<![if !supportEmptyParas]><![endif]>

Bruce Gillespie*

Site Operations Manager, Apache Point Observatory

<![if !supportEmptyParas]><![endif]>

<![if !supportEmptyParas]><![endif]>

Abstract

<![if !supportEmptyParas]><![endif]>

The Apache Point Observatory 3.5-meter telescope is a working model of a modern mid-sized telescope used primarily on a shared-night, remote-observing basis. After a decade of successful remote operation and scientific accomplishments, the Astrophysical Research Consortium, builder and owner of the telescope, is examining the role by which this university-owned instrument can best serve its constituency and astronomy at large in the coming years. Various "niche" scientific capabilities are described for the telescope, including fast-response observations of transient phenomena, synoptic observing programs, reactive queue-scheduled observations, temporal study programs, plus being a capable test bed for new instruments. While specialized uses of the telescope offer potential for major scientific discoveries, traditional observing capabilities need to be sustained for the ongoing and future research programs for the majority of the consortium astronomers and students, a large and diverse community. Finding an appropriate balance between the "unique and specialized" versus the "bread-and-butter" observing models is discussed, as is the role hands-on remote observing can serve to support the various operational models.

<![if !supportEmptyParas]><![endif]>

Keywords:Astrophysical Research Consortium, Apache Point Observatory, 3.5-meter telescope, remote observing

<![if !supportEmptyParas]><![endif]>

<![if !supportEmptyParas]><![endif]>

1. background

<![if !supportEmptyParas]><![endif]>

<![if !supportLists]>1.1<![endif]>Overview

<![if !supportEmptyParas]><![endif]>

The Astrophysical Research Consortium (ARC or 'consortium' hereafter) was established in 1984 to build and operate Apache Point Observatory (APO) at Apache Point, New Mexico, for the shared use and benefit of consortium astronomers and students. Two major ARC projects at the site are the 3.5-meter telescope, which has been in routine operation since 1994, and the 2.5-meter Sloan Digital Sky Survey (SDSS) telescope, in operation since 1998. The following ARC member institutions currently fund the operation and capital improvement projects for the 3.5-meter telescope: the University of Chicago, the University of Colorado, Johns Hopkins University, New Mexico State University, Princeton University, and the University of Washington. Telescope time is allocated to these institutions in proportion to their project contributions, and the telescope has a diverse and geographically widespread user community of more than 200 astronomers and students.

<![if !supportEmptyParas]><![endif]>

<![if !supportEmptyParas]><![endif]>

<![if !supportLists]>1.2<![endif]>3.5-meter telescope and instrument description

<![if !supportEmptyParas]><![endif]>

The 3.5-meter telescope has an f/10 Nasmyth optical design with a 30-arcminute field-of-view, a lightweight spun-cast 3.5-meter primary mirror, and lightweight secondary and tertiary mirrors. Rotating the tertiary mirror provides access to nine parfocal instrument ports. Direct friction drives enable precise pointing and tracking. The telescope structure and enclosure have low mass and are kept close to isothermal (at ambient temperature) by wind and fan-forced air. The secondary mirror is actively controlled to effect focus and tilt compensation, and the transformation between the alt-az telescope mount and celestial coordinates is derived from observational pointing models, while collimation is periodically adjusted using a Shack-Hartmann wavefront sensor. The telescope has excellent blind pointing performance, frequently provides sub-arcsecond images in the visible, and requires unscheduled repairs less than 2% of the time. Further telescope information is found at http://www.apo.nmsu.edu/Telescopes/eng.papers/eng.papers.html. An interior view of the telescope is shown in Figure 1.

<![if !supportEmptyParas]><![endif]>

<![if !vml]>

<![endif]>
Figure 1.The 3.5-meter telescope from inside enclosure, photograph by Dan Long

<![if !supportEmptyParas]><![endif]>

Several facility-class instruments are available for routine use on the telescope, including an infrared imager and grism spectrometer, a medium-resolution visible-light spectrograph and imager, a high-resolution visible-light CCD imager, and an echelle spectrograph. Also, a number of specialized visitor-supplied instruments have been successfully used with the telescope, some for limited observing runs and others on a continuing basis. Projects to upgrade existing instruments and for acquiring next-generation instruments have either been completed, are well underway, or are at the proposal stage. These include a detector upgrade to new low-noise CCDs for the medium-resolution visible-light spectrograph and imager (completed in 2002), a new IR imager with a medium field detector and Fabry-Perot etalon (under construction), a near-IR spectrograph (under construction), a new medium-field visible CCD imager (being designed), and a low-noise detector upgrade to the echelle spectrograph (proposed). Continued use of more specialized "visiting" instrumentation is expected and welcome, which include a Fabry-Perot tunable narrow-band visible imager, a laser lunar ranging experiment, and a Fourier Transform Spectrometer, among others. Additional information on most of the existing and future instruments can be found at http://www.apo.nmsu.edu/Instruments/.

<![if !supportEmptyParas]><![endif]>

<![if !supportEmptyParas]><![endif]>

<![if !supportLists]>1.3<![endif]>Operational overview

<![if !supportEmptyParas]><![endif]>

The telescope systems were designed and built for real-time remote operation through the Internet. Most (about three-quarters) of all observing is done remotely, with in-person visits to the site by astronomers being mainly for installation and testing of new instruments or for training purposes. Multiple independent science programs share the telescope in turn on the same night, often using more than one scientific instrument. Remote users also collaboratively use the telescope simultaneously from different off-site locations. Synoptic observing programs and rapid-response observations are frequently accommodated.

<![if !supportEmptyParas]><![endif]>

The manner in which observatories support users ranges from "help-yourself" to "full-service." The APO user support model is somewhere in the middle, and is often described as an equal partnership between the consortium astronomers and the observatory staff. User support is helped by easy telescope access afforded by the remote observing capability and on-line user information systems. Operational costs are kept low by allowing and encouraging the direct involvement of the astronomers in the operations and engineering projects. Also, consortium astronomers have first-hand involvement with the data acquisition process and can more readily understand and appreciate the quality of their data. The APO remote observing systems promote this hands-on observing approach while enabling flexible and semi-reactive telescope use, which saves a significant amount of astronomers' time (and funds) by greatly reducing the amount of travel necessary to observe. While Principal Investigators of the science programs are from ARC-affiliate institutions, collaborations with scientists and students from outside the consortium institutions are welcome and numerous.

<![if !supportEmptyParas]><![endif]>

The telescope is scheduled by quarters using proposals solicited and prioritized within each of the consortium institutions. Except for brief synoptic observations and target of opportunity programs, each night is typically divided into halves. These half-night blocks provide adequate on-target and calibration time, plus simplify the manual scheduling process. Several nights of Director's Discretionary Time are reserved in each quarter and are allocated ad hoc for a variety of special science programs, targets-of-opportunity, or engineering.

<![if !supportEmptyParas]><![endif]>

The observing proposal form is relatively simple, requiring only the science justification information necessary for selection and prioritization by the PI's institution, PI contact information, and brief observational and scheduling parameters. After selection and prioritization, the institutional schedulers electronically submit the proposals to the Director, who manually constructs a three-month schedule. Proposal priorities, lunar and other scheduling constraints, the balance between institutional allocations, and inclusion of engineering time, etc., are reasonably well accommodated. Most of the scheduled programs are allocated one or more half- or full-nights, unless they require a shorter observing duration such as synoptic or target-of-opportunity programs. These shorter programs are in the minority but do involve a significant amount of observing time. The proposal oversubscription rate varies somewhat between the ARC institutions, but typically runs about a factor of two.

<![if !supportEmptyParas]><![endif]>

<![if !supportEmptyParas]><![endif]>

<![if !supportLists]>1.4<![endif]>Observing modes

<![if !supportEmptyParas]><![endif]>

Typically, two or three observing programs are scheduled for adjoining time intervals in the same night. These science programs often employ different instruments, observers, and institutional affiliations. While observers can elect to either use the telescope in person or remotely through the Internet, most choose to observe remotely. Remote observing is conducted using a Macintosh-based application program, which is downloaded in advance to the remote observing stations from the site. A new platform-independent remote-observing software program is under development. Data communications with the site are handled through a dedicated T1 line from Apache Point to Las Cruces, New Mexico; from which point the data and command packets are transferred back and forth from the remote observing stations through Internet2.

<![if !supportEmptyParas]><![endif]>

The remote observer can carry out nearly all essential observing functions, including telescope and instrument control, guiding, quick-look quality assurance, and data retrieval.However, all telescope operations are attended and supported by an on-duty Observing Specialist, who is responsible for telescope safety as well as providing technical assistance to the on-site or remote astronomer who is using the telescope. Observing Specialists perform all instrument changes at night single-handedly, typically in a few minutes per change. The telescope design also enables some instrument changes to be performed quickly without human assistance, and allows for multiple instruments to be mounted on the telescope simultaneously and kept ready in a "standby" mode. For a number of programs, Observing Specialists are active collaborators in that they conduct the observing and participate in data reduction and analysis. In principle, the telescope could be operated without an on-site Observing Specialist, but the value of having staff experts assisting the remote astronomers has proven to be worth the cost for efficiency and telescope safety reasons-many remote astronomers use the telescope only occasionally and find they can use the telescope time more effectively when they work together with Observing Specialists, even though the partnership is virtual.

<![if !supportEmptyParas]><![endif]>

<![if !supportEmptyParas]><![endif]>

2. future considerations & plans

<![if !supportEmptyParas]><![endif]>

2.1 Overview

<![if !supportEmptyParas]><![endif]>

The present availability of 6- to 10-meter telescopes, with more and larger ones planned, will clearly influence the future role of mid-sized telescopes-the number of 4-meter size telescopes is probably less than the number of existing and planned 6- to 10-meter telescopes! Although the ARC 3.5-meter telescope is used mostly as a general-purpose facility for a wide range of smaller projects, it has potential capabilities and modes of operation that support unique and/or larger investigations not easily accommodated on the largest telescopes. One of the planning issues for ARC is whether a long-range strategy should maintain and emphasize the current mode of operation, or to place more priority and resources toward dedicated and specialized projects that are uniquely matched to the telescope and its instrumentation. Such projects include development of new specialized instrumentation, changes in telescope scheduling and observing modes, implementation of long-term survey projects, optimization of rapid follow-up, etc. The combination of remote observing access plus its fast instrument change capability enables new observing modes with the 3.5-meter telescope, opening unique classes of scientific exploration and educational experiences.

<![if !supportEmptyParas]><![endif]>

As attractive as these new observing modes and specialized projects seem, many of the consortium astronomers understandably wish to retain the current operational model generally as it is, adding and upgrading instruments and preserving most of the observing time for individual projects, especially those conducted by graduate students. These users feel that careful consideration of new observing modes and specialized projects should be given, but that the observatory should not be tempted into chasing "fad" science at the expense of established research areas. For these traditionally minded users, the cost-effective operation of the ARC 3.5-meter telescope and remote observing capability uniquely allow many small projects to get significant amounts of telescope time, possibly more than would be accessible with the new generation of larger telescopes for which the user demand and oversubscription are naturally greater.

<![if !supportEmptyParas]><![endif]>

ARC widely recognizes the importance of obtaining next-generation instruments for the 3.5-meter telescope-whatever balance of science priorities is chosen for the future use of 4-meter telescopes, instrumentation is generally what drives scientific productivity over the longer term. Many of the ARC astronomers have interests in wide-field imaging, multi-object spectroscopy, and infrared spectroscopy, observation types not provided by the current suite of facility instrumentation. Bringing the existing imagers and spectrographs to near state-of-the-art capability is already underway, a faster, cheaper way of keeping the instrumentation current. But part of the long-range planning for any modern telescope must include a program for replacement of instrumentation, except for those instruments dedicated to long-term survey and temporal study programs. As given later, upgrading and replacing facility instrumentation can involve making tradeoffs that work counter to specialization.

<![if !supportEmptyParas]><![endif]>

In the remainder of this section, some of the possible specialized science uses of the ARC 3.5-meter telescope are given and contrasted with a discussion of maintaining and upgrading the telescope for general-purpose observing programs.

<![if !supportEmptyParas]><![endif]>

<![if !supportEmptyParas]><![endif]>

2.2 Niche science opportunities

<![if !supportEmptyParas]><![endif]>

Several niche science and observing modes are possible with the ARC 3.5-meter telescope, which capitalize on its several unique strengths.

<![if !supportEmptyParas]><![endif]>

Fast-response observations of transient phenomena:The ARC 3.5-meter telescope is ideally suited for "fast-attack" observations of transient and target-of-opportunity programs. Remote operability, rapid instrument change capability, and good telecommunication with a broad user community promote the use of the telescope and instruments for rapid follow-up of time-critical phenomena. The telescope has been used successfully for optical observations of gamma-ray bursts, gravitational lens events, and extra-galactic supernovae. Coordinated observations with other ground- and space-based telescopes are often undertaken. As an example, a program is currently underway to obtain high-resolution echelle spectra of gamma-ray bursts using an automatic spacecraft alert to trigger an alarm at APO. If conditions warrant, the telescope is diverted from whatever program is running to take a high-resolution echelle spectrum of the gamma-ray source. The time from the alert to the start of the echelle exposure can be less than a few minutes.

<![if !supportEmptyParas]><![endif]>

Remote-control, queue-scheduled observations:Since it is possible to change instruments rapidly with the ARC 3.5-meter telescope, many remote (and on-site) users opt to not specify in advance which instrument or exact program they plan to use, but rather adapt their observing plans in real time based on observing conditions. This is tantamount to running a small queue-scheduled observing program by one astronomer. One limitation is that an individual astronomer's private queue of programs may not take full advantage of all possible observing conditions, nor contain any of the highest-ranked programs that require the most rare seeing or transparency conditions.

<![if !supportEmptyParas]><![endif]>

Larger, more complex, queue-observing models have been prototyped and tested on smaller and mid-sized telescopes, and queue observing has since been implemented on some of the largest new telescopes. "Service" queue scheduling (where the PI team is not directly involved with the data taking) can often make better use of the more expensive telescope resources, and it facilitates the completion of the higher priority science programs that require rare observing conditions. This is in contrast to the classical observing scenario where the scheduled astronomer attends the observing and makes best use of whatever the conditions and equipment allow. Running a large queue-observing program in service mode is expensive and labor-intensive, suitable for only the largest telescopes. For a full-blown queue operation, the on-site service observing staff must have extensive scientific expertise to make the numerous necessary tactical decisions, and the proposers need to prepare a fairly detailed explanation of their observing requirements. Complex software is needed to plan and prioritize the various observing program options between dozens, if not hundreds, of competing observing proposals. Satisfaction with the data products is not always universal, as the end customer is often unfamiliar with the trade-offs necessary to obtain the data, or is psychologically disinclined to trust others to take observations in their place.

<![if !supportEmptyParas]><![endif]>

Hands-on remote observing allows for a form of queue observing that would alleviate these problems. For those astronomers prepared to take their chances, a block of queue-scheduled time on a remote-accessible telescope would be allocated to a handful of remote astronomers who together have programs requiring a wide range of observing conditions and scientific priority. The astronomers would be on-call during weeklong blocks of time, and offered the time for their program(s) on a prioritized first-refusal basis if the observing conditions were suitable. If the highest priority program were untenable because of conditions, then the next astronomer needing the existing conditions would be given the option for remote use of the telescope. In the case where all the astronomers in the queue declined to use the time and if conditions allowed, the telescope would then be used for synoptic or other observatory programs with site staff conducting the prescribed observing. Although this relatively simple model of remote queue-observing has not yet been attempted at APO, in principle it is a way to optimize observations to better match changing observing conditions using a richer queue of programs than can be provided by a single observer.

<![if !supportEmptyParas]><![endif]>

Large (and backup) programs:Three of the recognized strengths of modern 4-meter class telescopes are wide-field imaging, spectroscopy of medium-faint objects, and high-dispersion spectroscopy of brighter targets. "Large" science programs of these types can require huge blocks of telescope time. Although it is difficult for individual ARC astronomers to win more than a few nights per year of time on the 3.5-meter telescope, groups of astronomers could submit proposals to conduct large projects on the telescope by pooling their allocations. In practice, however, the typical ARC astronomer seems to prefer smaller observing projects, usually with two or three collaborators at most, which use about a week of telescope time to complete over a year or so. Also, many of the ARC astronomers are already engaged in pursuing other major projects with data from the SDSS, which is in itself a large project.

<![if !supportEmptyParas]><![endif]>

A formidable, though attainable, goal is to build and promote large observing programs for the 3.5-meter telescope that would win the interest of sizeable groups of ARC users. As an example, one suggested large program would be to create a high-resolution echelle spectroscopic catalog of several thousand of the brightest stars. This program would be executed in "backup" service mode by observatory staff using observing conditions unsuitable for other projects. It is clearly the role of the Director to make a case for the importance of such programs and to solicit the collaborative involvement of the user community, especially if it is felt that the promotion of large programs is an important niche priority for the telescope's future. It is also conceivable that large observing programs would be undertaken with the 3.5-meter telescope in collaboration with other facilities.

<![if !supportEmptyParas]><![endif]>

Synergy with other telescopes, follow-up observations:In the past few years, synergism between the consortium's 3.5-meter telescope and the SDSS project has been a major scientific success. Many of the most spectacular and important scientific results produced by both projects have resulted from 3.5-meter follow-up observations of SDSS discoveries. The SDSS is continuing to provide the 3.5-meter telescope with a steadily growing source of fairly faint and interesting spectroscopic targets. Also, other telescopes could be effectively used in tandem with the 3.5-meter telescope in complementary ways. This is particularly true for follow-up observations, which usually can be quickly accommodated with the 3.5-meter telescope.

<![if !supportEmptyParas]><![endif]>

There is an ongoing community discussion about combining observatory resources in the U.S., both federal and private, into a coordinated system. Although a "national observing system" might avoid some duplication of capability between observatories, any such system would be compelled to impose complex and perhaps onerous requirements on instrument specification, user interfaces, and observing protocols. It could also hamper the important hands-on aspect of observing and instrument development by small groups or individuals, and might suppress innovation, experimentation, and philanthropy by individuals. In contrast, smaller-scale arrangements for sharing telescope and instrument time would be somewhat less cumbersome than a national system. At other private and consortium-based 4-meter telescopes, hands-on remote observing is under study and development. Telescope-sharing arrangements can be imagined where the user communities of two or more telescopes would have limited remote access to each other's telescopes at different locations and/or with different instrumentation in return for sharing comparable amounts of time between telescopes within the combined user communities. If the remote operation interfaces were largely similar and simple between the partner telescopes, users would have relatively seamless, though limited, access to a broader suite of instrumentation and possibly different sky coverage than would be available at their home facilities.

<![if !supportEmptyParas]><![endif]>

Surveys, synoptic and temporal monitoring programs:With its moderate aperture, wide field of view, and easy access to a variety of instruments, the ARC 3.5-meter would undertake several kinds of long-term studies. With a new wide-field visible or infrared imager, various kinds of imaging surveys are possible, including extensions of existing surveys taken with smaller telescopes. Multi-object spectroscopic surveys are also possible. In addition, studies involving the time domain of astronomical phenomena have become increasingly interesting. Using queue scheduling in service mode, the ARC 3.5-meter telescope would allocate small blocks of time on a frequent time interval for the execution of a variety of innovative synoptic and temporal monitoring programs.

<![if !supportEmptyParas]><![endif]>

Solar system programs:Studies of solar system objects are well served by 4-meter class telescopes with good imaging capabilities, easy user access, and appropriate instrumentation; it is often the case that a larger aperture telescope is unnecessary or would be difficult to justify its use. The ARC 3.5-meter telescope has conducted many solar system science programs, including cometary impacts on Jupiter, Venusian atmospherics, and Kuiper Belt Object discoveries. The ease in which prototype and specialized instruments are integrated with the telescope enables unique observational opportunities; an acousto-optic tunable infrared camera has been a frequent and successful APO visiting instrument for studies of planetary atmospheres, for example. Also a laser lunar ranging project has been proposed which would enable earth-moon relative distance measurements to an accuracy of about 1-millimeter, which could yield an order of magnitude improvement in the precision of three important tests of the basic properties of the gravitational interaction.

<![if !supportEmptyParas]><![endif]>

New instrument testing:The ARC 3.5-meter telescope design easily accepts new instrumentation. The Nasmyth instrument port has been used by several visiting prototype instruments over the past several years, and instrument developers have found the telescope well suited and convenient for commissioning new instruments. We intend in the future to continue encouraging the use of visitor instruments. This is motivated in part as "community" support for those instrument builders who need testing time for new instruments on a telescope of this size. Also, by encouraging the commissioning of these new instrument at APO, unique science opportunities are afforded as is the possibility of retaining the new instrumentation for general use by consortium astronomers as replacements or upgrades to aging facility instruments. Any mid-sized telescopes that can afford easy focal plane access for new instruments will be in demand for commissioning the next-generation instruments for the largest new telescopes.

<![if !supportEmptyParas]><![endif]>

<![if !supportEmptyParas]><![endif]>

2.3 General-purpose programs

<![if !supportEmptyParas]><![endif]>

To maintain the continuing support and involvement of the majority of its astronomers and students, the ARC 3.5-meter telescope will need to sustain a core capability for "facility-class" instrumentation and hands-on (both remote and on-site) observing. The complement of instruments originally commissioned with the telescope includes a medium-field visible imager/photometer, a narrow-field IR imager with grism spectrometer, a low- and medium-dispersion visible spectrometer, and a high-resolution echelle. As mentioned above, the consortium has found the resources to upgrade the detectors on the low/medium dispersion spectrometer, and to replace the IR imager with a device offering a larger field of view and lower noise. Other projects for the facility instruments include a low-noise detector upgrade for the echelle spectrograph, building a near-IR spectrometer, and the development of a low-noise multi-color visible imager.

<![if !supportEmptyParas]><![endif]>

A fundamental problem with astronomical instruments, both facility and specialized, is that they have relatively short service lifetimes. With the pace of advances in technology, instruments have a useful lifetime of less than a decade. If four or five facility instruments are to be maintained, upgraded, and replaced on this time scale, an observatory needs to plan and budget for a major upgrade or replacement averaging every few years. A typical instrument upgrade or replacement for a 4-meter telescope can cost about $1M, which is a substantial fraction of the annual operational and capital development budget for most telescopes in this class. A telescope can have a lifetime of more than 25 years, but if the instruments are not kept near the state-of-the-art, user communities will not be satisfied with the scientific potential of the observatory as a whole and may lose interest in its continued support.

<![if !supportEmptyParas]><![endif]>

Also, overall costs increase with the number of instruments supported. If a modestly funded observatory intends to keep four or five general-purpose instruments in prime operational condition, there is little room left in the staffing plan and budget to support any substantial program of specialized niche instrumentation. One approach to solving both problems is to encourage astronomers to independently propose and find resources to develop new instrumentation, both general-purpose and specialized. Several of the past, present, and planned ARC instrument upgrades have followed this approach, including both specialized instrumentation and general facility-class instruments. ARC will continue to solicit these kinds of grass-roots instrument developments and in return, offer instrument developers varying degrees of support such as cost sharing, telescope time and site support. In general, ARC assistance to the developers is provided at a negotiated level that is keyed to whether the instrument upgrade/replacement is generally useful to the larger ARC user community or is of more specialized value to a small number of users.

<![if !supportEmptyParas]><![endif]>

Although this approach has worked in the past, it can be somewhat inefficient. Being "market-driven" and time-consuming for ARC management, many false starts are suffered. On the other hand, experience has shown that in general, good instrumentation is rarely designed and built by committee. Many of the most productive astronomical instruments have been conceived and built by individuals or small groups with a particular scientific project in mind.

<![if !supportEmptyParas]><![endif]>

<![if !supportEmptyParas]><![endif]>

2.4 Outreach and teaching

<![if !supportEmptyParas]><![endif]>

Because of their lower on-sky operating costs, 4-meter and smaller telescopes are natural tools for public outreach and teaching opportunities-remote hands-on observing makes this even more tenable. The ARC 3.5-meter telescope has been in partnership with the Adler Planetarium for several years, participating in a successful remote-access public program. Planetarium and university astronomers remotely operate the 3.5-meter telescope on Friday evenings for public display, during evening twilight hours at APO that are otherwise not used for science observations. Even during normal science observations, it is conceivable that private or public eavesdropping would be supported over the Internet with little or no impact to the science programs.

<![if !supportEmptyParas]><![endif]>

Furthermore, the ability to provide astronomy students with hands-on access to research-grade telescopes is made easier by remote observing, especially for those students with limited travel support. Hands-on access to observatory facilities is declining with the increased use of observing modes such as queue scheduling and service observing on the new largest telescopes. Astronomy students in the coming years will have increasingly fewer opportunities to learn the hands-on aspects of the field, unless easy access to the smaller telescopes is encouraged. This is extremely important, because the future telescope and instrument builders will graduate from the ranks of these students who have been given extensive observatory experience.

<![if !supportEmptyParas]><![endif]>

<![if !supportEmptyParas]><![endif]>

3. Conclusions

<![if !supportEmptyParas]><![endif]>

The ARC 3.5-meter telescope and its baseline facility instrumentation have been incrementally improved during the first decade of operation to the level where the observing systems are mature, reliable, and routinely provide its user community with useful and often exciting data. Looking forward, the ARC community has begun discussions on what role this facility can have in the company of the new large-telescope facilities recently put into operation, and on the rapidly evolving scientific priorities in the field as a whole. While the future scientific value of 4-meter class telescopes will be partly determined by the vision and leadership of their management and governing bodies, the realization of the potentials and continued viability of the facilities will be largely driven by the scientific priorities of the users-especially if these astronomers and students are actively involved with the telescope and its instrumentation.

<![if !supportEmptyParas]><![endif]>

Two principal ARC priorities for the 3.5-meter telescope are excellence and innovation in science, and graduate student education and training. By identifying, maintaining, and building on the intrinsic strengths of the telescope, the consortium astronomers and students will continue to benefit from both its unique and generic scientific capabilities. Some of the more important future advances in astronomy will come from the innovative use of state-of-the-art hardware, computer technologies, and new observing modes-even with modest aperture telescopes. For APO, it is not presently clear how these kinds of initiatives can be allocated priority while continuing to serve the wider range of interests of the ARC user community. The dilemma for the consortium is to determine the balance between these paradigms, i.e., can they happily co-exist and be affordable. In the future, telescopes of this size with large-field, high-resolution imaging capabilities, state-of-the art CCD imagers and spectrometers, flexible scheduling, remote operations, and easy access for users, will undoubtedly play a significant part in the future advancements in astronomy.

<![if !supportEmptyParas]><![endif]>

<![if !supportEmptyParas]><![endif]>

4. Acknowledgements

<![if !supportEmptyParas]><![endif]>

The author wishes to thank the many ARC astronomers and APO staff who provided perceptive input for this article. Their continued active interest in the long-term planning process is essential and much appreciated.