Solar irradiance variations affect Earth's climate, but the magnitude of
the Sun's intrinsic variation is uncertain. Current observations cannot
reject the possibility that intrinsic variations played a major role in the
climate changes recorded over the past few millennia. Physical
understanding, based on images of the sources of irradiance variation, will
clarify the Sun's role in global climate change. From space-borne
bolometric radiometers, we know that during the 11-year sunspot cycle the
total solar irradiance (TSI) varies in proportion to local magnetic
fields. Why do we propose to study irradiance at the upcoming sunspot
minimum, when the local fields will be weakest? Only then can observations
detect other possible sources of TSI variation with the least confusion by
the large amplitude signals from local magnetic fields. This is the best
observational approach to physical understanding of the possible long-term
TSI variations. For the second flight we propose to operate the Solar
Bolometric Imager (SBI-2)
above Antarctica, where near-space conditions can be attained for 10-20
days. SBI will provide bolometric (wavelength-integrated
light) and color temperature images from which we may assess both the
irradiance signals and their underlying physical causes. Images are
necessary to characterize irradiance variations associated with subtle
magnetic structures, acoustic oscillations, pole-equator temperature
differences, and rotational-convective cells. The investigation builds on
the successful flight of the SBI-1 on September 1,
2003.
Scientific Goals and Objectives
Background
Solar variability is the main external driver in climate change. Solar
driving of climate may occur on timescales from decades to eons through a
variety of mechanisms (reviewed by Lean 1997; Rind 2002). Heating of the
troposphere by the total solar irradiance (TSI) is the most direct effect
of the Sun on the Earth. Driving of some Ice Age cycles is well explained
by the orbital changes described by the Milankovitch mechanism. A question
at the research frontier is: Does luminosity variation intrinsic to the Sun
drive climate change on the multi-decadal timescales relevant to the global
warming problem?
Space-borne full-disk bolometry (Hickey et al. 1980; Willson et al. 1981)
permits quantitative investigation of this crucial question. The
irradiance variation on solar rotational and 11-yr timescales has been
correlated successfully with photospheric magnetic structures such as
spots, faculae, and network (reviewed by Hudson 1988). The irradiance
variation on longer, multi-decadal timescales is not well characterized.
The observed TSI variation over the past 25 years is small and uncertain.
Correlative studies overlook mechanisms, as described below, that could
dominate long-term variation. Simple monitoring of TSI variations has
yielded controversial results on long-term
irradiance variations. Description of long-term irradiance impacts on
past and future climate depends on improved empirical and physical insight
into TSI variation (Foukal 2003).
For the observational part of our program, we will characterize the
bolometric signals from the known and predicted spatial structures on the
Sun. The observations will provide a critical measurement of the sources of
irradiance variation at solar minimum. They will also be a baseline for
understanding variation over the sunspot cycle and for comparison with
future sunspot minima.
We will critically challenge current models of solar intensity structure
and irradiance variation. These include: the standard model with its
emphasis on surface magnetic field contrasts; numerical models of
convective and large-scale structure; and analytic models of disturbance of
heat flow by magnetic fields. Validation of the models would provide
confidence in our understanding of the critical physical processes of
irradiance variation. Discrepancies between the observations and models
would indicate areas requiring further physical understanding or the
identification of new physical processes.
In the following sections we detail those goals and objectives.
The present model of irradiance variation and its limitations
For the past two decades, radiometric
observations have been available from many, overlapping TSI
instruments. Today the suite includes SOHO/VIRGO,
ACRIMSAT/ACRIM3, and SORCE/TIM. We have
learned that much of the TSI variation is correlated with surface magnetic
fields (e.g., Chapman 1986; Walton et al. 2003). In the standard empirical
model, all irradiance variation is caused by changes in the projected areas
and contrasts of photospheric magnetic structures - sunspots, faculae, and
network. Other known contributors to TSI variation such as acoustic
oscillations and convection simply produce irradiance noise of constant low
level. Global flows may have thermal structures with this property also.
The Sun's
surprising brightening around activity maximum is caused by the dominance
of bright magnetic faculae over spot darkening, in determining the 11-yr
irradiance cycle. The disappearance of solar magnetic activity at sunspot
minimum causes a return to a constant baseline value of TSI. No secular
variation of TSI from other sources is expected, according to the standard
model.
The standard model can be tested with time series of TSI observations and
measurements of the areas and contrasts of the magnetic structures. On the
rotational time scale, testing of the model is limited by current
uncertainties in the broadband photometric contribution (area and contrast)
of faculae. One goal of this proposal is to improve this accuracy for
comparison with radiometry. The comparison can teach us whether
photospheric magnetic structures account for essentially all irradiance
variation on the rotational time scale, or whether, even on this short time
scale, there is evidence for other influences from e.g. "convective
stirring" (Parker, 1994). This distinction would have fundamental
implications for the "standard" model.
What solar photometric properties are best studied at activity
minimum?
The standard model is incomplete, as it excludes nonmagnetic sources of
irradiance variation. These other sources are best studied at activity
minimum, when noise from magnetic structures is least. Nonmagnetic sources
include known photospheric structures such as convective cells and acoustic
oscillations and inferred structures such as pole-equator temperature
gradients and global variations of effective temperature.
We expect that all measurable sources of irradiance variation will have
distinct spatial patterns over the solar disk, just as surface magnetic
fields and convection cells have. Variation of TSI will occur from an
unbalanced summation of the local contrasts. For example, the correlation
of TSI with the disk-integrated magnetic flux is nil, while the correlation
with the sum of the local magnetic signals is high. Therefore we propose
to use bolometric images to understand the localized sources of irradiance
variation.
The above figure, generated from a SOHO/MDI magnetic and intensity image
pair, illustrates the variety of structures to be studied. From left to
right the structures
range from the observed limb-darkening with an amplitude of ±1000K
centered at 5250K; through surface
magnetic fields in form of spots and faculae signals, scaled -50±150K;
convection, and acoustic oscillations signals, scaled ±30K; predicted
global convective and flow structures like giant cells, pole-equator
gradient, and torsional waves, with predicted amplitudes of ±3K (far
right).
Known thermal structures have irradiance signals
The extraordinary precision of TSI observations permits the detection of
very weak signals from a variety of known thermal structures. Power
spectra quickly showed 3 mHz (5-minute) acoustic oscillations at a few
parts-per-million (ppm) per mode (Woodard and Hudson 1983). At lower
frequencies a continuous spectrum of irradiance fluctuations is seen,
identified with convective noise and active region signals (Pelletier 1996;
Vazquez Ramio 2002). Jimenez et al. (1999) made the correspondence between
the TSI signal and the spatially resolved acoustic patterns. No equivalent
correspondence between TSI and convective patterns exists, leaving an
obvious gap in characterizing irradiance variation, which we propose to
fill.
Predicted irradiance signals have not been detected in TSI
observations
Physical modeling of energy and momentum flow inside the Sun predicts
several large-scale thermal structures. The highest amplitude is a
pole-equator temperature difference (DeRosa 2002). The extant observations
are limited (Altrock and Canfield 1972) and do not address time variations.
Helioseismic observations are beginning to measure the form (Braun and Fan
1998) and variation (Chou and Dai 2001) of the meridional flow that is
coupled to a latitudinal temperature variation. We will search for this
pattern to test physical understanding of the solar interior and our
techniques for probing that volume.
The radial temperature gradient of the photosphere produces the largest
intensity signal − the limb-darkening. Limb-darkening is normally
treated as a constant, but the dynamic photosphere alters its profile
locally. Petro, Foukal, and Kurucz (1985) modeled the range of variations
expected, given changes in the source function, effective temperature, and
convection in the photosphere, and showed that irradiance variations might
be detectable with precise photometry. Separation of the contributions to
limb-darkening variation requires multispectral photometry (Foukal and
Duvall 1985). Typical fractional limits on the stability of limb-darkening
are 10-2 (Livingston and Wallace 2003); only Petro et al. (1984)
were able to approach 10-3, which corresponds to realistic TSI
signals. This figure
shows the first bolometric limb-darkening observation a full-disk mosaic
made during the SBI-1 flight. The general agreement with model values
indicates that data calibration is accurate. When the full flight data are
reduced, the precision of the bolometric curve will exceed that of the
models. On the proposed next flight we will achieve more precise
photometry, to understand the stability of limb-darkening and its
correlation to TSI variation. Multi-spectral images will diagnose the
photospheric temperature structure.
The high thermal conductivity of the solar interior redistributes much of
the heat flow disturbed by surface magnetic fields, but the redistribution
is not perfect. Some of the heat blocked by sunspots should leak out in
surrounding bright rings with amplitudes of 1 to 5K (Spruit 1977).
Ground-based observations have limits (Fowler, Foukal, Duvall 1983; Rast et
al. 2001) near the upper end of the predicted range. Large-scale thermal
structures are also predicted to form near large concentrations of surface
magnetic fields (Spruit 2003). The corresponding pressure gradients
accelerate mass flows that couple with the Coriolis force to generate the
flow pattern seen in the torsional wave (Vorontsov et al. 2002). The
predicted amplitude, ~0.3K, is too low to be detected from ground-based
photometry in the presence of seeing and transparency fluctuations. The
model also fails to explain the presence of the torsional wave during
sunspot minimum, when surface magnetism is weak. We propose to search for
thermal structures, to test our understanding of heat flow in the solar
convection zone.
Magnetic effects are complex
The success of the standard model in correlating TSI variation with surface
magnetic fields derives from the large feature contrasts. Large areas of
magnetic field in sunspots block convective flow, permitting the surface to
cool and appear dark. Smaller areas of magnetic field in faculae reach
horizontal pressure balance at lower gas density and opacity than the
nonmagnetic surroundings, enabling deeper, hotter layers to radiate excess
flux and appear bright. Despite its success, correlative modeling uses
constant, typical values for many parameters such as spot temperatures,
umbra/penumbra area ratio, center-to-limb contrast function, and so on. To
improve precision in understanding irradiance variation, we propose to
replace these values by direct measurement, in multispectral and bolometric
images.
The standard model segments the solar disk into sunspots, faculae, and
network, to assess their irradiance signals. Those structures are not
mutually exclusive, but may overlap. Some small sunspots and penumbral
regions appear dark in the continuum and bright in facular images. The
standard model counts those areas twice with both positive and negative
signals. Those areas are actually dominated by horizontal or rapidly
diverging magnetic fields and have an unusual temperature gradient in the
photosphere. Observations at any resolution will necessarily smooth over
such structures, but our proposed observations will do so in a
bolometrically correct manner. Our multispectral images will
simultaneously identify magnetic structures.
Further, the parameters used in the standard model may vary through the
sunspot cycle (Albregtssen, Joras, and Maltby 1984; Ermolli, Berilli, and
Florio 2003) in response to variations of the magnetic field vector
orientation, scale length (ephemeral regions vs active regions), flux
density, and so on. Even with detailed observations of magnetic structure,
our proposed bolometric images are needed to understand the impact on
irradiance variation.
References
Albregtsen, F., Joras, P. B., and Maltby, P., Limb-darkening and solar
cycle variation of sunspot intensities, Solar Phys., 90, 17 (1984).
Altrock, R. C., and Canfield, R. C., Observations of Photospheric
Pole-Equator Temperature Differences, Solar Phys., 23, 257 (1972).
Braun, D. C., and Fan, Y., Helioseismic Measurements of the Subsurface
Meridional Flow, Astrophys. J., 508, L105 (1998).
Chapman, G. A., and Boyden, J. E., Solar Irradiance Variations Derived from
Magnetograms, Astrophys. J., 302, L71 (1986).
Chou, D.-Y., and Dai, D.-C., Solar Cycle Variations of Subsurface
Meridional Flows in the Sun, Astrophys. J., 559, L175 (2001).
DeRosa, M. L., Gilman, P. A., and Toomre, J., Solar Multiscale Convection
and Rotation Gradients Studied in Shallow Spherical Shells, Astrophys. J.,
581, 1356 (2002).
Ermolli, I., Berrilli, F., and Florio, A., A measure of the network
radiative properties over the solar activity cycle, Astr. Astrophys, 412,
857 (2003).
Foukal, P., Can Slow Variations in Solar Luminosity Provide Missing Link
Between the Sun and Climate?, EOS, 84, 205 (2003).
Foukal, P., and Duvall, T., Differential Photometry of Magnetic Faculae,
Ap. J., 296, 739 (1985).
Fowler, L. A., Foukal, P., and Duvall, T., Sunspot Bright Rings and the
Thermal Diffusivity of Solar Convection, Solar Phys., 84, 33 (1983).
Lean, J., The Sun's Variable Radiation and Its Relevance For Earth,
Ann. Rev. Astron. Astrophys., 35, 33 (1997).
Livingston, W., and Wallace, L., The Sun immutable basal quiet atmosphere,
Solar Phys., 212, 227 (2003).
Hickey, J. R., Stowe, L. L., Jacobowitz, H., Pellegrino, P., Maschhoff,
R. H., House, F., and Vonder Haar, T. H., Initial Solar Irradiance
Determinations from Nimbus 7 Cavity Radiometer Measurements, Science, 208,
281 (1980).
Hudson, H. S., Observed Variability of the Solar Luminosity,
Ann. Rev. Astron. Astrophys., 26, 473 (1988).
Jimenez, A., Roca Cortes, T., Severino, G., and Marmolino, C., Phase
Differences and Gains Between Intensity and Velocity in Low-degree Acoustic
Modes Observed by SOHO, Astrophys. J., 525, 1042 (1999).
Parker, E. N., Theoretical properties of Omega-loops in the convective zone
of the sun. 2: The origin of enhanced solar irradiance, Astrophys. J., 440,
415 (1994).
Pelletier, J. D., Variations in Solar Luminosity from Timescales of Minutes
to Months, Astrophys. J., 463, L41 (1996).
Petro, L. D., Foukal, P. V., and Kurucz, R. L., Photospheric Limb-darkening
Signatures of Global Structure Variations, Solar Phys., 98, 23 (1985).
Petro, L. D.; Foukal, P. V.; Rosen, W. A.; Kurucz, R. L.; Pierce, A. K., A
study of solar photospheric limb-darkening variations, Astrophys. J., 283,
426 (1984).
Rast, M. P.; Meisner, R. W.; Lites, B. W.; Fox, P. A.; White, O. R.,
Sunspot Bright Rings: Evidence from Case Studies, Astrophys. J., 557, 864
(2001).
Rind, D. H., The Sun's Role in Climate Variations, Science, 296, 673
(2002).
Spruit, H. C., Heat Flow Near Obstacles in the Solar Convection Zone, Solar
Phys., 55, 3 (1977).
Spruit, H. C., Origin of the Torsionsional Oscillation Pattern of Solar
Rotation, Solar Phys., 213, 1 (2003).
Vazquez Ramio, H., Roca Cortes, T., and Regulo, C., Background Solar
Irradiance Spectrum at High and Low Phases of the Solar Activity Cycle, In
Solar Variability: From Core to Outer Frontiers, ESA SP-506, 897
(2002).
Vorontsov, S. V.; Christensen-Dalsgaard, J.; Schou, J.; Strakhov, V. N.;
Thompson, M. J., Helioseismic Measurement of Solar Torsional Oscillations,
Science, 296, 101 (2002).
Walton, S. R., Preminger, D. G., and Chapman, G. A., The Contribution of
Faculae and Network to Long-term Changes in the Total Solar Irradiance,
Astrophys. J., 590, 1088 (2003).
Willson, R. C., Gulkis, S., Janssen, M., Hudson, H. S., and Chapman, G. A.,
Observations of Solar Irradiance Variability, Science, 211, 700 (1981).
Woodard, M., and Hudson, H, Solar oscillations observed in the total
irradiance, Solar Phys., 82, 67 (1983).
