560 likes | 686 Views
Measuring the Milky Way with the VLBA. Mark J. Reid Harvard-Smithsonian Center for Astrophysics. Fringe spacing: q f ~ l /D ~ 1 cm / 8000 km = 250 m as Centroid Precision: 0.5 q f / SNR ~ 10 m as Systematics: path length errors ~ 2 cm (~2 l ) shift position by ~ 2 q f
E N D
Measuring the Milky Way with the VLBA Mark J. Reid Harvard-Smithsonian Center for Astrophysics
Fringe spacing: qf~l/D ~ 1 cm / 8000 km = 250 mas Centroid Precision: 0.5 qf / SNR ~ 10 mas Systematics: path length errors ~ 2 cm (~2 l) shift position by ~ 2qf Relative positions (to QSOs): DQ ~ 1 deg (0.02 rad) cancel systematics: DQ*2qf ~ 10 mas Measuring the Milky Way with the VLBA Comparableto GAIA & SIM
Where is the Galactic Center? Sun Harlow Shapley Galactic Center Globular clusters based on Cepheid distances Projected onto Galactic Plane
Galactic Center Stellar Orbits VLT: 2 mm • M = 4 x 106 Msun • R < 50 AU • Den. > 1017 Msun/pc3 5” Ghez et al / Genzel et al 0.04” What can radio observations tell us?...
Where is the Galactic Center? VLA l = 6 cm 1.3 cm 5” Zhao Jun-Hui Sgr A*
Where is the Galactic Center? Combined IR + Radio astrometry shows Sgr A* at focal position of stellar orbits (+/-10 mas) Stars move fast. Does Sgr A* move fast?
The Proper Motion of Sgr A* 2 Ref. QSOs: DQ ~ 0.7o Expect to see effect of Sun’s orbit around Galactic Center: 6 mas/yr along Gal. Plane Galactic Plane 90 cm VLA image: LaRosa et al
Proper Motion of Sgr A* • Parallel to Galactic Plane: 6.379 (+/- 0.024) mas/yr Qo/Ro = 29.5 km/s / kpc • Perpendicular to Gal. Plane: 7.2 km/s motion of Sun • Could re-define Galactic Plane Now: HI & Sun in plane New: LSR orbit & Sgr A* in plane • Sgr A*’s motion ^ to Gal. Plane -0.4 +/- 0.9 km/s ! IAU Gal. Plane fit to data Reid & Brunthaler (2004)
SMBH “Brownian motion” • ~106 stars orbiting Sgr A* • Calculate orbit segments • Stellar c.o.m. changes • Sgr A* moves to keep total system c.o.m. constant • Smi << MSgrA. … MV ~ mv • Smi ~ MSgrA. … MV2 ~ mv2
Latest Results: Sgr A* Proper Motion • IR Stellar Orbits: • MIR ~ 4 x 106 Msun • R < 50 AU • Radio Observations: • Sgr A* motionless • M > 10% of MIR • Observed size: • R < 0.5 AU • IR + Radio data combined: • Dark mass = luminous source • Density > 1022 Msun/pc3
Must Sgr A* be a SMBH? Object Density Method Mass within Radius (Msun/pc3) (Msun) M 87 106 HST 3x109 7 pc NGC 4258 1010 VLBA : H2O4x107 0.1 pc Sgr A* 1017 IR Star orbits 4x106 50 AU Sgr A* >1022 VLBA p.m. >4x105 0.5 AU SMBH 1024 3RSch4x106 3*0.08 AU 3RSch =30 mas @ 8 kpc VLBI (JCMT/SMA-ALMA-LMT-SMT-CARMA) @ 0.8 mm 20 mas News Flash: Fringes: HawaiiArizona @ 1.3mm (60 mas) ! Doeleman et al
Milky Way Viewed From Inside Sgr A* Thomas Lucas Productions, Inc. (www.tlproductions.com)
Lunar Parallax 7000 km baseline Hipparchus (189 BC) Pete Lawrence’s Digitalsky: http://www.digitalsky.org.uk
Geocentric • … Pythagoras, Aristotle … • Ptolemy • Heliocentric • Aristarchus (ca. 250 BC) • Copernicus • Stellar Parallax: clear test of heliocentric cosmology Geocentric vs. Heliocentric Cosmology
From: Galileo Galilei & Johannes Kepler To: Cosimo II de’ Medici, Grand Duke of Tuscany Re: Funds for a bigger telescope to measure stellar parallax “If the Earth moves around the Sun, we should see stellar parallax!” Grand Duke’s advisors point out: Referee 1) Che’ idea pazzesca! (What a crazy idea !) Referee 2) Even if the idea were correct, we don’t know how far away the stars are; so there is no way to estimate the probability of success. Referee 3) Null result doesn’t prove or disprove anything. Excerpts from a 1609 Proposal:
Centuries of failure…until 1838 • Friedrich Wilhelm Bessel • using a Fraunhofer telescope • 61 Cygni: p = 0.314” • (p = 0.287”) • Followed closely by • Thomas Henderson (a Centuri) • Wilhelm Struve (Vega) Parallax: The Race to Measure the Cosmos1 Friedrich Wilhelm Bessel 1Alan Hirshfeld 2001 (W.H. Freeman & Co., NY)
118,000 stellar parallaxes sp ~ 0.8 mas 10% accuracy at 125 pc… mapped solar neighborhood Next great space missions: GAIA & SIM: ~10 mas HIPPARCOS (Perryman et al 1997 A&A 323 49)
Practical Lessons • Narrow the angle: • finding nearby calibrators • Measure atmospheric delays: • “geodetic-blocks” or GPS • Measure reference position:
Finding Background Calibrators • Select candidates from NVSS • <1.5 degrees on sky from target • >20 mJy • Unresolved (<20”) • Observe with VLA at X- and K-band and require • unresolved • synchrotron spectral index (a < -0.5) • usually find 1 – 3 good sources • Re-observe in A or B config to get position (<30 mas)
Typical Background Sources • Sn = 5 80 mJy • nearly point-like
Effects of Atmospheric Delay Errors • Simulations: • 1 cm zenith path errors • 4 VERA stations • PA of target w.r.t. reference Honma, Tamura & Reid (2008)
Obs ~15 sources over sky • large ZA differences • 60 ICRF sources • positions <1 mas • >1000 obs, little structure • Trial schedule • skip randomly through list • generate fake data and fit • Pick best of ~1000 trails “Geodetic Block” Atmospheric Calibration
Geodetic-Block vs. GPS Calibration Geo-block GPS Honma, Tamura & Reid (2008)
Reference Source Position Crucial 1) Include 1 source with mas position in p-ref’ing OR 2) Fully calibrate data (except p-ref’ing) Make dirty image Estimate position from “speckles”
Reference Source Phases • Can get “raw’ phases within few turns over day • Similar to VLA - A data
Orion Nebular Cluster Parallax D = 414 +/- 7 pc 389 +/- 22 pc Sandstrom et al (2007) 437 +/- 19 pc Hirota et al (2007) 414 +/- 7 pc Menten et al (2007) Menten, Reid, Forbrich & Brunthaler (2007)
Milky Way Parallaxes Distance estimates: Kinematic = 4.3 kpc Photometric = 2.2 kpc (R. Humphreys 1970’s) W3(OH) T. Megeath (Spitzer Space Telescope)
Kinematic Distances • Doppler shift: • VDop = V - Vsun • Model (Gal. Rotation): • Vmod(Ro,Qo,l,D) • Adjust D to match • Vmod to VDop V V { D Qo Ro
W3OH Parallax p = 0.512 +/- 0.010 mas Xu, Reid, Zheng & Menten (2006)
W3OH Parallax • Dphoto ~ Dparallax • Dk way off • In Perseus Arm, not in Outer Arm • Large peculiar V Schematic Model of Milky Way: Taylor-Cordes / Georgelin & Georgelin
S 252 Parallax Maser 2 3 QSOs; Maser 1 p = 0.480 +/- 0.010mas Reid et al (2008)
Methanol Maser Parallaxes Kinematic distances (Dk): Problem: Dk > Dp Partial fix: Ro < 8.5 kpc and/or Qo > 220 km/s Sgr A* p.m. requires Qo/Ro = 29.5 km/s/kpc = 236 / 8.0 = 251 / 8.5 Brunthaler, Menten, Moscadelli, Reid, Xu, & Zheng Honma et al; Hachisuka et al
Peculiar Motions of Star Forming Regions • In rotating frame: Ro = 8.5 kpc Qo = 220 km/s • Clear systematic motions • Update Galaxy model: • Ro = 8.5 kpc • Qo = 251 km/s • Systematic motions smaller, but significant
Peculiar Motions of Star Forming Regions Galactic model: Ro = 8.5 kpc Qo = 251 km/s & Solar Motion: U = 8 km/s V = 18 km/s W = 10 km/s Residual motions considerably smaller
Solar Motion • Hipparcos & VLBA • U: Gal. Center • W: North Gal. Pole Maser parallax/P.M. Sgr A* proper motion Dehnen & Binney (1998) Hipparcos data (black) Excellent agreement between Hipparcos and VLBA
Solar Motion V Gal. Rot. “Asymmetric Drift:” V appears larger when measured against older stars with higher dispersion Asymmetric Drift Late B-type Maser p & p.m. Dehnen & Binney (1998) Hipparcos data in black Massive stars born rotating ~13 km/s slower than Galaxy spins; as they age, first speed up and then slow down again.
Massive Star Birth • Possible Sequence: • Molecular cloud in circular orbit • Hit by Spiral shock • Goes into elliptical orbit (near apocenter) • Compression triggers star formation
The Next 5 Years • Ro & Qo from parallax & p.m. data with 3% accuracy • Map of Milky Way spiral structure • Critically test spiral density wave theory
Extreme Supergiants Extreme red supergiants with L 106 Lsun “Fabulous 4”: NML Cyg, S Per, VY CMa, VX Sgr H2O and SiO masers in circumstellar envelopes VY CMa 0.4 < D < 1.8 kpc Association with NGC 2362 D = 1.5 kpc (M-S fitting) Parallax measured to be 1.1 kpc... L = 3 x 105 Lsun (quite reasonable) NGC 2362 cluster closer than thought
Local Group Proper Motions M33 • 1920s van Maanen claimed to see M33 spin! • mas/yr motions Spiral nebulae nearby (Galactic) • Hubble argued more distant (extra-galactic) • van Maanen’s error not found
Extragalactic Proper Motions M33 • Parallax accuracy: sD ~ 10% at 10 kpc can’t do galaxies yet • Proper motion: same techniques, but sm ~ T-3/2 • M33 & IC10 1) see spin (van Maanen) 2) see galaxy’s motion Andreas Brunthaler’s PhD Thesis
Extragalactic Proper Motions • M33/IC133 – M33/19 masers • VLBA: Dmx = 30 +/- 2, Dmy = 10 +/- 5 mas/yr • HI: Dvx = 106 +/-20, Dvy = 35 +/- 20 km/s • D = 750 +/- 50 +/- 140 kpc • Improvements in Rotation Model & 3rd maser source: • sD < 10% possible sm sv Brunthaler, Reid & Falcke
Extragalactic Proper Motions • M31’s motion unknown • Critical for L.G. dynamics
Tidal Heating of M33 • Try M31 proper motions; then calculate orbits • Tidal heating of M33 for trial proper motions of M31: mM31 = (100,-100) km/s ( -50, -50) km/s ( 0, 0) km/s mM31 ~ 0 km/s M33 destroyed mM31 ~100 km/s M33 OK Loeb, Reid, Brunthaler & Falcke (2005)
NGC 4258 • Seyfert galaxy • H2O masers in an edge-on, sub-parsec disk • Rotation speed ~1000 km/s • M ~ 3 x 107 Msun • Geometric model D = 7.2 +/- 0.5 Mpc • Used by Hubble Key Project to re-calibrate Cepheid PL relation Herrnstein, Moran, Greenhill et al (1999)
AGN Maser Distance Measurements A = V2 / R R = D q \ D = V2 / A q High Velocity Maser Map Red high-velocity masers V R = D q Blue high-velocity masers Drift of systemic masers over time Systemic velocity masers
Maser Distance Measurements (2) D = V2 / Aq V q
Maser Cosmology ProjectBraatz, Condon, Greenhill, Henkel, Lo & Reid • Goal: Ho accurate to 3%; constain Dark Energy Eq. of State • How: Geometric Distances to H2O masers in Hubble Flow GBT finds masers VLBA maps them