What Space - Based Telescopes Will be Available in 2010?

NASA has an ambitious program of missions planned and either funded or budgeted through to 2010. Most of these, and certainly the ones that are most relevant to the NGC, are carried out under the ``Astronomical Search for Origins" (ASO) program. The stated goals of the ASO program are to address the following scientific themes:

The major first-generation missions in this program are the Next Generation Space Telescope (NGST) and the Space Interferometer Mission (SIM), to be followed by the Terrestrial Planet Finder (TPF) and to be preceded by a number of more minor, or pre-cursor, missions, many of which have been in existence for some time.

HST
The HST is projected to continue through to 2010, albeit in low-cost non-refurbishment mode. The main instruments during this phase will be ACS (Advanced Camera for Surveys) operating in the visible and near-uv and an ultraviolet spectrograph.

FUSE
FUSE is a modest Explorer class mission that will be launched in late 1998 and which will carry out spectroscopy in the far-ultraviolet, with particular emphasis on determining the Deuterium content of the Universe.

WIRE
This is a fairly modest Explorer class mission to be launched in mid-1999. It will map of order 50 15#15 deg² patches of sky in the 12 and 25 3#3 wavebands with typical sensitivities of order 0.1 mJy. The primary aim is a survey for dusty objects at high redshift, not least as targets for SIRTF.

SOFIA
SOFIA is an infrared 3-m passively cooled telescope mounted in a modified Boeing 747 aircraft.

SIRTF
This is a 0.85m cryogenically cooled telescope that is currently in Phase C/D with launch planned in December 2001. It will cover the 3.5-180 3#3 waveband with broad-band imaging and low resolution spectroscopy (R = 50 - 600). SIRTF will be operated as a facility-class observatory, with an emphasis on early execution of a number of major ``Legacy'' science programs.

NGST
Without doubt, NGST is the space mission of most relevance to the NGC question, and therefore more details are provided below. Broadly speaking NGST is planned as a cooled filled-aperture 8-m telescope operating in the 1-5 3#3 waveband with possible extensions to 0.5 3#3 and to about 30 3#3. NGST will have a range of imaging and spectroscopic instrumentation, including likely multi-object and integral field spectrographs. Launch is planned for 2007 and the telescope will likely be at the Earth-Sun L2 point where refurbishment will not be possible.

SIM
SIM will be an interferometer operating in the visible waveband, with launch planned for 2004. Six 0.3m apertures on a 10m baseline will yield impressive performance for astrometry (4 3#3arcsec astrometry at V = 20 in 10⁵ sec), high resolution imaging (10 3#3arcsec images of the core of M87 in 10⁵ sec) and planet detection (with a nulling interferometer operating at 10⁴ rejection).

TPF
This second generation ASO instrument, intended to be launched in 2010+, combines technologies from NGST and SIM. TPF is planned as a set of large (4m?) cooled telescopes acting as an interferometer on a 100m baseline in the 7 - 17 3#3 waveband. This will give 0.025 arcsec resolution, 10⁶ null rejection for isolating the light from extra-solar Earth-like planets. The few hundred stars within 13 parsec will be surveyed. Earth could be detected broadband in 1 hr and an R=20 spectrum (sufficient in principle to give evidence of CO₂, H₂O and O₃) obtained in 1 week of observing.

One point should be made concerning this program. It is directed squarely at the search and study of extra-solar planets with an emphasis on interferometric techniques - the ASO program is also funding the Palomar Prototype Interferometer and the Keck Interferometer on the ground. NASA's ultimate goal is the Planetary Imager (PI), which would consist of multiple sets of 8m class telescopes acting as an interferometer on a 6000 km baseline, the technology for which is not even ``on the horizon''. At least some of the enthusiasm at NASA for NGST, the most traditional ``astronomical" mission, arises from the perception that the technology required for NGST will be applicable to the later very ambitious interferometers.

ESA's current approved space astronomy program is both spectrally broader and generally shorter term than the NASA program outlined above. The most relevant missions to the NGC question are the FIRST and Planck missions with XMM and INTEGRAL being somewhat peripheral.

FIRST
Now planned for launch in 2007, FIRST will be a cryogenically cooled 3.5m telescope operating in the 80 - 670 3#3 waveband, with imagers and spectrographs with R=2000 - 10⁷. This is aimed primarily at studying dust-enshrouded phases of galaxy evolution at high redshift as well as local star-forming regions.

Planck
As currently envisaged, Planck will share a 2007 launcher with FIRST. Planck will map the Cosmic Microwave Background (CMB) with 4 arcmin resolution and definitive sensitivity. Planck should extract essentially all of the available cosmological information available from the CMB.

XMM
This is an observatory in the soft X-ray region (200 eV to 17 KeV) to be launched in January 2000. XMM will host 3 scientific instruments: an imaging X-ray camera, an X-ray grating spectrometer, and an optical monitor. XMM will be targeted to the study of stellar coronae, cataclysmic variables, supernova remnants, normal galaxies, active galactic nuclei, and galaxy clusters.

INTEGRAL
INTEGRAL is a gamma-ray observatory, to be launched in 2001, dedicated to fine spectroscopy (16#16) and fine imaging in the energy range 15 keV - 10 MeV.

In the longer term, ESA is planning on participating in NGST, and in undertaking some or all of the following ``Horizon 2000+'' missions currently under study. Launch dates for these would be in the 2008, 2012, 2016 timeframe.

GAIA
A follow-on to Hipparcos, GAIA is a filled-aperture telescope with similar astrometric performance to SIM. The emphasis however would be on an all-sky survey of all 17#17 objects with V < 15.

IRSI (also known as DARWIN)
This is an IR interferometer with similar capabilities and science objectives to TPF.

XEUS
A large X-ray telescope aimed at spectroscopy of early Universe X-ray sources.

NGST is currently in a pre-phase A Study Phase with Phase A starting in mid-1999 and launch planned for 2007. The current design specifications for NGST are:

The sensitivity gains of NGST relative to ground-based 8m class telescopes in the near-IR and mid-IR arise from: (a) the reduced background relative to that produced by the atmosphere and by ambient temperature optics and (b) the attainment of diffraction limited images over a wide field (19#19 arcmin).

NGST achieves its remarkable performance by virtue of being above the Earth's atmosphere (which has strong OH emission lines between 0.8 and 2.0 3#3, strong continuum thermal emission at >2 3#3, and absorption which limits ground-based observations to atmospheric windows which are reasonably open shortward of 2 3#3 but narrow and only partly open at longer wavelengths). For NGC, the reduction in background varies between about a factor of ten in the near infrared to a factor of 10⁶ at 10 3#3, giving gains in sensitivity from this reduced background alone of factors of 3 to 10³, depending on wavelength. NGST also gains by the attainment of diffraction-limited performance without the need for Adaptive Optics, which is limiting in terms of field of view and has difficulties in the optical waveband.

The implication of the fact that NGST is essentially as large as any ground-based telescope is that, although the ground-based telescopes will have more up-to-date instrumentation (since NGST will not be astronaut-serviceable), NGST will be able to do any observation at wavelengths longer than 1 3#3 better than any present ground-based telescope, except for very high spectral resolution spectroscopy, R >> 10⁴, where the performance is limited by the detectors and only a limited wavelength range is usually required.

We look first at the aperture required to match NGST's performance at wavelengths longer than 1 3#3. The sensitivity gain in the diffraction-limited case is simply D²b^-0.5 (D is the relative diameters of the telescopes and b the relative backgrounds), or in the case where the image quality is independent of aperture, Db^-0.5. Thus, if a ground-based telescope is fully-diffraction limited with AO, the aperture required to match NGST is only 1.7 20#20 8m at 1 3#3, but increases past 2 3#3 to 10 20#20 8m at 3 3#3 and 30 20#20 8m at 10 3#3. Clearly observations at >2 3#3 are unattractive from the ground, since very large telescopes are required to even match NGST's performance.

At wavelengths less than 2 3#3, a diffraction-limited 10m with efficient AO is, in principle, already comparable to the performance of NGST (although NGST still avoids the highly featured OH background, the effects of the atmospheric transmission windows and isoplanetic effects associated with AO). Thus at these wavelengths, ground-based astronomy is still competitive with NGST.

The ``uniqueness space'', thus, for a very large ``next generation" ground-based telescope is in observations at wavelengths <2 3#3. Gains in the speed to reach a fixed point-source sensitivity in the background-limited case (i.e. imaging and low resolution spectroscopy) depend on both the aperture (as D²) and the continued shrinking of the diffraction-limited image (again as D²). Thus a diffraction-limited 25m represents a gain in speed of a factor of 40, or a gain in sensitivity of a factor of 6. In the non-background limited case (e.g. in very high resolution spectroscopy) the gain in speed is simply D². Even without the gains of AO diffraction-limited imaging, a 25m would represent as large a gain over the Keck as Keck itself was over previous generations of 4m class telescopes.

The most obvious need for a ground-based 25m telescope to exploit the gains of NGST would be in spectroscopy of very faint high redshift sources in the 1 to 2 3#3 regime. The Lyman-21#21 line is still shortward of 2 3#3 for all z < 15. Although NGST may be able to detect forming star-clusters at this redshift (it can detect a system forming one solar mass per year for 10⁶ years to z = 20 in a pessimistic cosmology, i.e. 22#22), it will not be able to spectroscopically confirm these redshifts - such objects have 23#23 (AB = 32.5). Very deep spectroscopy with NGST itself could reach to 24#24 with heroic efforts, i.e. to within about 3 magnitudes of the imaging limit, but to probe still deeper would require a larger collecting area (and still heroic efforts). In this sense the 25m would be to NGST as the Keck is to HST - in both cases the aperture increase is about a factor of four.

In the Figure below we summarize the above discussion by plotting the relative sensitivities of a 25m ground-based instrument compared to the NGST. This is done for both imaging and spectroscopy of point sources in the wavelength region 1 - 20 3#3. The calculations adopt the parameters specified in the paper by Gillett and Mountain (in Science With the Next Generation Space Telescope, eds E.P. Smith & A. Koratkar, ASP Conf. Proc 133, in press) which are reasonable ones for the respective instruments. What is important here is that both telescopes are likely to have segmented mirrors and so are given equivalent Strehl ratios in the calculations.

**Figure:** Point source sensitivity of a ground-based 25m telescope compared with an 8m NGST for both imaging and spectroscopy. Relative exposure times for a given signal/noise ratio will scale as the square of these sensitivity ratios.
25#25

For imaging point sources, NGST and a ground-based 25m telescope will have equivalent sensitivities at 1 3#3 with NGST superior by about a factor of 100 at 10 3#3. For wavelengths shorter than 1 3#3, the 25m ground-based telescope has superior performance. The gains of the 25m relative to NGST for spectroscopic applications are quite spectacular since these may be detector limited for NGST. For point sources observed with a spectroscopic resolution R = 1000 at 10 3#3, NGST is more sensitive by approximately a factor of 100, the sensitivities of the two telescopes are equal at about 2.3 3#3 and by 1 3#3 the 25m ground-based instrument wins by more than a factor of 5. Current design does not push the NGST sensitivity below 1 3#3 but even if it does extend to this short a wavelength, the 25m instrument will have absolutely no competition in the optical for either imaging or spectroscopy.

NGST will be operated as an facility-class observatory for a 5-10 year mission lifetime. The main science areas, as defined in the Design Reference Mission, encompass
(a) the detection of the very first stars and galaxies at redshifts z >> 5
(b) elucidating the physical origins of the Hubble sequence of galaxy types that is believed to be set up during the 1 < z < 5 range
(c) the study of the distribution of dark matter in the Universe through weak lensing effects
(d) the study of stellar populations in nearby galaxies
(e) the study of star-formation regions in our own and other galaxies
(f) the detection of low mass bodies in star-forming regions
(g) study of low mass bodies in the Solar System (especially Kuiper belt objects).

NGST is budgeted through to launch in the ASO program and enjoys strong political support in the U.S. It is therefore very likely to proceed as planned. ESA and CSA are planning to participate at the $200M and $50M levels respectively.

There are three main implications for an NGC in the timeframe of 2010 of the space missions outlined, particularly as a telescope to complement NGST.

In terms of sensitivity, NGST will likely surpass any ground-based facility of less than 25m aperture longward of 2 3#3. However, the field of NGST will be limited, and the instrumentation will be unchangeable and will represent 2003 technology. Furthermore, NGST's life will be limited to 5-10 years. In general terms ground-based facilities will be most competitive for

(a) the optical waveband, where the background reduction in space is minimised.
(b) high resolution spectroscopy, where the background is less important than the detector performance.
(c) wide-field imaging and spectroscopy, especially at high image quality.