardt, P. R. M., Mainzer, A. K., et al. 2010, AJ, 140, 1868
  • Zacharias et al. (2017) Zacharias, N., Finch, C., & Frouard, J. 2017, AJ, 153, 166
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
""
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
   
Add comment
Cancel
Loading ...
143026
This is a comment super asjknd jkasnjk adsnkj
Upvote
Downvote
""
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters
Submit
Cancel

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test
Test description