Avril DayJones University of Hertfordshire Collaborators David Pinfield University of Hertfordshire UH Federico MaroccoUH Ben BurninghamUH Maria Teresa RuizUniversidad de Chile UC ID: 271886
Download Presentation The PPT/PDF document "Measuring Brown Dwarf Properties from De..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Measuring Brown Dwarf Properties from Deep Surveys
Avril Day-JonesUniversity of Hertfordshire
Collaborators: David Pinfield (University of Hertfordshire, UH), Federico Marocco(UH), Ben Burningham(UH), Maria Teresa Ruiz(Universidad de Chile, UC), Zenghua Zhang(UH), Joana Gomes(UH), Leigh Smith(UH), Phil Lucas(UH), Hugh Jones(UH), James Jenkins(UC).Slide2
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide3
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide4
What is a ‘Deep’ Survey?
The advancement of brown dwarf science can be largely attributed to the rise of ‘deep’ optical and NIR surveys.
Until recently 2MASS was considered the ‘state of the art’ in the NIR, taking over as the Deepest NIR survey from the TMSS, going 100 million times fainter.This changed however with the advent of UKIDSS (and VISTA). The number, type and quality of properties that can be measured by these surveys is essentially a factor of Volume. As even though UKIDSS covers just 10% of the sky, it probes 4 magnitudes fainter, or 10 times volume and has discovered more T dwarfs than 2MASS, plus more to come…(see poster of
Skrypek
)Slide5
A comparison of deep surveys
Survey
YearBandLimiting magnitudeSky CoverageLT
Y
SDSS
2000-2012/13
u
’,
g
’,
r
’,
i
’, z’
22.0, 22.2, 22.2, 21.3, 20.5
14,555sq deg~380 +34 (Zhang’10)
~50-
DENIS
1995-2001R, I,K18.5, 16.5, 14.0
16,700sq deg~50
1
-2MASS
1997-2001J, H, Ks15.8,15.1,14.3
All Sky
~400~50
UKIDSS
1995-2013Z,Y,J,H,K
20.2, 19.6, 18.8, 18.2 (LAS YJHK)6800sq deg (All surveys)LAS 3600 sq deg
~501
+52 (Day-Jones’13)~20087 + 76
(Burningham’13) +10 (Day-Jones’13)
VISTA2010-presentZ,Y,J,H,K
21.2, 21.2, 20.6, 20.6 (VHS YJHS)22,200sq deg (All Surveys) VHS 20,000 sq deg
5
WISE2010W1,W2,W3,W4
16.5, 15.5, 11.2, 7.9All Sky~10
~90
+ 87 (Mace’13)
14
Brown dwarfs discoveredSlide6
A comparison of deep surveys
www.ukidss.orgSlide7
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide8
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide9
What brown dwarf properties can we measure from deep surveys?
Both optical and NIR surveys have played a large part in the discovery of over ~1400 brown dwarfs.
We can measure from imaging: Photometry (magnitudes, colours) and identify all the weird and wonderful basic observable characteristics of BDs. Astronomtry : Proper motions, parallaxes, and thus distances.…and from spectroscopy we can measure their chemical composition, gravity and temperature.However due to the unique M-L relation for brown dwarfs, we can’t measure other fundamental parameters, such as mass or age directly (at least the models are not yet robust enough for an accurate calculation).Other properties on the larger scale come from the sheer numbers of objects of different spectral type we can measure. From this we can measure a mass function and a formation history. Which up until the era of the latest deep surveys has remained unknown.Slide10
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide11
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide12
Individual property measurements
Benchmark
BDs can place limits on the mass (dynamical mass measurements), metallicity or age. The usefulness of a benchmark BD will be dependent on the accuracy of the property.In order to help constrain atmospheric models the one of the key ingredient is knowing what stage of its evolution it is currently in (i.e. age). Benchmarks physical properties (spectral and photometric properties) can then be studied, to enable trends and correlations in colours
and observable features to factors such as
metallicity
and age.
Such benchmarks could be identified as members of young cluster, moving groups or as binary/multiple systems.
However, in general, good benchmarks are rare!
Individual properties that aren’t directly measureable can be calculated using ‘Benchmark’ brown dwarfs.Slide13
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide14
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide15
Benchmark
brown dwarfs
Young (<1Gyr) benchmark BDs can be identified as members of open clusters or moving groups from their measured distances, proper motions and their position on a colour magnitude diagram, confirming their membership. Such clusters (ages >100Myrs) that have been studied/mined spectroscopically for
BDs
include
Hyades, Pleiades, Castor,
Ursa
Major and Alpha
Persei
.,
dozens have been confirmed.
The largest complement of ‘evolved’ benchmark
BDs
with ages >1Gyr come from members of widely separated multiple or binary systems with primary stars that have known ages derived from robust models.
These include main-sequence stars, M dwarfs,
subgiants
, giants and white dwarfs.
The age of the main-sequence primary can be calculated via several methods, including age-
metallicity
relations,
chromospheric
or coronal activity (
CaHK
, H
, X-ray), Lithium abundance, rotational or space velocity, or from their location on a HR diagram compared to theoretical isochrones.
However these ages often have large uncertainties associated with their ages, due to the convergence of model tracks on the main-sequence (Girardi et al. 2000; Yi et al. 2001).
The later stages of evolution, such as the
subgiant or white
dwarf primaries however can give at least minimum ages, but have the potential to reveal BDs with age accuracies of ~10% (if the WD is suitably high mass >0.7Mo)
. See Silvia Catalan’s talk on Wednesday!
The Young
And the old
.Slide16
Age benchmark brown dwarfsSlide17
Age benchmark brown dwarfs
L0 L5 T0 T5 Y0
Spectral TypeSlide18
Our own searches for benchmark
BDs
Identified ~380 candidate WD+BD systems in SDSS+UKIDSS (DR7/DR8).Discovered the first T dwarf companion to a white dwarf.Over 90% of these systems are now followed up for astrometry (proper motions) from 2700sq deg of sky, down to J=19; with our observing campaigns on the SOAR, Blanco and Magellan telescopes in Chile over the last 3 years. Where we will be able to measure the very wide binary fraction for WD+BD systems at a new unprecedented level.
T dwarf + WD binary (Day-Jones et al. 2011).
A new WD + L dwarf binary (Day-Jones et al., 2013b, in prep).
We have made several searches for benchmark
BDs
as members of binaries. Including common proper motion companions to white dwarfs. Slide19
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide20
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide21
Properties of the BD galactic population
Fundamental parameters such as the mass and initial functions, and the formation history, can be measured from a large population of
BDs. The initial mass function is measured from young clusters.Formation history (or birth rate) can be calculated from the older, disk population.Objects in very deep surveys can be used to measure the galactic scale height.Slide22
The galactic scale height of brown dwarfs
The newer deep surveys give an opportunity to measure the scale height for brown dwarfs.
The UKIDSS LAS can probe out to distances of ~350pc for early L dwarfs (H=18.8 limit). This changes to 260pc for L3-L5, 115pc for L7-T2 and 46pc for T6-T8. However there is significant non-uniformity at H>18.0, which limits the detection of changes in the scale height for early to late L types.The UKIDSS Deep eXtragalactic Survey (DXS) and Ultra Deep Survey (UDS), while it only covers ~35sq deg are sensitive down to J=22.5 (3 mags more sensitive than the LAS) and 100% complete. It will allow for reliable space densities of late L type and L/T transition objects out to ~300pc above the galactic plane.Lodieu et al. (2009) discovered the first late T dwarfs, a T6±1 and T7±1 in the DXS, at distances of ~60 and 80pc.Slide23
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide24
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide25
Simulations of the brown dwarf birth rate
Several groups have made simulations of the mass function and formation histories of sub-stellar objects.
Chabrier (2002) made the first estimation from monte-carlo simulations, using just 2 formation histories, a flat and a time-decreasing exponential form, for 2 different IMFs:IMF1=powerlaw (α=1.55)
IMF2=Lognormal/exponential
IMF2 yields a smaller number of bright
BDs
since they were formed several billion years ago and are now very faint.
Allen (2005) also took a
Baysian
approach to early 2MASS data. Using 3 models. A segmented power law, a log normal form-the same as
Chabrier
and a low mass cut off power law. With a uniform (solid), increasing (dot-dashed) and decreasing (dashed) age. But concluded that the L8-T5, or M
J
region 14.0-15.5 was incomplete and could not, at that point constrain the birth rate.
Chabrier
(2002)
Allen (2005)Slide26
Simulations of the brown dwarf birth rate
Burgasser et al. (2004).The region of transition between L and T dwarfs (~1100-1500K) is suggested to be the most sensitive temperature range for the birthrate.
1. A flat or constant formation (
simpliest
form).
2. Empirical formation, which yields a similar
Teff
distribution to the flat birthrate, the same as that measured for stars by Rocha & Pinto (2000).
3. A cluster formation history also gives essentially the same result, assuming a flat, but stochastic (
i.e
in a number of clusters) formation.
4. The exponential formation however, produces a rather different distribution in the
mid~L-mid~T
range, and is consistent with a star formation rate that scales with the average gas density (Miller 1979).
5. Finally, the halo formation results in a radically different
spt
distribution, and considers formation within a 1~Gyr burst, 9~Gyr in the past. Slide27
Simulations of the brown dwarf birth rate
Burgasser et al. (2004).The region of transition between L and T dwarfs (~1100-1500K) is suggested to be the most sensitive temperature range for the birthrate.
Simulations based on Deacon &
Hambly
et al. (2006)Slide28
The UKIDSS LAS and measuring the brown dwarf birth rate
We selected candidates from UKIDSS (DR7), based on NIR and NIR-red optical colours, where selections were derived from Schmidt et al (2010), who provide colours from an unbiased spectroscopic complete sample of L and T dwarfs.
262 candidate mid L – mid T dwarfs.
(Day-Jones et al. 2013)
. Slide29
X-Shooter has 3 arms UVB (300-550nm), VIS (550-1000nm) and NIR (1000-2500nm) that take long slit spectroscopy simultaneously.
Used Integration times of:J<17.0 800s, J<17.5 1200s, J<18.0 1600s,J<18.2 2000sS/N ~30, when binned to R=510 (NIR).
VIS arm S/N ~10, when binned to R=800.Negligible flux in the UVB arm.Our X-Shooter observing campaignWe have a project on X-Shooter/VLT (
26
nights, across
4
semesters,
7
observing runs). Started in 2010 (ESO
P86-91)
to observe our sample of
BDs
that sit in the 1100-1500K (L/T transition region)
.
We have observed ~200 of our candidates to date. Complete to J=17.6 for the LAS sky, and to 18.1 for a smaller 450sq degree area.Slide30
We used the ESO X-Shooter pipeline along with our own custom built IDL codes to reduce the data.
Spectral typing was done using template spectra from the SpeX
prism library using the standard spectral types of Kirkpatrick et al (1999) and Burgasser et al (2006) for L and T dwarfs, respectively.Our X-Shooter observing campaign
All the spectra is planned to be available in an online resource.Slide31
Unresolved binaries
Several
identified unresolved binary candidates were identified (based on the criteria defined by Burgasser et al. 2010.).Day-Jones et al. (2013).Slide32
Unresolved Binaries
Day-Jones et al. (
2013)Template spectra were used to make combined templates from the Spex prism library. And fits were made using Chi-squared and an F test
These need to be followed up with AO imaging and RV measurements to confirm their
binarity
.
Once confirmed they will play an important role in the accurate determination of the birth rate, but also allow us to place constraints on the close binary fraction, which has been suggested to range from 3-45%.Slide33
The brown dwarf birthrate from the UKIDSS LAS.
Taken a sub-sample of 68 L and T dwarf, which is complete for a set region of sky (RA=22-04hrs, DEC= -1- 10deg), corresponding to 495 sq deg (1/10 of the LAS), down to a magnitude limit of J=18.1.
Completeness: We are complete to 85% for L0-L3, 88% for L4-L6, 94% for L7-T0.99% for T1-T4 Histogram of density vs Spt.Maximum distances for each object selected calculated (MJ-SpT) using Marocco et al. (2010) and derived the volume sampled by each Teff bin.Malmaquist
and
Eddington
bias was also corrected for in the space densities, considering the mean scatter of the sample of known L and T dwarfs around the MJ-
SpT
relation. Representing an increase in the sample volume of 22%.
Also include the effects of the ‘good’ unresolved binaries.
These are compared to simulations by Niall Deacon, based on the code described in Deacon &
Hamley
(2006). Slide34
The brown dwarf birthrateSlide35
The form of the mass function is in agreement with the findings for late T dwarfs alone, suggesting a negative form of
α, such that -1< α < 0It isn’t currently possible to constrain the birthrate further than to say that a halo form is extremely unlikely.
With a full sample (260) we will be able to reduce the errors by ~50% and should be able to rule out at least a β=0 or β=-0.5.This work also represents the discover of a new population of L-T transition objects. An increase in the number of L4-T5 dwarfs by ~50% (for the first sub-sample of 68). Slide36
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide37
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide38
Future Deep surveys
Optical
DES (DECam) –g’r’,I’,z’, Y filters. VLT Survey Telecope-KIlo Degree Survey (KIDS)- u’,g’,r’,I’,z’.1500sq deg.Panstars – will be able to provide accurate parallaxes and identify many more benchmark primary stars.LSST-
0.3-1.1 micron over 6 bands, covering 20,000 sq deg.
Gaia
-
All sky V=6-20mag. Will measure very accurate astrometry, RV and
photospectroscopy
.
There are several missions both underway and planned for the future, that will discover many new, fainter and cooler brown dwarfs as we probe deeper than ever before. Some of these include:
Just how deep can we go?
The main problems that will arise from the detection of fainter
BDs
in these surveys is the increasing difficulty of following them up with spectroscopy, which will be challenging for even E-ELT/GMT/TMT type next generation instruments, such that measuring properties in the traditional ways will not be extremely difficult, or not possible. So we will need alternative ways of measuring their properties.
NIR/IR
VISTA VHS/ VIKING
EUCLID (ESA)–
1 optical + 3NIR band imaging (to 24.5mag) and
slitless Lowres
(R=250) spectroscopic capabilities.15,000sq deg.SPICA (ESA/Japan)- MIR
camera:Wide field Imaging and spectroscopy at 5-38microns. R up to 30,000.
JWST-Imaging and spectroscopy NIR (0.6-5microns) and MIR (5-28microns). Medium resolution.Slide39
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide40
Overview
What is a Deep Survey? Brief review of the Optical and NIR surveys that have been
successful in identifying BDS.
What Properties of
BDs
can we measure from Deep surveys?
Individual properties
Population properties
Benchmark brown dwarfs
The mass function & Formation history
Future deep surveys-how much deeper can we go?
Tasks for the futureSlide41
Tasks for the future
Properly constrain the birth rate
. This is a statistical game, which is reliant on large numbers of brown dwarfs across (critically) the L/T transition region to be identified.Reduce error bars on benchmark object properties. This can be somewhat reliant on models of other stars, and our understanding of age indicators in young clusters and moving groups.Identify larger numbers of benchmark brown dwarfs, including new Y dwarf benchmarks.Fully populate the benchmark space. Including ages, masses and metallicities.There are several tasks that need to be done in the near future in order to fully characterise brown dwarf parameter space and allow the calibration of models, which will be the hinging point for which the future measurement of accurate properties will rely upon. These can be done with continuing current and near future deep surveys (e.g. UKIDSS, SDSS, VISTA, WISE) and include some of the following:
With such populations of benchmarks identified, and a better understanding of the formation history, this will lead models to greater reliability and allow us to look at links with observable properties (photometry and astrometry) with other physical attributes (age, mass, gravity,
metallicity
), and allow us to estimate such physical properties with only basic observations –
Important as we go deeper and fainter!Slide42
Thank you.