Note: This article is an update of an article that originally appeared in ESA Bulletin No. 67 (August 1991).
The Infrared Space Observatory (ISO) satellite will be the world's first true astronomical observatory in space operating at infrared wavelengths. Astronomers will be able to choose specific targets in the sky and point ISO towards them for up to ten hours at a time to make observations with versatile instruments of unprecedented sensitivity. During its lifetime of 18 months, ISO will be used to observe all classes of astronomical objects ranging from planets and comets in our own solar system, right out to the most distant galaxies.
In the years since observational astronomy finally escaped from its confinement to the narrow visible range of the electromagnetic spectrum accessible from the ground, it has become clear that a full understanding of the properties and physics of astronomical sources can only be obtained by studying them across the widest possible frequency range. A good example is a nova. The behaviour, at visible wavelengths, of a star suddenly brightening dramatically over a period of only hours or days and then fading slowly over hundreds of days has been known for centuries. However, it was only with the advent of X-ray, ultraviolet and infrared observations that the true nature of such a nova outburst, and the underlying physics, began to be understood.
The infrared region of the spectrum is of great scientific interest, not only because it is here that cool objects (10-1000 K) radiate the bulk of their energy, but also because of its rich variety of diagnostic atomic, ionic, molecular and solid-state spectral features. Measurements at these wavelengths permit determination of many physical parameters of astronomical sources, such as energy balance,temperatures, abundances, densities and velocities.
Infrared astronomy and the study of dust are inextricably linked. Dust particles, ranging in size from a few hundred Angstroms to tens of microns, area very common phenomenon throughout the Universe. This dust absorbs visible and ultraviolet light and re-radiates it in the infrared. It is estimated that dust in theinterstellar medium accounts for approximately one third of the total luminosity of our Galaxy.
Detailed photometric and spectroscopic study of this emission by ISO will give astronomers a much clearer understanding of the energy balance of the Galaxy and of the composition of the dust (large molecules, carbon grains, silicate grains, etc.) in different parts of it.
Many astronomical sources are surrounded by clouds of dust and gas. These clouds act as an interstellar 'fog', obscuring the astronomical objects and making it very difficult to observe them with visible light. Owing to its longer wavelength, infrared radiation can pierce these dusty regions and bring astronomers information about the conditions inside. As an example, the centre of our Galaxy is hidden from optical telescopes by thick veils of dust. However, a clear view can be obtained even at a relatively short infrared wavelength of 2 microns and the Galactic Centre can, therefore, be best studied at infrared wavelengths. Figure 1 shows how the Galactic Centre appeared to an earlier infrared survey satellite, IRAS.
Figure 1. IRAS view of the centre of the Milky Way. This is a composite image made from data taken at three wavelengths and presented in false colour. The yellow and green knots and blobs scattered along the band are giant clouds of interstellar gas and dust (called HII regions) heated by nearby stars. Some are warmed by newly-formed stars in the surrounding cloud, and some are heated by nearby massive, hot, blue stars that are tens of thousands of times brighter than our Sun. The red areas represent regions dominated by cold gas and dust. The large yellow bulge near the middle is the centre of our Galaxy. (Courtesy of NASA/JPL)
The scientific potential of infrared astronomy has been amply demonstrated by observations made from both ground-based telescopes and those on high-flying aircraft and balloons. However, Figure 2 shows the two main limitations to these observations. irstly, the Earth's atmosphere is totally opaque at many wavelengths, absorbing all the incoming radiation and thus preventing the astronomer from viewing the celestial object. Work from ground-based telescopes is only possible through a number of narrow spectral 'windows'. Even at altitudes of 30 to 40 km, which are typical for balloon-borne telescopes, the atmosphere is not totally transparent.
The second problem is that the telescope and atmosphere are warm and emit infrared radiation themselves. Astronomical sources a million times fainter must be found against this undesired 'background' (really foreground) emission. This severely limits the sensitivity of ground-based observations.
Thus, for maximum sensitivity and wavelength coverage, it is necessary to cool the telescope and its instruments and to operate them in space. The first major step in this direction was taken with the highly successful Infrared Astronomical Satellite (IRAS), which surveyed nearly all the sky in four broad photometric infrared bands. Among the results of the IRAS mission (US/NL/UK) is a catalogue of over 250 000 sources. ISO will build on the results of IRAS by making detailed observations of selected sources. Compared to IRAS, ISO will have a longer operational life-time, wider wavelength coverage, better angular resolution, more sophisticated instru-ments and, through a combination of detector improvements and longer integration times, a sensitivity gain of up to several orders of magnitude.
Figure 2. Upper panel: Transmission of the terrestrial atmosphere as a function of wavelength. Note that, from the ground, observations are only possible through some 'windows', shown in black
Lower panel: Relative flux from a 300 K black body as a function of wavelength, showing that the thermal emission from the warm (approx. 300 K) telescope optics and atmosphere peaks around the 10-micron wavelength. This emission hampers observations and is at a maximum where the Earth's atmosphere is relatively transparent
ISO is a true astronomical observatory. It has a highly versatile and sensitive set of scientific instruments, capable of undertaking a wide range of scientific tasks. Time on this observing facility is available to all European, Japanese and US astronomers. The overall ISO system includes not only the scientific instruments and the spacecraft in orbit, but also its control centre on the ground.
Four instruments make up the ISO scientific payload: an imaging photopolarimeter (ISOPHOT), a camera (ISOCAM) with polarimetric capabilities, a short-wavelength spectrometer (SWS), and a long-wavelength spectrometer (LWS). These instruments were built by international consortia of scientific institutes for delivery to ESA. The technical aspects of the instruments are discussed in 'The ISO Scientific Instruments - Technical Highlights' in this issue; their main features are summarised in Table 1, while an overview of their scientific capabilities is given in Figures 3 and 4. In summary, observers are provided with a range of photometric, polarimetric, spectroscopic and imaging capabilities across the entire ISO wavelength range. These unique instruments will reach out to new frontiers, probing fainter sources with higher spectral and spatial resolution than ever before at these wavelengths inaccessible from the ground.
Table 1. Main features of ISO's scientific instruments Instruments Wavelenght (Principal investigator) Main function (µm)
ISOCAM Camera and polarimetry 2.5 - 17 (C. Cesarsky, CEN-Saclay, F) ISOPHOT Imaging photo-polarimeter 2.5 - 240 (D. Lemke, MPI für Astronomie Heidelberg, D) SWS Short-wavelength 2.4 - 45 (Th. de Graauw, spectrometer Groningen, NL) LWS Long-wavelength 43 - 198 (P. Clegg, spectrometer Qeen Mary & West Field College, London)
Figure 4. Imaging capabilities of the ISO scientific instruments. Each of the four instruments receives a circular 3 arcmin field of view (drawn to scale in the central part of the figure). The outer ring shows, in expanded scale, more details of the detector fields of view as projected onto the sky
In order to prevent the sensitivity of the scientific instruments from being degraded by their own thermal emission and that from the telescope, all parts of ISO 'seen' by the infrared instruments must be cooled to only a few degrees above absolute zero (-273 degC). Thus, the ISO satellite is, essentially, a huge Thermos flask designed to provide the extremely low temperatures necessary. ISO consists of a cryostat containing, at launch, over 2000 litres of liquid helium, and a cryogenically cooled telescope with an aperture of 60 cm. The telescope can be pointed anywhere on the sky to an accuracy of a few seconds of arc for a period of up to 10 h. The in-orbit lifetime of the satellite is limited by evaporation of the liquid-helium cooling fluid, but will be at least 18 months. The spacecraft is described in more detail in another article in this issue, The ISO Spacecraft , and the cryogenic system was presented by Davidson et al in ESA Bulletin No. 57.
An Ariane-4 launcher will place ISO into a highly eccentric orbit with an apogee of 70 600 km, a perigee of 1000 km and a period of 24 h. In this orbit, ISO will spend about 16 h per day outside the Earth's radiation belts. The infrared detectors in the scientific instruments are made from small pieces of silicon and germanium. If the energetic particles in the radiation belts (mainly electron and protons) hit these detectors, they release a large number of electrons, which prevents the ISO instruments from operating at full sensitivity.
The ISO operations will be carried out by a team of scientists and engineers located at ESA's satellite tracking station in Villa-franca, Spain. However, to achieve continuous communications with ISO during all its scientifically-useful observing time, a second ground station is needed. It is at Goldstone, USA, and is provided for ISO as an international collaboration with NASA and ISAS. Since ISO's in-orbit lifetime is strictly limited by the evaporation of its liquid helium, the efficiency of the operations is even more important than usual.
ISO will be used to make observations of specific objects in the sky that have been selected by individual astronomers via a process of proposal submission and approval. (This process is described in more detail in 'Using ISO' in this issue.) The detailed observing schedule will be planned on an orbit-by-orbit basis a few days in advance. During scientific operations, ISO will always be in real-time contact with the ground Control Centre. However, real-time modifications to the scientific observing programme will be minimised in order to maximise overall efficiency.
The downlinked data will be quality-checked upon receipt. They will then be subjected to sophisticated pre-processing before being sent on CD-ROM to the commissioning astronomer's institute for scientific analysis and interpretation. The results will also be placed in an archive for later use by the astronomical community.
Figure 5 gives a pictorial representation of the ISO operations.
ISO offers high sensitivity and sophisticated observing facilities for a difficult spectral region, and its scientific programme touches upon virtually every field of astronomy, ranging from solar system studies to cosmology. Some of the possible scientific highlights are summarised here.
Planets and their satellites
Like the Earth, most planets have atmospheres, composed mainly of molecules of various gases. ISO will be used to investigate the chemical composition and the physical nature of the atmospheres of the giant planets, together with Titan and Mars. A detailed inventory of the species present will be established, allowing for a better under-standing of the planets chemistry.
Titan is the only satellite in the solar system to possess a thick atmosphere. It is thought to be similar to the atmosphere originally possessed by the Earth. Studies of Titan's atmosphere are expected to lead to a better understanding of how the Earth's atmosphere evolved. Detailed studies of Mars's surface temperature and emissivity properties, their temporal variations and their relation with atmospheric dynamics (e.g. dust storms) will also be possible with ISO.
Comets are believed to retain, in the form of ice and trapped dust, the original content of the primordial solar nebula, from which our solar system condensed. Therefore, their study provides a unique probe into the history of our solar system and its relation to the interstellar medium. With ISO, it will be possible to detect comets at large heliocentric distances (5 AU), to study the onset of activity (emission of gas and dust) when a comet approaches the Sun, in particular to study the activity, evolution and composition of the coma. Cometary dust and nucleus have a low temperature and albedo, and are thus best detectable in the infrared. The spectral, spatial and sensitivity capabilities of ISO will allow a thorough comparison of the general interplanetary dust with the properties of dust close to its probable sources, comets (cometary trails) and asteroids (asteroidal bands) (Fig. 6).
Figure 6. An IRAS image (wavelength 60 microns) of the ecliptic plane, showing the central asteroid dust band, consisting of asteroid collision debris (wide band cutting across centre of picture). (Courtesy of M. Sykes, Univ. of Arizona)
The space between the stars is not empty (Fig. 7). It is a very active and violent space, containing objects such as gaseous nebulae, supernova remnants, dark molecular clouds, dust, and high-velocity winds from young stars. The material of the interstellar medium has an extremely wide variety of temperatures and densities.
Figure 7. An infrared view from IRAS of the well-known Ursa Major (Great Bear or Plough constellation). The familiar stars, which can be seen with the naked eye, have been circled for recognition. Note the gas and dust between and around the stars radiating at infrared wavelengths. (Courtesy of NASA/JPL)
IRAS revealed a new component of the interstellar medium - extended, fuzzy clouds which often have filamentary structures. These clouds range in angular size from tens of degrees down to a few arc minutes (the limiting spatial resolution of IRAS) and, because of their appearance, they have been named 'infrared cirrus'. ISO will explore the nature and composition of these puzzling clouds.
The processes by which stars form are not yet well understood. Much of the action is hidden by dust and more infrared observations are needed. Under the right conditions, some dense parts of molecular clouds can start to collapse upon themselves. Initially, these so-called 'protostars' radiate by virtue of the gravitational energy of the in-falling material and remain cold compared to the Sun. Eventually, their temperature rises sufficiently for nuclear reactions to start. When the 'burning' of hydrogen to form helium is underway, the protostar has become a star.
Stars are formed with a wide range of initial masses; a well-known example of a region of massive star formation is in the constellation Orion (Fig. 8). Among the many open questions on star formation to be addressed by ISO observations are: What triggers the collapse process? Does the accretion always involve a disk? What determines the relative numbers of large and small stars in the resulting cluster? What is the role of the high-velocity (several hundred km/s) mass outflows that are seen from young stars? and What are the properties of the embedded young stellar objects?
Figure 8. IRAS false-colour map of the sky around the constellation Orion. Well-known regions of star formation are apparent, such as the Orion Molecular Cloud (large feature dominating lower right of picture), located in and surrounding the sword of Orion. The large ring in the upper right of the image is a shell of gas swept up by the expanding gases around a young star. The bright region left of centre is the Rossette Nebula in Monceros. (Courtesy of NASA/JPL)
The interstellar medium, containing atoms like hydrogen, oxygen and carbon and molecules like carbon monoxide or water vapour, acts as a chemical factory. Atoms and molecules can collide and they can absorb radiation from nearby stars. By these two processes, other larger molecules may be formed. The physical conditions in interstellar space under which the formation of molecules takes place are extremely difficult to simulate in the laboratory. The ISO spectrometers will reveal the chemical processes in molecular clouds or thick envelopes around young stars.
Stars and stellar physics
Stars are dense gaseous spheres which, for most of their lives, burn or, strictly speaking, fuse hydrogen to form helium in their interior, like the Sun. As stars get older, other nuclear reactions start and, eventually, the star's life ends in a way that depends on its mass. Stars have been extensively observed at many wavelengths, but much important information on their structure and their evolution can only be extracted from infrared observations.
The nearby star Vega, or Alpha Lyrae, is the fifth brightest star at visible wavelengths and is still in its hydrogen-burning phase. It had been extensively observed at many wavelengths and its properties were thought to be well understood. It was, thus, a great surprise when infrared observations by IRAS showed brighter-than-expected emission at many wavelengths longer than 25 microns. These data indicate the presence of a disk of cool (around 85 K) material in orbit around the star. This disk may well represent an early stage in the condensation of a planetary system.
A number of other stars also have similar 'infrared excesses'. In one of these cases, Beta Pictoris, observations in the visible have actually revealed a thin disk of gas and dust around the star (Fig. 9).
Imaging, photometry and spectroscopy between 3 and 240 microns of these stars will also give us a deeper insight into the formation processes of our solar system. ISO will also be used to investigate how widespread the phenomenon of matter in orbit around these types of stars is.
Figure 9. Dust around a star. The image of Beta Pictoris, with the star itself masked, shows the presence of a disk (seen edge-on) of dust similar to that from which the Earth and other planets supposedly formed, in the vicinity of the newborn Sun. The disk was discovered by IRAS. (Courtesy of B. Smith, Arizona, and R. Terrile, JPL)
Stars contain enough hydrogen for the fusion to last for a long time, but not forever. The rate at which a star consumes its fuel depends on its mass; the more massive a star, the quicker it evolves. After around five to ten billion years, stars like our Sun evolve to become so-called 'red giants', i.e. very large cool stars. More massive stars, such as those with around 40 times the mass of the Sun, race towards the red-giant phase in only a few million years. During this giant phase, the stars lose a significant amount of their mass via outflows and winds from their atmospheres. A circumstellar envelope is thus built up and sometimes these envelopes are so massive that the stars can no longer be seen in the optical. At the end of this phase, stars can evolve into planetary nebulae (Fig. 10), small hot stars, called white dwarfs, surrounded by their expelled material, which is ionised by the ultraviolet radiation from the stars themselves.
With ISO, it will be possible to study those stars that are deeply embedded in their circumstellar envelopes. These are at the very end of the phase as a giant, and one open question is how the star evolves during the very short transition phase from a red giant to a planetary nebula and a white dwarf. During the phase of mass loss, the star returns its matter - now processed to include heavy elements - to the interstellar medium. This enriched inter-stellar medium is the source material for the next generation of stars, and its chemical composition is therefore of great interest. This will be deduced from measurements by the ISO spectrometers of the atomic and molecular spectra of planetary nebulae.
A massive star that fails to lose enough mass during its evolution is, then, doomed to end its life in a huge explosion, a supernova, such as that seen in our companion Galaxy, the Large Magellanic Cloud in 1987 (Fig. 11). This explosive event also returns the material from a dying star to the reservoir from which new stars may form. ISO will study the 'leftovers' (called 'supernova remnants') of such events, which are the source of very heavy elements, like iron.
Figure 11. Two colour photos showing the sudden appearance of the bright supernova 1987A (above the main body of the galaxy). The left-hand picture was taken before the supernova exploded and the right-hand one afterwards. (Courtesy of ESO)
Other galaxies, far distant from our own Milky Way Galaxy, have always attracted much observational attention. They have a variety of morphologies, many having spiral arms, interstellar matter and a core region or nucleus, thus reflecting the structure of the Milky Way. Study of these galaxies gives a 'bird's eye view' of processes occurring in the Galaxy, but difficult for us to see. Many galaxies are so far away from the Earth, and their light takes so long to reach us, that observing them is like looking back in time, thereby allowing an examination of the evolution of the Universe. The infrared properties of galaxies are extremely diverse; for example, far-infrared luminosities have been found that span a range of seven orders of magnitude.
There are many questions in extragalactic astronomy needing answers: What are the mechanisms that trigger and maintain the formation of stars in galaxies? Why are some galaxies producing large numbers of new stars in hugely energetic bursts? What is the energy source at the centres of the most luminous galaxies making them orders of magnitude more energetic than their quieter neighbours? These questions are central to understanding the processes by which galaxies evolved from their original formation to give us the Universe we see today. In order to answer these questions, it is necessary to discover the physical conditions prevailing in the diverse and often exotic sources that populate the Universe.
Using ISO, astronomers will seek to understand the properties of star-forming regions in nearby, normal galaxies (Fig. 12) by studying, spectroscopically and photometrically, the properties and spatial distribution of the dust produced there, the kinds of organic compounds that form in the interstellar medium, the energetics of the gas, and the mass distribution of stars produced there.
Figure 12. A 'wide-angle' view of part of the infrared sky showing the galaxies M31 (Andromeda nebula, top left) and M33 (lower right) on a background of varying infrared emission (Courtesy of NASA/JPL)
The results of these studies will be compared with observations of the same entities in radically different environments such as the nuclei of active galaxies, completely obscured by dust absorption at visible wavelengths, or at the heart of colliding galaxy systems, powerfully luminous at far-infrared wavelengths. With these observations it may be determined whether some galaxies with extremely luminous nuclei ('active' galaxies) are, in fact, the final stage in the development of galaxy mergers. In this scenario, two galaxies collide, precipitating a huge burst of star formation throughout their interstellar material. This would give rise to a far-infrared, ultra-luminous galaxy which finally decays to become an active galaxy with a massive black hole at its nucleus: a Seyfert galaxy or a quasar.
Since ISO's instruments can see emission from cold (a few Kelvin to a few hundred Kelvin) dark matter (i.e. material not luminous in the visible), it may detect the elusive population of low-mass stars thought to condense out of the streams of gas that flow from intergalactic space onto many of the large elliptical galaxies at the centres of galaxy clusters. These 'cooling flows' of gas, inferred from X-ray observations, produce no corresponding population of stars detectable at visible wavelengths.
ISO can address a number of questions of great cosmological significance. A particularly vital question concerns the total mass in the Universe. If this is greater than a certain amount, then gravitational force will eventually stop the expansion of the Universe and make it collapse into itself again. If the Universe is less massive than this 'closure' mass, it will go on expanding forever. The density of directly detected matter (self-luminous, light-reflecting, or light-obstructing) currently accounted for in the Universe is at most about 20% of the closure density. However, mass could be hidden in dark forms, invisible at optical wavelengths, but radiating in the infrared region.
One possibility is that some of this missing mass is hidden in the form of objects called brown dwarfs (Fig. 13). These are failed stars, i.e. bodies formed out of the interstellar material, but which were not massive enough to support nuclear burning in their cores. It has been suggested that such objects might constitute the unseen halos of galaxies, postulated in order to account for the detailed orbits of material around galaxy nuclei. It is hoped that the camera (ISOCAM) and the photopolarimeter (ISOPHOT) will be able to unambiguously detect, and confirm the existence of, such objects for the first time.
Figure 13. Artist's impression of a brown dwarf, silhouetted against the backdrop of the immensely rich star fields of the Milky Way. A brown dwarf is an object that started to collapse to become a star but was not massive enough to be able to initiate nuclear reactions. (Courtesy of NASA)
It is planned that ISOCAM and ISOPHOT will both perform very long observations intended to detect sources out to high red-shifts. The relative proportions of blue galaxies, merging galaxies, active galaxies and more typical galaxies found in such deep-source counts is an indicator of the mechanisms through which galaxies originally formed. Did they form at about the same time in a single great burst, or have they formed by a process of hierarchical merging of galaxies, so that they grow, and mergers become less common, as time goes on?
During its lifetime, ISO will offer astronomers a unique opportunity to study the Universe at the relatively unexplored infrared wavelengths. ISO's legacy to the future will be the database of its observational results, which will be used by astronomers long after the in-orbit mission has been completed. The science of ISO will build not just upon the results of the IRAS mission, but also on those from ground-based optical, infrared, submillimetre and radio telescopes. Observations with ISO will have a significant impact on all areas of astronomy. However, the most exciting aspect of the mission is that it is a voyage into largely uncharted waters, and no-one knows what will really be discovered. Hopefully, nature has a few surprises in store for us once again!
The scientific highlights described in this article are based, in large part, on the work of the Principal Investigators of the ISO instruments (Catherine Cesarsky for ISOCAM; Peter Clegg for LWS; Thijs de Graauw for SWS; and Dietrich Lemke for ISOPHOT) and their teams of astronomers, in defining their guaranteed-time observational programmes for the mission.
March 1979 Proposal to ESA for ISO 1979 Assessment Study 1980 Pre-Phase-A Study 1981-1982 Phase-A Study March 1983 Selection of ISO for inclusion in ESA Scientific Programme June 1985 Selection of Scientific Instruments Dec.1986 Start of Phase-B (Definition) March 1988 Start of Phase-C/D (Main development) April 1994 Release of 'Call for Observing Proposals' November 1995 Launch