Noao Long Range Plan Fy 1991

Transcript

National Optical Astronomy Observatories National Optical Astronomy Observatories LONG RANGE PLAN FY 1991 - 1995 March 26,1990 TABLE OF CONTENTS Executive Summary i I. Introduction and Plan Overview 1 n. Science at CTIO and KPNO 3 A. The Large-Scale Structure of the Universe 3 B. The Formation and Evolution of Galaxies C. Stellar Structure and Evolution D. Star-Formation 4 6 7 m. Initiatives: KPNO and CTIO 9 A. NOAO Goals for Large Telescopes B. Beyond 8-m Telescopes C. 4-m Projects 9 18 19 IV. Telescope Technology Program (TTP) 23 V. Observatory Operations: CTIO and KPNO A. Cerro Tololo Inter-American Observatory B. Kitt Peak National Observatory 24 24 27 C. Instrumentation for KPNO and CTIO 29 VI. Science at NSO 50 A. Internal Dynamics B. Magneto-Convection 50 51 Initiatives for Solar Astronomy A. Large Earth-based Solar Telescope (LEST) 55 55 B. SYNOP 57 Vm. Global Oscillation Network Group (GONG) Project 60 IX. Solar Observatory Operations A. NSO Operations 65 65 B. Instrumentation for NSO 69 NOAO Operations 76 VTJ. X. A. XI. Scientific Staff 76 B. Computer Support 77 C. 80 Facilities Maintenance Budget A. NOAO Observatory Operations B. Initiatives C. Future Telescope Technology D. Global Oscillation Network Group E. Management Fee G. Tables 85 85 85 85 . 85 85 86 EXECUTIVE SUMMARY The next decade will present unparalleled opportunities for making qualitatively new kinds of astronomical observations. With the launch of the Hubble Space Telescope (HST), the Gamma Ray Observatory (GRO), and later in the decade the Advanced X-ray Facility (AXAF) and the Space Infrared Telescope Facility (SIRTF), virtually the entire electromagnetic spectrum will become accessible. These new capabilities will place strong demands on ground-based facilities for supporting observations, including particularly spectroscopy. In order to respond to these new requirements, NOAO plans to carry out a complete revitalization of its observing facilities during the next ten years. This plan includes the construction of two 8-m optical/infrared telescopes and the Large Earth-based Solar Telescope (LEST). The 8-m telescopes will be used to attack a broad range of astrophysical problems, but two areas of research that will be enabled by these telescopes are the study of the evolution of galaxies in the early universe and the evolution of protoplanetary disks around young stellar objects. LEST will provide the high photon flux and low instrumental polarization needed to study fundamental astrophysical processes with the detail that only the Sun allows. The telescope will be equipped with adaptive optics to attain angular resolution approaching 0.1 arcsec on a regular basis. Both the 8-m telescopes and LEST will be built in partnership with other groups. All aspects of NOAO's plan for the next five years are designed to further the construction of these major facilities. In nighttime astronomy, the instrumentation programs for both KPNO and CTIO emphasize building instruments for the 4-m telescopes that can serve as prototypes of the instrumentation planned for the 8-m telescopes. The emphasis in the instrumentation program for existing telescopes at CTIO and KPNO will be on developing multiple-object spectroscopy, mosaicing of CCDs for wide-field imaging, providing infrared spectrometers with a range of resolutions, enhancing infrared imaging capabilities, and beginning work on 10 urn arrays. There is also a plan to develop and apply to near-term projects some of the technologies required for building 8-m telescopes. The recently completed renovation of the 2.1-m telescope at Kitt Peak made use of new concepts for mirror support and telescope control systems. Work on the 3.5-m borosilicate mirror will result in the technology necessary to produce 0.25 arcsec images. Critical technical issues include optical testing techniques, construction of an active mirror support system, and design and implementation of a thermal control system. The building of the WIYN telescope to house this mirror will provide experience in telescope construction. There are plans to place at least one additional 4-m class telescope at CTIO as well. If funding is provided at a rate that is consistent with the most rapid progress on technical issues, then the nighttime program would develop as follows: NOAO Nighttime Schedule 1991 1992 1993 1993 1995 1997 1999 Start construction - first 8-m telescope Complete polishing 3.5-m mirror First light - WIYN Start construction - second 8-m telescope First light - CTIO 4-m First light - first 8-m telescope First light - second 8-m telescope The program of the National Solar Observatory over the next several years will focus on two central themes-exploring the interior of the Sun through the techniques of helioseismology and understanding the structure and evolution of convection and magnetic fields. Helioseismology will be addressed by the GONG project Magneto-convection will be attacked by developing techniques for high spatial resolution imaging of the Sun. Key elements of this program are development of adaptive optics for the Vacuum Tower Telescope at Sacramento Peak, participation in the international effort to build the LEST, and exploitation of new diagnostics in the infrared. Key milestones in the solar program are as follows: NSO Schedule 1990 GONG: start construction 1991 Advanced Stokes Polarimeter and Near-Infrared Magnetograph: on-line 1992 1993 Adaptive Mirror: on-line GONG: network operational 1993 LEST: start construction 1997 LEST: first light In addition to building major new facilities, NOAO will continue to provide telescopes for use by the commumty. This plan outlines schedules for fabrication of instruments, acquisition of new computers, and maintenance of existing facilities. Progress in these areas requires an increment to the budget of $1,000,000 for acquisition of detectors and computers and for outside contracts for major repairs to buildings and domes. Operations of existing telescopes will be modified to meet the requirements of new types of research programs planned for the coming decade. Queue scheduling, service observing, and archiving will increase the throughput of existing telescopes and make the data more accessible to the community, as well as making it possible to carry out large-scale programs, synoptic programs, and programs that require special observing conditions. li I. INTRODUCTION AND PLAN OVERVIEW The three decades that have passed since the founding of AURA have been a remarkable period in astronomy. Quasars, pulsars, and the cosmic background radiation were all discovered since 1960. Access to space has led to the opening of the gamma-ray and X-ray windows. Infrared astronomy, first on the ground and then in space with the IRAS satellite, has revolutionized our view of starformation, of the structure of the interstellar medium, and of the energetics of active galaxies. There have been voyages of discovery to the planets and to Halley's comet With the measurements of solar neutrinos and oscillations, astronomers have developed diagnostic tools that allow them to probe the interior structure of the Sun. During the next decade and a half we can expect to see the launch of the Hubble Space Telescope and of long-lived observatories for gamma-ray, X-ray, infrared, and solar astronomy. These facilities, combined with immensely more powerful ground-based telescopes, will provide the tools necessary to resolve, or at least to begin to resolve, the fundamental questions raised by the many discoveries during the past 30 years. What is the large-scale distribution of matter in the Universe? To what extent does luminous material trace the distribution of matter? Did galaxies all form at about the same time, or are there young galaxies relatively nearby? Once formed, do galaxies evolve at a uniform rate, or does the pace of evolution depend on local conditions? What is the nature of the engine that powers the emission from quasars and active galaxies? What triggers star-formation? What determines the initial mass function, the frequency of binary and multiple systems, and the conditions under which planetary systems result? How well do models represent the detailed structure of the Sun, and to what extent must stellar models be modified to reflect the detailed knowledge now being acquired from solar observations? Ground-based observations will continue to play a central role in astronomy. The high density of spectral information in the optical and infrared regions of the spectrum makes these often the wavelengths of choice for analyzing the dynamics, physical conditions, and compositions of astronomical objects. Increasingly, however, the objects studied are selected from surveys carried out at very short or very long wavelengths. These objects may be quite faint in the optical and groundbased infrared, and so the demand for larger telescopes and more powerful instrumentation will only accelerate. The opportunities in astronomy now available have not been matched at any time in the history of the field. Advances in observing techniques and the feasibility of building much larger telescopes with much better image quality for both solar and nighttime astronomy are stimulating innovative thinking and attracting some of the most talented scientists of our generation. New regimes of high spatial resolution and new precision in spectroscopy can be attained. It has become possible to attack qualitatively new kinds of problems ranging from the interior structure of the Sun and stars to the formation and evolution of galaxies. NOAO has developed a plan that would permit it to meet the needs of the observing community for new capabilities, and this document describes that plan. For nighttime astronomy, the key program is the building of two 8-m telescopes, one in Chile and the other on Mauna Kea. A proposal for this initiative was submitted to the NSF in September, 1989. There are also plans to work with university consortia to build and operate a 3.5-m telescope on Kitt Peak and at least one additional 4-m class telescope in Chile. In solar astronomy, the GONG project continues to make good progress in developing the prototype system, but it is now essential that funding be provided to purchase the hardware required to build six complete observing stations. The major long range instrumentation goal for the solar program is to achieve new capabilities for high spatial resolution studies of the Sun, and NSO has joined an international consortium that is exploring the possibility of building the Large Earth-based Solar Telescope (LEST). In order to make it easier to follow the programs for solar and nighttime astronomy in their entirety, plans for each are presented separately. For each of these two areas of NOAO activities, we outline first some of the key scientific opportunities for the next several years. It is this analysis that forms the basis for the specific plan presented in this document The succeeding sections describe the initiatives in each area-8-m and 4-m telescopes for the nighttime program, LEST and stellar synoptic studies within NSO. The plan then describes the status of ongoing initiatives-work on borosilicate glass mirrors and GONG. The following portions of the plan describe the observatory operations within each of the divisions of NOAO and describe how the instrumentation at each site will evolve. The subsequent sections discuss programs relevant to NOAO as a whole, including issues relating to the scientific staff, to development of computing and archiving capabilities, and to maintenance of facilities. The final section of the plan presents the budget required to carry out the proposed programs. n. SCIENCE AT CTIO AND KPNO Astronomical research is the basic function of the NOAO. New and fundamental insights into the nature of the Universe are its "end product" The scope of the research is extremely broad, and over 9,600 scientific papers were published between 1961 and 1989 at NOAO. In this plan, we can only summarize key results in a few particularly active areas of research and indicate where we think the promise of rapid progress is the greatest. It is, of course, this assessment of the future course of our science that has determined our priorities for new instrumentation and for major initiatives. A. The Large-Scale Structure of the Universe. The earliest cosmologies began with the assumption that the Universe is homogeneous and isotropic, and this assumption has persisted until the present, in that current theories posit homogeneity and isotropy when averages are taken over a large-scale. In recent years, however, observations taken with NOAO and other telescopes have begun to pose serious challenges to current cosmological models. In 1983 R. Kirshner (Qr. for Astrophysics) and his collaborators used the KPNO telescopes to identify what appeared to be a gigantic void, or absence of galaxies, in the direction of the constellation Bootes. In 1986, this same group used the KPNO facilities to confirm the existence of this void, which occupies a volume of over one million cubic megaparsecs and is roughly spherical in shape with a diameter of about 120 megaparsecs. Also in 1986 V. de Lapparent, M. Geller, and J. Huchra (Qr. for Astrophysics) published results from the extended Center for Astrophysics redshift survey, which showed galaxies residing on the surfaces of contiguous bubble-like structures whose diameters are typically 25 megaparsecs with a maximum of 50 megaparsecs. A similar picture of the distribution of galaxies in space emerges from a 21-cm survey of 2,700 galaxies published in 1986 by M. Haynes (Cornell U.) and R. Giovanelli (Arecibo Obs.). Finally, a group of seven investigators (D. Lynden-Bell, S. Faber, D. Burstein, R. Davies, A. Dressier, R. Terlevich, and G. Wegner) have used KPNO and other telescopes to discover streaming motion of galaxies on an extremely large-scale. Their analysis of the motion of 400 elliptical galaxies has revealed a net streaming motion of galaxies over a region about 100 megaparsecs in size, with a velocity at the Sun of roughly 570 km/s over and above the uniform Hubble flow. The existence of a "Great Attractor" has been confirmed recently, and the mass associated with it is about 5 x 1016 Mq. Its center is located in the direction of the constellation Centaurus. All of these large-scale features in the galaxy population pose severe challenges to current cosmological models. The difficulties lie in the inability of the models to produce such large structures by purely gravitational means. The hot dark matter models use the hot particles to suppress excessive small-scale fluctuations, but they do not reproduce the galaxy-galaxy correlation function, and they have difficulty producing the giant mass fluctuation required to explain the large-scale streaming motions. Cold dark matter models cannot produce the large structures, and they yield too much small-scale structure unless ad hoc assumptions such as biased galaxy formation are introduced. Non-gravitational theories using shock waves to initiate galaxy formation may succeed in accounting for the observed structure, but they require as yet unobserved explosive events of enormous energy-about 1065 ergs per event, which is equivalent to the energy radiated by 10,000 galaxies over the age of the Universe. The dilemma faced by current theories has caused some consternation, particularly because the gravitational models have appealing features from the standpoint of high energy particle physics. It is clear that the continuing and widespread interest in cosmology and its implications for the large-scale structure of the Universe will motivate many programs in observational astronomy during the next five years. Further work is required to determine just how empty the voids really are, and it is also necessary to search for any gaseous intergalactic matter in the voids. How common is the large-scale streaming motion? Additional and more distant surveys need to be carried out, and one is currently being planned that uses rich Abell clusters of galaxies. The possibility of using another velocity independent distance indicator in the infrared is also being explored. A crucial element in cosmological models is the evolution of structure with time, and because galaxies are too faint to be detected at the relevant lookback times, surveys of quasar distributions are of great importance. Confirmation of these large-scale structures and a determination of their evolution in time will not only constrain cosmological models but will also lend insight into the effects of possible non-baryonic or "dark" matter components in the Universe. Moreover, detection of non-random features in the initial perturbation spectrum could indicate the presence of cosmic strings. Resolution of all of these issues requires a large database, which must be acquired in a systematic and self-consistent manner. Some programs have been started in this area, and more are planned in the future. In all of these programs, the new telescopes and instruments planned for use at the NOAO will be essential to their success. The planned construction of a 3.5-m telescope on Kitt Peak in cooperation with university partners will allow long-term, dedicated programs to be carried out. One such program is a massive redshift survey, which will supply detailed dynamical data over a sample large enough to address these problems. Further into the future, the large collecting area of 8-m telescopes, when coupled with the speed of fiber-fed multi-object spectrometers, results in an improvement of more than two orders of magnitude over current facilities. This capability will allow the study of faint quasars and clusters of galaxies at redshifts between one and three, which is an essential period in the study of the evolution of large-scale structure. B. The Formation and Evolution of Galaxies. The simplest view of galaxy formation would entail some epoch when galaxies formed their constituent stars, with all subsequent evolutionary effects being the result of the relatively slow and continuous process of stellar aging. Until quite recently this view was the prevalent one. Observations taken over the past several years, particulariy at KPNO and CTIO, have shown the need to modify this view. It is important to note that many of these observations simply could not have been made without the current generation of sophisticated solid state detectors and their accompanying instrumentation. One of the first indications of a more complex picture was revealed 10 years ago when H. Butcher (Kapteyn Obs.) and G. Oemler (Yale U.), using the KPNO telescopes, discovered that rich clusters of galaxies appear bluer as their distance increases. In an observing program spanning several years, Butcher and Oemler discovered that nearby rich clusters tend to be populated with old, quiescent elliptical galaxies, yet these observations suggested that at earlier epochs such galaxies were undergoing significant amounts of star-formation. The clusters surveyed extended out to a redshift of approximately 0.4, which corresponds to looking back approximately one quarter of the age of the Universe. This result which has since been confirmed, indicates a much more rapid evolution of elliptical galaxies in clusters than had been previously thought. A second set of relevant observations comes from a survey of the galaxies associated with strong radio sources. H. Spinrad (UC Berkeley) and S. Djorgovski (Ctr. for Astrophys.), using the KPNO telescopes, have obtained data on some of the most distant galaxies observed. Looking back to times half the age of the Universe or less, they find that these radio galaxies contain a great deal of hot, ionized gas which is in very rapid, turbulent motion. The galaxies themselves often appear distorted or moderately disrupted. This work has been confirmed and expanded by P. McCarthy (Mt Wilson) who has found this effect in a larger, more complete sample of radio galaxies via observations with the KPNO 4-m telescope. By way of contrast similar strong radio sources nearby are found to be associated with old elliptical galaxies, which contain very little gas and are regular and undisturbed in appearance. Hence there is strong evidence for major changes with time for this special class of galaxy. Further evidence for evolution was found by A. Tyson (Bell Labs.) in examining the environment of quasars at two different epochs. The CTIO and KPNO telescopes were used to observe quasars in the redshift ranges 0.1 - 0.5 and 1.0 - 1.5. Tyson found that the more distant quasars had 10 times more galaxies around them than did the nearby sample, indicating a drastic change in luminosity for these galaxies with time. Another, unbiased, sample was obtained by Tyson and P. Seitzer (CTIO) with deep CCD imaging. This survey found that galaxies at half the age of the Universe are consistently bluer and thus have much more active star-formation than at the current epoch. Perhaps the most striking case has been the recent discovery of a very large (100 kpc) cloud of ionized gas at a redshift of 1.82, which looks back to two-thirds the age of the Universe. No stellar population is found, though a radio source is present. The data are consistent with this being a galaxy in the process of formation. A final complication emerges from the work of D. Hamilton (CTIO, now at Caltech), which shows there to be a very old, unevolving population of red galaxies, which have basically been unchanged for about half the age of the Universe. In addition, S. Lilly (Hawaii) has shown that the rapidly evolving radio galaxies at large redshift may also have an underlying stellar population that is very old, consistent with an epoch of formation at a redshift of about five. This picture has been made more complex by very recent results from the KPNO two-dimensional infrared array. These K-band data indicate that the old stellar population in these galaxies may be elongated along the axis of the radio source. If this is found to be a general feature of these distant radio galaxies then models for star-formation in them will have to be significantly revised. Thus recent observations show that galaxy formation was far from coeval, that galaxies may have been forming throughout the age of the Universe, and that many, but not all, have undergone dramatic evolutionary changes in that time. The possible nature of these evolutionary forces is also becoming clearer through recent work at the NOAO and elsewhere. In many cases, those galaxies that show strong evidence for evolution are found to be in special circumstances; i.e., in clusters of galaxies, near quasi-stellar objects, or associated with strong radio sources. Membership in a cluster of galaxies provides several mechanisms that can perturb a galaxy and thus change the history of its starformation. Near encounters with other galaxies, stripping of gas by ram pressure of the intergalactic medium, or conversely accretion of such gas by cooling flows onto a massive central galaxy are all possibilities. It is by no means clear which of these processes is relevant, and much further work needs to be done. That the proximity of quasars to galaxies may be important was implied by the survey of Tyson. In addition R. Green (KPNO) and H. Yee (U. de Montreal) have used the facilities at NOAO and the Canada-France-Hawaii Telescope to show that the clusters of galaxies associated with quasars become much richer beyond a redshift of about 0.5. The exact nature of this interaction is again unclear, but the reality of the effect is not T. Heckman (U. of Maryland) and his collaborators have used NOAO telescopes to determine that galaxies associated with radio sources often reside in regions of higher galaxy density than do similar galaxies that are "radio quiet." In addition Heckman, Miley (ST Scl), van Breugel (UC Berkeley), and their collaborators have, in a multi-year program, established that many radio galaxies have copious amounts of hot ionized gas, enriched with heavy elements and spread throughout and beyond the galaxy. Such galaxies often show a distorted morphology. This body of data again indirectly suggests the presence of some form of interaction among galaxies. Thus it may be that encounters with other galaxies are a condition for rapid and dramatic evolution of galaxies. However, the deep CCD survey of Tyson and Seitzer implies that many isolated galaxies also have undergone significant evolution. A particularly elusive indication of galaxy formation and evolution comes from the "forest" of Ly-a absorption lines seen against distant quasars. These systems have been observed at NOAO by D. York (U. of Chicago) and by Green and J. Bechtold (U. of Arizona) but their exact nature remains in doubt A possibility is that these systems are primordial or protogalaxies lying along the line of sight to the quasar, but a better definition of their properties is needed to allow resolution of this question. Another group of galaxies that may be undergoing dramatic evolution are those recently detected by the Infrared Astronomy Satellite (IRAS) satellite. These objects emit up to 10 times the luminosity of a normal spiral galaxy, with the emission mostly in the far-infrared region. The presence of copious amounts of dust is suspected, but many more observations will be required to define these objects properly. An alternative candidate for a galaxy in the earliest stages of evolution is the hydrogen cloud discovered by Giovanelli (NAIC) and Haynes (Connell U.). This elongated starless cloud seems to offer support for the idea that disks of galaxies can form slowly throughout the history of the Universe. Although a wealth of data apparently exists concerning galaxy formation and evolution, much of it is very new and rather sparse. The questions are just now emerging, and much more observational material needs to be gathered before the relevant issues can be well defined. What is the role of the environment of galaxies upon their evolution? What are the important aspects of galaxy interactionsnear, distant, gaseous, and gravitational? If the implications from current data are true, what causes some galaxies to delay their formation and others not to do so? Is there a single cause of galaxy formation? Why do some galaxies become radio sources and others not? The answers to these and related questions will clearly involve many large-scale observing programs, which will require significant use of present and planned NOAO facilities. Fiber-fed spectrographs on medium and large aperture telescopes, two-dimensional infrared arrays, large format optical detectors, and enhanced stateof-the-art computing facilities will all be essential to these programs in the next five years. C. Stellar Structure and Evolution. Pursuit of questions concerning stellar structure and evolution not only reveals the nature of how stars evolve but also sheds light on such diverse topics as star-formation, the chemical and dynamical evolution of the galaxy, the calibration of the distance scale, and the age of the Universe. For example, J. Stauffer (Smithsonian Astrophysical Obs.) has obtained rotational velocities for stars in the Pleiades cluster, and he finds that nearly half of the stars observed have rotational velocities much higher than expected. This indicates that a major portion of the angular momentum of the protostellar cloud is retained during collapse and is not shed into a circumstellar disk. Stellar spectroscopy is also raising questions about the dynamics of star-formation in the galaxy. J. Hesser (Dominion Astrophysical Obs.), W. Harris (McMaster U.), and R. Bell (U. of Maryland) have found variations in the chemical composition of main-sequence stars in the globular cluster 47 Tucanae. Such stars are thought to have formed at the same epoch; hence these results raise questions about either the chemical homogeneity of the gas cloud that formed the cluster or about mixing within the stellar interiors. Much further work is needed in determining the compositions of stars in globular clusters. The chemical composition of another class of stars, M giants in the galactic bulge, has been studied at CTIO by J. Frogel (KPNO) and A. Whitford (Lick Obs.), who find them to be extremely metal-rich and unlike stars near the Sua Remarkably, these M giants are very similar to the major constituent stars in giant elliptical and SO spiral galaxies. This provides an opportunity to study close at hand a stellar population similar to that of the largest galaxies in the Universe. Spectroscopy of stars in our Galaxy is providing new and very interesting results that bear on the formation of the galaxy, and by implication, on the formation of all spiral galaxies. D. Geisler (CTIO) has used NOAO telescopes to study star clusters in the galactic disk in the direction of the anticenter. He finds a population of metal-poor stars that is younger than those found in other parts of the disk, and this important result implies that the entire galactic disk was not formed at the same time but rather that the outer portions formed much later. A complementary result has been obtained very recently by K. Gilroy and C. Sneden (U. of Texas), C. Pilachowski (KPNO), and J. Cowan (U. of Oklahoma) in their study of r and s process elements in halo stars. By coupling their results with known nucleosynthetic processes in stars of differing mass, these investigators have been able to determine that the extreme halo population of stars in the galaxy was formed in a very short time, about 10 million years. These two results have profound implications for models of galaxy formation, and they will provide motivation for additional observational and theoretical programs in the future. The above examples illustrate the broad range of stellar programs being carried out with NOAO facilities. Many of these are just now becoming well defined and will require further observations. In most cases, the next step is spectroscopy: of more stars, intrinsically fainter stars, more distant stars. Among young star clusters, more work must be done to understand the relation between starformation, stellar activity, and magnetic fields. Much work remains to be done in the area of stellar evolution through the study of the surface compositions of stars at different phases of evolution and to understand the role of mixing. Scientific programs in the area of spectroscopy of stars in clusters will benefit especially from multi-object capability, which will allow observations of many individual stars in a cluster simultaneously. Both CTIO and KPNO have multi-object fiber-fed spectrographs in operation. The designs of the 3.5-m WIYN telescope, which will be placed on Kitt Peak, and of the proposed 8-m telescopes have been optimized for fiber spectroscopy. D. Star-Formation. In view of their proximity and the number of years devoted to their study, surprisingly little is known about how stars form. Processes involving the formation and evolution of the parent clouds, their fragmentation and collapse, the role of angular momentum and magnetic fields, the establishment of the initial mass function, and the evolution of protostars and very young stars are all areas of active investigation. NOAO facilities have been used to observe regions of star-formation in our own and in nearby galaxies, and the advent of two-dimensional detector arrays that operate in the infrared is stimulating the growth of even more observing programs relevant to this topic. Moreover, these arrays are particularly well suited to be used in multi-wavelength, multi-observatory programs to address key questions in the area of star-formation. In particular, they will complement observations by HST, SIRTF, SOFIA, and millimeter and sub-millimeter telescopes and arrays. It has become well established that the collapse of protostellar clouds to form new stars is accompanied by an outflow of mass from the central region, and several programs have used NOAO observations to investigate the nature of these outflows and their implications for the star-formation process. These outflows are usually anisotropic, often bipolar, and sometimes very highly collimated. An important question, which has been examined by several groups, is the origin of this outflow and the mechanism for its collimation. In some cases the outflow is seen to be collimated to within 100 AU of the surface of the young stellar object, but it is still unclear if the wind is intrinsically bipolar or whether it is collimated by material around the star. Evidence that circumstellar disks are found around virtually all young stellar objects has been obtained by S. Strom (U. of Massachusetts) and his collaborators, which argues for an external collimation mechanism. However, additional data are needed. A successful understanding of this very common outflow phenomenon would provide valuable information about the role of angular momentum in the star-formation process, about the efficiency of stellar collapse and the role of circumstellar material, and about the conditions at the surface of the young star itself. Recent observations, again by the Stroms and their collaborators, suggest that the inner portions of disks around protostars, which have masses of about one percent of the mass of the Sun, are essentially cleared away over a period of about 107 years. What remains around main-sequence stars are disks with masses of about 10"6 Mq, and these disks do not extend all the way to the surfaces of the central stars. The obvious interpretation is that the disk material has aggregated to form planets. Again, detailed studies at high spatial resolution and low-resolution infrared spectroscopy are required to explore the evolution of stellar disks and hence to constrain models of the formation of planetary systems. Another phenomenon related to the star-formation process is the Herbig-Haro objects, which are emission line nebulae thought to originate from the interaction of matter ejected from a young star with the interstellar medium. Observations are consistent with highly supersonic outflow, often accompanied by a larger-scale, less rapid outflow in the parent molecular cloud. It is not clear if the rapid outflow is continuous or intermittent, nor is the mechanism known which produces the highdegree of collimation that is observed. Understanding of this phenomenon would shed light on the nature of both the young stellar object and its environment. Productive areas for future investigation include both the local phenomena of protostars, mass outflow, circumstellar shells, and accretion disks and such global aspects as the initial mass function and the variation of star-formation with changes in metallicity, turbulence, and magnetic fields. A major impetus for new and continuing programs in star-formation has come from the availability of two-dimensional infrared array detectors. Already these arrays have detected a circumstellar disk of molecular gas around a proto-stellar object, and they have also revealed rich groups of very young stars whose existence was heretofore only assumed. These early and tantalizing results are stimulating the formulation of many new observing programs. For the first time it will be possible to study in detail the star forming regions in other galaxies such as the Magellanic Clouds and members of the Local Group. In the galaxy itself, the problem of low mass star-formation and the lower mass region of the initial mass function can now be addressed. It will be possible to identify and study stars below the critical mass for nuclear ignition and to examine accretion disks around protostars and young stellar objects. The arrays will also make possible the search for particle disks around main-sequence stars, these being the presumed progenitors of planetary systems. in. INITIATIVES: KPNO AND CTIO It has been 20 years since the initiation of construction of a major new nighttime facility by NOAO. If ground-based astronomy in the U.S. is to remain competitive at the new levels of performance that are becoming available through major investments in Europe and Japan, then NOAO must build a new generation of facilities. The centerpiece of the nighttime program during the 1990s is construction of two 8-m telescopes, one at CTIO and the other on Mauna Kea. It is probable that this project will involve international partners. In addition, NOAO plans the construction and operation, again with partners, of one 4-m class telescope at each of its existing nighttime sites. These initiatives have been chosen after careful consideration of scientific opportunities, timing, and funding requirements. NOAO's pursuit of these initiatives reflects our assessment of their scientific merits and importance to the nation's astronomy effort; the time sequence for implementation of these projects is consistent with the pace of technology development and has been chosen to limit the number of concurrent activities. Construction of the 4-m telescopes is being pursued through collaborative efforts with universities. A proposal has been submitted to the NSF for construction of two 8-m telescopes. A. NOAO Goals for Large Telescopes. For the first time since the completion of the 5-m Hale telescope in 1948, it is technically feasible to build telescopes with significantly larger apertures. This alone would represent a major step forward for astronomy, but these new telescopes will be more than simply larger. By taking advantage of new technology for adaptive optics, interferometry, large format CCDs, infrared arrays, and high throughput, stable, multi-object spectrographs, it is possible to build facilities that will support qualitatively new kinds of science. In response to this opportunity, several groups both in the U.S. and abroad are planning to build telescopes with apertures in the range 8-m to 10-m. NOAO plans a two-step program of construction of large telescopes. The first step is to build two or more 8-m telescopes, with at least one in each hemisphere. The second step will be the development of an array of telescopes useful for interferometry. In 1989 NOAO formed an advisory committee on the best approaches to achieving high spatial resolution observations with large telescopes. Their input on adaptive optics and on interferometry is reflected in the proposal describing the NOAO 8-m telescopes project. In addition, we expect that some prototype interferometric arrays will be built during the next decade by university-based groups. Their experience will provide critical technical input to the definition of a national array. The goal of a nationally accessible, large aperture (specifically 15-m) optical and infrared telescope was originally defined by the Astronomy Survey Committee (the so-called "Field Committee" report, Astronomy and Astrophysics for the 1980s, 1, p. 15). Since that recommendation was made, several developments led NOAO to the conclusion that it is desirable to construct 8-m telescopes as a logical step toward a unique national facility. First extensive site surveys, interferometric measurements, and other data from Mauna Kea have established that the median full width half maximum image size is 0.4 arcsec, with 0.25 arcsec occurring about 10 percent of the time. No optical telescope now in operation can match the seeing on Mauna Kea. The construction of 8-m mirrors that can take advantage of this image quality is a major technical challenge. The detailed measurements of seeing on Mauna Kea yield an average image diameter that is about half the median actually observed at the telescopes already on the site. In order to realize the advantages of Mauna Kea, and of other sites as well, it is necessary to understand and correct the features of dome and telescope design that cause a degradation in image quality. If it is indeed possible to realize the full advantages of a site of the caliber of Mauna Kea, then much of the science originally planned for the 15-m can be carried out with an 8-m telescope. There is also a clear scientific requirement that NOAO expand its observing facilities in Chile. Certain astrophysically important objects, including most notably the Magellanic Clouds, are observable only from the Southern hemisphere. During the next decade, several space observatories will be launched that are certain to have a profound impact on astronomy, among them the Hubble Space Telescope, AXAF, the Gamma Ray Observatory, and SIRTF. Experience with previous orbiting telescopes has shown that most of the important discoveries from space require ground-based follow-up study, and about one-third of the objects surveyed by these all sky satellites cannot be observed from Northern hemisphere observatories. Strong community interest in increasing U.S. optical capabilities in Chile is already evident and several universities are considering locating 4-m telescopes on NOAO property in Chile. There is an additional private initiative to place an 8-m telescope on Las Campanas. NOAO believes that it is essential that there be a nationally accessible 8-m telescope in the Southern hemisphere as well. Accordingly, NOAO in cooperation with the astronomical community has proposed to the NSF to construct and operate 8-m telescopes in both the north and the south. These telescopes are designed to take full advantage of the characteristics of the best available sites. In addition to its superb seeing, Mauna Kea is the best infrared site developed so far. CTIO is characterized by good seeing and a large number of photometric nights. We are working closely with the other groups that have undertaken the construction of 8-m telescopes on design studies and will examine the feasibility of co-locating our telescopes and theirs to minimize infrastructure and operating costs. The advantages of working with other groups in developing 8-m telescopes include better use of national and international talent and resources, cost savings, and time savings. Commonality of design where appropriate can save money for both the private groups and NOAO. Complementarity in telescope performance specifications and in the selection of instrumentation can provide a much broader range of capabilities to the U.S. astronomical community as a whole than would be the case if these projects were uncoordinated. NOAO sees the construction of 8-m telescopes as a key step toward the development of arrays of telescopes with large effective aperture. The NOAO proposal consists of four volumes and covers all aspects of the project: the scientific justification; the technical description of the telescope system; a full complement of instruments; and the budget estimates and plans for project management. Here, we present a summary of the technical specifications, schedules, and cost estimates. Summary of Design Parameters for the 8-m Telescopes. Design Philosophy. Modem large telescope facilities consist of the telescope, usually with more than one well-instrumented focal position, housed in a protective enclosure, operated from remote consoles through computers, and preferably located on a dark, dry site where most of the nights are clear. This description applies to the 4-m telescopes now operated by NOAO at Kitt Peak and Cerro Tololo, and it will apply to the 8-m telescopes we propose to build on the summit ridge of Mauna Kea and on AURA property, probably Cerro Pachon, in Chile near Cerro Tololo. The design of both 8-m telescopes is illustrated in Figure 1. In planning to extend the national facilities to 8-m capability, we have been influenced by: 10 1) The belief that astronomy benefits immensely from open access on the basis of scientific merit to large telescopes possessing the light-gathering power and angular resolution to attack fundamental long-standing problems in astrophysics, and 2) The conviction that NOAO must take advantage of new technologies to push telescope performance at the national observatory to new standards of excellence. Figure l Figure 1 The 8-m telescope proposed for construction on Mauna Kea and a site in Chile. The design allows air to flow through the structure while retaining high stiffness properties. The top end can be extended to serve different optical configurations. 11 In the 20-year period since the NOAO 4-m telescopes were designed, many factors that influence telescope design have changed. We now recognize that the free atmosphere, undisturbed by interaction with the ground, delivers far better optical performance than was heretofore commonly believed. Moreover, as a direct result of our own site survey efforts, confirmed by a number of independent tests, we have identified at least one site-Mauna Kea~where the free-atmosphere performance can routinely be realized. We anticipate that conditions at Cerro Pachon will also frequently approach these same high performance levels. Recent efforts at many observatories have been directed toward identifying and minimizing the principal contributions to image degradation not associated with the free atmosphere-localized thermal pollution (dome and mirror seeing), tracking errors, and residual optics errors. These efforts have shown that our collective observing experience has generally been limited, even at the best of sites, by factors not directly related to atmospheric seeing. Over the past 20 years, structural analysis techniques have improved to the point where telescope system performance can accurately be predicted and designs can truly be optimized for best performance. Recent technological breakthroughs in a number of key areas related to telescope construction and design lead us to believe that 8-m telescopes can be constructed to take full advantage of the best performance that the atmosphere can deliver. For example, the testing of large optics can now be accomplished in a few minutes to accuracies impossible 20 years ago. The finite-element analysis technique now makes it possible to optimize mirror support and telescope structure in advance of fabrication. Microprocessor technology has brought about fundamentally different approaches to engineering design, enabling electronic coordination of widely separated functions that formerly were coordinated awkwardly at best (such as separate drive systems for the two altitude journals). Because of these and many other recent gains in technology, we believe that a doubling of the telescope size is possible, and at the same time the imaging performance can be improved. Toward achieving this goal, we have adopted the design guidelines listed below. 1) The telescopes and all of their subsystems will be designed to preserve the median image quality at Mauna Kea. The NOAO site tests, supported by other data, have shown that median upper atmosphere seeing (> 10 m above the ground) is below 0.5 arcsec FWHM. To avoid degrading this natural performance by more than 10 percent the telescope facility must be designed for 0.25 arcsec FWHM performance. We have adopted this goal for the 8-m telescopes, along with the corollary policy that any difficulty in achieving the required performance level in one subsystem will not justify relaxing the performance standards for other subsystems. (See Figure 2.) 2) Manmade seeing effects will be minimized by avoiding heat transfer to the air in and around the telescopes and by allowing air to pass naturally through the telescope enclosure to keep it flushed. Heat sources or sinks must be avoided. The telescopes will be positioned well above the naturally occurring ground-layer thermal turbulence. 3) The primary mirror will be circular and monolithic to simplify polishing and support. Guided by the present work on mirror blank manufacture, we have chosen 8 m as the largest size likely to be available in the early 1990s. Blanks of this diameter are expected to be available from the University of Arizona (Steward Observatory Minor Lab.), as well as from Corning Glass Works, and Schott Glaswerke. 12 Total Budget 0.25 r Telescope Enclosure 0.06 0.24 Residual Design Optical Surfaces Alignment Aberrations ^Tracking 0.17 0.14 0.10 f Primary Corrector Secondary 0.1 1 0.08 0.04 Polishing Thermal Effects Support 0.05 n no I |0.05 | < > Warpage due to Warpage due to Mirror Seeing Temperature Non-uniform 0.05 Gradients Expansion 0.05 Coefficient 0.05 1 .0 LU o Q CD CC O 15 um is dependent primarily on advances in detector technology. 15 Instrument proposed: • Infrared imager with array detectors plus options of filters, Fabry-Perot interferometer, grisms, or coronagraph. • High-resolution infrared echelle spectrograph, R = 20,000; up to R = 100,000 with adaptive optics. Scientific applications: • High-resolution imaging: search for planetary systems around nearby stars; search for brown dwarfs; distances to supemovae in other galaxies out to 1,500 Mpc; populations of star forming regions; mapping of molecules in gas clouds; nuclei of starburst and active galaxies. • Medium-resolution infrared spectroscopy: spectral evolution of galaxies; evolution of active galactic nuclei and QSOs; star-formation in distant galaxies; initial mass functions of populations in other galaxies. • High-resolution infrared spectroscopy: young stellar objects; star forming regions in nearby galaxies; active galactic nuclei. The Prime Focus Option. Although we do not plan to use the prime focus initially, the option of using it has been explored. A 14 arcmin FOV is possible with corrected image quality of about 0.33 arcsec FWHM for a wavelength range from 0.36 to 1 um. Instrument possibilities: • Direct imaging with CCDs; option of grisms. Scientific applications: • High-sensitivity imaging: calibration of distance scale, by Cepheids, out to distance modulus of -30, including Sculptor and Centarus groups of galaxies; planetary nebulae as distance indicators out to Virgo and Fornax clusters of galaxies and beyond; galaxy clustering properties as a function of redshift; color evolution of galaxies in clusters. The Interferometric Array Option. Another potential usage for the 8-m telescopes is as the major element of an interferometric array. Details of such a design remain as future endeavors, but we have considered the means to extract the light beam and feed it to a central beam correlator. Fibers could be used, but their limited transmissive properties, especially in the infrared beyond 5 um, are likely to be a continuing disadvantage. A conventional coude" beam extraction system using flat mirrors, coated for best reflectivity in the wavelength region of interest, appears to be the best solution at this time. The telescope design will enable future installation of this system. Adaptive Optics. The potential for adaptive optics correction of waveftont aberrations in the future is very great, and we have explored how this could be done. The system tentatively proposed has the following optical characteristics: 16 • Imaging down to the diffraction limit of the telescope. • Image scale: 2.52 mm/arcsec. • Wavelength coverage: infrared wavelengths > 1.6 um. • Final f/ratio (f/65) chosen to give critical sampling at 1.6 um for a detector with pixel spacing 50 um (= 0.02 arcsec). • Field of view: 2.5 arcsec for 128 x 128 detector arcay. Instrument available: • Infrared array detector with option of use with filters or grism. Scientific applications: • Detection of faint point-like sources against a bright background. • Relative photometry over small areas. • Spectroscopy of faint point-like sources. • Morphology of galaxies and QSOs out to faint limits. Schedule for the 8-m Telescopes. The proposal for the telescopes was submitted to the NSF in September 1989. According to the optimal funding schedule worked out with NSF, the project would be initiated in FY 1991 and would take six years until first light for the first telescope in FY 1997. Construction of the second telescope will require funding from sources outside the NSF. NOAO has been seeking partners and has had extensive discussions with the United Kingdom and Canada. If these two countries choose to join the project, then costs for both telescopes would be shared in the ratio 2:1:1 for the U.S., U.K., and Canada, respectively. The schedule that we are now working on assumes that in FY 1991 startup funding would enable the project group to be formed that will oversee the design and construction of the telescopes. Detailed design work can be started, as can the process of preparing for submission of bids by outside contractors who will be involved with the project Architectural work and site preparation efforts can also begin. An order for glass would be placed near the end of FY 1991. The detailed schedule for the 8-m telescopes specifies that contracts would be let for the major telescope systems such as the enclosure and the optical support structure in FY 1992. Major work on the instruments would also be started. As already indicated, we expect many of the instruments to be built outside NOAO. It is estimated that two years will be needed to produce the 8-m minor, polishing and testing will take another three years, so that the mirror will likely be the pacing item in the project. Approximately three years should be allowed for completion of the enclosure and up to three years for the optical support structure. These two parts of the project can proceed in parallel. In FY 1993 work will continue on all the telescope systems. By FY 1994 delivery of the mirror blank would be expected. The enclosure should be finished by early FY 1995 assuming work started in 17 early FY 1992 on the site preparation and architectural design. The telescope mounting could also be ready for shop testing by the end of FY 1995. In the following years the mounting and optical support structure would be installed and tested in the enclosure before the minor polishing is completed; this will facilitate the installation and testing of the optics once they are finished. By FY 1997 first light at the completed telescope can be expected. The second telescope will be constructed in parallel with the first one where possible. This will yield a saving in both time and cost compared to two separate projects. Being in different hemispheres, work on the enclosures for the two telescopes can proceed independently. Mechanical parts can be made in duplicate when it is efficient to do so. However, the fabrication of the mirror blanks is expected to proceed serially. It may be possible for the polishing of the mirrors to proceed concurrently, depending on what facilities are developed over the next few years. A summary of the proposed construction schedule for two telescopes follows. Fiscal years 1991 through 1995 are represented. FY 1991: project team will be set up; telescope specifications will be established, glass for the first 8-m minor will be ordered. FY 1992: instrument specs will be prepared; and bids will go out for the control facility, dome enclosure, mount, and coating chamber. FY 1993: the concrete work will begin and we will start to manufacture a control building. The primary minor blank will be produced. FY 1994: assembly of the control building will occur, and it then will be shipped to the site. This work will extend into FY 1995. Manufacturing and testing of the coating chamber will occur, as well as manufacturing of secondary cells and corrector optics. FY 1995: more assembly, manufacturing, and testing will occur. The control system will be fabricated and tested; software development will begin, and the coating chamber will be shipped and installed. Glass will be ordered for the second primary mirror, and the primary cell and supports will be manufactured and shipped. Preliminary cost estimates are approximately $85M for the first fully-instrumented 8-m telescope in 1989 dollars. The second 8-m telescope would cost approximately $58M. The primary reason for the difference is that developmental costs and the initial six years of staff costs have been allocated entirely to the first telescope. B. Beyond 8-m Telescopes. Scientific progress in many forefront problems of astrophysics demands very high angular resolution observations. This capability, when combined with high spectral resolution observations and the unprecedented light gathering power of large apertures, will permit breakthroughs in many scientific problems. It will also complement and extend observations from the new generation of space-based observatories. The scientific significance of high angular resolution observations increases enormously with increasing baseline. For example, the jump from 4-m apertures to 8-m apertures allows fundamental 18 progress in the studies of stellar and planetary formation. The further jump to interferometric baselines of 20 to 200 m permits the detailed imaging of stellar surfaces and the measurement of supernovae diameters in the nearest external galaxies. Interferometric baselines of about 1 km will allow detailed imaging of the inner regions of active galactic nuclei (AGNs) and quasi stellar objects (QSOs). Large aperture telescopes (D = 8 m) are the essential building blocks for achieving these goals because their high light gathering power will permit the necessary spectral and photometric sensitivity. The major impact of both adaptive optics and interferometric capability will be in the near-infrared (X = 2.5 um), where the background from both atmospheric airglow and thermal emission is relatively low. NOAO has outlined four key phases of technology development that if successful, would provide interferometric capability on Mauna Kea. We expect most of the technology development to take place outside NOAO. Once feasibility has been established, NOAO would assume responsibility for implementation on Mauna Kea. The phases are as follows: For Phase, I, the NOAO 8-m telescopes will provide diffraction-limited imaging up to their aperture limit by means of both passive and active observing techniques. In Phase II, we propose the installation of three or four telescopes of medium aperture (2 - 3 m) on transporters, movable along contours of constant elevation adjacent to the Mauna Kea 8-m telescope. This array would provide coverage of telescope separations up to approximately 100 m. In Phase III, we propose to incorporate one or more additional large apertures into the "backbone" of the array. These may be either additional NOAO telescopes or facilities of other institutions merged by agreement into the interferometric configuration. In Phase IV, additional independent telescopes on the Mauna Kea site will be merged optically into the ad hoc array, with very complete u,v coverage for separations up to 1 km. Optical fibers may be very suitable for retro-fitting heterogeneous facilities for interferometric operation. C. 4-m Projects. In addition to its program to develop 8-m and larger telescopes for the astronomical community, NOAO recognizes a continuing need for additional 4-m class telescopes. It is likely that telescopes of this aperture will become the workhorses of the future that 1-m to 2-m telescopes are now. The challenge is to provide access to 4-m class telescopes within the National Observatories without slowing the momentum toward the construction of larger telescopes. New 4-m class telescopes at NOAO will be designed from the outset to provide capabilities not now available to us. They will have limited instrumentation; changes between available instruments will be designed to be rapid; and the capabilities will be complementary to the existing 4-m telescopes with minimum duplication of instruments. NOAO's share of the observing time, which will be queue scheduled with staff doing most of the observing, will be used for surveys, synoptic programs, and projects that need simultaneous or near simultaneous observations at multiple wavelengths. The most pressing need in terms of specific instrumentation, based on both existing requests for observing time and on our assessment of the scientific problems that can be addressed by 4-m class telescopes, is for high and moderate-resolution multiple object spectroscopy. This choice is driven by 19 the significant gains to be made from observing many objects in a star cluster or cluster of galaxies, or many discrete points in an extended galaxy or nebula, simultaneously. The existing 4-m telescopes have been used to study selectively a small sample of objects of interesting classes; this approach is viable for some programs, but limits progress significantly for important astrophysical programs that -equire large-scale spectroscopic surveys. The evolution of clusters of galaxies will not be understood until many clusters, not just a few, are systematically studied; such data are essential for the determination of the cosmological constant q0. The large-scale structure of the Universe will not be known until hundreds of quasi-stellar objects, not just a handful, have been observed to map out the distribution of matter in the Universe. The evolution of stars in clusters and of stellar populations will not be well understood until large samples of stars can be surveyed. A new multiple object spectroscopic telescope can greatly expand our capability to obtain observations in such areas as QSO spectroscopy to study the large-scale structure of the Universe, studies of the evolution of clusters of galaxies, the study of stellar evolution through spectroscopy of stars in clusters, and the detailed analysis of stellar structure through the techniques of the rapidly expanding field of stellar physics. Specific examples of the types of programs that would benefit from multiple object spectroscopy include studies of the detailed element abundance changes with stellar evolution in members of star clusters of different ages, studies of the binary frequency of stars in clusters of different ages and of different stellar populations, or the monitoring of activity in many stars along the main-sequence of star clusters of different ages. Spectroscopy of many planetary nebulae in nearby galaxies out to 5 Mpc (such as the Sculptor group of galaxies or M 81) or of H II regions out to 10 to 15 Mpc will be possible. Observations of individual globular clusters in Virgo group galaxies to determine velocity dispersions and compositions can also be obtained efficiently using multiple object spectroscopy. To confine our considerations to current scientific need is, however, insufficient; past experience teaches that space observatories in orbit, in preparation for launch, or in the planning stages will open up exciting new fields that can only be fully studied using large ground-based instruments. As an example, follow-up observations of X-ray sources discovered with the Einstein satellite have over the past few years accounted for almost one-quarter of the visitor use of the CTIO 4-m and 1.5-m telescopes. At KPNO, requests for observing programs to follow-up on IRAS observations continue to increase dramatically. NOAO-University Projects. The clear requirement to move vigorously to meet the need of the astronomical community for access to 4-m class telescopes can be met by new approaches to funding and telescope construction. The technology being developed for very large telescopes is applicable to small ones as well, and along with advances in computer control, permits a significant lowering in the cost of construction. In actual dollars, the cost of building a 4-m telescope today is approximately equivalent to the cost of building the KPNO 4-m telescope. With allowance for inflation, this means that the real cost of construction has decreased by about a factor of three and has come within the reach of some universities or university consortia. While construction can often be financed by universities with relatively modest astronomy programs, only the largest groups are well equipped to assume the continuing burdens of operating and instrumenting telescopes. Telescopes of 4-m class aperture require substantial technical support for operation. Instrumentation, too, is becoming increasingly complex and costly to build. NOAO has in its Newsletter invited universities to join with either KPNO or CTIO to build and operate 4-m telescopes. The universities would assume responsibility for funding the construction phase, NOAO would operate the completed telescope for a fixed period of time, and the observing 20 time would be shared between the universities and the NOAO community of users in proportion to the financial commitments of the partners. Two projects are now in the planning stages, and we would expect to seek NSF approval to proceed with one or both during the next two years. The first project would place a 3.5-m telescope optimized for multiple-object spectroscopy on Kitt Peak. As part of the technology development program in support of new technology large telescopes, NOAO has received a 3.5-m, spin-cast, lightweight mirror, which has been cast at the Steward Observatory Mirror Lab. The mirror is being used to study the feasibility of fast, lightweight mirrors for large telescopes; the mirror is being polished to about f/1.75, and an active mirror support system is being developed to maintain the 0.25 arcsec images required by the best sites. The mirror, figured and with a support system, will be completed not later than 1992. The development and construction of a new telescope facility will allow us to put this test mirror into a real telescope and to evaluate its optical performance under real astronomical conditions. Three universities, Wisconsin, Indiana, and Yale have expressed interest in joining NOAO in completing a telescope that would make use of this mirror. Among these three institutions, a total of $8.5M has been committed over the time period, 1989 - 1994; engineering design studies are being undertaken. An agreement governing the construction and operation of the telescope is now being negotiated. If Wisconsin, Indiana, Yale, and NOAO meet the obligations detailed in the draft WIYN agreement, then the observing time remaining after the allocation of maintenance and discretionary time would be apportioned in the following way; Wisconsin 26%, Indiana 17%, Yale 17%, and NOAO 40%. A project manager has been hired and is undertaking the preliminary studies required for definition of the project. The telescope itself will be to a large degree a copy of the ARC telescope on Sacramento Peak, but some modifications to that design are necessary to meet the requirements for a wide field for multiple object spectroscopy and to provide an f/ratio suitable for fibers. The schedule for construction is likely to be controlled by cash flow. In contrast to most other projects, it appears likely that the mirror will not be the pacing item. NOAO will provide a multiple-object spectrograph for the WIYN telescope. If funding permits, a new spectrograph will be built based on the instrument cunently being fabricated for the 4-m Mayall Telescope. If, as appears likely, funding does not permit, the Mayall spectrograph will be transferred to the WIYN. The universities will be responsible for the remaining instrumentation, but the selection of specific instruments has not yet been made. Key milestones for the WIYN project provided cash flow can be maintained, are as follows: WIYN Milestones March 1989 June 1991 June 1991 December 1992 January 1993 May 1993 May 1993 June 1994 Received primary blank Begin telescope fabrication Begin construction of enclosure Enclosure completed Install mount Install optics First light Full operation 21 Three groups are currendy pursuing plans to build 4-m class telescopes at CTIO. The most advanced of these is the Southern Observatory for Astronomical Research (SOAR) telescope, with the University of North Carolina taking the initiative in this project. Columbia has recently indicated its interest in joining SOAR. As currendy planned, this telescope would employ a light-weight altitude-azimuth mount similar in concept to that of the ARC 3.5-m telescope. The primary mirror would be a 20-cm thick meniscus fabricated from low-expansion ULE glass by the Coming Glass Works. Instruments would be semi-permanentiy stationed at the two Nasmyth foci, and possibly also at two bent-Nasmyth foci. A rotating tertiary mirror would allow rapid access to any of the foci so that quick changeovers from one instrument to another could be accommodated on a given night Current strategy calls for North Carolina and its partner to build the telescopes and instruments, although some of the latter might be built at CTIO under contract. In return, NOAO would agree to operate the telescope for some specified period of time. The exact division of observing time among the three partners has yet to be specified, but the working plan is for this to amount to approximately one-third for each. The two other groups interested in building 4-m telescopes at CTIO are Harvard University and a consortium of astronomical institutes in Brazil. The planning for these telescopes is still at an early stage, but it is likely that their characteristics will be similar to those of the SOAR telescope. NOAO's participation in these projects would probably be more along the lines followed at Mauna Kea, with CTIO supplying a site but charging for other services at cost. In return, the NOAO community would receive something on the order of 15% of the observing time on both telescopes. These new 4-m telescopes would be located on one of the two remaining undeveloped summits on the AURA property in Chile-Ceno Morado or Cerro Pachon. Cerro Morado is located 4 km south of Ceno Tololo and has a very large, flat summit area with plenty of room for a large complement of telescopes. It is slightly lower than Tololo (~ 50 m) but almost certainly has site characteristics similar to those of Tololo. Cerro Pachon, however, is a potentially better site than Tololo due to its 500 m height advantage. Located approximately 10 km to the southeast of Tololo, Pachon can be routinely reached by an unimproved dirt road in 30 minutes. Although Cerro Pachon's summit is not as flat as Morado's, there is still room for a number of large telescopes. In order to determine more precisely the site characteristics of Cerro Pachon, a site survey recently was initiated there. With the help of financial contributions from the SOAR group, Harvard, and the Brazilians, the road to the summit was built and a shelter constructed for the observers in mid-1988. Site testing equipment from the NOAO Mauna Kea/Mt Graham surveys, including a weather station, infrared sky radiance monitor, micro-thermal tower, echosonde, and seeing telescope, has since been installed, and similar equipment is presently being set up on Tololo to provide a point of reference. Current plans call for the site testing of Cerro Pachon to proceed for two full years ending in 1992. 22 IV. TELESCOPE TECHNOLOGY PROGRAM (TTP) In March 1989, NOAO received the second 3.5-m diameter borosilicate honeycomb mirror casting produced by the Steward Observatory Minor Lab. (SOML). The first such casting, made to evaluate the casting furnace performance, was delivered to the Astrophysical Research Consortium (ARC) in June 1988 and will be installed in a telescope on Apache Point near Sacramento Peak when polishing work is completed. The NOAO mirror blank will be used to develop and test the methods needed for producing high quality 8-m minors. This program involves polishing, supporting, and thermally stabilizing the minor such that none of the individual error sources (e.g., residual polishing error, support effects, thermal warpage, mirror seeing, etc.) exceed about 0.05 arcsec FWHM. Strategies and testing methods for accomplishing this challenging goal have been part of the TTP program work since its inception. The NOAO 3.5-m mirror is being polished to a high-quality spherical surface. During the remainder of 1990 this minor will undergo a series of tests to evaluate performance of a controlled-force support and a thermal control system. Specifications for the thermal control system will emerge from tests already completed at NOAO on a 1.8-m diameter honeycomb mirror and from a joint project to test a "one-sixth of a 3.5-m mirror" mock-up at the Apache Point ARC Observatory. This latter activity combines the efforts of NOAO, Steward Observatory, the ARC, and the Magellan project. In 1991, after the developmental tests are complete, the 3.5-m mirror will be aspherized and tested again leading to its final completion in 1992. Concurrent with the development work related to the 3.5-m minor, the TTP group will coordinate design specifications with the Wisconsin-Indiana-YaleNOAO (WIYN) project so that the completed 3.5-m minor will be suitable for the telescope planned by that group. Milestones for the project are as follows: 3.5-m Mirror Project Major Milestones March 1989 Received mirror blank May 1989 February 1990 April 1990 May 1990 July 1990 March 1991 March 1992 August 1992 Completed generation of blank Advanced Technology Optical Telescopes Conference Complete thermal control test at Apache Point Finished spherical polishing Mirror cell structural fabrication completed Complete active optics test Finish aspheric polishing Complete full system test In 1992, after completion of work on the 3.5-m mirror, the TTP program will be eliminated. 23 V. OBSERVATORY OPERATIONS: CTIO and KPNO A. Cerro Tololo Inter-American Observatory. CTIO and KPNO each operate six telescopes for the benefit of the nighttime community. During the past year both observatories have focused on devising plans that would lead to the construction of major new telescopes and on laying out a long range plan for instrumentation. The plans for 8-m and 4-m telescopes have been described in earlier sections of this document. The following sections describe current operations and the plans for new instruments. During the next year, both observatories must begin planning for the changes in operations that will be mandated by the requirements of the new types of scientific programs that will be undertaken during the next decade. Greater use of queue scheduling, service observing, and archiving, along with improved opportunities to carry out large-scale surveys and synoptic programs, would permit NOAO to meet the diverse requirements of the community more effectively. The present plan describes some of these possibilities briefly. Future long range plans will detail the initiatives that we propose to undertake in the area of operations and define the resources required. Overview of the Facilities. CTIO operates six telescopes with a range of capabilities chosen to accommodate the research needs of U.S. astronomers in the Southern hemisphere. The multi-purpose 4-m telescope is still the largest optical telescope in the Southern hemisphere and has been equipped with the most modem instrumentation for imaging and spectroscopy. Coupled with the dark sky at Tololo, the 4-m telescope has been an important tool in the study of the most distant galaxies and quasars. The 1.5-m telescope is also multi-purpose and is CTIO's optimal telescope for the IR. It has been used extensively for infrared spectroscopy and imaging photometry and has produced excellent results for objects in the Magellanic Clouds. The 1-m and 0.9-m telescopes have been converted into dedicated facilities, performing the tasks most requested by observers, i.e., imaging and spectroscopy. The Yale 1-m has been equipped with a twodimensional photon counting spectrophotometer that is one of the observatory's most productive instruments. Ideally suited to studies of emission-line objects, it has been used for the study of supernovae, active galaxies, and mass transfer binaries. The 0.9-m is a very good imaging telescope and is used exclusively for CCD imaging photometry. The Michigan Curtis Schmidt is equipped for wide field photography and objective prism surveys. It has been the workhorse of such pioneering investigations as the discovery of carbon stars near the galactic center and the search for distant QSOs. The smallest telescope, the Lowell 24-inch, is used for single-channel aperture photometry and has been active in establishing photometric sequences. The full complement of telescopes and associated instrument configurations at CTIO are listed in Table I. 24 TABLE I CTIO Telescope/Instrument Combinations 4-m Telescope: ARGUS fiber-fed multi-object positioner + bench-mounted - R-C Spectrograph + CCD - Echelle Spectrograph + CCD (Air Schmidt camera) Automatic Single Channel Photometer Echelle Spectrograph + CCD (Air Schmidt or Long cameras) Infrared + D2 and/or D3 Photometers + Spectrometer + Imaging array Prime Focus Camera + CCD + Plates R-C Spectrograph + CCD (Air Schmidt cameras) + 2D-Frutti (Folded Schmidt camera) Rutgers Imaging Fabry-Perot + CCD Cassegrain Direct + CCDs 1.5-m Telescope: Automatic Single Channel Photometer Cassegrain Direct + CCD Cassegrain Direct + Plates Cassegrain Spectrograph + CCD Fiber-Fed Echelle Spectrograph + CCD (300 mm or Long cameras) Infrared + D2 and/or D3 Photometers + Spectrometer + Imaging array Rutgers Imaging Fabry-Perot + CCD 1-m Telescope: Cassegrain Spectrograph + 2D-Frutti Automatic Single Channel Photometer 0.9-m Telescope: Cassegrain Direct Imaging + CCD 0.6-m Telescope: Manual (standard) Photometer Curtis Schmidt: Plates, Direct or Prism 25 Site Survey. The Andean peaks of northern Chile have been shown to provide excellent Southern hemisphere observing sites, and both NOAO and other U.S. institutions plan to place more large telescopes in Chile. Since the summit of Tololo is now fully occupied, alternative locations must be explored. Fortunately, other peaks exist on AURA property which are already known from previous testing to be good telescope sites. The highest peak on AURA land is Cerro Pachon (8900 feet). Ceno Pachon appears to be the best choice of site for additional NOAO telescopes. Its higher elevation is an advantage for infrared observations. Tests in the 1960s indicated that the seeing was excellent. There is ample room for several telescopes. The site is close enough to CTIO to make use of much of the existing infrastructure. Ceno Pachon consists of a ridge about 1.5 km long, on which are located four distinct summits. The two most north-easterly summits are on AURA property; the other two are on adjacent privately owned property. Since the prevailing wind is from the northeast, AURA already owns the best sites; however it is advisable to try to purchase some of the adjoining land, especially since the present access road crosses the private property. The CTIO director is currently negotiating purchase of the land for AURA. Comparative site testing of Tololo and Pachon has just begun and will continue for three years. A road has been constructed to the summit A small cabin has been built to provide living quarters for the worker who is monitoring the survey equipment. Data are now being gathered daily by a microthermal tower, acoustic sounder, and an IR water vapor monitor. Identical 12-inch telescopes used as seeing monitors are being calibrated, and one is already making direct measurements of the atmospheric seeing. These data will be gathered at different locales on the Pachon ridge in order to identify the best sites for new telescopes. New Telescopes. Together with KPNO, Ceno Tololo is entering a phase where several important opportunities are presenting themselves that will have long-term impact on the observatory. As a result of staff discussions, CTIO has arrived at a consensus regarding the immediate directions it would like to proceed in for the next few years. The major decisions relate to participation in an 8-m project, collaboration with universities in building 4-m telescopes, and major instrumentation and detector development. The 8-m telescope project remains the highest priority of the observatory. Other initiatives should not be allowed to interfere in a major way. This imposes certain constraints on CTIO participation in consortia wishing to build and instrument telescopes in Chile. Because the scientific staff and engineering department are not large, CTIO is reluctant to commit itself to full participation in the design and instrumentation of such telescopes. It will consider the operation of the telescope if funds are available. Closing the smaller telescopes would allow sufficient funds to be re-programmed within CTIO to do so, and the staff and users favor this if NOAO were to receive at least 40 percent of the telescope time. Realistically, CTIO could at most think of operating one telescope in this fashion. Any other telescopes would have to be operated independently of CTIO, in the Mauna Kea style. The future of CTIO is directiy linked to new large telescopes, and therefore such projects are being encouraged strongly. 26 B. Kitt Peak National Observatory. Overview of the Facilities. The first of the AURA managed observatories was established at Kitt Peak, and telescopes at that site have been continuously in operation since 1959. KPNO now operates six nighttime telescopes. The 4-m telescope is a general purpose telescope that is used in the optical and infrared, for spectroscopy and imaging. With the initial announcement of the availability of infrared arrays, the 2.1-m telescope during bright time became the most oversubscribed of all the Kitt Peak telescopes. This telescope is also used for low-resolution spectrophotometry, CCD imaging, and coude- spectroscopy. The 1.3-m telescope is used for photometry during dark time, and with its chopping secondary is particulariy well suited to measurements of extended objects with low surface brightness. In bright time it is used for infrared astronomy. The Bunell Schmidt is the only nationally accessible Schmidt in the Northern hemisphere that is equipped with prisms for spectroscopic surveys; this telescope is operated jointly by KPNO and Case Western Reserve. The coude" feed telescope was designed to send light to the coude" spectrograph at the 2.1-m telescope, thereby allowing use of the spectrograph when the 2.1-m is being scheduled for other programs. To help reduce costs, the #1 0.9-m will be closed in the summer of 1990. The #1 0.9-m telescope was used almost exclusively for CCD imaging, and this function will be taken over by the #2 0.9-m telescope. The #2 0.9-m now has two instruments~a single channel photometer system and an intensified Reticon scanner for spectrophotometry. The availability of very large apertures at the #2 0.9-m makes this telescope suitable for spectrophotometry of galaxies with large angular sizes. Table II lists the Kitt Peak telescopes and summarizes the instrumentation available for each. 27 TABLE H KPNO Telescope/Instrument Combinations 4-m Telescope: Prime Focus Camera + Photographic Plates + Cass Direct CCD Imaging R-C Spectrograph + CCD (UV Fast Camera) + Cryogenic Camera Echelle Spectrograph + CCD (UV Fast Camera) Infrared Imager (IRIM) Infrared Cryogenic Spectrometer (CRSP) Fourier Transform Spectrometer (InSb and Si) Multi-object Fiber-optic Feed (NESSIE) 2.1-m Telescope: Gold Spectrograph + CCD (Wynne CJ Camera) Cassegrain Direct CCD Imaging Infrared Imager (IRIM) Infrared Cryogenic Spectrometer (CRSP) Infrared Photometer (Blue Toad 1-5 um InSb) Fiber Optic Echelle Spectrograph + CCD (Visitor Instrument) Coude" Spectrograph #5 or #6 Camera + CCD Coud6 Feed: Fiber Optic Echelle Spectrograph + CCD (Visitor Instrument) Coude" Spectrograph #5 or #6 Camera + CCD 1.3-m Telescope: Mark HI Photometer Infrared Infrared Infrared Infrared Photometer (OTTO 1-5 um InSb) Bolometer (2-20 um Ge:Ga) Imager (IRIM) Cryogenic Spectrometer (CRSP) Burrell-Schmidt Telescope: Direct Photographic or Choice of 5 Objective Prisms + Photographic Plates or CCD #2 0.9-m Telescope: Automatic Filter Photometer White Spectrograph + Intensified Reticon Scanner Direct CCD 28 Approximately 600 astronomers use KPNO telescopes each year. About 20 percent of the users are from institutions that have major telescopes of their own but wish to make use of unique instrumentation at KPNO. At any given time there are about 20 students carrying out observations for Ph.D. dissertations. There are several other telescopes on Kitt Peak in addition to the KPNO nighttime facilities. NSO operates two solar telescopes on Kitt Peak-the Vacuum Telescope, which provides much of the basic monitoring data for the Sun including daily magnetograms, and the McMath, which is used for both solar observations and monitoring of stellar activity. Other telescopes are operated by the National Radio Astronomy Observatory (NRAO), including one of the antennas for the Very Long Baseline Array (VLBA) on the southwest ridge; by Steward Observatory of the University of Arizona; and by the MDM Observatory, which serves the University of Michigan, Dartmouth College, and the Massachusetts Institute of Technology. These facilities all benefit from the fact that KPNO pays for most of the costs of maintaining the mountain infrastructure. Site Quality. NOAO believes that Kitt Peak can and should remain one of the premier sites for U.S. astronomy. Stimulated by theoretical calculations by R. Garstang (University of Colorado) that showed that Kitt Peak should still be a good dark sky site, we have undertaken a series of measurements of sky brightness. In the V magnitude, measurements show that Kitt Peak is only about 0.10 mag brighter than the value expected for a completely dark sky, and Garstang's calculations based on predictions of population growth by the State of Arizona suggest that even in 2035 Kitt Peak will be only 0.26 mag brighter than the natural sky background at the zenith. These calculations do not take into account the effect of the lighting ordinances. The extensive publicity given to problems of light pollution has induced all of the relevant counties in Arizona to adopt lighting ordinances requested by the astronomy community. Kitt Peak will, therefore, remain satisfactory well into the next century for most programs, and will be especially well suited for high-resolution spectroscopy and near-infrared observations. New Telescopes. While we do not propose to build 8-m class telescopes on Kitt Peak, we do believe it would be cost effective to upgrade the facilities offered to the Kitt Peak user community. As described elsewhere in this plan, we are actively exploring the construction, in collaboration with universities, of a 3.5-m telescope, which NOAO would use primarily for multiple-object spectroscopy. C. Instrumentation for KPNO and CTIO. One of the results of the formation of NOAO has been a closer coordination of instrument development at the two nighttime sites. The responsibility for designing and fabricating TV acquisition systems, CCD controllers, and infrared instrumentation, for establishing high-speed data links, and for testing and characterization of detectors has been assigned to one of the observatories, which has then carried out the work on behalf of both. Even in cases where the observatories have adopted different approaches to instrumentation, as in the case of fiber-fed spectrographs, certain technology developments common to both projects have been shared. This document presents a five year plan for instrumentation at each observatory. In practice, these plans cover the first four years in detail and assume continued level of effort in the fifth year. Since it takes approximately two years to complete a major instrument, four years is a natural planning cycle in that it encompasses this round of instruments and the next The plans are presented separately so that it is clear what the evolution of instrumentation at the separate sites will be. Close coordination 29 of the two programs will continue, and the separate five year plans have served to identify areas where cooperation is most likely to be both necessary and effective. Infrared Instruments at KPNO. Infrared astronomy at KPNO experienced a technological revolution in 1987 with the introduction of a camera based on the SBRC 58 x 62 InSb array. This camera and, more recentiy, a spectrometer based on the same detector chip, have provided significant new capabilities to the community. Our ability to lead this revolution stems directly from our vigorous Research and Development program. We expect that the research and development effort will continue to drive the KPNO infrared program for the foreseeable future. Research will focus on acquisition, evaluation, and operation of new detectors. In the 1 - 5 um wavelength interval, dramatic improvements in both format size and noise performance are confidently expected. At 10 um and beyond, we will take our first steps towards the operation of detector anays. Implementation of new detectors in the cameras and spectrometers described below will emphasize, where possible, commonality of design-based on the novel concept of the "Cryogenic Optical Bench." Designs will be modular, will support plug-in detector upgrades (including the anticipated format size increases), and will utilize closed-cycle refrigeratioa Plumbing for the closed-cycle systems will be required at the 1.3-m, 2.1-m, and 4-m telescopes. Upgrades in computer hardware and software, and in the speed of instrumental electronics, will be required in order to provide reduction and archiving capabilities consistent with these innovations. A new f/15 secondary mirror at the 4-m telescope, and small active secondaries capable of "fast guiding" at the 4-m and 2.1-m telescopes, will be needed for optimal operation. The infrared program is strong and vigorous; there is no shortage of ideas for exciting new instruments, nor of expertise to build them. For this forward-look period, progress will be dictated mainly by the level of support available. For example, level funding for the next five years (the assumption used in the preparation of this document) imposes a projected rate of approximately one new instrument per year, at this rate KPNO will not offer a 10 um array-based user instrument before 1995. Although collaborative ventures or the success of the 8-m telescopes proposal may provide some relief from this prognosis, the most straightforward way to accelerate deployment of new infrared capabilities is by increased support for the program. The FY 1990 total reflects the budget figures given in the FY 1990 provisional program plan. Instrumentation projects are planned only through FY 1994; beyond that, we are less certain about what will be undertaken. 30 TABLE m IR Program Current Year (FY 1990) Project Labor Scientist (mm) Capital ( $K ) Detectors ( $K ) FY 1990 Four-Color Camera Joyce, et al. 70 15 Cryo Echelle Spectrometer Cryo Optical Bench Hinkle, et al. Probst et al. 30 35 Gadey, et al. 20 25 145 35 Detector/Other R&D 188 100 15 20 15 Total 103 Five Year Plan FY 1991 Cryo Echelle Spectrometer Cryo Optical Bench Near IR Camera 4-m IR f/15 Secondary Medium-Resolution Spectrometer Detector/Other R&D Hinkle, et al. Probst et al. Pilachowski, et al. Gadey, et al. Joyce, et al. Gadey, et al. 5 5 10 ]0 5 35 25 90 165 240 Misc. KP Projects 70 Total FY 1992 Cryo Echelle Spectrometer Cryo Optical Bench Medium-Resolution Spectrograph Fast Two-Axis Secondary Four-Color Camera Electronics Upgrade Detector/Other R&D Hinkle, et al. Probst et al. 20 5 40 5 Joyce, et al. Ridgway Joyce, et al. Gadey, et al. 40 40 20 25 50 30 40 50 Detectors (Echelle/COB) 200 Total 185 180 200 FY 1993 Medium-Resolution Spectrometer Ten um Spectrometer/Camera Fast Two-Axis Secondary 2nd Gen. Low-Resolution Spectrometer Fiber-Fed IR Spectrograph Detector/Other R&D Joyce, et al. 80 70 Gillett, et al. 10 30 Ridgway Joyce, et al. 60 40 5 5 IR Staff Gadey, et al. 5 5 25 85 185 235 Detectors (MRS/Spare) 100 Total 31 100 IR Program (Continued) FY 1994 Medium-Resolution Spectrometer Ten um Spectrometer/Camera 2nd Gen. Low-Resolution Spectrometer Fiber-Fed IR Spectrograph 2nd Generation Cryo Optical Bench Detector/Other R&D Joyce, et al. 25 Gillett, et al. 90 90 Joyce, et al. 40 40 5 5 IR Staff 40 Probst et al. 5 2 Gadey, et al. 25 87 Total 190 2~64 Total 190 497 Detectors (ten um/LRS) 200 FY 1995 TBD 32 200 A timeline for the development of instrumentation is given below. A description of each instrument and of outstanding engineering issues follows: • Infrared Imager (IRIM). Broad- and narrow-band imaging at 1 - 5 um with a 58 x 62 SBRC InSb array is carried out with the infrared imager. IRIM has been used successfully for speckle observations, polarimetry, and as a stellar coronagraph. Tests of grism operation and of raster mapping with this instrument were important in specifying the parameters of the Four-Color Camera (SQIID) and the Cryogenic Optical Bench (BUFO). • Infrared Grating Spectrometer (CRSP). The infrared grating spectrometer is a long-slit spectrograph which operates with a 58 x 62 SBRC InSb array in the 1 - 5 um range. Two resolutions are available, R = 200 and R = 1,500 per pixel. The CRSP was constructed by retro-fitting an array detector into an existing prototype instrument; as a result of the very large improvement in detector performance, various additional modifications also proved necessary, primarily to reduce the ambient radiation level inside the cryostat. Thus commissioning of CRSP provided an invaluable object lesson and set the engineering standards for future instruments. • Near-IR Ge Camera. Collaboration between the R&D effort and the Hughes Technology Center has raised the possibility of using 256 x 256 Germanium arrays for astronomy. Detectors will be received for evaluation in early 1990, and, should they prove satisfactory, will be deployed soon after in a standard CCD dewar by the O/UV group. This new camera will bridge the "gap" between traditional optical and infrared regimes, operating from 0.7 - 1.5 um. This project is a good example of the potential for low cost innovation through collaboration. • Four-Color Camera (SQIID). This camera, made up of four 256 x 256 PtSi array detectors, will simultaneously view the same area of sky at wavelengths J, H, K, and L (1.2, 1.6, 2.2, and 3.5 um). This large field of view camera will be used primarily in a "telescope raster" mode to build up images of a square degree or more in size. On-line data reduction will be emphasized. The capability for simultaneous J, H, K imaging polarimetry has already been designed into SQIID; this function is enabled by the addition of a warm super-achromatic half-wave plate and a cooled analyzer. Closed cycle coolers will remove the need for liquid cryogens in this and all subsequent instruments. • Cryogenic Echelle Spectrometer. The cryogenic echelle spectrometer is a long-slit spectrometer of resolution 100,000 operating over the 1 - 5 um range. The collimator minors are now under construction in the KPNO optics shop, and the grating is already in hand. The mechanical design effort-scheduled for early 1990--will start with an analysis of the collimator assembly, with particular emphasis on the elimination of flexure. Because of the "modular engineering" approach adopted for the IR instruments, future outstanding engineering issues will closely resemble previously solved problems. Initial performance tests will be carried out with PtSi devices, as we monitor and investigate the ongoing rapid improvements in detector technology. Identification of the optimum detectors) both for this instrument and for BUFO is a major goal of the R&D effort. 33 Cryogenic Optical Bench (BUFO). The cryogenic optical bench is a versatile combination of broad- and narrow-band filters, tunable solid etalon Fabry-Perot and order-sorting "linear variable filter" (also capable of stand-alone operation), grism, polarimeter, and stellar coronagraph. The Fabry-Perot and "linear variable filter" were designed and fabricated by OCLI to our specification. The final choice of detector is yet to be made. First light will probably employ PtSi, but ultimate installation of a high quantum efficiency device is definitely required in order to exploit the narrow-band modes. Medium-Resolution Spectrograph. The anay-based infrared spectrograph (CRSP) was a retro-fit of an existing instrument further upgrades cannot be accommodated because of space and configuration constraints. A new design is required because 256 x 256 format detectors are already available and even larger formats are planned. A medium-resolution spectrograph for 1 - 5 um with a resolution about 10,000—intermediate between CRSP and the Cryogenic Echelle Spectrometer-is a high priority instrument for the proposed NOAO 8-m telescopes. This instrument would support both long-slit and cross-dispersed modes. Ten um Spectrometer/Camera. The R&D group has in hand, and has made successful preliminary tests of, 64 x 20 IBC arrays. In broad-band imaging applications at 10 um, the high levels of ambient background radiation will saturate the detector in only 300 usees; much faster electronics are therefore required for a 10 um camera than for any of the instruments construaed to date. The first KPNO 10 um instrument will combine spectroscopic and imaging capabilities, and, as with the rest of our proposed instrument complement, will be based on the "cryogenic optical bench" concept. Fast Electronics. Fast electronics are an absolute necessity to deal with the high background levels encountered at wavelengths around 10 um. They also become necessary for detectors operating at somewhat shorter wavelengths, 3 - 5 um, as the detector format size grows larger. We anticipate the need to operate 256 x 256 InSb detectors in broad-band cameras well before the end of this forward-look period. Even now, the existing imager (IRIM) does not work at optimal efficiency at 3 - 5 um because of limitations imposed by the speed of the system. Initial tests of the next generation fast electronics will be performed at the telescope with IRIM, and immediate benefit to the observatory will accrue from this upgrade. The third application we propose for the fast electronics is in the area of improved image quality. Experiments by Ridgway (KPNO) and others in speckle interferometry have shown that, in the near-infrared, much of the radiation in the image of a point source is actually contained in a single speckle. Consequentiy, a substantial improvement in image quality can be realized by a "fast guiding" system, which simply removes the translational component of image degradation. The proposed "fast electronics" are required, quite simply, to provide the requisite positional feedback for the "fast guider." Infrared array detectors gain much of their sensitivity by virtue of long "on-chip" integrations. We can satisfy the apparently antithetical requirements of "fast guiding" and "long on-chip integrations" in an instrument like SQIID by reading only one channel rapidly, using that information to steer a "fast guiding mirror," while continuing to integrate on-chip in the other channels of the instrument New Secondary Mirrors for the 4-m and 2.1-m Telescopes. Implementation of this "fast guiding mirror" project requires installation of suitable minors at the 4-m and the 2.1-m telescopes. A focal ratio in the range f/60 - f/120 is appropriate, and we envisage the "fast guiding" mirror as a (small) secondary, utilizing the two-axis control concepts developed in the construction of the f/15 IR secondary at the 2.1-m. 34 We also plan to install an f/15 secondary mirror at the 4-m telescope for routine operation of SQIID, BUFO, and other infrared instruments. The cell and unfigured blank for this project exist from the original 4-m design; the blank is in the KPNO optics shop awaiting figuring. The rationale for this new secondary is simple; a radical simplification of instrument design and operation results if we standardize to a single focal ratio throughout the observatory. At the 2.1-m and 1.3-m telescopes we operate at f/15, and the necessary parts exist for the 4-m, making this the cost effective way to implement the standardization. Computer Hardware/Software. As the sophistication of the instruments increases, the need for real time data reduction will become increasingly apparent We have placed heavy emphasis on satisfying this need in the development of future IR instruments for KPNO. For example, the SQIID will pass pictures immediately to the Sun for distortion removal (from the transmissive optics), flat-fielding, mosaic assembly, and multi-color display. The user will then be in a position to continue analysis using IRAF. We anticipate that continued efforts in real time reduction and analysis will be a major concern for the foreseeable future, and we place a high priority upon such developments. We also identify a need for the establishment of appropriate archiving capabilities, both shortlong-term, for the large amounts of data that will be produced by the proposed instruments. and Detectors Present and Future. On completion of the present program of acquisition and laboratory testing of the 256 x 256 PtSi arrays for SQIID, the lab test dewar will be converted for evaluation and testing of the 20 x 64 As:Si IBC arrays. The IBC arrays, which have been in hand for two years, are sensitive out to a wavelength of about 27 um; they will be the arrays for the initial 10 urn and 20 um imager/spectrometer, the prototype of which would use a retrofitted IRIM system. It is crucial that NOAO continue to support a vigorous research and development effort in detector array acquisition, evaluation, and operation. A list of detectors currendy under consideration for future use is given in Table IV. TABLE IV Infrared Detectors Material Current Format 1991? Wavelength Manufacturer Range (u.m) Format PtSi 256 x 256 512x512 1 -3.5 Hughes-Carlsbad InSb 58x62 256 x 256 1 -5 Hughes-Santa Barbara Ge 256 x 256 256 x 256 0.7 - 1.6 Hughes-Carlsbad HgCdTe 256 x 256 256 x 256 1 -2.5 Rockwell 20x64 58 x62 8-27 IBC Hughes-Carlsbad Rockwell 35 All of the IR instruments will profit enormously from utilization of high quantum efficiency detector arrays. Very high performance HgCdTe 256 x 256 arrays for wavelengths less than 2.5 um have been produced for the NICMOS instrument; these detectors will be available for purchase from Rockwell almost immediately. Longer wavelength versions of the HgCdTe array (sensitive out to 4 or 5 um) should also be available on a slightiy longer timescale. SBRC is currendy developing a commercially available high performance 256 x 256 InSb anay, which is expected to be available for order early in 1990. It is essential that one or the other of these array types be acquired very early in FY 1992 in order to have tested and understood devices for the Echelle, for BUFO, and for possible refitting of the other instruments. InSb would be the first choice if SBRC succeeds in fabricating good devices of this type. Further in the future, the subsequent generation of IR arrays will advance in two areas: larger format arrays for the 10 and 20 um regimes with large pixel capacity to accommodate better the very high photon background, and 512 x 512 anays for the 1 to 5 um regime. Both array types will be of great interest to NOAO, and we should plan for acquisition of such devices in FY 1993. Advanced Mirror Coatings Test. The performance of all the IR instrumentation is affected by the reflectivity and emissivity of the telescope optics. The optical design of the telescope system, the intrinsic properties of the mirror coating, the contamination of the mirror surfaces by dust and other residues, and the deterioration of the mirror surfaces are all important factors in determining the system emissivity, and thus the ultimate performance of the IR instrumentation. In the thermal IR, notably around 4 um and 11 um (and perhaps even in the K band), the thermal emission of the telescope system is the dominant source of background at the detector. We propose to initiate a program to evaluate the performance of telescope mirror coatings. We will apply an advanced coating to the optics of the existing 13-inch f/15 low background configuration telescope, a telescope cunendy stored in the IR lab, adapt the telescope tube so that it can be attached to the top surface of the IRIM and CRSP, store the telescope in the 1.3-m dome, and make measurements of the 13-inch telescope emissivity, zenith sky emissivity, and 1.3-m telescope emissivity at 4 urn each time the IRIM (and possibly the CRSP) is mounted on, or removed from, the 1.3-m under clear weather conditions. Use of the 13-inch telescope for these tests will allow us to isolate sources of background emission in the telescope, recoat the mirrors, change coatings, and make mirror cleaning and scattering tests-all of which would be rather impractical with the 1.3-m telescope itself-while at the same time providing a realistic approximation to a real telescope environment and configuration, as well as a comparison baseline with the aluminum coatings on the 1.3-m. The initial coating would be thorium fluoride/silver/chromium, which the NOAO coating lab is now able to produce. This prescription has been in use at the 4-m FTS for several years. If the testing proposed here indicates substantially improved durability and performance of advanced coatings, it will be important to develop the capability to lay down silver undercoated with copper and overcoated with sapphire and tantalum oxide, die preferred prescription in the NOAO supported study to develop a broad wavelength, improved durability, minor coating. The study was performed at the Optical Sciences Center (Song et al., A. O., 24, p. 1164, 1985). Collaborative Ventures. BUFO and SQIID are being produced in collaboration with members of the staff of ST Scl. We expect such collaborations to be highly productive and will actively seek to make additional such arrangements as appropriate. 36 Instrumentation for 8-m Telescopes. The KPNO Infrared Group expects to build instruments for the NOAO 8-m telescopes. The instruments listed in Table III would all be suitable for such an application; we have proposed that the "cryogenic optical bench" approach be adopted, and that the present generation of KPNO instruments be regarded as prototypes for the 8-m telescopes. Optical-Ultraviolet (O/UV) Instrumentation at Kitt Peak. During the last few years, the Kitt Peak O/UV instrumentation program has focused on the development of a multi-object spectroscopic capability for the KPNO 4-m telescope. Our first effort was the adaptation of the Lockheed camera at the R-C focus to hold fiber optic plugboards. Two fiber optic cables, optimized for the red and the blue spectral regions, were also fabricated. Following the commissioning of "Nessie" we began the design and construction of an automatic fiber positioner for the 4-m R-C focus, using an x-y stage and a robot gripper to locate magnetic fiber buttons in the telescope focal surface. The specific design was chosen after a review of similar projects at other observatories. A multi-object spectrograph designed for use with optical fibers and large-format CCD detectors is also under construction for use with the automatic fiber positioner. We expect to complete the fiber positioner and the first phase of the fiber optic bench spectrograph in FY 1990. A plan for the next five years has been defined through discussions among Kitt Peak staff members and with the user community. We are aware of our responsibility to provide instrumentation of use and interest to the community, and we believe that the program we propose will satisfy both ourselves and the community we serve. Our resources are extremely limited. We have chosen to focus on a small number of projects and do them well, rather than trying to complete many new instruments and succeeding on none. In practice, this means that many worthy projects will be delayed or not undertaken at all. The plan outlined in this document will, of course, evolve as we continue to reevaluate our scientific needs and technical resources. Our program beyond FY 1990 identifies three major efforts to bring new capabilities to Kitt Peak and to upgrade existing instrumentation. These include a major effort to enhance the imaging capability at the KPNO 4-m telescope (and perhaps also at the KPNO 2.1-m and the 0.9-m telescopes), to continue to upgrade our CCD arsenal with new devices offering better performance, and to replace our CCD controller systems with modem architecture to take advantage of the improvements in the new generation of CCDs. As resources permit, we will also undertake smaller projects to improve performance of existing instruments. 37 TABLE V KPNO O/UV Program Current Year (FY 1990) KPNO Project Scientist(s) FY 1990 Fiber Positioner Labor (mm) Capital ($K) CCD Development Program Fiber R&D Program 15 30 13 25 2 5 30 10 20 Near-IR Dewar 12 3 Bench Spectrograph New CCD Controller 8 HRIS Camera Instrument Improvement Projects 5 3 7 Total 105 83 Five Year Plan FY 1991 4-m Wide Field Corrector CCD Development Program New CCD Controllers (2) Fiber R&D Program Instrument Improvement Projects Total 5 25 25 30 10 15 2 5 30 72 38 113 FY 1992 4-m Wide Field Corrector 2x2 CCD Mosaic 8 2 16 405 CCD Development Program 25 30 New CCD Controllers (3) 15 15 Fiber R&D Program Instrument Improvement Projects 2 5 30 96 10 467 2x2 CCD Mosaic 4-m R-C Fast Guider 34 212 15 10 CCD Development Program New CCD Controllers (3) Fiber R&D Program Instrument Improvement Projects 25 30 15 15 Total FY 1993 Total 38 2 5 10 40 101 312 KPNO O/UV Program (continued) FY 1994 2x2 CCD Mosaic 4-m R-C Fast Guider CCD Development Program Fiber R&D Program Instrument Improvement Projects Total 52 43 20 30 100 10 5 10 40 52 147 215 147 216 FY 1995 TBD Total 39 Imaging at the KPNO 4-m Telescope. CCD mosaics offer the only way to image significant areas of the sky with large telescopes. NOAO needs to gain experience in the fabrication of CCD mosaics and in the operation and readout of many CCDs simultaneously. Perhaps the best way to gain this experience and to prepare for the 8-m optical imager is to build a CCD mosaic imager for the KPNO 4-m telescope. When 8-m telescopes come on line, the 4-m will be used more heavily for imaging; this would be a valuable instrument to "find" objects to observe spectroscopically with the 8-m. At the 4-m prime focus, a 2 x 2 mosaic of detectors, each with 2048 x 2048 pixels will provide sky coverage of approximately 32 arcmin on a side, or a quarter of a square degree. (This assumes a pixel size of 24 um square, a current "best guess" at what Tektronix will be able to provide.) The scale of such a device will provide approximately 0.5 arcsec per pixel. The scientific motivation for a wide field CCD mosaic is strong, especially for galaxy photometry and deep sky survey programs. For these programs, sky coverage may be more important than spatial resolution, so that larger pixels are the preferred choice. The 2x2 mosaic developed for the 4-m prime focus could also be used at the R-C focus to provide a wider field with higher spatial resolution. The implementation of a CCD mosaic at the 4-m prime focus will require the design and fabrication of a new corrector. Even with a single 2048 x 2048 CCD, the existing doublet conector does not provide a large enough field of view with good imaging. The first step in designing a new prime focus conector is the definition of the operational characteristics we need. The design prepared for the new CTIO corrector is specified, for example, to work with optical fibers over a 50 arcmin field of view, but a new corrector for the Kitt Peak 4-m telescope would be used for imaging. For imaging, we can limit the wavelength range of the corrector to only 1,000 A at a time without refocusing, and we may be able to relax the imaging tolerance somewhat in the ultraviolet. The ADC may become somewhat simpler, since we can tolerate refocusing for different filters. It may also be possible to trade broad-band simultaneous quality for higher throughput at 3,200 A. Nevertheless, the CTIO corrector design is an excellent starting place for a new corrector for the KPNO 4-m. The same mosaic detector may be used at the KPNO 2.1-m and the 0.9-m telescopes if the guiders are adapted to allow a wider field of view. This effort will be included in the overall program through instrument improvement projects to modify existing guiders. The use of the mosaic imager on several telescopes will allow us to get more use out of an expensive instrument. A second direction we will investigate is the development of a high spatial resolution imager, using a fast guider and perhaps a fast shutter to improve image quality. The scientific motivation for high spatial resolution imaging is strong. Examples are the resolution of luminous O and B stars in young clusters in external galaxies and the study of the nuclear regions of galaxies. A detector with pixels of 20 - 27 um (the same choice as for 4-m prime focus) would be appropriate and a variety of options is appearing in the marketplace. On nights of excellent seeing we would expect to achieve images better than 0.7 arcsec FWHM. Similar instruments are being developed at the CFHT (see the CFHT Newsletter, No. 20, p. 5) and at ESO (see "Very Large Telescopes and their Instrumentation," p. 751), and by a group at Johns Hopkins for use on Carnegie's Dupont Telescope. We will review these designs critically for adaptation to our own program. Experience with this technology would naturally be valuable for the NOAO 8-m telescopes as well. While the foremost argument for providing a fast guiding camera is the good science it will do, several other arguments support this choice for the O/UV program. Such observations are of interest 40 to many members of the Kitt Peak scientific staff, and this direction takes advantage of many of the strengths within the O/UV group, both scientific and engineering. The experience we gain with this technology will be of value for the NOAO 8-m telescope program. Adaptive optics are increasingly recognized in the community as an essential technology for the next decade, and the effort to build a fast guiding camera will lead us in this direction. Higher resolution spatial imaging is of interest to the KPNO IR group, and we may expect to share some of the necessary development work with the IR group. Implementation of rapid guiding at the 4-m is contingent on success at near infrared wavelengths and on improvements to overall image quality at the 4-m through changes in the thermal environment, since rapid guiding is most effective if the seeing is already good. The CCD Development Program. The acquisition, characterization, evaluation, and implementation of new CCD detectors will remain a high priority, since all other aspects of our instrumentation program depend on our detectors. We need to work actively to obtain new detectors, which will offer improved quantum efficiency, cosmetic quality, read-noise, and format. NOAO is one of the few organizations able to test CCDs on a significant scale. The recent "CCDs in Astronomy" conference established clearly that several new types of CCDs are now becoming available. These new devices offer a range of physical size, pixel size, and wavelength range. We already have access to large pixel, large-format devices from Tektronix through NOAO's original purchase orders and through contracts with the STIS project. We are negotiating with Tektronix to convert our order for seven more 512 x 512 pixel devices to 1024 x 1024 devices to provide for wider field imaging. The new opportunities to obtain CCDs through a "foundry" approach is being explored joindy with CTIO and Steward Observatory, and we expect to place a joint order in FY 1990. High on our priority list are the needs to replace the TI device in the Cryogenic Camera with a larger format flatter, higher blue sensitivity detector (perhaps a Thomson or Ford Aerospace device), and to offer a 2048 x 2048, small-pixel detector on the R-C spectrograph, which would yield about the same resolution as the TI devices with a factor of 2.5 increase in wavelength coverage. This same detector, if available at prime focus on the 4-m, would offer spatial sampling of about 0.25 - 0.3 arcsec per pixel, with a sufficient field to survey distant galaxies for planetary nebulae, for example. Data Acquisition and CCD Control. KPNO has a chronic shortage of dewars and controllers and must put effort into building more modem controllers and dewars. It is essential that the associated CCD hardware (controllers, displays, etc.) be upgraded to handle the new, large-format, low noise detectors with four-quadrant readout and multi-phase pinned operation. Our effort to upgrade controllers will continue over the next five years to meet new standards of performance. The telescope and computer environment for data acquisition and reductions is a third area to which we need to devote some effort, especially for large-format CCDs. We need to streamline the flow of information to the observer through improved and more interactive acquisition and quick-look software (e.g., standard star magnitudes, transformation coefficients, the "flats" command at CTIO, automatic chip formatting for standard star observations, etc.). The observing and data reduction procedures need to be more integrated into the IRAF environment, which is available through the more modern controller architecture. Other observatories are taking the lead in the development of observing software and user interfaces; we will learn from their efforts and adopt new approaches for Kitt Peak. 41 Fiber Optics R&D Program. Fiber optic spectroscopy offers a substantial increase in observing efficiency. In FY 1991 we will concentrate our fiber development effort on evaluation and research (scrambling, fiber connectors, focal ratio degradation, etc.), and gain some experience with the bench spectrograph, which will be completed in FY 1990. By FY 1992, we will be under some pressure to provide a blue fiber cable to the bench spectrograph at the 4-m, and we are optimistic that by then blue fiber performance will have improved significandy through development efforts now underway at vendors. We will also need to investigate techniques for guiding with fibers for use with the WIYN telescope. For the NOAO 8-m telescopes we need to develop a tiltable gripper and learn to deal with the very large area of the focal surface. Miscellaneous Projects. In addition to the four areas outlined above, we will continue to support at a modest level instrument improvement projects to enhance performance at all of our telescopes. We hope to complete one or two of these smaller projects each year, subject to the availability of resources. The specific choices will be made on an annual basis depending on scientific merit and urgency. The list below is not in priority order and is by no means inclusive; it is only intended to demonstrate the many tasks ahead of us. • The throughput of the R-C spectrograph does not appear to be as high as we would like, in comparison to the Goldcam and to the performance of the CTIO R-C spectrograph. Barden, Massey, DeVeny, and Armandroff (KPNO) are currendy investigating the throughput of the 4-m R-C and Echelle spectrographs. Pending the results of this study, we may want to consider implementing a new high-throughput camera for the R-C and making better use of anti-reflection coatings on gratings and optics. A factor of two improvement in throughput may be possible. The use of a camera like the air Schmidt at CTIO with a dedicated CCD may be the answer. We may also wish to consider modifications to the R-C spectrograph to allow the use of an echellette, perhaps with cross dispersion provided with a grism. • Following the completion of the 4-m bench spectrograph in FY 1990, we anticipate that several upgrades will be needed to bring the instrument to its full level of performance. These include the addition of a long focal length camera, a blue optimized camera and a blue fiber cable, a "superdensepak" cable, cross dispersion, and the completion of the Schmidt collimator. Another possibility is the addition of a fourth fiber cable with smaller diameter fibers. These upgrades will take place in the period FY 1992 - 1994. • The Intensified Reticon Scanner at the KPNO 0.9-m telescope is aging, and we would like to replace it with a CCD camera and detector analogous to the Goldcam at the KPNO 2.1-m telescope. The camera already exists in the form of the retired JJDS camera from the gold spectrograph, but we need to acquire a dedicated detector and fabricate a dewar. Small pixels are essential. One alternative might be to use the TI CCD retired from the Cryogenic Camera when that device is replaced. • The guiders on the KPNO 2.1-m and 0.9-m telescopes need to be modified to accommodate a larger field of view to make use of the CCD mosaic imager to be built for the 4-m prime and R-C foci. Use of this camera on these smaller telescopes will provide large fields especially for imaging relatively nearby galaxies. At the 0.9-m, the CCD mosaic will cover a field of view of at least 46 arcmin square with a sampling scale of 0.7 arcsec per pixel. At the 2.1-m telescope, the field of view will be at least 21 arcmin square with a sampling scale of 0.3 arcsec per pixel. Use of the 42 CCD mosaic on the 2.1-m and 0.9-m telescopes will allow us to make more frequent use of this expensive piece of hardware. • The consolidation of the #1 and #2 0.9-m telescopes into a single 0.9-m facility will impact severely conventional photometry. If a CCD is used as a detector to replace the photomultiplier for all-sky photometry, the improved ability for sky subtraction available with an area detector may allow us to achieve fainter limiting magnitudes in bright time than we do now with a photoelectric photometer in dark time. As reported at the "CCDs in Astronomy" conference, CCDs are capable of milli-magnitude precision photometry if properly used. The disadvantage of CCDs for photometry is the need for image processing to derive magnitudes, and the relatively low efficiency associated with reading out a full format image. During the next five year period, we hope to develop a CCD photometer that automatically sky subtracts and extracts a stellar magnitude from a small format image in nearly real time, leaving the observer with a simple magnitude measurement With the Sun computer available at the 0.9-m telescope and the new CCD controller architecture integrated into IRAF, the process can be automated to a high degree. Observing efficiency will be a high priority for this system. Initially, this project will emphasize software development The Burrell Schmidt may also be a place where one could do CCD photometry during bright time. If we could develop the hardware and software to do all sky photometry efficiendy and quickly without the need for the astronomer to become involved in image processing, we might be able to make up in part the loss of the #1 0.9-m telescope. For a CCD photometer to work efficiendy at the Burrell Schmidt we would have to address die problem of pointing and acquisition with this telescope for a small field. • The recent success of cross dispersion with grisms at the coude" spectrograph at the 2.1-m telescope leads us to propose the design, fabrication, and installation of new grisms there. If we can provide a detector with enhanced blue sensitivity for the Cryogenic Camera, that instrument will also need new grisms for the blue spectral region. • We expect to continue to encourage astronomers who develop specialized instrumentation to bring it to Kitt Peak for use by visitors. This allows us to provide a wider range of instrumentation and scientific capability despite the restricted resources available to build new instruments within KPNO. Examples of this approach are the Perm State Fiber Optic Echelle and the Goddard Fabry-Perot. Some resources from the O/UV group may be required to install these new instruments on Kitt Peak. • During the next five years we expect the new WIYN telescope to come on line, and we must make provision in the instrument program to provide the multi-object spectroscopic instrumentation for the telescope. This instrumentation may in fact be the bench spectrograph and automatic fiber positioner now being fabricated for the KPNO 4-m telescope. If resources are available, we may prefer to copy this existing instrumentation rather than move it from the 4-m. In any case, some modification or upgrade of the spectrograph, the fiber cables, and the positioner will surely be required for use on the WIYN telescope, and this work must figure in our long range planning. About six months before WIYN is complete we will need to modify the positioner or build a copy. Schedule. We anticipate the completion of one segment of the 4-m imaging program each year from FY 1992 1995. We expect to complete the new wide field prime focus corrector in FY 1992, the mosaic dewar 43 in FY 1994, and the R-C fast guider in FY 1995. Parallel with this effort there must be an ongoing program to evaluate new CCDs and upgrade existing detectors, implement the new CCD controllers, and develop optical fiber technology. Instrument improvement projects, of the type oudined above, may be completed at the rate of one to two per year. Such projects should be of limited scope and be tighdy defined, so that completion within one year is possible. Priorities for such projects will be redefined annually. The CTIO program for optical instrumentation closely parallels the one at KPNO. The two groups will work together with one observatory or the other taking the lead in specific aspects of the program. CTIO. During FY 1991 - 1995, CTIO's work in instrumentation will take advantage of technological advances in two main areas: detectors and communications. In both the visible and infrared, we expect a new generation of larger detector arrays to become available; these will increase the number of pixels in the focal plane by an order of magnitude or more. In some cases, the larger-format arrays can be retrofitted into existing instruments, but in others new instruments must be constructed. Advances in fiber optics, microwave technology, and the availability of relatively inexpensive satellite communications mean that remote observing and large-scale computer networking are realistic possibilities for CTIO. We intend initially to explore these capabilities on an experimental basis, subsequendy implementing those that appear useful and cost-effective. In addition, CTIO is currendy in the midst of a process of upgrading instrument control and data acquisition capabilities to make use of present-day computer technology. Our goals are improved data handling, better user interfaces, more internal standardization of hardware and greater overall reliability. Auxiliary instrumentation, such as TV acquisition and guiding capabilities, are being upgraded also. In the long-term, a third generation of infrared arrays is likely to become available. In the visible, improvements in packaging and decreases in costs will permit construction of very large-format arrays by means of mosaics of large-format CCDs. The next generation near-infrared imager and spectrometer will be designed to handle not only the second-generation arrays but the somewhat larger formats anticipated thereafter. In the optical, our original instrumentation has up until now been able to handle advances in detectors. With the advent of large-format arrays and mosaics this will no longer be the case and replacement of instruments will be required. This process has already begun with construction of the new CTIO 4-m PF corrector/ADC and the large-format PF CCD imager and will continue with the Faint Object Spectrograph/Camera and the mosaic imager. Table VI presents an overview of the five year plan for CTIO instrumentation. Specific projects are described in more detail below. The FY 1990 total reflects the budget figures given in the FY 1990 provisional program plan. 44 TABLE VI CTIO Instrumentation Program Current Year (FY 1990) CTIO Project Scientist(s) Manpower (months) Capital ($K) FY 1990 4-m PF Corrector/ADC Ingerson 23 5 4-m Large-Format PFCCD 2nd Generation IR Imager CCD TV Acquisition Cameras Walker 32 11 Elias, Gregory 22 28 Walker, Baldwin 24 34 Ingerson, Walker Ingerson, Elias, Gregory Ingerson 32 32 9 6 9 5 CCD Controllers IR Anay Controller Motor/Device Controller Telescope Control (1.5-m, 0.9-m) Site Survey Weller 3 1 Staff 0 45 154 167 16 2 Total Five Year Plan FY 1991 4-m Large-Format PFCCD 2nd Generation IR Imager 4-m Seeing Improvements CCD TV Acquisition Cameras CCD Controllers Rutgers F-P Upgrade Remote Observing Optical/TR Focal Reducer Walker Elias, Gregory 4 19 30 41 Walker, Baldwin 16 36 Ingerson, Walker Phillips, et al. Ingerson, et al. 60 25 25 14*** Heathcote, et al. Total 5 5 5 5 161 147 FY 1992 2nd Generation IRS CCD Mosaic Implementation of Large-Format CCDs SOAR Imager Instrument Upgrade Project Computer improvements Elias, Gregory 58 20 50** Walker, et al. 26 55 Staff Staff Staff 50 70* 10 15 0 164 356 Walker, et al. Total 96 70 FY 1993 CCD Mosaic Walker, et al. 20 SOAR Imager Active-Optics Camera Staff 50 Suntzeff, Ingerson, Weller Elias, Gregory 35 Walker, et al. 26 55 Staff Staff Total 10 55 Ten um Camera Implementation of Large-Format CCDs Instrument Upgrade Projects SOAR Telescope Commissioning 45 10 14 165 65** 120* 5 90** 0* 390 CTIO Instrumentation Program (continued) FY 1994 Ten um Camera Implementation of Large-Format IR Array Implementation of Large-Format CCDs Instrument Upgrade Projects SOAR Telescope Commissioning Computer Improvements CTIO Project Scientist(s) Manpower Capital (months) ($K) Elias, Gregory Elias, Gregory 15 5** 15 80 Walker, et al. 20 90 Staff Staff Staff Total 55 155 0 80 139 410 Total 139 431 34 0* FY 1995 TBD * Total costs are shown for SOAR projects. We assume that we will be able to hire manpower equivalent to 2 FTEs in FY 1992, with SOAR funds, and that SOAR will cover all direct capital costs. ** Joint project with KPNO or duplication of KPNO instrument Only CTIO resources are indicated. *** Joint project with Rutgers. Only CTIO resources are indicated. 46 Optical Instruments. The acquisition of larger and better CCDs continues to be a central theme when planning for future optical instrumentation. NOAO has on order a small number of 2048 x 2048 Tektronix detectors, but these, when delivered, will only fill the most pressing needs for large-format CCD detectors. Given their high price tag~$100K each-and the difficulties in production, it seems unlikely that all requirements can be filled through additional purchases from Tektronix. Instead, as noted earlier, CTIO plans to join with KPNO and Steward to obtain custom CCDs fabricated by (e.g.) Ford Aerospace. The cost of a production run is likely to be $50K, with another $50K of effort for packaging and thinning and a likely (but not guaranteed) yield of 10 - 20 useful devices. Thus, NOAO takes some risk but stands to gain detectors with the characteristics we want, in reasonable quantity and at lower unit cost than is possible by any other route. Participation in a consortium offers the potential of participating in more than one production run, thereby reducing the risk. The instrument development plans described below presume the availability at CTIO of roughly a dozen large-format CCD detectors over the next three to five years. The first project is the construction of a new prime focus imager. In its initial form it will accommodate CCDs as large as the Tek 2048. The design is flexible enough to allow imaging with a mosaic of CCDs up to 110 mm across. Two additional capabilities will be provided: a low-resolution (slidess) spectroscopy capability and "short scanning," i.e., clocking out the CCD a line at a time while simultaneously the CCD is moved relative to the star by the same amount. The basic instrument is scheduled for completion in FY 1990, and the additional capabilities will be added as needed, probably in FY 1991 - 1992. The contract has now been let for purchasing the materials and fabricating the recentiy designed CTIO 4-m Atmospheric Dispersion Corrector. We hope to see the PFADC tested and installed by early FY 1991. By the end of FY 1991, it should be in routine use. CCDs larger than the 2048 pixel (55 mm) square Tektronix chip are at least several years in the future, given the apparent lack of any commercial or military incentive for their production. Therefore in order to cover wider fields at a resolution sufficient to sample star images adequately in subarcsec seeing, a mosaic of CCDs will be required. As a design goal we are considering a mosaic of four Tektronix 2048 CCDs in their normal packages. This mosaic will be used on the CTIO 4-m at both the Cassegrain and prime foci. Since one Tek 2048 will generate 8 MB of data, preprocessing of data from the mosaic is mandatory. Tyson (Bell Labs.) has designed a dedicated preprocessor capable of generating in near real time a composite full field frame from which all instrumental signatures have been removed. We will either duplicate Tyson's hardware or produce a functionally equivalent module within the framework of our CCD/R controller projects. We expect to build the system in FY 1990 - 1991. We are actively seeking potential partners outside NOAO to share in the large capital costs. A need has long been perceived at CTIO for an instrument similar to the extremely successful European Southern Observatory (ESO) instrument, EFOSC. This instrument provides both an imaging and spectroscopic mode, with the possibility of switching from one mode to the other in a matter of seconds. Multi-object capabdity is provided by means of aperture plates that can be prepared, in near real time, from exposures taken in the imaging mode. Since an instrument with characteristics similar to those proposed here is under consideration for the Southern Observatory for Astronomical Research (SOAR) 4-m telescope, a particularly attractive route would be to construct "CTIO-FOSC" in collaboration with the other members of the consortium and use it on the CTIO 4-m, pending completion of the SOAR telescope. 47 There has recentiy been a considerable growth of interest in "synoptic" programs. These typically require the collection of only modest amounts of data per night, but the observations must be repeated at regular-often nighdy-intervals over an extended period of time. CTIO is exploring the possibility of establishing a synoptic/service direct imaging program. In order to establish a parallel spectroscopic program we intend to construct a low-cost, dedicated "synoptic" spectrograph, in the form of a fiber-fed bench-mounted echellete spectrograph for the 1-m telescope. Recent instrument development at the Dominion Astrophysical Observatory (DAO) has produced a simple "active-optics camera" that can improve the average seeing from 0.8 to 0.5 arcsec at the Canada-France-Hawaii Telescope (CFHT). During FY 1992, we intend to produce a similar instrument to be mounted at prime focus or Cassegrain on the CTIO 4-m telescope. We do not plan to construct any major new fiber-fed instruments in the mid-term, but we will continue to work to improve the efficiency and flexibility of the bench mounted spectrographs. For example, we believe that fiber technology will continue to improve, allowing us to replace our present fibers with fibers which will work farther into the ultraviolet (UV). Infrared Instruments at CTIO. The present generation of infrared instruments at CTIO uses arrays with 58 x 62 pixels. CTIO's instrumentation plans in the infrared are focused on making full use of larger format arrays, which are also expected to have smaller pixel sizes, lower read-noise, and lower dark current. CTIO's current complement of array-based instruments comprises a 1 - 5 um imager and a 1 - 5 |im spectrometer. We intend to replace these with second-generation instruments and to add a 10 um imaging capability and an "IR-CCD" capability during the next three years. In the longer term, we expect that 512 x 512 arrays will become available, either as single devices or as buttable arrays. We will design our second-generation instruments to accept these larger arrays whenever possible. CTIO will start construction during FY 1990 of a second-generation imager. The dewar design will be copied from the present IR spectrometer, limiting the additional mechanical design work to necessary modifications and the internal optics. There will be a full complement of discrete filters. In addition, there will be one or two internal linear variable filters. It now appears likely that a low-resolution (R - 700) linearly variable Fabry-Perot etalon can be produced. Following construction of the second-generation imager, we will construct a second-generation IR spectrometer. The new instrument will have an optical design better matched to the larger formats and smaller pixels of second-generation arrays. We anticipate only a small increase in the physical size of the instrument. The spectrometer would be optimized for the CTIO 4-m telescope. It would have resolutions (2-pixel) from R = 200 - 3500, plus a non-dispersing imaging mode. Slit length would be nearly 3 arcmin on the CTIO 4-m telescope and 8 arcmin on the CTIO 1.5-m telescope. It is our intention to design the instrument to be usable on the KPNO telescopes, which would permit KPNO to copy the design in FY 1992. KPNO intends to begin design of a 10 um imager in FY 1992. CTIO will construct a duplicate of the KPNO instrument A 10 um imager should provide diffraction-limited (0.5 arcsec) imaging on the CTIO 4-m telescope. Although the infrared is commonly thought of as beginning at 1.0 um, where silicon CCDs cease to be sensitive, conventional visible wavelength instruments should work out to nearly 2 um, since it is only at this wavelength that thermal emission from the sky, telescope, and instrument become significant 48 when compared with airglow. The ideal detector material is germanium, which has a wavelength cut-off around 1.8 um. If Ge arrays that are to be delivered to NOAO prove to have good performance, we would install them in dewars that can be used with our visible wavelength spectrographs and imagers. Most needs could be met by Ge arrays in a CCD direct dewar attached to a copy of the red air Schmidt camera, which is now used on the CTIO 4-m R-C spectrograph. Additional Projects. The next three years will be an intensive period of integration and consolidation of new communications equipment, which is now being installed. The La Serena-Cerro Tololo Microwave System should soon become operational. We hope to have this link fully debugged by mid-1990. NASA has agreed to pay for an experimental test of a wide-band (56 kbaud) satellite link from Cerro Tololo to the Internet Their purpose is to improve communication with astronomical institutions in the Southern hemisphere and to support experiments in remote observing with telescopes on Ceno Tololo; NASA will pay all the costs of this link for three years. We expect it to become fully operational in mid-1990. We must leam how to use all these new tools efficiendy. During the three year duration of the NASA experiment we expect to refine our ability to do both local and distant remote observing so that we will have the knowledge and experience that will be required to build and support future large telescopes. Within three years, our goal is to have a fully integrated computer network, which will allow local remote observing to be done easdy and transparentiy from any "local" site and as efficiendy and rapidly as the bandwidth permits from any "distant" site. We are now building a new generation of detector control systems. These new controllers are far more powerful than what we are now using and will also be more flexible and reliable, as well as cheaper to replicate. They will also permit a high degree of compatibility between IR and CCD controllers. Over the next three years, we intend to replace all of our current CCD and IR controllers with the new system. Part of the specification of the new Detector Control System is a mechanism of peripheral instrument control (motors, shutters, filter bolts, spectrographs, etc.). During the next three years, we will be gradually transferring control of all such devices to the new system. This should gready increase reliability and flexibility as well as providing greater integration between processes of signal detection and control of ancillary devices. During FY 1991, we will finish upgrading the telescope control systems by cloning our new TCS to the 1-m and the 0.9-m telescopes. CTIO has recentiy completed a prototype CCD-based acquisition camera, similar in concept to the Lick CCD-TV, but substantially more compact The prototype will be repackaged for greater ease of use, and will be replicated beginning in FY 1990. Our intention is to equip all the telescopes on the mountain with this camera by the end of FY 1991. 49 VI. SCIENCE AT NSO The study of the nearest star continues to advance our knowledge of its structure and evolution, but our understanding of the Sun is far from complete. Current research programs are investigating the internal dynamics of the Sun and exploring the structure of its photosphere and upper atmosphere. A. Internal Dynamics. One of the most active areas of research is that of helioseismology, which uses acoustic waves to map the interior structure of the Sun. The objectives and program of the Global Oscillations Network Group (GONG) are described elsewhere in this plan, but other programs in helioseismology are also being carried out While the GONG project is the primary path by which NSO is probing the solar interior, there are several other programs whose main aims are the study of the upper convection zone. Since we are now at the maximum of the solar cycle, the majority of these projects are focused on the relationship between solar activity and the solar oscillations. The projects fall broadly into two classes-the effects of activity on the properties of the oscillations and the use of the oscillations to probe the subsurface structure of the activity. It was discovered recendy, from observations at the Vacuum Telescope on Kitt Peak, that sunspots absorb a significant fraction (up to 50 percent) of incident p-mode acoustic wave energy. In addition, small sunspots (pores) and areas of enhanced magnetic activity (plage) also exhibit appreciable absorption. Observations obtained with the Vacuum Tower Telescope at Sacramento Peak suggest that the absorption of the oscillation energy by a sunspot disappears while a flare is occurring. NSO plans to observe oscillations over the next five years during the peak of the activity cycle. During this time, the effect of solar flares on the oscillations may be studied with a higher success rate than in the past. It will also be possible to obtain more statistics on the interaction of the p-modes with sunspots. The absorption of acoustic waves by sunspots and the probable emission of waves by a flare hold out the hope of studying the subsurface structure of sunspots and active regions by a form of acoustic tomography. Cunendy, a number of theoretical groups are working to uncover the mechanism causing the absorption effect. Once this and the emission problem are solved, we will be able to study the evolution of solar active regions from a three-dimensional point of view. In addition to using the absorption/emission as a probe of active region structure, it may also be possible to use these effects to detect the presence of spots on the back side of the Sun. The feasibility of the technique is cunently under investigation and, if proven valid, may enable us to study active regions throughout their lives. These phenomena will be observed through use of the spectromagnetograph and High-/ helioseismograph at the Vacuum Telescope on Kitt Peak. Simultaneous observations with the Fourier Tachometer and the Universal Birefringent Fdter at Sac Peak will also be used to address the same issues. In addition to the absorption and emission of p-mode oscillations by active regions, the oscillations may be used to probe the horizontal velocity fields below active regions. Recendy, a three-dimensional analysis of oscillation data has been developed in which the signature of the oscillations is a set of trumpet-shaped surfaces in the f, - L - v volume. When the surfaces are sliced at a constant temporal frequency, a set of rings appears. The central positions of these rings are proportional to an average over depth of the two horizontal components of the velocity of the solar plasma. By analyzing rings from data obtained at different heliocentric coordinates, the horizontal flows can be determined inside and outside active regions. Observations from the Vacuum Tower Telescope at Sac Peak show that the presence of an active region produces a 20 m/s difference in both meridional and rotational velocities. With the aid of inverse theory, we can determine maps of the 50 horizontal velocities as a function of depth beneath active regions. These maps should be of great value in the study of the evolution of active regions. Coupled with a tomographic determination of the subsurface magnetic field structure, a three-dimensional picture of active regions will emerge. The ring diagrams can also be used to study the global pattern of convection in the Sun. A mosaic of ring diagrams covering the solar disk will help to determine whether the largest scales of cells in the convection zone, if they exist, are zonal, sectoral, or tesseral. Synoptic observations obtained with the High-/ Helioseismograph over the course of a solar cycle will be essential to build up the statistics that are needed to answer this question and to determine if the cell shapes evolve with the solar cycle. Studies of flows associated with active longitudes can also be made, and the possibUity that the formation of active regions is presaged by a certain flow pattern in the convection zone holds out the hope that a long-range forecast for specific solar activity may be in reach. While we have known of the existence of the solar cycle for a century, we still do not understand its origin. The most developed explanation is dynamo theory which, until recendy, was observationally constrained mainly by the solar cycle period and the structure of the butterfly diagram. The results of helioseismology have shown that the interior rotation rate is not constant on cylinders as some numerical convection zone models had predicted. As a consequence, some theorists have hypothesized that the seat of the solar cycle is at the base of the convection zone. However, the precision of current determinations of the solar oscillation spectrum is still too poor to distinguish between some very different patterns of the interior rotation rate. The observational programs that NSO will undertake in the next five years will provide information that will be of great importance to dynamo theorists. GONG will provide the most accurate determination of the internal rotation rate ever obtained. The High-/ Helioseismograph will yield maps of the horizontal flows as a function of both depth and longitude in the convection zone. Solar interior tomography will allow the subsurface evolution of active regions to be followed. These prospective observations are exciting, as they will provide unprecedented calibration for dynamo theory. Such observational advances must be accompanied by corresponding theoretical progress in order to increase our understanding of the solar cycle. B. Magneto-Convection. One of the primary modes of energy transport in stars involves compressible, turbulent, three-dimensional flow in a strongly radiating medium. It has proved to be extraordinarily difficult to model. The Sun-the only star on which we can see surface detail-provides an excellent opportunity for developing better theories of convection. Convective motions penetrating into convectively stable regions can propagate as internal gravity waves. Theory suggests that their wavelengths are less than one arcsec, that they remain unresolved in current velocity observations, and that they broaden line profiles. These waves could carry considerable fluxes of non-radiative energy, which may contribute significandy to the heating of the upper atmosphere. Since the causes of photospheric and chromospheric heating remain unidentified, the detection of these waves poses a challenge to solar astronomers. The solar magnetic field is responsible for the structure and dynamics of most of the phenomena visible in the solar atmosphere. Magnetic flux appears in the solar atmosphere predominandy (some researchers believe almost exclusively) as subarcsec elements, or flux tubes, with field strengths around a kilogauss. The emergence, transport, equilibrium, coalescence, submergence, and decay of these elementary flux tubes are topics of intense interest in solar physics. Waves along the flux tubes and dissipation by reconnection and anomalous plasma processes must certainly contribute to the heating of the upper solar atmosphere. These higher altitude processes, in turn, are driven by the interaction of strong flux elements and small-scale convection in the photosphere. The evolution of 51 sunspots also involves a great variety of flux transport and reconnection processes at the subarcsec level. The storage of energy in active regions and its release and transport during flares involve complex magneto-hydro dynamics (MHD) and plasma processes that are only partially understood for lack of sufficient spatial resolution. The release of flare energy is also believed to occur on extremely small scales. Aside from their intrinsic astrophysical interest, solar flares represent an important feature of solar activity. They affect the Earth by disrupting communications, long range power transmission, and manned and other space flight operations. Since the initial flare process is predominandy magnetic in character, the same techniques (subarcsec polarimetry) may be used as for the quiet Sun. The small-scale dynamics of the chromosphere, prominences, and corona appear to be controlled by magnetic flux tubes, driven by photospheric motions below. Observations from space (e.g., SMM, HRTS, rocket flights) have given tantalizing glimpses of the complex behavior of the plasma and fields in the upper atmosphere, but improved understanding of the physical processes involved requires more regular observations at high spatial resolution. Space platforms will eventually provide the ultimate in spatial resolution of the Sun. However, observations at the highest possible resolution from space are still years in the future. A wellconceived long-term program of high-resolution solar observations from the ground is an important predecessor and partner for any space program apart from allowing many critical experiments. Ground observations serve to frame incisive questions for space experiments, to develop plasma diagnostic techniques, and to provide the time history of the solar magnetic cycle that is necessary to place the space experiments into proper perspective. During the operational period of a high-resolution solar space mission, the NSO can play a critical complementary role, as was amply demonstrated with Skylab, OSO-8, and the Solar Maximum Mission (SMM). The NSO intends to play an active role in coordinating ground-based observations during the flight of the planned Orbiting Solar Laboratory and other space missions such as SOHO, and it will continue to provide support for rocket and balloonborne experiments. Most of the basic physical processes in the solar atmosphere occur on spatial scales below one arcsec. The interaction of convective motion with the small-scale magnetic field pattern, which generates most of the fine structure we observe and which heats the overlying layers, can be studied in detail only with solar images that attain at least 0.3 arcsec resolution for at least several hours. NSO plans to mount an intensive program of instrument development, centered on adaptive optics, to meet this stringent requirement The successful implementation of adaptive optics is a prerequisite for any consideration of new large aperture solar telescopes. The NSO program to achieve high spatial resolution imaging of the Sun includes the development of observing techniques, including speckle and active optics reconstruction of solar images; development of techniques to compensate telescope polarization using Liquid Crystal Devices to improve measurements of vector magnetic fields on the Sun; development of flare prediction algorithms based on the relationships between vector magnetic fields, chromospheric geometry, and observed flare locations and magnitudes. Planned programs include observations and modeling of the effects of magnetic fields on energy transport to provide data on the storage and release mechanisms that control solar activity; analysis of the transition in the solar atmosphere from convective and radiative (photosphere) to wave and radiative (chromosphere) energy transport and heating, by means of the recendy developed correlation tracker system used in conjunction with agile (rapidly tiltable) minors to stabilize the solar image; and use of very narrow band filter systems (20 mA bandpass) to observe 52 the height dependence of convective overshoot and the concentration of magnetic flux tubes into larger structures. Infrared observations offer unique advantages for probing the magnetic and thermal structure of the solar atmosphere. For example, because the splitting of spectral lines in the presence of a magnetic field increases direcdy as the square of the wavelength, it is possible in the infrared to measure the intrinsic strength and orientation of solar magnetic fields with a sensitivity that is difficult or impossible to achieve in the visible region. The discovery at NSO of numerous atomic emission lines near 12 um has made this possibility a reality, but the work of exploiting these lines has just begun. The vibration-rotation bands of carbon monoxide (CO) at 2.3 and 4.6 um are a sensitive thermometer that can be used to probe temperature inhomogeneities in the upper photosphere direcdy and to test the validity of widely-used atmospheric models. Moreover, the CO lines show prominent intensity oscillations that can be used to study the penetration of solar p-modes into the upper atmosphere. The infrared continuum is also a valuable diagnostic. The 1.6 um continuum arises deeper in the solar atmosphere than any other observable wavelength and provides a unique window on magnetic fields and convection below the photosphere. Because the infrared continuum approximates the Rayleigh-Jeans portion of a blackbody curve (with intensity direcdy proportional to temperature), it is straightforward to study temperature variations both vertically and laterally. For years, infrared observers struggled with single-element detectors while array detectors changed the face of optical astronomy. The era of infrared arrays has now arrived, and their revolutionary effect has already been felt. A small-format (58 x 62) array for the 1 - 5 um region has been successfully used at the McMath Telescope during the last year, and large-format (256 x 256) infrared detectors are currendy undergoing tests at NOAO. The McMath Telescope is particulariy well-suited to solar infrared observations. Its all-reflecting optics provide high transmission, low scattering, and low instrumental polarization at all infrared wavelengths accessible to ground-based observation. Its large aperture is a major advantage both for angular resolution (0.25 arcsec at 1.6 um, less than 2 arcsec at 10 um) and for light-gathering power (the number of solar photons per Doppler line-width is some 20 times smaller at 10 um than at 0.5 um). The scientific potential and emerging technical capabilities have stimulated broad interest in an enhanced solar infrared facility. Studies indicate that an upgrade of the McMath to 4-m aperture is feasible and would be a cost effective way to respond to this need. A new era in coronal and prominence studies is dawning at NSO, due to several technological advances during the past decade. The application of diode array detectors and video cameras to coronal emission line spectroscopy, for example, permits more accurate subtraction of the sky background, shorter exposures (because of the higher quantum efficiency of solid state detectors) and much greater flexibility in real time or post-observational image processing. As a result, it has recendy become possible to record accurate emission line profiles in coronal holes at intensity levels below a millionth of the disk brightness. This achievement opens the prospect of detailed studies of the physical parameters of coronal holes from ground-based telescopes, with important implications for solar and solar terrestrial physics. The importance of small-scale structures in the corona (with scales of a few arcsec or less) has been emphasized recendy, both by observational and theoretical studies. The heating of the corona, for example, has been postulated to occur through a continuous process of magnetic field reconnection. The free energy contained in the coronal field is thought sufficient to maintain the corona's high temperature. If this process occurs, it must take place at small spatial scales. Evidence for such field reconnection, with associated heating, has recendy been obtained at NSO, with the 20-cm coronagraph. 53 Further observations and study of this process and others like it will be vigorously pursued with this and with future instruments. The maximum of the current solar activity cycle is expected in 1991. NSO's plans for coordinated flare research during the coming solar maximum as part of the "Max '91" program suffered a major setback with the collapse of NSF support for this program. However, with existing instrumentation NSO will continue to provide users with facilities for flare observations as a part of regular visiting scientist support. For example, the process of magnetic energy build-up in preflaring active regions can be followed with the multi-channel magnetometer on Kitt Peak and with imaging spectroscopy at Sac Peak. Ground-based flare observations, taken in coordination with rocket, balloon, and Solar-A experiments, will continue to provide users with the best available data on the impulsive energy release process in flares. Hydrodynamic and magnetohydrodynamic processes dominate many of the phenomena observed on the Sun. To study these, high spatial resolution is essential. NSO plans to augment the focal plane instruments at the Vacuum Tower Telescope (VTT) at Sac Peak for high-resolution imaging and spectroscopy of the solar chromosphere. In collaboration with the High Altitude Observatory (HAO), an Advanced Stokes Polarimeter will be built and used at the VTT for studies of the build-up and release of magnetic energy in flares, for research on the dynamics of spicules and on MHD wave phenomena in sunspots, and for many other projects. 54 Vn. INITIATTVES FOR SOLAR ASTRONOMY The key initiative within NSO during the coming decade is the construction, with international partners, of a new facility for high-resolution studies of the Sun. There is in addition wide-spread interest in developing a capability for synoptic observations of the analog of solar activity in other stars. A. Large Earth-based Solar Telescope (LEST). Background. In order to provide a ground-based facility capable of studying fundamental astrophysical processes with the detaU that only the Sun allows, a consortium of nine nations intends to build a 2.4-m solar telescope at a site with superb seeing. This Large Earth-based Solar Telescope (LEST) will provide high photon flux and low instrumental polarization and will be equipped with adaptive optics to attain angular resolution approaching 0.1 arcsec on a regular basis. The telescope will cost approximately $46M, of which the U.S. share will be one third. A drawing of the LEST is shown in Figure 3. Status. The conceptual design of the LEST was selected in 1984 and endorsed by the Scientific and Technical Advisory Committee of LEST. Since then, more than 30 studies have been carried out confirming and elaborating the concept. These studies include detailed optical design mounting and tower analysis, wind tunnel and helium tests, polarization analysis, and studies of focal plane instruments (including adaptive optics, polarimeters, spectrographs, and new technology filters). 55 Figure 3 ••l«M»»tl Weawitter . D»B* DriM 1 Utwti •••! • • o n t«i * tiwwth 0>i*« N*l«1i*t Figure 3: Telescope tube, mounting and dome structure. 56 AURA and UCAR have joined as U.S. representatives to the LEST Foundation, the governing body for LEST. A Scientific and Technical Working Group has been established to coordinate U.S. efforts. This group has endorsed the LEST management's decision to let a contract for the detailed design of the telescope to the same group that designed the Nordic Optical Telescope. The design phase will extend from 1990 to 1993 at a cost of $3M of which $2M is already available from member contributions. The resulting design will be critically reviewed by the solar community. A conceptual design, developed joindy by NSO and LEST staff, has been presented to the U.S. Scientific and Technical Working Group. Adaptive optics are expected to play a key role in enabling LEST to reach its full scientific potential. R. Smithson's ground-breaking experiments with a prototype adaptive mirror at NSO/Sacramento Peak have confirmed that substantial improvement of image quality is possible even with existing designs but more development is needed. The LEST consortium has agreed, therefore, to pursue a dual-track strategy: to design the telescope and to identify work packages, including adaptive optics, which will be implemented by individual groups. NSO and HAO are preparing a proposal to funding agencies for the U.S. share of these activities (approximately $2.1M over three years) and that proposal will be submitted in the spring of 1990 to AURA for review prior to submission to the NSF. Future Activities. During the next five years, NSO proposes: • to continue to develop, test, and employ adaptive optics systems, • to mobilize and inform the U.S. solar community concerning the progress of LEST, through workshops, newsletters, and electronic mail, • to seek, with HAO, the U.S. share of construction funds, • to play a leading role in the design and construction of focal plane instruments for the LEST, • to participate in developing the data processing, archiving, and distribution systems necessary for LEST, • to participate actively in the LEST Scientific and Technical Advisory Committee (NSO member R. Dunn) and the U.S. Scientific and Technical Working Group for LEST (NSO members: S. Keil, D. Rabin). B. SYNOP. Scientific Rationale. Synoptic programs represent a key element in research on the variety of time-dependent magnetic phenomena exhibited by stars. Among these phenomena are the long-term activity cycles seen in stars that are analogues of the familiar 11-year sunspot cycle in the Sun, and short-term displays of magnetic activity such as spots, flares, and active plage-like regions that occur on timescales of a stellar rotation period or less. In addition to these surface manifestations of an interior magnetic dynamo, it is probable that a large number of late-type stars exhibit a rich spectrum of oscillations, thus making possible a major new technique for the observational study of stellar interiors. Indeed, such oscillations have recendy been detected in Procyon. 57 The observation of stellar oscillations with a new, advanced spectrograph developed according to the guidelines of asteroseismology is a natural extension of the GONG experiment A wealth of information concerning stellar masses, radii, and structure can be deduced even from the relatively small number of modes of oscillation that can be observed in stars. The internal distribution of density and temperature can only be ascertained through this observational technique. Progress in the very young field of asteroseismology will require precise radial velocity measurements of many spectral lines of solar-type stars during observing runs extending over at least one week with minimal interruptions. The parallel study of the origin of magnetic activity cycles and the nature of the dynamo can be productively pursued through monitoring of the chromospheric K-line in a large sample of solar-type stars. The successful HK monitoring program at Mt. Wilson is based on this observational approach. However, the sequential observation of program stars is a slow, inefficient process. As a result, measurements of cycle periods exist for only a handful of solar-type (G2-G4 V) stars. But by simultaneously observing numerous stars in selected open clusters with a wide-field K-line filter, our knowledge of cycles in solar-like stars can be significandy enhanced. Through the use of cluster stars, we gain the important advantages of a stellar sample that is homogeneous in age and chemical composition. In recognition of the need for a new national facility dedicated to solar-stellar synoptic programs in high-resolution spectroscopy and spectrophotometry, the NSO established the SYNOP (S/TVoptic Observing Program) initiative in FY 1987. The organizational framework of SYNOP consists of a Scientific Steering Committee, an Instrument Design Committee, and a Telescope Options Committee to, in total, guide the development of the SYNOP program within the context of the scientific goals, design a spectrograph and K-line camera, and examine the telescope options for the eventual placement of this advanced instrumentation as the key elements of a new synoptic facility. SYNOP will therefore emphasize the development of two major instruments that will form the basis of the program. These include (a) an advanced, visible light, fiber-fed spectrograph characterized by specifications that meet the observationally demanding requirements of stellar seismology (asteroseismology), and (b) a narrow-band, wide-field K-line filter capable of observing, simultaneously, large numbers of solar-type stars in open clusters. Stellar K-Line Filter. Support for the design and construction of a prototype filter has been allocated by the NOAO. R. Dunn, M. Giampapa, and R. Radick (NSO) are leading this project. The optical components have been purchased and they are now being tested to verify the specifications. The thermally controlled mechanical enclosure for the filter has been designed and is now under construction. We anticipate completion of the K-line filter during FY 1990 followed by tests at the McMath telescope. Spectrograph Plans. R. Dunn (NSO) and L. Ramsey (Pennsylvania State U.) are investigating design concepts for the SYNOP spectrograph. The present design goal is to obtain simultaneously as large a portion as possible of the 4500 - 9000 A range on a single 2048 x 2048 CCD chip at a spectral resolution of at least 100,000. A separate spectrograph will be available to observe the spectral region below 4100 A at 60,000 resolution, and the full spectral range with lower resolution. The spectrograph will be fiber-optically fed so as to be adaptable to a variety of telescopes. Following completion of the engineering design stage, a consortium of interested astronomers from the community will prepare and submit a proposal for the construction of the SYNOP spectrograph and its 58 placement at a suitable site, such as the Mt Wilson 100-inch telescope. If the Mt. Wilson option is adopted, then the newly-founded Mt. Wilson Institute would serve as the principal investigator as well as the administrative and managerial agent for both SYNOP proposal submission and any SYNOP operations that would involve the use of the facilities at Mt Wilson. As a member of the SYNOP consortium, the NSO would serve, in this scenario, as a subcontractor to the Mt. Wilson Institute for work involving instrument design and construction, as well as subsequent research activities. NSO does not plan to divert any of its existing core funding for stellar synoptic programs, which are fully committed to operating the McMath, toward operation of either Mt. Wilson or a new SYNOP facility or toward fabrication of instruments for SYNOP. The total budget for the spectrograph is difficult to estimate in advance of an engineering design for the instrument. We expect that the entire SYNOP facility will take about three years to build following the initiation of funding. 59 VTJI. GLOBAL OSCILLATION NETWORK GROUP (GONG) PROJECT The Global Oscillation Network Group (GONG) is an international project to study the internal structure and dynamics of the closest star by measuring resonating waves that penetrate throughout the solar interior~a technique known as helioseismology. To overcome the limitations of current observations imposed by the day-night cycle at a single observatory, GONG is developing a six-station network of sensitive and stable solar velocity mappers located around the Earth to obtain nearly continuous observations of the "five-minute" oscillations, as well as direct measurements of the "steady" motions of the solar surface itself. To accomplish its objectives, GONG is also establishing a distributed data reduction and analysis system to facilitate a coordinated analysis of these data. The primary analysis will be carried out by a dozen or so teams, each focusing on a specific category or problem. Membership in these teams is open to all qualified researchers; there are cunendy 119 members, representing 50 different institutions. NSO is carrying out the GONG project in close collaboration with the community. Milestones for the project are as follows: GONG Milestones April 1984 September 1984 October 1992 First Community Workshop Project Proposal Submitted to NSF Interim Technology Development Project Start Full Site Survey Network Operational Begin Integrated Light Tests - Doppler Imager Breadboard First Doppler Images - Doppler Imager Breadboard Begin Field Station Capital Acquisitions Prototype Design Review First Light Full Prototype System Data Reduction and Analysis Hardware Ordered System Design Review Site Selection Completed Begin Integration of Field System Components Data Reduction and Analysis Center Operational Begin Site Installation Begin Observations September 1995 September 1996 End Observations End Data Reduction October 1985 October 1986 November 1986 January 1987 September 1988 May 1989 February 1990 March 1990 January 1991 February 1991 March 1991 April 1991 March 1992 April 1992 Historical Perspective. In the early 1980s, a flow of exciting new results demonstrated the growing number of crucial questions that helioseismology could address: stellar internal rotation and oblateness, the neutrino deficit and the efficiency of convection. At the same time, the limitations of observations from a single site, or from the relatively brief campaigns at the South Pole, were becoming more and more apparent, and pressure to achieve long, continuous observations was mounting. In April 1984, NSO sponsored a workshop to explore the scientific questions and to investigate the technical issues. As a result of the interest and support manifested at the workshop, NSO staff, in collaboration with the scientific community, established the GONG to coordinate efforts towards the design, construction, and utilization of a network of stations, distributed in longitude and dedicated 60 Figure 4 rf^ V -- \ \ 61 to obtaining a uniform set of data. A proposal was generated and submitted to the National Science Foundation (NSF) in late 1984, by the NSO on behalf of the entire scientific community. In FY 1985, an examination of worldwide climatic data was undertaken to identify potential sites. Simulations that allowed for equipment failures and weather history from suitably located observatories indicated that a minimum of six sites, spaced roughly equally in longitude, would be required to achieve the design objective-a minimum of three years of nearly unbroken data. A robust, automated, digital sunshine monitor for a survey of candidate sites was developed. Alternative technologies for making the solar velocity measurements of the requisite precision were investigated, and the nature of the overall investigation began to take shape. In FY 1986 an interim technology development program led to the design of a variable path length Michelson interferometer, with a narrow prefilter isolating a single solar line, to provide a stable, and sensitive velocity measurement. Early in the year, the site survey started operation. It now has grown to include 14 locations: Arizona Western College (Yuma), Big Bear Solar Obs. of the California Institute of Technology, the Institute for Astronomy of the U. of Hawaii sites at Haleakala and Mauna Kea, the High Altitude Obs. site on Mauna Loa, the Australian Ionospheric Prediction Service's Learmonth Solar Obs., the Udaipur Solar Obs. (India), the Instituto de Astrofisica de Canarias site at Izafia (Canary Islands), the Las Campanas and the Cerro Tololo Inter-American Observatories (Chile), the Urumqi Astronomical Station (China), the King Abdul Aziz Gty of Science and Technology (Saudi Arabia), and the Centre Nationale de Recherche (Morocco). The project obtained a go-ahead from the NSF in FY 1987 and is directed toward completion of a prototype instrument in FY 1990. The breadboard instrument is under test, and the prototype is under construction. Meanwhile, results from the site survey indicate that we may anticipate observing duty cycles well in excess of 90 percent The site survey will continue, with a selection of the final sites planned for 1991. The Instrument The five-minute oscillation is a subtle effect. Individual modes exhibit velocities of less than 20 cm/s, while the sum of all the modes is only a few hundred m/s. The ultimate intention is to have the measurements be limited by the Sun's "random" surface motions. This means developing six stable instruments capable of making imaged velocity measurements with a precision of significandy better than 1.0 m/s. The basic idea is to isolate a single solar absorption line and determine its precise (Doppler shifted) wavelength. The instrument chosen for this task is called a Fourier Tachometer, similar to the one developed in collaboration with the High Altitude Observatory, which is currendy in routine operation in Tucson. Based on a Michelson interferometer, it processes the light from all parts of the solar disk simultaneously to produce a velocity sensitive image of the Sun. This is picked up by a 256 x 256 pixel solid-state detector and stored in a data acquisition computer's memory. At the conclusion of each 60 to 75 second acquisition cycle, three intensity images, differing in modulation phase by 120°, are summed, differenced, and divided to produce a single velocity image, as well as an intensity and a line strength image. The resulting velocity image is then stored on a magnetic tape cartridge, along with an intensity image and other information, for subsequent reduction at the central data reduction and analysis facility. 62 The entire field instrument will reside in an environmentally controlled shelter building, which houses an external light feed, Fourier Tachometer, control and acquisition computers, and data recording equipment. A drawing of the shelter is shown in Figure 4. The light feed itself will be a fully automatic system, which will turn itself on each day, locate and track the Sun, continue to track using an ephemeris during cloudy periods, and supervise and report its environment and operational status. Adaptive software is being developed which will even create a real world ephemeris for the actual site and conditions as deviations from the "pre-canned" ephemeris are noted. Data Management and Analysis. Even at first glance it is clear that the real challenge in the areas of GONG data reduction and analysis are presented by two factors: (1) a monumental volume of data; and (2) a long sequence of complex computing tasks. Each station in the network will produce at least 200 megabytes of data every day. The whole network will generate a gigabyte a day, seven days a week, for three years. Over this time the total accumulation of field data will exceed one terabyte! The reduction process itself is not trivial. Each individual 64 K pixel frame of each station must be adjusted, pixel by pixel, for a variety of instrumental, photometric, and geometric effects. Furthermore, the Doppler effect of the known motions of the Earth and the Sun also must be removed. As many as three adjacent GONG stations may observe simultaneously for periods of several hours. These data must be merged into a single stream of the best frames attainable at each moment. Once these preliminary tasks are complete, several more computationally intensive reductions will be performed. For example, the decomposition of the image data into time series of spherical harmonic coefficients and their subsequent reduction to frequency spectra will be standard processes. Finally, the raw field data, the ultimate reduced data sets, and several intermediate stages must all be placed in long-term storage in computer-based archives using a combination of optical disks and rotary head tapes. Scientific analysis of the data will generally proceed from these archived data sets. While researchers may choose to write their own specific analysis programs, the GONG project and its scientific teams are establishing a central users' library of contributed analysis software, which will be available for general use. This library will include the data access and display system as well as basic analysis tools, which will be fully supported and highly transportable, so researchers can pursue work from their home institutions if they wish. The central GONG computing facility will feature a main computer in the "super-mini" to "mini-super" class. In addition to performing routine data reduction tasks, this system will also be available for research by both on-site visitors and by remote access. Many of the GONG data management tasks are very similar to those of the Stanford/Lockheed Solar Oscillations Investigation (SOI) experiment. This experiment will be launched on the NASA/ESA SOHO spacecraft in 1995. GONG expects a major collaboration with the Stanford group to provide for joint development of reduction, archival, and analysis software common to both projects. Current Status and Future Directions. The development of the field instrument is underway, and this will lead to a complete prototype field system in early 1990. Initial construction of the six field stations began in the summer of 1989. Installation of these units at the six sites is scheduled for 1992, with full network operation to begin late in 1992. 63 The data reduction and analysis system is a major component of the program and is being developed concurrendy. Even before the installation of the network, data acquired during the validation of the prototype should provide the basis for important research. The design and construction of the needed software tools will require the efforts of both the project staff and many other members of the solar physics community. The principal hardware for this system will be procured and installed in early 1991 and will be in full operation later that year. Coordination with the scientific community continues to be assured by regular consultation with a five member Scientific Advisory Committee, a quarterly newsletter, and an annual GONG meeting (with some 80 participants from nearly 40 institutions and 15 countries in 1989). The GONG project is slated to gather data for a minimum of three years. Full scale data analysis activities will continue for at least one year beyond the end of the data gathering phase. If the results of these studies indicate a need, the continuation of data gathering for a longer fraction of the solar cycle remains a possibility. In any case, beginning in the mid-1990s we can look forward to some truly exciting developments in our understanding of our nearest star. 64 IX. SOLAR OBSERVATORY OPERATIONS A. NSO Operations. The National Solar Observatory operates facilities on two sites-Sacramento Peak and Kitt Peak~for both solar and stellar astronomy. In addition, NSO has taken the lead in formulating and implementing the GONG project which is designed to probe the interior structure of the Sun by analyzing solar oscillations. NSO, in partnership with HAO, is the U.S. representative to the LEST Foundation, an international consortium to consider construction of the next generation, large ground-based solar telescope. The GONG and LEST projects are described elsewhere in this Long Range Plan. The facdities operated by NSO play a unique role in solar astronomy in this country and indeed the world. The U.S. solar community is largely dependent on NSO facilities for observational solar physics, and they remain the largest and best instrumented anywhere in the world. In accepting this significant responsibility, NSO has forged important partnerships with major U.S. solar research interests, such as the U.S. Air Force Systems Command Geophysics Lab., NASA, NOAA, and HAO to support various aspects of these activities. In concert with its several partners, NSO has developed a decadal scale Strategic Plan to guide the evolution of its program and facilities. Scientifically, this plan places its major focus on two areas of study: the solar interior, and the interaction between solar magnetic fields and plasmas. Operationally, the plan emphasizes the fullest utilization of existing facilities through enhancements to their focal plane instrumentation and improvements to their basic capabilities. Overview of the Facilities. The Vacuum Tower Telescope (VTT) at Sac Peak consists of an evacuated reflecting telescope with a 1.6-m primary mirror (76-cm entrance window). It is designed to provide high spatial resolution images and spectroscopy of the Sun. Primary analyzing instruments consist currendy of a 12-m echelle spectrograph, a universal spectrograph, a universal birefringent filter, other specialized birefringent filters, a multi-diode array, and some small, specialized instruments. Horizontal and vertical optical benches are available at two exit ports for mounting temporary experiments. The high quality spatial resolution possible with the VTT allows important studies of fine details of the solar atmospheric structure, including prominences. Most experiments seek high spatial/spectral resolution spectra, and/or high spatial resolution, narrow spectral-band images. Quiet Sun studies are concerned mostiy with energy transport and atmospheric heating as produced by small and large-scale convection and wave motions in the photosphere, chromosphere, and transition region. Small-scale, intense magnetic fields and associated velocity flows in the quiet Sun are an important area of research. The complex plasma processes that charaaerize active regions remain poorly understood. In evolutionary terms they are characterized by flux emergence, adjustment of active region structure to reanangement of the sub-photospheric field, and decay. The formation of sunspots and pores in active regions and their morphological characteristics continue to pose many fundamental questions in solar physics. Flare processes are even more complex, because of the rapid, localized release of large amounts of energy. In this area of research also, the magnetohydrodynamic complexities are such that considerable interpretive uncertainties remain, and more research needs to be done. The John W. Evans Solar Facility (previously called the Big Dome) at Sac Peak has two 40-cm aperture emission-line coronagraphs, other smaller telescopes mounted on an 8.2-m photoelectrically guided spar, and a 30-cm coelostat. The main coronagraph and the coelostat are each designed to feed 65 light to a variety of analyzing instruments including a 13-m Littrow spectrograph, a spectroheliograph, a universal spectrograph, and a coronal photometer. Anay detectors are avadable for spectrographs and imaging. As with the Vacuum Tower Telescope, experiments carried out at the Evans Solar Facility cover a broad range of solar phenomena, with emphasis on observations of the emission corona, prominences, and disk features where the low scattered light of a coronagraph is essential. Certain observations are routinely recorded on a daily basis and provide a record of changes on the Sun-measurements important both for short-term and long-term studies. Coronal studies cover practically the full regime of visible coronal emission phenomena. Examples are the physics of loops (heating mechanisms, electric fields, flow velocities, stability, evolution, reconnection, polarization), coronal holes (morphological characteristics, flows, temperature, sector boundaries), transients, streamers, and general morphology. Measurements of the emission of three coronal lines representing different coronal temperature regimes are transmitted daily to national centers that are concerned with forecasting solar activity. The low polarization and low scattered light instrumental characteristics of the main coronagraph permit studies of the polarization of prominence emission and deduction of vector magnetic fields. The same techniques can be applied to the study of sunspots. The Hilltop Facility at Sac Peak has a wide variety of small telescopes dedicated to synoptic programs, mounted on a common spar. The instrumentation includes a white light flare telescope and a 20-cm aperture, two-emission-line coronagraph. The set of Hilltop instruments automatically records flares, sunspots, coronal, and other solar phenomena on a daily basis from sunrise to sunset. These data are archived for use as a record of solar activity and are available to research scientists. The McMath Telescope complex at Kitt Peak contains three telescopes in one inclined enclosure (the 1.5-m main and two .76-m auxiliaries) permitting three conjoined or independent research projects to be run at one time. The Vacuum Telescope on Kitt Peak and the small Razdow patrol instrument comprise the second solar complex on Kitt Peak. At the McMath Telescope, the available solar instrumentation includes a long-focal-length, highdispersion spectrograph and the Fourier Transform Spectrometer (FTS). The FTS has a unique combination of spectral resolving power (typically 0.3 - 1.0 x 10^), total spectral range (0.25 - 18 um), simultaneous spectral coverage (up to a factor of four in wavelength), wavelength accuracy (vacuum wavenumbers to better than one part in 109 if enough photons are available) and freedom from scattered light. In addition, a large and active program of laboratory spectroscopy is carried out at the FTS. Additional instrumentation can be brought to any of the telescopes. The work at the McMath facility centers around its capability for high spectral resolution and for infrared work. Freedom from scattered light makes the McMath Telescope particularly valuable for solar and planetary research. Because of its large aperture and all-reflecting design, the McMath Telescope is ideal for infrared work in the wavelength region 1 - 15 um. Infrared observations have been re-emphasized at the McMath facility during the past year, yielding exciting new results on photospheric magnetic fields and the thermal structure of the solar atmosphere. The solar-stellar community makes use of a dedicated spectrograph and a high signal-to-noise CCD spectrograph for the study of solar type phenomena that appear in stars like the Sun. The Vacuum Telescope on Kitt Peak, supported in part by NASA/GSFC, is used for daily magnetic observations of the solar surface and the acquisition of Helium 10830 A spectroheliograms and it is 66 being upgraded for the accurate measurement of velocity fields. Oscillation observations are also carried out there. The Fourier Tachometer is also in daily operation, utilizing a heliostat on the roof of the Tucson office building, to obtain synoptic helioseismic observations, as well as to support specific visitor campaigns. The Fourier Tachometer was built, and continues to be operated, in collaboration with the High Altitude Observatory of the National Center for Atmospheric Research (HAO/NCAR). Table VII summarizes these facilities operated by NSO. 67 TABLE Vn NSO Telescope/Instrument Combinations Vacuum Tower Telescope (SP): Echelle Spectrograph Slit Jaw Camera System Universal Birefringent Filter Universal Spectrograph Stabilized Spectrograph Horizontal and Vertical Optical Benches for visitor equipment Correlation Tracker Branch Feed Camera System John W. Evans Solar Facility (SP): 40-cm Coronagraphs (2) 30-cm Coelostat Littrow Spectrograph Spectroheliograph Electrograph Dual Camera System Universal Spectrograph Coronal Photometer Hilltop Dome Facility (SP): Ha Flare Monitor White Light Telescope Full Limb Coronagraph Mirror Coronagraph Fabry-Perot Magnetograph Sunspot Drawing Telescope White Light Flare Polarimeter McMath Telescope Complex (KP): Vertical Spectrograph 150-cm Main Unobstructed Telescope 76-cm East Auxdiary Telescope 76-cm West Auxiliary Telescope Infrared Imager Universal Birefringent Filter Magnetograph Image StabUizers 1-m Fourier Transform Spectrometer Stellar Spectrograph System 3 Semi-Permanent Observing Stations for visitor equipment Vacuum Telescope (KP): Spectromagnetograph High /-Helioseismograph Fourier Tachometer (T) Razdow (KP): Ha patrol instrument 68 B. Instrumentation for NSO. Science Requirements. Study of the nearest star continues to yield important results about its structure and evolution, with significance for astrophysics and solar-terrestrial physics, but most observed phenomena remain poorly understood. Over the next 10 years or so, the NSO Strategic Plan calls for two principal scientific foci: studies of the solar interior and studies of the interaction between solar magnetic fields and plasmas. Thus, proposed research programs cover the internal structure and dynamics of the Sun, the origin and evolution of its magnetic field, its influence on the structure of the atmosphere, and related topics. Internal Dynamics. One of the most active areas of research is that of hehoseismology, which uses acoustic waves to map the interior structure of the Sun. Past and present efforts in hehoseismology have revealed much about the internal rotational characteristics of the Sun. The objectives and program of the Global Oscillation Network Group (GONG) are described in Section VIII of this plan, but other programs in helioseismology are also being carried out at NSO. The Sac Peak Vacuum Tower Telescope and the Kitt Peak Vacuum Telescope are both used by NSO staff and visitors in studies of the global oscillations of the Sun. In addition, oscillation observations are made with the Fourier Tachometer located in Tucson. The Fourier Tachometer, developed and operated in collaboration with the High Altitude Observatory (HAO), is ideally suited for full-disk, low-amplitude velocity studies of p-mode oscillations in the Sun. This instrument is considered to be an interim solution for providing the community with a specialized velocity measuring device in a clear site until the GONG becomes operational in a few years. The High-/ Helioseismograph will obtain high quality observations of p-modes with high-degree and medium to high frequency at the Vacuum Telescope on Kitt Peak. Previous observations of these modes have been limited by poor determinations of the spatial wavelengths of the oscillations. The data from the High-/ Helioseismograph is complementary to that produced by the GONG project, which has much coarser spatial resolution. The High-/ Helioseismograph will also serve as a pathfinder for the high-resolution Solar Oscillations Investigation (SOI) on-board the SOHO spacecraft, a project in which NSO will be playing a major role. The High-/ Helioseismograph, unlike either the GONG or the SOI, is intended to run synoptically throughout the solar cycle. In this way, scientists can study the evolution of the flows in the convection zone throughout the cycle. Studies will also be made of the absorption and emission of the p-modes by active regions. The utility of the South Pole as a helioseismology observing site has been amply and frequendy demonstrated since 1980 by NSO scientists and collaborators. Advantages include high altitude, an extremely clean sky, constant Sun elevation, and the possibility of nearly uninterrupted observations for periods on the order of 100 hours. NSO has collaborated with Bartol Research Inst, NASA/GSFC, and Photometries, Ltd., during three austral summers with successful results each time. Much helioseismology remains to be done from the South Pole. A possible project for the 1990 austral summer season is to operate the 1988 equipment, but with two arcsec pixels in a 1024 x 1024 format. This will require adding a fast guiding system and other improvements to the existing equipment. Such a project, undertaken near the maximum of the current solar activity cycle, also offers the possibility of major contributions to the emerging field of solar interior tomography. 69 Magneto-Convection. The NSO/NASA Spectromagnetograph will provide an advanced two-dimensional detector and real time data processing system for the calibrated measurement of solar magnetic and velocity fields at the Vacuum Telescope on Kitt Peak. Changes in line shape and line position will be well separated in the final data stream so that thermodynamic, velocity, and magnetic variations in space and time can be properly measured and compared. The Spectromagnetograph will thus become a powerful new tool for study of solar magnetohydrodynamics over temporal and spatial scales ranging from synoptic variations over the solar cycle to impulsive flare-related phenomena. Daily synoptic observations will also be made to build up a record of persistent global velocities, such as differential rotation, giant cells, and torsional oscillations. In collaboration with the Air Force Systems Command Geophysics Laboratory, NSO will develop an LCD FUter for high-resolution vector magnetic field measurements at the Sac Peak Vacuum Tower Telescope (VTT) and for a filter system that could eventually form the foundation of a Solar Synoptic Network (SSN). A narrow-band filter system (a 20 mA Fabry-Perot in conjunction with the Universal Birefringent Filter) is being developed in collaboration with groups at Florence and Naples for measuring velocity and fields as a function of height in the photosphere. The High Altitude Observatory (HAO) is developing a Stokes spectrograph, to operate in conjunction with an NSO light-feed and polarization compensator at the Sac Peak Vacuum Tower Telescope. The Air Force is also supporting the development of a vector magnetograph by Johns Hopkins University/APL for use at the Hilltop Dome at moderate-resolution. One of the primary goals for future work within NSO is the development of adaptive optics instrumentation for the VTT. Its success is crucial to meet the current goals of high-resolution solar atmosphere studies. The system that is now planned will provide a flexible and powerful research tool in conjunction with this telescope and its valuable auxdiary instrumentation. The scientific potential for such an improvement is enormous. In addition to providing increased resolution for observations by staff and visitors, the adaptive optics program will provide an important part of the NSO contribution to the international LEST effort to construct a new generation, ultra-high-resolution, large aperture solar telescope. The NSO adaptive mirror system builds on the experience of Lockheed and others. We intend to incorporate a continuous faceplate, rather than segments, longer life actuators (> 2,000 hours), more actuators than on the Lockheed minor (the average Rq at SP is 9.6-cm), more computer intervention, a flexible computer for the calculation of phase, and a waveftont sensor that will track the granulation and accommodate various types of mirrors. NSO plans to take advantage of the large-aperture, all-reflecting optics of the McMath Telescope at Kitt Peak to develop new instrumentation that exploits the unique scientific potential of solar infrared observations. To realize the full potential of the McMath Telescope for infrared astronomy, it must be mated to modem infrared detector packages and data systems. What follows is a schematic list of needed auxiliary instrumentation and some of the scientific capabilities it will provide. • Near-infrared camera. Equipped with a 256 x 256 indium antimonide array, this instrument will provide three new capabilities: i) spectropolarimetry of Fe I 1.565 um for maps of magnetic field strength; ii) direct imaging in the 1.63 um and 3.70 um continuum windows, reaching from 40 km below to 40 km above the base of the photosphere; iii) area spectroscopy of the CO fundamental vibration-rotation bands near 4.67 um to infer the lateral temperature structure of the temperature 70 minimum region. The camera will also accept one of the 256 x 256 platinum sUicide arrays already ordered by NOAO. • 5 - 25 um camera. This instrument would incorporate a 20 x 64 IBC array (now being tested at NOAO). It will provide: i) area spectroscopy and spectropolarimetry of the 12 um lines for analysis of magnetic flux tubes; and ii) direct imaging of the 11 um continuum window, probing 130 km above the base of the photosphere. • Cooled spectral isolators beyond 2.5 um. For high-resolution spectroscopy it is important that only a rather narrow bandpass reach the detector so that undispersed background light does not overwhelm the dispersed solar signal; similar considerations apply to Fourier transform spectrometry. The spectral isolator must itself be cool. Depending on the application, a grating-spectrometer postdisperser, tunable Fabry-Perot filter, or sets of narrow band interference filters, might be preferred. • Data systems with real time processing. In typical solar applications, low-noise infrared arrays will be saturated in a fraction of a second. To build up the signal-to-noise ratios that will often be required, it will be necessary to average tens or hundreds of frames of data at each spectral/spatial location. There is a simUar requirement in nighttime work, so the effort will be NOAO-wide. Considerable interest has arisen in the ground-based monitoring of small solar irradiance fluctuationsfor both their astrophysical and solar terrestrial importance-and NSO is preparing to support a community initiative to develop specialized instrumentation to carry this out. The field of solar corona physics has advanced to a point where many critical and fundamental questions are posed, but observational answers are inaccessible due to inherent limitations of existing instrumentation. A major technical breakthrough in optical polishing promises a factor of 10 increase in the sensitivity with which we can measure the faint outer solar atmosphere-the corona-where the solar wind is heated and accelerated. With a highly successful miniature prototype instrument (Minor Advanced Coronagraph-MAQ now operational, we are exploring ways to realize the potential of this "super-polishing" technique in a major new Advanced Reflecting Coronagraph (ARC). With more advanced instrumentation, many outstanding and fundamental problems of astrophysics can be critically studied, such as: The earlier idea of acoustic heating by waves propagating from the photosphere has now been substantially discarded since propagation of these waves is inefficient, and none have been observed. Processes related to electric current dissipation or MHD wave generation and damping are now receiving more attention, but observations of small-scale dynamical phenomena are needed to support and offer more development of these ideas. Flow and condensation of the coronal plasma are known to exist above active regions and around prominences. Flows are particularly impressive in the case of post-flare loops, where rapid evolution is present Condensation of the coronal plasma is often assumed to explain the formation of prominences but has never been direcdy observed, perhaps because either very small-scale processes are inferred or very faint emission is involved. 71 Small-scale explosive or impulsive events are believed to produce fast ejections which propagate along the open magnetic field lines of a coronal hole, producing the fast solar wind. However, no detailed observations exist of the ejection process. Coronal Mass Ejections (CMEs) are observed propagating through the outer white-light corona. We know relatively little about the basic physics. For this problem, high-resolution images, spectra, and magnetic field measurements of the inner corona are required. Following the successful development and operation of the 5-cm aperture MAC, a larger instrument (~ 35-cm aperture) is now in the planning stage. In a joint agreement with the Institut d'Astrophysique (IAP), part of the construction will be carried out by LAP. It is anticipated that this second instrument will be completed within the next two years. It will be used both as a research-quality coronagraph and also as a "proof-of-concept" for an eventual 2-m class telescope. Optical design studies for the large-aperture telescope are in progress. Extended (full-day) observations with the existing coronal photometer on the Evans Solar Facility spar have detected subde transients in the emission-line corona. This discovery indicates the existence of a previously unknown regime of frequent low level coronal activity. To investigate these transient events systematically we need a dedicated coronagraph feed. We therefore will be recommissioning the spare coronagraph as a dedicated feed for the Coronal Photometer. The nature of the activity cycle is not understood, and much of the effort in this field goes into long-term synoptic studies simdar to those in the solar-stellar area. The aging collection of solar synoptic telescopes currendy used to provide the bulk of the low-resolution synoptic observations should be replaced by a standardized, research grade, worldwide network of small synoptic instruments to form a Solar Synoptic Network (SSN). There is considerable interest in this concept on the part of several government funding agencies, and also in several other countries that support research in this area. NSO will continue to play a leadership role in the definition of this woridwide program, and these efforts may result in a separate initiative, to be submitted primarily to agencies other than the NSF, in this area in the future. 72 TABLE VBLI NSO Five Year Instrumentation Plan Summary Project Labor Scientist (mm) Capital ($K) FY 1991 Sacramento Peak Adaptive Optics* Main Lab Computer Upgrade/Phase B Correlation Tracker One Shot Coronagraph Upgrade VTT Universal Spectograph CHIRP Dunn 22 Colley Colley 11 Smartt 3 3 Dunn, Stauffer Dunn, Stauffer 3 3 Total 38 45 38 14 20 Tucson/Kitt Peak IR Magnetograph/Arrays Rabin Stellar K-line Filter Giampapa Main Spectrograph Gratings Upgrade Spectromagnetograph** High-/ Helioseismograph** Pierce 3 15 Jones 5 20 J. Harvey 2 27 Total 25 82 3 1 FY 1992 Sacramento Peak Adaptive Optics* Main Lab Computer Upgrade/Phase C Correlation Tracker One Shot Coronagraph Upgrade VTT Universal Spectograph PE Phaseout (VTT, Evans) Dunn 22 Colley Colley 11 Smartt 2 Dunn, Stauffer Hull, Stauffer 6 44 45 47 Rabin 14 17 Pierce 1 J. Harvey 2 Pierce 1 3 Jones 7 12 25 52 Dunn 30 20 Dunn 3 3 TBD 13 26 46 49 Total 3 1 Tucson/Kitt Peak IR Magnetograph/Arrays Main Spectrograph Gratings Upgrade High-/ Helioseismograph** McMath Heat Exchanger Study Video Magnetograph** Total 20 FY 1993 Sacramento Peak Adaptive Optics* R0 Meter (VTT) New Project A (TBD) Total 73 Tucson/Kitt Peak IR Magnetograph/Arrays McMath Heat Exchanger Video Magnetograph** SYNOP Telescope Design New Project A (TBD) Rabin 13 4 Pierce Jones 20 13 6 Giampapa, Dunn 3 8 14 ~26 55 Dunn 22 Smartt 13 20 5 Total FY 1994 Sacramento Peak Adaptive Optics Large Reflecting Coronagraph (Phase A Study) New Project A (TBD) TBD 13 26 48 51 Rabin 14 20 Jones 3 Total Tucson/Kitt Peak IR Magnetograph/Arrays Video Magnetograph** SYNOP Telescope Design New Project A (TBD) Giampapa, Dunn 6 TBD 4 59 27 90 Dunn 22 20 Smartt 15 14 Total 11 FY 1995 Sacramento Peak Adaptive Optics Large Reflecting Coronagraph (Phase A Study) New Project B (TBD) TBD Total 13 18 50 52 14 20 Tucson/Kitt Peak IR Magnetograph/Arrays SYNOP Telescope Design New Project A (TBD) New Project B** (NASA - TBD) Rabin Giampapa, Dunn 6 15 TBD 4 61 Jones 4 Total * Carried out joindy with USAFSC/GL ** Carried out joindy with NASA/GSFC 74 28 91 Solar-Stellar. Several exciting and important areas in solar-stellar physics can be advantageously pursued in the infrared. Illustrative examples include stellar magnetic field measurements, where the direct dependence on wavelength of Zeeman splitting relative to thermal Doppler widths results in a higher sensitivity to complexes of magnetic activity on stellar surfaces. Spectroscopic signatures of cool star spot umbrae are more evident in the infrared due to their increased brightness relative to the sunounding quiet photosphere. The near-infrared He I triplet feature at 10830 A is a valuable proxy for coronal X-ray emission in solar type stars. This line is not within reach of our current CCD system, but a suitable IR detector would enable synoptic programs that emphasize this key diagnostic to be implemented. Exploratory programs in stellar seismology using CO line intensities could be pursued in the infrared. In order to realize the potential in the infrared fully, an array detector must be installed at the McMath for nighttime use. The detector may eventually be coupled with a cryogenic echelle, thus establishing a uniquely powerful capabdity for solar-stellar synoptic spectroscopy. Current capabdities in visible light spectroscopy will be enhanced to achieve a Doppler velocity precision in the few meters per second range. The study of stellar convection requires resolving powers (A/AX) of 1.5 - 2 x 105. Exploratory programs in stellar seismology and extra-solar planetary detection, each using Doppler spectroscopy, demand the simultaneous acquisition of numerous spectral lines. The cross-dispersing of our echelle and the new sheer system represent significant steps toward increased spectral range and enhanced spectral resolution for the McMath nighttime program. The addition of a gas cell in the optical beam that would allow the superposition of reference features onto the stellar spectrum would increase the precision that can be attained by essentially providing a measure of spectrograph stability. 75 X. NOAO OPERATIONS A. Scientific Staff. NOAO is a service organization, charged with the responsibility of budding and operating state-of-theart facilities for the community. In order to carry out this task, NOAO must necessarily also be a research organization of the first rank. The quality of service that is provided to the community of observers that we serve is direcdy linked to the quality of the scientific staff. It is the scientific staff that must work with the community to identify major opportunities in astrophysical research and on that basis define the future course of NOAO. They prepare proposals for major new telescopes, oversee the design and construction of new instruments, and monitor the performance of existing facilities. The scientific staff provides the link between NOAO and the users of its observatories and in this way evaluates how well NOAO's programs match the needs of the astronomical community. The staff is also an important repository of the technical information necessary in order to design and budd new telescopes. Recent interactions with various university groups concerning construction of 4-m class telescopes have shown that NOAO expertise is essential in the early stages of project definition before funding becomes avadable to hire experienced staff. During the next five years we plan to maintain the scientific staff at approximately its cunent size. We expect between two and four retirements during this time period and plan to fill those vacancies with people who can play key roles in the development of the new facilities now being planned by all three divisions. An important improvement in the scientific environment has been the revitalization of the post doctoral program. In FY 1989, CTIO had two post-docs, while NSO had one located at Sac Peak. KPNO had five, a reduction of one from the previous year. As soon as budget levels permit, CTIO and KPNO will have three post-doctoral positions each, and NSO will have two plus a student program. We plan to stabilize the program at this level. NOAO is in the process of developing stronger policies and procedures for post-tenure review, which already includes annual evaluations by the appropriate director. Salaries are not competitive with university astronomy programs of comparable stature, and over the next several years this disparity will be eliminated where performance merits improved salaries. We will implement an emeritus program to make it possible for members of the scientific staff who retire to remain active in research. We wiU also review and clarify the career track for those members of the scientific staff who devote most of their time to the construction of innovative instrumentation. B. Computer Support. Downtown Tucson Computer Support. The computer facilities run by CCS in the Tucson office complex provide three general needs for NOAO-Tucson: data reduction and analysis for the scientific staff and visitors, general computing for all staff members, and LRAF development. Over the next five years we expect to continue with two main themes: making the transition to a distributed computer environment ("a workstation on every desk") and replacing old, inefficient expensive to maintain systems with new hardware. In the process of carrying out these replacements, we expect to achieve a major upgrade of capabilities, responding to increasing data bandwidths, more powerful data analysis algorithms, and the need for more sophisticated scientific and engineering modeling. 76 A high capacity data link will be installed between Kitt Peak and the Tucson office, enabling data archiving, remote observing, and system-wide software maintenance, and supplying Kitt Peak visitors with access to the world of electronic communications. KPNO and CTIO propose to initiate archiving of astronomical observations. While the archives will likely be indexed through the NASA Astrophysics Data System, it will be necessary to develop parallel archiving activities at both Kitt Peak and Cerro Tololo. The FY 1990 and 1991 totals in Table IX reflect the budget figures given in the program plans. TABLE IX CCS Schedule of Major Capital Expenditures Current Five Year Plan Year FY1990 FY1991 FY1992 FY1993 FY1994 FY1995 Distributed Computing VAX 8600 Replacement Research Grade Film Hardcopy 100K 50K 65K 275K 345K 386K KP-Tucson Data Link 1.54 Mb 25K 25K 500K 50K 25K Item Archiving - Capital Total 125K 75K 640K 25K 25K 25K 100K 50K 400K 420K 30K 441K The total expenditures in this table, which are for NOAO Tucson only, are in a similar range to the capital investment in computer equipment NOAO-wide in recent years. If adequate support is to be provided for CTIO, NSO, and the KPNO mountain, significandy increased investment is required in this area. If, instead, spending continues at a constant level, it seems likely the VAX replacement and the archiving would be defened. The computer costs for the other sites are reflected in their instrumentation budgets. Ceno Tololo Computer Support. Computer planning at Cerro Tololo is similar to that of Central Computer Services in Tucson. Over the next five years, we need to complete the establishment of a distributed computer network, fully integrated first between Cerro Tololo and La Serena and later between CTIO and KPNO. This will require the purchase of additional individual scientific workstations, upgrading the ones which we have as technology advances and improving the speed with the network functions so that it will not become clogged by the ever-increasing amount of traffic which wiU be passing over it The latter will be done by replacing our two current Ethernets with Fiber Distributed Data Interconnection (FDDI) systems. We intend to begin archiving astronomical data in synchronization with KPNO, using wed defined standards so that data wiU be fully interchangeable between sites and easily accessible to users, probably using the NASA Astrophysics Data System. Routine storage of data will begin during FY 1992 with all data archived and then made accessible to the astronomical community once standards have been established in conjunction with KPNO. A dedicated microwave link has already been established between La Serena and Ceno Tololo, with a TI (1.544 Mb/sec) data channel scheduled to begin high speed connection of the mountain and downtown computer networks within the first half of 1990. This system was funded as an experiment and has neither backup nor spares. Because repairing equipment of this type in Chile is slow, as this 77 link becomes increasingly important to our operation it wiU become necessary first to purchase spares and later equipment sufficient to install a full backup channel. A backup channel will also double the data transmission capacity, which will become necessary as the traffic on die network increases. NASA has funded a three year experiment to connect CTIO direcdy to the Internet via a 56 Kbaud dedicated satellite uplink facility on Cerro Tololo to the PanAmSat This experiment is designed to demonstrate the value of having CITO wed connected to the U.S. networking system and to allow experiments in remote observation. By the end of this three year period, we expect that the experience acquired will demonstrate the need and allow us to establish a simdar satellite link of at least 64 Kb direcdy between CTIO and KPNO for a relatively modest cost Such a link will allow greater integration of the operation of the two observatories, permitting routine remote observing and joint data archiving. TABLE X chedule of Major Capital Expenditures Current Year Item Distributed Computing Equipment Upgrades Five Year Plan FY1990 FY1991 FY1992 FY1993 FY1994 75K 40K 50K 25K 75K 75K 75K 50K 50K 70K 80K 50K CTIO-KPNO SateUite Link Ethemet-FDDI Conversion 15K 15K Microwave Spares/Backup 30K 30K Data Archiving-Capital 125K ♦Expenditures to 65K 40K 60K 50K 210K 310K 245K FY1995* 260 be determined. Kitt Peak Computer Support. The facdities on Kitt Peak use computers for control of instruments and telescopes, for data reduction (using IRAF), and for the general computing needs of the mountain staff. Over the next five years, we wiU continue two processes already started: replacement of the control computers, and the provision of sophisticated TRAP systems for all observers. In addition, new or augmented facilities will be required to provide the mountain network with a connection to the planned high-speed data line to Tucson, and to provide a central database server that will allow computer access to an extensive collection of astronomical catalogues as well as lists of individual observer's program objects. 78 TABLE XI KPNO Schedule of Major Capital Expenditures Current Year Item Five Year Plan FY1990 FY1991 25K 25K Telescope Control Instrument Control IRAF Data Stations and 50K 30K upgrades to existing systems Computer spares 15K 15K Database Server 1992 FY1992 FY1993 30K 30K 20K 20K 25K 35K 45K 50K 50K FY1995 50K 25K 50K Ethernet-FDDI Conversion 15K High Speed Communication Upgrades Total FY1994 90K 70K 135K 110K 15K 30K 30K 30K 145K 80K The spares money will allow us to handle computer fadures immediately instead of waiting for repair under the Sun maintenance contract, which could take as long as a couple of days since we are contracted for only weekday coverage. (Cannibalization of working computers elsewhere in NOAO is another way to provide immediate repair, but this is a course of action likely to be protested.) NSO Sac Peak Computer Support. The computer facility at NSO/SP is comprised of three facilities; Main Lab (ML), Evans Solar Facility (ESF), and the Vacuum Tower Telescope (VTT). The ESF and VTT computer systems are mainly used for telescope control and data collection and limited analysis. They are not set up for general computing. The ML facility is used for data reduction, analysis, and general computing. Our main objectives over the next five years are to: 1. Replace old, inefficient, and expensive systems with new hardware. This not only includes computer hardware, but also electrical and environmental systems. 2. Upgrade offices with a distributed workstation environment 3. Upgrade to a "compute server" to handle the increase in data acquisition at the observing sites. With new instruments coming on line in the near future, there will be a dramatic increase in the amount of data coUected. This data increase wid require a major increase in computational power to analyze the data. 4. Network NSO/SP with a high bandwidth communication system to handle the increase in data traffic between the observing sites and the "compute server," as well as allow more flexibility. Specific aspects of this plan include: 1. Replace the Perkin-Elmers at the VTT and ESF. 79 2. Replace the motor generators (MG) and A/C systems with more efficient UPS and cooling systems at the ML. 3. Complete the fiber optic network to the VTT and ESF. 4. Upgrade our Internet connection to include a dedicated network gateway and higher speed modems. 5. Install workstations in scientific staff offices. 6. InstaU a "compute server" to handle the increase in data codection. This would be in the 100 MFLOPS cpu(s) 10 Gigabytes disk space range. 7. Upgrade the fiber optics network to FDDI. This is needed to handle the increase in data traffic. The minimum connection needed would be between the ML and the VTT. TABLE XII NSO Schedule of Major Capital Expenditures Item P/E Replacement MG & A/C Replacement Fiber Optics Network Gateway FY1991 FY1992 FY1993 FY1994 80K 10K 10K 8K 15K Distributed Workstations 45K 45K Compute Server 45K 300K FDDI Network Total FY1995 24K 18K 150K 45K 300K 69K C. Facilities Maintenance. One consequence of inadequate operating budgets during the past decade has been the deterioration of facilities at all four sites. Routine maintenance is carried out by the staff of NOAO, but there are many projects which require either expertise not found within the observatory or require major capital expenditures that cannot be completed at current funding levels. Some of these projects, such as replacement of transformers containing PCBs and underground fuel storage tanks, are mandated by federal law; other projects, such as guardrails at CTIO, mirror handling equipment at the McMath, and fire alarm systems at various locations, have been highlighted by review committees as safety issues; other projects, such as painting the McMath and resurfacing roads at Sac Peak, are essential to maintaining facilities over a long period of time. As soon as the budget permits, it is our intention to begin to work on the projects in order of urgency. If we were to budget $500,000 per year for these projects, it would take five years to complete work on the existing backload. The maintenance requirements identified to date for all sites are summarized in Table XIII. Comments on specific issues at each of the sites foUow: 80 TABLE Xm Maintenance Requirements Maintenance and Safety Summary CTIO $ 612K KPNO 585K NSO/KP 222K Tucson 198K NSO/SP 830K $2,447K Total Maintenance and Safety CTIO Replace transformers containing PCBs Replace metal water pipes on Tololo Replace corrugated metal warehouse and shop buildings on Tololo Replace road grader, dump truck, bulldozer Rework drive-train/declination gear Curtis-Schmidt Guard raUs/Tololo road $ 32K 20K 190K 165K 75K 130K Maintenance and Safety KPNO Underground storage tanks RUPS servicing Paint 4-m budding and dome $ 45K 20K 60K Power line 6K Sewer line repair Seal and chip surfacing of roads Telephone line replacement Roofing 4-m building systems Rotator guider 20K 90K 50K 8K 44K 150K Central fire alarm 30K 4-m ingress/egress (visitors/staff-fire) 4-m pedestrian roadway control 50K 12K Maintenance and Safety NSO/KP Modify #3 mirror mount to allow safe handling Ice shield over east door of Vacuum Telescope Guard to prevent fall from Vacuum Telescope elevator 10K 5K 7K Comprehensive survey and overhaul of McMath electrical system 100K Paint McMath 100K 81 Maintenance and Safety Tucson Electrical system - east wing Boder replacement - west wing Roofing/stucco repairs Underground fuel tanks Safety lighting Upgrade fire protection Air conditioning system Cooling tower Resurface parking areas/work yards Painting $ 8K 10K 35K 15K 5K 15K 50K 25K 15K 20K Maintenance and Safety NSO/SP Fire detection system Fire suppression systems Replace a segment of water mains Paint redwood buildings Paint facdities maintenance buddings Resurface roads, driveways, and parking lots Storage/parking building Modem vehicles/equipment Hilltop Dome electronics replacement Motor Generator replacements Plating room upgrade Video and computer network replacement $ 100K 100K 40K 60K 15K 90K 45K 175K 75K 70K 35K 25K Facilities' Maintenance. Ceno Tololo. The CTIO infra-structure, both in La Serena and on Tololo, is in need of renovation. This includes the buried electrical lines, the water distribution system, the telephone plants, and the roads. A substantial fraction of the wires and pipes have become corroded over the years, and must be replaced. We intend to accomplish this gradually over the next five years by replacing 20 percent of the water and electrical systems each year. The well that supplies water to the La Serena compound has been operating near full capacity all summer, and a second well will be required to meet demand in the next few years. The telephone plants are old technology and cannot accommodate more lines. Integration of the CTIO phone system into the new microwave system will require the purchase of new phone plants. The roads need constant maintenance, both paved and dirt, and a major filling and grading of the Tololo road will soon begin. Guard rails need to be placed at certain critical areas of the road, and some wiU be installed to ensure safety in the next few years. The electrical and telephone problems wiU be resolved with the existing operations budget. New funds required to deal with the other problems are listed in Table XIII. Facilities' Construction, Cerro Tololo. CTIO is establishing a broad-based data link to the U.S. for all of our communications, including telephone and e-mad, in cooperation with NASA. The sateUite up-link facilities will be situated on the Tololo summit where we will construct two identical ground stations, one of them to be a "hot" spare, and each consisting of a 4-m dish with support structure and electronics. Although the La Serena 82 offices and labs will need expansion in the future, the most pressing need at present is for two buildings on Tololo and Pachon. The current mountain instrument and machine shop is housed in inadequate, make-shift quarters in the ground floor of the 1.5-m telescope. In order to provide for a shop that can properly support the mountain maintenance needs, a new shop budding must be constructed near the telescopes. Furthermore, the increasing presence of CTIO on Cerro Pachon will in the near future require a utility budding for storage and a small workshop, situated on the Pachon summit. Facilities' Maintenance, KPNO. During the past three years, KPNO has corrected a large number of facihties maintenance problems. The maintenance needs are growing, however. As soon as funds allow, we must paint the 4-m building and the McMath telescope, chip seal ad the roadways, replace the solar dorm roof, upgrade the 4-m telescope primary mirror air supply, overhaul the 4-m dome trucks, provide a cleaner environment for the 4-m aluminizing chamber, correct the McMath telescope mirror handling equipment so it is not a safety hazard, upgrade the 4-m guider/rotator to state-of-the-art operation, and improve seeing in the 4-m telescope by better insulating the control room, cooling the oil, and improving the cage power cooling. Facilities' Construction, KPNO. Two large construction projects are expected to be completed during this period. Both of these projects are non-NSF funded. They are the WIYN telescope and expansion of the Kitt Peak Visitor Center. The latter will be supported through donations from the public. Since the majority of the work will be accomplished through contractors, the impact on mountain staff wiU be minimal. AdditionaUy, there are a number of smaller construction projects which will be completed by our staff. During the summer of 1990, the present #1 0.9-m telescope site will be cleared. Site preparation and construction of the WIYN telescope building will follow. It is anticipated that first light will occur in 1993. A 0.4-m visitors' telescope building will be constructed in 1990. Funding is available from contributions by the public. The expansion of the Kitt Peak Visitors Center is projected for 1991 1992, depending on avadability of donated funds. This project will also include the remodeling of the exhibit area. Examples of smaller construction projects that we believe fit the must-be-done category are replacement of our eight, 20-year-old plus, underground fuel tanks with EPA-approved equipment to comply with EPA regulations, replacement of approximately 500-feet of under roadway sewer line which is ftdl of tree roots, replacement of a large portion of our nearly 30-year-old, rapidly fading underground telephone cable, and the installation of a mountain-wide fire alarm system. Less critical, but nonetheless important other new projects include correction of the 4-m building structure ingress/egress for fire safety, installation of a reliable uninterruptable power source for the 1.3-m telescope, and a large number of AURA safety survey-identified safety improvements. Facilities' Maintenance. NOAO Tucson. NOAO Tucson Central Facilities Operations (CFO) maintains nearly two city blocks of buildings and parking facilities. During the past three years, significant progress has been made regarding the upgrading of the roof structure and power for the NOAO Tucson headquarters building. However, due to many years of growth, this budding has been subdivided to a point where air conditioning ceases to function properly in certain areas, and power must be carefully measured before allowing the installation of equipment in both offices and labs. These problems have solutions, and they are being 83 corrected. In FY 1990, a pilot project will make use of a computerized preventive and recurring maintenance program. The intent is to modernize the approach used for maintenance of facdities, reduce O & M costs, and increase reliability. We wiU hire heating/cooling and power consultants to review building systems. We wiU also bring our building drawings up to standard by properly documenting ad as-built drawing changes. Examples of other expected facilities maintenance include stucco repair to the main budding, repavement of the service yard, improved exit/safety lighting, upgrade of the fire protection system, modification of the shop area bay lighting, replacement of all floor covering trip hazards, repainting of interior and exterior buiddings, and exterior landscaping. Facilities' Construction, NOAO Tucson. It is not possible to provide sufficient office and lab space within the NOAO Tucson Headquarters building for ad proposed as wed as ongoing projects. For the 8-m telescope project alone, it will be necessary to create additional office, lab, and parking space. Proposals are being formulated to provide 5,000 to 32,000 square feet of new office and laboratory space. We must also provide approximately 50 additional spaces for parking. In 1992, the AURA Corporate Office is projected to accommodate the GONG Computer Center and will be configured by our maintenance staff as necessary. NOAO Tucson must also comply with the new EPA, state, and local requirements for replacement/upgrading of underground fuel storage tanks. It wiU also be necessary to replace one of our air conditioning cooling towers. Facilities' Maintenance, Sacramento Peak. Considerable progress has been made in correcting minor safety deficiencies at NSO/Sac Peak, but it is imperative that major items be addressed in the future. The AURA safety survey team recommended that the 1968 LaFrance fire truck be replaced and that a comprehensive fire protection survey be conducted. The survey, performed by a professional company, led to many recommendations relating to fire detection and suppression systems. The existing roads, driveways, and parking lots all show moderate to advanced deterioration and need to be sealed before irreparable damage occurs. Boilers need to be replaced and the redwood buildings need to be painted. All of the special-use vehicles such as front-end loaders and the fork lift are so old that it is becoming ever more difficult to find replacement parts. We have reached the point where major maintenance can no longer be deferred. Facilities' Construction, Sacramento Peak. The latest site plan for Sacramento Peak shows the construction of an addition to the main lab, the addition of a visitor center, and the construction of a storage/parking budding. Existing office and workspace in the main lab is totady inadequate to support staff and visitors and to house existing data reduction facdities. NSO/Sac Peak is probably the only national observatory that does not have a visitor center. The number of visitors has been increasing steadily (to 30,000 in 1989) and on some days they completely overflow the designated parking area and are congesting the facilities maintenance area. Increased parking space and the addition of a visitor center in a less congested areas would benefit the observatory as wed as visitors. As funds become avadable, a storage building with an attached covered parking area for maintenance vehicles wdl be instaded and the main lab building expanded. 84 XI. BUDGET The core budget and staffing shown in the tables include three components: observatory operations which includes scientific staff and support, operations and maintenance, instrumentation, and management fee; the TTP program, and GONG. These sections are discussed separately below. A. Observatory Operations. We show a budget that provides for 5% real growth for the FY 1991 - 1995 period. The growth increment is reflected in the budget tables. B. Telescope Technology Program (TTP). Funding for this program wiU continue to go towards polishing, testing, mechanical support, and thermal control of the 3.5-m mirror that has been cast in the Steward Observatory Mirror Lab. The program will be stopped after FY 1992 when work on the mirror is complete. C. Global Oscillations Network Group (GONG). Activities GONG intends to undertake during the period of this Plan with the proposed funding are to order data reduction and analysis hardware; complete the site selection; install equipment at the sites; and carry out the observations. D. Initiatives. The funding schedule for the 8-m telescopes project is based on the proposal submitted to the NSF, with aUowance for inflation. The funding covers one-half of the costs of a two telescope project. No funds are budgeted for 4-m telescopes since support for their construction wiU be sought through collaborative efforts with universities. The funding for LEST is the total estimated U.S. share of the project. Proposals for funding will be submitted to the NSF in late FY 1990. Review of the proposals may be assigned to the Division of Atmospheric Sciences, in which case some or all of the funding might come from that division. E. Management Fee. The management fee charged by AURA covers the cost of the corporate office and staff; the travel and meeting expenses for committees of the Board; required legal audit and consulting services; and incidental costs conneaed with AURA's management of NOAO and ST Scl. These include corporate memberships in the American Astronomical Society and the Astronomical Society of the Pacific, and contributions to meeting of these societies and to educational and training programs in astronomy. AURA seeks to keep corporate costs low by maintaining a small permanent staff-only five (and this includes the President and Vice President)-and by relying on volunteer service by members of the Board in lieu of corporate staff and in lieu of consultants for expert administrative and other services where possible. As a result the AURA fee for the operation of NOAO is held to less than 2% of the value of the contract with NSF. 85 TABLE I RATIONAL OPTICAL ASTRONOMY OBSERVATORIES FY 1991 - FT 1995 LONG RANGE PLAN BUDGET SUMMARY (Aeounts in Thousands) Observatory Operations Scientific Staff I Support Operations t Maintenance Instrumentation FY-1990' fY-1991 FJL1292 ft'"?? F $ 4,401 14,898 2.688 S 4.676 15.770 S 4.911 16.707 4,102 $ 5.156 17,543 3,850 $ i2Z 407 S22.S94 123,633 Management Fee Subtotal Observatory Operations 2.780 USAF I NASA Support of NSO Total Obs. Operations-NSF Funds m $26,147 448 $26,997 $ <737> $21,725 S22.964 $25,445 $ $ $ $26,260 $ CD Telescope Technology Progrsa Global Oscillation Network Group Subtotal 884 906 475 1-500 i°oo _LfifiZ _2J42 _ $24,109 $25,770 $28,007 $28,403 $ $ 4,000 $16,000 $17,000 $ INITIATIVES 8-aeter Telescope Solar Nigh Resolution Telescope (LEST) Design 1,500 500 200 Construction 1-000 1.999 5.500 Total Budget - NSF Funds $24.109 122*22 $4' $51.103 FY-1990 Program Plan, Revision I (tentative) includes new funds of $23,800K, $259K carried forward from FY-1989, and $50K new REU funds. $ IASLE II MIIOMM. OPTICAL ASICOJHNY OBSEIVAIOBIES FY 1991 - FY 1995 LONG NANCE PUM SUSMNY BY FfjOGtAtt (Ajsounta in Thousand*) ft im' fT im ft mi ft 1W1 Personnel Other Personnel Other Personnel Other Personnel Other Perso twit Costa Costs Coats Costs Costs Costs Costs Cos 111 SI, 2 Carro lololo Inter-Aaarican Observatory Scientific Staff t Support Operations I Maintenance SI,121 1,759 147. —ill _IL2Z2 M463 m Inatruaantatlon Total CTIO 1,665 SI,172 1,980 S H.1H 140 Operation* t Maintenance SI,042 1,477 Inatruaantation _Jil S 109 728 ALm »3.*37 93 SI,HI 1,726 SI,087 1,644 H.W S3,$27 ~l HI 737 12° Total NSO OO liZ 101 S3,935 » S 106 1.864 SI,293 2.183 hi S 1,957 __M $2-458 S5,963 $5,448 »5,155 National Solar Observatory Scientific Staff a Support 1,737 SI,231 2,079 S $A $6,277 98 799 SI,198 1,812 S 103 839 SI. 1 _22 2W J24 303 Itt $3.156 rl.WI $3-313 $1.079 S4.3B9 »3 254 SI,864 3,614 S SI,957 3,795 S $2 3 S4.157 w Kitt Peak National Observatory Scientific Staff t Support Operation* t Maintenance SI,648 3,243 1.1W Inatruaantation Total KPNO tt.MI 252 1,245 SI,775 3.442 —HI 1.336 S ALM S7.B49 M.cH S 1,245 1.403 m S8.40S turn *6-M1 s S 267 1,367 280 1,435 1-473 W «.«1 S9.362 iLiii- _L t7. S9.567 Central Offices Director's Office Scientific Staff 8 Support S Operation* t Maintenance 2 S ifi Inatruaantation Totel Director's Office 55 386 3 <73> S 56 384 <47> S 481 S <59> S S 27 688 110 S 4 225 S 825 S 483 S 135 647 S S S 12 <36> S 507 S 3 S 142 679 S 204 130 7? 62 S <51> 423 <49> tf 11 43 " 59 403 1} 47 <37> S 532 S <38> S 3 S 149 713 S 3 225 S 214 «o 100 Central Coaputer Services Scientific Staff 1 Staiport Operations 1 Maintenance Inatrusentetion Total Central Coaputer Services S 7? 12} 354 s 857 s 337 Central Adeinistrative Services 943 405 992 404 Central Facilities Operations 555 590 584 779 S 400 628 SI, 1,042 424 1,094 445 1 613 843 644 885 900 s 857 S 962 S Table II, continued P.r.onn.|fT ""othsr Co^T £2,1, P.r.onnel Cost. ^ Other toju Central Emjineerlns 8 Technical Services Total C.r.l Office, *°° ' 12fi _JZ ^ $ MJ $ —21 —"• _Jtt —8Z I1.W ^^tlsttf —U* tti.fWL, $23,63J W.HI W.ffl $26,147 ^WsW W^^ ififtf* $21,725 $22,964 <702> <737> S25.U5 $26,260 ™ «' 27V J29 '** *2 2*2 2*J fit LB m Teleecope Solar Niffc Nasoiution Telescope (LEST) S24.109 Lm ZH 125^72 WJSL 128*422 4,000 16,000 17.000 5ofl 1.500 0"t»n r.ooo 1.000 —uxal S32.270 50fl 500 1.000 1000 —LL!ctx $45.W 1 FY-1990 Proora. Plan. Sevi.ion I (tentative) includes new funds of S23,800IC, S259K carried forward free. FT 1989. and S50K new SEU funds. — S15.64J$22,394 «.« *» TOTAL BUDGET - NSF FUNDS —* —448 IHITInJIYai Co , _42Z I24J02 Perso , „8 t l.ttt „^.rh MM ns.iiu.s-. $»t weup Construction 6U »LW ^'2.481 T.t.c^I.telwfrsr- •-•star CosU __*02 TOTAL OBSESV. OPEMIIONS-NSF FUNDS Subtot.l Other Co,» _iSZ USAf $ NASA Support of NSO S £otU FT 1993 '" t 1.M ^ILiM Faa Subtotal Obe.rv.torv Operation. Personnel Cost, f * Publication. ».nfo™.,ion.MourcM FT 1992 Personnel '™ Other ,.,Xcw ;„ tv toon* ' Includes credit for estiaaled indirect costs recovered froa non-NSF projects (S110K in FY 1WU). 200 200 ?.?W -2^22 £U21 M. W l TABLE 111 NATIONAL OPTICAL ASTRONOMY OBSERVATORIES FY 1991 - FY 1995 LONG RANGE PLAN STAFFING PLAN (In Full Time Equivalents) By Budget Category FY Scientific Staff t Support Operations t Maintenance 1990 FT 1991 FY 1992 fT 1°°? 64.00 321.95 65.00 321.95 49.65 65.00 321.95 50.65 321.95 51.65 FY 1994 Instrumentation 49.65 Telescope Technology Prograa) 12.50 64.50 321.95 49.65 12.50 Global Oscillations Network Group 19.00 1V.09 24.00 25,09 25.00 467.10 467.60 466.10 462.60 463.60 10.00 13.00 17.00 21.00 479.60 484.60 64.00 66.50 32.10 75.50 Subtotal 65.00 5.50 1H|T|AT1VES 8-meter Telescope CO Solar High Resolution Telescope (LEST) Total Staff Scientists Engineers I Scientific Progrsa Administrators t Supervisors Clericsl Workers Technicians Maintenance t Service Workers Subtotal 467.10 63.00 69.00 63.50 69.00 64.00 69.50 64.00 66.50 32.10 75.50 130.60 32.10 75.50 130.60 r?,°9 °*.?9 32.10 74.50 129.10 96.90 32.10 75.50 127.60 96.90 128.60 96.90 467.10 467.60 466.10 462.60 463.60 6.00 1.00 7.00 7.00 2.00 8.00 1.00 1.00 1.00 1.00 3.00 4.00 7.00 10.00 477.60 479.10 479.60 404.60 INITIATIVES 8-meter Telescope Engineers t Scientific Progri Administrators t Supervisors Clerical Workers Technicians Total Staff 467.10 2.00