Presentation is loading. Please wait.

Presentation is loading. Please wait.

35 th COSPAR Scientific Assembly 18-25 July 2004 Probing the Solar Dynamo by observing the Sun’s high latitude field, flows and seismic waves Thierry Corbard.

Similar presentations


Presentation on theme: "35 th COSPAR Scientific Assembly 18-25 July 2004 Probing the Solar Dynamo by observing the Sun’s high latitude field, flows and seismic waves Thierry Corbard."— Presentation transcript:

1 35 th COSPAR Scientific Assembly July 2004 Probing the Solar Dynamo by observing the Sun’s high latitude field, flows and seismic waves Thierry Corbard Observatoire de la Côte d’Azur France

2 35 th COSPAR Scientific Assembly July 2004 Plan  Links between the Solar interior and magnetic cycle  Observed Surface Magnetic fields  Solar Dynamo models  Current helioseismic probe of the dynamo and limitations  Internal angular velocity from Global helioseismology  3D Subsurface flows from local helioseismology  Perspectives for Solar Orbiter

3 35 th COSPAR Scientific Assembly July 2004 Observed Surface magnetic Fields Courtesy: D. Hathaway; NASA/MSFC Equatoward migration of sunspot-belt Variation of amplitudes, duration etc..

4 35 th COSPAR Scientific Assembly July 2004 Observed Surface magnetic Fields Courtesy: D. Hathaway; NASA/MSFC Weak diffuse fields outside the sunspot drift poleward Polar reversal takes place during sunspot maximum Polar field take the sign of the following spots of the current cycle

5 35 th COSPAR Scientific Assembly July Ingredients for Solar Dynamo Models Parker 1970 Helical Turbulence (  -effect) Differential rotation Ω-effect Poloidal Field Toroidal Field + Meridional Circulation. + Flux Tube dynamics, buoyancy and stability Parker 75, Fan et al. 93, D’Silva & Choudhouri 93, Caligari et al. 95…

6 35 th COSPAR Scientific Assembly July Ingredients for Solar Dynamo Models Helical Turbulence (  -effect) Differential rotation Ω-effect Poloidal Field Toroidal Field + Meridional Circulation + Flux Tube dynamics, buoyancy and stability 2 Main Class of Dynamo models  Flux transport Dynamo (Wang et al. 91, Choudhuri et al. 95, Durney 95…)  Interface Dynamo (Parker 93, Tobias 96, Charbonneau & Mc Gregor 97….) Dikpati & Charbonneau 99, Petrovay & Kerekes 2004

7 35 th COSPAR Scientific Assembly July 2004 Evolution of Magnetic Fields in a Flux Transport Solar Dynamo Model Courtesy M. Dikpati + Addition of a tachocline  -effect is needed to satisfy Hale’s polarity rule (Dikpati & Gilman 2001)

8 35 th COSPAR Scientific Assembly July 2004 Global Helioseismology Angular degree l Frequency (mHz) n=0 n=1 n=2 n=3 : Internal Rotation breaks spherical symmetry => Multiplets of 2 l+ 1 frequencies => 2D Inverse Problem observedStandard Solar Model Unknown

9 35 th COSPAR Scientific Assembly July 2004 Global Helioseismology Results: Internal Rotation Tachocline Equator Pole Period (days) “Polar Jet” ? Corbard 98Schou et al. 2002

10 35 th COSPAR Scientific Assembly July 2004 Global Helioseismology Results: Torsional Oscillations Pattern found to persist from the surface down to (0.82R) Howe et al Indications of even deeper penetration at high latitudes Vorontsov et al likely a back reaction from the Dynamo generated magnetic field Adapted from Corbard & Thompson 2001

11 35 th COSPAR Scientific Assembly July 2004 Global Helioseismology Limitations  Sensitive only to the part of the rotation that is symmetric about the equator.  Kernels Independent of longitude (e.g. cannot detect effect of active region on rotation)  Limited access to the polar zone  Impossible to separate the spherically asymmetric effects other than rotation (meridional circulation, magnetic fields, structural asphericity)

12 35 th COSPAR Scientific Assembly July 2004 Local Helioseismology Analysis Principles  Small areas (16ºx16º) are tracked (typically between 8hrs and 28hrs)over the solar disk at a rate depending on the latitude of their center.  These areas are remapped using a Postel or transverse cylindrical projection that tend to preserve the distance along great circles.  => Data cubes (Latitude – Longitude – time)

13 35 th COSPAR Scientific Assembly July 2004 Local Helioseismology Main Methods 1.Time-distance analysis (Time domain) (Duvall et al. 1993, 1996) 2.Ring-Diagram analysis (Fourier domain) (Gough & Toomre 1983; Hill 1988) 3.Acoustic Holography (Phase sensitive) (Lindsey & Braun, 1990, 1997) Based on the local observation of global acoustic modes.

14 35 th COSPAR Scientific Assembly July 2004 Time-distance helioseismology Principles  First developed by Duvall 1993  Employs waves travel times observed between different surface location on the Sun.  These travel times are then “inverted” to deduce:  Flow speed and direction,  Sound speed perturbations  [Magnetic field perturbations] Along the ray paths of the observed modes.

15 35 th COSPAR Scientific Assembly July 2004 Time-distance helioseismology Measuring Travel times  These travel times are obtained by cross-correlating the observed surface oscillations.  Because of the stochastic nature of excitation of the oscillations, the cross-covariance function must be averaged over some areas on the solar surface to achieve a S/N sufficient for measuring travel times. =>Averages are made on rings around the central point Kosovichev & Duvall., 1997  

16 35 th COSPAR Scientific Assembly July 2004 Time-distance helioseismology Interpretation of travel time perturbations Fermat’s Principle and ray approximation theory Phase travel time perturbation Ray path in the unperturbated medium Perturbation of the wave number Gough 1993 Kosovichev 1996  Bogdan 97: wavepacket are not confined to the raypath  Development of wave-theoretical sensitivity kernels: Jensen et al. (2000), Birch & Kosovitchev (2000), Jensen & Pijpers (2002), Gizon & Birch (2002)  First comparisons (e.g. Couvidat et al. (2004)) show reasonable agreement for depths above 15Mm

17 35 th COSPAR Scientific Assembly July 2004 Time-distance helioseismology Interpretation of travel time perturbations Dispersion relation:  Temperature perturbations => do not depend on the direction of propagation  Flow perturbations => waves move faster along the flow  Magnetic fields perturbations => waves traveling perpendicular to the field lines are the most sensitive to B (wave speed anisotropy not yet detected) Ryutova & Scherrer D velocity fieldAcoustic cut off frequency Alvén velocity (Kosovichev et al. 1997)

18 35 th COSPAR Scientific Assembly July 2004 Time-distance helioseismology Inverting Travel Times Maps Annulus ranges: 1º19 - 1º598 Zhao Different annulus ranges sound different depths Inversions for sound speed perturbations and 3D flows as a function of depth

19 35 th COSPAR Scientific Assembly July 2004 Time-distance helioseismology Results: 3D Flows below sunspot 0-3 Mm6-9 Mm 9-12 Mm Zhao, Kosovichev,Duvall 2001 Downward flows Upward flows

20 35 th COSPAR Scientific Assembly July 2004 Time-distance helioseismology Sub sunspot dynamics and structure Courtesy SoHO/MDI The cluster Model of Sunspot Parker 1979 Illustration of sound speed variations and subsurface flow patterns of a sunspot

21 35 th COSPAR Scientific Assembly July 2004 Ring Diagram Analysis kxkx kyky  3D FFT ( , ,t)  (k x,k y,  ) 3D Power spectra Fitting 2 x 1D Inversions V z may then be computed using the divergence of the horizontal flow and assuming mass conservation. (Komm, Corbard et al. 2004)

22 35 th COSPAR Scientific Assembly July 2004 Ring Diagram Analysis Flow Maps

23 35 th COSPAR Scientific Assembly July 2004 Synoptic Maps of Vorticity – Kinetic Helicity etc.. Komm, Corbard et al (ring diagrams) Zhao & Kosovichev 2003 (time-distance)

24 35 th COSPAR Scientific Assembly July 2004 Time and depth variations of Meridional Flows Basu & Antia 2002 Gonzalez Hernandez et al. 2004

25 35 th COSPAR Scientific Assembly July 2004 Controversial detection of a counter cell at “high latitudes” in the Northern hemisphere Haber et al Not found by Basu and Antia 2003 And Time- distance analysis Zhao and Kosovichev 2003

26 35 th COSPAR Scientific Assembly July 2004 Dynamo Models and Polar Reversal N-S asymmetry in meridional flow speed during and the appearance of a reverse, high-latitude flow cell in the N- hemisphere during caused the N-pole to reverse ~1 yr before the S- pole Dikpati et al. 2003

27 35 th COSPAR Scientific Assembly July 2004 Perspectives for The Solar Orbiter  The Visible Light-imager and Magnetograph (VIM) will provide high resolution Magnetograms and Dopplergrams from outside the ecliptic plane (up to 30º).  Immediate benefit: Reduced foreshortening and Increased sensitivity to Magnetic Fields (factor 5)  Possibility of stereoscopic observations with SDO or Earth (120º)  Observations will provide the missing parts of solar dynamo models  Polar rotation rate and torsional oscillation pattern (High latitude Jets?, lower rotation bands migrating poleward ?)  Polar meridional flows (counter cell, returning flow ?)  Maps of high latitude vorticity / kinetic helicity / Small scale Dynamo  Direct observations of polar reversal and flux cancellation processes  Access to the deep layers (tachocline) from stereoscopic observations  Access to the far side of the Sun (continuous probe of spots / far side imaging technique)

28 35 th COSPAR Scientific Assembly July 2004 Acoustic Holography Far Side Imaging Lindsey & Braun (1997) A local acoustic depression at the focal point will shift the phase of the ingression-egression correlation.

29 35 th COSPAR Scientific Assembly July 2004 Concluding Remarks  The field of Local Helioseismology is fairly new  A lot of efforts have been made during the last few years in:  Understanding the forward problem (wave-theoretical sensitivity kernels / Building of artificial data)  Making comparison between the different Methods (group LoHCO)  Understanding errors, methods resolution and correlations  Understanding effects of image misalignment / geometric effects (P-angle / Bo angle)  First Results are very encouraging and positive  First attempt of “archeo-helioseismology” using Mt Wilson data in order to infer the meridional circulation in the past two solar cycles  There is no doubt that a lot more progress will be achieved in developing and understanding these methods before the launch of Solar Orbiter


Download ppt "35 th COSPAR Scientific Assembly 18-25 July 2004 Probing the Solar Dynamo by observing the Sun’s high latitude field, flows and seismic waves Thierry Corbard."

Similar presentations


Ads by Google