Transcript Un universo tridimensional con Gaia - Gaia-UB

Un universo tridimensional con
Gaia
Francesca Figueras
On behalf of the UB- Gaia team
Universidad de Barcelona (Spain)
Index
•
•
•
•
Basic principles and data reduction strategy
What Gaia will observe: the Gaia Simulator (GUMS and GOG)
Scientific challenges: some examples
On ground spectroscopic surveys complementary to Gaia
Gaia astrometric accuracy
Gaia astrometric accuracy
~10 as/yr
Earth – Moon
Gaia astrometric accuracy
Basic Principles
Scanning law
Scanning Space Astrometry: to transform positional information into timing data
Scanning Space Astrometry:
Red Photometer CCDs
Blue Photometer CCDs
42,35cm
to transform positional information into timing data
Radial Velocity
Spectrometer CCDs
Absolute parallaxes
Obtained from differential along-scan measurements between the two FoV.
The error depend on the Γ angle (0.5 μas accuracy)
Parallactic displacement along the great cicle Sun-Star
Sensitivity AL is proportional to sin ξ sin Γ
ξ = Sun-spin axis angle = 45º for Gaia
Γ = basic angle = 106.5º for Gaia
Optimal values between astrometry requirements - that call for a large angle - and
implementation constraints - such as payload shading and solar array efficiency
Instrumental calibration
Each CCD:
-calibrated geometrically
-Calibrated photometrically
Chromaticity correction:
Asymmetric aberrations, the
diffraction image of Gaia is
wavelength-dependent.
The polychromatic image
centroid is shifted:
- aberration (FoV position)
- Source SED (BP, RP spectra)
Reduction Strategy
AGIS Core Processing
100 million “well-behaved” stars
Intermediate Data Update
Astrometric Global Iterative Solution
• Requirements at 20 μas level (10-10 rad)
• Colours are needed for PSF, LSF (centroiding)
• Radial velocities enter in astrometric model
• relativistic model (orbit, ephemerides, time)
• Models for stars not fitting the model:
• Binaries
• Variable stars
• Data Volume
• ~500 TB – 1PB for 5 years
• 1020-21 flop
• CPU time: 1sec/star – 30 years
What Gaia will observe?
the Gaia Simulator:
GUMS and GOG
The Gaia Simulator(s)
Gaia Universe Model Simulator
Published: Robin et al. 2012 A&A
Available at CDS
The Gaia Simulator: MW stars
Besançon Galaxy Model
Drimmel et al. (2003) 3D extinction
Variable stars:
Stars per square-deg (log10)
(Y,X)
(Z,X)
(Z,Y)
The Gaia Simulator: galactic populations
The Gaia Simulator: extragalactic objects
Unsolved Galaxies
~3.8·106
Z < 0.8
Quasars
~ 5·105
Z<4
GOG: Gaia Object Generator
An attempt to simulate Gaia products
GUMS + a model for Gaia errors
Goals:
To fill the Gaia Archive
For Science Exploitation
Sky map of the mean parallax error
Scanning law (large
number of transits)
The Galactic Center
(large number of
faint stars)
Equatorial coordinates, units: mas (Palmer, Luri et al., 2013, in prep)
Gaia and the Magellanic Clouds
The distance to LMC and SMC
GUMS: Based on a real catalogue, 7.5·106 (LMC), 1.5 106 (SMC) with G<20
Gaia data:
Large error in individual distances
Maximum Likelihood techniques are mandatory (Luri et al., 1996)
Relative error in mean distance: 0.5% (LMC), 1.5% (SMC)
No 3D map
SMC with OGLE (Haschke et al., 2012):
Cepheids (2522 stars): 63.1  3.0 kpc , 4.7 % accuracy
RR Lyrae (1494 stars): 61.5  3.4 kpc , 5.5 % accuracy
GAIA’s view
R136 (LMC)
Transverse velocities
~1-2 km/s accuracy
G=12-15 mag
(~10 as/yr)
de Bruijne and de Marchi, 2011
R136, the star cluster in the Tarantula (30 Doradus)
Gaia (GIBIS)
HST
GIBIS: Gaia Basic Image Simulator
Stellar density at G<20 ~1.4 106 stars /sqdeg
de Bruijne and de Marchi, 2011
Gaia and the distance scale
1912: Henrietta Leavitt discovered the key for
the distance scale of the Universe
Period-Luminosity relation
(25 cepheids in the SMC)
33
The Cepheids
Gaia will observe ~9000 Cepheids (extrapolated from Berdnikov et al. cat)
Gaia: Metallicity dependence of the PL relation
Windmark et al., 2011
Hipparcos suspected binarity in Cepheids
~ 50 % binaries, the companion star affect the brightness and motion
Gaia will treat several of them as binaries (astrometric orbit)
Hipparcos: no allowance for
binarity, thus the motion
along the orbital arc could
falsify the deduced parallax
value.
It is remarkable that all
negative parallax values
plotted in the figure belong
to known binaries (open
clicles).
Hipparcos vs. 'ground-based' parallax
(Laszlo Szabados, 2010)
~500 000 Quasars
- The Reference Frame
- How to detect them?
- Lensed QSO in Gaia survey?
- Astrometry for astrophysics
-…
Inertiality of the Gaia Celestial Reference Frame
Accuracy of the residual rotation (units: as/yr), Mignard (2011)
QSO Detection
SVM (Super Vector Machine) using BP/RP and astrometry
Bayler-Jones et al. (2008)
Confusion matrix
QSO with low EWs emission lines
removed , they are confused with cool
(4000–8000 K) highly reddened
(AV = 8–10) stars
BP/RP spectra for QSO (red) and stars (blue)
Number of lensed Quasars in Gaia Survey?
0.6% of quasars will consist of multiple lensed QSO images (~3000)
Finet et al. 2011
AGNs
A set of selected AGNs revealed photocenter jitters at mas level
perturbations in the accretion disk emissivity? (no, Popovic et al. 2012)
energetic processes occurring therein (SN, GammaRays)?
Radio-quiet 1620+172 (Mrk 877), R=16, z=0.112, 2pc/mas (Antón et al., 2011)
The transient sky by Gaia
Ground contact: 8h /day
Analysis and anomaly detection at Cambridge
Alerts issued in 24-48 hours after observations
See: Wyrzykowski et al. 2013, IAU 298
The Red Clump Stars
The Galactic Bar (s) and the Spirals
Galactic disk space distribution function
1/ is a biased estimates of
the true distance!!
Red clump surface density
Does our Galaxy have one/two bars?
work in the space of the observables!
distances
parallaxes
Romero-Gómez, Aguilar et al.
Does our Galaxy have one/two bars?
work in the space of the observables!
Extinction is critical: A new 3D new map using Gaia and IR
Red Clump: accuracy in tangential velocities
Gaia data
Gaia + IR distances (10%)
Mandatory to combine Gaia and IR data
Hyper-Runaway Stars
HIP 60350 (Runaway, B-star, 3.5 kpc)
At the moment, the quality of the observational data is insufficient to pinpoint
the precise origin of the star within the spiral arm (cluster birthplace?)
Gaia parallax accuracy ~10 as (G~11), 3 % accuracy in the relative parallax
Was the star originated some ≈15 Myr ago, in the Crux-Scutum spiral arm?
Irrgang, A., et al. 2010
Solar System minor bodies
~100 TNO: orbit, binarity detection, …
~2·105 asteroids: orbit, rotation, shape, …
Can Gaia discover new Earth Trojans?
Can Gaia discover Earth Trojan?
Solar system fossils
Solar system star formation
Oct-2010: 1st 2010TK7
Difficult from Earth:
Rather close to the Sun
Very dispersed on sky
Region of highest probability for
Detection (Todd et al. 2011 )
Can Gaia discover Earth Trojan?
ET simulations
Todd et al. 2012
Gaia advantages:
Gaia disadvantages:
Earth’s L2 Lagrangian point
Co-orbital nature with ET
Continuous scanning mode
Regions surveyed multiple times
Down to a Solar elongation of 45º
No limitation on local zone
No limitation on airmass
Only up to G=20
High along and across scan velocity
(loss of signal, out of CCD window)
Is 2010 TK7 the largest?
If yes, prob. detection is low
On-ground Spectroscopic Surveys
complementary to Gaia
Gaia-ESO survey (GES)
Public large spectroscopic survey with FLAMES@VLT
Started Feb/2012 + 5 years (300 nights)
Ips: Randich, Gimore + ~300 Co-Ips
All stellar populations: Halo, Bulge, Thick/Thin disk + open clusters
Products:
105 Giraffe spectra (R~16000-25000)
104 UVES spectra (R~47000)
 + ESO archive
An optical Multi-Object-Spectrograph (2017)
WEAVE@ WHT
Canary Island
Radial velocities 2 km/s V=20
Abundances V17
The Gaia
Archive
Data Archive: Goals
I. A validated and well documented set of Gaia data
II. A functional, single point access to all Gaia science data
III. A defined API to allow the development of high throughput access
applications and visualization tools.
IV. A set of advanced applications for data manipulation and visualization
V. Tools and content for outreach (social impact of the mission)
IP: X. Luri (Univ. Barcelona)
~300 Co-IPs
Gaia and other (future)
large surveys
LSST and Gaia:
complementary for studying the Milky Way
Gaia will provide calibration checks to astrometric LSST data
LSST will extend the Gaia survey four magnitudes deeper.
end
Networks and coordinated projects
Scientific Exploitation:
GREAT-FP7: European Union ‘Initial Training Network (32 institutions )
REG: Red Española de Explotación Científica de Gaia (140 members, 30 institutions)
GREAT-ESF: European Science Foundation (2011-2015) (17 countries, 90 institutions)
On Ground Complementary Data:
GES: Gaia ESO Survey (2012-2016)
Gaia Data Achive:
CU-9: Gaia DPAC (~300 participants)
GENIUS-FP7:
General aspects
Proper Motions at 20 μas/y (V=15)
• 20 μas/y = 10 m/s a100 pc (planets around 0.5 milion stars;
Júpiter motion is 15 m/s)
(MV=+10)
• 20 μas/y = 1 km/s at 10 kpc (slower star’s motions detected at 10 kpc)
(MV=0)
• 20 μas/y = 5 km/s at 50 kpc (internal LMC kinematics as the local kin.
(MV=-3.5)
today, 5 km/s = 2.5 mas/y at 400 pc)
• 20 μas/y = 100 km/s at 1 Mpc (curva de rotación en M31?)
(MV=-10)
• Parallax accuracy 20 μas (V=15) = 1% in distance at 0.5 kpc (6D structure of the
Orion complex 2pc resolution)
Can Gaia discover Earth Trojan?
Fossils: Solar system formation
Oct-2010: 1st discov. 2010TK7
Difficult from Earth:
Rather close to the Sun
Very dispersed on sky
Probability of existence of Earth trojans
on stable orbits (Earth at longitude
=0), Todd et al. 2011
Gaia i els exo-planetes
Resultats esperats:
•
•
•
•
•
~2000 exo-planetes (sistemes
simples) detectats astrometricament. 1.2 Planète : r = 100 mas P = 18 mois
Dd(")
1/07/02
~300 sistemes amb diversos
1
planetes.
1/01/03
òrbites ben determinades per ~1000 0.8
1/07/01
1/01/02
0.6
sistemes.
1/01/01
~5000 trànsits planetaris observats 0.4
1/07/00
Planetes amb masses fins per sota de0.2
1/01/00
10 MTerra a 10 pc.
0
0
Terceres Jornades d'Astronomia al Montsec
0.2
0.6
0.4
Dacos d (")
0.8
1.0
Gaia capabilities / products
• Positions, proper motions and parallaxes for 1 billion stars (G < 20)
• Low resolution spectrophotometry for 1 billion stars, allowing estimations of
Teff, logg, Av and [Fe/H]
• Radial velocities for 150 million stars (G < 16)
• Atmospheric parameters, reddening and rotational velocities for 5 million
stars (G < 12)
• Detailed chemical abundances for 2 million stars (G < 11)
Proper direction:
Orbital data prec.: 150m, 2.5 mms-1
Field angles
Field angles: ,
Proper direction: u
Sistema instrumental:
z: Eje de spin
x: Bisectriz de las dos
direcciones astrométricas
Attitude
• The attitude relates the SRS to the CoMRS (esentially the ICRS)
• Is given by A(t):
X,Y,Z components of u in CoMRS
x,y,z components of u in SRS
Expresed in quaternions:
Instrumental calibration
Each CCD row calibrated geometrically
- (in,x,y), D(in,x,y)
Each CCD photometrically defined
- bias, flatfield, etc.
in=0,1
x,y: position on the focal plane
PSF, LSF calib.
Global Iterative Solution
Comparison of observed and calculated field angles
Differences explained by a lineal model as a function of a set
of parameters, depending on what you want to measure
Global Iterative Solution
Attitude: B-spline coefficients.
Time interval, all the observations
Calibration: all the observations on a given column and time
interval
Global Iterative Solution
Source: six astrometric parameters
All the observations of a given source
Global:
All the observations
Global Iterative Solution
Source: six astrometric parameters
All the observations of a given source
Global:
All the observations
Data reduction
• 1012 Individual measurements (transits)
• 1010 Unknowns (all related, simultaneous determination)
• 5·109 stars(pos, pm,par)
• 1.5·108 attitude
• 10 - 50 106 calibration
• Some tens of “global”
• Requirements at 20 μas level (10-10 rad)
• Colours are needed for PSF, LSF (centroiding)
• Radial velocities enter in astrometric model
• relativistic model (orbit, ephemerides, time)
• Data Volume ~500 TB – 1PB for 5 years, 1020-21 flop
• CPU time: 1sec/star – 30 years
WEAVE: optical MOS
• For WHT
• FoV = 2o
• MOS + mIFU + LIFU
• R= 5k, ~1000 fibres
V=20 (R=5k, SNR=10, 1h)
• R= 20k from grating change
V=16 (R=20k, SNR=50, 1h)
• D = 0.37-1.00 μm
Status:
• Concept Study, 01/2011, first Science Case completed 2/2012
• 03/2011 ASTRONET partners recommend North(WHT)+South(ESO) MOS
• 06/2011 UK, NL commit funding for WEAVE to PDR (expected 2013)
• 09/2011 IAC support construction of WEAVE
• 2016: Instrument First Light
WEAVE & Gaia
• Formation scenarios for Galactic stellar halo. In-situ or
accreted?
– Total mass of the Milky Way out to 200 kpc
– The shape of the Galactic gravitational potential within 50–100 kpc
– Lumpiness of the Galactic dark matter distribution within 20–50 kpc
• The dynamics of the Galactic disk & chemical labeling
– Configuration space and global phase-space constraints
– Local substructures in phase-space, resonances, and stochasticity
– Chemo-dynamical constraints
• Galactic open clusters
– Formation and disruption
– Tracers of chemical evolution of the disk
Gaia and LSST performances
Field angles: ,
Ellipse: instantaneous scan great-circle
Rectangles: the two Gaia FoV (BAM  = 106.5º)
Sun: always 45º from the positive z axis
Global Iterative Solution
Comparison of observed and calculated field angles
Differences explained by a lineal model as a function of a set
of parameters, depending on what you want to measure
Astrometric Global Iterative Solution
The Gaia Simulator: stars with planets
The Gaia simulator: extragalactic objects
Stars per square-deg (log10)
GOG
• Provides:
– Epoch (transit) and combined (end-of-misssion) data
– True data, data as observed by Gaia and their errors
• Is based on:
– A model of the Gaia instruments
– Error models provided by the CUs (DPAC) (final Gaia
data will be more complex)
• Has two main simulation modes :
– The GUMS universe model (integrated in GOG)
– An external list of sources provided by the user
Proper motion accuracy (End-of-mission)
units: as/yr (Palmer, Luri et al., 2013, in prep)
The colour scale represents log density of objects in a bin size of 80mmag by 2 as/yr
R136, the star cluster in the Tarantula (30 Doradus)
Gaia (GIBIS)
HST
de Bruijne and de Marchi, 2011
QSO Detection
SVM (Super Vector Machine)
using BP/RP and astrometry
QSO with emission-line EWs less than
5000Å removed (they are confused
with cool highly reddened stars)
Està al revés?
QSO, galaxies and stars: astrometry
Astrometry hardly improves the results
Bailer-Jones et al. (2008)
Gaia QSO catalogue
QSOs : crucial targets to define the Gaia Celestial Reference Frame
(GCRF)
Unprecedented precision in photo-center position:
astrometric stability of QSOs and the possible physical consequences
inner quasar structure and physical processes
AGNs
Radio emitter AGNs:
defining the quasi-inertial International Celestial Reference Frame (ICRF).
will help in the alignment between optical (Gaia) and radio reference frame
A set of selected AGNs revealed photocenter jitters at mas level
perturbations in the accretion disk emissivity? (no, Popovic et al. 2012)
energetic processes occurring therein (SN, GammaRays)?
Radio-quiet 1620+172 (Mrk 877), R=16, z=0.112, 2pc/mas (Antón et al., 2011)
The Supernovae
≃ 6000 SNe (Type Ia + CC) SNe over the 5-year mission (G<19)
Most distant observed type Ia SNe will be at ≃ 500Mpc (z ≃ 0.12).
About 1/3 of the SNe is expected to be observed before maximum
Science Alert programme stablished
The Gaia Scanning Law
The Gaia Scanning Law
End-of-mission number of
transits per source
Microlensing events during the
Gaia mission
The Gaia Simulator: microlenses
Prediction of astrometric
microlensing events during the
Gaia mission
43 astrometric microlensing effect during the Gaia mission.
The effect allows a precise measurement of the mass of a
single star that is acting as a lens
The 2 events that should be observable with Gaia are
plotted as green squares