Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Principal Science Opportunities and Initiatives for Ground-Based Solar Research Advances in experimental and observational techniques now make it possible to observe aspects of the Sun that were previously unknown or unappreciated. The observations reveal a star of complex and mysterious behavior. Neutrino astronomy; helioseismology; high-resolution observa- tions of the solar surface; radio, infrared, UV, X-ray, and gamma-ray observations of the outer atmosphere; vector magnetic field observations; and spacecraft observations of the secular changes in the solar luminosity have all uncovered new and puzzling aspects of the Sun. These fundamen- tal investigations have been possible only because of the proximity of the Sun. One may infer that other stars are equally mysterious, but they cannot be resolved in the telescope and are too far away for the necessary close scrutiny. In this chapter the committee explores in greater detail the principal needs and most promising opportunities for investigation over the coming 5 years in the four research areas at the forefront of solar physics today: (1) probing the solar interior, (2) the physics at small spatial scales, (3) the mechanisms underlying the solar cycle, and (4) the physics of transients. The committee interviewed leading solar physicists from all major solar physics research centers in the United States and solicited oral and written comments from the solar community at large. 16
17 PROBING THE SOLAR INTERIOR The Basic Issues Information from the interior of the Sun is needed to understand fluctuations in the Sun's radiative and nonradiative outputs, to verify the theory of stellar structure and evolution, to help develop an understanding of fluid motions in realms beyond laboratory and theoretical modeling, and to advance several areas of basic physics. Recent work suggests that significant revisions are required in our current concepts of all these topics and that the ramifications may extend far beyond the traditional range of solar physics. One of the triumphs and major foundations of astrophysics is the theory of stellar structure and evolution. Much of what we understand about the universe derives from this theory. It is now possible to critically test the predictions of the theory for the case of the Sun, and the results are disturbing. The flux of neutrinos produced in the solar core has been measured since 1968 in a celebrated experiment located deep in the Homestake mine in South Dakota. Only one-third the flux of neutrinos predicted by the best models of the solar interior has been measured. A new experiment located at Kamioka, Japan, was started in 1987; the first results confirm that the neutrino flux is less than that predicted by solar models. This "neutrino problem" is larger than can be explained by current understanding and uncertainties of the relevant physics. Another prediction from the theory of stellar structure concerns the frequencies of the normal modes of oscillation of the Sun. Helioseismolog- ical observations have measured these frequencies with a precision of a few parts per hundred thousand. There is a systematic discrepancy between the observations and the predictions of a few parts per thousand. Again, this discrepancy is larger than can be explained by current understanding of the relevant physics. The theory of stellar evolution predicts that the Sun should have brightened by about 30 percent since the formation of the solar system. Geological and climatological evidence suggests that the change in solar luminosity has been much smaller. One proposed solution to this problem is to mix the solar interior to provide fresh fuel to the energy-generating core. Mixed models seem to be ruled out by current helioseismology results. Evolutionary theory also suggests that the interior of the Sun should be rotating much more rapidly than the surface layers, which have been braked by angular momentum transfer to the solar wind. Instead, helioseismology indicates that the interior is rotating very much as the surface rotates. Theoretical understanding of circulation and convection inside the
18 Sun is not well advanced because of the intrinsic difficulty of the relevant physics, an inability to construct and run realistic numerical models, the large extrapolations required from laboratory experience, and the relative lack of observations of the solar interior to provide guidance. Existing models of motions within the convection zone have not been confirmed by observations. Predictions of a polar vortex, giant circulation cells, and strong variations in rotation rates with depth and latitude in the convection zone have not been supported by observation. Evidence from a variety of observations suggests that nearly all stars with a mass of less than about 1.5 times the solar mass (and this means most stars) exhibit activity of the type that we observe in the Sun. We do not yet have a good understanding of how magnetic activity is produced even within the best observed star- the Sun. Much has been learned from observations of other stars that have a range of physical parameters different from those of the Sun. Probing the interior of the Sun can provide additional information about how stellar and solar activity is generated. Initial results from helioseismology indicate that the subsurface structure of sunspots and active regions does not agree with that described by current models. New models of the solar magnetic dynamo, which is thought to generate the solar activity cycle, are under development based on helioseismology. The discrepancies between current models and current observations listed above have challenged many researchers to suggest innovative solu- tions. Some of these suggestions extend into the realm of exotic physics. A typical example is the hypothesis that there may exist weakly interacting massive particles (WIMPS or cosmions) within the solar interior (and else- where). Such cosmions could reduce the central temperature of the Sun and thereby explain the neutrino deficit. It is worth noting that a model of the Sun that includes cosmions predicts p-mode oscillation frequencies that are significantly closer to observations than are those predicted by stan- dard solar models. It has also been suggested that neutrinos have a small rest mass, and even a magnetic moment, that could explain aspects of the neutrino problem. Laboratory results on this important physics question are conflicting, but future solar observations should help to verify or deny these suggestions. Initiatives and Impacts Researchers in the United States have led or participated in most investigations involving the solar interior. The United States has been particularly strong in observational work and can maintain a leading role in some areas and a presence in others by balancing support for continuing facilities and programs, initiating selected new programs, and collaborating with international partners where appropriate.
19 Theory and Modeling The study of the solar interior depends intimately on predictions from theory and modeling. It is essential that support for this activity be accorded as much priority as observational programs. The United States can continue among the world leaders in this field by initiating and supporting collaborative as well as domestic work. A good example is a 6-month workshop planned for 1990 at the Institute for Theoretical Physics of the University of California, Santa Barbara. Neutrino Obse~vaiions A survey of recent publications and plans for future projects clearly shows that the United States is heavily involved in neutrino observation, although not always in a leadership role. The committee urges that the United States maintain its presence in the field by continuing a few key experiments and supporting U.S. participation in international projects. Leading opportunities for initiatives include the following: 1. Continue operation of the 37C1 experiment through the next solar maximum expected in 1991, and continue support of the Kamiokande II experiment, whose results are an important consistency check for the 37C1 experiment. This will allow tests of suggested correlations of neutrino flux with the solar activity cycle and, more speculatively, with the Earth's heliocentric latitude. A confirmation of modulation of the neutrino flux will have a profound impact on solar physics, astrophysics, and particle physics. 2. Support U.S. participation in additional new international mea- surements of neutrinos from the Sun. Leo experiments may be considered as examples. The first is the proposed 2H experiment (Sudbury Solar Neu- trino Observatory), a Canadian, U.S., and U.K experimental collaboration, which will measure a variety of FIB neutrino properties, including their spectrum; the second is a 40A experiment, led by the Italians, which will provide an independent measurement of the SIB neutrino properties. In addition to determining the production rates and spectra of the neutrinos, these experiments will address the question of the mass of the neutrino and the hypothesis that the neutrino problem is due to a change of one type of neutrino to other types in transit to the earth. 3. Support U.S. participation in international measurements of neu- trinos from the main nuclear reaction that produces solar energy, although given the already existing strong international support for these projects, support might be more modest than the support for the preceding ex- periments. This next generation of experiments will enlarge the scope of neutrino measurements beyond that of the current experiments, which sample neutrinos from a minor nuclear reaction in one region of the solar core. Ibro experiments using 7iGa are in preparation to do this. One is
20 primarily a European experiment, and the other is a Soviet experiment with limited U.S. participation. These experiments will detect neutrinos from the most common reaction that produces energy in the solar core. Results will indicate whether the neutrino problem originates in physics or astronomy. 4. Complete an experiment to deduce the average neutrino flux over the last several million years. Sometime in 1989, results are expected from 981t extracted from about 20 boxcars of molybdenum ore mined from the Henderson mine in Colorado. I-his isotope is produced by absorption of neutrinos that have penetrated the 1500-m depth of the mine. Since the half-life of the isotope is a few million years, this difficult experiment may be able to measure the constancy of neutrino flux over the past several million years. If evidence for a changing flux is found, orthodox views of stellar evolution will need to be changed. Helioseismology Projects The United States is the world leader in helioseismology observations utilizing solar imagery. Europe leads in helioseismology of the Sun ob- served as a star. Both approaches enjoy unusually strong and stimulating international contributions and cooperation by observers and theoreticians. As a result, the field of helioseismology has expanded rapidly since its beginning in 1975. The literature comprised about 700 papers in mid-1987 and has doubled every 3 years. Work in this field (see, for example, Figure 3.1) has already answered some long-standing questions about the solar interior but has raised new questions of potentially wide-ranging signif- icance throughout astrophysics and physics. fib maintain leadership and momentum in this field, the United States should pursue a number of initiatives: 1. Support exploration of new observational methods and techniques. Groups at the California Institute of Technology; Stanford University; the Universities of Arizona, Delaware, Hawaii, and Southern California; the National Solar Observatory (NSO); the High Altitude Observatory (HAO); National Aeronautics and Space Administration (NASA)/Goddard; and elsewhere are advancing state-of-the-art observational helioseismology. A good example is the tomography of sunspot structure developed recently by researchers from NASA and the University of Hawaii using NSO/Kitt Peak facilities. The result of supporting innovative observational helioseismology will be the development of new and improved methods for probing the solar interior. 2. Support the Global Oscillations Network Group (GONG) project. This is a community project initiated by the NSO to provide continuous solar oscillation data for a period of 3 years. It was a response to an
21 ~7,: , 35 days 30 days \ ) / - - / ~ ~26 days '`l ~ \ \ MY ( / 25 days FIGURE 3.1 Contour plot: cross section through the Sun, with contoum of constant rotation period as a function of latitude and depth in the solar interior. The dashed line marks the base of the solar convection zone. The picture is based on measurements of oscillations of the Sun's surface, which are a manifestation of sound waves traveling through the solar interior. The rotation rate is determined by comparing waves that travel east-to-west and west-tc'+ast at different depths inside the Sun. Because of limitations in the current measurements, the results here are only accurate for radii larger than 0.4 solar radii. The results indicate that the Sun's surface rotation persists throughout the outer 30 percent of the Sun, where it is probably driven by large-scale convection. Below the convection zone the Sun appears to rotate nearly rigidly (with the possible exception of the deep intenor) at a period of about 27 days. This picture is based on helioseismology data obtained by K. Libbrecht at Big Bear Solar Observatory and on inversions of the data by J. Christensen-Dalsgaard and J. Schou, as well as by P. Goode and W. Dziembowski. (Reproduced flay permission of the California Institute of Technology.)
22 obvious scientific need for such data and to an invitation by NSF for innovative projects. Motivation for the project is reduction of the noise and confusion introduced by nightly gaps in solar oscillation data obtained from single observatories. Continuous observation by means of a network of six sophisticated instruments around the world promises to reduce this problem by at least an order of magnitude. The impact of this project will be great improvements in the precision of p-mode oscillation frequencies and amplitudes for degrees up to about 300. This will permit definitive determinations of the temperature stratification and large-scale motions of most of the solar interior. It is important that funding also be provided to assist the helioseismology community to analyze and interpret the data from the GONG project. 3. Support the U.S. helioseismology experiment on the European Space Agency's (ESA) SOHO spacecraft. This experiment was selected by NASA and ESA as one of the major tasks for the SOHO spacecraft expected to be launched in 1995. Aside from the important advantage of continuous sunlight afforded by an orbit around the Ll Lagrangian point, the lack of atmospheric distortion will present unique opportunities to study oscillations of both high degrees and long periods. The impact of these observations will be a definitive determination of the stratification and motions of the upper layers of the convection zone, where our current understanding of the physics is quite uncertain. Investigation of the Interiors of Other Solarlike Stars The study of the solar interior gives us information about one star. It would be naive to think that we can safely-extrapolate that information to other stars without some verification. Similarly,-comparison of some of the characteristics of the solarlike stars, such as age, chemical composition, and rotational velocity, would provide a considerably sharper test of the theory of both solar and stellar structure and evolution. For example, the study of the depletion of light elements in a wide range of stars is a sensitive indicator of the maximum temperature to which convecting material is exposed in the outer layers of a star. In the Sun and several other stars, the outer layers appear to have been exposed to higher temperatures than can readily be explained by standard theory. While neutrino radiometry of other stars is currently beyond the capabilities of foreseeable technology, the prospects are good for seismic probing of solarlike stars. Already the first steps have been taken on both observational and theoretical fronts and have shown considerable promise. On the observational side, what is needed is a highly stable echelle spectrograph, fed by a several-meter- aperture telescope, and large blocks of contiguous night scheduling. A recent experiment involved 2 weeks of observing time with the Soviet 6-m
23 telescope. A dedicated facility would be optimum because of the peculiar requirements of large amounts of observing time to do seismology of other stars, but a facility shared with other observing programs is also a feasible solution. Another way of approaching this problem is to attempt precise photom- etry of members of stellar clusters. Although such work is best done from space, it may be possible to obtain sufficient accuracy with ground-based equipment. Such experiments should be supported. The impact of work in this area will be to allow confident application of what we learn about the solar interior to other stars. The theory of stellar structure and evolution will be tested over a broader range of parameters than can be done using the Sun alone. There will also be feedback of information about other stars into the total picture of the solar interior. THE PHYSICS AT SMALL SPATIAL SCALES The Basic Issues It is now well known that magnetic fields play a central role in the dynamics of the solar surface layers (for example, by ordering local trans- port coefficients such as thermal conductivity in an anisotropic fashion, by blocking convective transport, and by carrying the "mechanical" energy and momentum flux required for coronal plasma heating and acceleration of the solar wind); hence solar magnetic activity largely defines the interaction between the Sun's interior and atmosphere, and between the Sun's atmo- sphere and the heliosphere and terrestrial magnetosphere. The detailed physics by which the magnetic activity both arises in the solar interior and ultimately couples to the outer solar atmosphere and heliosphere remains a matter of active research. It is nevertheless clear that the answers lie in an understanding of the interaction between magnetic fields and tur- bulent conducting fluids and of the equilibrium and stability properties of magnetized plasmas, and in the realm of collective plasma behavior. These issues of physics are intimately connected and are, furthermore, of great interest both to space physicists and terrestrially bound plasma physicists. Thus issues of plasma confinement (and their attendant problems of magnetohydrodynamic equilibrium and stability) and plasma heating (by wave and/or particle beam and plasma interaction) and transport are central to fusion plasma efforts. It should therefore not be surprising that, for example, current models for solar plasma heating borrow heavily from recent advances in the laboratory domain, and that, conversely, some of the early work on plasma confinement schemes grew out of work originally carried out in the astrophysical domain. Because the phenomenolo~ of the solar surface layers is so rich, one
24 cannot hope to summarize fairly the entire range of current theoretical and observational work; hence the following represents an outline of what the committee perceives as the most exciting current research directions, with an emphasis on those that exemplify various aspects of the interaction between solar magnetohydrodynamics and plasma and space physics, plasma astrophysics in general, and the terrestrial fluid dynamics and laboratory plasma domains. Magnetic Field Generation and Intermittency Solar magnetic fields are striking in two very distinct respects: they persist in spite of the observed rapid diffusion of surface magnetic fields, and they are whenever observed spatially highly concentrated and in- homogeneous. It is commonly believed that these circumstances can be understood by appealing to the interaction between magnetic fields and turbulent shear flows. Thus much of the observed phenomenology associ- ated with the solar magnetic cycle can be reproduced by kinematic magnetic dynamo models, and spatial intermittence is thought to result from "sweep- ing" initially homogeneous magnetic fields into regions of stagnant flow by organized cellular flows (viz, classic Benard convection cells). Unfortunately, solar magnetic fields are relatively strong, so that it is dubious whether kinematic theories are an appropriate description of the physics underlying the solar dynamo; furthermore, the solar convection zone is far from laminar in behavior (the Rayleigh number is far above critical, and the Reynolds number exceeds unity by many orders of magnitude), so that it is unclear whether results from laminar theory can be immediately adopted. It is therefore not surprising that these issues are currently being attacked via sophisticated numerical simulation schemes, which include the effects of magnetoconvection and buoyancy. What is particularly fascinating about this work for (solar) fluid dynamicists and plasma physicists is that the Sun at present provides the only "laboratory" for testing theories of flux concentration and enhanced (turbulent) diffusion of magnetic fields. Equilibnum and Stability Theory Because magnetized plasma structures in the outer solar atmosphere- ranging from cool prominences to million-degree coronal "loops"~an show both periods of great quiescence and intervals of highly intermittent activity, there has been a concerted effort to understand the equilibrium configurations and stability properties of magnetic-pressure-dominated plas- mas. There are of course obvious parallels to related work in the plasma- fusion community, and indeed the early solar studies anticipated related laboratory plasma studies. Stability calculations are being actively pursued today in the solar context, with substantial input from the now classic work
25 from the laboratory domain. This includes use of the Bernstein "energy principle" and the concept of "line-tying" as applied to magnetic field lines entering the high-density photosphere from the overlying tenuous chromo- sphere and corona; application of helicity conservation in construction of equilibria; studies of the existence of equilibria under specified (realistic) boundary conditions; and study of field line stochasticity. Rapid Magnetic Field Reconnection The role played by collective effects in the solar atmosphere was first appreciated in the impulsive phenomenon known as the solar flare, com- monly believed to occur when oppositely directed magnetic fields in the solar corona "reconnect," thereby releasing energy in the form of heat, particle acceleration, and induced rapid flows. It has long been evident that the observed short time scale of impulsive energy release demands a breakdown of the classical (high electrical conductivity) magnetohydrody- namic picture normally used to describe the solar outer atmosphere. As a result, a blossoming of interest in magnetic reconnection (driven also by observations of related impulsive phenomena in the terrestrial magnetotail) has occurred: steady-state fluid theory has been placed on a robust, for- mal footing; calculations have been extended to the collisionless domain; and extensive efforts at numerical simulation and laboratory modeling of reconnection are currently being conducted. From the solar perspective, one needs to understand the geometric configuration of the reconnection site; to understand the conditions under which sudden energy release occurs; and to be able to estimate the energy released into fast particles, direct plasma heating, and flow acceleration. These questions are indeed common to the various disciplines in which field reconnection plays a role; the contribution of solar studies will be to extend significantly the parameter regimes in which reconnection can be studied. The Physics of Thermal Heat; Conduction The rapid rise of the gas temperature above the solar photosphere to several million degrees within a few thousand kilometers has raised many questions, not the least of which is how one is to calculate the thermal transport coefficients properly. The classical Spitzer-Harm thermal conduction is inherently a linearization, entailing asymptotic expansion in the ratio of the thermal mean free path to the temperature gradient scale length. This has been shown to fail in laboratory studies of heat transport in hot plasmas for very small values of this ratio. In addition, inertial confinement studies suggest that microturbulent effects may also come into play. These terrestrial laboratory results are only now finding
26 their way into the solar plasma physics domain, and it seems inevitable that rather significant changes in our understanding of the interchange of energy between the solar corona and the underlying photospheric gas will result. The impact of these applications is in our understanding of the fol- lowing: previous calculations of the (transition region) thermal heat flux may be in error; the large mean free path of coronal electrons may signif- icantly alter the ionization balance of cooler, lower-lying layers (and thus upset standard plasma diagnostic techniques); and the nonlocal character of heat transport by long-ranging suprathermal electrons may vitiate previous hydrodynamic studies based on local theory. Plasma Diagnostics, Heating, and Monons The current' state of the art in remote-sensing plasma diagnostics finds solar plasma physics at the forefront. From the astronomical perspective, this is by design, for' the Sun provides physical conditions that are not unlike those encountered in much of the rest of the universe Tut at inaccessible distances) and reduces demands on instrumentation (because its proximity leads both to the availability of copious numbers of photons throughout the electromagnetic spectrum and to some useful degree of spatial resolution of the activity itself). Thus the Sun has been studied not only for its own sake but also as a test case for exploring new instrumentation and diagnostic concepts in a more familiar and accessible context. day's frontiers of solar plasma diagnostics lie in the direction of nonequilibrium studies and in the exploitation of high-spectral-resolution observations, combined with high spatial and temporal resolution (particularly in wavelength domains heretofore relatively poorly explored with spectroscopic tools). This frontier area includes efforts to diagnose departures from ionizational equilibrium (using, for example, satellite lines of strong resonance lines) by observing detailed line profiles formed at transition-region and coronal temperatures (which allow one to test for Doppler broadening from the systematic motion of hot plasma associated either with' flows or with quasi-periodic motions resulting from propagating or standing waves3. The latter studies have particular relevance to tests of theories for atmospheric plasma heating, to studies of mass exchange between the solar photosphere and the hotter overlying layers (as can occur during the course of solar flares), and to the classic problem of solar wind heating and accel- eration in the immediate solar vicinity. High-resolution spectroscopy, when combined with high spatial resolution and the ability to measure polariza- tion states (i.e., the Stokes parameters), also allows direct measurement of vector magnetic fields in the solar atmosphere and hence determination of the magnetic field topology in the solar corona. At very high photon
27 energies, high-resolution hard X-ray and gamma-ray spectroscopy allows one to test detailed particle acceleration models (through interaction be- tween these fast particles and ambient matter), whereas in the infrared, high-resolution (spectral and spatial) spectroscopy takes advantage of the fact that atomic line Zeeman splitting is proportional to the square of the line center wavelength to enable exploration of the magnetic field structure in the lower photosphere and chromosphere. Initiatives and Impacts It is evident from the foregoing discussion that studies of the physics of the Sun's outer layers will very likely involve substantially greater inter- action with the laboratory and magnetospheric plasma physics communities and increasingly greater contact with observers and plasma theorists dealing with astrophysical plasmas in general. The rapid development of instrumen- tation capable of extremes in high spatial, temporal, and spectral resolution will challenge the modeling abilities of theorists; and, as has been the case in the magnetospheric domain, large-scale numerical simulations will play an increasingly important role. Because these research activities place solar physicists at the forefront of both experimental techniques and com- putational needs, the committee considers it important to ensure that the opportunities available in solar physics research are realized. Thus, whereas the problem areas discussed above define the direction of research into the physics of the solar surface in the immediately forsee- able future, it is of considerable importance to note that the success of these studies is predicated on the existence of the instrumentation to carry out these studies. Because the most promising directions in experimental research of the solar surface involve state-of-the-art technology and hence require both a cadre of highly qualified scientists and technologists and a significant investment in high-technology laboratory facilities (including computational resources), it is crucial to define, implement, and maintain a well-defined, long-range observational program. ~ simply maintain exist- ing equipment without an active program for developing and implementing new instrumentation is a strategy ultimately certain to cripple the science. High-Spatial-Resoluiion Visible and Infrared Obse~vaiions One of the great puzzles of solar physics is the observed clumping of magnetic field structures. An essential element in studying these structures is of course their observation. This task requires telescopes with high spatial resolution (well below 1.0 arcsecond), extremely well characterized polarization effects (to a level less than 1 percent), and high-photon- collection capability. Recent experiments at NSO/Kitt Peak have also shown the substantial benefits to be gained from infrared observations; at these
28 long wavelengths, atomic line Zeeman splitting is sufficiently large that the pi and sigma components can be readily separated, with relatively little modeling effort needed to produce good magnetic field measurements. The missing ingredient is high spatial resolution: the NSO/Kitt Peak facilities have limited spatial resolution, and the Sacramento Peak Vacuum Tower telescope can produce subarcsecond resolution in the near infrared but cannot be used beyond a wavelength of 2.4 microns. Efforts to improve this situation (such as the HAO/NSO Advanced Stokes Polarimeter project) must be supported in order to advance in this area. However, the key next step is to plan now for future observing capa- bilities that can provide a significant and necessary advance over what is currently available. With this goal in mind, the HAO scientists, acting as representatives of the interests of U.S. solar astronomy and recently joined by NSO scientists, have been involved in discussions with scientists from nine other countries on building a large-aperture ground-based telescope, whose goal is to obtain both high throughput and diffraction-limited images (the latter with the use of adaptive optics). This Large Earth-based Solar Telescope (referred to as the LEST project) addresses in a complementary fashion many of the scientific issues that are at the heart of NASAs Orbit- ing Solar Laboratory (OSL), a moderate-aperture, free-flying, visible- and UV-light space telescope. Able 3.1 provides some points of comparison for these two telescopes: the freedom from atmospheric distortion that allows the OSL to image relatively large structures on the solar surface with high angular resolution is traded off against the difficulty of placing very large aperture mirrors in space (the latter allowing for high-photon-collection capability and for the ultimate in diffraction-limited spatial resolution). In both cases, many of the scientific issues discussed above- including the structure of magnetic field concentrations and of convective overshoot, and the interaction between convection and magnetic fields are directly addressed. Infrared Telescope Instrumentation for Imaging and Spectroscopy The infrared offers some unique physical diagnostic opportunities that have not been exploited. Because imaging improves substantially as one enters the infrared, there are substantial benefits to observing at these wavelengths from the ground. Indeed there are solid reasons for believing that optical interferometric measurements (including speckle interferomet~y and imaging) are best carried out in the infrared. At present, only the NSO/Kitt Peak facilities have any capability in the world in this regard, but this capability is highly compromised because available instrumentation is not optimized for such observations. Development of instrumentation to exploit these unique scientific advantages should receive continued support.
29 TABLE 3.1 Companson of LO and OSL Telescopes Property LEST OSL Field of view a few arcsecondsa 3 arcminutes to 3 arcmis~utas Angular resolution ~ 0.1 arcsecond 0.13 arcsecond (approx.5,000 angstroms) to >0.Sarcsecond b Wavelength range 3,500-24,000 2,200-10,000 Aperture 2.4 m 1.0 m Polarization low compensated Flexibility of focal plane instrument exchange/redesign high none Continuous observing capability at shorter wavelengths no yes Lifespan decades a few years bAssumes successful implementation of adaptive optics. Upper wavelength limit set by instrumentation, not by optics. High-Spatial-Resolution Microwave Instrumentation The near-term potential for extremely high-resolution imaging of coro- nal and chromospheric structures is nowhere as great as at radio wave- lengths; this is of course a consequence of the coherence of radio wave- lengths over very large baselines, so that ground-based interferometric observations can relatively easily reach subarcsecond spatial resolution; in addition, with the aid of spectral resolution and polarimetry, it is possible to infer the structure of magnetic fields in the atmosphere overlying the solar surface. This area of research is only now coming into its own; and the possibilities of correlating radio emission structures with structures that will be seen in the UV and soft X-ray region by spacecraft now under construc- tion in the United States, Europe, and Japan offer a totally novel way of understanding the structure of that part of the solar atmosphere that most directly influences the variability of our terrestrial magnetosphere and near- space environment. For these reasons, high-spatial-resolution microwave instrumentation requires support Theory and Modeling The new high-spatial-resolution observations of the Sun are leading theorists into heretofore unexplored realms of hydrodynamics and magne- tohydrodynamics. Without strong theoretical support, these observational programs are likely to lead to a plethora of data but only a modicum of understanding. In the case of space-based programs, the NASA Solar- lbrrestrial Theory Program and the NASA-supported workshop series for
30 the OSL go a long way toward providing the needed theoretical support and fostering the essential experimenter and theorist interactions. NSF should similarly ensure that the experimental programs it supports receive critical support in the theoretical area as well. MEClIANISMS UNDERIXING THE SOILER CYCLE The Basic Issues The longer time scales of solar variability reflect the presence of ill- understood phenomena in the deep interior that link rotation, convection, and magnetism. Cyclic variations of magnetic activity occur in many other solar-type stars, but we still lack satisfactory theoretical explanations of the origin and development of stellar magnetic fields. Interest in solar variability has recently been stimulated by the discovery that solar luminosity varies on these longer time scales, evidently in step with the general level of magnetic activity. This discovery suggests that some past variations in terrestrial climate may have occurred in response to variations in the total solar luminosity as well as to the very large variations in the UV and in X rays. These harder radiations cause enormous variations in stratospheric temperature; with complicated and still not~fully understood effects on the troposphere. The conditions that support human life may be directly affected. On shorter time scales, the solar UV flux is known to control phenomena such as the orbital lifetimes of artificial satellites in low Earth orbit because it warms and inflates the upper atmosphere, producing increased drag. On- longer time scales, we do not have a sufficiently long quantitative data base to definitively establish the terrestrial effects of solar variability. The Causes of Solar Vanabili~ We have known since Galileo's time of the imperfections of the Sun, and these hint at luminosity variations. We now have data that show these variations directly Figure 3.2~. Several different mechanisms affect luminosity, and each gives some information about the interior structure that produces the perturbation. The new, precise measures of the total solar irradiance shown in Figure 3.2 have given us several types of solar variability. Ibble 3.2 briefly describes the currently known contributors to these solar luminosity variations, as observed by the ACRIM instrument on board the Solar Maximum Mission spacecraft. The tiniest variations yet observed are due to the global solar normal- mode oscillations (see section above, "Probing the Solar Interior"~. The amplitude of a single p-mode is a few parts per million of solar luminosity.
31 C`2 \ it, 1368 v <I: i_ P: 1_ ED o ED 1366 1 ,1 1 1 1 N~> : ,1 1 1 1 _ ~1 1 1 . I ~ , , , 1 , , 1 1 1 , , , , 1 , , , 1 1 1 1 1 o 1000 2000 3000 DAY NUMBER, 1980.0 1990.0 FIGURE 3.2 Daily values of the total solar irradiance (the "solar constant") as observed bar the ACRIM instrument on board the Solar M~Darrnun Mission satellite since 1980. The data show striking dips of a few days' length due to the presence of large sunspots on the visible hemisphere. A general decline toward solar minimum, a flattening dunug the minimum yeam 1984 to 1987, and an upturn most recently suggest the existence of a solar~yde modulation of about 0.1 percent in the solar bolometnc luminosity. The data prior to day number 1600 have a reduced precision due to a spacecraft malfunction; Shuttle astronauts repaired it in orbit in 1984. (Courtesy of the National Aeronautics and Space Administration.) Despite this small amplitude, an individual p-mode frequency can be mea- sured to an accuracy approaching 0.001 percent. The distribution of sound speed throughout the solar interior is the main determining factor for the frequency of a p-mode, and the resulting sound-speed integrals represent the most precise information about interior structure and dynamics. Mil- lions of normal modes of oscillation exist and appear to be permanently excited in the solar interior. The sunspot cycle is perhaps the best studied of the solar variations, since some of the phenomenology has been known for hundreds of years.
32 TABLE 3.2 Identified Components of Solar Luminosity Vanability Cause Time Scale Amplitude p-modes 5 min a few ppm (rms) per mode Granulation 1 hr 0.05% rms broadband Sunspots a few days < 0.2% peak to peak Faculae a few weeks c 0.05% peak to peak Long term ll years? : about O.l~o peak to peak In terms of solar luminosity variability, the solar cycle appears to produce a variation of some 0.1 percent, due to effects of solar magnetism that are at present poorly understood. The Nature of Solar Magnetism As time extends the record of variability, its interpretation becomes steadily more important in studies of solar interior dynamics. The mech- anisms that create the solar magnetic field and distribute it through the interior and atmosphere present some of the most fascinating challenges of astrophysics; the solar dynamo, if understood quantitatively, might have analogs in regions as exotic as accretion disks around black holes. Obser- vations of the solar global structure and its evolution, on active-region and solar cycle time scales, represent an observational prerequisite to solving this problem. The Influences of the Sun on the Earth Solar magnetic activity produces hard radiation that affects the Earth's atmosphere and has significant social and economic consequences. These effects include the inflation of the Earth's upper atmosphere in proportion to the degree of solar activity, with attendant orbital and pointing disrup- tions of low-altitude satellites, the disruption of electrical power distribution caused by ionospheric surges, disturbances of navigation systems, and haz- ards for spacecraft and astronauts via solar flare energetic particles. Also, solar variability must be studied in the context of its linkage to climate and climate change. Much of the interest in applied solar physics centers on the need to predict solar activity for applications in the communications, naviga- tion, electrical power, pipeline, oil exploration, and space industries. This problem is reminiscent of weather forecasting, but there are certain simpli- fications that might make the prediction of solar activity easier. In principle, we can obtain solar data of uniform quality across an entire hemisphere
33 with good calibration and regular, frequent sampling. However, there are intangible uncertainties connected with the unknown physics of the sub- photospheric magnetic fields; this latter problem impels us to study basic solar physics as vigorously as possible. Initiatives and Impacts Solar global observations include synoptic data, in which various tracers of solar activity are followed through the years in a semiquantitative manner. The conduct of such observations tends not to interest research-oriented solar physicists (nor most astronomers), in part because very long- time scales are necessary to achieve results. Perhaps for a similar reason, potential commercial users have not stepped forward to support the creation of improved data bases. In spite of this lack of attention, the U.S. program of synoptic solar data compilation and distribution by the National Oceanic and Atmospheric Administration's (NOAA) National Geophysical Data Center is the world's finest, presenting a large body of useful data ranging from white-light to cosmic-ray observations. Unfortunately, many of these data are of the qualitative classical type, benefiting little from recent technological developments in detectors and data-handling systems and reflecting an inadequate degree of access to the stable observing conditions of space. A modern program of synoptic solar observation and data management is long overdue. Such a program would have interdisciplinary consequences, linking stellar, solar, and terrestrial researchers and applications users. It would also represent an interagency effort, since components of the current synoptic data come from the Department of Defense, NOAA, NSF, NASA, and other sources, none of whose mission responsibilities specifies an adequate program of solar data management. The components of a new, comprehensive program of quantitative data management should conform to the policy and guidelines of the National Research Council's Committee on Geophysical Data report titled Geophys- ical Data: Policy Issues (National Academy Press, Washington, D.C., 1988) and the Committee on Solar-Terrestnal Research report titled Long-Term Solar-Terrestrial Observations (National Academy Press, Washington, D.C., 1988~. The synoptic solar observations would include, at a minimum, solar imagery at moderate spatial resolution in a number of key wavelengths, with a network of automated and carefully calibrated telescopes situated so as to minimize gaps (synoptic observations from space would also be an extremely attractive possibility). The time resolution of the observations should be high enough to permit characterization of rapid fluctuations (e.g., flares and p-modes). Grants for theoretical work, to be carried out hand- in-hand with the observational programs, should focus on solar interior
34 dynamics, including dynamo problems, which include the processes that lead to the variability of the solar constant. THE PHYSICS OF TRANSIENTS The Basic Issues A large solar flare releases as much as 1032 ergs in times as short as 100 to 1000 s. Much of this energy appears in the form of high- energy particles and hot plasma. It is believed that the flare energy comes from the dissipation of- the nonpotential components of strong magnetic fields~oronal current systems in the solar atmosphere, possibly through magnetic reconnection, but the details of the energy release process as well as the mechanism of particle acceleration are still only poorly understood. The interactions in the solar atmosphere of accelerated electrons produce radio emissions, and gamma-ray and hard X-ray bremsstrahlung, and the interactions of protons and nuclei produce gamma-ray line and neutron emissions. The combined observation of the time profiles of these emissions is one of the best-known tools for studying the temporal development of acceleration processes in astrophysics. For example, flares on the Sun are one of the very few astrophysical sites where it has been possible to study simultaneously the acceleration of electrons and protons. Solar flares are also among the few astrophysical sites from which the escaping accelerated particles can be directly detected. Furthermore, many closely correlated lower-energy phenomena (soft X-ray, EUV, UV, and radio emissions), some of which are the direct consequence of the interactions of accelerated particles, can be observed as well. These lower-energy observations reveal the properties (e.g., temperature, density, and magnetic configuration) of the ambient plasma prior to, during, and after the flare. The imaging of flares in hard X rays, the detection of gamma-ray lines and continua from many flares, and the direct detection of solar neutrons are the particularly significant results obtained from NASAs Solar Maximum Mission satellite and the Japanese Hinotori satellites. It has been known for some time that the hallmark of impulsive energy release in flares is the acceleration of electrons to tens of keV, as evidenced by hard X-ray emission. These nonrelativistic electrons probably contain a large fraction of the total flare energy. The hard X-ray images have shown that at least some of this energy is-deposited at the footpoints of magnetic loops. The simultaneous brightening of distant footpoints suggests that energy released in the loops is transported to the footpoints by electron beams. The observation of impulsive gamma-ray emission from many flares has shown that the acceleration of protons and relativistic electrons is also a common properly of the impulsive energy release. The gamma-ray and
35 neutron observations have provided independent (albeit indirect) evidence that the accelerated particles interact at the footpoints. Both the hard X-ray and gamma-ray observations show that the acceleration is very impulsive. Gamma-ray observations have demonstrated that protons are accelerated to GeV energies and electrons to energies of tens of MeV in less than a few seconds. Furthermore, the acceleration of the protons and the electrons is practically simultaneous. Particles in solar flares could be accelerated by shocks, turbulence, and large-scale electric fields, as well as by a variety of other possible mechanisms. Although it is quite clear that acceleration phenomena on the Sun occur -at many sites and produce particle populations of widely different observational characteristics, it is not known whether a single mechanism is responsible for all of the observed acceleration phenomena or whether different mechanisms operate at different sites. Furthermore, it is not known whether particles are accelerated directly from the ambient plasma by a single mechanism or whether the particles are preaccelerated in a process that is distinct from the acceleration mechanism. And perhaps most importantly, it is not clear at all how any of the above mentioned mechanisms can accomplish the very rapid and efficient acceleration that is indicated by the observations. Particle transport in magnetic flare loops is an interesting and exciting problem. The preferential detection of high-energy gamma-ray continua from flares close to the solar limb suggests that the angular distribution of relativistic electrons in the interaction region is anisotropic, peaking at directions tangential to the photosphere. The required anisotropy could result from magnetic mirroring and losses in a convergent chromospheric magnetic flux tube, provided that the awns of the tube is perpendicular to the photosphere. There is as yet no information on the angular distribution of the ions. Such information could be obtained by observing the shapes of gamma-ray lines with detectors having good energy resolution and by observing neutrons from flares at different locations on the Sun. Many of these observations remain to be carried out. Abundances Observations of X-ray and gamma-ray lines from solar flares have pro- vided new techniques for determining abundances in the solar atmosphere. The time-dependent flux of the 2.223-MeV line can provide information on the abundance of 3He in the photosphere, where it has not been obtained by any other method. This line results from neutron capture of hydrogen in the photosphere; 3He is an important neutron sink, and therefore the observed 2.223-MeV line intensity and time profile depend on the 3He/iH ratio in the photosphere.
36 Nuclear deexcitation line fluences are directly proportional to the abundance of elements in the interaction region of the accelerated particles. This region is most likely located at chromospheric densities in flare loops. The abundances obtained from nuclear line spectroscopy can be compared with abundances obtained by atomic spectroscopy of the photosphere and corona. The results indicate that the abundances of carbon and oxygen relative to magnesium, silicon, and iron, as derived from the gamma- ray data, are suppressed in comparison with the corresponding ratios in the Photos there. It has been suggested that this suppression could be due to charge-dependent mass transport from the photosphere to the chromosphere. In addition to these elements, the abundance of neon has also been determined by gamma-ray spectroscopy. A surprising result Is that the neon/oxygen ratio deduced from the gamma-ray data is significantly higher than the corresponding ratio in the corona or in the solar wind. The origin of this difference is not yet understood. It should be noted, however, that the photospheric neon abundance has not yet been measured. Ra~liophysics and Plasma Dynamics Electromagnetic radiation in the vast domain of radio astronomy has demonstrated great potential for diagnostic characterizing of solar struc- tures and dynamics, especially in the corona. In addition to pure elec- tromagnetic waves, the corona generates several other types of radiation, including hydromagnetic waves, Langmuir waves, and whistler waves. These radiations, although not propagating to Earth, still have important roles in energy transport and possibly in particle acceleration for many of the phenomena of solar activity. The electromagnetic radiation sources include continua from the free- free, free-bound, gyroresonance, and synchrotron mechanisms; in addition, there may also be weak emission lines formed by upper-level transitions in hydrogen or other ions. The gyroresonance and synchrotron mechanisms exhibit strong dependence on the magnetic field intensity and orienta- tion; in general, at centimeter and longer wavelengths the corona may become optically thick during flaring. For millimeter waves, unity optical depth occurs in the upper photosphere in normal free-free opacity. The submillimeter-far infrared spectrum then scans through the photospheric layers down to the opacity minimum at about 1.6 microns. Plasma waves in general have a much more complex physics and may serve to couple particles and electromagnetic waves in coronal processes such as the Type I-V radio bursts obseIved in meter-wave dynamic spectra of the Sun. Further, masering cyclotron waves driven by anisotropic distri- bution functions have been implicated in flare energetics. In general, the numerous plasma wave modes provide a link to the distribution functions of
37 energetic particles, one of the key links between laboratory plasma physics and astrophysical plasma applications. Coronal Dynamics The solar corona displays a wide variebr of transient phenomena, in- cluding the radio bursts mentioned above. These constitute some of the most dramatic forms of solar activity and have yet-unresolved associa- tions with the physics of classical solar flares. In some cases, a powerful flare will follow a white-light coronal transient and produce a clearly de- fined blast wave that propagates into the interplanetary medium. In other cases, transient phenomena occur in the corona without any soft X-ray or H-alpha manifestation at all and sometimes from solar latitudes far above the sunspot zones. Current observations are extremely deficient in both synoptic and diagnostic leverage on these coronal phenomena, which is unfortunate because of the rich physics they could reveal. This physics all occurs in a region that is transparent to the observer, since the corona is optically thin. Thus we are not afflicted with the subtleties of radiative transfer theory; on the other hand, stereoscopic observations capable of to- mographic reconstructions of the full three-dimensional geometry of these regions are quite feasible in principle. The coupling of more ordinary flares to the corona also remains a frontier research area. We understand approximately what happens in closed magnetic loops, with powerful energy release signaled by the hard X-ray "impulsive phase," followed by the ablation of dense chromospheric gas up into these loops. Less intelligible are the processes associated with the open field lines known to exist from radio observations. Some flares, often those relatively deficient in hard X-ray and gamma-ray production, couple strongly into the interplanetary medium. Such flares are often associated with one form of "solar cosmic ray" acceleration and coronal transients responsible for geomagnetic perturbations. New observations comparing white-light coronal data with impulsive X-ray signatures suggest that the bunk coronal motions may often precede the impulsive burst, thus raising the possibility that the origin of the flare resides in large-scale magnetic field motions. Initiatives and Impacts Hard X rays and Gamma Rays In the hard X-ray and gamma-ray area, observations with high spatial and energy resolution are needed. Hard X-ray imaging spectroscopy of sufficiently good sensitivity, energy coverage, and angular resolution will al- low researchers to trace the evolution of the electron spectrum throughout
38 the source so that they can determine the accelerated electron distribution, study the magnetic field geometry, and test theoretical transport models. Hard X-ray and gamma-ray spectroscopy with the keV-energy resolution now possible with high-purity germanium detectors will determine the angular distribution of the accelerated ions and electrons, measure abun- dances, and determine the temperature and density of the ambient plasma in flares. There are no plans to carry out such high-resolution observa- tions with satellite-borne detectors during the upcoming solar maximum. Long-duration balloon flights therefore present an important alternative. Neutron Monitors Ground-based neutron monitors have turned out to be very useful for observing the neutrons produced at the Sun by accelerated particle inter- actions. Escaping neutrons have been directly detected from one flare by several neutron monitors (e.g., monitors on Jungfraujoch in Switzerland). Escaping neutrons were also detected by the Solar Minimum Mission satel- lite; furthermore, the protons resulting from the decay of the neutrons in interplanetary space were also observed. There are no U.S. plans to operate neutron monitors during the upcoming solar maximum. However, neutron monitors in Europe, the Soviet Union, Japan, and China will be used for solar flare study during this solar maximum. Radio Obse~vadons Radio observations allow us to study the solar atmosphere from ap- proximately the temperature minimum region out to 1 AU, roughly corre- sponding to the wavelength range from 1 mm to 100 km. Observational technique at these wavelengths has sharpened to the point that interfer- ometry can produce images at milliarcsecond resolution on the Sun. One milliarcsecond corresponds to only some 700 m! At present, no large U.S. radio facilities are dedicated to solar obser- vations, although glimpses with nonsolar instruments such as the very large array (VLA) radio telescopes have produced wonderful results. These re- sults, together with the innovative "frequency-agile" observations at Owens Valley and U.S. and Japanese millimeter-wave work have shown that short- wavelength radio astronomy should be considered seriously in the planning of new observations both for the active and the quiet Sun. A strong case can be made for a "mini-VLA" dedicated to frequency-agile microwave observations at arcsecond resolution; such a facility would be well within the technical state of the art. MAX-91 MAX-91 is a coordinated program of great value for the study of so- lar activity during the upcoming solar maximum. The core of the program
39 consists of Japan's Solar A satellite, NASAs Gamma Ray Observatory (with limited solar observing time), and the Global Geospace Program's WIND spacecraft. In addition, long-duration balloon flights will have a very impor- tant role in this program. These flights could carry out the hard X-ray and gamma-ray imaging and high-resolution spectroscopy observations that will not be earned out by the satellite-borne instruments. In addition, a variety of ground-based observations are planned to support these efforts. The sci- entific objectives of the MAX-91 effort are the study of energy buildup and flare onset and the characterization of energy release and transport mecha- nisms. As indicated earlier, these issues involve fundamentally high-energy phenomena intimately related to the problems of particle acceleration and transport. Fast Optical Spectroscopy Flares, and perhaps other high-energy events, accelerate particles downward into the chromosphere and lower regions of the solar atmo- sphere with dramatic effect. Observations of these events provide important information about energy and momentum balance in the flare process. The physics of shock formation, explosive evaporation of the chromosphere, and thermal conduction along magnetic flux tubes is vital to the flare process but is still not well understood. If we had a more thorough understanding of the physics of particle acceleration and propagation, we could interpret the older, classic observations such as H-alpha images of flares in new and more relevant ways. The key observations are optical spectra of hare emission in Balmer and other chromospheric lines with arcsecond spatial resolution and subsec- ond time resolution. Only a few observations with relatively poor angular resolution and especially poor time resolution have been obtained up to the present time. These have been sufficient, nonetheless, to revolutionize our understanding of the lower parts of flares. Many more observations with better angular and temporal resolution are required. The develop- ment of improved instruments to rapidly obtain optical spectra should be supported. The impact of this initiative will be a better understanding of the chromospheric and upper photospheric parts of flares, improved quan- titative information about energy and momentum balance in flares, and new diagnostics of thermal conduction along magnetic flux tubes. Vector Magnetographs It is evident that magnetic fields are responsible for nearly all solar activity. As a result, a great deal of effort has been devoted to observing magnetic fields in order to understand and predict solar activity. This effort has centered largely on observation of the line-of-sight component of the
40 photospheric magnetic field because it is a fairly easy and robust measure- ment to make. Such observations give only one component at one level of a vector field that extends in three dimensions. Ideally, observations that yield the full vector field measured in a three-dimensional volume as a function of time are needed. This is a formidable observational task The problem is complicated by requirements for excellent angular and spectral resolution, high sensitivity to linear and circular polarization, and freedom from polarization effects in telescopes. These requirements have frustrated most earlier efforts to observe the vector magnetic field, in spite of the importance of the task. Now we have improvements in detector technology, new image stabilization techniques, and powerful analysis programs. Thus the promise of obtaining useful vector field measurements, at least in the photosphere, is brighter than ever before. This has led to development activity at nearly every solar observatory. These activities should be sup- ported, and the construction of a few of the most promising instruments should be fully funded. The impact of a successful vector magnetograph on the study of high- energy solar phenomena will be profound. It will be possible to follow the buildup and storage of magnetic energy, which is thought to power high- energy events. It will also be possible to localize electric current systems, which may trigger explosive energy release. These capabilities offer the potential of predicting solar flare activity with far higher reliability than is currently possible.' Advanced Coronal Observations As noted previously, it is now suspected that coronal transient'phe- nomena result from the relaxation of stressed coronal magnetic fields. The coupling between this relaxation and the occurrence of associated solar flares or eruptive phenomena is currently unclear. Understanding the transient process requires high-temporal-resolution observations of the solar corona and a more precise understanding of the spatial and tem- poral relationship between coronal activity and near-surface phenomena. Current instrumentation is inadequate. Revolutionary new coronal obser- vations will be carried out by instruments on board the SOHO spacecraft, and it is within our technical capability to extend coronal remote-sensing observations to 1 AU or beyond. Limb observations of the innermost corona, with high spatial and spec- tral resolution, can be obtained effectively with a ground-based k-coronameter. A suitable k-coronameter will feature a sensitive system capable of observing the corona to within 50,000 lan of the solar surface with high spatial and temporal resolution, employing the polarization se- lectivity necessary to observe the corona in the presence of sky light and
41 its fluctuations. Such an instrument is required to examine the nature of the origin of coronal transient phenomena and the relation between tran- sients and surface solar activity. The impact of these observations will be to illuminate the nature of the evolution of the solar large-scale magnetic field and the role of that evolution in generating solar activity.
