Gravitational redshifts, and other wavelength shifts in ...dainis/Presentations/Vilnius_2011/... ·...

Gravitational redshifts, and other wavelength shifts in stellar spectra

Dainis Dravins – Lund Observatory

www.astro.lu.se/~dainis

KVA

Exactly 100 years ago

A.Einstein, Annalen der Physik 340, 848 (1911)

Predicted effects by gravity on light

Foreground galaxy cluster manually retouched away

STRONG LENSING WEAK LENSING MICROLENSING

Already long before Einstein…

John Mitchell (1784)

Predicting gravitational effects on light

Predicting gravitational effects on light

John Mitchell (1784)

Predicting gravitational effects on light

John Mitchell (1784)

Predicting gravitational effects on light

Pierre-Simon Laplace (1796)

Verifying Einstein ?

Klaus Hentschel: Erwin Finlay Freundlich and Testing Einstein’s Theory of Relativity, Archive for History of Exact Sciences 47, 243 (1994)

Freundlich’s attempts to verify relativity theory (I)

Erwin Finlay Freundlich (1885-1964) worked to experimentally verify the predictions from Einstein’s theory of relativity and the effects of gravity on light.

Freundlich’s attempts to verify relativity theory (II)

Klaus Hentschel: Erwin Finlay Freundlich and Testing Einstein’s Theory of Relativity, Archive for History of Exact Sciences 47, 243 (1994)

Einsteinturm, Potsdam-Telegrafenberg

Nature 306, 727 (1983)

Unethical falsifications in astronomy ?

QJRAS 26, 279 (1985)

Controversial interpretations of history

Actual gravitational redshift in white dwarfs

X-ray spectra for early (top), and late phases of the bursts from neutron stars.

Red: Absorption lines from the circumstellar medium

Blue: Photospheric absorption lines at redshift Z = 0.35

J.Cottam, F.Paerels, M.Mendez: Gravitationally redshifted absorption lines in the X-ray burst spectra of a neutron star, Nature 420, 51

Gravitational redshift in neutron stars

Gravitational redshift in the laboratory

R.V.Pound & G.A.Rebka: Gravitational Red-Shift in Nuclear Resonance, Phys.Rev.Lett. 3, 439 (1959)

Robert Pound

Glen Rebka

Stellar spectroscopy

EFFECTS ON SPECTRAL LINES

L.Lindegren & D.Dravins: The fundamental definition of “radial velocity, A&A 401, 1185

Mechanisms causing wavelength shifts

The process includes: motion of the object; its emission of an electromagnetic signal; its propagation through space; motion of the observer; and the reception of the signal.

R.F.Griffin: Spectroscopic Binaries near the North Galactic Pole. Paper 6: BD 33° 2206, J.Astrophys.Astron. 3, 383 (1982)

Expected gravitational redshifts

Expected gravitational redshifts

D. Dravins

IAU Symp. 210

Radial velocities without spectroscopy

Astrometric radial velocities I

Dravins, Lindegren & Madsen, A&A 348, 1040

Astrometric radial velocities II

Dravins, Lindegren & Madsen, A&A 348, 1040

Astrometric radial

velocities from

perspective acceleration

Dravins, Lindegren & Madsen, A&A

348, 1040

Astrometric radial velocities III

Dravins, Lindegren & Madsen, A&A 348, 1040

STARPATHS (200,000 y)

Madsen, Dravins & Lindegren, A&A 381, 446

Pleiades from Hipparcos Proper motions over 120,000 years

Hyades from Hipparcos D. Dravins, IAU Symp. 215 (2004)

Hyades lineshifts Madsen, Dravins & Lindegren, A&A 381, 446

S.Madsen, D.Dravins, H.-G.Ludwig, L.Lindegren: Intrinsic spectral blueshifts in rapidly rotating stars?, A&A 411, 581

Apparent radial velocity vs. rotation?

Differential velocities within open clusters

B.Nordström, J.Andersen, M.I.Andersen: Critical tests of stellar evolution in open clusters II. Membership, duplicity, and stellar and

dynamical evolution in NGC 3680, Astron. Astrophys. 322, 460

Different “velocities” ─ giants vs. dwarfs?

Dean Jacobsen, astrophoto.net

M 67

L.Pasquini, C.Melo, C.Chavero, D.Dravins, H.-G.Ludwig, P.Bonifacio, R.De La Reza:

Gravitational redshifts in main-sequence and giant stars, A&A 526, A127 (2011)

Searching for gravitational redshifts in M67

M67 (NGC 2682) open cluster in Cancer contains some 500 stars;

age about 2,6 Gy, distance 850 pc.

M67 color–magnitude diagram with well-developed giant branch.

Filled squares denote single stars.

Dean Jacobsen, astrophoto.net

L.Pasquini, C.Melo, C.Chavero, D.Dravins, H.-G.Ludwig, P.Bonifacio, R.De La Reza:

Gravitational redshifts in main-sequence and giant stars, A&A 526, A127 (2011)

Searching for gravitational redshifts in M67

Radial velocities in M67 with a superposed Gaussian centered on

Vr = 33.73, σ = 0.83 km s−1

Radial velocities in M67: No difference seen between giants (red)

and dwarfs (dashed)

Real line formation

Solar disk

June 12, 2009

GONG/Teide

AN ”IDEAL” STAR ?

Granulation near the limb (towards the top) at 488 nm; Swedish 1-m solar telescope, La Palma

A REAL STAR

Solar Optical Telescope on board HINODE (Solar-B) G-band (430nm) & Ca II H (397nm) movies

Spectral scan

across the solar surface.

Left: H-alpha line

Right: Slit-jaw image

Big Bear Solar Observatory

“Wiggly” spectral lines

of solar granulation

“Wiggly" spectral lines in the solar photosphere

inside and outside a region of activity, reflecting

rising and sinking motions in granulation (wavelength increases to the right).

The central part crosses a magnetically active region

with reduced velocity amplitudes. (W.Mattig)

“Wiggly” spectral lines

of stellar granulation

(modeled)

Disk-center Fe I profiles from 3-D hydrodynamic model of the metal-poor star HD 140283 in NLTE and LTE.

Top: Synthetic “wiggly-line” spectra across stellar surface. Curves show equivalent widths W along the slit.

Bottom: Spatially resolved profiles; average is red-dotted.

N.G.Shchukina, J.Trujillo Bueno, M.Asplund, Astrophys.J. 618, 939 (2005)

Spatially resolved line profiles of the Fe I 608.27 nm line (exc = 2.22 eV) in a 3-D solar simulation.

The thick red line denotes the spatially averaged profile.

The steeper temperature structures in upflows tend to make their lines stronger (blue-shifted components).

M.Asplund: New Light on Stellar Abundance Analyses: Departures from LTE and Homogeneity, Ann.Rev.Astron.Astrophys. 43, 481

Solar-type granulation causes

convective lineshifts on order 300 m/s

”INVERTED” GRANULATION & CONVECTIVE REDSHIFTS ?

Simulated (top) and observed (Dutch Open Telescope) solar images in continuum (left) and

Ca II H wing (right).

(Leenaarts & Wedemeyer-Böhm, Astron.Astrophys. 431, 687, 2005)

Other stars

A.S.Brun, A.Palacios: Numerical Simulations of a Rotating Red Giant Star. I. Three-dimensional Models of Turbulent Convection and

Associated Mean Flows, ApJ 702, 1078

3-D modeling of stellar hydrodynamics

Cool supergiant (”Betelgeuse”)

Bernd Freytag (Uppsala)

STELLAR CONVECTION – White dwarf vs. Red giant

Snapshots of emergent intensity during granular evolution on a 12,000 K white dwarf (left) and a 3,800 K red giant. Horizontal areas differ by dozen orders of magnitude: 7x7 km2

for the white dwarf, and 23x23 RSun2 for the giant. (H.-G. Ludwig)

CORRUGATED STELLAR

SURFACES ?

Synthetic images

[negative] of

granulation in four

stellar models

From top:

Procyon (F5 IV-V),

Alpha Cen A (G2 V),

Beta Hyi (G2 IV), &

Alpha Cen B (K1 V).

Disk center (=1),

and two positions

towards the limb

D.Dravins & Å.Nordlund

Stellar Granulation IV.

Line Formation in

Inhomogeneous

Stellar Photospheres

A&A 228, 84

Fe I-line bisectors in Sun and Procyon

(F5 IV-V) Average bisectors for Fe I lines of different strength,

produced from a time-dependent 3-D model

C.Allende Prieto, M.Asplund, R.J.García López, D.L.Lambert: Signatures of Convection in the Spectrum of Procyon:

Fundamental Parameters and Iron Abundance, Astrophys.J. 567, 544

Bisectors of the same

spectral line in different

stars Adapted from

Dravins & Nordlund, A&A 228, 203

From left:

Procyon (F5 IV-V),

Beta Hyi (G2 IV),

Alpha Cen A (G2 V),

Alpha Cen B (K1 V).

Velocity [m/s]

Non-LTE effects on spectral line wavelength

shifts

Adapted from

Dravins & Nordlund, A&A 228, 184

L.Pasquini, C.Melo, C.Chavero, D.Dravins, H.-G.Ludwig, P.Bonifacio, R.De La Reza:

Gravitational redshifts in main-sequence and giant stars, A&A 526, A127 (2011)

Searching for gravitational redshifts in M67

Gravitational redshift predictions vs. mass/radius ratio (M/R) (dashed red)

do not agree with observations.

Calculated convective wavelength shifts for Fe I lines in dwarf (red crosses)

and giant models (squares).

Spatially resolved stellar spectroscopy

Solar granulation near the limb (upward on the image)

Filtergram at 488 nm; Swedish 1-m Solar Telescope on La Palma (G.Scharmer & M.G.Löfdahl)

Simulated intensities approaching

the solar limb

Mats Carlsson, Oslo; in

Å.Nordlund, R.F.Stein, M.Asplund:

Solar Surface Convection, Living Reviews in Solar

Physics, 2009

Exoplanet transits

EFFECT OF TRANSIT GEOMETRY ON THE ROSSITER EFFECT FOR HD209458

Left: Geometry of the crossing for 3 cases

Right: Radial- velocity anomalies for these cases, for different stellar rotation Vsini

D.Queloz, A.Eggenberger, M.Mayor, C.Perrier, J.L.Beuzit, D.Naef, J.P.Sivan & S.Udry

Detection of a spectroscopic transit by the planet orbiting the star HD 209458

Astron.Astrophys. 359, L13

RADIAL-VELOCITY

RESIDUALS FOR

THE STAR HD209458

TOP:

Outside planetary transit,

residuals are random errors.

BOTTOM:

During transit

(dashed line),

Rossiter effect appears.

ELODIE observations

D.Queloz, A.Eggenberger, M.Mayor,

C.Perrier, J.L.Beuzit, D.Naef

J.P.Sivan & S.Udry

Detection of a spectroscopic transit by

the planet orbiting the star HD 209458

Astron.Astrophys. 359, L13

Exoplanet transits (Rossiter-McLaughlin effect)

When an exoplanet transits a stellar disk,

successive center-to-limb contributions

to the integrated line profile are occulted,

with different effects for stars of different

rotational velocities.

Hiva Pazira (Lund Observatory)

Spatially resolved stellar spectroscopy

Synthetic line profiles

across stellar disks

Examples of synthetic line profiles from

hydrodynamic simulations of a solar-type star.

Curves are time-averaged line profiles, at

different positions and angles on the stellar disk.

Black curve is line profile at disk-center;

gradually lower intensities of blue, green, and

red curves reflect the limb darkening.

Solar model; Fe I, 620 nm, 1 eV.

Disk locations µ=1, 0.87, 0.59, 0.21;

azimuthal angles 0, 90, 180, 270°

Simulation by Hans-Günter Ludwig (Landessternwarte Heidelberg)

Spatially resolved stellar spectroscopy

Doppler imaging

DOPPLER

IMAGING

OF

STELLAR

SURFACES

S.S.Vogt, G.D.Penrod,

PASP 95, 565

Doppler imaging of stellar surfaces

For a star with a dark spot close to

the equator, spectral line profiles

are affected throughout their whole

Width, as the spot is carried around

the star by rotation.

(Jean-François Donati)

Doppler imaging of stellar surfaces

If the spot is located close to the pole,

spectral lines are only affected in their

core regions; the difference with the

previous case enables one to recover

information on both latitude and longitude

of starspots.

(Jean-François Donati)

Spectroscopy at very & extremely large telescopes

Visual high-resolution spectrometers at 8-10 m telescopes

Telescope SALT Keck I VLT

Kueyen

HET Subaru LBT

Diameter [m] 10 10 8.2 9.2 8.2 2 8.4

Spectrometer HRS HIRES UVES HRS HDS PEPSI

Maximum R 65,000 84,000 110,000 120,000 160,000 320,000

Wavelengths [µm] 0.37– 0.89 0.3 – 1.0 0.3 – 1.1 0.39 – 1.1 0.3 – 1.0 0.38 – 0.91

Potsdam Echelle Polarimetric and Spectroscopic Instrument @ Large Binocular Telescope

2080 cm R4 echelle grating for

PEPSI

HE 1523−0901: A strongly r-process-enhanced metal-poor star, [Fe/H] = − 2.95 , with detected uranium

(Figure by Klaus Strassmeier)

At B=12.1, PEPSI @ LBT needs 9 hours to reach S/N=850 at R=300,000

Resolving power and

spectral range of proposed

E-ELT spectrographs

Pasquini et al.: CODEX: the high resolution visual spectrograph for the E-ELT

Proc. SPIE 7014, 70141I (2008)

Optical arrangement of multi-camera CODEX design

40-m

European

Extremely

Large

Telescope

E-ELT on Cerro Armazones (artist’s impression)

Hiva Pazira (Lund Observatory)

Spatially resolved spectroscopy with ELTs

Left: Hydrodynamic simulation of the supergiant Betelgeuse (B.Freytag) Right: Betelgeuse imaged with ESO’s 8.2 m VLT (Kervella et al., A&A, 504, 115)

Top right: 40-m E-ELT diffraction limits at 550 nm & 1.04 μm..

Grand challenge: Design an efficient

R = 1,000,000 high-fidelity

spectrometer for E-ELT !