sign in sign up
<<
Home , Magnetogram, Picard (satellite), Solar observatory, Solar Sentinels, Stereo, Vector magnetograph

Solar Dynamics Observatory

“…to understand the and source of the variations that affect life and society.”

Report of the Science Definition Team Solar Dynamics Observatory Science Definition Team

David Hathaway John W. Harvey K. D. Leka Chairman National Colorado Research Division Code SD50 P.O. Box 26732 Northwest Research Assoc. NASA/MSFC Tucson, AZ 85726 3380 Mitchell Lane Huntsville, AL 35812 Boulder, CO 80301

Spiro Antiochos Donald M. Hassler David Rust Code 7675 Southwest Research Institute Applied Physics Laboratory Naval Research Laboratory 1050 Walnut St., Suite 426 Johns Hopkins University Washington, DC 20375 Boulder, Colorado 80302 Laurel, MD 20723

Thomas Bogdan J. Todd Hoeksema Philip Scherrer High Altitude Observatory Code S HEPL Annex B211 P. O. Box 3000 NASA/Headquarters Stanford University Boulder, CO 80307 Washington, DC 20546 Stanford, CA 94305

Joseph Davila Jeffrey Kuhn Rainer Schwenn Code 682 Institute for Astronomy Max--Institut für Aeronomie NASA/GSFC University of Hawaii Max Planck Str. 2 Greenbelt, MD 20771 2680 Woodlawn Drive Katlenburg-Lindau Honolulu, HI 96822 D37191 GERMANY

Kenneth Dere Barry LaBonte Leonard Strachan Code 4163 Institute for Astronomy Harvard-Smithsonian Naval Research Laboratory University of Hawaii Center for Astrophysics Washington, DC 20375 2680 Woodlawn Drive 60 Garden Street Honolulu, HI 96822 Cambridge, MA 02138

Bernhard Fleck Judith Lean Alan Title ESA Space Science Dept. Code 7673L Lockheed Martin Corp. c/o NASA/GSFC Naval Research Laboratory 3251 Hanover Street Code 682.3 Washington, DC 20375 Palo Alto, CA 94304 Greenbelt, MD 20771

Richard Harrison John Leibacher Roger Ulrich CCLRC National Solar Observatory Department of Astronomy Chilton, Didcot P.O. Box 26732 UCLA, 9831 MSB Oxfordshire OX11 0QX Tucson, AZ 85726 Los Angeles, CA 90024 UNITED KINGDOM

Barbara Thompson Project Scientist Code 682 NASA/GSFC Greenbelt, MD 20771 Table of Contents 1 Executive Summary ...... 1 2 Overview...... 3 3 Current Scientific Understanding and Outstanding Questions ...... 8 3.1 Solar Influences on Global Change and ...... 8 3.1.1 Irradiance Variations...... 8 3.1.2 Energetic Particles from Flares and CMEs ...... 13 3.1.3 Coronal Structure and Solar Variations...... 17 3.2 Mechanisms of Solar Variability...... 20 3.2.1 The ...... 21 3.2.2 Active Region Evolution...... 24 3.2.3 Small-Scale Magnetic Structure...... 27 4 Required Observations ...... 32 4.1 Helioseismic Images...... 32 4.2 Longitudinal ...... 33 4.3 Atmospheric Images...... 34 4.4 EUV Spectral Irradiance...... 35 4.5 Photometric Images ...... 35 4.6 Vector Magnetograms ...... 36 4.7 UV/EUV Spectra...... 38 4.8 Coronagraphic Images...... 39 4.9 Total Irradiance ...... 40 4.10 Coronal Spectroscopy...... 41 4.11 Heliometry ...... 42 5 Potential Instruments and Allocation of Resources...... 42 6 Mission Concept...... 43 6.1 Orbit Selection...... 44 6.2 System ...... 45 6.3 Data and Communication System ...... 46 6.4 Spacecraft Power...... 47 6.5 Instrument Module ...... 47 6.6 Ground System...... 47 6.7 Mission and Science Operations ...... 48 7 Concurrent Observations...... 49 7.1 STEREO ...... 49 7.2 Solar-B...... 49 7.3 Solar Probe...... 49 7.4 SORCE ...... 51 7.5 GOES/NPOESS ...... 52 7.6 SOLIS ...... 53 7.7 ATST ...... 53 7.8 ...... 54 7.9 ...... 54 7.10 FASR ...... 55 8 Acknowledgements ...... 54 1 Executive Summary advance in modeling, quantifying, and perhaps eventually predicting solar variability over the The Solar Dynamics Observatory (SDO) is a relevant timescales. Variable solar outputs nec- cornerstone mission within the Living With a essarily impact parallel efforts aimed at under- (LWS) program. SDO’s mission is to un- standing, modeling, and predicting the behavior derstand the nature and source of the solar vari- of the - system. Accordingly, we also ability that affects life and society. As such, its speak to the scientific issues and problems that principal functions are two-fold. First, it must arise from inadequate knowledge of solar radia- make accurate measurements of those solar pa- tive, particulate, and magnetic outputs. rameters that are necessary to provide a deeper physical understanding of the mechanisms that Together, the solar and terrestrial science ques- underlie the Sun’s variability on timescales tions and ongoing research activities point to ranging from seconds to centuries. Second, specific observables that must be supplied by through remote sensing, it must monitor and re- SDO for the successful operation of the LWS cord those aspects of the Sun’s variable radia- science program. This document identifies a set tive, particulate, and magnetic plasma outputs of required observables that best addresses the that have the greatest impact on the terrestrial dual objectives of SDO. In so doing, it provides environment and the surrounding . the basic arguments and the supporting evidence that leads to the identification of the essential Our Sun is an active star. This activity impacts observations. The nature of the required meas- planet Earth and human society in numerous urements in turn drives the choice of a geosyn- ways. Terrestrial climate, ozone concentrations chronous (GEO) orbit for the spacecraft. A po- in the stratosphere, and atmospheric drag on sat- tential suite of instruments is described which is ellites all respond to variations in the Sun’s ra- in principle capable of acquiring the required diative output. , airline passengers, observables, while at the same time satisfying and electronics are all imperiled by the the logistical constraints imposed on the SDO. energetic particles produced in solar flares and coronal mass ejections (CMEs). Electrical power The overarching science questions to be ad- to our homes and businesses, communications, dressed by the SDO are: and navigation systems can all be interrupted by geomagnetic storms driven by blasts in the solar · What mechanisms drive the quasi-periodic wind. SDO will study the mechanisms of solar 11-year cycle of solar activity? variability – through a broad spectrum of tempo- · How is active region magnetic flux synthe- ral, spatial, and energetic scales – to provide the sized, concentrated, and dispersed across the tools and scientific understanding that will en- solar surface? able us to improve the quality of forecasts of · How does on small solar activity. SDO will also provide the meas- scales reorganize the large-scale field topol- urements that are critical as input to studies of ogy and current systems? How significant is the geospace environment at their point of origin it in heating the corona and accelerating the in the Sun-Earth system in order to quantify the ? Sun’s influence on global change and improve · Where do the observed variations in the our characterizations and forecasts of space Sun’s total and spectral irradiance arise, and weather. how do they relate to the magnetic activity cycles? In this report we discuss several key science · What configurations lead to questions that derive from our present incom- the CMEs, filament eruptions, and flares that plete knowledge of the physical underpinnings produce energetic particles and ? of the Sun’s variability. They are selected for the · Can the structure and dynamics of the solar promise they hold in catalyzing a significant wind near Earth be determined from the magnetic field configuration and atmos- listed below satisfy the data requirements stated pheric structure near the solar surface? above while remaining within the financial and · When will activity occur, and is it possible logistical limitations placed upon the SDO mis- to make accurate and reliable forecasts of sion. space weather and climate? · Helioseismic/Magnetic Imager The answers to these pressing science questions · Atmospheric Imaging Array are to be found in direct observations of the · EUV Spectral Irradiance Monitor relevant solar activity and the interpretation of · these data. To this end, the complement of SDO · Photometric Mapper instruments should supply the following basic · UV/EUV Spectrometer data types (ordered from solar interior outward, · not by priority). The first three instruments are of highest prior- · Full-disk dopplergrams of appropriate spa- ity. SDO must include instruments such as these tial and temporal resolution, duration and to fulfill its mission. All three provide data with continuity to permit accurate helioseis- proven value that are not likely to be supplied mological inferences of conditions in the through other programs. The final four instru- solar interior. ments are of high priority. The data they obtain · Full-disk magnetograms capable of charac- are needed for the SDO mission but they may be terizing the surface magnetic field and its supplied in some other, albeit compromised, evolution, and monitoring the emergence form by other programs, may be more specula- and processing of magnetic flux. tive in nature, or may tax the SDO resources. A · Full-disk precise photometric images to ex- Total Solar Irradiance Monitor is also consid- plore the temporal and spatial variability of ered to be of highest priority, but two TSI the solar irradiance and determine couplings Monitors are expected to fly on other platforms with the magnetic structures. concurrently with SDO. · Full-disk filtergrams recorded simultane- ously in a variety of visible and EUV band- The remainder of this document amplifies on the passes to assess the dynamics and energetics considerations that went into this selection of of the solar atmosphere on global and active questions, critical data, and instrument comple- region scales. ment. It provides the traceability from the scien- · Sun-as-a-star EUV spectral irradiance meas- tific questions to the instrument requirements urements to monitor and record temporal and mission design. variations of radiative outputs crucial for gauging ionospheric, mesospheric and ther- mospheric responses to solar forcing. · Restricted field of view UV/EUV slit spectra to make precise diagnoses of plasma dy- namics and energetics. · White- polarization brightness images of the solar corona to record and monitor co- ronal evolution and re-structuring important for generating geoeffective interplanetary disturbances.

To provide a context and a guide for proposals submitted in response to the forthcoming AO associated with SDO, we have devised a poten- tial suite of generic instruments. The instruments

2 observations, potential instruments, and a 2 Overview mission design.

The primary goal of SDO is to understand the nature and source of the solar variations that affect life and society. This broad goal leads to two objectives. One objective is to understand the mechanisms of solar vari- ability as characterized by three processes that operate on three different timescales – the solar cycle (months to centuries), active region evolution (hours to months), and small-scale magnetic element interactions (seconds to hours). The second objective is to understand the solar influences on global change and space weather as characterized by three different sources – irradiance varia- tions, energetic particles from flares and CMEs, and plasma disturbances from solar Figure 2.1. Recent solar cycle variability. wind structures. number and solar 10.7 cm radio flux (top two panels) are well-correlated indicators of solar variability. With the exception of the slow evolutionary Solar flares (third panel) and total solar irradiance changes in solar structure over the last 4.5 (fourth panel) generally follow the solar cycle. Geo- billion years, all solar variability is magnetic magnetic variability (bottom panel) has a component in phase with the cycle but also shows considerable in origin. The solar cycle is a magnetic cycle activity near attributed to the effects in which the Sun’s magnetic poles reverse of high-speed solar wind streams. with a periodicity of approximately 11 years and intense magnetic fields erupt through What mechanisms drive the quasi- the surface in whose numbers wax periodic 11-year cycle of solar activity? and wane with the cycle. Solar flares and CMEs occur when magnetic fields are The solar cycle controls the long-term be- stressed beyond their limits. The very havior of solar activity and the resultant structure of the corona and the solar wind is modulation of the Sun’s electromagnetic, determined by the structure of the magnetic particulate and magnetic plasma emissions field. The heating of the Sun’s corona and that affect the Earth. Solar magnetic fields the acceleration of the solar wind are with their associated forces and electric cur- thought to be due to interactions between rents are recognized as being responsible for small-scale magnetic elements. SDO will the Sun's activity, but the underlying proc- help us to understand the mechanisms of esses which create and then dissipate these solar variability by observing how the mag- fields in an 11-year cycle are poorly under- netic field is generated and structured and stood. Although has re- how this stored magnetic energy is released vealed flows and thermal structures related into the heliosphere and geospace. Our cur- to the magnetic variability, present theoreti- rent scientific understanding leads to a series cal models based on these observations can of outstanding questions that must be ad- only broadly reproduce the observed mag- dressed by SDO. These questions lead to netic evolution and are far from having pre-

3 dictive capability. Historical records suggest dent as faculae in the , in that the strength of the cyclic magnetic the and large loop structures variations may have been different from in the corona – alter electromagnetic radia- what is observed today and that there may tion at all . The evolution of ac- have been associated terrestrial climate tive regions depends upon the structure of changes. Furthermore, sun-like are ob- the emerging magnetic flux, the local flow served to have a wider range of activity than patterns, and the magnetic connections in is seen in the Sun, suggesting that current the solar atmosphere. solar behavior could be misleadingly steady. SDO will have the capability to study active SDO will examine the processes that control regions and their evolution by obtaining the solar cycle. The SDO Magnetographs measurements unavailable from other mis- will measure the structure and evolution of sions or observatories. The Helioseismo- the magnetic field itself. The Helioseismo- graph will measure the local flow patterns graph will measure the relevant fluid flows using observations with unique spatial at levels within the Sun from the surface to resolution and coverage. It will have the the deep interior with unprecedented resolu- ability to “see” active regions as they de- tion and coverage. By extending the volume velop on the far side of the Sun and will re- of the solar interior accessible to helioseis- solve structures just below the surface. It mic probing and increasing its sensitivity in may also be able to detect magnetic struc- the crucial regions, SDO will provide data tures before they emerge at the surface, a critical for understanding the 11-year activ- capability shared with the Photometric ity cycle. The Photometric Mapper will Mapper. The Magnetographs and Atmos- measure temperature and radiance variations pheric Imaging Array will provide critical on the solar surface that are associated with information on the complexity of the mag- the deep-seated drivers of the solar cycle netic structures in active regions. The resul- while the EUV Spectral Irradiance Monitor tant fluctuations recorded by the EUV measures the dramatic variations in the Spectral Irradiance Monitor will capture the EUV. disk-integrated effect of the emergence, evolution and rotation of all active regions. How is active region magnetic flux syn- thesized, concentrated, and then dis- How does magnetic reconnection of solar persed across the solar surface? magnetic fields on small spatial scales re- late to coronal heating, solar wind accel- The evolution of active regions controls the eration and the transformation of the behavior of solar activity on timescales from large-scale field topology? hours to months. Sunspots cause decreases of a few tenths percent in total in solar irra- The small-scale magnetic elements control diance. Energetic particles are released solar activity on short timescales. As these through magnetic reconnection associated elements erupt through the photosphere they with the evolution of active regions. Al- interact with each other and with larger ex- though simple isolated regions rarely pro- tant magnetic structures such as those asso- duce flares or CMEs, complex sunspot ciated with active regions. The magneto- groups with complicated and stressed mag- dynamic nature of small-scale magnetic flux netic field elements frequently do produce may be the basis for short-term solar vari- eruptive events. Bright active regions – evi- ability. They may provide the triggers for

4 eruptive events and their constant interac- Sun’s EUV radiation cause dramatic fluc- tions may be a key source of coronal heating tuations in the density of the Earth’s outer- and solar wind acceleration. They may also atmospheric layers and the contribute to irradiance variations in the density in the , affecting the con- form of enhanced network emission. trol and operation of earth-orbiting space- craft, and communication and navigation SDO will make critical observations of the systems. Magnetic features – sunspots, ac- small-scale magnetic elements, their inter- tive regions, network – that alter the tem- actions, and the resulting transformation of perature and composition of the solar at- the large-scale field topology. The Mag- mosphere are primary sources of solar irra- netographs will have sufficient spatial and diance variability. temporal resolution and coverage to follow the evolution of these elements. The Helio- SDO will make direct measurements of the seismograph will determine the nature of the solar irradiance, map the sources of the irra- photospheric and sub-photospheric flows diance variations, and provide observations that control their motions. The Atmospheric of the physical characteristics of these Imaging Array will follow the magnetic sources. The SDO EUV Spectral Irradiance connections within the atmosphere by si- Monitor will provide the first continuous, multaneously providing images of coronal high time resolution measurements of the loops at a series of different temperatures. EUV irradiance variations that are critically The EUV Imaging Spectrometer will deter- important for changes in the Earth’s upper mine the physical conditions associated with atmosphere and ionosphere. The measure- these features including temperature and ments will be made on a timescale of sec- bulk velocity. The Coronagraph will provide onds while the SDO mission provides cov- information on the large-scale field topology erage over the solar cycle. The Photometric and the associated solar wind. Mapper will provide unique imaging capa- bility in bolometric intensity and in visible Where do the observed variations in the and IR bandpasses that allow for the positive Sun’s total and spectral irradiance arise, identification of the solar sources of the total and how do they relate to the magnetic irradiance variations. The Atmospheric Im- activity cycles? aging Array will provide coincident images of the full solar disk made in EUV radiation The Sun’s electromagnetic radiation is the at selected emission temperatures so as to primary energy input to the Earth. This ra- identify the sources of the observed EUV diation varies at all wavelengths, and on all variations. SDO will also provide measure- timescales observed thus far. Total solar ir- ments to aid in our understanding of the radiance varied by about 0.1% during recent sources of the irradiance variations. The 11-year sunspot cycles. While these varia- EUV Imaging Spectrometer will examine tions are thought to be too small to have a the physical conditions of key features. dominant impact on climate, there is never- Magnetograms will provide information on theless considerable evidence in both con- the magnetic nature of the features and their temporary and paleo-climate records of ap- interactions while Helioseismic images pro- parent solar-related variability. Variations in vide information on related sub-surface the solar UV are larger and have a direct and structures and flows. significant effect on stratospheric ozone concentrations. Still larger variations in the

5 What magnetic field configurations lead organize populations of energetic particles to the CMEs, filament eruptions, and in the Earth’s , radiation flares that produce energetic particles? belts, and ionosphere. Storm-induced elec- tric currents can flow through power lines, The Sun emits energetic particles during overpower circuit breakers and transformers, solar flares and CMEs. The energetic parti- and ultimately disrupt power distribution. cle flux in a large flare can be lethal Sudden changes in the electron density and to an outside the protective enve- structure of the ionosphere can hamper lope of Earth’s magnetosphere. Significant global positioning systems and radio com- particle fluxes penetrate to aircraft altitudes munications. During maxima of the solar where they pose a health risk to passengers cycle CMEs are the primary contributors to and crew. Showers of these energetic parti- these solar wind disturbances. Near minima cles degrade and destroy electronic compo- the disturbances are dominated by variations nents on commercial, military, and research in the structure of the solar wind itself. Solar . They also cause short-term deple- wind structures also modulate the flux of tions of the ozone layer, especially in galactic cosmic rays, allowing a larger flux regions. The solar eruptions that produce of cosmic rays at the Earth during periods of large fluxes of energetic particles are also lower solar activity. The resultant cosmo- magnetic in nature, frequently resulting from genic isotopes – 14C and 10Be – when stored the reorganization of magnetic fields in the in tree rings and ice cores produce unique outer solar atmosphere. archives of long-term solar variability that are crucial to untangling solar influences on SDO will provide continuous and improved Earth’s past climate. There is even specula- observations of the solar conditions that lead tion that galactic cosmic rays may alter cli- to eruptive events such as flares and CMEs. mate by creating cloud condensation nuclei. Magnetographs will provide images of the underlying magnetic field. The Atmospheric SDO will observe the structure and dynam- Imaging Array will reveal the resulting ics of both the corona and the source region structures in the chromosphere and corona at of the solar wind. The Coronagraph will ob- the spatial and temporal resolution of the serve both CMEs and solar wind structures. TRACE imager but with the full-disk cover- The Magnetographs and Atmospheric Im- age of the SOHO/EIT instrument. The Co- aging Array will reveal the related magnetic ronagraph will detail the structure and evo- structures closer to the surface. The EUV lution of the corona and CMEs while the Imaging Spectrometer will examine the EUV Imaging Spectrometer examines the physical conditions that accompany the ac- physical conditions at key positions associ- celeration of the solar wind, the heating of ated with the energetic particle events. the corona, and the initiation of CMEs. Co- ronal spectroscopy would be an important Can the structure and dynamics of the tool for detailed measurements of velocity, solar wind be determined from the mag- composition, and temperature. netic field configuration and atmospheric structure near the solar surface? When will activity occur, and is it possible to make accurate and reliable forecasts of Disturbances in the solar wind rack the space weather and climate? Earth’s magnetosphere and drive geomag- netic storms. These storms enhance and re-

6 Within the LWS Program SDO must pro- The Helioseismic/Magnetic Imager would vide the observations that are crucial for un- extend the capabilities of the SOHO/MDI derstanding and predicting solar variability instrument by going to higher spatial and on all timescales. SDO must provide the temporal resolution with continuous full- data that serves as inputs to, or boundary disk coverage. The Atmospheric Imaging conditions for, the other systems within the Array would be similar to the SOHO/EIT LWS program. The extent to which space and TRACE instruments but with several weather and climate can be predicted will be independent providing simulta- recognized only after we gain a better un- neous observations with the spatial resolu- derstanding of the sources and nature of so- tion of TRACE, an order of magnitude in- lar variability by addressing the preceding crease in temporal resolution and the full- questions. disk coverage of EIT. The EUV Spectral Irradiance Monitor would provide long- SDO Science. In the following section term, continuous measurements of the solar (Section 3) we outline our current under- irradiance from 1 nm to 120 nm. The Co- standing of solar variability itself – irradi- ronagraph would be similar to the ance variations, energetic particle emission SOHO/LASCO instrument but with far bet- by flares and CMEs, and solar wind varia- ter spatial and temporal resolution along tions – along with the mechanisms that drive with extended spatial coverage from 1.1 this variability – the solar cycle, active re- RSun to 15 RSun. The Photometric Mapper gion evolution, and small-scale magnetic would be unique to SDO. It would provide element interactions. photometric and bolometric images of the solar radiance. The UV/EUV Imaging SDO Instruments. The outstanding scien- Spectrometer would be a high resolution, tific questions that arise can be addressed by high time cadence imaging spectrometer de- a set of observations as described in Section signed to complement the Atmospheric Im- 4. A potential set of instruments to acquire aging Array and Magnetographs. The Vec- these observations on SDO includes: tor Magnetograph would provide both the strength and direction of the magnetic flux · Helioseismic/Magnetic Imager across the full solar disk at 5-minute inter- · Atmospheric Imaging Array vals. · EUV Spectral Irradiance Monitor · Coronagraph SDO Mission Concept. The high resolu- · Photometric Mapper tion, rapid cadence, and continuous cover- · UV/EUV Imaging Spectrometer age required for the observations lead to an · Vector Magnetograph SDO Mission design that places the satellite into a geosynchronous orbit. This allows for A Total Solar Irradiance (TSI) Monitor continuous contact at high data rate with a should also be included if redundant obser- dedicated ground station. A geosynchronous vations are not available concurrently with orbit can also be used to minimize the data SDO. The detailed instrument capabilities interruptions caused by the satellite passing are traced back to the scientific questions in through the Earth’s and the ’s shad- the following sections. However, a brief de- ows. The SDO Mission Concept (Section scription of the instruments should provide 6) calls for a launch in late 2006 or early some initial guidance. 2007 on a Delta booster. The mission would be for a nominal 5 years with an additional

7 5-year extended mission. The SDO mission influences on global change and space will help us make great strides toward un- weather represent one broad area. There are derstanding the solar variability that affects three significant types of solar influences: life and society. The instruments will pro- (irradiance), energetic particles (as- vide the observations that lead to a more sociated with flares and CMEs), and plasma complete understanding of the solar dy- (coronal structure and solar wind variabil- namics that drive variability in the Earth’s ity). The mechanisms of solar variability environment. represent the other broad area. While these mechanisms are all magnetic in nature, they 3 Current Scientific Under- can be neatly separated according to the standing and Outstanding timescales associated with each: long-term Questions (the solar cycle), mid-term (active region evolution), and short-term (small-scale magnetic element interactions). SDO must We have learned much about the Sun and address all these areas to advance our under- solar variability in the last few years. The standing of the nature and source of the solar scientific discoveries from missions such as variations that affect life and society. SOHO and TRACE make headlines in the popular press and are occasionally featured on the nightly news. The very nature of sci- 3.1 Solar Influences on Global entific inquiry, however, teaches us that the Change and Space Weather more we learn the more we find that we don’t understand. We have learned much Understanding the solar influences on global about the solar influences on global change change and space weather is increasingly and space weather, but we have also found important as society becomes more reliant that we do not fully understand all sources on technology. Global warming and ozone of the irradiance variations and we can’t re- depletion can be influenced by solar irradi- liably predict energetic particle eruptions or ance variations. The health and safety of sat- solar wind variations. Likewise, we have ellites and astronauts can be influenced by learned much about the structure and dy- energetic particles from solar flares and namics of the solar interior and the evolution CMEs. Electrical power distribution to our of active region magnetic fields, but we still cities can be interrupted by geomagnetic don’t understand the and can’t storms induced by variations in the solar reliably predict the size of the next solar cy- wind. While our understanding of these ef- cle or the emergence of the next active re- fects and their solar origins has improved gion. The understanding of the mechanisms greatly over the past decade, many out- of solar variability that we have gained from standing questions remain. previous missions, ground-based observa- tions, and theoretical studies leads us to ad- 3.1.1 Irradiance Variations ditional questions that require new observa- tions. In the following subsections we de- The radiation that the Sun provides to the scribe our current scientific understanding Earth varies at all wavelengths and on all and list some of the outstanding questions timescales. While the Sun's fundamental en- that must be addressed by SDO. ergy source is the nuclear conversion of hy- drogen to helium in the core, the immediate We have divided the science into two broad radiative energy source for the Earth’s sur- areas with three subdivisions each. The solar face and lower atmosphere is the solar pho-

8 tosphere. The chromosphere and corona produce the UV, EUV and X-ray that are the primary source of energy for the Earth’s upper atmosphere and ionosphere. With million-year diffusion times in the core, the emergent luminosity at its outer boundary should be effectively constant on Figure 3.2. Total solar irradiance (W/m2) as a func- any total solar irradiance measurement tion of time. Total solar irradiance varies by about timescale. Given the observed variability, it 0.1% over a solar cycle and by several times that on follows that the mechanisms that deliver en- shorter (solar rotation) timescales. ergy to the photosphere must be variable. The spectrally integrated ‘total’ irradiance, of relevance to climate on Earth, varies by a few parts per million in association with p- mode oscillations, a few tenths percent dur- ing solar rotation, about 0.1% in recent solar cycles (Fig. 3.2), and possibly by a few tenths percent on longer timescales. The UV radiation from 160 to 300 nm that produces stratospheric ozone and modulates the mid- dle atmosphere, varies by 1 to 20% during a solar rotation, about twice that during the solar cycle, and possibly more on longer Figure 3.1. Solar spectral irradiance and its solar cycle variability as functions of . The so- timescales. Although the radiative output at lar cycle variability at wavelengths shorter than about wavelengths less than 400 nm 100 nm is typically greater than 100%. constitutes only 8% of the total radiative output, the stronger variability at these The magnitude of the irradiance variability shorter wavelengths makes a significant depends on the timescale and the wave- contribution (30%) to the total solar irradi- length of the radiation (Fig. 3.1). Most vari- ance variations. The EUV radiation from 1 able is EUV radiation at the shortest wave- to 120 nm that produces the ionosphere and lengths. Emitted from outer layers of the is the primary heat source for the thermo- solar atmosphere, this radiation exhibits sphere varies by 50% to factors of two or rapid short-term order-of-magnitude fluc- more during solar rotations and the solar cy- tuations during solar flares and more gradual cle. Significant variations are expected dur- variations of a few orders of magnitude ing solar flares, but are presently unknown. during the solar cycle. Least variable is the near infrared spectrum, emitted from the deepest layers of the photosphere, where 3.1.1.1 Irradiance Variations – flare-related variability is negligible and so- Current Understanding lar cycle changes are thought to be a few tenths of a percent, at most. Magnetized features in the solar atmosphere are associated with solar irradiance varia- tions across the entire electromagnetic spec- trum. Magnetic fields in sunspots redirect the upward flow of energy from the convec-

9 tion zone and affect the emergent radiation Ground-based measurements of the solar locally in the photosphere. In bright mag- radius exist over the last 300 years, but these netic regions, specifically active region results are controversial and inconsistent. plage, faculae and dispersed network, irradi- Historical measurements show that the Sun's ance is locally enhanced. The interplay of radius may even have been larger during the locally “dark and bright” features causes Maunder Minimum - during extremely cold variations in photospheric radiation (wave- periods in Europe and the Atlantic regions. lengths > 200 nm) on timescales ranging The French CERGA radius measurements from days to a solar cycle. Bright, magnet- detected a larger solar radius during solar ized features largely control the variability minimum. Others have found the opposite of chromospheric emission that includes positive correlation between apparent radius some of the strongest EUV emission lines. changes and the solar activity cycle. There In the outer solar atmosphere, magnetic are also hints of periodic solar radius varia- fields in active regions expand to form coro- tions over timescales of 1,000 days to 80 nal loop structures that at times cover a sig- years, but taken as a whole these measure- nificant fraction of the global solar atmos- ments generally are neither consistent nor phere. Density and temperature enhance- conclusive. It is apparent that the results to ments alter the emission in coronal magnetic date have been severely limited by the at- loops and their presence causes significant mosphere. increases in X-ray and lines, especially during flares. Because the nuclear energy generation rate is effectively constant on solar cycle time- Fluctuations in total (spectrally integrated) scales the Sun must be storing a fraction of irradiance are traceable to both sunspot and this energy flux in thermal, gravitational, or facular sources on short (solar rotation) magnetic forms in order for its net luminos- timescales. The movement across the visible ity to vary. Each of these mechanisms leads solar disk of a large sunspot group can cause to distinct perturbations in the equilibrium the total irradiance to decrease by a few stellar structure and potentially detectable tenths of a percent. These short-term varia- changes in the solar diameter. It follows that tions are superimposed on an overall total a sensitive determination of the solar radius solar irradiance increase from sunspot fluctuations can help reveal the cause of the minimum to maximum. Empirical models of solar cycle changes. sunspot darkening and facular brightening account for recent solar irradiance cycles, Quantitative understanding of the magnetic but the models require a factor of two more sources of EUV irradiance variations is far facular brightening than sunspot darkening less developed. Until recently, the only during maximum. It is not clear that all the models of EUV irradiance variability were variation can be found in the faculae. Mod- simply parameterizations of the 10.7 cm ra- els for the flow of heat through the convec- dio flux. In reality the database of reliable tion zone suggest that heat flow blockage by EUV irradiance observations is simply too sunspots should restructure the convection limited to realistically constrain even the zone and produce bright rings around active most simple of empirical models. This has regions. Recent observations of bright rings lead to a new approach of using differential around active regions and changes in the emission measures (combined with atomic solar radius add to the controversy. parameters of the emitting species) of spe- cific magnetic features (active regions, co-

10 ronal holes) and the quiet background Sun, total irradiance variations attributed to them. to construct full-disk EUV irradiance. Solar An accurate (~10%) equality would be an images (e.g., Ca K, SOHO/EIT, / important test of the models. Equality of the SXT) provide estimates of the fractional year-to-year photometric and radiometric volume covered by the respective features at contributions probably offers the cleanest any time during the solar cycle. This ap- discriminator between the enhanced net- proach depends on having reliable emission work, and other explanations of the 11-year measure distributions. Unfortunately, we component of irradiance variation. Merely have only a small number of suitable, cali- showing proportionality leaves open the brated, high spectral and spatial resolution possibility that more global changes also observations covering the needed wide proportional to activity level might be play- range of emission temperatures. Neverthe- ing a significant role. less, the new approach to quantifying mag- netic sources of EUV irradiance variability [Text highlighted in gray provides the - from a more physical understanding of the ability between the scientific goals, the temperature, density and atomic parameters measurement characteristics, and the re- is promising. The major challenge here is to quirements on potential instruments for obtain better data in the EUV lines of coro- SDO.] nal and transition region origin. Many of these lines arise in structures whose areas The total solar irradiance must continue to are poorly estimated by traditional chromos- be monitored to address these questions. Ef- pheric indices such as Ca K and Mg II. forts toward identifying the sources of these variations are of little use without knowl- edge of the variations themselves. 3.1.1.2 Irradiance Variations – Outstanding Questions Photometric images, photometrically accu- rate narrowband images and broadband bo- lometric images, are needed to determine the Where do the observed variations in the actual sources of the irradiance variations. Sun’s total and spectral irradiance arise, and how do they relate to the magnetic These images must have the spatial resolu- tion to resolve sunspots and faculae and the activity cycles? This is the overarching question for the solar irradiance variations. accuracy and precision to determine their contribution to the total irradiance. It leads to a series of more focused questions in specific areas. Magnetograms with comparable spatial What are the sources of the total irradi- resolution are also needed in order to dis- ance variations? The projected areas of criminate between the magnetic and the non- sunspots and faculae are well correlated with magnetic irradiance sources. the total solar irradiance variations. How- ever, alternative sources such as luminosity Helioseismic images will aid in measuring changes due to radius variations associated variations in the size of the Sun that should with sunspots and faculae can also give this be related to luminosity variations. correlation. Present photometry of spots and faculae is not accurate enough to demon- Heliometry – precise measurements of the strate that the irradiance effects are equal to size and shape of the Sun – will also aid in – rather than merely proportional to – the measuring variations in the size of the Sun

11 that should be related to luminosity varia- ance observations (to specify the actual tions. variations and to verify the understanding developed from analysis of the images). How does the EUV spectral irradiance change on timescales of seconds to dec- Atmospheric images like those obtained ades? There is a severe lack of knowledge with SOHO/EIT and Yohkoh/SXT will aid of solar spectral irradiance variability at in the identification and quantification of the wavelengths shorter than 120 nm, where the spectral irradiance variations. Full-disk, nar- database is characterized by intermittent ob- rowband EUV images at a series of EUV servations made with poorly calibrated in- wavelengths should be taken simultaneously struments. Solar cycle EUV irradiance vari- with the EUV spectral irradiance measure- ability amplitudes are uncertain by 50% or ments. These images should have the spatial more, and variations associated with flares resolution to identify the nature of the fea- are essentially unknown. Only for the total tures and the temporal resolution to follow irradiance and for the spectral irradiance the rapid variations in flares. The measure- from 120 to 400 nm do continuous, well- ments must be made in a sound radiometric calibrated databases exist. NASA’s Office of way that eliminates spurious instrumental or Earth Science (OES) has made these latter spacecraft effects. The spectral bands should observations during the past decade at be chosen such that emission measures longer wavelengths for understanding natu- spanning a range of temperatures from 104K ral variations in climate and ozone. to a few 106K can be constructed for indi- vidual or classes of magnetic elements. The EUV spectral irradiance shortward of 120 nm must be monitored. The observa- UV/EUV spectra are needed to provide de- tions should cover the neglected wavelength tailed information about the density and range from 1 nm to 120 nm with enough temperature of the plasma in individual spectral resolution to be useful to solar, up- magnetic regions on the Sun's disk. It is evi- per atmospheric and ionospheric studies. dent that a major contributor to irradiance The temporal resolution should be short changes are the variations in UV/EUV emis- enough to follow flares, and the temporal sion lines associated with regions of strong coverage should extend through a solar cy- magnetic fields in the photosphere. Without cle. These measurements should have suffi- sufficient spectral and spatial resolution it is cient short-term precision and long-term not possible to understand the physical stability to monitor the EUV variations on mechanisms responsible for the variations. timescales of flares to the solar cycle. High spectral resolution would provide sig- nificantly more reliable emission measure What are the sources of the spectral irra- determinations, and the coordination with diance variations? Although magnetic re- the full-disk images would provide global gions are understood to be the primary cause context for these spatially resolved observa- of EUV and X-ray irradiance variations, the tions. details of this association have yet to be de- termined. This is because of the complete Were solar irradiance variations larger in absence of simultaneous well-calibrated full- the historical past? Present specifications disk images (that resolve the spatial features of solar total and spectral irradiance vari- with adequate spectral resolution) and inde- ability for terrestrial research are based pendent, spectrally-compatible, EUV irradi- largely on multi-component empirical mod-

12 els. Although these models can be quite complex, they represent a very simplified view of the structure of the solar atmos- phere.

The ability to understand solar variations outside of the current era requires the devel- opment of physical models based on obser- vations of the solar sources of the total and spectral irradiance variations at the neces- sary spectral and spatial resolution. Figure 3.3. LASCO images of the solar corona be- fore (left) and just after (right) the large flare of July 14, 2000 (the Bastille Day event). Energetic particles 3.1.2 Flares and CMEs as the from the flare eventually overwhelmed the detector Source of Energetic Parti- and made observations impossible for some time. cles 3.1.2.1 Flares and CMEs – Cur- rent Understanding Flares and CMEs are the primary sources of energetic particles from the Sun. Solar flares and coronal mass ejections are the largest Solar flares have long been recognized as a explosions in the , and are the key component of solar variability on short root cause of much of the Earth’s geomag- timescales. The advent of space-based ob- netic and ionospheric disturbances. The en- servations has also added CMEs and shocks ergetic particles produced in these explo- in the solar wind as significant components. sions can be harmful to humans and elec- Much has been learned about these phenom- tronic instruments in space. Figure 3.3 ena and about how they accelerate particles shows LASCO images of the solar corona to cosmic ray energies. Our understanding before and after the flare of July 14, 2000. and predictive abilities have improved but The energetic particles produced in this not to the point where we can reliably pre- event showered the LASCO detector and dict the occurrence and geoeffectiveness of made observations impossible for some these events. time. The fundamental physics of these events remains elusive and predictions must Solar flares can be detected at virtually all become more reliable. These extremely wavelengths. The NOAA GOES satellites complex phenomena require the explosive include solar X-ray monitors that are used to conversion of magnetic energy into acceler- detect and classify solar flares. The occur- ated particles, bulk mass motions, light, and rence of flares is well correlated with the heat. SDO must examine the evolving mag- magnetic complexity of active regions. Delta netic configurations associated with these spots, sunspots with opposite magnetic po- events and identify their precursors. larities included within a common penum- bra, often produce flares. Detailed meas- urements of the magnetic field also reveal stressed magnetic field configurations in flaring active regions. Figure 3.4 shows a flaring active region along with the observed magnetic field and the stress-free field de-

13 rived from the observed line-of-sight field component. The magnetic field vectors are Traditionally CMEs have been observed twisted or sheared in the region where this with white light using polariz- flare occurred. Other indicators of stressed ers and broadband filters. With the launch of field configurations include sigmoid, or S- SOHO, considerable progress toward a de- shaped, coronal loops observed in images tailed description of CMEs is being made obtained in the UV/EUV. While these mag- through a combination of observations with netic structures indicate the likelihood of a the EIT, LASCO and UVCS instruments. flare, they are not highly reliable and cannot SOHO observations with the LASCO co- indicate when a flare will occur or how ronagraphs show that CMEs have a wide strong it will be. Detailed determinations of range of speeds from 100 km/s to greater the magnetic configurations in many flaring than 1000 km/s. EIT observations show that and non-flaring active regions are needed for a large CME can empty a substantial frac- systematic studies of these and other flare tion of the corona. We know typical masses indicators. (1016 g) and total energies (1032 ergs) of CMEs, but not what causes the variations of these parameters from one CME to another. We need to know how the magnetic field energizes and accelerates CME material.

There is evidence that initially slower-speed CMEs that tend to have higher accelerations are related to prominence eruptions while the faster CMEs are associated with flare events but the correlation is not exact. SMM observations show that the range of CME speeds does not appear to be related to the amount of thermal input into the CME, so the variation must be related to the nonther- mal energy from the magnetic field. UVCS observations indicate ionization states formed at temperatures over a broad range (from 104 to 107 K) and heat input compara- ble to the kinetic energy in the ejected mate- rial.

By their explosive nature, CMEs also reflect short-term changes in the configuration of the coronal magnetic field. On short times- cales, magnetic reconnection of field lines is often seen just before and sometimes after a CME has erupted, but whether reconnection Figure 3.4. Stressed magnetic field configuration in a is a cause or an effect of the CME is still flaring active region (off-band Ha - top panel). The debated. There has been some success in observed magnetic field vectors (center panel) are using the presence of sigmoid magnetic twisted relative to the stress-free field vectors (bot- fields in the low corona as a precursor for tom panel) in the region of the flare. CME eruptions but it is not yet possible to

14 predict which sigmoid structures will erupt increasing helicity) may be about the same or when. as the amount of free energy removed con- vectively by CMEs during a solar cycle. The geoeffectiveness of CMEs that strike While there is good evidence for the ex- the Earth is known to depend, in part, on the pected flux and helicity transport in the past shock strength and on the direction of the three decades of solar wind data, quantita- magnetic field. Those CMEs that have tive inferences about magnetic helicity from southward directed interplanetary magnetic solar observations have been difficult to ob- fields are known to produce the largest tain. geomagnetic storms. Studies of the relation- ship between the direction of the magnetic field in CMEs and their geoeffectiveness conclude that accurate storm forecasting re- quires a means of predicting the magnetic field configuration in CMEs.

The process of magnetic reconnection is thought to be the principle mechanism by which magnetic energy is converted into thermal energy in flares and CMEs. A pri- mary result of magnetic reconnection is the creation of new field line connectivities. On Figure 3.5. Helical structure in a CME observed with large spatial scales, the new plasma loops LASCO on June 2, 1998. are often seen to connect pre-existing photo- spheric flux elements with newly emerged photospheric flux elements. The rate at 3.1.2.2 Flares and CMEs – Out which these new loops appear is on the order standing Questions of 0.1% to 10% of the Alfvén time, indicat- ing that this occurs in the regime of fast re- connection. The high electrical conductivity What magnetic field configurations lead of the coronal plasma implies that small to the CMEs, filament eruptions, and spatial scales are often needed in order for flares that produce energetic particles? magnetic reconnection to occur at a reason- Our current understanding of flares and able rate. These spatial scales are below the CMEs suggests that the magnetic field con- resolution limit of current instrumentation. figuration and its evolution determine the development of these explosive events. The Besides providing clues about the storage photospheric magnetic fields often indicate and release of magnetic energy, CMEs pro- the existence of stressed magnetic field con- vide a major avenue (in addition to the fast figurations. However, the lack of significant and slow speed wind) for coupling the Sun's changes in the photospheric fields after an atmosphere to the more extended helio- eruption suggests that the actual restructur- sphere. CMEs may be the fundamental ing of the field occurs high in the solar at- means by which the Sun sheds both mag- mosphere where the magnetic forces domi- netic flux and helicity (Fig. 3.5) over the 11- nate. Significant restructuring of the corona year solar cycle. The rate of free energy is often observed before, during, and after built up over the solar cycle (in the form of these events.

15 Vector magnetic flux measurements are ciated with the magnetic field pattern evolu- required to determine the configuration of tion. the surface magnetic fields. While longitu- dinal magnetograms provide some indica- Is the magnetic helicity removed from the tion of the field's complexity, vector mag- Sun by CMEs consistent with the rate of netograms in the photosphere (and ideally helicity generation under the surface and the chromosphere where the fields are closer with the rate of helicity transport in the to a force-free state), are required to deter- solar wind? The Sun has a weak preference mine the presence and magnitude of the for left-helical fields in the north and right- magnetic stresses associated with these helical fields in the south but the cause of eruptions. Useful observations require ade- this hemispheric pattern is not yet under- quate spatial resolution (~1 arcsec) to re- stood. Measurements of the patterns and the solve the small-scale elements that may trig- net amount of helicity generation would ger the eruptions, adequate temporal resolu- place important constraints on solar dynamo tion to follow the evolution (minutes) and models. From a global point of view, CMEs the spatial and temporal coverage to capture are an inevitable response to helicity gen- all events and to determine long-range inter- eration, but we do not know whether the connections between magnetic regions. helicity is primarily generated by the dy- namo, by convection, or by shearing of co- Atmospheric images are required to deter- ronal fields. A better understanding of mine the configuration of the magnetic helicity generation and removal is crucial to fields in the lower corona associated with understanding CMEs. flares and CMEs. These images provide di- rect evidence of the occurrence of these Can the magnetic field direction and twist events and provide critical information on of CMEs be related to their geoeffective- their spatial, temporal, and physical charac- ness? The direction of the interplanetary teristics. magnetic field in the solar wind is known to be a good indicator of the geoeffectiveness Coronagraphic images are required to de- of the disturbance. To what extent can this termine the configuration of the magnetic field component be determined from obser- fields in the outer corona associated with vations of CMEs close to the Sun? flares and CMEs. These images also provide direct evidence of the occurrence of these Vector magnetic flux measurements are events and provide critical information on required to determine the helicity of the sur- their spatial and temporal characteristics. face magnetic fields.

UV/EUV spectra are required to determine Atmospheric imaging and UV/EUV spec- the physical conditions at the site of the troscopy can provide information on the eruption – both before and after the events sign of the magnetic helicity, and the field themselves. These measurements may help geometry for the twisted magnetic fields in to identify the triggering events for these CMEs. eruptions. Where do CMEs form shock waves, and Helioseismic images are required to meas- how are coronal shocks related to the ure the surface and sub-surface flows asso- fluxes of energetic particles at the Earth? Energetic particles are accelerated in shocks

16 associated with CMEs. The locations and interactions eventually include the formation development of these shocks must be stud- of shocks that accelerate particles which ied in more detail. contribute to the space weather environment.

EUV coronal spectroscopy would provide important information on the physical con- ditions inside of CMEs as they evolve dur- ing their through the outer corona. Velocity distributions, line of sight veloci- ties, and the detection of high temperature spectral lines can be used to determine Figure 3.6. Geomagnetic fluctuations during the de- shocks and current sheets within CME clining phase of Cycle 22. Disturbances recurring at 27-day intervals are due to high-speed streams in the structures. solar wind.

What is the role of magnetic reconnection 3.1.3.1 Coronal Structure and in the heating and acceleration of CMEs? Solar Wind Variations – Some models of CMEs give a prominent role to a reconnection current sheet con- Current Understanding necting the CME flux rope to a post-CME flare arcade. In the high-temperature, low-density ex- tended corona the plasma beta (ratio of gas UV/EUV spectra can provide evidence of pressure to magnetic field energy density) is extremely hot structures at the reconnection low enough that motions of the coronal sites with observations of high temperature plasma are usually dominated by the mag- spectral lines and measurements of the netic field configuration. We know that the plasma temperature, density, and heating high-speed wind originates in the open mag- rate. The presence of predicted MHD waves netic field regions in coronal holes and that created at reconnection sites can be detected the slow-speed wind emanates from above by their effects on observed streamers. profiles for comparison with models of the reconnection process. Magnetic field extrapolation models, using photospheric magnetic fields as input, can 3.1.3 Coronal Structure and now fairly reliably reproduce the large-scale Solar Wind Variations structure of the inner corona (Fig. 3.7). These models have become more sophisti- cated by including non-potential fields and Solar wind variations also produce geomag- MHD effects. Maps of regions where the netic disturbances. This is particularly evi- magnetic field is opened to the heliosphere dent late in each solar cycle when flares and correspond well to coronal holes and the CMEs are less likely and the Sun’s global source of high-speed wind streams. Yet, the field geometry is more stable. As the Sun’s physics involved in the heating of the corona rotation brings around the sources of high- and the acceleration of the solar wind is still speed streams we see recurring geomagnetic hotly debated. disturbances at ~27-day intervals (Fig. 3.6). Corotating Interaction Regions (CIRs) form While much has been learned in recent years within the heliosphere when such streams about coronal heating and solar wind vari- interact with ambient slow solar wind. These ability, no unifying theory about the physi-

17 cal mechanisms responsible has emerged. tures in spite of the lack of complete (back side) Theories of coronal heating and solar wind magnetic field information. acceleration generally fall into two broad categories: (1) heating and acceleration by While Alfvén waves with frequencies high Alfvén or other MHD waves and (2) heating enough to probe cyclotron resonances (10 to and acceleration by the impulsive release of 10,000 Hz) have not yet been observed di- energy at sites of magnetic reconnection. rectly in the solar wind or corona, UVCS Observations from current solar missions observations have given rise to (e.g. SOHO, Yohkoh, and TRACE) have renewed interest in the theoretical investiga- provided evidence for both types of mecha- tion of resonant wave dissipation. Spectral nisms. line widths indicating large anisotropic velocities have been observed. It is not known if the fluctuations originate primarily at the coronal base or are generated continu- ously in the acceleration region of the wind (via turbulent cascade or plasma instabili- ties).

Figure 3.8. Coronal structures from 1.0 to > 5 R (August 11, 1999 LASCO C2 and total eclipse photo from S. Koutchmy). Solar wind speed can be measured using moving features in the outer corona but much of the restructuring takes place much closer to the surface.

SOHO/MDI observations suggest that a “magnetic carpet” of twisted magnetic fields that reconfigure on times- Figure 3.7. Coronal structures derived from an MHD cales of minutes may be responsible for field extrapolation model (top and middle panel, heating the quiet corona. Thin current sheets SAIC) for the August 11, 1999 eclipse (bottom are generated between the twisted and panel). There are good agreements with many struc- braided field lines, and magnetic reconnec-

18 tion occurs stochastically to relax the system 3.1.3.2 Coronal Structure and to a lower-energy state (with net particle Solar Wind Variations – heating). “Microflares” from the reconnec- Outstanding Questions tion sites on small loops seen in Yohkoh/SXT images and vector magneto- grams may provide enough energy to heat Can the structure and dynamics of the the corona at least locally in active regions. solar wind be determined from the mag- Whether these local heating processes can netic field configuration and atmospheric heat the extended corona above a few tenths structure near the solar surface? The tem- of a solar radius is still an open question. peratures and outflow speeds of the large polar coronal holes at solar minimum have Recent progress has also been made on the been shown to differ from those of the coro- slow speed wind and its relation to coronal nal holes that appear primarily at low lati- streamers. UVCS observations of high tem- tudes at . For the low-speed perature high in streamers suggest pref- wind, the large equatorial streamer belt at erential heating of ions with high mass-to- solar minimum seems to be hotter than the charge ratios. These results, while not con- more sporadic streamers that form at all clusive, are consistent with ion cyclotron latitudes at solar maximum. Precisely how resonance absorption that could lead to a these thermal changes depend on the coronal steady expansion of material into the slow field morphology is not well understood. speed wind. Other streamer observations The solar corona as a whole is brighter in its made with LASCO show moving “blobs” or EUV emission at solar maximum than it is at density enhancements that appear to emerge solar minimum. Whether this arises because from the tips of streamers. This suggests that of an overall density increase, a change in some of the mass in the slow speed wind is the radial and latitudinal distribution of co- released by intermittent reconnections of the ronal heating, or is related to photospheric open and closed magnetic field lines in brightness changes, is not yet known. streamers. This view is consistent with the results of earlier investigations that link the Magnetograms are required to determine origin of the slow speed wind to the the configuration of the surface magnetic streamer belt. However coronal abundance fields associated with the coronal and solar measurements show that most of the slow wind structures. Magnetic measurements at speed wind probably originates from the the boundaries between streamers and coro- streamer edges, since these regions have nal holes will elucidate the role of open and similar depletions of high First Ionization closed fields in producing the slow speed Potential (FIP) elements that are found in wind. Longitudinal magnetograms are the the slow speed wind. More detailed meas- minimum required observation, and to fully urements are needed to establish, with con- support the science outlined above vector fidence, what fraction of the slow speed photospheric magnetic flux measurements wind comes from streamer edges versus the are required. While longitudinal magneto- closed-field regions of streamers. grams provide some indication of the field configuration, vector magnetograms in the photosphere, and ideally in the chromos- phere where the fields are closer to a force- free state, are required to provide a more accurate determination.

19 Atmospheric images are required to reveal wind? Solar wind variations have a direct the location, morphology, and thermody- impact on space weather. We must better namic characteristics of atmospheric fea- understand the processes involved in pro- tures associated with the outer coronal ducing these variations. Heating and accel- structures. The locations of polar plumes eration of the corona is ultimately related to and the edges of streamers can be identified the magnetic field. Observations are needed in such images. They also assist in the de- to differentiate between the alternative termination of the magnetic field configura- mechanisms of MHD wave absorption and tion. impulsive Joule heating. Different proposed mechanisms of wave generation (i.e., turbu- Coronagraphic images are required to de- lent cascade or local micro-instabilities) de- termine the positions of streamers and coro- pend in different ways on the evolving coro- nal holes, the solar wind speeds associated nal area, photospheric field with them, and to follow their evolution. strength, and plasma beta.

UV/EUV spectra are required to determine UV/EUV Spectra are needed to character- the physical conditions associated with the ize the density, bulk velocity, energy state, solar wind structures. Temperatures, densi- ionization, and elemental abundance of solar ties, flow speeds, and the presence of waves wind source regions. A key means of identi- and turbulence need to be determined within fying the dominant physical processes will these structures. Spectroscopic measure- be to observe how the measured plasma pa- ments of bulk coronal outflow velocities at rameters and inferred wave properties the boundaries between streamers and coro- evolve on timescales from days to solar cy- nal holes will elucidate the role of open and cles. Reconnection sites on the boundaries closed fields in producing the slow speed of streamers can be identified by measuring wind. spectral line widths in these regions. Mag- netic reconnection sites should also produce EUV coronal spectroscopy would also ions in highly excited charge states and their provide important information on solar wind characteristic spectra. structures. Velocity distributions of ions and could be measured to determine 3.2 Mechanisms of Solar Vari- the spectrum of waves and turbulence in co- ability ronal structures. Doppler shifts and ion abundances could be measured to charac- Magnetic structure is produced in the Sun terize the source regions of the high-speed through a broad range of spatial and tempo- and low-speed solar wind structures. ral scales. The global field reverses on a so- lar cycle timescale, while magnetic elements Helioseismic images are required to meas- at the limit of resolution evolve on times- ure the surface and sub-surface flows. These cales of minutes or seconds. Ultimately, the flows control, to some extent, the magnetic source of this magnetic activity can be field pattern evolution and the subsequent traced to the heat released by nuclear reac- restructuring of the corona and solar wind. tions in the . However, the mecha- nisms by which the activity is produced in- What are the fundamental processes re- volve the fluid flows driven by this heating sponsible for the heating and acceleration and the magnetic field itself. Our ability to of the high-speed and low-speed solar understand and ultimately predict solar ac-

20 tivity hinges on our understanding of how 3.2.1.1 The Solar Cycle – these magnetic structures are produced. Pre- Current Understanding dictions of the size and timing of the next solar cycle will not be reliable until we un- derstand how the cycle is produced. Predic- The solar cycle is a magnetic cycle in which the polarity of sunspots and the Sun’s global tions of flares and CMEs will not be reliable until we understand how active regions form magnetic field reverse approximately every 11 years (Fig. 3.9). The source of this mag- and evolve. Predictions of solar wind varia- tions will not be reliable until we understand netic cycle is widely believed to be a mag- netohydrodynamic dynamo seated within the how the wind is accelerated. SDO will pro- vide crucial observations needed to improve solar interior. Magnetic fields are effectively “frozen” into the highly conducting ionized our understanding of the mechanisms that produce solar magnetism. plasma within the Sun. Shearing motions within the plasma can amplify the embedded magnetic field. Regions with strong mag- 3.2.1 The Solar Cycle netic field become buoyant in the and rise rapidly to the surface. Mag- netic loops emerge at the surface to form bi- The solar cycle has proven notoriously diffi- polar active regions with a characteristic tilt cult to predict. Once a cycle is well under- imposed by the solar rotation that leaves the way its smoothed behavior can be predicted following polarity closer to the poles than with some reliability using statistical models the leading polarity. At the surface a meridi- for the shape of the cycle. Predictions prior onal flow slowly transports magnetic ele- to the start of a cycle are, however, much ments from the active latitudes to the poles less reliable and longer range predictions are while cellular flows like supergranules dis- virtually useless. Currently, all methods of perse them randomly across the surface. cycle prediction are empirical in nature. While we understand many of the processes The cycle is completed through magnetic involved in producing the solar cycle we do reconnection between elements of opposing not have a physical model that will take ini- polarities. These ingredients (rotation, shear tial conditions and predict future behavior. flow, meridional flow, cellular flows, rising Long-range predictions of the solar cycle (as tilted loops, and reconnection) are thought to well as extensions backward in time) are be critical components of the dynamo. Ob- important for understanding the climate servations and theoretical models exist for connection, for predicting satellite drag, and each of these but conflicts and questions re- for predicting the frequency of eruptive main and a comprehensive model containing events such as flares and CMEs. SDO will all these ingredients is yet to be formulated. improve our understanding of the solar cycle by providing continuous observations and extended spatial coverage of the two main components of the solar cycle – the mag- netic field and the fluid flows within the Sun.

21 Figure 3.9. Longitudinally averaged magnetic field as a function of latitude and time. Active region magnetic flux is evident in the “butterfly” pattern within about 50° of the equator. The reversal of the magnetic polarities from cycle-to-cycle is evident at the poles and in the equatorward edges of the butter- fly patterns.

The well-known seen on the surface of the Sun is one example of a shear flow – latitudinal shear. Radial shear has long been thought to be the principal Figure 3.10. Internal rotation rate as a function of driver for the dynamo. Early kinematic dy- latitude and depth. The (dashed line - where the rotation velocity changes rapidly) is namos for the Sun required a radial shear thought to be the location of the magnetic dynamo from a rotation rate that increased towards associated with the 11-year cycle. the interior while hydrodynamic models of the convection zone gave a rotation rate that The torsional oscillation pattern represents a decreased. Both of these models were shown shear flow on a smaller scale. Bands of en- to have problems when helioseismic obser- hanced rotation rate appear near the poles vations revealed the true nature of the inter- and migrate toward the equator by the mid- nal rotation profile. The rotation rate is dle of each cycle. This torsional pattern pro- nearly constant across the bulk of the con- duces enhanced latitudinal shear in the ac- vection zone with the shear layers at the top tive latitudes. The detection of the torsional and bottom. The region of high shear found pattern in the outer 5 percent of the Sun's at the base of the convection zone (Fig. radius by helioseismic probing makes it evi- 3.10) is thought to be the source of the solar dent that the structure is more than a super- cycle. The rate of shear across this “tacho- ficial process. In layers below this outer 5 cline” can wrap a magnetic field line once percent, the nature of the velocity structures around the Sun’s equator every 16 months. becomes more obscure. Recent observations suggest that there is a periodic variation in the rotation rate at the A meridional flow is observed both on the tachocline with a timescale of 1.3 to 1.8 surface and in the interior. The strength of years. Others analyzing the MDI and GONG this flow, its latitudinal structure, and its data only find evidence for chaotic varia- variations in time are still somewhat uncer- tions. Should there be additional time peri- tain. Tracking features on the surface gives a ods associated with the solar cycle, the con- flow speed about half that given by the di- straints on models will tighten. rectly observed Doppler velocities. Helio- seismic studies also give higher flow speeds and show that the poleward flow extends well into the convection zone. “Flux trans- port” models for the large-scale surface magnetic field have been developed over the last 40 years. These models do a remarkable job of reproducing the magnetic flux distri- bution using only surface flows and the

22 eruption of magnetic flux in active regions. However, the meridional flows employed in How is the differential rotation pro- these models are quite different from the duced? The processes that produce the dif- observed flows in both strength and struc- ferential rotation are still uncertain. Hydro- ture. There are serious concerns about why dynamical models for the solar convection the models of the polar cap size and field zone reproduce the latitudinal differential strength work as well as they do given that rotation but not its variation with depth or the poleward flow systems vary as much as time. Within these models the differential is observed. rotation is maintained by angular momen- tum fluxes driven by large-scale convective flows. The detection of these cellular flows The hierarchy of velocity structures in the and confirmation of the presence of the Sun begins with differential rotation and me- momentum flux would help our under- ridional circulation at the largest scale. At standing of the solar cycle. smaller scales is the torsional oscillation pattern. Superposed on the torsional oscilla- What is the structure of the meridional tions is a persistent cell-like or wave-like flow and how does it vary? The meridional pattern with a geometric scale of roughly 10 flow itself has been exceedingly difficult to percent the solar diameter. At still smaller measure. Its form, in both latitude and ra- scale are supergranules. The supergranule dius, is still quite uncertain and its time de- pattern dominates the smaller scale distribu- pendence is unknown. The poleward surface tion of the magnetic field and clearly plays a flow extends well into the interior but the role in moving magnetic flux across the sur- equatorward return flow has not been de- face. tected. Better characterizations of the me- ridional circulation will aid our understand- 3.2.1.2 The Solar Cycle – ing of the solar cycle. Outstanding Questions What role does the torsional pattern play in the solar dynamo? The extension of the What mechanisms drive the quasi- torsional oscillation pattern well below the periodic 11-year cycle of solar activity? surface suggests that this phenomenon is This key question leads to others directly more than a small perturbation in rotation concerning the relevant processes them- due to the presence of active region mag- selves – differential rotation, meridional cir- netic fields at the surface. While it is proba- culation, torsional oscillations, and cellular bly not the driving force behind the dynamo, flows. it may very well play a significant role in that process and may signal variations in the How are variations in the solar cycle re- behavior of the solar cycle. lated to the internal flows and surface magnetic field? The amplitude of the solar Are the small latitudinal surface bright- cycle can vary dramatically from one cycle ness variations related to the torsional to the next. Can we identify an early indi- pattern or the meridional flow? Latitu- cator of solar cycle magnitude? We expect dinal bands of enhanced surface brightness these variations to be related to differences are seen in MDI data and the extreme polar in the internal flows (e. g., differential rota- regions appear darker. These features may tion and meridional flow) and the magnetic be a thermal signature related to the meridi- field itself (e. g. polar field strength).

23 onal circulation or the torsional pattern. 3.2.2 Active Region Evolution Changes in their structure and position may hold important clues to understanding the Active regions are the most visible evidence solar cycle. of solar variability. The intense magnetic fields in active regions are the source of Magnetic Images are needed to follow the flares and most CMEs. The chromospheric evolution of the surface magnetic field. The and coronal structures that accompany ac- magnetograms themselves should cover the tive regions are the source of irradiance visible disk to allow mapping the field to all variations in the UV, EUV, and X-rays. Un- latitudes and longitudes. The sensitivity derstanding and predicting the structure and should be sufficient to follow changes in evolution of active regions is key to our un- quiet areas as well as in active regions and derstanding and predictions of explosive to accurately measure the polar flux. events and spectral irradiance variations. Detection of emerging magnetic structures Helioseismic Images are required for stud- in the interior and understanding their physi- ies of the structures and flows in the solar cal properties can make important contribu- interior. Helioseismic probing of the critical tions to space weather forecasting. near surface layers (a region that all acoustic waves travel through and, because of the 3.2.2.1 Active Region Evolution lower sound speed, spend much of their time – Current Understanding in) requires high spatial resolution and rapid cadence observations. Probing of the three- dimensional structures deep in the convec- The probability of the appearance of an ac- tion zone requires good spatial resolution tive region can be given as a function of the near the limbs to allow for widely separated size of the region for a given latitude and the measurement areas. phase and amplitude of the solar cycle. The number distribution of active region flux has Atmospheric Images are needed in con- been well measured through the solar cycle. junction with the magnetograms to deter- It is an exponential function of the flux that mine the large-scale magnetic field configu- does not change slope over the solar cycle. ration associated with the solar cycle. The amplitude of the function, and hence the total active region flux, changes by about a Coronagraphic Images are also needed in factor of eight from solar minimum to conjunction with the magnetograms to de- maximum depending upon the amplitude of termine the large-scale magnetic field con- the cycle. figuration associated with the solar cycle. Once an active region emerges, there is a high probability that additional eruptions of Photometric Images are needed to deter- mine the actual source of the solar cycle flux will occur in the neighborhood, or even in the same region (activity nests, active variations in irradiance. These images may also reveal thermal structures associated longitudes). There is some weak evidence that activity tends to concentrate around the with the solar dynamo process. edges of giant cellular patterns. Once an ac- tive region starts to become visible, its ulti- mate size and complexity can be estimated by the rate of flux growth and the complex- ity of its early appearance. However, none

24 of these predictions are based on physical convection zone it is hard to understand how models for the emergence of active region the active regions that result from flux flux. There is a clear need for a better under- emergence can long remain connected to standing of active region formation. their origin in the tachocline.

The source of the active region flux is MDI high-resolution measurements have thought to be in the tachocline, the shear demonstrated that it is possible, using time- layer just below the convection zone. Dif- distance helioseismology and acoustic to- ferential rotation should amplify the field mography, to form images of the critical top until the magnetic energy density is so large 10,000 km. Time sequences of such images that it becomes buoyant and rises to the sur- should do much to increase our understand- face. The magnetic fields must be amplified ing of active region flux emergence and the 5 to about 10 Gauss to become sufficiently subsequent detachment of this flux from the buoyant to begin their rise. Typically, active deep convection zone. The behavior of the region magnetic flux emerges through the flux in the top 10,000 km of the Sun may be surface in the form of sunspot groups and key to the understanding of the solar dy- plage. The spots grow in size over the namo process that generates the active re- course of days while the preceding and fol- gion flux. lowing spots drift apart in longitude. Active regions have characteristic tilts with respect As to the prediction of specific active re- to lines of constant latitude and asymmetries gions at specific times and places, very little in the morphology of the preceding and fol- can be done at present. Searches for thermal lowing spots. shadow precursor signals in advance of the eruption of an active region have been in- Models of buoyant flux tubes have been conclusive using currently available obser- constructed to follow their rise from the vations. Helioseismic tomography can be bottom of the convection zone to within used to probe the flow field beneath active about 10,000 km of the solar surface. Above regions (Fig. 3.11). These techniques might that height the rate of change of pressure is detect emerging active structures in the deep so great that current computers are not suffi- interior before they are visible. However, a cient for the calculations required. (Six of clear case of a precursor signal has not been the fifteen scale heights between the bottom observed. Initial results using the MDI high- and top of the convection zone occur in the resolution data indicate that the flux propa- top 10,000 km.) These models show how the gates rapidly, in less than 8 hours, through effects of solar rotation can produce active the top 10,000 km of the convection zone. regions with both tilt and morphological One could speculate that this rise time is suf- asymmetry between preceding and follow- ficiently rapid and the cadence of existing ing spots. However, these models neglect observations is too slow to detect a precur- the turbulent motions within the convection sor signal. Alternatively, the precursor sig- zone. Even neglecting turbulence it is diffi- nals may simply be too weak to detect. Sur- cult to create a stable rising flux bundle. face Doppler measurements do, however, show persistent convergence at the sites of Observations of the interior rotation (Fig. active regions prior to their emergence. 3.10) show that there is a sharp decline in These observational techniques provide the angular rotation rate over the top 10,000 great promise for furthering our under- km of the convection zone. Given the high standing of active regions. shear between the surface and the deeper

25 Figure 3.11. The material flows at the depth of 0-3 Mm (a) and at 6-9 Mm (b). The outline of the sunspot umbra and penumbra is shown. The colors are verti- cal motion with positive values (red) corresponding to downflow. The arrows are horizontal motion. Measurements like these may reveal active regions before they emerge.

Another recent success is the new ability to make crude images of sound speed anoma- lies near the surface on the far side of the Sun (Fig. 3.12). Thus, it has become possi- ble, in principle, to predict that a large active region will rotate onto the visible disk days to weeks before this happens.

Figure 3.12. Helioseismic maps of the Sun. This set of three maps constructed 13.5 days apart shows that the “big spot” (AR9393) was seen on the far side before it was seen on the Earth side.

3.2.2.2 Active Region Evolution – Outstanding Questions

How is magnetic flux synthesized, concen- trated, and transported to the solar sur- face where it emerges in the form of evolving active regions?

Can active region magnetic flux be ob- served before it erupts through the sur- face? Helioseismic studies can see changes in the sound speed below active regions but have not seen these variations before they erupt through the surface.

26 To what extent are the appearances of dimensional structures deep in the convec- active regions predictable? An important tion zone requires good spatial resolution goal for SDO is to develop an understanding near the limbs to allow for widely separated of the locations and times of magnetic re- measurement areas. gion eruption that goes beyond simple sta- tistics. Can we go beyond empirically identi- Atmospheric Images are needed in con- fying the locations of activity? junction with the magnetograms to deter- mine the magnetic field configuration asso- What roles do local flows play in active ciated with active regions and their evolu- region evolution? Some aspects of active tion. Simultaneous observations in several region evolution are obviously dependent temperature regimes are needed to follow upon the characteristics of the emerging flux the evolution of the overlying loop systems. itself. Other aspects depend upon the sur- The rapid evolution (seconds) of these loops rounding flows that are themselves influ- requires rapid cadence. enced by the presence of the active region. Coronagraphic Images are also needed in Magnetic Images are required to determine conjunction with the magnetograms to de- the configuration of the surface magnetic termine the magnetic field configuration as- fields associated with active regions. Vector sociated with active regions and their evolu- magnetograms are required to provide a re- tion. alistic assessment and to indicate the pres- ence of electrical current systems and non- Photometric Images may reveal thermal potential magnetic field systems. Vector structures associated with the active regions magnetograms will also give evidence as to before they emerge. whether flux leaves the surface predomi- nantly as bubbles, or whether it is princi- UV/EUV Spectra are needed to provide in- pally the outcome of local annihilation of formation on the physical characteristics of fields of opposing polarity. High cadence atmospheric structures within active regions. MDI observations have shown that the mag- These measurements may provide important netic configuration changes in intervals as information on reconnection events associ- short as minutes. Therefore the fields must ated with the evolution of active region be measured on a comparable cadence in magnetic fields. order to separate temporal and spatial varia- tions. 3.2.3 Small-Scale Magnetic Structure Helioseismic Images are needed to con- struct subsurface maps with sensitivity and While the large-scale distribution of mag- turnaround time superior to present capa- netic flux on the Sun determines the cou- bilities. We also need helioseismic mapping pling from the Sun to the heliosphere, the of the far side of Sun with fast turnaround small-scale mixed polarity flux may provide time. Helioseismic probing of the critical most of the energy that heats the corona. In near surface layers (a region that all acoustic order to understand the distribution of fields waves travel through and, because of the on the Sun we must identify the types of lower sound speed, spend much of their time sources of magnetic flux, the mechanisms in) requires high spatial resolution and rapid for the spreading the flux, and the manner in cadence observations. Probing of the three- which flux disappears from the surface. The

27 various mechanisms of flux disappearance characterized as a “salt and pepper” pattern are key to understanding the local heating of or the “magnetic carpet”. How it is created, the outer atmosphere, the acceleration of the how it evolves, and how it is destroyed are solar wind, and the contribution these ele- all uncertain. Presumably it is a manifesta- ments make to the solar irradiance varia- tion of a local dynamo at work in the upper tions. photosphere, but its relation to other scales of magnetic activity is not known.

Observations during the last few years have established that a significant fraction of the observed flux on the solar surface emerges in ephemeral regions – bipolar regions with lifetimes of hours. The ephemeral region number distribution, like the active region number distribution, is also exponential. Although it has not been as well measured as the active region flux over the cycle, there is strong indication that the amplitude of the Figure 3.13. Small-scale magnetic elements (red and cyclic variation is a factor of two or less. blue contours) and the coronal network bright points Consequentially, the ratio of total active re- (SOHO/EIT image). The bright coronal features are gion to ephemeral region flux varies consid- well correlated with magnetic elements in general erably during the cycle. Another component and magnetic dipoles in particular. of the flux is the inter-network field. This component has not yet been well character- 3.2.3.1 Small-Scale Magnetic ized. There is evidence that the field strength Structure – Current Un- of the inter-network field is 500 Gauss or derstanding less, while the active region and ephemeral region magnetic structures, excluding sun- Away from active regions, most of the pho- spots, have field strengths of 1100 ±100 tospheric magnetic flux appears to exist in Gauss. the form of flux tubes. These structures have complex dynamics. They form when a suffi- When active region flux emerges it forms cient amount of flux is aggregated by local plage as well as sunspots. Plage are regions velocity flows to affect the transport of in which the average magnetic flux is 125 ± thermal energy. A cooling and subsequent 25 Gauss. Since the typical magnetic ele- collapse of the mass in the tube leads to a ments in plage have field strengths of about compression of the field into a kiloGauss- a kilogauss, the magnetic filling factor for strength configuration. This structure per- plage is about 12%. Quite remarkably, plage sists until torn apart by convection. maintain their average field strength as the active region grows and finally disappears. An unknown amount of flux exists outside Outside plage, in the quiet Sun, the average of active regions and the strong flux tubes. field flux is about 5 Gauss. Because about Some estimates indicate that the total flux 90% of the flux in the quiet Sun also has may be as much as or more than that con- field strength of a kilogauss, the quiet Sun tained in active regions over the solar cycle. filling factor is about 0.5%. Subsurface im- This pattern of mixed polarity has been aging should collect data that will allow in-

28 sights into the structure of the flow and flow, and the latitude that the meridional magnetic fields below the plage that give flow terminates, these linear models produce raise to the remarkable bimodal character of remarkable agreement with low-resolution plage and quiet Sun. observations of the flux distribution over the surface. Unfortunately, the value of the dif- fusion coefficient that is required by the models is higher by a factor two than the measured values reported in the literature. This is not very serious because the majority of the measurement techniques for deter- mining the diffusion constant strongly weight high concentrations of flux. Recent MDI observations have shown that the dis- persal of flux is a function of the magnitude of the size of the flux concentration and that larger flux concentrations disperse more slowly than small ones. (Of course, these results mean that the uses of a fixed diffu- sion constant and consequently linear model of the surface spreading of magnetic field are not sufficient for an accurate description Figure 3.14. Histogram for the number of magnetic of flux dispersal.) Much more serious is that regions as a function of size (area). Ephemeral re- both the strength and the latitudinal structure gions (open circles) appear as an extension of the of the meridional flow used in the models distribution function for active regions (closed cir- cles). are not consistent with the observations. The poleward flow speed and the latitude of the Once field elements leave the plage, they are upper cutoff control both the diameter and dispersed over the surface by the diffusive the magnetic flux of the polar cap field. action of convection, by the differential ro- There are strong indications from both tation, and by meridional flows. Linear ground and MDI measurements that these models of flux distribution by diffusion, ro- poleward current systems may also vary tation, and flows have been developed over both in longitude and time. the past 40 years and have done an excellent job of predicting the mean characteristics of Kinematic modeling that includes the super- the surface flux distribution. In particular, granulation flow shows that there must be a given the observed plage strength, the mean constant eruption of new flux to maintain field with a spatial resolution of a few arc- the measured total flux observed with a minutes (see Fig. 3.9) and the mean field at resolution of a few arcseconds. Constant the pole can be predicted. However, these eruption of flux is also required to maintain models do not properly reproduce the num- the observed quiet Sun network. Simula- ber distribution of flux, the total flux with tions show that the magnetic network would resolutions of a few arcseconds, or the spa- disappear in days without continuous re- tial distribution of flux over the surface. placement. The most recent measurements of the ephemeral flux emergence rate yields By proper choice of the value of the diffu- about 5 1023 Mx/day, which is equivalent to sion coefficient, the speed of the meridional 10 to 20 large active regions. This rate is

29 sufficient to replace the quiet Sun network in less than a day and the flux in an active Local dynamo action is extremely interest- region, if the emergence rate is the same as ing as a process that has the capacity to con- in quiet Sun, in a few weeks. Because the tinuously generate energy that can then be rate of emergence of ephemeral flux is so used to heat the chromosphere, transition high, 10 to 30% of the quiet Sun flux may region and corona. In addition models sug- reside in the cell centers just because the gest that local and global dynamos can cou- travel time to the boundary is only a small ple and thus produce variations in the length fraction of the total flux lifetime on the sur- and amplitude of the solar cycle. face.

Ephemeral region flux emerges inside su- 3.2.3.2 Small-Scale Magnetic pergranulation cells and is transported to the cell boundary where is cancels with existing flux in the magnetic network. This means that ephemeral flux is not just the result of magnetic loops emerging through and then Structure – Outstanding sinking back below the solar surface. Rather ephemeral regions must disappear by some Questions type of reconnection process. There are a number of methods for estimating the en- How does magnetic reconnection of solar ergy released in the reconnection. But using magnetic fields on small spatial scales re- the initial field strength and the foot point late to coronal heating, solar wind accel- separation of the emerging loop to estimate eration and the transformation of the a magnetic volume energy, the result is 3x large-scale field topology? There are indi- 107 ergs/cm2/sec. This energy emergence cations that the interactions between oppo- rate is two orders of magnitude more than site polarity elements contribute to coronal necessary to heat the corona, but most of the heating. These interactions may also provide energy maybe dissipated in the chromos- the triggers for eruptive events that restruc- phere and transition region. ture the large-scale field. How are the small-scale magnetic features There is clear evidence from very high- produced? There is now good evidence for resolution filter magnetograms that magnetic magnetic dynamo activity in the thin layer field emerges on the scale of granulation. just below the photosphere. There is strong Spectropolarimetry indicates that there is a shear across this layer and the rate at which component of the field with a strength that is magnetic flux appears and disappears sug- probably less than 500 Gauss, and that there gests local generation. is a significant amount of weak horizontal field in quiet Sun. Taken together these ob- What are the dynamics of magnetic struc- servations constitute a strong suggestion that tures on very small spatial scales? Several there are surface magnetic fields associated mechanisms may be involved in the appear- with every form of convection. Numerical ance, disappearance, and interactions be- simulations indicated that a weak seed field tween these small magnetic structures. High- is amplified sufficiently by convective flows resolution vector magnetograms show that to generate a mixed polarity field that con- many magnetic elements seen in longitudi- tains 25% of the kinetic energy originally in nal magnetograms may just be intersections the convection.

30 of undulating flux ropes with the photo- sphere.

Magnetic Images are required to identify and to follow the evolution of these ele- ments. Full-disk observations with enough spatial resolution (~1 ) to identify them will be needed to determine their character- istics. The temporal cadence should be fast enough to follow their evolution and inter- actions (~ 5 min is probably the best that can be achieved). Longitudinal magnetograms should be sufficient for most studies in- volving the small-scale magnetic elements but vector magnetograms will be required to understand the flux ropes forming the small- scale elements.

Helioseismic Images are needed to deter- mine the nature of the fluid flows that sur- round these elements and to examine their sub-surface structure. Since flows close to the surface will be the most significant, the oscillation images will need the spatial (~1 ) and temporal (<50 sec) resolution to resolve them and their temporal evolution. Atmospheric Images are needed to deter- mine the effects of these elements on coro- nal and chromospheric heating. These im- ages should also have the spatial resolution (~1 ) to identify the elements and the tem- poral resolution (~10 sec) to follow their evolution. These images need to cover a wide range of temperatures to determine the effects of these elements on atmospheric heating and dynamics.

UV/EUV Spectra are needed to provide in- formation on the physical characteristics of atmospheric structures above these small- scale magnetic elements. These measure- ments should provide important information on events associated with coronal heating by these magnetic elements.

31 return from SDO but should not be included 4 Required Observations at the expense of higher priority observa- tions. An array of observations is needed to ad- dress the scientific questions posed for SDO. The highest priority observations include Each of these observations and their char- (from the interior outward): acteristics can be traced back to one or more · Helioseismic Images of the outstanding questions stated in the · Longitudinal Magnetograms previous section. (Note that all six of the · Atmospheric Images scientific areas discussed in the previous · EUV Spectral Irradiance Section are deemed critical to the SDO Mis- · Total Solar Irradiance (should con- sion. The scientific links among the different current measurements not be avail- areas suggest that all would suffer from the able from other platforms) omission of any one.) Since single instru- The high priority observations include: ments might be designed to provide more · Photometric Images than one measurement (e.g., helioseismic · Vector Magnetograms images and magnetograms), we present the · UV/EUV Spectra requirements for the measurements them- · Coronagraphic Images selves in the following sub-sections. Poten- The important observations include: tial instruments and the possible allocation · Coronal Spectroscopy of resources are given in Section 5. · Heliometry Further refinement of priorities within these All of these observations are important for groups is unproductive at this phase of the one or more of the scientific areas to be ad- mission definition. The actual resources re- dressed by SDO. However, some are quired to obtain the measurements, the ac- deemed to be of higher priority than others tual instruments to be flown on other mis- for a variety of reasons. Some of these ob- sions, and other programmatic considera- servations are either unique to SDO or have tions could easily change those priorities. requirements that can be uniquely met by SDO (e.g. full-disk spatial coverage, con- tinuous time coverage over several years, 4.1 Helioseismic Images rapid cadence, high bandwidth). Other ob- servations may be provided by other pro- Helioseismic images are required for studies grams, may be more speculative in nature, of the Sun’s internal flows and structures. or may be more difficult to obtain with cur- This impacts all aspects of SDO science. rent technology and resources. We recognize Helioseismic images provide information on the need for setting priorities for SDO and solar radius changes for irradiance studies have done so by grouping the observations (Section 3.1.1.2) and on sub-surface flows and their respective instruments into three that drive magnetic field evolution associ- priority categories: highest priority – SDO ated with flares and CMEs (Section 3.1.2.2) must make these observations; high priority and with coronal structures and solar wind – due effort should be made to include these variations (Section 3.1.3.2). Helioseismic observations but for one reason or another images provide critical measurements of the (as discussed in the following sub-sections) internal flows that drive the solar cycle the mission could go on without them; im- (Section 3.2.1.2), active region evolution portant – these would enhance the scientific (Section 3.2.2.2), and the dynamics of the small-scale magnetic elements (Section

32 3.2.3.2). One of the highest priorities for Field Of View. In order to detect fields and SDO is to obtain the helioseismic observa- flows nearer the base of the convection tions required to provide a detailed view of zone, we need simultaneous observations the upper 10-20 Mm of the convection zone. across the full disk (again 75° from disk These measurements are of highest priority center is a reasonable goal, this is 0.97 in for SDO. They will not be available with the units of the radius). required spatial resolution and full-disk cov- erage from any other source. Completeness. Gaps in the data stream compromise the helioseismic objectives. Accuracy. Signals near the base of the con- Experience with MDI and GONG indicate vection zone likely generate wave travel that nearly complete coverage (more than time variations as small as 0.01 second. The 95% of the time) is needed. total travel time approaches 2 hours so the timing accuracy must be on the order of a Table 4.1: Measurement Characteristics part in 106 over intervals of hours to days. for the Helioseismic Imager

Dynamic Range. Previous experience Observable Osc. Time Series shows that the best signal-to-noise can be Accuracy (Clock) 10-6 obtained with Doppler observations as com- Dynamic Range 13 km/s pared to brightness observations in the 1 to 5 Time Cadence < 50 sec mHz region. Therefore dynamic range is Spatial Resolution 1 arcsec discussed here in terms of velocity. The in- Field of View Full Disk strument must have a velocity dynamic Duration 10 years range large enough to accommodate both the Completeness 99.99% coverage spacecraft orbit and the Sun’s rotation. The 95% of time orbit will have a range of about 6 km/s and the Sun adds about 7 km/s from East to 4.2 Longitudinal Magneto- West limbs giving a dynamic range of about 13 km/s. grams

Time Cadence. There is useful wave power Magnetic images are critical for all of the up to about 8 mHz and measurable power to scientific areas to be addressed by SDO. at least 12 mHz. A sample with a 10 mHz Longitudinal magnetograms are needed: to Nyquist frequency leaves the spectrum clean distinguish between magnetic and non- up to 8 mHz. This gives a time cadence of magnetic sources of the irradiance variations about 45 seconds per measurement. (Section 3.1.1.2); to determine the configu- ration of the magnetic fields associated with Spatial Resolution. A reasonable goal is to flares and CMEs (Section 3.1.2.2) and solar sample the wave-speed profile and flows wind variations (Section 3.1.3.2); to follow with about 1 Mm depth resolution in the up- the evolution of the magnetic field over the per 10-20 Mm of the convection zone. This solar cycle (Section 3.2.1.2) and as active requires sample points 3 Mm apart. At 75° regions evolve (Section 3.2.2.2); and to from disk center, this corresponds to about 1 identify and follow the small magnetic ele- arcsec resolution. This can be provided with ments (Section 3.2.3.2). Longitudinal mag- 0.5 arcsec detector pixels. netograms are of highest priority to SDO. Full-disk, high-resolution magnetograms will not be obtained with the required tem-

33 poral resolution and coverage by any other means. 4.3 Atmospheric Images

Precision. A pixel-by-pixel precision be- Atmospheric images are required for all as- tween 5 and 50 G over a 5 minute integra- pects of the SDO mission. They indicate the tion is needed to follow the evolution of the sources of the EUV irradiance variations small-scale magnetic elements. (Section 3.1.1.2), the occurrence and nature of flares and CMEs (Section 3.1.2.2), and Accuracy. The disk integrated magnetic the structure and dynamics of the low co- flux should be measured with an accuracy of rona (Section 3.2.3.2). They reveal the mag- about 0.1 G. This level of accuracy allows netic structures associated with the solar cy- for studies of active region flux imbalances cle (Section 3.2.1.2), active region evolution and determinations of the global field con- (Section 3.2.2.2), and small-scale magnetic figuration. elements (Section 3.2.3.2). These observa- tions are of highest priority to SDO. Full- Dynamic Range. Sunspot umbrae harbor disk images at the required spatial and tem- the strongest magnetic fields with peak val- poral resolution will not be obtained by any ues of about 3000 G. A dynamic range of 6- other means. 7 kG would capture even the strongest flux concentrations. Precision/Accuracy. A photometric accu- racy of about 10% is needed for these im- Time Cadence. Individual magnetograms ages to be useful as a tool for the determin- should be obtained at a cadence of at least 1 ing the sources of the spectral irradiance. minute to account for the Doppler shifts due Similar accuracy is needed in making useful to the 5-min oscillations. ratios of the images to determine physical Spatial Resolution. High spatial resolution characteristics of the atmospheric features. (~1 arcsec) is required to identify the small These imagers must be properly calibrated intranetwork magnetic elements. to insure the stability and repeatability of the measurements at the 10% level. Field Of View. Full-disk measurements are Dynamic Range. A high dynamic range required to provide the global field configu- 3 ration, to monitor all active regions, and to (10 or greater) is needed to provide obser- provide complete coverage for studies of vations with sufficient signal to noise and to irradiance and small magnetic features. span the range of emission features from intra-network loops to bright active regions. Table 4.2: Measurement Characteristics Time Cadence. An important diagnostic of for the Longitudinal Magnetograph coronal activity may lie with transient flow patterns that are perhaps created by short- Observable Longitudinal B term heating events. The flow velocities on Precision 5-50 G / 5 min the order of 100 km/s, near the sound speed, Accuracy 0.1 G are often observed. A cadence of 10 sec- Dynamic Range Several kG onds would be sufficient to track these flows Time Cadence ~ 1 min from resolution element to resolution ele- Spatial Resolution 1 arcsec ment between consecutive exposures. Field of View Full Disk Duration 10 years

34 Spatial Resolution. We know from meas- needed to effectively span this range is un- urements with HRTS and SUMER that the certain. The minimum is probably four but atmospheric structures are highly filamen- seven would be far more useful. tary. TRACE observations show that with 1 arcsec spatial resolution, solar structures are Table 4.3: Measurement Characteristics recorded with sufficient clarity that much of for the Atmospheric Imagers the confusion produced by the fine structure is overcome. CCDs with a 4096x4096 for- Observable Intensity mat will soon become available and would Precision/Accuracy 10 % provide a 1.2 arcsec resolution over the 40 Dynamic Range 103 – 105 arcmin field of view. Time Cadence 10 sec Spatial Resolution 1.2 arcsec Field Of View. Full-disk images are re- Field of View 40 x 40 arcsec2 quired to observe flares and CMEs and to Spectral Resolution l/Dl ~ 20 identify sources of EUV irradiance variabil- Temperature Range 0.02 – 4 MK ity. CMEs are often large-scale events that can involve a significant portion of the solar 4.4 EUV Spectral Irradiance disk. In addition, it would be useful to ob- serve the initial acceleration of CMEs near The EUV irradiance and its variations influ- the solar limb with some overlap with the coronagraph. Consequently, a field of view ence ionospheric and thermospheric compo- sition, density, and temperature. The need (FOV) of 40 arc-min (1.25 R ) is speci- Sun for continuous observations of these varia- fied. tions (Section 3.1.1.2) drives the SDO re- quirements for an EUV Spectral Irradiance Spectral Resolution. Relatively high spec- Monitor. These observations are of highest tral purity (resolving power of 20 or so) is priority to SDO. While EUV spectral irradi- needed so that the individual images refer to ance observations will be obtained by the only a small temperature range. SEE instrument on the TIMED mission, these observations will not be concurrent Temperature Range. Solar activity often with the SDO mission and are of lower involves the nearly simultaneous rapid evo- spectral and temporal resolution. The SDO lution of plasmas over a wide range of tem- EUV observations play a critical role in peratures, such as hot flare plasmas, the helping us to understand how solar variabil- eruption of cool prominence material and ity influences the ionosphere and thermo- the reconfiguration of coronal loops during a sphere. Tracking the solar cycle variations, flare or CME event. To understand the dy- and the possibility of longer-term secular namics of the coronal field, we must be able variations is desirable. The observations to trace the loops as they evolve in tem- should be obtained continuously to maintain perature. That requires narrow pass bands a calibrated time series, and to provide con- because only those allow the clear identifi- tinuous inputs for geophysical studies un- cation of loops while minimizing the line of dertaken elsewhere within the LWS pro- sight confusion. The ability of the Atmos- gram. pheric Imagers to observe plasmas over a wide range of temperature (0.02 – 4 MK) Precision/Accuracy. The absolute uncer- will help untangle the relevant physical pro- tainty (accuracy) translates into uncertainties cesses. The number of image bandpasses in geophysical parameters (densities, tem-

35 peratures) calculated by geophysical models photoelectron production and heating, re- that use these inputs, so the goal is - quires input spectra with 0.1 nm resolution. mum accuracy. Accuracies of 10% reflect the state of the art irradiance calibrations Spectral Range. Understanding ionospheric using synchrotron irradiance standards. Deg- and neutral density responses to solar vari- radation of the optics and detectors forces a ability requires knowledge of the solar en- requirement for some on-board calibration ergy inputs over a spectral range from 1-120 as well. The measurements should have nm in both lines and continua. long-term stability on the order of 5% for useful solar cycle studies. Table 4.4: Measurement Characteristics for the EUV Spectral Irradiance Monitor Dynamic Range. The solar cycle variability ranges from 50 percent to two orders of Observable Spectral Irradiance magnitude over this spectral range, with Precision/Accuracy 10 % even larger variations (a factor of up to 1000 Dynamic Range 103 in some lines) occurring during flares. A dy- Time Cadence 10 sec namic range of about 1000 can be obtained Spatial Resolution None with a combination of detector dynamic Field of View 1° range and different exposures. Spectral Resolution Dl ~ 0.1 nm Spectral Range 1 – 120 nm Time Cadence. Synchronization with the SDO Atmospheric Imagers is desirable for tracing the flux to the magnetic (and possi- 4.5 Photometric Images bly other) sources of its variability and for formulating physical mechanisms. This in- Photometric image data is essential to find- dicates a similar time cadence of about 10 ing the origins of solar irradiance and lumi- sec for the EUV irradiance measurements. nosity variability (Section 3.1.1.2). These images may also reveal thermal structures Field Of View. A slightly larger FOV than associated with the solar cycle (Section the solar disk will minimize pointing error 3.2.1.2) and active region evolution (Section problems. The FOV should be sufficient to 3.2.2.2). These observations are of high pri- capture off-limb radiation that may contrib- ority for SDO, as they address three impor- ute to the flux of some hot coronal lines. tant areas. Data of high photometric sensi- tivity or extended bandwidth will only be Spectral Resolution. Solar emission lines obtained from space and would be unique to that are close in wavelength can nevertheless SDO. However, empirical modeling of the have quite different variabilities if they have sources of the irradiance variations suggests different emission temperatures i.e., sources that the sources are well understood and he- in different solar atmospheric regions. At lioseismic determinations of solar radius wavelengths longer than 10 nm, line and variations indicate that these variations continuum energy inputs to the upper at- probably play only a minor role in produc- mosphere and ionosphere must be separately ing the irradiance variations. The ability to measured and distinguished to achieve the image thermal perturbations associated with required geophysical understanding. AURIC active regions before they emerge is some- (Atmospheric Ultraviolet Radiation and what speculative but may prove to be revo- Ionization Code), which is used to calculate lutionary.

36 It is likely that the photometric mapper (PM) Table 4.5 Photometric Mapper Charac- will operate with two channels. In its first teristics channel it will achieve the highest possible photometric accuracy and spatial resolution. Channel Photometric Bolometric The second channel captures the widest pos- Observable Surface Bolometric sible spectral coverage (approaching total Intensity Intensity solar irradiance bolometric spectral cover- Precision 0.1 % 3 % age) with sufficient spatial resolution to Dynamic >103 30 identify photospheric magnetic contributions Range as small as faculae. Time Cadence 1 min 1 min Spatial 1 arcsec 10 arcsec Precision. In its first channel the PM must Resolution have photometric stability and sensitivity Field of View Full Disk Full Disk sufficient to detect faint diffuse brightness Spectral Narrow Broad changes associated with surface magnetic Resolution band band features like sunspot bright rings, facular Completeness 95% 95% shadows and the network irradiance pertur- bations. Detecting these subtle features will 4.6 Vector Magnetograms require a precision of about 0.1%. In its sec- ond channel the PM must have the ability to Vector magnetograms are needed to deter- tally the contributions to the irradiance mine: the magnetic stresses and current sys- variations from the various photospheric tems associated with flares and CMEs (Sec- features. A precision of 3% at each pixel tion 3.1.2.2); the nonpotentiality of coronal should be sufficient for the number of pixels magnetic fields (Section 3.1.3.2); and the given by the spatial resolution requirement. large-scale magnetic field pattern evolution associated with the solar cycle (Section Dynamic Range. The dynamic range at 3.2.1.2) and active region evolution (Section each pixel should be sufficient to capture the 3.2.2.2). These are high priority observa- range of intensities from the darkest sunspot tions for SDO. They address critical needs in umbrae to the brightest plage and faculae. several areas. However, full-disk vector magnetograms with the required spatial Time Cadence. The images should be ob- resolution may be difficult to obtain with tained at about 1 minute intervals to resolve current technology and mission resources. the p-mode oscillation component of the ir- Limited field of view observations at higher radiance variations. resolution will be acquired with instruments on Solar-B (Section 7.2). Full-disk meas- Spatial Resolution. Photospheric faculae urements from the ground will be acquired should be resolved in these images. A spatial with the SOLIS instruments (Section 7.6). resolution between 1 arcsec and 10 arcsec These alternative observations do not, how- should be sufficient. ever, fully satisfy the SDO requirements. Solar-B has a very limited field of view and Field Of View. The full photospheric disk SOLIS will have poorer spatial resolution must be imaged for the mapper to provide and temporal coverage. the observations needed to determine the sources of the irradiance variations. Precision. The precision of the transverse field direction measurements should be a

37 few degrees. This translates into a polariza- -4 tion precision of ~10 . The precision of the 4.7 UV/EUV Spectra field strength measurements should be better than 50 G for a 10-minute integration time. The UV/EUV Imaging Spectrometer will provide spectral images and measurements Accuracy. The vector field measurements that reveal quantitatively the dynamics of should yield the correct vector field direc- the solar atmosphere, from the photosphere, tion in each resolution element to within five through the transition region to the corona. degrees. The longitudinal component accu- The UV/EUV Imaging Spectrometer will racy should be consistent with a disk inte- provide quantitative constraints on the grated flux accuracy of about 0.1 G. physical mechanisms associated with: the sources of the spectral irradiance variations Dynamic Range. Sunspot umbrae harbor (Section 3.1.1.2): the impulsive release of the strongest magnetic fields with peak val- energy that results in flares and CMEs (Sec- ues of about 3000 G. A dynamic range of 6- tion 3.1.2.2); and the sources of high-speed 7 kG would capture even the strongest flux and low-speed solar wind (Section 3.1.3.2). concentrations. UV/EUV spectra are also needed to provide information on the physical characteristics Time Cadence. Polarimetric scans should of atmospheric structures in active regions be obtained at a cadence that allows the (Section 3.2.2.2) and above small-scale Doppler shifts due to the 5-minute oscilla- magnetic elements (Section 3.2.3.2). These tions to be removed or compensated for. observations are of high priority to SDO. Full-disk vector magnetograms should be They are required to address several out- obtained at a cadence of about 6 per hour. standing questions. Although there is an EUV Spectrometer planned for the upcom- Spatial Resolution. High spatial resolution ing Solar-B mission (Section 7.2), it is not (~1 arcsec) is required to identify the small ideally suited to the SDO scientific objec- magnetic elements. tives for two reasons. First, the Solar-B EIS is primarily a coronal spectrometer, with Field Of View. Full-disk measurements are only one strong emission line below 106 K. required to provide the global field configu- Second, the estimated count rates of the So- ration, to monitor all active regions, and to lar-B EIS are too low to match the temporal provide complete coverage for studies of cadence of the Atmospheric Imager Array. irradiance and small magnetic features. The SDO UV/EUV Imaging Spectrometer nicely complements the Atmospheric Imag- Table 4.6: Measurement Characteristics ing because it provides quantitative obser- for the Vector Magnetograph vations with the spatial and temporal resolu- tion necessary to resolve ambiguities seen in Observable Vector B the imager observations. Transverse Precision 50 G (~3°) -4 Polarimetric Precision ~10 Precision/Accuracy. Spectral intensity Dynamic Range Several kG should be measured to within 10% to con- Time Cadence ~ 10 min tribute to the spectral irradiance studies. Spatial Resolution 1 arcsec Line widths should be measured to within Field of View Full Disk about 10% to provide useful information on Duration 10 years non-thermal broadening associated with

38 physical mechanisms in the relevant fea- tures. Doppler velocities should be measured Table 4.7: Measurement Characteristics to within 1-5 km/s to determine the nature of for the UV/EUV Imaging Spectrometer the dynamic events. Observable Line Profiles Dynamic Range. Spectral intensity varies Precision/Accuracy Intensity 10 % widely from one spectral feature to another Width 10% and from quiet sun to active plage. A dy- Velocity 1-5 km/s 3 5 namic range of 103 to 105 should capture all Dynamic Range 10 – 10 these features. Time Cadence 10 sec Spatial Resolution ~1 arcsec Field of View 16 to 34 arcmin Time Cadence. Ideally, the time cadence of the UV/EUV Imaging Spectrometer should Spectral Resolution l/Dl ~ 30,000 match the 10 sec cadence of the Atmos- Temperature Range 0.02 – 4 MK pheric Imaging Array. This can be accom- plished with limited FOV raster motions 4.8 Coronagraphic Images about a given location (e.g. central meridian) or target (e.g. active region), repeated Coronagraphic images are important for throughout the majority of the observing pe- many aspects of the SDO Mission. These riod. Synoptic observing programs provid- images indicate the occurrence of CMEs ing raster images of the entire disk may also (Section 3.1.2.2) and reveal the structure of be performed periodically. the corona and presence of solar wind varia- tions (Section 3.1.3.2). They also provide Spatial resolution. A spatial resolution of important information on changes in the 1.3 arcsec should be sufficient to resolve the magnetic configuration of the corona associ- facular elements associated with the spectral ated with the solar cycle (Section 3.2.1.2) irradiance variations. and active region evolution (Section 3.2.2.2). The coronagraphic images are of high priority to SDO. They provide critical Field Of View. The routine raster FOV will information on events and processes associ- be limited by the necessity to match the time ated with several outstanding questions. cadence of the Atmospheric Imaging Array. However, the STEREO mission (Section However, it will also be necessary to ob- 7.1) is designed to address many of these serve the full solar disk and inner corona by questions. Nonetheless, coronagraphic im- slit rasters or other techniques. ages from SDO will be needed to span the duration of the mission and provide infor- Spectral Resolution. A spectral resolution mation on the occurrence of CMEs and solar (resolving power) of 30000 should be suffi- wind variations. cient to measure the required line profile information. The SDO coronagraphic images will proba- bly require two separate channels in order to Temperature Range. The measurements map the structure of the corona from 1.1 should include spectral features that cover RSun to 15 RSun. The stray light suppression the temperature range of the Atmospheric requirements are such that it is not practical Imaging Array – 0.02 to 4 MK. for a single coronagraph channel to be used for the entire range of heights. Further, a de-

39 sired spatial resolution of 12 arcsec or better nels respectively, are required and sufficient

at the inner edge of the FOV (1.1 RSun) re- for accurate determinations of liftoff times, quires an internally occulted coronagraph helical motions, and speed profiles for dif- system. The channel for the outer corona ferent parts of coronal mass ejections. should provide overlapping coverage with the other coronagraph channel. That implies Spatial Resolution. Observations of CMEs that this channel has to be an externally oc- should have a high enough spatial resolution culted system. The requirement for halo to discern their small-scale structures. CME detection is a high priority for the Plumes and polar rays are on the order of a LWS mission, in general, and SDO, in par- few arc-minutes in width but they have fine- ticular since these are CMEs that affect scale structure on the sub arc-minute level Earth and have their origins from regions that is probably related to their magnetic directly observed on the disk with the At- field configurations. A 2048 x 2048 CCD mospheric Imaging instruments. detector would give a spatial resolution of 6” for the inner coronagraph and 30 arcsec Precision. The polarization brightness for the outer coronagraph. This would give should be measured with a precision of SDO twice the resolution of LASCO C1 and about 10% to provide useful information on a comparable resolution to C2; it is more coronal structures and their variations. than adequate to meet the SDO science re- quirements. Dynamic Range. A dynamic range of about 103 captures most of the observed variations Table 4.8: Measurement Characteristics in the inner corona. A somewhat wider dy- for the Coronagraph namic range of about 104 captures most of the observed variations in the outer corona. Channel Inner Outer Observable Polarized Polarized

Field Of View. An inner height of 1.1 RSun Intensity Intensity is required to overlap the FOV of the At- Precision 10 % 10 % 3 4 mospheric Imaging instruments and thus Dynamic Range 10 10 allow for continuous tracking of events like Time Cadence 1 min 5 min CMEs from their initiation in disk imagers Spatial Resolution 6 arcsec 30 arcsec through their development in the extended Field of View 1.1-3. RSun 2.5-15 RSun corona. An outer height of about 15 RSun is Spectral Range 400-700 400-700 needed in order to allow detection of halo nm nm CMEs that are directed toward the Earth. Experience from LASCO observations shows that beyond 15 RSun the propagation properties of CMEs don’t change much and 4.9 Total Irradiance so this outer limit would be acceptable. The total solar irradiance must be accurately Time Cadence. High time cadence will be and precisely monitored to determine the required to study the dynamics of coronal nature and source of the irradiance varia- disturbances (CMEs, streamer blowouts, tions (Section 3.1.1.2). These observations eruptive prominences, etc.). SOHO/LASCO are of highest priority to SDO but will likely observations have shown that 1-min and 5- be obtained from both SORCE (Section 7.4) min cadences, for the inner and outer chan- and GOES/NPOESS (Section 7.5) during

40 the SDO mission and are therefore not in- Table 4.9: Measurement Characteristics cluded as part of SDO. If, however, it ap- for a Total Solar Irradiance Monitor pears that these observations will not be provided by these alternative sources then a Observable Total Irradiance Total Solar Irradiance Monitor should be Precision/Accuracy 0.01% placed on SDO. Repeatability 0.001% per year Time Cadence 1 min Precision/Accuracy. The absolute uncer- Duration Solar Cycle tainty (accuracy) translates into uncertainties Completeness continuous in the energy input to the terrestrial climate Field of View 2° system. The goal is maximum accuracy to minimize uncertainties in climate models, and also to ensure that the instrument is 4.10 Coronal Spectroscopy properly characterized to achieve the needed high repeatability. The actual Spectroscopic measurements in the extended input to the radiometer depends on the en- corona are needed to for characterizing the trance aperture, but is typically of order 100 mechanisms that accelerate CMEs (Section milliWatt. The ability to measure the change 3.1.2.2) and the solar wind (Section 3.1.3.2). in this signal due to changes in total solar These observations are important for SDO irradiance depends, in part, on the noise but are limited in scope and would require floor of the radiometer electronics. This ra- significant resources. Spectroscopic meas- tio sets the dynamic range and it must be urements in the corona up to about 2 R sufficient to enable the required repeatability Sun can be accomplished with either an inter- and uncertainty, at a cadence of 1 observa- nally occulted coronagraph design (see, e.g. tion per minute. LASCO-C1) or a very sensitive “wide an- gle” EUV spectrograph. However spectro- Time Cadence. The 5-minute oscillations scopic measurements in the extended corona affect total solar irradiance, and the instru- (beyond 1.5 R ), require an externally oc- ment must be capable of resolving these in Sun culted (see, e.g. SOHO/UVCS) time. due to the rapid decrease with height of co- ronal emission line intensities. A large ap- Duration. Tracking the solar cycle varia- erture, externally occulted coronagraph- tions, and the possibility of longer term spectrometer system would have the proper secular variations, is desirable. stray light suppression and sensitivity to al- low for the measurement of dozens of faint Completeness. The instrument should be coronal lines out to heliocentric distances of operated continuously to maintain a cali- 10 R . Such an instrument would be capa- brated time series, and to provide continuous Sun ble of characterizing sites of magnetic re- inputs for geophysical studies undertaken connection and shock formation in CMEs by elsewhere within LWS. observing high charge state ions and non- thermal line broadening. It will also measure Field Of View. A FOV slightly larger than velocity distributions of H, He, electrons, the solar disk will minimize pointing error and heavy ions to determine the power problems. spectrum of resonant MHD waves that may be responsible for heating and acceleration. Helical 3D velocities can be determined

41 from Doppler shifts and Doppler dimming. “acoustic” radius fluctuations. The Photo- Coronal source regions for CME and solar metric Images may be obtained in a manner wind plasma can be determined from abun- that also allows for heliometry. The magni- dance determinations. tude of the radius fluctuation, compared to the irradiance change during a solar cycle While advanced coronagraph-spectrometers contains important information on where would be desirable, we felt that such instru- and how energy is stored. If W is the ratio of ments could not be accommodated with the relative radius and irradiance changes, then present SDO spacecraft resources. How- physical models predict a wide range of val- ever, if an opportunity does arise for flying ues for W. Depending on the mechanism and this type of instruments it would be a valu- depth of the interior perturbation estimates able complement to the SDO Mission. of W range from 2´10-4 to 7.5´10-2. Given a solar cycle irradiance change of about 10-3, Table 4.10: Measurement Characteristics a desirable goal for SDO is to achieve a ra- for a UV Coronagraph Spectrometer dius sensitivity of at least 10 milliarcsec on solar rotation time-scales. Such measure- Observable Line Profiles ments will allow us to clearly discriminate Precision/Accuracy Intensity 15 % between competing solar cycle luminosity Width 10% variation models. Velocity 5 km/s Dynamic Range 105 Time Cadence 10 min Spatial Resolution 4” 5 Potential Instruments and Field of View 4~30” ´ 2000” Allocation of Resources Range of View 1.1-10 RSun Spectral Resolution l/Dl ~ 10,000 We have examined several generic instru- Spectral Range 28-140 nm ments for inclusion in the SDO payload to estimate the total mass, data rate, power and 4.11 Heliometry volume required to accommodate them. These estimates are included in Table 5. Measurements of small changes in the solar Several assumptions have been made in ar- radius and limb shape (heliometry) are riving at these estimates. The masses do not needed to determine how and where the include electronics boxes, mounting, radia- emergent solar luminosity is gated and tors, etc. The data rates assume the use of stored (Section 3.1.1.2). These measure- image compression. The range of masses, ments are important for SDO but should not data rate and volume for the Atmospheric require the resources of an additional in- Imaging Array is due to the number of pos- strument. The Helioseismic Images from sible telescope tubes. SDO will be important for observing solar

42 Table 5. Allocation of Resources

Instrument Mass Data Rate Power Volume (kg) (Mbps) (W) (cm3) HMI 40 25 60 90x40x25 Atmospheric 40-70 20-50 45 100x15x30 Imaging Array 100x45x60 EUV SIM 20-30 <1 45 44x48x21 Coronagraph 30 1 35 135x17x17 UV/EUV 40-60 15-30 40 160x60x30 Spectrometer Photometric 30 5 50 100x30x30 Mapper Vector 10+ 5 20+ 90x40x40 Magnetograph Total 210-270 62-117 295

6 Mission Concept downlink. The large data rate, along with the strict limitations on on-board storage capac- This section details the spacecraft, launch ity, result in an effective requirement of vehicle, ground system and data system for continuous contact. An inclined geosyn- SDO. These items have been discussed as chronous (GEO) orbit will allow nearly being sufficient to support the science as continuous observation of the Sun, and can defined in this Science Definition Team re- downlink data to a single dedicated ground port. Future adjustments and variations will station. undoubtedly occur, as the mission concept matures into the design and development The mission will launch into a geosynchro- phase. nous transfer orbit (GTO) and then use an apogee kick motor (AKM) or other orbital The science of the Solar Dynamics Obser- transfer mechanism to boost the spacecraft vatory optimally will be performed on a into geosynchronous orbit. The spacecraft spacecraft that allows nearly continuous ob- will be three-axis stabilized and will main- servations of the Sun and a scientific data tain solar pointing with occasional maneu- rate well in excess of 100 Megabits per sec- vers to unload the momentum accumulated ond. These two requirements drive the orbit in the reaction wheels. Twice a year SDO and spacecraft specification and the defini- will undergo “eclipse seasons,” which will tion of the SDO Mission. Nearly continuous last 2-3 weeks, with a maximum Earth observations can be obtained from other or- eclipse period of approximately 70 minutes. bits, such as a low Earth orbit (LEO), but a Spacecraft maneuvers, eclipses, and occa- LEO orbit would require on-board storage of sional ground system outages will interrupt large volumes of scientific data pending the scientific observations.

43 6.1 Orbit Selection Missions of the program are designed to perform investigations of the The orbit of the Solar Dynamics Observa- long-term variations of the Sun-Earth con- tory will allow a high science data rate (160 nected system. The SDO mission will be Mbps) and nearly continuous contact via a designed for a 5-year baseline with expend- single dedicated ground station. This ground ables to last an additional five years of an station can be built and operated at a fraction extended mission. of the cost of a mission that would rely on existing ground contact networks. The Table 6.1 illustrates the mass breakdown ground track of a geosynchronous orbit with used in a preliminary definition study for an inclination of 28.5 degrees orbit is shown SDO. Compared are the mass estimates for a in Figure 6.1, projected onto 102 degrees custom-built spacecraft and a spacecraft Earth longitude. from the Rapid Spacecraft Development Of- fice (RSDO) catalog. An instrument module mass of 200 kg, an Apogee Kick Motor (AKM) mass of 668 kg, and the RSDO catalog mass estimates were used in the study calculations. Modifications of a cus- tom-built bus would include a lighter bat- tery, a lighter structural composition of the bus, and adjustments to the propulsion and communication systems.

Table 6.1: SDO Preliminary Study Bus and Payload Mass Breakdown (kg)

Custom RSDO ACS 54 42 Power 60 92 Harness 15 19 RF comm. 35 72 w/ C&DH Figure 6.1: Ground track of geosynchronous orbit C&DH 11 with an inclination of 28.5 degrees. Propulsion 10 61 Hydrazine 20 20 The disadvantages of this orbit include AKM Stage 10 8 launch and orbit acquisition costs (relative to Balance Wt. 10 7 LEO) and eclipse (Earth shadow) seasons Thermal 60 twice annually. During these 2-3 week Mechanical 34 114 w/ therm eclipse periods, SDO will experience a daily Subtotal 319 434 interruption of solar observations. The maximum duration of these interruptions is Instruments 200 200 70 minutes, during which solar observations AKM 668 668 will be interrupted. The spacecraft attitude Total 1187 1302 control system (ACS) and power system must recover from each of these eclipse pe- riods, which may involve a longer duration

44 of the interruption of the scientific observa- ous launch vehicles, using an AKM such as tions. Three lunar shadow events also occur the Star-30E and a spacecraft mass of 634 annually, with durations of approximately kg. The Delta 2925 meets the launch mass 30 minutes. The total duration of these inter- criteria, but other vehicles such as the Delta ruptions is approximately 45 hours annually. Lite and the Taurus would require use of a Eclipse and shadow periods are shown in much lighter spacecraft and AKM, or would Figure 6.2. require the consideration of an alternate or- bit profile. During the design of the SDO mission, contingency and mass margins as well as mass and spin balance requirements must be taken into consideration; these items will constrain the total mass able to be ac- commodated.

Figure 6.2: Annual periods of Earth and lunar shadow (minutes per day).

Figure 6.4: Mass margins for various launch vehi- cles. The total mass of the spacecraft and AKM is shown in gray. The launch vehicle capabilities (in mass to GTO) are shown in blue. A Delta 2925 has a margin of nearly 500 kg. Figure 6.3: Illustration of a launch to geosynchro- nous transfer orbit (GTO) and orbit circularization 6.2 Attitude Control System with an Apogee Kick Motor. Most of the requirements of the SDO mis- SDO's orbit can be achieved with a launch sion can be met by a geosynchronous space- into a geosynchronous transfer orbit (GTO), craft from the RSDO catalog, with some with an apogee kick motor (AKM) to circu- necessary adjustments. The standard geo- larize the orbit at geosynchronous altitude synchronous ACS would include a star- (Figure 6.3. The total mass to be lifted to tracker, course and fine sun sensors, gyro- GTO includes the spacecraft and instru- scopic sensors and reaction wheels. Four ments as well as the AKM. Figure 6.4 com- reaction wheels in a pyramid configuration pares the mass capabilities to GTO of vari- will allow efficient unloading of accumu-

45 lated momentum and will provide redun- based on the field of view and the design of dancy for a long-term mission design. Atti- the coronagraph, the Sun must be centered tude control system considerations include behind the occulting disk to within a desir- the adaptation of a geosynchronous space- able tolerance. The accuracy and stability of craft to maintain a sensitive three-axis stabi- the roll of the spacecraft is specified in rela- lized solar pointing. These modifications tive terms; the HMI requires a stable view- would include a measurement from a ing angle, though the angle does not neces- guidescope in the instrument package which sarily have to be aligned at solar North. will link to the ACS to meet the stability re- quirements required by the instrument pack- Attitude Knowledge: Because of its re- age, and the use of low-jitter reaction wheels stricted field of view, the spectrometer (such as the SMEX Lite IRWA) to restrict would require a knowledge of a fraction of the jitter introduced to the system. the instrument resolution. Therefore, the knowledge in pitch and yaw would be set at The spacecraft ACS requirements which a fraction of an arc second (which can be were discussed were as follows: provided by the ACS or by the instrument package). The HMI requires a roll knowl- Jitter: The instruments with higher resolu- edge of 30-60 arcsec (3 sigma), corre- tion place restrictions on the amount the sponding to a fraction of a pixel at the solar pointing can vary during the collection of limb. data. The HMI and AIA require that the jit- ter in both pitch, yaw and roll not vary by SDO will have several modes of operation more than a fraction of a pixel over the in- including: Science mode - 3-axis zero- terval of time required to collect the data. momentum control, pointing roll axis toward Therefore, the image must be stabilized to the Sun, run wheels in bias speed, using .25 arcsec (3 sigma) over a few seconds measurement from guide telescope for pitch (determined by the image collection time) in and yaw and star tracker measurement for pitch and yaw. It is likely that the spacecraft roll; Calibration mode - includes possible ACS will not be able to satisfy this require- offsets and maneuvers for instrument cali- ment, and will have a pitch/yaw jitter near 5 bration; Safehold mode - uses CSS and arcsec over 45 seconds. The instruments wheel to point solar array normal to the Sun, would incorporate image stabilization using similar to Sun acquisition; Delta V mode - data from the guidescope. Because the im- for orbit maintenance; and Delta H mode - age stabilization system cannot correct for to unload momentum using propulsion jitter in the roll axis, the roll jitter must be held to within 50 arcsec (3 sigma) over a These operational modes may need to in- period of 45 seconds. clude periods of recovery from Earth shadow periods, and maneuvers to assist in Accuracy/Stability: It is estimated that an instrument calibration. image stabilization system cannot function properly at angles greater than 10-15 arcsec; 6.3 Data and Communication therefore, if the spacecraft introduces 5 arc- System sec jitter, the greatest deviation from the overall pointing must be within 5-10 arcsec Commands to the science instruments will (3 sigma) in pitch and yaw. A coronagraph be scheduled for 1 contact per day. The high requires a similar accuracy in pitch and yaw; rate science data downlink, at 160 Mbits/sec,

46 and the low rate housekeeping data will be 6.5 Instrument Module continuously maintained between the ground and the spacecraft. An instrument module must accommodate the SDO instruments and provide an accept- able launch environment. The instrument module consists of a mounting structure, instrumentation, cables, heaters and radia- tors. The mounting structure shall be de- signed to serve as an optical bench to allow for mounting and alignment of the sensitive SDO instruments. Figure 6.6 shows an il- lustration of the launch configuration of SDO, including the instrument module, AKM, and solar panels in a 9.5-foot Delta fairing.

Figure 6.5: SDO CD&H system.

6.4 Spacecraft Power

The instruments were assumed to require a combined power of 250 Watts, with 50 Watts for survival. (The estimates were based on 24% efficient Triple Junction GaAs cells, with 2 spacecraft wings, each with solar array area of 3.2 square meters, including losses from UV and energetic par- ticle radiation, thermal cycling, assembly Figure 6.6: SDO instrument module launch configu- losses, and losses from SA to battery, bat- ration tery to load, & SA to load.) The estimates of the spacecraft power requirements are listed 6.6 Ground System in Table 6.2 The SDO mission utilizes an inclined geo- Table 6.2. Spacecraft bus eclipse power synchronous orbit to take advantage of a (W) single dedicated ground station. This ground station would ideally be located in a location Custom RSDO with minimal effects of rain attenuation. X- ACS 84 53 band frequencies are becoming less avail- Power 21 52 able to missions in the space sciences; it is RF comm. 100 41 likely that the primary downlink of SDO C&DH 35 12 data will use Ka- or Ku-band frequencies. Propulsion 5 12 For space research 37 GHz is currently allo- Thermal 50 91 cated, with a possibility of allocation at 21 Harness 23 or 27 GHz. The Ku and Ka frequencies suf- Total 295 284 fer greater rain attenuation than X-band, but the antennae are much smaller and can be

47 built inexpensively, and backup stations can be considered to meet data completeness requirements.

Data latency (including delivery to final us- ers) requirements may drive ground system considerations; several of the LWS partners may require latency significantly less than an hour. Additionally, the high science data rate indicates that only the health and safety data would be able to be stored on board in the event of a contact interruption. The data gathering architecture will be selected to maximize long continuous streams of valid Figure 6.7: SDO Mission Operations and Ground System Network data. The high data completeness require- ment will also be a design driver for the The Mission Operations Center, Science ground system and the downlink margins. Operations Center and Ground System could be distributed or centralized. Centralization During the nominal mode of spacecraft op- requires less transport of data over the Inter- eration, the ground system can operate as a net and greater communication between the semi-autonomous system, allowing standard mission operations and science teams. How- operations to proceed on a 40 hour per week ever, the cost of data transport at the time of schedule. The geosynchronous orbit and on- the SDO mission is unknown, and the cost board safing systems make the use of a advantage gained from less transport re- semi-autonomous ground system a low risk quires further study. Moreover, the cost of and low cost way of meeting the mission temporary storage of SDO data at the requirements. ground system site (to compensate for tem- porary network and distribution failures) is 6.7 Mission and Science Opera- dependent on the future price of bulk stor- tions age.

Most of the SDO planned instruments are The ground station will be able to strip the full-disk instruments and the observations of science data packets from the downlink the mission are not “event-driven,” (e.g. re- stream and send them to the storage sites, sponses to flares or CMEs). There is consid- either at PI institutions or at a centralized erable value in collecting observations as data service. The packets will be assembled routinely, and under as stable observing into Level Zero data sets and may receive conditions and repeatable operation scenar- additional processing prior to being made ios, as possible. Thus, the nominal science available for distribution. mode of operation can be accomplished with daily command loads similar to those used by the TRACE mission. Instead of near- real-time commanding, daily command loads compiled by the PIs will be sent by the Mission Operations Team.

48 Table 7: Concurrent Observations

Facility Measurements Date STEREO Atmospheric Images 2006- Coronal Images 2011 Solar-B Atmospheric Images 2005- Magnetograms UV/EUV Spectra Solar Atmospheric Images 2010 Probe Coronal Images 2015 Magnetograms Figure 6.8: SDO Distributed Data System SORCE Total Irradiance 2002- Spectral Irradiance 2007 Data generated by LWS missions are to be GOES/ Total Irradiance 2010- free and publicly available for analysis. NPOESS Spectral Irradiance Proper support of an open data policy re- SOLIS Magnetograms 2001- quires the provision of the development of Spectra 2025 software and analysis tools and calibration ATST Magnetograms N/A algorithms by the instrument team. These Solar Atmospheric Images N/A services will be provided through the Princi- Sentinels Helioseismic Images pal Investigations. Solar Atmospheric Images 2009- Orbiter Coronal Images Magnetograms 7 Concurrent Observations UV/EUV Spectra FASR Coronal N/A Several space-based and ground-based in- Magnetograms struments will be operational at times during the SDO mission. Observations from some 7.1 STEREO of these instruments [particularly those on other LWS missions] will, at the very least, STEREO (Solar-TErrestrial RElations Ob- complement those from the SDO instru- servatory) will provide new perspectives on ments. In some cases (e.g. total solar irradi- the structure of the solar corona and CMEs ance) these observations may replace those by moving away from our customary Earth- that otherwise would need to be obtained bound vantage point and by using two with SDO. The following sub-sections de- spacecraft to provide information on three- scribe the instruments as they were specified dimensional structure. Both spacecraft carry at the time of this report. Future changes in a suite of instruments including: two co- these specifications may influence the final ronagraphs (covering the range 1.25 - 4 choices for SDO instruments. Table 7 lists RSun, and 2-15 RSun), an extreme ultraviolet the facilities along with their anticipated imager (full-disk, 1 arcsec pixels), a helio- dates of operation and key measurements. spheric imager (an externally occulted co- ronagraph that can image the heliosphere from 12 RSun to beyond Earth’s orbit), an interplanetary radio burst tracker, in situ particle and field detectors, and a plasma composition experiment.

49 modulates the total solar output and creates This mission is designed: to further our un- the driving force behind space weather. derstanding of the origins and consequences of CMEs, to determine the processes that The spacecraft will accommodate three ma- control CME evolution in the heliosphere, to jor instruments: a large solar optical tele- discover the mechanisms and sites of solar scope (SOT), an X-ray telescope (XRT), and energetic particle acceleration, to determine an EUV Imaging Spectrograph (EIS). All the 3D structure and dynamics of coronal three major instruments will give extremely and interplanetary plasmas and magnetic high resolution observations of a field of fields, and to probe the solar dynamo view on the Sun that is restricted to an active through its effects on the corona and the he- region spatial scale. The SOT includes liosphere. plane instruments that will provide longitu- dinal magnetograms with 1-5G sensitivity The STEREO Mission spacecraft are ex- and vector magnetograms with 30-50G sen- pected to be launched in December 2005. sitivity to transverse fields. The XRT pro- The two spacecraft will slowly drift apart in vides atmospheric images at wavelengths longitude with a total separation of from 2 to 6 nm. The EIS provides UV/EUV 45° after the first year and 90° after the sec- spectra with a resolving power of about ond year. The mission is expected to last 10,000 over wavelength ranges from 17-21 two years at minimum and, more likely, five nm and 25-29 nm (covering temperatures years. from 105 to 107 K but with only one strong emission line below 106 K). STEREO will overlap with SDO and pro- vide coronagraphic images like those needed The Solar-B spacecraft is scheduled for for SDO. STEREO’s lifetime, however, is launch in the fall of 2005. It will be placed shorter than SDO’s and during parts of the in a polar, sun-synchronous orbit about the STEREO mission the STEREO instruments Earth. This will keep the instruments in will be viewing solar regions far removed continuous , with no day/night cy- from the SDO observations. The STEREO cling for nine months each year. Solar-B coronagraphs will complement the meas- will address some of the same problems that urements made from SDO but cannot be SDO will address but with higher spatial used to replace them. resolution and a smaller field of view. The combination of vector magnetograms, at- 7.2 Solar-B mospheric images, and UV/EUV spectra should provide the measurements needed to Solar-B is a Japanese mission proposed as a answer many of the outstanding questions follow-on to the highly successful Ja- concerning the initiation of flares and pan/US/UK Yohkoh (Solar-A) collaboration. CMEs. The mission consists of a coordinated set of optical, EUV and X-ray instruments that 7.3 Solar Probe will investigate the interaction between the Sun's magnetic field and its corona. The re- The Solar Probe mission is an unprece- sult will be an improved understanding of dented exploration of the inner heliosphere, the mechanisms that give rise to solar mag- which will achieve unique science by flying netic variability and how this variability over the pole of the Sun and as close to the Sun's surface, through the solar corona, as is

50 technologically feasible today. It will first vations of the polar magnetic field and polar travel to for a , leave the sub-surface flow patterns are essential to ecliptic plane, fly over the Sun's poles to answering fundamental questions related to within 8 solar radii, and reach perihelion the solar dynamo and the origins of the solar over the equator at 4 solar radii. A unique cycle, which are central to the SDO primary aspect of the Solar probe orbit is that the scientific objectives, but which cannot be trajectory is orthogonal to the Sun-Earth line obtained with the SDO spacecraft. during perihelion passage so that there is continuous radio contact throughout the flyby. Two perihelion passes are planned, 7.4 SORCE the first near the 2010 solar maximum and the second near the 2015 solar minimum. SORCE (SOlar Radiation and Climate Ex- This orbit ensures that the mission will periment) is a program within NASA’s Of- probe both the high speed solar wind fice of Earth Science (OES) for measuring streams and the equatorial low-speed solar irradiance. The solar-pointed SORCE streams. spacecraft carries five instruments to meas- ure both total and spectral solar irradiances. The results from SOHO and have These are the Total Irradiance Monitor focused our understanding of the solar co- (TIM), Spectral Irradiance Monitor (SIM), rona to the point where in situ measurements two identical Solar Stellar Irradiance Com- are now necessary for further progress. Both parison Experiments (SOLSTICE) and the imaging and in situ measurements will pro- X-ray Photometer System (XPS). vide the first three-dimensional view of the corona, high spatial and temporal resolution The TIM measures total solar irradiance us- measurements of the plasma and magnetic ing electrical substitution radiometers. The fields, as well as helioseisology and mag- observations have an uncertainty goal of netic field observations of the solar pole. 0.01% and a long-term repeatability of 0.001% per year. The Solar Probe Nadir-Viewing Imagers will provide the only full view of the Sun's The SIM measures solar UV, visible and IR Poles, imaging both the North and South spectral irradiance from 200 nm to 2000 nm Pole within one day. The Solar Probe Mag- with spectral resolution ranging from a few netograph/Helioseismograph will provide nm at UV wavelengths to tens of nm at IR out-of-the-ecliptic observations of the polar wavelengths. It uses a prism for wavelength magnetic field and polar sub-surface flow dispersion and a bolometer and diodes for patterns essential to answering fundamental signal detection. The SIM spectral irradi- questions related to the solar dynamo and ances have uncertainties of 0.03% and long- the origins of the solar cycle. term repeatabilities of 0.01% per year.

The observations of the solar corona and The two SOLSTICE instruments measure solar wind that Solar Probe will provide are the solar UV spectral irradiance from 120 to critical to understanding fundamental proc- 300 nm with spectral resolution of about 0.1 esses that can be obtained in no other way, nm, using grating spectrometers that have and will provide a set of measurements that the capability also to observe bright blue complement, but are distinct from, the SDO stars for calibration tracking. The SOL- observations. Finally, the Solar Probe obser- STICE UV spectral irradiance uncertainties

51 are in the range 3% to 6% and repeatabilities ESS, expected to commence around 2010, are 0.5%. depending on existing resources.

The XPS is a bank of broadband X-ray Measurements of solar EUV radiation in photometers in the range 1 – 31 nm with five broad bands in the range 10 to 130 nm spectral bands 5 to 10 nm, uncertainty of are planned to be made from future GOES 12% and repeatability of 3% per year. platforms commencing in late 2002, de- pending on the health and status of existing SORCE will be launched in mid 2002, with GOES spacecraft. The EUV broadband expected mission duration of 5 years. It will fluxes are recorded every 10 seconds, with overlap with ACRIMSAT, which has pro- an uncertainty of 10% and a repeatability of vided total solar irradiance data since 2000, 5% over 7 years. The instrument uses dif- thereby extending the continuous record of fraction gratings to disperse the light, thin- total solar irradiance that commenced in late film filters to further remove unwanted 1978. NASA OES plans to continue the so- wavelengths, diode detectors to col- lar irradiance measurements with a follow- lect the light and tantalum shielding to on solar irradiance mission in the time frame eliminate effects from radiation. Two GOES of 2006-2011 that measures total solar irra- spacecraft are planned to be operational at diance (e.g., TIM) and spectral irradiance any one time providing redundancy and from 200 to 2000 nm (e.g., SIM). cross-calibration for the EUV sensor. With the launch of each new GOES EUV Sensor 7.5 GOES/NPOESS every (2-5 years), a new calibration will be applied to the data set. GOES (Geosynchronous Operational Envi- ronmental Satellites) and NPOESS (National In addition to these major resources, at least Polar-orbiting Operational Environmental two European programs are planning to Satellite System) are satellite systems de- measure solar irradiance during the next ployed by NOAA to monitor the environ- decade. Three solar instruments (ACES, ment. SOVIM and SOLSPEC) will measure the total solar irradiance and the solar spectrum Solar irradiance will be monitored by NPO- from the EUV to the IR. The solar package ESS. Measurements of total solar irradiance is presently scheduled as part of the Colum- and of the solar spectral irradiance in the bus module of the International Space Sta- wavelength range from 200 to 2000 nm with tion, to be launched during 2004-2005 with accuracies and precisions similar to those of a priority of solar minimum observations. the TIM and SIM instruments on SORCE , a small spacecraft carrying in- are specified. The solar irradiance measure- struments to measure the total solar irradi- ments are designated for flight on one of the ance and solar diameter, is under develop- three NPOESS spacecraft. Since the priority ment in France, for launch by CNES around for the solar measurements is relatively low 2006 with a 2-3 year mission duration. among other NPOESS measurements, fail- ure of the spacecraft would not trigger an These measurements may facilitate the cali- immediate replacement of the solar instru- bration of the SDO EUV Spectral Irradiance ments. This means that gaps may occur in measurements but will not have the spectral the total solar irradiance measured by NPO- resolution or coverage to replace them.

52 7.6 SOLIS els. Quick-look SOLIS data will be avail- able on the Web within 30 minutes and more SOLIS (Synoptic Optical Long-term Inves- accurately reduced data within 24 hours. tigations of the Sun) is a ground-based proj- Special campaigns and user-proposed pro- ect of NSO to provide regular observations grams can be interleaved with the regular of the Sun for at least 25 years. It will re- synoptic observations. The synoptic data place many of NSO's current synoptic fa- will be openly available. cilities. The primary science objectives are to increase understanding of solar activity SOLIS and SDO will complement each and its effects on earth by means of obser- other in a number of ways. Lightweight vations of the Sun's full vector magnetic space magnetographs will be filter based and field and the dynamics and evolution of so- it will be very useful to compare results lar changes related to the magnetic field. from such an instrument with the higher Aside from basic research on the solar ac- spectral resolution SOLIS measurements to tivity cycle, the observations will also be seek out systematic errors in both types of used as inputs to test models that are alleged observations. Similarly, the higher spatial to be able to forecast activity. resolution space observations will permit a study of how ground-based observations are The first SOLIS facility is nearing comple- degraded by terrestrial seeing. In case a tion and the recent NAS/NRC report “As- vector magnetograph is not flown on SDO, tronomy and Astrophysics in the New Mil- the SOLIS vector observations will provide lennium” recommends building two addi- an obvious enhancement. The same holds tional systems to be located at longitudes true for the chromospheric magnetograms. different from the US. This would increase SOLIS will provide regular monochromatic the average 24-hour duty cycle from about measurements of intensity and dynamics in 30% to about 80%. The US system includes the cool solar atmosphere that will be valu- three instruments: a vector spectromagneto- able for use with the SDO high temperature graph (VSM), a full-disk monochromatic measurements. Perhaps the best comple- imager, and an integrated sunlight spec- mentarity would be one that we cannot pre- trometer for sun-as-a-star spectroscopy. dict, namely, some SDO discovery that stimulates follow-up observations with SO- The VSM is the most unique instrument and LIS, or vice versa. would be the common network instrument. The VSM is a 50-cm telescope and a high- 7.7 ATST resolution spectrograph. It can provide full- disk vector magnetograms in about 15 min- ATST (Advanced Technology Solar Tele- utes with a polarimetric noise level of about scope) is a proposed ground-based telescope 3 x10-4 using one arcsecond pixels. It will facility designed for high-resolution studies also provide high-sensitivity photospheric of the Sun. The proposal is for a 4-meter and chromospheric line-of-sight component telescope operating in the visible and infra- magnetograms as well as He I dynamics im- red (0.3 to 35 microns) with very high reso- ages. lution (0.1arcsec or better) and a large pho- ton flux for sensitive polarimetry. The facil- The monochromatic imager provides inten- ity will have the spatial resolution and sen- sity and Doppler images in a number of so- sitivity to study the ubiquitous weak mag- lar spectrum lines using one arcsecond pix- netic field elements in the photosphere and

53 measure magnetic fields in the corona. It to follow the development of structures will have the capability to examine waves in within the Sun. magnetic flux tubes to test models of chro- mospheric and coronal heating. It will have The fleet will consist of four Inner Helio- a 5 arcmin field of view that will allow spheric Sentinels in heliocentric orbits studies of active region evolution and the ranging between 0.5 and 0.95 AU, a FarSide initiation of flares and CMEs. Sentinel in a 1 AU orbit opposite Earth, on the far side of the Sun, and a single L1 Sen- ATST will address some of the same prob- tinel to provide solar wind input information lems that SDO will address (e.g. small-scale to the geospace components. These elements magnetic elements) but with higher spatial will work together to track solar distur- resolution, smaller field of view, and less bances as they evolve and transit the inner complete coverage. The proposal suggests heliosphere. The inner heliospheric sentinels that operations begin in about 2008. are spinning satellites. The FarSide Sentinel is three-axis stabilized.

7.8 Solar Sentinels Solar Sentinels will supplement SDO obser- vations to provide continuous and whole The Solar Sentinels will consist of a fleet of surface imaging of the photosphere and solar spacecraft distributed throughout the helio- corona, hence allow the study of evolution sphere. They will help to improve the accu- of solar active regions. The observations racy of models of CMEs and other solar complement those from SDO but with little wind transients. They will resolve geoeffec- or no redundancy. tive solar wind structures and map them back to solar features. They will search for the locations and mechanisms of energetic 7.9 Solar Orbiter particle acceleration, and provide tomo- graphic images of the Sun. When fully de- The Solar Orbiter was selected by ESA at ployed, the Sentinels will increase the lead- the end of 2000 as a “flexi”- mission, for time and accuracy of geospace forecasts. launch in the 2008-2013 time frame. The key mission objectives of the Solar Orbiter The Inner Heliospheric Sentinels will make are: (a) to study the Sun from close-up (45 in situ observations of the heliospheric vec- solar radii, or 0.21 AU), and (b) to provide tor magnetic field, the solar wind plasma images of the Sun's polar regions from he- properties, and the spectrum of high-energy liographic latitudes as high as 22 degrees particles. They will make remote measure- during the nominal mission, and over 30 de- ments of the propagation of interplanetary grees during the extended mission. shocks by tracking radio bursts. A Farside Sentinel will also provide EUV images of Solar Orbiter’s unique heliosynchronous, the solar corona along with photospheric near-Sun trajectory will allow, in conjunc- magnetograms. Radio occultations will be tion with concurrent high-resolution remote employed along with these images and those sensing observations, in situ investigations from SDO to identify the birthplace of tran- of the energetic particle environment in sients. Helioseismic measurements will be close proximity to different source regions, made in conjunction with those from SDO such as active regions, flare locations, CMEs and associated shocks. Solar Orbiter’s

54 unique high-latitude trajectory will allow it and chromospheric fields, and in this regard to determine the longitudinal extent of promises to be an important supporting in- CMEs and provide, in conjunction with strument for SDO studies of the magnetic SDO and ground-based observatories, full field in the corona. FASR has been rated coverage of the entire Sun over 360º in lon- highly by the NRC panel on the future of gitude. ground-based solar astronomy, and recom- mended as a moderate-sized initiative by the The potential payload includes two instru- decadal NRC Astronomy and Astrophysics ment packages: the Heliospheric in situ in- Survey Committee. The telescope will be a strument package and the Solar Remote solar-dedicated instrument providing excel- sensing instrument package. The Helio- lent images of the full Sun with arcsecond spheric in situ instruments include: a solar spatial resolution at a wide range of fre- wind analyzer, radio and plasma wave ana- quencies nearly simultaneously, with both lyzers, a , energetic particle targeted research and synoptic capabilities. detectors, an interplanetary dust detector, a neutral particle detector, and a solar neutron Radio observations are able to measure detector. The Solar remote sensing instru- magnetic fields due to the gyroresonance ments include: an EUV full-Sun and high effect: electrons spiraling in the coronal resolution imager, a high-resolution EUV magnetic fields provide opacity at radio spectrometer, a high-resolution visible-light wavelengths at low harmonics of the elec- telescope and magnetograph, an EUV and tron gyrofrequency, 2.8 ´ 10-3 B GHz where visible-light coronagraph, and a radiometer. B is measured in Gauss. A given observing frequency and sense of circular polarization We hope that the interested parties within is sensitive to a single value of magnetic ESA will do everything possible to ensure a field strength; by changing frequencies launch in a timely manner, i.e. in 2009 or FASR will be sensitive to magnetic fields in soon after. From recent solar and helio- the range 100 - 2000 G. Surfaces of constant spheric physics missions, e.g. SOHO, magnetic field strength show up in radio Yohkoh, TRACE, and Ulysses, we have maps at the appropriate frequency as bright learned that by far the best scientific return regions: they have coronal brightness tem- from missions is through efficient coordina- peratures because gyroresonance opacity tion. A launch of Solar Orbiter in 2009, or makes them optically thick, whereas the sur- soon after, would provide a significant rounding atmosphere with lower magnetic overlap with SDO. The combination of field strength is optically thin and has much Earth-orbit high-resolution observations lower brightness temperature. The depend- from SDO with the close-encounter and po- lar observations from Solar Orbiter would ence of opacity on viewing angle introduces allow an invaluable, thorough analysis of some complications but the theory is well many aspects of solar activity and its influ- understood, and radio data have proven to be ence on the Earth. excellent diagnostics for testing extrapola- tions of photospheric fields into the corona. 7.10 FASR 8 Acknowledgements FASR (Frequency Agile Solar Radiotele- scope) will provide radio observations of The Science Definition Team would like to coronal magnetic fields that complement acknowledge the following people for their optical/IR measurements of photospheric

55 assistance in the production of this report: John Leon, Paul Caruso, Sahag Dardarian and Ron Miller from the GSFC Project team for their support during the discussions of mission requirements; Art Poland, Dick Fisher, Karel Schrijver and the LWS Sci- ence Architecture Team for LWS program science support; Members of the SDO Mis- sion Science Preformulation Team: Leon Golub, Russ Howard, Steve Kahler, Dana Longcope and Vic Pizzo; Military Space Weather advisers during preformulation: Lt. Col. Michael Bonnadonna, Maj. Peter En- gelmann, Capt. Riley D. Jay, Maj. Phyllis Kampmeyer and Lt. Col. Erwin Williams; Joe Gurman and Terry Kucera at GSFC for mission development; Jennifer Rumburg for information systems; Scientists who assisted in the instrument study and discussions: Rock Bush, Darrell Judge, Don McMullin, Jesper Schou, Jake Wolfson and Tom Woods.

56