Galactic Surveys Astrometry and photometry

(1)

Carlos Allende Prieto IAC

Galactic Surveys Astrometry and

photometry

(2)

Overview

• Astronometry: Hipparcos and Gaia

• Photometry: DSS, SDSS, 2MASS …

• Fitting data to models

Tuesday, August 27, 2013

(3)

Basic astronomical

measurements (from light)

Astrometry > positions of stars in the sky, proper motions, parallaxes

Photometry > colors and brightness (stellar properties)

Spectroscopy > radial velocities, line strengths, stellar properties

(4)

Gaia’s Three Elements

Astrometry (V < 20)

completeness to 20 mag ⇒ 10⁹ stars

accuracy: 10–25 µarcsec at 15 mag

scanning satellite, two viewing directions

principles: global astrometric reduction (as for Hipparcos)

Photometry (V < 20)

Low-dispersion spectrophotometry 0.3 - 1 µm

Radial velocity (V < 16–17)

slitless spectroscopy near Ca II triplet (847–874 nm⁾

third component of space motion,

dynamics, population studies, binaries

spectra: chemistry, rotation

(5)

Astrometry

• Positions and motions of stars provide full 3D maps of the near universe around us – first the solar neighborhood, now more distant

parts of the Milky Way and even the nearest local group galaxies

• First parallax measured in 1838 by Bessel (61 Cygni, 0.3 arcsec)

• The Hipparcos mission measured parallaxes for 1e5 stars with mas precision and 1e6 stars with lower precision between 1989 and 1993.

• Gaia is Hipparcos successor, with all-around enhancements

(6)

Hipparcos Gaia

Magnitude limit 12 20 mag

Completeness 7.3 – 9.0 20 mag

Bright limit 0 6 mag

Number of objects 120 000 26 million to V = 15 250 million to V = 18 1000 million to V = 20 Effective distance

limit

1 kpc 50 kpc

Quasars None 5 x 10⁵

Galaxies None 10⁶ – 10⁷

Accuracy 1 milliarcsec 7 µarcsec at V = 10

10-25 µarcsec at V = 15 300 µarcsec at V = 20 Photometry

photometry

2-colour (B and V) Low-res. spectra to V = 20

Gaia: Complete, Faint, Accurate

(7)

Stellar Astrophysics

Parallaxes and photometry imply a comprehensive luminosity

calibration



distances to 1% for ~10 million stars to 2.5 kpc



distances to 10% for ~100 million stars to 25 kpc



parallax calibration of all distance indicators e.g. Cepheids and RR Lyrae to

LMC/SMC



accurate parallaxes imply accurate surface

gravities and age

(8)

Stellar Astrophysics

An unbiased survey implies a detailed Galactic census

solar neighbourhood mass function and luminosity function

e.g. white dwarfs (~200,000) and brown dwarfs (~50,000)

initial mass and luminosity functions in star forming regions

rare stellar types and rapid evolutionary phases in large numbers

Statistics on variability across the board (~40 (RVS) - 100 (AS,XP) visits per object)

(9)

One Billion Stars in 6-d will Provide …

 in our Galaxy …

 the distance and velocity distributions of all stellar populations

 a rigorous framework for stellar structure and evolution theories

 a large-scale survey of extra-solar planets (~20,000)

 a large-scale survey of Solar System bodies (~ few 100,000)

 … and beyond

 definitive distance standards out to the LMC/SMC

 rapid reaction alerts for supernovae and burst sources (~20,000)

 QSO detection, redshifts, microlensing structure (~500,000)

 fundamental quantities to unprecedented accuracy: γ to 10^-7(10^-5 present)

(10)

0 0.2 0.4 0.6 0.8 1 1.2

∆α χοσ δ (∀)

1 /0 1 /0 0

1 /0 1 /0 1

1 /0 1 /0 2

1 /0 1 /0 3

1 /0 7 /0 0 1 /0 7 /0 1

1 /0 7 /0 2

∆δ(∀)

0 0.2 0.4 0.6 0.8 1.0

Πλαντε : ρ = 100 µασ Π = 18 µοισ

Exo-Planets: Expected Discoveries

 Astrometric survey:

 monitoring of hundreds of thousands of FGK stars to ~200 pc

 detection limits: ~1M_J and P < 10 years

 masses, rather than lower limits (m sin i)

 multiple systems measurable, giving relative inclinations

 Results expected:

 ~20,000 exo-planets (~10 per day)

 orbits for ~5000 systems

 masses down to 10 M_Earth to 10 pc

 >1000 photometric transits

Figure courtesy François Mignard

(11)

 Asteroids etc.:

 deep and uniform (20 mag) detection of all moving objects

 ~ few 100,000 new objects expected (357,614 with orbits presently)

 taxonomy/mineralogical composition versus heliocentric distance

 diameters for ~1000, masses for ~100

 orbits: 30 times better than present

 Trojan companions of Mars, Earth and Venus

 Kuiper Belt objects: ~300 to 20 mag (binarity, Plutinos)

 Near-Earth Objects^:

 Amors, Apollos and Atens (2249, 2643, 406 known today)

 ~1600 Earth-crossers >1 km predicted (937 currently known)

 detection limit: 260–590 m at 1 AU, depending on albedo

Studies of the Solar System

(12)

Satellite and System

• ESA-only mission

• Launch date: late 2013

• Launcher: Soyuz–Fregat

• Orbit: L2

• Lifetime: 5 years

• Ground station: New Norcia and Cebreros

• Downlink rate: 4–8 Mbps

• Mass: 2120 kg (payload 700 kg)

• Power: 1720 W (payload 735 W)

Figures courtesy EADS-Astrium

(13)

(14)

Payload

(15)

Payload and Telescope

Two SiC primary mirrors 1.45 × 0.50 m² at 106.5°

SiC toroidal structure (optical bench) Basic angle

monitoring system

Combined focal plane

(CCDs) Rotation axis (6 h)

Figure courtesy EADS-Astrium

Superposition of two Fields of View

(FoV)

(16)

Focal Plane

Star motion in 10 s

Total field:

- active area: 0.75 deg² - CCDs: 14 + 62 + 14 + 12 - 4500 x 1966 pixels (TDI) - pixel size = 10 µm x 30 µm

= 59 mas x 177 mas

Astrometric Field CCDs

Bl ue

P h ot

o m et er

C C D s

Sky Mapper CCDs

104.26cm

R ed

P h ot o m et er C C D s

Radial-Velocity Spectrometer

CCDs

Basic Angle Monitor Wave Front Sensor

Basic Angle Monitor

Wave Front Sensor

Sky mapper:

- detects all objects to 20 mag - rejects cosmic-ray events - FoV discrimination

Astrometry:

- total detection noise: ~6 e^-

Photometry:

- spectro-photometer - blue and red CCDs

Spectroscopy:

- high-resolution spectra - red CCDs

42.35cm

Figure courtesy Alex Short

(17)

On-Board Object Detection

Requirements:

unbiased sky sampling (mag, colour, resolution)

all-sky catalogue at Gaia resolution (0.1 arcsec) to V~20

Solution: on-board detection:

good detection efficiency to V~21 mag

FPA CCDs generate Gbps thus windows needed

(18)

Sky Scanning Principle

Spin axis 45^o to Sun Scan rate: 60 arcsec/s Spin period: 6 hours

45^o

Figure courtesy Karen O’Flaherty

(19)

Astrometric Data Reduction Principles

Sky scans (highest accuracy

along scan)

Scan width: 0.7°

1. Object matching in successive scans 2. Attitude and calibrations are updated 3. Objects positions etc. are solved 4. Higher terms are solved

5. More scans are added 6. System is iterated

Figure courtesy Michael Perryman

(20)

Light Bending in Solar System

Movie courtesy Jos de Bruijne

Light bending in microarcsec, after subtraction of the much larger effect by the Sun

(21)

Gaia imaging

91 CCDs (4000 x 2000 pixels each) Distances for 1.000.000.000 sources!

(22)

The Radial Velocity Spectrometer

TDI spectroscopy!

(23)

Stellar motions

(24)

Photometry Measurement Concept

Figures courtesy EADS-Astrium

Blue photometer:

330–680 nm Red photometer:

640–1000 nm

(25)

Photometry Measurement Concept

Figures courtesy Anthony Brown

Blue photometer

300 350 400 450 500 550 600 650 700

0 5 10 15 20 25 30 35

AL pixels wavelength (nm)

0 5 10 15 20 25 30 35 40

spectral dispersion per pixel (nm) .

Red photometer

600 650 700 750 800 850 900 950 1000 1050

0 5 10 15 20 25 30 35

AL pixels wavelength (nm)

0 2 4 6 8 10 12 14 16 18

spectral dispersion per pixel (nm) .

RP spectrum of M dwarf (V=17.3) Red box: data sent to ground White contour: sky-background level

Colour coding: signal intensity

(26)

Ideal tests

Shot, electronics (readout) noise

Synthetic spectra

Logg fixed (parallaxes will constrain luminosity)

G=18.5 G=20

S/N per pixel

Bailer-Jones 2009

GAIA-C8-TN-MPIA-CBJ-043

(27)

(Spectro-)photometry

ILLIUM algorithm (Bailer-Jones 2008). Dwarfs:

G=15 ([Fe/H])=0.21

(Teff)/Teff=0.005

G=18.5 ([Fe/H])=0.42

G=20 ([Fe/H])=1.14

G=20

(28)

RVS S/N ( per transit and ccd)

3 window types: G<7, 7<G<10 (R=11,500), G>10 (R~4500)

 √ (S + rdn²)

Most of the time RVS is working with S/N<1

End of mission spectra will have S/N > 10x higher

G magnitude

Allende Prieto 2009, GAIA-C6-SP-MSSL-CAP-003

(29)

Sample RVS spectra

(mission end, black line)

G=10.5 G=12.3 G=15.8

B5V

G2V

Metal- poor K1III

Allende Prieto 2009

(30)

RVS produce

Radial velocities down to V~17 (10⁸ stars)

Atmospheric parameters (including overall metallicity) down to V~ 13-14 (several 10⁶ stars)

Chemical abundances for several elements down to V~12-13 (few 10⁶ stars)

Extinction (DIB at 862.0 nm) down to V~13 (e.g.

Munari et al. 2008)

~ 40 transits will identify a large number of new

spectroscopic binaries with periods < 15 yr (CU4, CU6, CU8)

(31)

RV performance

Spec. for late-type stars

1 km/s at V<13

15 km/s down to V=17

(32)

Atmospheric parameters

^(Ideal

tests)

Solid: absolute flux

Dashed: absolute flux, systematic errors (S/N=1/20)

Dash-dotted: relative flux

Allende Prieto (2008)

(33)

Photometry

• Gaia will not be the first full-sky photometric survey

• Palomar photographic plates (POSS)

• HST needed a full-sky pointing catalog, which was prepared from digitized photographic

plates (DSS)

• 2MASS: first full-sky ground-based near-IR photographic survey (J H Ks filters)

• SDSS provided a large/area (14,000 sqr. deg) optical survey (ugriz system) using CCD

detectors

• Others: GALEX (NUV), WISE (IR), UKIDS …

(34)

Usual photometric systems

• Johnson (-Cousins) UBVRI

• Ströngrem ubvy

• Near-IR Y J H Ks

• SDSS ugriz

• GALEX FUV/NUV

• …

• System responses usually include

(approximate) atmospheric extinction

(35)

SDSS

• First massive solid-state optical photometric survey (some 14,000 deg2 and 150 million stars down to r ~ 22 mag)

• 2% photometry – 1% in stripe 82

• Highly-uniform observations (single site/telescope/instrument)

• Carefully designed filters (though issues with u-band)

(36)

SDSS imaging

• 6 x 5 CCDs

• Running in TDI

(37)

The world’s biggest

picture

_• 26 Gigapixels!

(38)

2MASS

• 2 automated 1.3m telescopes (one in Arizona, one in Cerro Tololo, Chile)

• 3 channel (J, H, Ks) cameras, each with a 256x256 HgCdTe detector

• 7.8s exposures

• 4 years of operation

• PSC: ~ 300 million stars down to J/H/Ks of ~ 16,15,14

• About 1 million extended sources

(39)

(40)

UKIDSS

started in 2005

some 400 papers already published

uses WFCAM on 4-m class UKIRT (four 2048x2048 Rockwell devices)

7500 sqr. Deg down to K~18

(41)

VHS

• 19,000 sqrt. deg

• About 4 mag. deeper than 2MASS

• Using ESO’s VISTA telescope

(42)

The Future: LSST

• A wide-field 8.4m telescope

• A 3.2 Gpix camera

• Imaging the whole (accessible sky) every few nights

• Starting in 2018

• Tens of TB of data each night

(43)

LSST

(44)

The variable sky

• Most of the transients are fairly near, but most exciting ones are far away

(45)

The variable sky

(46)

The variable sky

• We do not know what is out there

• Lots of room for classification

algorithms

(47)

Zero-point

• Absolute calibration of astronomical sources is non-trivial

• Good lab reference sources hard to observe through telescopes as if they were at infinity

• Atmospheric extinction/distortion gets in the way

• Traditional reference source is Vega, which sets zero-point tied to lab sources (Tungsten lamps or black bodies; see Hayes 1985,

Megessier 1995), but Vega is not easy to model

• Spectrophotometric calibration nowadays tied to DA white dwarf models (Bohlin 2010 and

prev. refs.)

(48)

White dwarfs

• DA white dwarfs are fairly simple: just two parameters (Teff,logg), pure-H physics,

NLTE but good agreement among models

Allende Prieto, Hubeny &

Smith 2009

Examples From SDSS

(49)

HST DAs

• 3 DA white dwarf stars constitute the basis for HST calibration (see papers by Bohlin)

• Good to 1-2%

• Calibration consistent

for Vega

V=0.023 +/- 0.008

Allende Prieto, Hubeny &

Smith 2009

(50)

HST DAs analyzed with different models

Allende Prieto, Hubeny & Smith 2009

(51)

A-type stars

 Not pure hydrogen, but spectrum dominated by it in optical and IR

(continuum and lines); exception FeII lines in UV

 Three parameters (Teff,logg,[Fe/H])

 Reddening needs to be accounted for (also true for faint WDs)

 Brighter and more common than WDs

(52)

A-type stars

 Not pure hydrogen, but spectrum dominated by it in optical and IR (continuum and lines)

 Three parameters (Teff,logg,[Fe/H])

 Reddening needs to be accounted for (also true for faint WDs)

 Brighter and more common than WDs

Allende Prieto & del Burgo (in prep). Spectra from NGS (Gregg et al.)

(53)

Vega

Allende Prieto & del Burgo (in prep). HST spectrum Gilliland & Bohlin Fast rotation

(54)

Zero-point

• HST flux calibration: using zero-point V

magnitude for Vega (not quite zero, V= 0.023 mag) and 3 DA WDs models

• Vega STIS spectrophotometry calibrated in that way compares well with a model

atmosphere for Vega and leads to a

consistent zero-point based on photometry performed on the model

• System seems robust to 1-2% level

• STIS spectrophotometry (calspec, NGSL) now being used to set zero points for photometric systems

(55)

Halo turn-off stars

 F-type, metal-poor: H continuum + lines and few metal lines (not so few in the

blue/UV)

 Again 3 parameters (+ reddning) but now higher impact of [Fe/H] due to electrons

forming H-

 Many of them (just leaving the main sequence), easy to pick up from colors

 Choice used for SDSS

 BD +17 4708 is the prototype

(56)

Halo turn-off stars

 F-type, metal-poor: continuum H and H-, H lines and few metal lines (not so few in the blue/UV)

 Many of them (just leaving the main sequence), easy to pick up from colors

 Choice used for SDSS

 BD +17 4708 is the prototype

Ramírez et al. 2009; HST data from Bohlin and colleagues

(57)

Fitting models to data

• Understanding the structure of the Milky Way is critical

• Starcounts are the most fundamental (and easy) measurement: just photometry and coarse astrometry (star positions)

• Distances must be estimated, but parallaxes to a few percent available for only some 1e5 stars (Hipparcos)

• Photometric parallaxes derived for dwarfs based on models or semi-empirical

relationships (e.g. clusters)

M-m = 5 -5 log(d)

(58)

Color – absolute mag

relationships for dwarfs

Tuesday, August 27,

2013 Juric et al. 2008

(59)

Standard candles

• Dwarfs outnumber giants in most cases (not always). Their luminosities depend on metal content but weak dependence on age at a given color

• An alternative is to use stars at specific evolutionary stages that can be identified with certainty, and with reliable theoretical (or semi-empirical) luminosities

• Examples include cepheids, red-clump stars, RR Lyrae

(60)

Standard candles

• For example, red-clump giants have been

very useful in the obscured parts of the Milky Way

• An approximate extinction relation between colors and passbands can be adopted

Cabrera-Lavers et al. 2008

Babuiaux and Gilmore 2005

(61)

Red-clumb giants in the bulge

Cabrera-Lavers et al. 2008

(62)

Large data sets

• One can either fit the data with models, e.g.

Larsen & Humphreys (2003), Robin et al.

(2003)

• Or derive density maps, which are subsequently fit to infer the model parameters, e.g. Juric et al. 2008

• Models involve a number of std. Milky Way stellar components: a disk (or two), a halo, and a bulge (plus other non-std. such as a bar or streams as needed)

• Nowadays, more complex orbit-family-type or numerical-simulations available, but

parametric models provide a fast and useful path to start

(63)

Density maps

• Photometric parallaxes are derived first

• Positions on the sky and distances are used to

create a binned density map (Juric et al. 2008)

• Photometric [Fe/H]

estimates can be used

(64)

Density maps

• This approach allows to clean-up the density maps before we fit radially symmetric models

Juric et al. 2008

(65)

Fitting models to data

• Typical 3-4 component stellar Milky Way models involve 6-8 parameters: relative

densities, halo exponent, disk scale height(s) and length(s)

• Extinction needs to be

included in disk and bulge

• Parameters constrained by optimization algorithm

algorithm

(66)

Tools

• Besançon model (Gaia universe model)

• M. Cohen’s model

• TRILEGAL (see Girardi’s lectures)

• GALFAST (Juric)

• Jordi Molgo’s simulator

• …

(67)

What’s next

• Back to Gaia…

• Starcounts soon to be suplemented with trig.

Parallaxes (replacing photometric ones)

• and spectrophotometric metallicities (Gaia) plus spectroscopic metallicities and more detailed abundances for a fraction of the

sample (APOGEE/SDSS, Gaia-ESO, GALAH…)

• Further work on map construction desirable

• Idem for tools for evaluating simple, parametric, Milky Way models