text
stringlengths
0
44.4k
take ve ltergrams across the Fe I line at 6173 A, separated by 69 m A (Borrero54 Gizon, Birch & Spruit
Figure 24: The HMI instrument to be
own in 2010 onboard NASA's Solar
Dynamics Observatory. Courtesy of Philip Scherrer.
et al. 2007). This line has a Lande factor g= 2:5 and therefore is better suited
for the measurement of the vector magnetic eld (than e.g. the Ni 6768 line). A
picture of the HMI
ight model is shown in Figure 24 . The HMI instrument was
delivered in November 2007 and has been integrated onto the SDO spacecraft.
At the time of writing, the launch of SDO is scheduled for February 2010 from
Cape Canaveral.
11.2 Solar Orbiter
Solar Orbiter is the next solar physics mission of the European Space Agency
(ESA) and a logical step after SOHO. The target launch date is 2015. Solar
Orbiter will use multiple gravity assist manoeuvres at Venus and the Earth such
that the inclination of the orbit to the ecliptic will incrementally increase during
the course of the mission (about 10 years) to reach heliographic latitudes of at
least 30. The elliptical orbit will have a minimum perihelion distance of 0 :22 AU.
The scienti c payload will include a remote sensing package that will deliver
0:5 arcsec pixel images of the solar photosphere (intensity, Doppler velocity, and
magnetic eld).
While the exact details of the orbit (and observation windows) are still be-
ing discussed, it is clear that Solar Orbiter will o er unique opportunities forLocal Helioseismology 55
helioseismology (Woch & Gizon 2007). First, it will be possible to study the sub-
surface
ows and structure in the polar regions, which is not possible today and
is important to understand the solar cycle. Second, Solar Orbiter will enable us
to test and apply the concept of stereoscopic helioseismology. Stereoscopic helio-
seismology combines observations from di erent vantage points. Solar Orbiter's
orbit is particularly interesting as it will o er a large range of spacecraft-Sun-
Earth angles. With observations from two widely di erent viewing angles (Solar
Orbiter and another Earth or near-Earth experiment), it becomes possible to
consider acoustic ray paths with very large separation distances (see Figure 6b).
This is useful in local helioseismology to probe structures deep into the Sun, and,
in particular, at the bottom of the convection zone.
12 SUMMARY AND OUTLOOK
Local helioseismology exploits the information contained in the local dispersion
relation of the acoustic and surface-gravity waves (ring-diagram analysis) and
in the correlations of the random wave eld (time-distance helioseismology and
related methods) in order to study the subsurface structure and dynamics of the
Sun in three dimensions. The high-quality observations from the GONG network
and the SOHO satellite have made possible the study of the properties of the
upper layers of the convection zone and their variations with the solar cycle.
Local helioseismology has not reached maturity and there are many open ques-
tions about data analysis methods and interpretation. The observational results
which, in our view, are the most robust and physically sensible are sketched in
Figure 25 and listed in the Summary Points below. Local helioseismology
measures e ects that are subtle, such as velocities of only a few m s1. In addi-
tion to approximations in the data interpretation, it is important to keep in mind
that several sources instrumental errors can a ect the measurements, e.g., plate
scale errors and optical distortion (Korzennik, Rabello-Soares & Schou 2004) or
uncertainties in the orientation of the image (Giles 2000).
An important challenge for future work in local helioseismology is to detect
signatures of magnetic elds at the base of the convection zone, where the eld
is expected to be ampli ed by di erential rotation and stored until erupting to
the surface as active regions. Direct detection through their e ect on wave prop-
agation properties is unlikely. Because of the high pressure at the base of the
convection zone, the contrast in propagation speed is very much lower than in
surface structures like sunspots, even at the inferred eld strengths of 105G.
More promising is the prospect of detecting systematic
ows that might be as-
sociated with magnetic structures at the base of the convection zone. Easiest
to detect would be azimuthal
ows (variations in rotation rate), such as have
already been reported on the time scale of the solar cycle. Even if the sensitivity
of helioseismic methods turns out insucient to detect such deeply seated struc-56 Gizon, Birch & Spruit
Figure 25:
tures, it may well be sucient to rule out certain less-preferred classes of models
for the solar cycle, such as convective dynamo models acting throughout the con-
vection zone or in a shallow surface layer. An important class of
ows would be
the geostrophic
ows caused by thermal e ects of magnetic elds (see Section 7).
Such disturbances are much easier to detect through their thermal winds than
directly by their temperature contrast. They might be turned into a diagnostic
of magnetic elds in deeper layers that can be probed with helioseismology.
The availability of powerful computers provides exciting opportunities to de-
vise, validate, and optimize improved methods of local helioseismology. Exploring
these possibilities will be key to taking full advantage of the observations of solar
oscillations. In Sections 4.2 and 6, we have shown examples of the usefulness of
numerical simulations of wave propagation through prescribed reference sunspot
models. Simulations of wave propagation in spherical geometry (e.g. Hanasoge
et al. 2006) have been used in time-distance studies of the deep convection zone
(e.g. Zhao et al. 2009) and to validate far-side imaging (Hartlep et al. 2008).Local Helioseismology 57
Figure 26: Radiative MHD simulation of a sunspot by Rempel et al. (2009). ( a)
Bolometric intensity (black and white) and subsurface magnetic eld strength
on a vertical cut through the center of the sunspot (in the range 0 { 8 kG). See
Supplemental Movie 11 . (b) Power spectrum of the surface oscillations in the
simulation. The blue line is the phase speed at the bottom of the box, above
which the model is not realistic.
It is now becoming possible to simulate the near surface layers of the Sun, in-
cluding pores and sunspots, by numerically solving the radiative MHD equations
(Rempel et al. 2009). Figure 26ashows a snapshot of the intensity and the
magnetic eld for a sunspot simulation. In this simulation, the solar oscillations
are naturally excited by the convection (see Figure 26b). This type of simulation
provides a means for computing realistic time series of Dopplergrams, which can
be used as input to all the methods of local helioseismology. With this type of
data set, it will be possible to resolve some of the outstanding issues, for example
regarding sunspot subsurface structure (Section 6.4). The recently achieved con-
vergence of observations and realistic 3D radiative MHD simulations of sunspots