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.
6 Search for Life Outside the Solar System INTRODUCTION Life as we know it is a planetary phenomenon: its origin appears to have required interactions among liquid water, a gaseous atmosphere, and miner- als provided by a solid planetary surface. The energy required to produce the appropriate chemical reactions was available from solar ultraviolet light, bombardment by charged particles, meteoritic impacts, local volcanism, hydrothermal vents, lightning discharges, coronal discharge, and even acous- tic shocks. A nearly circular orbit about a stable star promotes fairly uniform condi- tions for the billions of years required (at least on Earth) for life to evolve from single cells with no nuclei to multicelled intelligent organisms. The evolution and dispersion of life on Earth have radically altered the surface, oceans, and atmosphere of this planet in ways that are discernible from a remote observational vantage point. During the past 1.5 to 2 billion years, a distant observer would have found presumptive evidence for life on Earth in the oxygen-rich nonequilibrium chemistry of the Earth's atmosphere. More recently, the microwave signals generated by human technology could pro- vide that same remote observer with circumstantial evidence for the exis- tence of some form of intelligence on the planet Earth. These two examples of life detection should apply to other planets as well. Isolating the light or thermal radiation of a distant planet from the brilliance of its star would enable us to examine the spectrum of the planet and search for chemical evidence of the existence of life. This evidence would take the form of some massive departure from chemical equilibrium, a large amount of free oxygen being the most obvious example and proba- bly the most observationally traceable. However, the detection of trace gases such as CH4, NH3, N2O (nitrous oxide), and CS2 (carbon disulfide) in 105
106 THE SEARCH FOR LIFE'S ORIGINS excess of the amounts predicted by chemical equilibrium and plausible nonequilibrium sources might also provide presumptive evidence for life if the 4- to 11-,um region of the spectrum could be accessed. Detection of signals generated by an extraterrestrial technology would be even more compelling evidence that life is not uniquely confined to Earth. (Although searches for such signals have become known as SETI [the search for extraterrestrial intelligence], in reality such searches could detect only those intelligent forms that utilize an electromagnetic technology.) In this chap- ter, both of these approaches to the detection of life outside the solar system are discussed. GOAL: To understand the nature and distribution of life in the uni verse. The underlying scientific goal in searching for life beyond the solar system is the same as the goal of the Viking biology experiments on Mars or the study of astrophysical influences on the evolution of life on the planet Earth. As described in previous chapters, there are compelling reasons to look for evidence of the origin of life on Mars during a more clement epoch of that planet's history. On the other hand, the prospects for detecting extant life on Mars seem remote. To find such life, our vision must expand beyond the realm of the known planets and moons. Recent strategy reports from this committee (SSB, 1981, 1986) empha- sized the dynamic and complex interactions between life and its terrestrial environment. Those reports focused on the opportunities afforded by suit- able airborne and orbital platforms for the study of the terrestrial biosphere as a single, complex but closely coupled system. Now it is time to extend this perspective to other planets and satellites in other planetary systems. The same kind of evidence that reveals the presence of life on Earth can, in time, be sought in other planetary systems. If the search is successful, it will mark the beginning of a new science. It will then be possible to examine life on Earth as just one of several examples of a remarkable property of matter, rather than the only example. This chal- lenge was recognized by the authors of the 1981 SSB report: "The emer- gence of complex societies capable of extraterrestrial communication is an evolutionary biological phenomenon. Thus, the problems of the search for, and attempt to communicate with, extraterrestrial life lie at least in part within the province of planetary biology and chemical evolution." APPROACHES TO THE DETECTION OF LIFE OUTSIDE THE SOLAR SYSTEM If we are to attempt to sense the impact of life on distant planets from this remote vantage point, those planets must first be located. At present,
SEARCH FOR LIFE OUTSIDE THE SOLAR SYSTEM 107 there is no unambiguous evidence of the existence of an extrasolar plane- tary system, although there are many tantalizing clues. The detection of even a single example of an extrasolar planet that has been modified by the evolution of life would have extraordinarily profound consequences. It would then be possible to bring to bear the full power of the scientific method in extracting laws of nature from several examples of the same phenomenon. Despite its extraordinary diversity, life on Earth offers just a single example from which it is impossible to generalize to other systems. To escape from this dilemma, those other systems must be found in order to achieve the fundamental goal of understanding the distri- bution of life in the universe. The basic techniques for detecting extrasolar planets are well understood (a detailed discussion of this subject is given in reports of the SSB, 1988a,b). Individual planets, once formed, will have a number of perturbing effects on the motion and apparent brightness of their host star, but these effects will be very small and will require extreme measurement accuracy. For example, the accuracy required to detect Jupiter-mass planets around the nearest stars has been achieved with heroic efforts during the past two decades. Recent technological advances now allow previous accuracies to be obtained more rapidly and make additional detection schemes feasible. Planets with the mass of Neptune may become detectable within the decade, but the detection of Earth-mass planets will require dedicated space-based systems, not yet approved by any funding agency. It is appropriate for the purposes of this report to consider what systems are available now, which are planned for the immediate future, and, especially, what must be studied to ensure that instruments of significantly improved capability become avail- able to the next generation of planet seekers. These instruments must be systematically and exhaustively employed for long periods of time if a census of even the nearest stars is to provide a basic understanding of the frequency of planetary systems. Furthermore, we must be able to extend what is learned from studies of nearby planetary systems to the more distant stars around which we cannot hope to detect planets by these simple tech- n~ques. Programs to search for other planets have been endorsed repeatedly. These endorsements by astrophysicists and planetary scientists are concurred in by scientists interested in studying the origin and evolution of life. In this regard, it should be noted that the desire of exobiologists to examine spec- troscopically the global chemistry of a distant planet imposes the most extreme instrumental requirements. In parallel with this search for passive evidence of the existence of life beyond the solar system, a different kind of search can be carried out: that is, the search for evidence of technologically advanced civilizations. The unambiguous detection of an extrasolar technology would have profound implications. Not only would we know that life in the universe is not
108 THE SEARCH FOR LIFE'S ORIGINS unique to the planet Earth, but we could contemplate the possibility of communicating with, and learning from, the intelligent life that created the detected technology. It is conceivable that the distribution of life in the universe could come to be understood in the larger context of the evolution of the cosmos. These possibilities have previously been endorsed by the National Research Council ( 1972, 19821. Since the 1940s, there has been a rapid growth in the techniques and technology of communications and radio astronomy that can improve the chances of detecting signals at great distances. These tools have not yet been brought to bear SETI in a systematic way. It is now possible, by using state-of-the-art instrumentation, to expand the search domain for extrater- restrial signals far beyond anything that has been done to date. These searches for deliberately generated signals should be complemented by searches for other indirect manifestations of a distant technology. Although the two approaches to searching for life beyond the solar sys- tem can provide very useful inputs to each other, they may be pursued independently and simultaneously at a pace set by the rate of maturation of the requisite technologies. NASA is the leading agency in the development of most, if not all, of the necessary instrumentation. Furthermore, NASA is the agency responsible for providing the orbital platforms required by some of the observational techniques (discussed below). The readiness of some of the requisite technology and the concurrent development of orbital astro- nomical facilities invite rapid implementation of search programs utilizing these capabilities. SCIENTIFIC OBJECTIVES To achieve an understanding of the nature and distribution of life in the universe, a number of discrete scientific objectives must be carried to com- pletion. In many cases, these objectives overlap those already enunciated by the Astrophysics and Solar System Exploration program offices within NASA. However, it is important to remember that the desire of exobiolo- gists to study both the nature and the distribution of life in other potential habitats is likely to place more stringent technical requirements on the rele- vant instrumentation than demanded by other objectives. It is necessary to ask, in a cosmic context, how unique are this solar system, the Earth, and terrestrial life? The precise mechanisms that led to the formation of our solar system, with its small, dense inner planets and its more distant gas giants, are the subject of much debate. Debatable as well is the time scale over which these processes occurred. Consistent with current theories of planetary formation, there does not seem to be anything unique about the protosolar nebula; other nebulae around other stars should also form planets by the same methods, whatever they are. Comparative
SEARCH FOR LIFE OUTSIDE THE SOLAR SYSTEM 109 studies of protostellar nebulae currently in the process of collapse and, perhaps, planet formation could provide a bound on the length of the pro- cess, as well as an estimate of the mass ratio between the flattened nebula and the centrally condensed protostar. The latter quantity is the key to distinguishing among current models of planetary formation. There are a number of Earth-orbital telescopes for infrared and submillimeter observa- tions under design, or in the planning stage, but as currently conceived, none will provide the resolution of 0.01 to 0.1 arc see necessary to observe dimensions corresponding to our planetary system within distant protostel- lar nebulae. The recent study Space Science in the Twenty-First Century (SSB, 1988b) noted this same deficiency and recommended that advanced technology programs be pursued to achieve this accuracy in future systems. Perhaps the closest we can get to the process of planet formation is to study the extended disks of dust and gas surrounding some young stars that have just reached the main sequence. These stars were discovered in the analysis of the IRAS four-color data. A significant infrared excess in the 60-,um band has been correlated with the existence of a dusty disk compo- nent circling the distant star. These disks are typically much larger than our solar system. Whether there is a selection effect favoring large disks, whether the size of the disk is determined by the age of the star alone, and whether planets have formed, are about to form, or have failed to form in these large disks are questions whose answers are still unknown. These are certain to be topics for thorough investigation in the coming years. Studies of these disks and their frequency of occurrence have not yet produced any general agreement among theorists as to how planets form. During the coming decade it may not be possible to determine observa- tionally the processes by which other planetary systems (and by extension our own) formed. It should, however, be possible to verify observationally that other planetary systems have indeed formed. The unambiguous detec- tion of the first extrasolar planet will validate the hypothesis that planetary formation is not uniquely related to the Sun. OBJECTIVE 1: To determine the frequency and morphology of nearby planetary systems. For the purposes of exobiological understanding, a significant planetary census will be required. Only this can provide a meaningful estimate of the frequency of occurrence of terrestrial-mass planets at orbital distances from the primary star that are suitable for the maintenance of surface tempera- tures amenable to life. A survey of nearby stars must be made initially to determine the fre- quency of planetary-system formation in our local galactic neighborhood. This survey must be augmented with a study of distant protoplanetary disks and protostellar nebulae to allow for extrapolation of the frequency of
110 THE SEARCH FOR LIFE'S ORIGINS planetary-system formation throughout the galaxy. Once detected, the distri- bution of extrasolar planet masses as a function of the distance from the parent star (location in the protostellar accretion disk) must be determined and correlated with stellar type to better understand the processes that con- trol the formation of planets and the role of the central condensed protostar. From observations of protostellar nebulae and early stellar disk systems, an evolutionary sequence of events and time scales should be developed to predict the rate at which planets are formed within the Milky Way. The sensitivity of available instrumentation will undoubtedly dictate that these surveys proceed in a stepwise fashion. The detection or (equally signifi- cant) nondetection of Jupiter-mass planets in the vicinity of the nearest few stars and the study of the largest dust disks around young stars will come first. These should be followed by the search for lower-mass planets (down to terrestrial size). Investigation of protoplanetary formation in progress around distant protostars can commence whenever instruments of sufficient angular resolution permit. To implement these investigations properly, advanced technology studies are needed that will lead to a new generation of instruments capable of detecting low-mass planets and investigating solar- system-scale phenomena within protoplanetary disks in nearby regions of star formation. OBJECTIVE 2: To determine the frequency of occurrence of condi- tions suitable to the origin of life. The actual surface temperature of any particular planet will depend upon the abundances and chemical nature of its atmospheric constituents as well as its distance from the host star. Obtaining information on surface tem- perature and the chemical composition of an atmosphere requires direct imaging and spectroscopic analyses of each distant planet body. The tech- nology required for such spectral studies is not yet available and may not be available when a statistically interesting number of extrasolar planets are first recognized. Models for their atmospheric composition will have to be constructed on the basis of comparative planetology (including the major planetary satellites) in our own solar system, together with what is known about the origin of the Earth's current and precursor atmospheres. Predic- tions made on the basis of outgassed and accreted atmospheric constituents must be compared with spectral data as soon as instrumentation permits. The goal is to understand just how often conditions suitable for life occur in the Milky Way galaxy. Once terrestrial-mass planets have been detected, they must be analyzed spectroscopically to assess their atmospheric composition as well as their surface temperatures and pressures. This will provide an estimate of the frequency of occurrence of conditions similar to those that are presumed to have given rise to life on this planet. To appreciate the intrinsic difficulty of this task, it is necessary to remember that we have only been able to
SEARCH FOR LIFE OUTSIDE THE SOLAR SYSTEM 111 accomplish a similar assay of Titan in the past few years and are still waiting for a probe mission to resolve speculations about a liquid organic ocean there. Titan presents a far bigger and brighter target than a distant Earthlike planet and one that can be observed without confusion from the solar luminosity. For these studies, therefore, it is important to continue the development of technologies for supersmooth mirror production, low-light- scattering telescopes, and large-orbital infrared, submillimeter, and optical telescopes that may eventually permit the direct imaging and gross atmo- spheric characterization of distant planets. OBJECTIVE 3: To search for presumptive evidence of life in other planetary systems. The probability that life, once started, will evolve to intelligence depends on many things, one of which may well be the fortunate location of a planetary system within the parent galaxy now or at some past epoch. If evolution toward intelligence everywhere requires billions of years, then changes in the astrophysical environment may play a central role in the process. Extinction events and evolution itself have been episodic on Earth (see Chapter 51. There is strong circumstantial evidence for an astrophysi- cal connection in at least some of these episodes. Changes in the character- istics of the solar system have been most convincingly cited, but changes in our galactic location have also been suggested as causative agents. In the foreseeable future, detection and spectroscopic study of extrasolar planets will be confined to the immediate galactic neighborhood. The pecu- liar velocities of our stellar neighbors will ensure that some of them have sampled far different galactic environments than Earth has in the past bil- lion years. Given a sufficiently large sample of planets examined for signs of life and sufficiently accurate models for galactic dynamics, it might be possible to draw some conclusions regarding the probability of life arising and evolving as a function of galactic locale. Any potentially suitable terrestrial-type planets must be studied in detail to search for signs of nonequilibrium chemical constituents, possibly signi- fying the action of some form of active biological system. Although the overabundance of molecular oxygen in the Earth's atmosphere and its coex- istence with methane is an example drawn from Earth that correlates with life as we know it, life as we do not know it is likely to be no less surpris- ing than the soil chemistry of Mars. The key to a distant atmosphere may lie in the coexistence of two other highly reactive components requiring a source function for which no natural planetological explanation can be found. OBJECTIVE 4: To search for evidence of extraterrestrial technology. Because the instrumentation for detecting evidence of extraterrestrial technology is far more mature than the instrumentation necessary for exam- ining distant planets minutely, another technology (and, by inference, an-
2 THE SEARCH FOR LIFE'S ORIGINS other biology exhibiting intelligence) may be detected before any other evidence is found for extraterrestrial life. The examination of distant plan- ets first requires the identification of such planets, but searches for other technologies can be made in the direction of plausible targets without a priori knowledge of the existence of a suitable planetary abode. Further- more, searches can also be indiscriminate with respect to direction if the technology of another advanced civilization is sufficiently "loud." Thus, several different search strategies, based on different concepts of what con- stitute the most detectable features of a distant technology, may have to be employed in conducting an exhaustive search for signs of extraterrestrial technology. Such searches can attempt to detect purposeful communication signals, whether intended for Earth or for some other receiver. Because such signals are intentional, they can be expected to present a high signal- to-noise ratio for whatever communication scheme is utilized. In this case, it is possible to define what constitutes an exhaustive search. One must also use current terrestrial technology as a paradigm and formulate the sensitivity required to "see" the leakage radiation that is generated on Earth from across the galaxy. In the case of intentional beacons, it is not neces- sary to achieve this extreme sensitivity because intentional beacons could be much brighter. However, it is necessary to define this limit and use it as a standard against which to measure the significance of negative results. Searches can also be conducted to detect the by-products of noncommu- nicative technologies of other civilizations. Intentional amplification is not expected to be imposed on these signals, and it is far more difficult to define what constitutes a definitive search for such evidence. Nevertheless, searches that can achieve some well-defined detection sensitivity should be pursued. MEASUREMENT REQUIREMENTS For the recommendations and scientific objectives outlined in the previ- ous section to be accomplished, certain measurement accuracies must be achieved in a number of different observational technologies. It is appro- priate for this report to compare the required accuracies to those likely to be achieved by various facilities, including mature missions nearing launch and those whose planning is well under way. In those areas where the available accuracies match or exceed the required ones, the research objec- tives of exobiologists can probably be accommodated by the appointment of specialists in this field as interdisciplinary scientists and by widespread distribution of announcements of opportunity. Facilities whose near-term available accuracies fall short of requirements will entail basic technologi- cal development so that subsequent generations of instrumentation will be able to provide the necessary capabilities for exobiological purposes.
SEARCH FOR LIFE OUTSIDE THE SOLAR SYSTEM Detecting Extrasolar Planets 113 An extensive literature exists on the techniques for carrying out searches for extrasolar planetary systems and their current or expected capabilities. Recent discussions may be found in Space Science in the Twenty-First Century (SSB, 1988b). With the exception of direct detection techniques, little has changed since these comprehensive reviews, and this report only briefly summarizes the measurement accuracies required to detect either Jupiter or the Earth in orbit about the Sun if these systems were located at a distance of 10 parsecs (30 light-years). Extrasolar planets may be imaged directly or detected by one of three indirect techniques: astrometry, spectroscopy, and photometry. indirect Detection 1. Astrometric detection: Astrometry uses distant"fixed" stars as a reference frame against which to measure, over long periods of time, the relative position of a candidate star. A star and planet orbiting a common center of mass introduce a barely discernible reflex motion, or "wobble," into the stellar trajectory. The maximum amplitude of the reflex is 0.5 milliarc see (500 microarc see) for Jupiter and the Sun observed at a dis- tance of 10 parsecs, and 0.3 microarc see for the Earth and the Sun at the same distance. Atmospheric turbulence limits the obtainable ground-based measurement accuracy to something in excess of 100 to 300 microarc sec. Although nearby Jupiters can be detected from good ground-based sites, astrometric detection of even the closest Earthlike planet will require a space-based platform. Any instrumental system must possess extreme sta- bility over very long time scales commensurate with planetary orbits, be- cause it is the periodic nature of this reflex that ultimately distinguishes measurement from noise. 2. Spectroscopic detection: The same wobble in the star's motion in- duced by a star and planet orbiting about their common center of mass can be seen spectroscopically if the plane of the orbits is nearly parallel to the line of sight. In this case the change in relative velocity introduced as the stellar candidate moves toward or away from the observer can be seen as a periodic Doppler shift in the absorption line spectrum of the stellar photo- sphere. The effect of Jupiter on the Sun is to cause a maximum shift in velocity of 12 m/s, whereas the maximum effect induced by a terrestrial mass planet is 0.1 m/s. These small velocity shifts must be detected on top of photospheric turbulence and the peculiar velocities of the stars moving through space at tens of kilometers per second with respect to the observer. Except for accuracy limits imposed by the brightness of the star, this tech- nique is independent of the distance to the star. Again, the measurements
4 THE SEARCH FOR LIFE'S ORIGINS demand extreme instrumental stability over long periods of time and a peri- odic signature associated with the velocity shift to distinguish signal from noise. Because the atmosphere does not seriously impede these spectro- scopic studies, they can be conducted from the ground, and several facilities have already achieved instrumental accuracies sufficient to detect the ef- fects of a Jupiter-mass planet. The recent announcement of seven stars that exhibit spectroscopic shifts >10 m/s is extremely exciting. Measurements must be continued over the next decade to identify the underlying periodic- ity before definitive conclusions can be drawn. The limiting factor in this approach is the lack of detailed knowledge about any periodic or quasi- periodic processes in stellar photospheres that might induce Doppler shifts of this order internal to the star. Continued observations of the Sun's whole disk are required to determine the eventual sensitivity limit of this method. 3. Photometric detection: Astrometric and spectroscopic detection schemes have a long history, and both have recently benefited from a new generation of instrumentation; The next generation or two of astronomical instrumentation may enable a third detection scheme. If the orbit of a planetary system is very well aligned with the observer's line of sight, then it is theoretically possible to detect the periodic decrease and slight change in color of the stellar luminosity produced when all or part of the stellar disk is occulted by a planet. This measurement requires extreme accuracy, far beyond current capabilities. Absolute photometry to a precision of 1 ppm in two different color bands is required to conclude reliably that the diminution of stellar luminosity is due to an occultation and not intrinsic fluctuations. The periodic nature of the effect is also a necessary characteristic. Since this method of detection was reviewed (SSB, 1988), ground-based measurements have achieved a photometric accuracy of 2 x 10-5. Observations from Skylab indicate that Jupiter occulting the Sun would be relatively easy to detect from afar but that the Earth would be detectable only in quiet phases of solar activity. Because it is impossible to know in advance the inclination of a planetary system's orbital plane or when an occultation might occur, this technique is intrinsically a statistical one. Many stars must be monitored simultaneously and more or less con- tinuously. Debatable, and rather optimistic, calculations suggest that moni- toring 4000 stars with the requisite photometric accuracy will provide a detection rate of about one planet per month. Because the atmosphere se- verely degrades photometric precision, this detection scheme requires an in- strument in Earth orbit. The double differential photometer required for the job is at least one generation of technology away, but this particular method may be worth pursuing because it is best suited to the detection of the small-orbit, short-period planets within any system. The real utility of this approach may eventually be to search for the inner planets of planetary systems already detected by other methods. In general, all the other meth-
SEARCH FOR LIFE OUTSIDE THE SOLAR SYSTEM 115 oafs work best for low-mass stars and high-mass planets. It is Jupiters that will be found (if anything) in the near future, not other Earths. Direct Detection To image a planet directly it is necessary to resolve its combined internal luminosity and reflected starlight from the overwhelming luminosity of the nearby star. For the Jupiter-Sun system at 10 parsecs this requires being able to achieve a brightness contrast ratio of 2 x 10-9 at visible wavelengths (or 10-4 at infrared wavelengths) just 0.5 arc see away from the star. The Earth-Sun system at 10 parsecs would require a contrast ratio of 2 x 10-'° (or 10-6) at a distance five times closer to the star (0.1 arc see). No system in existence on Earth or planned for orbit comes close to achieving these accuracies. In the near future the only planets to be directly imaged will be massive, in large orbits about low-mass stars, or much closer than 10 par- secs. Since the SSB reviewed the potential for direct imaging and recom- mended consideration of low-light-scattering requirements for future tele- scopes (A Strategy for the Detection and Study of Extrasolar Planetary Materials: 1990-2000, SSB, in press), there have been new studies yield- ing promising results. The combination of a coronagraph to reduce the diffraction of starlight by the telescope and a supersmooth mirror to reduce scattered light holds great promise for the future. One system currently under study is called the Circumstellar Imaging Telescope (CIT). This telescope should greatly reduce diffracted light in the wings of an image of a point source (Airy pattern) over the entire field of view of the instrument. Such an instrument would have to orbit the Earth to get above the detrimen- tal effects of atmospheric viewing. A 2-m-diameter instrument in Earth orbit would allow the direct detection of a Jupiter-sized planet around a solar-type star out to a distance of approximately 10 parsecs. A larger in- strument 10 m in diameter would be required to detect the Earth out to these distances, and the technology for such a large mirror has not been demon- strated. In the CIT, diffraction is controlled by a Lyot coronagraph. Combining a Lyot coronagraph with anodizing occulting masks in the first focal plane can probably reduce the contribution from diffracted starlight by a factor of 1000. However, this factor of improvement in diffracted light can be util- ized only if the scattering due to figure error in the primary mirror and to surface dust or scratches is at least 1000 times less than the diffraction of a conventional mirror. Supersmooth mirrors have been manufactured for use in the fabrication of microelectronics. These mirrors (0.5-m-diameter spherical mirrors) have been characterized as approximately five times smoother than the Hubble
116 THE SEARCH FOR LIFE'S ORIGINS Space Telescope mirror. During the fabrication of supersmooth mirrors, figure errors decrease monotonically until the desired specification is reached and then polishing stops. The metrology used to measure the figure of the mirrors can go beyond the current specifications, and there appears to be no inherent reason why much smoother mirrors could not be fabricated. This promising technology should be pursued in coming years. Studying Nebulae and Disks The study of protoplanetary nebulae and young stellar disk systems also requires direct imaging techniques. Estimates of the required measurement accuracies for studying solar-system-scale objects at 10 parsecs or at 140 parsecs (Taurus molecular cloud star-forming region) follow. To resolve a linear dimension of 1 AU at 10 parsecs requires a resolution of 0.1 arc see, whereas the diameter of our solar system (100 AU) can be resolved with a resolution of 10 arc sec. In the Taurus cloud, the corresponding resolu- tions become 0.007 arc see and 0.7 arc see, and resolving something even as large as a dusty protoplanetary nebula for a 1-solar-mass star in the Taurus cloud still requires a spatial resolution of 80 arc see (1.33 arc min). To study dusty disks around young stars or protoplanetary nebulae in the process of collapse requires more than just spatial resolution; spectral reso- lution and good sensitivity are necessary. Resolving powers (~//~) as high as 105 may be needed to use far-infrared, submillimeter, and millimeter molecular lines as tracers of the kinematics and chemistry of the collapsing nebulae. To locate the inner cutoff (if any) of a dust disk around a young star and to characterize the particle size distribution and density as a func- tion of radius require the ability to suppress the luminosity of the central star, as in the case of individual direct planet detection. Once again, corona- graphs can be used, but their physical size will ultimately determine how close to the star the measurements can be made. Ground-based interfer- ometers at millimeter, submillimeter, and infrared wavelengths may prove to be the instruments of choice to study the radial dependence of the contin- uum emission from the dust. Long baselines, large collecting area ele- ments, or many elements are required for sensitivity. Spectroscopic Analyses of Planetary Atmospheres To achieve the most ambitious objectives of remotely detecting, charac- terizing, and searching for evidence of life on an extrasolar planet, direct light from the planet must be examined without contamination from the light of the parent star. Only direct methods of planet detection are suitable for obtaining these observations. In the future, large systems employing coronagraphs, anodization, and supersmooth mirrors with diameters 210 m
SEARCH FOR LIFE OUTSIDE THE SOLAR SYSTEM 117 may allow full characterization of the temperature, pressure, atmospheric composition, etc., of extrasolar planets. These large-aperture instruments may be monolithic but will most likely be composed of multiple phased elements. Therefore, additional technologies involved in stabilization of the point spread function of a multiple-element telescope will be needed. An alternative to a very large-aperture telescope has recently been sug- gested for imaging a distant Earthlike planet and spectroscopically analyzing its atmosphere in search of oxygen and ozone: an optical aperture-synthesis interferometer. In particular, a 25-element array of 2.8-m anodized mirrors with surface accuracies comparable to the Hubble Space Telescope might be able to achieve sub-arc-second resolution sufficient to isolate planet and star. The stellar light can be satisfactorily suppressed by the combined ef- fects of anodization and precise positioning of the star in a null of the inter- ferometer. Very preliminary calculations suggest that 60 hours of integration time would be required for a 10-sigma detection of the 7600-5 oxygen A- band absorption feature for an Earthlike planet at the distance of Tau Ceti (~1 parsec). If the distance is increased to 10 parsecs, the required integra- tion time increases to 300 hours. Whether such a scheme is feasible will depend ultimately on the accuracy with which the elements of the array can be maintained in relation to one another so that the star can be kept pre- cisely at the interferometer null and the signal phase can be maintained. The positional requirements are extraordinary, far in excess of today's capabilities. Unless they are fortuitously close, the direct imaging and spectroscopic analysis of Earthlike planets are not likely to occur within the time scale envisaged in this report. Continued development and research are required to ascertain whether very large smooth mirrors or extreme positional accu- racy in space can actually be demonstrated by the next generation of tech- nology. The search for primitive life may not yet be timely, but the devel- opment of technology to meet required measurement accuracies is. The problem of finding life beyond the solar system may become more tractable if there exist extraterrestrial technologies engaged in activities that can be detected remotely. Searching for Extraterrestrial Technologies Technologies That Generate Electromagnetic Signals One very practical test of the idea that intelligent life exists beyond our solar system is based on the postulate that other technologies have transmit- ted (either deliberately or unintentionally) electromagnetic signals that can be received and recognized with extant technology here on Earth. In 1959, it was proposed that transmissions in the neighborhood of the spectral line
118 THE SEARCH FOR LIFE'S ORIGINS of neutral hydrogen (1420 MHz) might be a means by which extraterrestrial technologies communicate with each other over interstellar distances. More than two decades of scientific debate and review have expanded this idea into a plan to systematically search through the terrestrial microwave win- dow for signals originating from an extraterrestrial source of intelligence. The plan calls for the use of existing radio telescopes, mature microwave technology, and very large special-purpose multichannel spectrum analyz- ers and signal-processing systems to carry out a promising set of explora- tory search strategies. The entire sky is to be scanned with moderate sensi- tivity over the frequency range of 1 to 10 GHz. A set of about 800 nearby solar-type stars will be targeted for much more sensitive searches over the 1- to 3-GHz frequency range. The types of signals sought are those be- lieved never produced by any natural process; they are compressed in fre- quency and perhaps in time as well. If implemented soon, this plan (which was recommended by the Astronomy Survey Committee in Astronomy and Astrophysics for the 1980s, Volume 1, National Research Council, 1982) should occupy most of the decade to be covered by this document. The measurement accuracy appropriate for such a search strategy may be determined by establishing the sensitivity required to detect the artificial microwave signals generated by current terrestrial technology, if the planet Earth was assumed to be located across the Milky Way galaxy from us. Since we do not deliberately transmit signals intended for communication with another technology, the Earth model must consist of signals generated for our own purposes that leak into interstellar space. The weakest, but by far the most numerous, signals are the narrowband carriers for radio and television broadcasts; these are rated at 106 to 107 W of effective isotropic radiated power (EIRP). The single strongest, though very infrequent, signal transmitted is the Arecibo planetary radar at 10~3 W EIRP. If these trans- mitters were located across the galaxy at a distance of 20 kiloparsecs (60,000 light-years), detection would require a sensitivity ranging from 10-37 to 10-3° W/m2 for signals with EIRP ranging from 106 to 10~3 W. For comparison, a typical molecular line survey made by radio astronomers achieves a sensi- tivity of 10-2° W/m2. The planned NASA Sky Survey will achieve 10-23 W/ m2 and, at its most sensitive, the planned NASA Targeted Search will achieve 10-27 W/m2. Successful detection by this planned search will require the existence of extraterrestrial transmitters that are more powerful (perhaps intentional) or closer than the other side of the Milky Way. Although the planned microwave search will not conclude until nearly the end of the next decade, advanced planning for follow-on searches should be conducted concurrently. The planned microwave search may fail to detect any signals, either because the strategy is flawed or because the sensitivity and coverage of parameter space were inadequate. The micro- wave region of the spectrum is "preferred" for such signal detection be-
SEARCH FOR LIFE OUTSIDE THE SOLAR SYSTEM 119 cause the natural astrophysical background radiation is least at those fre- quencies. The naturally quiet microwave window extends to at least 100 GHz, and signals from orbital transmitters may occupy the upper end of the window. The planned microwave search is restricted to the lower portion of the window by the increased atmospheric noise inherent to any terrestrial ground-based search. Future searches will require access to space to extend the search to higher frequencies and to escape the increasing interference generated by terrestrial communication technology. Conclusions concern- ing the best possible signal-to-natural-noise ratio as a function of frequency (if space-based transmitters and receivers are assumed) depend upon scaling laws for the construction of orbital structures. Under certain scaling laws, the infrared appears to provide the best signal-to-noise ratio; under others, the microwave region is still preferred. Experience with on-orbit construc- tion in the coming decades will allow an empirical determination of how the size of an antenna scales with wavelength. Therefore, it is essential to consider and review other plausible strategies for the detection of electro- magnetic signals from technologies that generate them and to support devel- opment of those that are worthwhile, should they be required. Technologies That Do Not Generate Electromagnetic Signals Even a highly technological civilization might not deliberately generate signals that are detectable over interstellar distances. The question is then whether the technology of such a civilization could be detected in some other way. It is impossible to extrapolate the likely progress of technology, here or elsewhere, with any degree of certainty. Although speculative, however, it is appropriate to consider those technological activities in which we now engage and ask how they might appear from afar if they were increased in scale and intensity as some scientists and engineers have pro- posed. Energy production or transformation will probably be an important con- cern for any advanced technology. Collection of stellar radiation on a planetary-system scale, large-scale stellar disposal of fissionable waste, and tritium leakage from orbital fusion plants might produce anomalous radia- tion in the infrared, in the optical lines associated with rare earth elements, and at microwave frequencies, respectively. Thus, there could be observ- able signatures of energy production by advanced civilizations, but it is not possible to make any quantitative predictions. However, in the event of unexpected or unusual observational results during the pursuit of more tra- ditional research, such explanations should be kept in mind if known astro- physical causes cannot be found easily. No separate or discrete efforts seem to be indicated in this line of research. It is important to lend support to the development and deployment of new and more capable infrared in-
120 THE SEARCH FOR LIFE'S ORIGINS strumentation for the study of protoplanetary nebulae and dusty disks around young stars. Continuation of the classical study of stellar spectra and im- proved instrumentation to make the process more efficient are important in their own right. Detection of another technology might be a serendipitous bonus. Interstellar travel and colonization, if undertaken by a distant technol- ogy, might well have extraordinary observational consequences. However, because this takes us beyond the realm of extrapolation of our current tech- nology, it is not appropriate for this report. Available Measurement Accuracies Now that the measurement accuracies required have been derived to de- tect extrasolar planets, spectroscopically scan their atmospheres, and detect evidence of extraterrestrial electromagnetic technologies, it is necessary to compare them with what is likely to be available in the near future. Tables 6.1 through 6.4 provide a summary of existing instrumental capa- bilities and possible future prospects for the four methods of extrasolar planetary detection discussed above. Table 6.5 is a summary of the obser- vational capabilities for NASA's proposed Microwave Observing Project. Innovations, breakthroughs, and perhaps completely new technologies are required to directly image terrestrial planets, perform chemical assays TABLE 6.1 Instruments for the Detection of Extrasolar Planets: Astrometric Methods Resolution Telescope/Spacecraft Launch Instrument (arc see) Allegheny Observatory Now MAP 10-3 Lick Observatory Now Image detector 10-3 Hubble Space Telescope 1990 WFC 10-3 FGS 1 0-3 Dedicated new ? MAP 10-4 astrometric telescope on Mauna Kea Imaging astrometric ? MAP 10-4 free flyer ATE on Space Station 1995 MAP 10-5 Free-flying astrometric ? Moving grating 10-6 interferometer NOTE: ATE = astrometric telescope facility; FGS = fine guidance sensor; MAP = multichannel astrometric photometer; WFC = wide-field camera.
SEARCH FOR LIFE OUTSIDE THE SOLAR SYSTEM TABLE 6.2 Spectroscopic Detection 121 Telescope/Spacecraft Launch Instrument AV (m/s) Ground-based facilities (6) spectrometer Now Radial velocity 10 TABLE 6.3 Photometric Detection Telescope/Spacecraft Launch Instrument /~\lk Space Station 1999 Double 10-5 Block II differential spectrometer TABLE 6.4 Direct Detection Resolution Telescope/Spacecraft Launch Instrument (arc see) Contrast Ground-based Now IR CCDs 1-2 5 x 10-2 speckle and (Keck coronagraphs in 1991) 0.5 10-3 Hubble Space Telescope 1989 FOS and 1 2 x 10-7 coronagraph Hubble Space Telescope, 1994 NICMOS or 0.2 10-5 2nd generation HIMS 0.1 10-5 ISO 1992 ISOCAM 6 10-3 SIRTF 1995 MIPS 6 10-3 CIT 1995 Coronagraph 1 10-9 FIRST ? 1 10-3 LDR ? 1 10-3 Dedicated IR ? 0.5 10-4 interferometer Optical ? 0.3 10~~° interferometer NOTE: CCD = charge coupled device; CIT = Circumstellar Imaging Telescope; FIRST = Far-Infrared Space Telescope; FOS = faint-object spectrograph; HIMS = Hubble imaging Michelson spectrometer; IR = infrared; ISO = Infrared Space Ob- servatory; ISOCAM = Infrared Space Observatory camera; LDR = large deployable reflector; MIPS = multiband imaging photometer; NICMOS = near-infrared camera and multiobject spectrometer; SIRTF = Space Infrared Telescope Facility.
22 THE SEARCH FOR LIFE'S ORIGINS TABLE 6.5 Search for Technological Civilizations That Generate Electromagnetic Signals Dual Mode Search Strategy for NASA's Microwave Observing Project Targeted Search 800 nearby solar-type stars Up to 1000 s of observation per star at each frequency band 1- to 3-GHz frequency coverage Dual circular polarization High resolution and sensitivity Sensitivity to a wide variety of signals pulsed and drifting continuous wave (COO) On-line RFI subsystem 120 million channels (1-, 2-, 4-, 8-, and 16-Hz and 74-kHz resolution band- widths) Sky Survey All-sky coverage 0.3 to 3 s of observation per beam at each frequency band 1- to 110-GHz frequency coverage Dual circular polarization Moderate resolution and sensitivity Primary sensitivity to CW signals RFI data base 15 million channels (30-Hz and 74-kHz resolution bandwidths) thereon, search for leakage radiation throughout the galaxy, and move the search for extraterrestrial technologies to higher frequencies or to other search concepts. Advances will be needed in technologies required for interference pro- tection of a large dedicated SETI facility in high Earth orbit or on the lunar farside; for large-scale optical, infrared, and submillimeter arrays in Earth orbit or on the lunar farside for direct imaging and spectroscopic examina- tion of extrasolar planets and protoplanetary nebulae; and for advanced data-processing techniques.