Skip to sections.
Located on the summit of Mauna Kea on the island of Hawaii, the National Astronomical Observatory’s large-scale optical infrared telescope (the Subaru Telescope) is equipped with several different observing instruments, such as the HDS (High Dispersion Spectrograph), which was introduced in the previous article (Part 1). One of these instruments, FOCAS (the Faint Object Camera and Spectrograph), has proved to be effective in the observation of distant galaxies in the furthest reaches of space. Nikon took charge of the manufacture of the main body and the principal lens assembly of FOCAS. On this occasion, we asked Professor Masanori Iye of the National Astronomical Observatory about the structure of FOCAS, the significance of the detection of distant galaxies, the next-generation observing instruments due to be installed on the Subaru Telescope, and the Extremely Large Telescope (ELT) project that is currently under research.
FOCAS mounted at the Cassegrain focus of the Subaru Telescope. FOCAS enables the basic visible-light observation of the Subaru Telescope, such as imaging observation, spectroscopic observation, and polarization observation.
The strategy used for studying the Universe on large is as follows. To start with, survey photographs are taken, and these are then used to search for interesting objects, in our case, distant galaxies in the furthest reaches of space. Only extremely faint light reaches us from these objects; however, what little light there is captured by the 8-m-aperture Subaru Telescope, and analyzed using a spectroscopic technique.
The HDS (High Dispersion Spectrograph) processes light from comparatively close, bright stars that are within our own galactic system (the Milky Way). Since the light from these stars is abundant, the spectrographic resolution is increased, and they can be studied in detail. However, our humble distant galaxies are too dark to observe with the HDS. For this reason, the detailed examination of the spectrum was abandoned, and the idea of FOCAS (the Faint Object Camera and Spectrograph), an observing instrument that can obtain information of faint objects using low-to-medium-dispersion spectroscopy, was conceived.
Prior to the construction of FOCAS, we constructed a similar instrument for use on the 2-m-aperture telescope at the Okayama Astrophysical Observatory. Associate Professor Toshiyuki Sasaki constructed an instrument known as OOPS (the Okayama Optical Polarimetry and Spectrometry system) at the National Astronomical Observatory especially for use on a 91-cm-aperture telescope. OOPS could be described as the prototype for FOCAS. OOPS came onto operation just as computer technology was rapidly developing, and the system allowed computer control of functions ranging from turning observing instruments on and off to saving observation data. We made use of our experience with OOPS when we designed the FOCAS control system. Likewise, since we had expertise on the CCD camera that was to be mounted on FOCAS, we decided to construct it. However, there was no way that we could construct a sophisticated lens system for FOCAS, which is why we asked Nikon to manufacture lens systems and the overall structure.
Research using an optical telescope requires three basic instruments: an observing instrument that takes photographs, a high-dispersion spectroscopic observing instrument, and a high-sensitivity low-dispersion spectroscopic observing instrument. On the Subaru Telescope, the Prime Focus Camera takes photographs, the HDS handles high-dispersion spectroscopy, and FOCAS handles low-dispersion spectroscopy. This three-instrument suite was described in the planning document for putting the concept of the Subaru Telescope into practice (commonly known as the Blue Book), which was drafted in 1989. This is still today believed to have been the correct strategy.
FOCAS is mounted at the Cassegrain focus, on Subaru’s hip (or backside) (laughs). FOCAS is one of four instruments at the Cassegrain focus—the others being the Infrared Camera and Spectrograph (IRCS), which was developed jointly with the University of Hawaii, the Cooled Mid-Infrared Camera and Spectrometer (COMICS), used for long-wavelength optical observation, and the Coronagraphic Imager with Adaptive Optics (CIAO), which enables the observation of dark celestial objects by masking out the light from bright celestial objects that are close to them.*1 Last year, the Multi-Object Infrared Camera and Spectrograph (MOIRCS) also became available for use and IRCS was moved to the Nasmyth focus to be used in conjunction with the new Adaptive Optics system. The Cassegrain instrument flange of the Subaru Telescope is used as a common mechanical interface for all the Cassegrain observational instruments. The instruments can be interchanged semi-automatically by means of a moving wheeled platform that runs on magnetic tape affixed to the floor. Since the interchange takes at least an hour, in principle this operation is not carried out on nights when observation is possible. In the room that houses the observing instruments, there are four standby platforms that are similar to the one at the Subaru’s Cassegrain focus. These standby platforms supply electrical power and network connectivity to the observing instruments, and keep the instrument detectors ready in a cooled condition, enabling observation to be carried out at any time.
First, there were some limitations imposed by the Subaru Telescope on observing instruments so that they can be mounted at its Cassegrain focus. Instruments could be no more than 2 meters in diameter, 2 meters high, and 2 tons in weight. These limitations bounded the design of FOCAS. One outstanding feature of the FOCAS structure is the pipes that are formed into a truss shape. These lightweight trusses are made of carbon fiber reinforced plastic (CFRP) and are designed to support the entire FOCAS assembly at its center of gravity, to minimize the mechanical flexure of FOCAS under its own weight. If the main optics of FOCAS were supported directly by the telescope flange plate with which it is attached to the Subaru Telescope, it would flex considerably as the orientation of the telescope changes to follow the target. In the initial planning stage, there was a proposal to install a system to automatically regulate this flexion; however, this truss system to support the main FOCAS modules at its center of gravity has successfully reduced the degree of flexion to approximately one-tenth. Nikon took charge of this structural design, the simulation-based optimization for reducing strain, and the main optical system for FOCAS, which is described below.
The FOCAS control screen: the status of each observational mechanism can be monitored and controlled using the GUI.
The remote monitoring room at the National Astronomical Observatory in Mitaka, Tokyo. The status of the entire Subaru Telescope, the status of each instrument, and observation data can be monitored in real time.
The internal structure of FOCAS is designed to allow a straight light path without using a mirror to avoid any loss of the precious photons captured by the Subaru Telescope. The multi-slit mechanism of FOCAS first extracts only the light from the target objects. The light introduced to FOCAS is then expended and transformed into a parallel beam using a set of collimator lenses. The optical elements that can be inserted in the collimated beam are filters that only extract light of certain wavelengths, dispersive grisms that are a combination of a diffraction grating and a prism, and polarizers that analyze the light in terms of its polarization components. This is where the light from a celestial object is processed according to the objectives of observation. Finally, the treated light is gathered by a set of camera lens and is photographed by a CCD camera.
One distinctive feature is the internal turret structure. In accordance with the observation program, the timely interchange of various optical components such as filters and grisms can be carried out in under a minute. The system is controlled by computer in combination with sensors. The graphic user interface on the control screen helps observers to recognize which filters are currently in place and which set of optical elements are to be used next. This control system was developed on the OOPS system.
Incidentally, the operational status of FOCAS and the entire Subaru Telescope can be monitored from here at the National Astronomical Observatory in Mitaka. In addition, actual observation data is transmitted to Mitaka in real time and archived. Since the Pacific Ocean communication lines linking Hawaii and Japan are still technically insecure, we do not issue commands for moving the Subaru Telescope or the observing instruments from here. However, since we are continuously monitoring the parameters necessary for maintaining the instruments, when there is a problem of some kind, we can resolve it from here.
It is intended to study distant galaxies. FOCAS was built to show us what the universe looked like at its inception.
The universe began with the Big Bang, and during the first three minutes it expanded enormously. During this period protons and helium nuclei were formed through strong interaction. However, the atomic nuclei of carbon, nitrogen, and oxygen that make up the human body were not formed, since the universe was expanding and the temperature was falling too quickly. Comprised only of hydrogen and helium atoms, the universe continued to expand in this fashion. Approximately 380 thousand years after the Big Bang, the universe cooled down to a temperature of around 3000K, and as a consequence protons and electrons combined to form neutral hydrogen atoms. Since the neutral hydrogen atoms do not interact with photons, they cannot be observed optically today. There are also many aspects of this time that are not understood theoretically, and this is why it is called the Cosmic Dark Ages.
380 thousand years after the Big Bang, the universe was filled with neutral hydrogen atoms. Subsequently the first stars were formed and generated ultraviolet rays. The ionization of the neutral hydrogen atoms enabled them to be observed optically.
After the elapse of several hundred million years, the concentrated neutral atomic hydrogen gas gathered together due to gravity. Within the clouds of collected gas, the first generation of stars were then formed and galaxies were created. Among the stars formed were extremely massive stars that emitted strong ultraviolet rays that heated the intergalactic medium and ionized the neutral hydrogen atoms. This marked the end of the Cosmic Dark Ages and this phenomenon is referred to as the reionization of the universe.
This phenomenon of reionization of the universe is thought to have taken place between approximately 300 million years after the Big Bang (corresponding to a red shift of 14) and approximately 950 million years after the Big Bang (a red shift of 6). However, this is not yet known for certain. When reionized hydrogen gas cools, it emits Lyman alpha photon, a characteristic emission line spectrum in the ultraviolet region with a wavelength of 121.6 nanometers. For this reason, galaxies that have just come into existence and that are continually producing large stars ought to emit strong Lyman alpha photons. The further away a galaxy, the faster it is receding; hence, the wavelengths of the light from it are lengthened, and the above-mentioned Lyman alpha photons undergo a Doppler shift (or red shift) toward longer wavelengths. Ultraviolet radiation will turn into optical wavelengths or even infrared radiation.
In relation to this, the Prime Focus Camera can capture a wide field of view in one shot—something other telescopes cannot do—and by setting a special filter that only admits a narrow wavelength band of infrared radiation, we can search for galaxies that are visible with this filter but not visible or very faint with other filters. This is because such young galaxies emit red-shifted Lyman alpha photons that fall within the wavelength band of the filter. The faint light from these candidate galaxies is then spectroscopically observed over a long exposure time by FOCAS, to ascertain whether this specific light is Lyman alpha emission.
Given the wide field of view of the Prime Focus Camera and the ability of FOCAS to perform spectroscopy on a larger number of celestial objects within a practical observing time, the Subaru Telescope is a world leader in research fields that involve the observation of distant Universe. Of the top ten most distant galaxies discovered as of July 2007 based on spectroscopic confirmation, discoveries with the Subaru Telescope account for nine of them. Of these, the very furthest galaxy was detected by us. Up until now, many of these galaxies have been referred to by number; however, since this one happened to be the first galaxy whose existence we could verify, we took the liberty of naming it IOK-1, using the initials of the surnames of the people principally involved in its discovery—Iye (me), Ota, and Kashikawa (laughs).
The top ten most distant galaxies (as of 14 September, 2006). With the exception of No. 8, which was detected by the Keck Telescopes, they were all discovered by the prime-focus camera of the Subaru Telescope and verified using FOCAS.
IOK-1, which was detected by Iye, Ota, and Kashikawa. The most distant celestial object (galaxy).
Up until now, distant galaxies that have been detected have exhibited a red shift of 6.6. However, IOK-1 exhibits a red shift of approximately 7.0, by far the highest recorded so far. The detection of distant galaxies is absorbing enough in itself; however, as it takes a long time for the light from distant galaxies to reach earth, even more exciting is the fact that distant galaxies also represent the face of ancient galaxies. Going back a further 60 million years or so, before the records that have been compiled using Subaru, the records for IOK-1 make it possible to see whether or not cosmic reionization had finished in this period. If we conduct further research into this period, we will be able to see the state of affairs at the time that the first stars came into existence and the universe was being heated up.
The Subaru Telescope came into operation in the year 2000, and there is a lease agreement on the land with the Hawaii State Government that runs until 2033. As the life span of a large-aperture telescope is regarded as between 30 and 50 years, we would like to continue operating it until around 2050, if possible. As for the next ten years, there is an on going program of upgrades in progress, and we are considering replacing the existing Prime Focus Camera with one that has a field of vision that is ten times larger, in order to find many more dark celestial objects. The development of an optical fiber-based spectroscope for multiple celestial objects is also progressing. If all goes according to plan, in ten years time there will be no one who can match us in this field (research into deep space).
Diagram showing the principle of Laser Guide Adaptive Optics. An artificial guide star is generated by illuminating the sodium layer using a laser beam. The light distortion caused by the earth’s atmosphere is measured by analyzing the wavefront from the artificial guide star, and is cancelled out in real time.
I would also like to draw attention to “Laser Guide Star Adaptive Optics,” which I am leading and promoting. This is intended to significantly enhance Subaru’s imaging capabilities. Testing was completed last year and full operation is due to commence next year or so.
An active optics system is installed on the Subaru Telescope. It uses actuators fitted to the underside of the mirror to correct deformation in the primary mirror under its own weight due to the orientation of the telescope. A basic feature of the Subaru Telescope, adaptive optics (AO) represents miniaturized and accelerated active correction technology that compensates for shimmering in the atmosphere. This system features a thin flexible mirror 10 cm in diameter and 0.1 mm thick attached at the telescope’s focal point. Real-time control is performed 2,000 times a second using tiny actuators attached to the underside of the mirror (consisting of 188 bimorph Piezo elements) at an amplitude of approximately one-tenth the wavelength of the light.
Although the light from a celestial object travels tens of billions of light years in a dead straight line, it ends up being distorted when passing through the earth’s atmosphere. For this reason, even on the Subaru Telescope (which is supposed to be capable of capturing an extremely small segment of 0.06 seconds of arc, according to the theoretical diffraction limit for an 8-m aperture telescope), the resolution achieved for observation is 0.6 seconds of arc on average.
With adaptive optics, this disturbance in the atmosphere is cancelled out. Up until now, however, it has only been possible to measure the fluctuation if there is a bright star close to the object being observed. The Laser Guide Star Adaptive Optics that we have constructed involves the combination of adaptive optics with a new device that points a laser beam adjusted to specific wavelengths with a precision of 1 in 10 million at the thick layer of sodium atoms 90-100 km above earth, stimulating the sodium atoms and making them light up (to form the laser guide star). We generate an artificial star in the sky in the direction in which the telescope is pointed, and use the light from it as a basis for correcting the disturbance in the atmosphere. As such, this system can be also used for observation of dark and extremely distant celestial objects that have no bright star in their vicinity.
In the testing phase for this system, the Subaru’s diffraction limit of 0.06 seconds of arc was verified. Fascinating new worlds await us in the future, so please share in our anticipation.
Orion trapezium of 0.06 seconds of arc obtained in test observations of Laser Guide Adaptive Optics on October 9, 2006 (left). On the right is a trapezium (of 0.6 seconds of arc) photographed in 1999, immediately after the completion of the Subaru Telescope.
Conceptual diagram of the next-generation Extremely Large Telescope (ELT). This consists of a 30-m-aperture reflecting telescope comprised of multiple reflecting mirrors. In the basic research for the ELT, Nikon is cooperating in areas such as raw materials research and surface polishing of reflecting mirrors using new materials.
The Subaru Telescope was initially conceived in 1984 and was completed in 2000. Hence, the idea for the telescope did not come to fruition until 16-17 years after it was first seriously considered. Although the Subaru Telescope’s main unit and its observing instruments are functioning extremely well and are yielding abundant scientific results, it is still necessary to consider what will come next. Thus for the past two or three years, we have been seriously considering the possibility of constructing a 30-meter-class Extremely Large Telescope (ELT) to take over the baton. However, it would be impossible for Japan to undertake such large-scale construction and management on its own. For this reason there has recently been much activity to lay the groundwork for international collaboration, and I have been flying around the United States and Europe.
Although we are thinking about a 30-meter telescope to succeed Subaru, it will take at least 10 years to complete. This means that the generation that built the Subaru Telescope will be retiring, and the younger generation that built the observing instruments will be working extensively on the next telescope. We are hoping they will make use of the experience that they acquired building HDS and FOCAS in the construction of the observing instruments for the next-generation telescope. Provided that our Subaru observing instruments upgrade plans go well, the Subaru Telescope will be hugely effective in locating interesting celestial objects in the universe, such as distant galaxies. After that, the celestial objects that it has found will be observed more closely by the 30-m telescope constructed jointly by the USA, Europe, and Japan. This being the case, we think that we can assume leadership of observation and research using the 30-m telescope.
With an altitude of 4,200 meters, an atmospheric pressure of 0.6, and a temperature of 0oC, the summit of Mauna Kea, where the Subaru Telescope is located, might be described as an extremely harsh environment for moving precision optical equipment such as FOCAS. Nobunari Kashikawa, associate professor at the Subaru Telescope, the National Astronomical Observatory, who was in charge of the development of FOCAS, describes the difficulties encountered during development—from determining instrument specifications to carrying out on-site adjustment.
Testing on the Cassegrain simulator at the National Astronomical Observatory in Mitaka.
During the development of FOCAS, the required components were constructed at the Astronomical Observatory’s research center and at the Nikon factory. They were then integrated and mounted on the Cassegrain simulator set up at the National Astronomical Observatory in Mitaka. Together with the Nikon engineers, we then subjected them to repeated tests of all kinds. This process was completed in 1998. We then had to partially dismantle the instruments and transport them to Hawaii. Since this was the first time that we had ever sent observing instruments like FOCAS from Japan overseas, it took longer to get them through customs than we had imagined. I waited a month in Hawaii for them to arrive (laughs).
We then mounted the instrument on the Cassegrain simulator installed at the Hilo base camp facility*2 and repeated the same tests that we had carried out at Mitaka once again. The Cassegrain simulator consists of a platform section that emulates the movement of the Subaru Telescope’s Cassegrain focus, whose inclination changes in accordance with the orientation of the telescope. It also offers power supply and network connectivity.
The reason for this was that once the equipment was on the summit of Mauna Kea, it would be extremely difficult to repair anything, should there be a problem. In addition, once the instrument was mounted on the Subaru, it would occupy the Cassegrain focus, making it impossible to conduct tests on other observing instruments. Our overall policy in the development of the Subaru Telescope was to try and minimize the amount of work that would be performed at the summit.
We again encased the instrument in special air-tight packing designed to reduce vibration, and gingerly carried them to the summit—taking an entire day to traverse a route that normally requires only two hours. This was because the atmospheric pressure differs between the lowlands and the summit, causing the humidity to change significantly as you pass through the clouds on the way up. On this occasion we were also accompanied by Nikon engineers, who provided us with advice on various issues.
FOCAS headed for the Subaru Telescope after the completion of testing at the Hilo base camp facility.
For this reason, we would make a record of our difficulties and the conditions at the summit, and go back down to the Astronomers’ Mid-Level Facility*3 at Hale Pohaku, which is at an altitude of 2,800 meters. Here, we would discuss matters with everyone, look for ways to solve the problems, and try them out the following day in a continuous process that lasted for approximately two weeks. It was also extremely difficult for us to establish the necessary observation parameters that would enable the use of FOCAS by ordinary researchers who knew nothing about it. During the first stage—known as Engineering First Light—we repeatedly conducted ordinary observations, and developed the observation process, while writing the manual and evaluating the capabilities of FOCAS at the same time.
Afterwards, the developers were awarded “guaranteed time,” in which they could use FOCAS freely as a form of compensation for their efforts; however, this time too was devoted to performing adjustments. From 1999 to 2000, we were conducting operational testing of FOCAS round the clock—including Christmas and New Year (laughs). We finally made it to Scientific First Light—the acquisition of observational data—and we were elated when we finally obtained such a beautifully clear image of M82. This made us want to get the adjustments out of the way as soon as possible and conduct our own observations.
After completion of installation at the Cassegrain focus and adjustment. Commemorative picture featuring all members of staff (February 6, 2002).
FOCAS First Light image (observation date: February 2, 2000). A distinct image showing red light, Hα radiation, given off by ionized hydrogen gas spreading out from the central portion of M82 (NGC3034).
With the development of the Subaru Telescope, there was a first time for everything and when a problem occurred you could never predict what the cause might be. All you could do was to investigate patiently and thoroughly, and solve the problem. Right now, I only go to Hawaii for my own observation time—four or five times a year, with a stay of about a week each time. However, when we were starting up, I was probably trekking back and forth to the summit for about four months straight.
FOCAS is now running in an extremely stable fashion. This is probably because, in the first place Nikon’s basic optical design is sound, and because Nikon considered the issue of flexion in the main body in advance. I think that this is probably also why the Multi-Object-Slit mechanism (MOS) and CCD cameras that we built are working. When data that has been obtained from a target celestial object as anticipated appears on the monitors in the observation room inside the Subaru Telescope, I feel an enormous sense of elation.
This website features data in Flash format.
To view this data requires Adobe Flash Player 7 or higher, which can be downloaded for free.
Posted January 2008
