Magnetic field and convection in Betelgeuse

Magnetic field and convection in Betelgeuse

M. Aurière, J.-F. Donati, R. Konstantinova-Antova, G. Perrin, P. Petit, T. Roudier

Roscoff, 2011 April 6

Outline

• Dynamo(s) in the Sun and cool stars

• The case of Betelgeuse • Spectropolarimetric detection of stellar magnetic fields • The cool supergiant Betelgeuse • Systematic field measurements in supergiant stars • Perspectives

The large-scale solar dynamo Helical motions

Differential rotation surface

QuickTime™ et un décompresseur TIFF (non compressé) sont requis pour visionner cette image.

QuickTime™ et un décompresseur TIFF (non compressé) sont requis pour visionner cette image.

tachocline

poloidal

toroidal

Combination of both effects (both linked to solar rotation)

toroidal

poloidal

Solar cycle Parker 1955

Some open questions about the solar dynamo

• Toroidal field generation : differential rotation (Parker 1955)  tachocline alone ?  convective zone as a whole ? (Brown et al 2010, Petit et al. 2008)

 what about the subsurface shear layer ? (Brandenburg 2005)

Poloidal field generation :  cyclonic convection ? (Parker 1955)  decay of active regions + meridional circ. ? (Dikpati et al. 2004)

Small-scale magnetism and solar dynamo Origin of small-scale (intranetwork) magnetic elements : • decay of active regions ? But: no or very limited variation over solar cycle • small-scale dynamo (Meneguzzi & Pouquet 1989, Cattaneo 1999 etc) ?

QuickTime™ et un décompresseur sont requis pour visionner cette image.

Lites et al. 2008 (Hinode observations)

Small-scale magnetism and solar dynamo Origin of small-scale (intranetwork) magnetic elements : • decay of active regions ? But: no or very limited variation over solar cycle • small-scale dynamo ? (Meneguzzi & Pouquet 1989, Cattaneo 1999 etc)

QuickTime™ et un décompresseur sont requis pour visionner cette image.

Vögler et al. 2007

Play with other stars to tune parameters

• How to make sure that small solar magnetic elements are not residuals from active regions, generated by the large-scale dynamo ? Observe a star without rotation (no global dynamo)

• How to resolve magnetic elements at the convective scale on a distant star ? Observe a star with huge convective cells

Betelgeuse : basic facts

Cool supergiant star • Teff = 3600 K • R = 600 - 800 Rsun , e.g. Perrin et al. 2004 (first stellar diameter ever measured, Michelson & Pease 1921)

• M ~ 15 Msun

• Prot ~ 17 yr (from space-resolved UV Doppler shifts)

HST/FOC

Convection in Betelgeuse

Giant convection cells (a few tens of cells on visible hemisphere vs ~ 106 cells on solar hemisphere)

• largest cells seen in nIR, lifetime ~ years • smaller cells in visible, lifetime ~ weeks (e.g. Schwarzshild 1975, Chiavassa et al. 2010, 2011)

Magnetic fields in Betelgeuse ? Prot ~ 17 yr

Ro = Prot/tconv >> 1 no solar dynamo expected

Convective dynamo simulations predict strong fields (500 G) with small filling factors (Dorch 2004) UV radius > optical radius (hot material above photosphere, Gilliland et al. 1996) … and : Radio radius > optical radius (cool material above photosphere, Lim et al. 1998)

Cool extended atmosphere coexists with hot extended atmosphere Ayres et al. 2003 report strongly absorbed lines of highly ionized species « Buried » coronal loops

Zeeman detection of stellar magnetic fields

J=1

J=0

Splitting of spectral lines in a magnetized atmosphere (proportional to field strength, unsensitive to field orientation) Zeeman 1896, Hale 1908 for the Sun, Babcock 1947 for a star

Zeeman detection of stellar magnetic fields

Zeeman splitting in a sunspot

Zeeman detection of stellar magnetic fields

J=1

J=0

Generally, B too weak to produce Zeeman splitting … but still able to polarize light in spectral lines

Zeeman detection of stellar magnetic fields

J=1

J=0

(Zeeman 1896) Light polarization controlled by strength and orientation of B

Extracting Zeeman signatures

• Generally, polarized Zeeman signatures signatures too weak to be detected in individual lines multi-line analysis (cross-correlation).

QuickTime™ et un décompresseur codec YUV420 sont requis pour visionner cette image.

Instrumental constraints

• Largest polarized Zeeman signatures in cool stars : V ~ 10-2Ic • For low-activity stars (e.g. solar twins) : V ~ 10-5Ic • Linear polarization (Q and U) ~ 10-2V ~ 10-7Ic for solar twins •

•

optimize the instrumental throughput (ESPaDOnS/NARVAL : about 15% including sky & detector) use large reflectors (ESPaDOnS/HARPSpol : 4m) perform accurate polarimetric analysis

•

resolve spectral lines (R > 30,000)

•

TBL, Pic du Midi NARVAL (2007) QuickTime™ et un décompresseur sont requis pour visionner cette image.

QuickTime™ et un décompresseur sont requis pour visionner cette image.

CFHT, Hawaii ESPaDOnS (2004)

QuickTime™ et un décompresseur sont requis pour visionner cette image.

La Silla, Chile HARPS (2010)

The magnetic field of Betelgeuse

Field detection using 15,000 photospheric atomic lines (note : thousands of molecular lines ignored)

QuickTime™ et un décompresseur TIFF (non compressé) sont requis pour visionner cette image.

B ~ 1 Gauss

Aurière et al. 2010

The magnetic field of Betelgeuse

QuickTime™ et un décompresseur TIFF (non compressé) sont requis pour visionner cette image.

QuickTime™ et un décompresseur TIFF (non compressé) sont requis pour visionner cette image.

Field variability < 1 month • much faster than stellar rotation • consistent with convective timescales (giant cells) Likely similar to « Quiet Sun » magnetism

Aurière et al. 2010

Velocity fields

Asymmetric Zeeman signatures generated by vertical gradients of magnetic fields & velocities QuickTime™ et un décompresseur TIFF (non compressé) sont requis pour visionner cette image.

Qui ckTime™ et un décompr esseur sont r eq ui s pour visionner cette image.

(Lopez Ariste 2002)

… seen also in solar intranetwork : QuickTime™ et un décompresseur sont requis pour visionner cette image.

Viticchié & Sanchez Almeida 2011

Are all cool supergiants magnetic ?

Grunhut et al. (2009) observed 30 late-type supergiants with 30% magnetic detections (weak fields) probably 100% of magnetic supergiants (assuming 5x better S/N) Quic kTime™ et un déc ompres s eur s ont requis pour v is ionner cett e image.

What happens to the 5-10% of strongly magnetic, main-sequence massive magnetic stars ? organized, strongly magnetic evolved stars (inclined dipole with ~500G field) Aurière et al. 2008 for EK Eri

Magnetic field often ignored in proposed processes creating highly structured wind to be reconsidered ?

Kervella et al. 2009 (NACO observations)

Perspectives

• Look for periodicities in field variability • Classical magnetic mapping prevented by long rot. period (17 yr) use simultaneous interferometry and spectropolarimetry use future ground-based solar facilities like ATST, EST. (AO + spectropolarimetry)

• Combine optical spectropolarimetry and UV spectroscopy UVMAG project (ask Coralie about that)