5
Report of the Panel on Stars and Stellar Evolution

SUMMARY

The science frontier for stars and stellar evolution is as close as the Sun and as distant as exploding stars at redshift 8.3. It includes understanding processes of exquisite complexity that connect the rotation of stars with their magnetic fields and areas of nearly total ignorance about phenomena that have been imagined but not yet observed, such as accretion-induced collapse. Because astronomers understand stars well, they have the confidence to use them as cosmic probes to trace the history of cosmic expansion; but because this understanding is not complete, there is much to learn about the subtle interplay of convection, rotation, and magnetism or the not-so-subtle violent events that destroy stars or transform them into neutron stars or black holes. Although the topics of stars and their changes over time comprise great chunks of introductory astronomy textbooks, and although the tools for these investigations are tested and sharp, many of the simplest assertions about the formation of white dwarfs, mass loss from giant stars, and the evolution of binary stars are based on conjecture and a slender foundation of facts.

The future is promising. X-ray and radio observations allow astronomers to probe stars where strong gravity is at work. These settings stretch the understanding of fundamental physics beyond the range of laboratory investigations into unknown areas of particle interactions at higher densities than those produced in any nucleus or terrestrial accelerator. By testing three-dimensional predictions against the evidence, more-powerful computers and programming advances put astronomers on the brink of understanding the violent events that make stars explode



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 207
5 Report of the Panel on Stars and Stellar Evolution SUMMARY The science frontier for stars and stellar evolution is as close as the Sun and as distant as exploding stars at redshift 8.3. It includes understanding processes of exquisite complexity that connect the rotation of stars with their magnetic fields and areas of nearly total ignorance about phenomena that have been imagined but not yet observed, such as accretion-induced collapse. Because astronomers under- stand stars well, they have the confidence to use them as cosmic probes to trace the history of cosmic expansion; but because this understanding is not complete, there is much to learn about the subtle interplay of convection, rotation, and magnetism or the not-so-subtle violent events that destroy stars or transform them into neu- tron stars or black holes. Although the topics of stars and their changes over time comprise great chunks of introductory astronomy textbooks, and although the tools for these investigations are tested and sharp, many of the simplest assertions about the formation of white dwarfs, mass loss from giant stars, and the evolution of binary stars are based on conjecture and a slender foundation of facts. The future is promising. X-ray and radio observations allow astronomers to probe stars where strong gravity is at work. These settings stretch the understand- ing of fundamental physics beyond the range of laboratory investigations into un- known areas of particle interactions at higher densities than those produced in any nucleus or terrestrial accelerator. By testing three-dimensional predictions against the evidence, more-powerful computers and programming advances put astrono- mers on the brink of understanding the violent events that make stars explode 207

OCR for page 207
Panel rePorts—new worlds, new HorIzons 208 and collapse. Advances in laboratory astrophysics lead to a better understanding of the underlying nuclear, atomic, and magnetohydrodynamic (MHD) processes. New technology for optical and infrared (IR) spectropolarimetry and for interfer- ometry open up the possibility of seeing magnetic fields and resolving the disks of stars. When well-sampled imaging is coupled to powerful systems for processing vast quantities of data sampled over time, subtle features of stellar interiors can be inferred. Similarly, rare and rapid transients that have eluded surveys to date will surely be found, and may be connected with gravitational waves. These advances are certain to open up a new and unexplored world of investi- gation on timescales from seconds to decades. In this report, the Astro2010 Science Frontiers Panel on Stars and Stellar Evolution sketches the most fertile opportuni- ties for the coming decade in the field of stars and stellar evolution. The panel is confident that it will prove a fruitful decade for this field of astronomy, with the resolution of today’s questions producing many new problems and possibilities. As requested by the Astro2010 committee, the panel formulated its report around four science questions and one outstanding discovery opportunity. The panel is under no illusion that this short list is complete: the field is so rich that there will surely be advances in areas not emphasized here. The panel does, however, have every reason to believe that these questions capture some of the most prom- ising areas for advances in the coming decade. The four questions and discovery opportunity are as follows: 1. How do rotation and magnetic fields affect stars? 2. What are the progenitors of Type Ia supernovae and how do they explode? 3. How do the lives of massive stars end? 4. What controls the mass, radius, and spin of compact stellar remnants? 5. Unusual discovery potential: time-domain astronomy—in which the tech- nology on the horizon is well matched to the many timescales of stellar phenomena. The subsections below summarize the main points. How Do Rotation and Magnetic Fields Affect Stars? There’s an old chestnut about a dozing theorist at the weekly colloquium who opens his eyes at the end of every talk and rouses himself to ask, to great approba- tion for his subliminal understanding, “Yes, all very interesting, but what about rotation and magnetic fields?” Astronomers are now in a position to address this question in a serious way. In the Sun, the effects are visible; in many other stars they are likely to be much more important. It is not sufficient to think of rotation and magnetism as perturba- tions on a one-dimensional star. These are fundamental physical phenomena that demand a three-dimensional representation in stars.

OCR for page 207
rePort Panel stars stellar evolutIon 209 of tHe on and Astronomers are poised to learn how stars rotate at the surface and within and how that rotation affects mass loss and stellar evolution. They seek a better under- standing of how magnetic fields are generated in stars across the mass spectrum and of how these fields power the chromospheres and coronae that produce observed magnetic activity. Finally, the origin of highly magnetized main sequence stars, in which surface fields approach 104 gauss, remains mysterious, and the investigation of these stars promises to shed light on the star-formation process that produced them as well as on the origin of even more highly magnetized compact objects. The prospects for progress on this question in the next decade stem from the emergence of greatly improved tools for measuring magnetic fields from po- larization, for resolving the atmospheres of some stars with interferometry, for probing the interiors of stars through their vibration spectra, and for extending observations into X-rays and gamma rays. When combined with more thorough understanding of the static and dynamic properties of magnetic atmospheres, astronomers will learn how stellar atmospheres really work and how rotation and magnetism affect the evolution of stars. What Are the Progenitors of Type Ia Supernovae and How Do They Explode? Many lines of evidence converge on the idea that Type Ia supernovae are ther- monuclear explosions of white dwarfs in binary systems. Because of their high luminosity, and with effective empirical methods for determining their distances from light-curve shapes, Type Ia supernovae have acquired a central role not just in stellar astrophysics but also in tracing the history of cosmic expansion and in revealing the astonishing fact of cosmic acceleration. Because this result points to a profound lack in the understanding of gravitation, a problem right at the heart of modern physics, completing the astronomical story of Type Ia supernovae is a pressing priority for the coming decade. First of all, the provenance of the exploding white dwarfs seen in other galaxies is not known with certainty. The prevailing picture is that Type Ia explosions arise in binary systems in which a white dwarf accretes matter until it approaches the Chandrasekhar limit, simmers, and then erupts in a thermonuclear flame. But it is not known how this picture is affected by chemical composition or age, two essen- tial ingredients in making a precise comparison of distant events with those nearby. Events that are precipitated by the merger of two white dwarfs are not excluded. The Type Ia supernovae in star-forming galaxies and in ellipticals are at present treated in the same way, but this is the result of small samples, not of evidence that they should be analyzed together. More broadly, these gaps in knowledge illuminate the need for a better understanding of the evolution of interacting binary stars, which are responsible for a variety of crucial, yet poorly understood, phenomena. It can be expected that both theory and improved samples will place this work on a firmer foundation. The complex, turbulent, unstable nuclear flame that

OCR for page 207
Panel rePorts—new worlds, new HorIzons 210 rips through the star and incinerates its core is at the present time impossible to compute fully in three dimensions. But the prospects for achieving that goal in the coming decade are intriguing. Samples today amount to a few hundred objects at low redshift and similar numbers beyond redshift 0.5. Much larger and significantly more uniformly discovered samples are coming soon through targeted aspects of time-domain surveys. They will create a much tighter connection between chemis- try, binary stellar populations, and supernova properties. Inferences on dark-energy properties are at present limited by inadequate understanding of the intrinsic properties of Type Ia supernovae as distorted by interstellar dust. Observing in the rest-frame infrared will expand the basis for comparing observations with compu- tations and provide more accurate measurements of dark energy. How Do the Lives of Massive Stars End? Ninety-five percent of stars will end their lives as white dwarfs. For the rest, stellar death is spectacular and dramatic: these massive stars can explode as su- pernovae, emit gamma-ray bursts (GRBs), and collapse to form neutron stars or black holes. The elements that they synthesize and eject become the stuff of other stars, planets, and life. The energy and matter that they produce are crucial for the evolution of galaxies and clusters of galaxies. Despite a basic understanding that gravity is the energy source for these events, a clear connection between the mass and metallicity of the star that collapses, the nature of the collapse and explosion, and the properties of the compact remnant remain mysterious. The rotation of the progenitor and its mass loss, areas of un- certainty highlighted in the panel’s first question, seem to be essential aspects of the link between core collapse supernovae and GRBs. Exploring these frontiers will require continued thoughtful analysis and full-blown first-principles calculations. The full range of outcomes for stellar deaths may not be well represented in current observational samples. Deeper, faster, wider surveys will surely detect rare or faint outcomes of stellar evolution that have not yet been seen. These could include pair-instability supernovae and other types of explosions that have only been imagined. The role of massive stars in the evolution of the universe is coming into view. The fossil evidence of massive stars is embedded in the atmospheric abundance patterns of our galaxy’s most metal-poor stars. The most distant object measured so far is a gamma-ray burst, presumably from a massive star, at redshift 8.3. In the coming decade, the direct observation of the first generation of stars, which are predicted to be exceptionally massive, will be within reach with the James Webb Space Telescope (JWST). Massive stars could be the source of gravitational wave signals, a neutrino flash, or nuclear gamma-ray lines. All of these novel messages from stars are within reach

OCR for page 207
rePort Panel stars stellar evolutIon 211 of tHe on and for very nearby cases, and could be exceptionally important in shaping the future understanding of the deaths of massive stars. What Controls the Mass, Radius, and Spin of Compact Stellar Remnants? Unanswered questions about the magnetic fields and rotation of stars carry through to similar questions about the exotic remnants that they leave behind as neutron stars and black holes. These are exceptional places in the universe where understanding of physics is extended beyond the reach of any laboratory. For example, the equation of state for nuclear matter sets the relation between mass and radius for neutron stars. Theoretical understanding of the forces at work is uncertain where the density exceeds that of the densest nuclei. Prospects for measuring masses for radio pulsars and neutron star radii from X-ray techniques promise a glimpse into the strange world of quantum chromodynamics and the possibility of hyperons, deconfined quark matter, or Bose condensates. The spins of neutrons stars and black holes are rich areas for future work. It is known that millisecond pulsars are spinning much faster than when the neutron stars were formed, and it is understood how accretion in a binary system can ac- celerate their rotation, but the mechanism that limits how fast these neutron stars can whirl is not known. The answer is expected to come from new pulsar surveys that are less biased against detecting the fastest pulsars. Similarly, there are now plausible measurements that imply that black holes are spinning rapidly. It seems very likely that these black holes are telling us about the conditions in which they formed, during the collapse of a massive star—one of the key points in the panel’s third question. In the coming decade, X-ray spectroscopy should be a powerful technique for expanding the slender sample of spinning black holes, all identified in binaries. Most stars surely become white dwarfs, but present understanding of the white dwarf mass for a main sequence star of a given initial mass is seriously incomplete. How stars lose mass is not understood well enough to predict which stars will become white dwarfs. Important details of the white-dwarf population remain unsolved and could lead to types of supernovae that have not yet been recognized. Large surveys will be powerful tools for finding these objects, making it possible to fill in these embarrassing gaps in understanding. Discovery Area: Time-Domain Astronomy For poets, stars are symbols of permanence. But astronomers know that this is not the whole story. Stars reveal important clues about their true nature by their rotation, pulsation, eclipses and distortions, mass loss, eruptions, and death. Across a wide range of timescales from seconds to years, stars are changing, and scientific

OCR for page 207
Panel rePorts—new worlds, new HorIzons 212 knowledge has been obtained through narrow windows of time set by practical matters of telescope time, detector size, and the ability to sift the data. The panel foresees a rich flood of data from specialized survey instruments capable of explor- ing this new frontier in astronomy, across the electromagnetic spectrum. These instruments will provide new time-domain data, with the potential for major im- pact on stellar astronomy, ranging from the precise understanding of stars through seismological data and the periodicities that rotation produces, to the detection of rare transient events that have not yet been revealed in extant surveys. An example provides a glimpse of the excitement: wide, deep, and frequent surveys will be the way to find the electromagnetic counterparts of gravitational wave events. The broad problems of binary star evolution, about which so much is assumed and so little is known, can be sampled by such an undertaking, perhaps advancing the knowledge of the progenitors of thermonuclear supernovae from being a plausible story to becoming an established fact. The range of stellar phenomena that will be addressed with large, accessible, time-domain databases goes far beyond the four questions of the panel. Summary of Panel’s Conclusions The conclusions of this panel report are summarized in Table 5.1. INTRODUCTION The advent of quantum mechanics and the study of nuclear fusion in the 1930s led to the first successful models for energy generation in stars and to a basic un- derstanding of the distribution of stars in the Hertzsprung-Russell diagram. This work, along with the study of stellar atmospheres, laid the foundation for modern astrophysics. In just the past decade, the resolution of the solar neutrino problem demonstrated conclusively the presence of new neutrino physics and the accuracy of solar models originally developed in the 1960s. The success of this work relied in part on the approximate spherical symmetry of stars, which enabled accurate one-dimensional models of their structure. The opening of new parts of the electromagnetic spectrum dramatically broadened astronomers’ views of stellar phenomena, leading to a number of breakthrough discoveries. Newly discovered radio pulsars in binaries, includ- ing the unique double-pulsar system, provided some of the most stringent tests of general relativity. Advances in X-ray astronomy have led to new discoveries related to accreting neutron stars and black holes, compact remnants of massive stars, and energetic phenomena, such as coronae and flares, on normal stars. The characterization of brown dwarfs, cool low-mass objects at the border between stars and planets, was due to the development of new infrared surveys. And the

OCR for page 207
rePort Panel stars stellar evolutIon 213 of tHe on and TABLE 5.1 Summary of Conclusions of the Panel on Stars and Stellar Evolution Question 2: Question 4: Question 1: What Are the What Controls the How Do Rotation and Progenitors of Type Ia Question 3: Mass, Radius, and Spin Magnetic Fields Affect Supernovae, and How How Do the Lives of of Compact Stellar Stars? Do They Explode? Massive Stars End? Remnants? Current and ATST, SDO, HST, 4-m PTF, PanStarrs-1, KAIT, PTF, PanStarrs-1, EVLA, ALMA, LOFAR, expected and 8-m telescopes, PAIRITEL, JWST, Swift, Swift, NuSTAR, EVLA, Gaia FAST, LIGO, FRIB, facilities Kepler, CoRoT, Gaia HST, Chandra, 8- to 8- to 10-m telescopes, Chandra, XMM, Suzaku, 10-m telescopes HST, GALEX, Chandra RXTE, Fermi New High spatial and OIR time-domain, Multiwavelength (radio Large-area decimeter- facilities synoptic solar large-FOV, high- to X-ray) time-domain, wavelength telescope; needed magnetometry; helio- cadence observations; large-FOV, high- large-effective-area X-ray and asteroseismology; precise IR follow-up; cadence observations; timing and spectroscopy; OIR interferometry; X-ray spectroscopy; post-Swift GRB gravitational wave OIR time-domain, 20- to 30-m telescope studies; X-ray observatory large-FOV spectroscopy; neutrino observations; high- and gravitational wave resolution multiobject observatories; 20- to OIR spectroscopy; 30-m telescope plasma physics experiments Crucial Detailed solar and Panchromatic Large-scale three- High-sensitivity X-ray capabilities stellar studies of spectroscopy of a dimensional timing and spectral internal rotation and representative sample; simulations; discovery observations of known magnetism; surveys large sample for of broad range neutron stars and of stellar surface finding diverse objects of transients and black holes; laboratory rotation, activity, and correlations with multiwavelength measurements of nuclear magnetism, and mixing environment; advanced follow-up; nuclear equation of state; deep diagnostics; three- three-dimensional data and oscillator radio pulsar searches; dimensional MHD simulation capability; strengths; abundance star-cluster white-dwarf simulations; pulsation progenitor surveys; studies of extremely searches; gravitational theory; UV and X-ray nuclear cross sections low metallicity stars wave detection of spectroscopy compact binary inspirals Other Dedicated follow-up for inferring fundamental stellar properties; progress on abundance priorities determinations; laboratory measurements of opacities; support for basic theory and computational astrophysics NOTE: Acronyms are defined in Appendix C. conclusive determination that long-duration gamma-ray bursts are associated with the deaths of massive stars showed that the central collapse of some massive stars can produce relativistic jets. In parallel with the continued exploration of the electromagnetic spectrum, high-energy neutrino and gravitational-wave views of the universe will likely be unveiled in the next decade. Observational techniques such as astrometry, interfer-

OCR for page 207
Panel rePorts—new worlds, new HorIzons 214 ometry, and time-domain surveys, all of which are well suited to studies of stellar phenomena, will reach maturity. Given this influx of new data, there are tremen- dous opportunities in stellar astrophysics, from the lowest-mass stars to compact objects. In “traditional” topics such as stellar structure and evolution and stellar seismology, it is crucial that the United States support the intellectual infrastruc- ture—observational, experimental, and theoretical—to take full advantage of these new opportunities. Stellar astrophysics has informed many other areas of physics, including nu- clear physics, particle physics, and general relativity. Moreover, an understand- ing of stellar astrophysics is needed for many other problems in astronomy. The study of galaxies at high redshift relies critically on an understanding of the stellar populations that make up those galaxies. The formation of stars, galaxies, and the intracluster medium in galaxy clusters is strongly influenced by the heavy elements, ionizing photons, and explosions produced by massive stars. The Sun continues to be a working template for understanding magnetohydrodynamics and plasma physics “in practice”—physics that is crucial in many other arenas, including that of compact objects. The most distant known object in the universe is now a GRB. These GRBs are (temporarily) much brighter than quasars in the optical-ultraviolet (UV), allowing unique studies of the intergalactic medium at high redshift. A distinct and complementary probe of star formation at these early times is pro- vided by the discovery of nearby extremely metal-poor stars, which constrain the nucleosynthetic products of the first generations of stars. Just as stellar astrophysics has a significant impact on other branches of physics and astrophysics, it also requires input from other disciplines, notably laboratory experiments. For example, the stellar interior models now in use rely on purely theoretical opacity calculations, but these will be tested by laboratory data in the next decade. Next-generation solar-neutrino experiments can measure the central solar temperature and potentially constrain solar abundances. Key nuclear cross sections are needed for a quantitative understanding of nucleosynthesis and energy generation in stars and stellar explosions. More recently, experiments focused on studying complex hydrodynamic processes have provided key insights into aspects of stellar physics. For example, laboratory experiments and space-physics mea- surements contributed to the recognition that fast magnetic reconnection occurs only in plasmas with low collision rates, with potential applications to stellar and accretion disk coronae. Continued laboratory studies of basic physical processes important in stellar astrophysics, such as reconnection, angular-momentum trans- port, and combustion, would complement more traditional observational and theoretical work. More generally, a transition to models grounded in experimental data has the capability to open up new realms of precision stellar astrophysics that could have a major impact on astronomy and physics.

OCR for page 207
rePort Panel stars stellar evolutIon 215 of tHe on and One-dimensional stellar models have yielded considerable insight and are quantitatively sufficient for many applications. However, there are also crucial aspects of stellar structure and evolution that require a three-dimensional ap- proach. In these areas, a combined effort involving numerical simulations inform- ing observations, and vice versa, is needed for progress. Studying these inherently three-dimensional problems computationally has become feasible in the past 10 to 15 years, with the tremendous advances in computational resources. Similarly, ob- servational progress on many of these problems is just now feasible. This includes the realization of interferometric techniques in the past decade to illuminate the rotationally distorted shapes of massive stars, the enlarged radii of active M-dwarfs, and the details of how stars lose mass. In addition, observations of the Sun show that convection (inherently multidimensional) plays a critical role in shaping both its rotation profile and magnetic structure. The observational and theoretical study of the solar case provides a point of departure for the study of rotation and mag- netism in other stars (science question SSE 1). Other problems in which a full three-dimensional understanding is crucial are the thermonuclear explosions of white dwarfs in Type Ia supernovae, and the core collapse and explosion of massive stars (science questions SSE 2 and SSE 3). These topics will benefit from the development of new time-domain surveys (SSE discovery area). Already, surveys are finding supernovae with extreme properties in terms of luminosity and energy (both high and low). The evolution leading up to stellar explosions—in particular, mass loss and the dynamics of binaries—is crucial for an understanding of the variety of observed explosions. The study of compact stellar remnants has become mature, but fundamental questions remain about the basic properties of compact objects, such as what determines their masses, spins, and radii (science question SSE 4). The entire subject of stellar astrophysics will benefit tremendously from large time-domain surveys, given that the variability of the timescales to which large surveys are sensitive matches well with those of many stellar phenomena (see the section “SSE Discovery Area: Time-Domain Surveys”). In addition to explosive events often related to compact objects, a variety of variable and binary stars can be studied. Many time-variable events may be found that have been predicted but not yet discovered (e.g., orphan afterglows of gamma-ray bursts). THE SCIENCE FRONTIERS SSE 1. How Do Rotation and Magnetic Fields Affect Stars? The standard models of stellar structure are traditionally one-dimensional. Observational and theoretical results in the past decade have demonstrated that

OCR for page 207
Panel rePorts—new worlds, new HorIzons 216 the study of rotation and magnetic fields, coupled with convection, demands a dynamic, three-dimensional approach that is now achievable. This complex reality is most vividly illustrated by the Sun’s bewildering array of multiscale inhomoge- neity, which is now seen with remarkable magnetic sensitivity, and with temporal and spatial resolution, and understood with the help of detailed simulations. It is sobering that this rich structure and its few-gauss average dynamo fields result from the interaction of only a few-kilometers-per-second surface rotation with comparable subsurface convective velocities. The much broader range of stellar rotation, convection, and magnetism poses both a challenge and an opportunity to complement the detailed views of the Sun. Beyond the Standard Picture Rotation is now recognized as fundamental to the understanding of stellar evolution. Photometric monitoring programs have mapped the mass-dependent range of pre-main sequence rotation rates and have shown that star-disk interac- tions (planet and star formation processes) are key to the subsequent evolution of stellar rotation. It is known that Sun-like stars spin down from magnetized winds. Yet recent large samples of active stars now show that the stellar-mass dependence of these winds is uncorrelated with the boundary at which stars become fully convective, contrary to theoretical expectations. Helioseismology has revealed the internal solar rotation, invigorating debate over dynamo theories and confirming the strong coupling between the radiative core and convective envelope. However, mechanisms for angular-momentum transport in stellar interiors, the related mixing, and relevant timescales are still poorly understood. Nascent asteroseismic constraints on internal stellar rotation exist, but crucial knowledge, such as the core rotation rate in pre-supernova stars, is lacking. Understanding rotation-induced mixing in stellar interiors and its effect on stellar and chemical evolution demands that these uncertainties be resolved. It is known that magnetic fields in cool stars are generated by a dynamo mecha- nism, but its precise nature, and the heating mechanism for chromospheres and coronae, remain subjects of vigorous debate. Computational advances now allow three-dimensional, radiative MHD models of the outer solar atmosphere that are consistent with observations showing no meaningful “average” or homogeneous chromosphere. Such models are finally making progress on the important ques- tion of how magnetic energy from the convective regions makes its way into the magnetically dominated layers. Strong magnetic fields are also detected in hot, higher-mass stars without strong surface convection, and in their stellar remnants. Their origin and evolution, possibly from relic fields, is a mystery. The prospect of exciting breakthroughs in the understanding of these issues arises from great leaps in stellar observational capabilities, such as interferometry,

OCR for page 207
rePort Panel stars stellar evolutIon 217 of tHe on and spectropolarimetry, magnetometry, asteroseismology, and extended waveband sur- veys from radio and infrared to X-rays and gamma rays. Stars can now be imaged and modeled as three-dimensional objects. Realistic three-dimensional atmosphere simulations are a fruitful new frontier in the understanding of the envelopes of stars. Powerful diagnostics of internal properties are available, and time-domain surveys will allow the study of stellar rotation in samples of unprecedented size, precision, and duration. The complementary approaches of large surveys and detailed studies of smaller samples promise new observational constraints on the origin, nature, and consequences of stellar rotation and magnetic fields. Gaia will provide precise astrometric data and spectroscopic information for an unprec- edented sample, which will be invaluable for characterizing stellar properties. These will be combined with unprecedented new capabilities for high-resolution solar observations, particularly with the Advanced Technology Solar Telescope (ATST) and the Solar Dynamics Observatory (SDO). On the theoretical side, fully three- dimensional, time-dependent MHD solar simulations will provide the template for more realistic two-dimensional and three-dimensional stellar models. Against this backdrop, the following subsections describe five important problems involving stellar rotation and magnetic fields that seem particularly ripe for progress. How Are Magnetic Fields Generated in Stars? Helioseismology, spectropolarimetry, and radio gyrosynchrotron observations offer new tools for measuring the complex global magnetic structures in the Sun and stars. Local helioseismology with the SDO (a satellite to be launched in early 2010) will revolutionize astronomers’ vision of subphotospheric magnetic fields, while sensitive polarimetry obtained with ATST (now under construction) will unveil the dynamic magnetized atmosphere from photosphere to corona (Fig- ure 5.1). Lower-spatial-resolution coronal magnetometry over the full disk using microwave gyrosynchrotron or IR Zeeman measurements could effectively sample intermittent solar atmospheric explosive events. This would also be a powerful trig- ger for more sensitive ATST observations. This new generation of high temporal- and spatial-resolution magnetometry will help disentangle the influence of flares and mass ejections on chromospheric and coronal structure. For stars, accurate spectropolarimetry with high Stokes Q/U/V sensitivity is the demonstrated key to measuring surface magnetism. A significant increase in the stellar sample size and duration of these observations could be made with just a single, dedicated 4-m- class telescope and suitable instrumentation equipped with a spectropolarimeter. A new generation of sophisticated simulations that include realistic treatments of radiation and non-equilibrium ionization can be harnessed to interpret these data. The panel urges the provision of support for such theoretical investigations and the extension of detailed solar models to the broader stellar regime. This re-

OCR for page 207
Panel rePorts—new worlds, new HorIzons 236 radiation which, if correct, would be of major importance to current and future gravitational-wave-detection facilities such as the Advanced Laser Interferometer Gravitational Wave Observatory (LIGO). Identifying the correct spin frequency distribution of MSPs will help to determine both the maximum spin rates of neu- tron stars and the limiting physical processes. Current and future radio and X-ray timing surveys have many fewer selection effects toward rapidly rotating pulsars than those in the past. If such systems are detected, they would directly limit the neutron-star equation of state by determining the maximum radius of the neutron star as a function of its mass for which it does not shed material at its equator. New constraints on neutron-star physics would come from the detection of a neutron star spinning more rapidly than 1,000 Hz. Astrophysical black holes are completely described by just two quantities, their mass and spin. Although the masses of stellar black holes in X-ray binaries have been measured dynamically for decades, it is only in the past few years that it has become possible to constrain the spins of black holes. The spin is constrained by determining the inner radius of the accretion disk, either by fitting the thermal disk component of the X-ray continuum spectrum, or through the relativistically broadened shape of the Fe K disk fluorescence line. The radius inferred by these methods is believed to be comparable to that of the last stable orbit in general relativity, but there are systematic uncertainties in this association that limit the precision of current constraints on black hole spin. Inferences about spin have now been made in 10 systems (using a variety of X-ray missions, most recently including Chandra, XMM-Newton, and Suzaku) and most are rotating significantly, with a wide variety of black hole spins measured, and several are believed to be spinning near the maximal amount allowed by general relativity. A slowly spinning, disk- accreting black hole must double its mass in order to spin rapidly, which is impos- sible for a black hole in an X-ray binary; thus, the measured spin distribution is essentially sampling the birth properties of these black holes. An alternative way of measuring black hole spin is through the spin-orbit coupling of a pulsar with a black hole. This method will require pulsar searches to discover pulsars with black hole companions. Gravitational-wave detection of stellar black-hole/black-hole or black-hole/neutron-star inspirals offers still another promising method for mea- suring black hole spins (and masses). A larger sample of black hole spin measurements will provide very strong con- straints on models of massive star evolution and core-collapse supernovae. More broadly, an improved understanding of black hole spin can be used to address a number of important issues, including the role of black hole spin in producing jets and in powering GRBs. Accurate knowledge of the black hole spin distribution is also crucial for designing theoretical search templates required for the direct detec- tion of gravitational waves from black-hole/black-hole and black-hole/neutron- star mergers. In order to make continued progress, soft X-ray continuum spectros-

OCR for page 207
rePort Panel stars stellar evolutIon 237 of tHe on and copy and medium- to high-resolution X-ray line spectroscopy of a larger sample of black holes in X-ray binaries is needed, requiring a more sensitive telescope than Chandra or XMM-Newton. In addition, further numerical and theoretical work on general-relativistic MHD models of black hole accretion disks is essential for interpreting X-ray continuum observations, and better theoretical models of the X-ray irradiation and fluorescent Fe line emission from the inner accretion disk are needed to interpret Fe line spectra. What Determines the Initial-Final Mass Relation Connecting Progenitors to White Dwarfs? Observational constraints on the initial-final mass relation come primarily from white dwarfs in open clusters and require accurate main-sequence turnoff ages plus white-dwarf cooling ages (to infer the initial mass) and precise final masses (Figure 5.9). The largest source of error is the variance in theoretical mass- lifetime relationships for stars with convective cores (and, by extension, open cluster ages). Gaia will provide precise distances and membership information for open clusters, and asteroseismology from missions such as CoRoT and Kepler may constrain the sizes of convective cores. Uncertainties in the white-dwarf cooling timescale also need to be addressed, in particular the properties of the atmospheric “blanket” that governs heat transport and is a poorly understood by-product of asymptotic giant branch (AGB) evolution. The mass of the blanket is a result of the processes that end the AGB evolution of the star. The recent discovery of carbon- atmosphere white dwarfs is as yet unexplained and points to interesting discovery areas in white-dwarf formation. Also crucial to understanding white-dwarf prop- erties is the onset of crystallization, which alters the internal energy structure and causes an abrupt change in effective temperature and luminosity. Both have been constrained by asteroseismology on a small number of stars, giving a partial picture and great promise for future advances. The mass of a white dwarf originating from a single star such as the Sun is related to the luminosity of the star as it leaves the AGB. This luminosity, and thus the white-dwarf mass, is determined by the mass-loss process. Theoretical and ob- servational studies of the dependence of mass-loss rates on stellar parameters have not reached a consensus, and prescriptions advocated and used differ dramatically from one another. Empirical and theoretical formulas span a wide range of slopes. To match observed initial-final mass relations with evolutionary and population models, empirical laws have been “corrected” with a variety of tuning parameters. However, there is no widely accepted or demonstrably correct mass-loss formula for the mass loss that produces white-dwarf stars, and thus no predictive power for extrapolating to understudied populations (such as young, low-metallicity cases). To measure mass-loss rates for large numbers of stars, infrared surveys and

OCR for page 207
Panel rePorts—new worlds, new HorIzons 238 FIGURE 5.9 The fates of stars—the relationship between their initial and their final mass—for stars that lose mass and thereby (if they start with >1.4 M◉) avoid exploding as supernovae, are observationally constrained by the study of white dwarf stars in clusters of known age. Considerable uncertainty exists for intermediate-mass stars (2 to 8 M◉), including how much initial composition affects the result. With the detection of many more white dwarfs in clusters, enough information to constrain mass-loss models and perhaps enough to extrapo- late to unobservable populations (low metallicity, high mass—as in the early universe) may become available. SOURCE: J.S. Kalirai, B.M.S. Hansen, D.D. Kelson, D.B. Reitzel, R.M. Rich, and H.B. Richer, The initial-final mass relation: Direct constraints at the low-mass end, Astrophysical Journal 676(1):594, 2008, reproduced by permission of the AAS. Courtesy of Jason Kalirai, Space Telescope Science Institute. molecular-line surveys have already proven useful, although the IR measures re- quire reliable gas/dust ratios and the radio lines measure mass loss at a different time (farther out in the flow); many mass-losing stars have variable outflows. The modeling of mass-loss processes requires non-local thermodynamic equilibrium hydrodynamics with shocks, non-equilibrium chemistry, and grain nucleation and growth. Strong tests of models are coming from interferometric studies of the structures of the atmospheres of mass-losing stars, as, for example, the discovery of “molecular shells” at about twice the stellar radius and coincident with the region where dust is expected to form. Molecular lines are ideal for mass-loss studies be- cause (1) they trace the gas, and (2) they carry velocity information. CO has been widely used in our galaxy. The Atacama Large Millimeter Array (ALMA) should be

OCR for page 207
rePort Panel stars stellar evolutIon 239 of tHe on and able to detect CO and thus measure mass-loss rates for AGB and red giant branch (RGB) stars in the Magellanic Clouds, a vital laboratory for the study of these populations. Additionally, CO traces mass loss for both C-rich and O-rich giants, which is not the case for tracers currently used to study the highest-mass-loss-rate objects in the Small Magellanic Cloud and LMC. Most white-dwarf stars are believed to be composed of carbon and oxygen. Ob- servations of novae show that some O-Ne-Mg white dwarfs are formed, presumably from relatively high mass progenitors, with details of the formation channel(s) as yet unclear. There are also He white dwarfs, including a surprisingly large popula- tion of He white dwarfs in very metal-rich clusters and a population of single-field He white dwarfs. Explaining the origin and evolution of these different classes of objects should be illuminating. The formation of He white dwarfs is understood only in the context of binaries, so further understanding of other ways of forming them potentially by single stars or through disrupted binaries or ejections from dense star clusters is needed. Stars that ignite C in their degenerate cores before losing their envelopes to mass loss may also produce unusual thermonuclear super- novae. Very little is known about this potential channel, but large surveys should yield valuable information. SSE DISCOVERY AREA: TIME-DOMAIN SURVEYS Astronomical timescales evoke the long stretches of time, reckoned in giga- years, that characterize cosmic expansion and most phases of stellar evolution. For these phenomena, a single comprehensive survey can reveal the essential facts, as in a Hertzsprung-Russell diagram for a cluster. But there are phenomena of rotation and pulsation, of orbiting binaries, of explosions and mass loss, and most spectacu- larly, of stellar death, for which the physical timescales are measured in seconds, days, or months. For a wide range of stellar events, knowledge has been obtained by observing through narrow windows in time, often set by single-investigator observing strategies or by the technical capabilities of the detectors being used. Narrow windows produce limited views. The panel anticipates that in the coming decade the burgeoning technological change due to detector development, fast computers, automated pipelines, and the ability for the entire community to in- teract with large-volume public databases (from a distance, over the Internet) will lead to significant scientific progress in revealing and exploring a wide range of stellar phenomena. For these reasons, time-domain surveys represent a significant discovery potential for the study of stars and stellar evolution. Discovery in the time domain in the next decade will be driven by detectors with large fields of view, which scan the sky with approximately daily-weekly cadence and provide all-sky data sets. In addition to unanticipated discoveries (Figure 5.10), there are expected events; Table 5.2 gives a sample of the wide range

OCR for page 207
Panel rePorts—new worlds, new HorIzons 240 FIGURE 5.10 An unusual optical transient with a nearly symmetric light curve was discovered during the Hubble Space Telescope Cluster Supernova Survey. Most explosive events have a rapid rise and a slower decline; this symmetry is anomalous. From its redshift and comparison with other recent supernova discoveries, the object was found to be an unusual, luminous supernova. Surprising, rare events continue to be discovered as time- domain surveys expand in reach and duration and improve in cadence and precision. SOURCE: K. Barbary, K.S. Dawson, K. Tokita, G. Aldering, R. Amanullah, N.V. Connolly, M. Doi, L. Faccioli, V. Fadeyev, A.S. Fruchter, G. Goldhaber, A. Goobar, A. Gude, X. Huang, Y. Ihara, K. Konishi, M. Kowalski, C. Lidman, J. Meyers, T. Morokuma, P. Nugent, S. Perlmutter, D. Rubin, D. Schlegel, A.L. Spadafora, N. Suzuki, H.K. Swift, N. Takanashi, R.C. Thomas, and N. Yasuda for the Supernova Cosmology Project, Discovery of an unusual optical transient with the Hubble Space Telescope, Astrophysical Journal 690(2):1358, 2009, reproduced by permission of the AAS. Courtesy of Kyle Barbary, University of California, Berkeley, Lawrence Berkeley National Laboratory. of stellar science that can be addressed with this type of time-domain survey, at many wavelengths. Follow-up observations of various types are often essential to carry out the science goals. For example, evolved giants and brown dwarfs observed interferometrically show evidence for time-variable spatial structures, possibly associated with dust-cloud formation and weather-like phenomena; follow-up observations with new interferometric facilities will provide important constraints on the physical mechanisms behind the observed time variations. Other kinds of time-domain studies not mentioned in Table 5.2 will also be valuable—particularly, continuous monitoring observations with high cadence for extended duration on individual objects, as in the case of asteroseismology. This panel’s four science questions all mention time-domain observations as

OCR for page 207
rePort Panel stars stellar evolutIon 241 of tHe on and TABLE 5.2 Time-Domain Surveys, Large Field of View, All-Sky Coverage, Daily-Weekly Cadence Science Survey Capabilities Follow-up Variable Stars and the Sun Starspots and rotation Optical/multicolor Spectroscopy, spectropolarimetry Massive stars—LBVs Optical/IR/UV Spectroscopy, photometry, spectropolarimetry Eclipsing binaries Optical/multicolor, long Photometry, spectroscopy, radial velocities duration Clouds and weather on brown Optical/IR Multiwavelength photometry, spectroscopy; dwarfs improved models Pulsating variables—classical, Optical/IR, long duration, Milky Photometry, spectroscopy; interferometry; rare Way and nearby galaxies, improved models helio- and asteroseismology Rare stages of stellar evolution Optical/IR, long duration, Photometry, spectroscopy; improved models; clusters and nearby galaxies interferometry Stellar mergers on dynamical or Optical/IR, long duration, Photometry, spectroscopy; improved models thermal timescale resolved stellar populations Stellar flares Optical (blue, U/u filter), UV, Time-resolved, multiwavelength photometry, X-ray, radio spectroscopy; improved models Solar corona Optical/IR/radio magnetometry ATST follow-up of energetic events Pulsars—rare types Radio Radio X-ray, γ-ray Magnetar flares Multiwavelength photometry, spectroscopy Variable Accreting Systems CVs, novae Optical, UV Spectroscopy—optical, UV, IR Tidal disruption of stars Optical, UV, X-ray, radio Multiwavelength photometry, spectroscopy; host galaxy properties LMXBs (black-hole/neutron- X-ray, wide-field Time-resolved X-ray photometry, spectroscopy star novae, X-ray bursters, superbursters) Supernovae and GRBs Optical/IR, n, gravitational Milky Way supernova Multiwavelength photometry, spectroscopy; waves, γ-ray, IR, radio rapid response Supernova searches including Optical/IR Multiwavelength photometry, spectroscopy rare forms, optical transients Shock breakout in supernovae UV, X-Ray Multiwavelength photometry, spectroscopy Types II and Ibc Electromagnetic counterparts to Optical/IR, UV, X-ray Photometry, spectroscopy—radio; rapid gravitational wave sources response γ-ray, X-ray Gamma-ray bursts Multiwavelength photometry, spectroscopy; rapid response Orphan afterglows of GRBs Optical/IR, radio Multiwavelength photometry NOTE: Acronyms are defined in Appendix C.

OCR for page 207
Panel rePorts—new worlds, new HorIzons 242 essential to making progress in the next decade (e.g., supernova searches, rotation, episodic mass loss from massive stars, pulsar searches). A few illustrative examples of additional stellar topics from Table 5.2 that have particular discovery potential are described in more detail below. Accretion-Induced Collapse (Rare Type of Supernovae) How often does a white dwarf approaching the Chandrasekhar mass in a bi- nary undergo an accretion-induced collapse (AIC) to form a neutron star, rather than blowing up as an SN Ia? An understanding of this question is essential for understanding the evolution of white dwarfs in binary systems and would dramati- cally constrain the allowed progenitors of SNe Ia. AIC has also been proposed as one of the most promising sites for third-peak r-process nucleosynthesis. AICs are predicted to be accompanied by the ejection of up to approximately 0.01 to 0.03 M◉ of Ni, produced in an accretion disk around the newly formed neutron star. This outflowing Ni produces a short, approximately 1 day, optical/near-IR SN-like transient with a peak luminosity of 1041 to 1042 ergs/s, significantly fainter and of shorter duration than ordinary SNe Ia. Electromagnetic Counterparts to Gravitational Wave Sources In the coming decade, it is likely that transient gravitational wave sources will be discovered by experiments such as Advanced LIGO and VIRGO (a gravitational wave detector at the European Gravitational Observatory), with lower-frequency gravitational wave sources perhaps becoming detectable toward the end of the decade. To optimize the astrophysics that results from such detections, it is criti- cal to have nearly simultaneous electromagnetic observations. Wide-field-of-view cameras are essential given the rather poor localizations provided by gravity-wave detectors (fractions of a square degree). A unique electromagnetic counterpart temporally and spatially coincident with a gravitational wave source would provide more confidence in the gravitational wave detection. Combined gravitational wave and electromagnetic observations could potentially provide detailed information about stellar sources, including neutron-star/neutron-star mergers, black-hole/ black-hole mergers, short gamma-ray bursts, and (perhaps) core-collapse super- novae. This information is unique given the gravitational wave constraints on the masses and spins (magnitudes and direction) of the objects. Rare Stages of Stellar Evolution (He Core Flash) At the tip of the first-ascent RGB, for stars of M < ~2.5 M◉, He ignition in the degenerate core leads to the He core flash. Very little is known observationally

OCR for page 207
rePort Panel stars stellar evolutIon 243 of tHe on and about what happens during the He core flash: for example, does the star lose mass? The ultimate fate of Earth—into the Sun or backing away—depends on whether the Sun will lose 20 percent of its mass before or during this event. Yet one cannot point to a single object in the sky that is currently undergoing an He core flash. The probable duration of this phase is about a thousand years, assuming that the star’s response resembles its response to He shell flashes, evolving on a thermal timescale. One probable observational signature will be erratic pulsation with rapid period changes. Large samples, including stars in clusters of various ages, are needed to identify individual objects in this critical stage of stellar evolution. These data will provide essential input to theoretical models of late-stage stellar evolution. Eclipsing Binaries and Binary Star Evolution Eclipsing binary stars are powerful diagnostics of stellar structure and evolu- tion, and they are relatively easy to find in time-domain surveys (Figure 5.11). They also provide a secure way to measure masses and radii for stars of all spectral types, metallicities, and ages. Longer-period eclipsing binaries, although geometri- cally less favorable, should be found in large surveys with long durations. Such stars will be valuable because their evolution is less likely to be impacted by the presence of the companion (synchronous rotation, enhanced activity), which may influence the stellar radius. Large samples will permit tests of the mass ratios and distributions of orbital separations of close binaries, which will in turn inform theories of binary-star formation and evolution. Studies of interacting binaries will also benefit, particularly for unusual and rare systems such as contact binaries and common-envelope systems, probable precursors to stellar mergers (leading, for example, to blue stragglers). Investigations of white-dwarf/white-dwarf and white-dwarf/massive-star systems are important for understanding the origin of SNe Ia and cataclysmic variables. For very wide eclipsing binaries and stars with large planets, detailed information on the resolved stellar surface can be obtained during the eclipse/transit. Radio Transients Stars of all kinds produce a surprisingly wide variety of nonthermal radio emission from timescales of nanoseconds (the giant pulses from the Crab pulsar), to months (the radio afterglows of supernovae). Recently, several relatively small- scale radio surveys have uncovered new forms of transients from known sources, such as extremely rare millisecond-duration pulses from rotating neutron stars (the so-called RRATs, or rotating radio transients), and bright coherent emission from brown dwarfs. Other surveys have found unidentified radio transients in extraga- lactic blank fields and toward the galactic center. Yet these surveys have covered

OCR for page 207
Panel rePorts—new worlds, new HorIzons 244 1.1 1.1 1.0 1.0 0.9 0.9 f/f0 f/f0 0.8 0.8 0.7 0.7 Ks H 0.6 0.6 -0.2 0.0 0.2 0.4 0.6 -0.2 0.0 0.2 0.4 0.6 orbital phase orbital phase 1.1 1.1 1.0 1.0 0.9 0.9 f/f0 f/f0 0.8 0.8 0.7 0.7 J z 0.6 0.6 -0.2 0.0 0.2 0.4 0.6 -0.2 0.0 0.2 0.4 0.6 orbital phase orbital phase 1.1 1.1 1.0 1.0 0.9 0.9 f/f0 f/f0 0.8 0.8 0.7 0.7 i r 0.6 0.6 -0.2 0.0 0.2 0.4 0.6 -0.2 0.0 0.2 0.4 0.6 orbital phase orbital phase 1.1 2.0 1.0 1.5 0.9 1.0 f/f0 f/f0 u 0.8 0.5 0.7 g 0.6 0.0 -0.2 0.0 0.2 0.4 0.6 -0.2 0.0 0.2 0.4 0.6 orbital phase orbital phase FIGURE 5.11 Multicolor light curves for an eclipsing binary, from Sloan Digital Sky Survey and 2 Micron All Sky Survey data. The two stars are low-mass, M0 + M1, dwarfs with an orbital period of 2.639 days. Time-domain surveys will provide large numbers of new eclipsing binary systems for stars across the Hertzsprung-Russell diagram, allowing for much-improved basic data for unusual as well as common types of stars. Follow-up for systems such as this one—for example to get radial velocities—will require large telescopes and/or substantial telescope time. SOURCE: Reprinted with permission from A.C. Becker, E. Agol, N.M. Silvestri, J.J. Bochanski, C. Laws, A.A. West, G. Basri, V. Belokurov, D.M. Bramich, J.M. Carpenter, P. Challis, et al., Two-Micron All-Sky Survey J01542930+0053266: A new eclipsing M dwarf binary system, Monthly Notices of the Royal Astronomi- cal Society 386:416, 2008, copyright 2008 Royal Astronomical Society.

OCR for page 207
rePort Panel stars stellar evolutIon 245 of tHe on and either only tiny fractions of the sky or a very small range of timescales. The situ- ation is improving with the development of the Low Frequency Array (LOFAR), the Long Wavelength Array, the Murchison Widefield Array, and ATA-42 (the Allen Telescope Array, configured with 42 radio-telescope dishes). As radio fields of view continue to increase and computing capability grows to allow wide-field, rapid- cadence, radio imaging, new surveys will uncover many more transient events of both known and unknown origin. These events have the potential to tell about particle acceleration, stellar magnetic fields and rotation, strong-field gravity, the interstellar and intergalactic media, the violent deaths of stars, and possibly physics beyond the standard model. Summary of SSE Discovery Area In summary, the time domain represents great discovery potential well matched to the timescales that are relevant for stellar phenomena during their lifetimes and their death throes. Astronomers look forward to the next decade as a period of renaissance for stellar astronomy as time information is added to the new advances in three-dimensional spatial resolution and the idealization of a star as a static, spherical object is put to bed.

OCR for page 207