Mem SAIt Vol 85 637ccopy SAIt 2014 Memorie della

Gaia limits for brown dwarf studies

L M Sarro12 D Barrado3 C Carrion1 and J A Caballero3

1 Dpto Inteligencia Artificial ETSI Informatica Universidad Nacional de Educacion aDistancia (UNED) Juan del Rosal 16 28040 Madrid Spain e-mail lsbdiaunedes

2 Dpto Astrofısica Centro de Astrobiologıa (INTA-CSIC) ESAC Campus PO Box 78E-28691 Villanueva de la Canada Spain

Abstract In this work we present a study of the expected properties of the Gaia sample ofultracool dwarfs (UCDs) This is defined preliminary (at a stage where we do not have realGaia data yet) by analysing the detectability by Gaia of known UCDs in the Dwarf ArchiveWe complement the observations listed in the Dwarf Archive (J- H- and K- band magni-tudes) with SDSS magnitudes and compare them with the BT-Settl model library in orderto assess detectability We also discuss the fraction of UCDs that will have accompanyingRVS spectra in the Calcium triplet wavelength region

Key words Methods data analysis ndash Stars brown dwarfs ndash Stars fundamental parameters

1 Introduction

Gaia was successfully launched on 14 Dec2013 and at the time of this writing the com-missioning phase is coming to an end with thefinal set of EPSL (Ecliptic Pole Scanning Law)orbits From this period we have learned thatcondensations in the mirrors and stray lightwill slightly degrade the instrument perfor-mances especially in the RVS (Radial VelocitySpectrometer) data Despite these uncertain-ties this is a good time to review the expectedcharacteristics of the Gaia data base with re-spect to its brown dwarf content

To that end we review the nominal instru-ment performances at the time of launch inSect 2 and apply the resulting limits to theDwarf Archives1 set of brown dwarfs to deter-mine the detectability of each source thereinand assess the expected properties of Gaiabrown dwarf data set (Sect 3)

1 httpdwarfarchivesorg

2 Gaia instrument performances

The nominal instrument performances canbe found in the official Gaia mission website2 These performances come from the GaiaMission Critical Design Review in 2011 andthus do not incorporate the performance degra-dation mentioned in the previous section andstill under quantification Frequent updates onthe status of the commissioning phase canbe found at httpblogsesaintgaiaand a more complete and technical report is ex-pected at the end of the phase

Sarro et al (2013) detailed the technicalaspects of the software module dedicated tothe detection and characterization of browndwarfs in the Apsis (Bailer-Jones et al 2013)chain of the Gaia data processing and anal-ysis pipeline The aspects of instrument per-formance most directly linked to the proper-

2 httpwwwcosmosesaintwebgaiascience-performance

638 Sarro et al Gaia limits for brown dwarf studies

ties of the Gaia ultracool dwarf sample (here-after GUCDS) are (i) the dependence of the de-tectability of brown dwarfs with distance andspectral type (ii) the expected astrometric un-certainties for very red objects in the faint de-tection limit and (iii) the expected signal-to-noise ratio for Red Photometer spectra at thatlimit

The first aspect in the list was addressed inSarro et al (2013) and depends on the assumedrelationship between spectral type agegravityand absolute G magnitude In that work weused BT-Settl models Allard et al (2012) andthe nominal G band transmission to infer amaximum distance at which a given ultracooldwarf (UCD) could be detected In terms of theabsolute I band magnitude we can summariseit with a few values The maximum distancefor a M6ndash7 dwarf is of the order of 190 pcfor spectral type M8ndash9 120 pc for spectraltype L3ndash4 35 pc We refer the reader to thatwork and Caballero et al (2008) for details onthe derivation The expected astrometric uncer-tainty for an M6 dwarf at the detection limitG = 20 mag is 100 microas and is not expected tochange significantly for later spectral types

The nominal RVS magnitude limit wasGRVS asymp 16 mag Ultracool dwarfs have typicalV minus I colour indices in the range between 4 and75 mag which translate into GminusGRVS between175 and 21 mag according to the parametriza-tion in Jordi et al (2010) and the set of BT-Settlmodels used in this work Assuming the localspatial densities of Caballero et al (2008) itimplies that RVS spectra would be availablefor a few hundred UCDs (mostly of the ear-liest spectral types) down to G = 181 magas shown in Fig 4 of Sarro et al (2013)Unfortunately the RVS spectrometer measure-ments are the ones most affected by the straylight detected during the commissioning phaseand a one magnitude shift in instrument perfor-mance is a reasonable estimate at the time ofwriting therefore severely reducing the num-ber of potentially available RVS spectra of ul-tracool dwarfs Figure 1 shows the BT-Settlspectra in the RVS range for five temperaturesbetween 500 and 2500 K At present there islittle hope for the utility of RVS spectra of

UCDs for the estimation of physical parame-ters

3 Properties of the (expected) Gaiabrown dwarf sample

In this Section we aim at predicting whatknown UCDs will be detected by Gaia andwhat would be the distribution of the GUCDSif the sample of known UCDs were completedown to the nominal Gaia detection limit Inorder to do so we focus our study in the DwarfArchives catalogue of 1281 L T and Y dwarfs

The first step in our analysis consists inestimating the apparent G magnitude of theUCDs in the Dwarf Archives sample The Gaiadetection limit is determined by the apparent Gmagnitude with a threshold nominally set at G= 20 mag although on-going studies may lowerthe magnitude limit to G = 21 mag Given thepredominantly red spectral energy distributionof UCDs and the relative availability of pho-tometric measurements in the SDSS system ofbandpasses with respect to other systems weattempt to parameterise the apparent G magni-tude in terms of the i and z bands We use theBT-Settl models to define a parametrization de-fined by

G minus i = 013 + 143(i minus z) minus 120(i minus z)2 +

035(i minus z)3 minus 006(i minus z)4 (1)

We subsequently search the SDSS3 cata-logue to complement the Dwarf Archives database with i and z photometric measurementsDwarf Archives include infrared photometry inthe J H and K bands but these do not producenarrow monotonous correlations with the op-tical broad band filter that defines the G mag-nitude We find potential counterparts to 948sources in the Dwarf Archives catalogue 587below 2 arcsec

Figures 2 show the distribution of the sam-ple of UCDs in the Dwarf Archives cataloguewith cross-match candidates within 2 arcsecin several colour-colour diagrams In eachplot we show the photometric measurements

3 httpwwwsdssorg

Sarro et al Gaia limits for brown dwarf studies 639

8000 8200 8400 8600 8800 9000

01

23

45

λ

Fλ

Fig 1 BT-Settl spectra of UCDs of effective temperatures equal to 500 K (blue) 1000 K (green) 1500 K(yellow) 2000 K (orange) and 2500 K (red) The vertical dashed lines represent the blue and red limits ofthe Gaia RVS spectrograph All five spectra have been degraded to the nominal RVS spectral resolution

with corresponding error bars in a gray scaleand the collection of BT-Settl models withcolours The colour code for the error barsgoes from black for small cross-match separa-tions to white at the largest allowed separationof 2 arcsec The BT-Settl models on the otherhand are coded according to the correspond-ing effective temperature as shown in the colorscale to the right of the plots

These figures should be compared to thepreliminary results with SDSS data by Hawleyet al (2002) Their sample extends to spectraltypes significantly earlier than ours but thereis a region of overlap between both samplesin the sense that their cooler sources corre-spond to our hotter ones Thus we have ex-tended the dataset of UCDs with complete pho-tometric measurements in riz and JHK bands

to fainter cooler spectral types Our results co-incide with those of Hawley et al (2002) in theregion of overlap but show significant discrep-ancies with the BT-Settl models in all diagramsinvolving the r-band magnitude

This data set of sources with completemeasurements in SDSS and 2MASS bands canbe used to estimate the corresponding appar-ent G magnitudes using the parametrizationin Eq 1 and predict which of these DwarfArchives UCDs will be brighter than the Gaiadetection threshold We find a total number of124 Dwarf Archive sources brighter than G =20 mag and 320 brighter than G = 21 mag Ifwe apply a simplistic correction to account forthe fraction of Dwarf Archive sources with-out counterpart available in the SDSS pass-bands (43 of the Dwarf Archive data set)

640 Sarro et al Gaia limits for brown dwarf studies

minus1 0 1 2 3

minus1

5minus

10

minus0

50

00

51

01

52

0

JminusH

Hminus

K

500

1000

1500

2000

2500

3000

3500

0 2 4 6 8

minus2

02

46

iminusz

rminusi

500

1000

1500

2000

2500

3000

3500

0 2 4 6 8 10

02

46

iminusJ

iminusZ

500

1000

1500

2000

2500

3000

3500

minus2 minus1 0 1 2 3 4 5

minus1

0minus

05

00

05

10

15

20

rminusi

JminusH

500

1000

1500

2000

2500

3000

3500

Fig 2 Colour-colour diagrams for the Dwarf Archives UCDs with riz magnitudes obtained fromthe SDSS database BT-Settl models are superimposed with a colour code that represents effectivetemperatures according to the scales to the right of each plot

these numbers transform to 288 sources downto G = 20 mag and 744 sources down to G =21 mag The first figure is similar to the es-timate by Smart (2014) while the second issignificantly larger than his but still less thanexpected given the volume factor expansionin going from G = 20 to 21 mag We believethat the difference between Smartrsquos estimatesand ours arises because he is using the calibra-tion between Gaia and SDSS by Jordi (2010)which is based on the spectral library of O-to M-type dwarf and giant stars by Pickles(1998) Therefore it is of limited applicabilityfor the coolest spectral types L and T which

would explain the discrepancy when going tothe fainter G = 21 mag limit

Figure 3 shows the number of sources inapparent G bins estimated from the local densi-ties by Caballero et al (2008) (gray) and fromthe Dwarf Archives catalogue (corrected forthe fraction of sources without available crossmatches in the SDSS bands violet) It seems toindicate that either the local densities result in a03ndash05 dex overestimate of the counts per binor the Dwarf Archives catalogue is incompleteby that same factor or a mix of both effectsmay be present at the same time According tothe work presented by Smart (2014) there is

Sarro et al Gaia limits for brown dwarf studies 641

Fig 3 Predicted UCD counts per magnitude bin inthe GUCDS according to the BT-Settl models and(i) the local spatial densities from Caballero et al(2008) (gray) and (ii) the Dwarfs Archive samplecorrected for the fraction of entries measurementsin the SDSS photometric system

evidence supporting an incompleteness factorof at least two in the Dwarf Archive under theassumption of isotropy and local homogeneityin the local distribution of UCDs which wouldexplain part of the discrepancy

If we convert the i minus J colour index intoa spectral type using the calibration by Hawleyet al (2002) we obtain the distribution of typesshown in Fig 4

If we apply a similar analysis to a set ofnearby star forming regions and clusters as-suming the corresponding set of distances andages and the BT-Settl family of synthetic spec-tra we can estimate the I-band magnitude thatcorresponds to the Gaia detection limit of G =20 mag and the corresponding values of masseffective temperature and spectral type Theresults are summarised in Table 1 Gaia willdetect a significant number of UCDs with spec-tral types from mid-M to late M and few L0Note that in the case of very young nearby as-sociations the masses will be close or even be-low the deuterium burning mass limit Thussome ldquoisolated planetary-mass objectsrdquo mightbe detected

M1 M5 M9 L3 L7 T1 T5

Spectral Type

Cou

nts

050

100

150

200

Fig 4 Distribution of spectral types of sources inthe Dwarf Archive that can be detected by Gaia(with estimated G magnitudes brighter that 20 mag)

4 Conclusions

We have used the Gaia pre-launch perfor-mances as assessed during the Critical DesignReview carried in 2011 to infer what can beexpected of the GUCDS in particular the dis-tribution of spectral types and the detectabilityof known UCDs in the Dwarfs Archive database In the process of inferring apparent Gmagnitudes for these Dwarf Archives UCDswe have extended the set objects with observa-tions in both riz and JHK bands well beyondthe coolest examples in the previous work byHawley et al (2002) We obtained SDSS pho-tometry for 587 UCDs in the Dwarfs Archivedatabase reaching spectral types in the T se-quence

As in Sarro et al (2013) these conclusionsrely on many uncertain assumptions They de-pend on the final performance of the satelliteinstruments in the first place but also criti-cally on the calibration of the relationship be-tween magnitudes in different passbands ob-tained from the BT-Settl models or the calibra-tion of spectral types with the iminusJ colour indexIn particular we show that the predictions fromthe Dwarfs Archive and from local spatial den-sities disagree by a factor at least two We rec-ommend caution when using these estimates

642 Sarro et al Gaia limits for brown dwarf studies

Table 1 An estimate of the coolest spectral type detectable with Gaia in several young nearbystar forming regions and clusters

Region d Age Dimmest MI Mass Teff Sp type[pc] [Myr] [mag] [MJup] [K]

ρ Ophiuchi 120ndash145 1 1305 10 2200 L0Taurus 140 1ndash2 1270 16 2400 M9Serpens 260 2 1138 27 2600 M8Chamaeleon I II 140 1ndash3 1270 18 2400 M9Lupus I II III 140 1ndash3 1270 18 2400 M9IC 348 385 2ndash4 1054 37 2700 M7Trumpler 37 800 4 783 201 3200 M4σ Orionis 400 2ndash4 1045 42 2800 M6Collinder 69 400 5ndash10 1045 49 2800 M6Upper Scorpius 145 5 1262 21 2400 M8NGC 7160 800 10 783 317 3400 M3IC 2391 175 30ndash55 1221 45 2600 M7IC 2602 160 30ndash50 1241 42 2600 M7IC 4665 350 45 1077 91 3000 M5α Persei 185 80 1209 60 2700 M7Blanco 1 270 100 1131 83 2900 M6Pleiades 150 125 1255 58 2600 M7Hyades 50 600 1499 37 2200 L0Praesepe 187 800 1207 72 2800 M6

since at the time of writing this manuscript theimpact of condensations and stray light on thefinal performances of the Gaia instruments isstill uncertain On the other hand the on-goingstudy about a potential extension of the Gaialimiting magnitude down to G = 21 mag wouldhave an important and positive effect on thecompleteness of the GUCDS

Acknowledgements The authors wish to acknowl-edge the Coordination Unit 2 of the Gaia DPAC forthe use of the GOG simulator and Rosanna Sordofor their kind help and guidance with the simula-tion of both synthetic and empirical spectra at theGaia instrumental characteristics This research hasbeen supported by the Spanish Ministry of Sciencethrough grants AyA2011-24052 AYA2011-30147-C03-03 and AYA2012-38897-C02-01

References

Allard F Homeier D Freytag B Jun 2012Royal Society of London PhilosophicalTransactions Series A 370 2765

Bailer-Jones CAL Andrae R Arcay Bet al Nov 2013 AampA 559 A74

Caballero JA Burgasser AJ Klement RSep 2008 AampA 488 181

Hawley SL Covey KR Knapp GR et alJun 2002 AJ 123 3409

Jordi C Gebran M Carrasco JM et al Nov2010 AampA 523 A48+

Jordi C May 2012 httpwwwrssdesaintcslivelinkopen2760608

Pickles A J July 1998 PASP 110 863Sarro LM Berihuete A Carrion C et al

2013 AampA 550 A44Smart R L 2014 MmSAI 85 649

640 Sarro et al Gaia limits for brown dwarf studies

minus1 0 1 2 3

minus1

5minus

10

minus0

50

00

51

01

52

0

JminusH

Hminus

K

500

1000

1500

2000

2500

3000

3500

0 2 4 6 8

minus2

02

46

iminusz

rminusi

500

1000

1500

2000

2500

3000

3500

0 2 4 6 8 10

02

46

iminusJ

iminusZ

500

1000

1500

2000

2500

3000

3500

minus2 minus1 0 1 2 3 4 5

minus1

0minus

05

00

05

10

15

20

rminusi

JminusH

500

1000

1500

2000

2500

3000

3500

Fig 2 Colour-colour diagrams for the Dwarf Archives UCDs with riz magnitudes obtained fromthe SDSS database BT-Settl models are superimposed with a colour code that represents effectivetemperatures according to the scales to the right of each plot

these numbers transform to 288 sources downto G = 20 mag and 744 sources down to G =21 mag The first figure is similar to the es-timate by Smart (2014) while the second issignificantly larger than his but still less thanexpected given the volume factor expansionin going from G = 20 to 21 mag We believethat the difference between Smartrsquos estimatesand ours arises because he is using the calibra-tion between Gaia and SDSS by Jordi (2010)which is based on the spectral library of O-to M-type dwarf and giant stars by Pickles(1998) Therefore it is of limited applicabilityfor the coolest spectral types L and T which

would explain the discrepancy when going tothe fainter G = 21 mag limit

Figure 3 shows the number of sources inapparent G bins estimated from the local densi-ties by Caballero et al (2008) (gray) and fromthe Dwarf Archives catalogue (corrected forthe fraction of sources without available crossmatches in the SDSS bands violet) It seems toindicate that either the local densities result in a03ndash05 dex overestimate of the counts per binor the Dwarf Archives catalogue is incompleteby that same factor or a mix of both effectsmay be present at the same time According tothe work presented by Smart (2014) there is

Sarro et al Gaia limits for brown dwarf studies 641

Fig 3 Predicted UCD counts per magnitude bin inthe GUCDS according to the BT-Settl models and(i) the local spatial densities from Caballero et al(2008) (gray) and (ii) the Dwarfs Archive samplecorrected for the fraction of entries measurementsin the SDSS photometric system

evidence supporting an incompleteness factorof at least two in the Dwarf Archive under theassumption of isotropy and local homogeneityin the local distribution of UCDs which wouldexplain part of the discrepancy

If we convert the i minus J colour index intoa spectral type using the calibration by Hawleyet al (2002) we obtain the distribution of typesshown in Fig 4

If we apply a similar analysis to a set ofnearby star forming regions and clusters as-suming the corresponding set of distances andages and the BT-Settl family of synthetic spec-tra we can estimate the I-band magnitude thatcorresponds to the Gaia detection limit of G =20 mag and the corresponding values of masseffective temperature and spectral type Theresults are summarised in Table 1 Gaia willdetect a significant number of UCDs with spec-tral types from mid-M to late M and few L0Note that in the case of very young nearby as-sociations the masses will be close or even be-low the deuterium burning mass limit Thussome ldquoisolated planetary-mass objectsrdquo mightbe detected

M1 M5 M9 L3 L7 T1 T5

Spectral Type

Cou

nts

050

100

150

200

Fig 4 Distribution of spectral types of sources inthe Dwarf Archive that can be detected by Gaia(with estimated G magnitudes brighter that 20 mag)

4 Conclusions

We have used the Gaia pre-launch perfor-mances as assessed during the Critical DesignReview carried in 2011 to infer what can beexpected of the GUCDS in particular the dis-tribution of spectral types and the detectabilityof known UCDs in the Dwarfs Archive database In the process of inferring apparent Gmagnitudes for these Dwarf Archives UCDswe have extended the set objects with observa-tions in both riz and JHK bands well beyondthe coolest examples in the previous work byHawley et al (2002) We obtained SDSS pho-tometry for 587 UCDs in the Dwarfs Archivedatabase reaching spectral types in the T se-quence

As in Sarro et al (2013) these conclusionsrely on many uncertain assumptions They de-pend on the final performance of the satelliteinstruments in the first place but also criti-cally on the calibration of the relationship be-tween magnitudes in different passbands ob-tained from the BT-Settl models or the calibra-tion of spectral types with the iminusJ colour indexIn particular we show that the predictions fromthe Dwarfs Archive and from local spatial den-sities disagree by a factor at least two We rec-ommend caution when using these estimates

642 Sarro et al Gaia limits for brown dwarf studies

Table 1 An estimate of the coolest spectral type detectable with Gaia in several young nearbystar forming regions and clusters

Region d Age Dimmest MI Mass Teff Sp type[pc] [Myr] [mag] [MJup] [K]

ρ Ophiuchi 120ndash145 1 1305 10 2200 L0Taurus 140 1ndash2 1270 16 2400 M9Serpens 260 2 1138 27 2600 M8Chamaeleon I II 140 1ndash3 1270 18 2400 M9Lupus I II III 140 1ndash3 1270 18 2400 M9IC 348 385 2ndash4 1054 37 2700 M7Trumpler 37 800 4 783 201 3200 M4σ Orionis 400 2ndash4 1045 42 2800 M6Collinder 69 400 5ndash10 1045 49 2800 M6Upper Scorpius 145 5 1262 21 2400 M8NGC 7160 800 10 783 317 3400 M3IC 2391 175 30ndash55 1221 45 2600 M7IC 2602 160 30ndash50 1241 42 2600 M7IC 4665 350 45 1077 91 3000 M5α Persei 185 80 1209 60 2700 M7Blanco 1 270 100 1131 83 2900 M6Pleiades 150 125 1255 58 2600 M7Hyades 50 600 1499 37 2200 L0Praesepe 187 800 1207 72 2800 M6

since at the time of writing this manuscript theimpact of condensations and stray light on thefinal performances of the Gaia instruments isstill uncertain On the other hand the on-goingstudy about a potential extension of the Gaialimiting magnitude down to G = 21 mag wouldhave an important and positive effect on thecompleteness of the GUCDS

Acknowledgements The authors wish to acknowl-edge the Coordination Unit 2 of the Gaia DPAC forthe use of the GOG simulator and Rosanna Sordofor their kind help and guidance with the simula-tion of both synthetic and empirical spectra at theGaia instrumental characteristics This research hasbeen supported by the Spanish Ministry of Sciencethrough grants AyA2011-24052 AYA2011-30147-C03-03 and AYA2012-38897-C02-01

References

Allard F Homeier D Freytag B Jun 2012Royal Society of London PhilosophicalTransactions Series A 370 2765

Bailer-Jones CAL Andrae R Arcay Bet al Nov 2013 AampA 559 A74

Caballero JA Burgasser AJ Klement RSep 2008 AampA 488 181

Hawley SL Covey KR Knapp GR et alJun 2002 AJ 123 3409

Jordi C Gebran M Carrasco JM et al Nov2010 AampA 523 A48+

Jordi C May 2012 httpwwwrssdesaintcslivelinkopen2760608

Pickles A J July 1998 PASP 110 863Sarro LM Berihuete A Carrion C et al

2013 AampA 550 A44Smart R L 2014 MmSAI 85 649

Sarro et al Gaia limits for brown dwarf studies 641

Fig 3 Predicted UCD counts per magnitude bin inthe GUCDS according to the BT-Settl models and(i) the local spatial densities from Caballero et al(2008) (gray) and (ii) the Dwarfs Archive samplecorrected for the fraction of entries measurementsin the SDSS photometric system

evidence supporting an incompleteness factorof at least two in the Dwarf Archive under theassumption of isotropy and local homogeneityin the local distribution of UCDs which wouldexplain part of the discrepancy

If we convert the i minus J colour index intoa spectral type using the calibration by Hawleyet al (2002) we obtain the distribution of typesshown in Fig 4

If we apply a similar analysis to a set ofnearby star forming regions and clusters as-suming the corresponding set of distances andages and the BT-Settl family of synthetic spec-tra we can estimate the I-band magnitude thatcorresponds to the Gaia detection limit of G =20 mag and the corresponding values of masseffective temperature and spectral type Theresults are summarised in Table 1 Gaia willdetect a significant number of UCDs with spec-tral types from mid-M to late M and few L0Note that in the case of very young nearby as-sociations the masses will be close or even be-low the deuterium burning mass limit Thussome ldquoisolated planetary-mass objectsrdquo mightbe detected

M1 M5 M9 L3 L7 T1 T5

Spectral Type

Cou

nts

050

100

150

200

Fig 4 Distribution of spectral types of sources inthe Dwarf Archive that can be detected by Gaia(with estimated G magnitudes brighter that 20 mag)

4 Conclusions

We have used the Gaia pre-launch perfor-mances as assessed during the Critical DesignReview carried in 2011 to infer what can beexpected of the GUCDS in particular the dis-tribution of spectral types and the detectabilityof known UCDs in the Dwarfs Archive database In the process of inferring apparent Gmagnitudes for these Dwarf Archives UCDswe have extended the set objects with observa-tions in both riz and JHK bands well beyondthe coolest examples in the previous work byHawley et al (2002) We obtained SDSS pho-tometry for 587 UCDs in the Dwarfs Archivedatabase reaching spectral types in the T se-quence

As in Sarro et al (2013) these conclusionsrely on many uncertain assumptions They de-pend on the final performance of the satelliteinstruments in the first place but also criti-cally on the calibration of the relationship be-tween magnitudes in different passbands ob-tained from the BT-Settl models or the calibra-tion of spectral types with the iminusJ colour indexIn particular we show that the predictions fromthe Dwarfs Archive and from local spatial den-sities disagree by a factor at least two We rec-ommend caution when using these estimates

642 Sarro et al Gaia limits for brown dwarf studies

Table 1 An estimate of the coolest spectral type detectable with Gaia in several young nearbystar forming regions and clusters

Region d Age Dimmest MI Mass Teff Sp type[pc] [Myr] [mag] [MJup] [K]

ρ Ophiuchi 120ndash145 1 1305 10 2200 L0Taurus 140 1ndash2 1270 16 2400 M9Serpens 260 2 1138 27 2600 M8Chamaeleon I II 140 1ndash3 1270 18 2400 M9Lupus I II III 140 1ndash3 1270 18 2400 M9IC 348 385 2ndash4 1054 37 2700 M7Trumpler 37 800 4 783 201 3200 M4σ Orionis 400 2ndash4 1045 42 2800 M6Collinder 69 400 5ndash10 1045 49 2800 M6Upper Scorpius 145 5 1262 21 2400 M8NGC 7160 800 10 783 317 3400 M3IC 2391 175 30ndash55 1221 45 2600 M7IC 2602 160 30ndash50 1241 42 2600 M7IC 4665 350 45 1077 91 3000 M5α Persei 185 80 1209 60 2700 M7Blanco 1 270 100 1131 83 2900 M6Pleiades 150 125 1255 58 2600 M7Hyades 50 600 1499 37 2200 L0Praesepe 187 800 1207 72 2800 M6

since at the time of writing this manuscript theimpact of condensations and stray light on thefinal performances of the Gaia instruments isstill uncertain On the other hand the on-goingstudy about a potential extension of the Gaialimiting magnitude down to G = 21 mag wouldhave an important and positive effect on thecompleteness of the GUCDS

Acknowledgements The authors wish to acknowl-edge the Coordination Unit 2 of the Gaia DPAC forthe use of the GOG simulator and Rosanna Sordofor their kind help and guidance with the simula-tion of both synthetic and empirical spectra at theGaia instrumental characteristics This research hasbeen supported by the Spanish Ministry of Sciencethrough grants AyA2011-24052 AYA2011-30147-C03-03 and AYA2012-38897-C02-01

References

Allard F Homeier D Freytag B Jun 2012Royal Society of London PhilosophicalTransactions Series A 370 2765

Bailer-Jones CAL Andrae R Arcay Bet al Nov 2013 AampA 559 A74

Caballero JA Burgasser AJ Klement RSep 2008 AampA 488 181

Hawley SL Covey KR Knapp GR et alJun 2002 AJ 123 3409

Jordi C Gebran M Carrasco JM et al Nov2010 AampA 523 A48+

Jordi C May 2012 httpwwwrssdesaintcslivelinkopen2760608

Pickles A J July 1998 PASP 110 863Sarro LM Berihuete A Carrion C et al

2013 AampA 550 A44Smart R L 2014 MmSAI 85 649

642 Sarro et al Gaia limits for brown dwarf studies

Table 1 An estimate of the coolest spectral type detectable with Gaia in several young nearbystar forming regions and clusters

Region d Age Dimmest MI Mass Teff Sp type[pc] [Myr] [mag] [MJup] [K]

ρ Ophiuchi 120ndash145 1 1305 10 2200 L0Taurus 140 1ndash2 1270 16 2400 M9Serpens 260 2 1138 27 2600 M8Chamaeleon I II 140 1ndash3 1270 18 2400 M9Lupus I II III 140 1ndash3 1270 18 2400 M9IC 348 385 2ndash4 1054 37 2700 M7Trumpler 37 800 4 783 201 3200 M4σ Orionis 400 2ndash4 1045 42 2800 M6Collinder 69 400 5ndash10 1045 49 2800 M6Upper Scorpius 145 5 1262 21 2400 M8NGC 7160 800 10 783 317 3400 M3IC 2391 175 30ndash55 1221 45 2600 M7IC 2602 160 30ndash50 1241 42 2600 M7IC 4665 350 45 1077 91 3000 M5α Persei 185 80 1209 60 2700 M7Blanco 1 270 100 1131 83 2900 M6Pleiades 150 125 1255 58 2600 M7Hyades 50 600 1499 37 2200 L0Praesepe 187 800 1207 72 2800 M6

since at the time of writing this manuscript theimpact of condensations and stray light on thefinal performances of the Gaia instruments isstill uncertain On the other hand the on-goingstudy about a potential extension of the Gaialimiting magnitude down to G = 21 mag wouldhave an important and positive effect on thecompleteness of the GUCDS

Acknowledgements The authors wish to acknowl-edge the Coordination Unit 2 of the Gaia DPAC forthe use of the GOG simulator and Rosanna Sordofor their kind help and guidance with the simula-tion of both synthetic and empirical spectra at theGaia instrumental characteristics This research hasbeen supported by the Spanish Ministry of Sciencethrough grants AyA2011-24052 AYA2011-30147-C03-03 and AYA2012-38897-C02-01

References

Allard F Homeier D Freytag B Jun 2012Royal Society of London PhilosophicalTransactions Series A 370 2765

Bailer-Jones CAL Andrae R Arcay Bet al Nov 2013 AampA 559 A74

Caballero JA Burgasser AJ Klement RSep 2008 AampA 488 181

Hawley SL Covey KR Knapp GR et alJun 2002 AJ 123 3409

Jordi C Gebran M Carrasco JM et al Nov2010 AampA 523 A48+

Jordi C May 2012 httpwwwrssdesaintcslivelinkopen2760608

Pickles A J July 1998 PASP 110 863Sarro LM Berihuete A Carrion C et al

2013 AampA 550 A44Smart R L 2014 MmSAI 85 649