Gemini Near-Infrared Integral-Field Spectrograph (NIFS)
|
|
|
Instrument Requirements
1.1 Scientific Context
The Gemini 8-m telescopes are designed to achieve unprecedented ground-based image quality using adaptive optics (AO) techniques. Diffraction-limited images have been demonstrated at 2.2 mm and should be achievable over most of the 1.0-2.5 mm wavelength range in good observing conditions. High spatial resolution is the new astronomical frontier. The Hubble Space Telescope (HST) has explored spatial scales to below 0.1² in the optical and more recently in the near-infrared with NICMOS, but to only a tantalizing extent in the latter case. Adaptive optics systems on ground-based 4 m telescopes have begun to realize the potential of this technique, but it will only be with facility AO systems, such as ALTAIR, on 8 m telescopes that the scientific potential of high spatial resolution ground-based astronomy can be exploited fully. Most AO instruments to date have been imaging systems. This is expected for a technique in its infancy. However, scientific returns are amplified many-fold by access to spectroscopic data. The light gathering power of an 8 m telescope is essential to perform high spatial resolution observations at even moderate spectral resolution, and an integral-field unit (IFU) is needed to efficiently sample complex spatial structures. With these facts in mind, it was decided to explore designs for a fully-sampled, diffraction-limited, moderate spectral resolution, near-infrared, integral-field spectrograph for Gemini North. In a rapidly evolving field, it is essential that instrument concepts be relatively simple and lead-times be minimized in order to maximize short-term scientific returns. This demands a relatively simple, fast-tracked instrument that can be made available to the community on the shortest possible timescale, and at modest cost. The Gemini Near-infrared Integral-Field Spectrograph (NIFS) is this instrument.
NIFS was proposed by the Research School of Astronomy and Astrophysics (RSAA) of the Australian National University for use with ALTAIR on Gemini North. The primary purpose of NIFS is to study moderate surface brightness structures around discrete objects that are revealed at high spatial resolution by ALTAIR. As such, NIFS is required to completely sample a small 2-D field on the sky. NIFS will do this using a reflective IFU to divide its 3.0²´3.0² field-of-view into 29 slitlets each 0.1² wide and 3.0² long and to simultaneously obtain spectra for each 0.04² pixel along each slitlet. NIFS will operate in the wavelength range from 0.94-2.50 mm where ALTAIR delivers its greatest gains. The diffraction full-width at half maximum (FWHM) of the Gemini telescopes varies from ~0.03² at 0.94 mm to ~0.08² at 2.50 mm. The NIFS spatial sampling is chosen to sample at close to the Gemini diffraction-limit while maintaining a modest field-of-view. An optical AO guide star will be required for all NIFS observations with ALTAIR. ALTAIR passes a 120² diameter circular field to NIFS. Initially, natural guide stars brighter than R~15 mag located within ~20² of the science object will be needed (Figure 1). This requirement will be relaxed when the laser guide star upgrade to ALTAIR is implemented, possibly towards the end of 2002, but a tip-tilt guide star close to the science object will still be required. The Strehl ratios expected from the telescope/ALTAIR/NIFS combination vary dramatically over the 0.94-2.50 mm range; ~0.2 at J, ~0.4 at H, and ~ 0.6 at K in median seeing (Figure 2). Clearly, NIFS will be most effective for high spatial resolution studies in the K band (1.95-2.50 mm). The laser guide star upgrade to ALTAIR will ensure that these Strehl ratios are achieved over a larger fraction of the sky, but the use of laser guide stars will not improve the best performance achieved by ALTAIR. Most NIFS programs will require good natural seeing to perform high spatial resolution observations. These programs will benefit from queue scheduling on Gemini.
Figure 1: Degradation of H band Strehl ratio achieved with ALTAIR (x=6.5 km) for increasing offset to a natural guide star (NGS) and laser guide star (LGS) in median seeing (ALTAIR CDR).
Figure 2: Sky coverage probability for ALTAIR with laser guide stars (left) and natural guide stars (right) in good (heavy lines) and average (light lines) seeing (ALTAIR CDR).
NIFS will address a wide range of science from studies of star formation and the Galactic center in our galaxy to the nature of disk galaxies at z~1. We identify studies of the demographics of massive black holes in galactic nuclei and studies of the excitation conditions in Seyfert galaxy narrow-line regions (NLRs) as two core NIFS programs. Emissions in the near-infrared spectral region arise primarily from late-type stars, either singly or integrated in galaxies, and from the interstellar medium. Late-type stars exhibit strong CO first-overtone absorption bands longward of 2.3 mm in the K band and CO second-overtone absorption bands near 1.6 mm in the H band. These absorptions characterize the near-infrared spectra of young stellar populations older than ~107 yr containing late-type supergiant stars and old stellar populations in which the near-infrared light is dominated by evolved late-type giant stars. The near-infrared CO absorption bands are also powerful probes of stellar dynamics which will allow NIFS to study massive black holes in the obscured nuclei of nearby late-type spiral galaxies and Seyfert galaxies, and to measure mass-to-light ratios, M/LK, in these objects and more distant starburst and ultra-luminous IRAS galaxies. M/LK varies from ~0.01 to ~2.5 as a stellar population ages from ~107 yr to ~1010 yr. Consequently, determining mass-to-light ratios in galactic nuclei on mass scales dominated by the stellar population provides information on the evolutionary histories of the central regions of galaxies. Dynamical estimates of the total mass in starburst galaxies constrain the low end of the stellar initial mass function in these regions. Near-infrared emission-lines of H I Pa 1.875 mm (accessible at z > 0.05), H I Pb 1.282 mm, H I Brg 2.166 mm, [Fe II] 1.257 mm, [Fe II] 1.644 mm, and H2 1-0 S(1) 2.122 mm are well-known and well-studied emission-lines arising in star formation regions, H II regions, planetary nebulae, supernova remnants, starburst galaxies, and AGN, and are diagnostic of ionization and excitation conditions and kinematics of circumstellar and interstellar gas under a variety of conditions. These emission-lines will be used to measure rotational velocities of LINER-like circumnuclear disks to reveal the presence of massive nuclear black holes, and to determine the excitation and dynamics of NLR clouds in nearby Seyfert galaxies. At high redshift, optical diagnostic emission-lines are redshifted into the near-infrared (Figure 3). Ha will be accessible to NIFS at z>0.43 and is of paramount importance beyond z~1.0 when Hb and [O III] l5007 are no longer accessible in the optical. Redshifted Ha emission will be used to measure rotation curves for disk galaxies at z~1. Hb and [O III] l5007 are redshifted into the K band at z>3.0, with [O II] l3727 accessible in the H band (1.49-1.80 mm). A goal will be to measure dynamical information from these emission-lines in Lyman break galaxies. NIFS should deliver the spectral resolution and sensitivity to study these emission-lines efficiently and to reveal dynamical information about the emitting regions.
Figure 3: Wavelengths of strong absorption and emission features as functions of redshift. The features shown are Mg II l2800, [O II] l3727, Hb l4861, [O III] l4959,5007, Ha l6563, Ca II l8498,8542,8664, CO (Dv=3) 1.6 mm, CO (Dv=2) 2.3 mm. Regions measured by the moderate resolution NIFS gratings are indicated by dashed lines.
NIFS will benefit from the fact that extinction due to interstellar dust is lower in the near-infrared than in the optical; it is frequently quoted that the extinction at K is one-tenth that at V. This permits near-infrared observations of many optically-obscured regions such as Galactic star formation regions, the Galactic center, the nuclei of many spiral galaxies and other starbursting and active galaxies. However, ground-based near-infrared observations at wavelengths shortward of ~2.2 mm are also plagued by OH airglow emission from the Earth’s atmosphere. NIFS will have sufficient spectral resolution to permit measurements between these emission-lines where the detected sky background is greatly reduced. Spectral resolving powers in excess of ~3000 are needed for this. Studies of stellar dynamics in galaxies require spectral resolving powers of ~3000-6000 (corresponding to 50-100 km s-1) for rotational studies and 4000-5000 (corresponding to 60-75 km s-1) for velocity dispersion measurements in the coolest stellar systems. These requirements are met with the two-pixel spectral resolving powers of 4000-5000 that the NIFS moderate resolving power gratings will deliver.
NIFS will also be used without ALTAIR for moderate resolving power near-infrared spectroscopy of extended objects through a 3.0²´3.0² aperture. These observations will benefit from the lower emissivity and higher throughput of the direct telescope beam, and from the large NIFS entrance aperture which will virtually eliminate slit losses even in poor seeing conditions while maintaining moderate spectral resolving power. Indeed, NIFS may be the most efficient Gemini instrument for obtaining near-infrared spectra of extended objects and the most appropriate for use in poor seeing conditions.
1.2 Technical Implementation
NIFS will realize the requirement for fast-tracked development by re-using many of the designs already developed for the Gemini Near-Infrared Imager (NIRI) by the Institute for Astronomy (IfA) of the University of Hawaii. The NIFS science instrument will be mounted in a duplicate of the NIRI cryostat and use a duplicate of the NIRI On-Instrument Wavefront Sensor (OIWFS) to sense tip-tilt and focus corrections. The NIFS science instrument will use the same mechanism encoding architecture used in NIRI and will use the same cryogenic stepper motors adopted for NIRI. This commonality of mechanical designs means that the NIRI mechanism control system hardware and temperature control system hardware can also be duplicated for NIFS with minimal change. The EPICS Instrument Sequencer (IS), Components Controller (CC), and engineering interface software developed for NIRI and the CC for the OIWFS can also be re-used for NIFS with only minor modification. The re-use of NIRI designs for NIFS permits major economies in budget and schedule that will allow NIFS to be commissioned in the shortest possible time and at low cost. NIFS will also be convenient to maintain and to support on Gemini North because of its commonality with NIRI.
The heavy reliance of NIFS on its NIRI heritage makes it efficient to share the construction effort between RSAA and IfA. IfA will be subcontracted by RSAA to duplicate the NIRI cryostat, OIWFS, and control system hardware. It is anticipated that this work will begin soon after successful completion of the NIFS Conceptual Design Review and will take place while the NIFS science instrument design is being developed further. This fast-tracked approach is suggested because of the low risk associated with duplicating components from an existing Gemini instrument, and because of the imperative on reducing the instrument development time.
Operation of a near-infrared instrument at high spatial and moderate spectral resolution places severe constraints on detector format, performance, and stability. NIFS will use a 2048´2048 detector array with 2048 pixels in one dimension devoted to spatial information and 2048 pixel in the other dimension devoted to spectral information. The detector array will be chosen from one of several options available, or possibly available, from the Rockwell Science Center. The baseline detector is a 2.5 mm cut-off HAWAII-2 HgCdTe/PACE device – the current state-of-the-art. The Rockwell Science Center are also developing 5 mm cut-off HgCdTe detectors on CdZnTe substrates using Molecular Beam Epitaxy (MBE) technology for NASA as part of their Next Generation Space Telescope (NGST) program. These devices have lower dark currents, better quantum efficiency, and lower persistence than HAWAII-2 PACE devices. A possibility exists that one of these detectors may become available to NIFS on loan. The Rockwell Science Center also plan to develop 2.5 mm cut-of HgCdTe/CdZnTe MBE devices. These devices should have even lower dark current, but the timescale for their development may be inconsistent with the fast-tracked development required for NIFS.
Use of a Rockwell HgCdTe detector places different requirements on the NIFS detector controller compared to NIRI which operates to 5 mm using a SBRC 1024´1024 InSb ALADDIN array and Wildfire controller developed by NOAO. NIFS will use a San Diego State University SDSU-2 array controller. These controllers have been used extensively by RSAA and many other groups to operate optical CCDs and by IfA to operate Rockwell HAWAII-1 and recently HAWAII-2 detectors. The Gemini optical CCDs will also use SDSU-2 controllers, so NIFS will benefit from the EPICS software effort already invested in this development, and Gemini observatory staff will be familiar with the operation, maintenance, and support of this controller. The detector controller system will be developed by RSAA as a modular part of NIFS.
These approaches to the implementation of NIFS will ensure that the Gemini community has access to a state-of-the-art facility instrument on a short timescale. Compatibility with other Gemini instruments and adherence to Gemini instrument requirements are largely ensured by the NIRI legacy from which NIFS will benefit. Consequently, the primary emphasis of this Conceptual Design Review Documentation is on demonstrating that the NIFS science instrument will achieve a similar high level of performance and result in an instrument package that is equal to, or better than, NIRI in all operational respects. Foremost amongst these are the design requirements for outstanding image quality, high throughput, and low instrumental background.
1.3 Baseline Instrument Parameters
The baseline instrument parameters for NIFS are summarized in the following list:
· Wavelength range: 0.94-2.5 mm.
· Spatial pixel size: 0.04²´0.10² on sky.
· Spectral resolving power: ~ 5300 (two pixels).
· Focal plane occulting masks: 0.1², 0.2², 0.5² diameter circular.
· Order blocking filters: J1, J2, H, K, J, HK.
· Gratings: J1, J2, H, K, J, HK.
· Detector: Rockwell 2048 ´ 2048 HgCdTe HAWAII-2, 18 mm pixels.
· Near-infrared on-instrument wavefront sensor.
- Wavelength range: 0.95-2.5 mm.
- Steerable over 180² FOV (120² used with ALTAIR).
- Image scale: 0.170²/pixel.
- Filters: J, H, K.
- Instantaneous field-of-view: <10².