Radio astronomy is the study of the universe through observations of the radio waves emitted by cosmic objects. Everything in the universe radiates radio waves, and modern radio telescopes are capable of detecting these waves from almost all known objects. Because the physical processes giving rise to radio emission are sometimes different from or more powerful than the processes giving rise to other types of radiations, objects can often be detected more easily by radio observations. In addition, important aspects of objects, such as the strength of magnetic fields, can often be determined only by radio observations. As a result, radio astronomy complements optical, X-ray, gamma-ray, infrared, and ultraviolet astronomy. All of these studies combine to reveal the true nature of celestial objects. (See \Tastronomy and astrophysics\t; \Tgamma-ray astronomy\t; \Tinfrared astronomy\t; \Tultraviolet astronomy\t; X-RAY \Tastronomy\t.) Radio astronomy was born in 1932 with the discovery by Karl \TJansky\t at the Bell Telephone Laboratories in New Jersey that radio waves were coming from the sky. His discovery was followed by the work of Grote Reber, a radio engineer at Wheaton, Ill., who in 1937 constructed a steerable parabola radio telescope and used it to map the distribution of radio emission in the sky. This study was followed by work in many countries during and immediately after World War II; radar technology, originally developed for military purposes, was used for pioneering radio astronomy observations as well as for making radar observations of neighboring planets (see \Tradar astronomy\t). The radio emission from the Sun was discovered in this way, and radio telescopes in Australia established the existence of discrete sources of radio emission. These discoveries led to the construction throughout the world of ever bigger and more sensitive radio telescopes during the 1950s and 1960s. Particularly active countries were England, Australia, the Netherlands, and the United States. By the mid-1960s radio astronomy had become the cutting edge of modern astronomy, leading the way to the majority of major new discoveries such as the existence of \Tradio galaxies\t, \Lquasar\ls, \Lpulsar\ls, \Tbackground radiation\t from the primordial fireball, and a host of complex molecules in interstellar space (see \Tastrochemistry\t). CAUSES OF RADIO EMISSION As late as 1950 some astronomers thought that radio observations would play no important role in the study of the universe. They believed that the only significant source of radio waves was thermal radiation caused by the heat of a body. Calculations indicated that the thermal radiation at radio wavelengths would be so faint that no useful radio measurements could be made of anything except perhaps the Sun and planets. The radio intensities observed by Jansky and Reber were, however, approximately 10 million million times brighter than might have been expected from the thermal emission of the stars and galaxies. This result implied that far more powerful mechanisms for the production of radio waves existed in the universe. Nonthermal Radiation The only important nonthermal mechanism of radio emission, and the one that almost always produces the observed radiation, is known as \Tsynchrotron radiation\t. This is radio emission from very energetic nuclear particles orbiting in a magnetic field. Scientists believe that these are, in every case, electrons orbiting in the magnetic field of, for example, a galaxy. To produce synchrotron radiation these electrons must have a total energy much higher than the energy associated with their mass (m), which is mcc, or 0.5 million electron volts. In space, electrons have energies as great as 100 billion billion electron volts, which means that magnetic fields must be accelerating them far beyond the energies produced in manufactured nuclear accelerators. In the interstellar magnetic field, which may be only one-millionth as strong as the magnetic field of the Earth, the electrons may complete only a few orbits a second. With their very high energies, however, the synchrotron mechanism causes the electrons to radiate primarily at radio frequencies and with very great intensity. Because of the large release of energy, the overall radiation from celestial objects is likely to be relatively great, allowing astronomers to detect and study many of the objects and to observe them at very great distances. Thus radio astronomy has given the astronomer a means to trace the history of the universe, even perhaps back to the creation of the universe. Thermal Radiation Although the nonthermal radio emission and the objects associated with it have been of the greatest interest, thermal radiation has turned out to be a valuable source of information also. With it astronomers have studied the gas clouds of our galaxy and other galaxies, the emission nebulae, and the planets. RADIO SOURCES WITHIN THE GALAXY When the sky is scanned with a radio telescope, a very bright band of radio emission is found that coincides with the Milky Way. The brightness of this band relative to the other cosmic sources of radio emission is much greater than the relative brightness of the Milky Way as compared to the ordinary stars as seen with the unaided eye. This radiation from the disk of our galaxy is, at most radio frequencies, synchrotron radiation from cosmic-ray electrons spiraling in the magnetic fields of our galaxy (see \Tcosmic rays\t). In addition to the bright radiation of the Milky Way, the sky is filled with distinct sources of radio emission. Some of these come from solar system objects, while others arise in more distant parts of the galaxy or outside of the galaxy. None of the bright radio sources is associated with any of the bright, well-known optical stars; instead, these sources of radio emission are associated with objects that in all cases are rather faint optically but radiate large amounts of radio energy due to synchrotron radiation or very high temperature. Aside from the solar system, the most prominent sources of radio emission within the galaxy are supernova remnants, pulsars, most ionized emission nebulae, and interstellar atoms and molecules (see \Tinterstellar matter\t). The Solar System The Sun is an interesting and complex source of radio waves. It radiates a steady level of thermal radio emission because of its high temperature. When observed at long radio wavelengths, however, the radio emission observed is very intense and is typical of a hot object whose temperature is 1,000,000 K rather than the 5,800 K temperature of the solar surface. This high-temperature radiation comes from the solar \Tcorona\t, which becomes the visible "surface" of the Sun when observed at long radio wavelengths. In addition to this steady emission, the Sun exhibits at least six other kinds of complicated, time-variable radio emissions. All of these seem to be associated with \Tsunspot\t activity and the \Lsolar flare\ls that accompany this activity. One type of radio emission varies slowly as the Sun rotates. Another is noise storms, which last hours to days. Also occurring are at least four different types of solar radio bursts--large increases in radio emission from a small region of the Sun, usually near a sunspot. One type of burst may last only a few seconds, while another type may persist for minutes or hours. Often the radio emission making up a burst occurs only at a very limited range of frequencies, a range that changes as time goes on, usually to lower frequencies. This occurrence provides evidence that the burst is caused by material ejected from the Sun that moves up in the solar atmosphere to less dense regions where the typical frequencies for radio emission are lower. Radio observations of Venus have shown that the temperature of its surface is much higher than once imagined, about 480 deg C (900 deg F), a result since confirmed by spacecraft. This high temperature is maintained by a \Tgreenhouse effect\t caused primarily by the dense carbon dioxide atmosphere and the clouds of Venus. Radio observations of Mercury have shown that temperature on its surface rises as high as 425 deg C (800 deg F). Observations of the Moon have shown that the mean temperature of the lunar surface is only slightly less than that of the Earth, but that the temperature of the surface plunges to about -100 deg C (-150 deg F) at night. Similarly, the temperature on the Martian surface, which barely reaches the freezing point of water during the day, approaches -100 deg C (-150 deg F) at night. The biggest surprises in solar-system research have been provided by Jupiter, which exhibits three different types of radio emission. First, radio emissions are caused by thermal emission from the body of the planet. Higher and higher temperatures are observed as observations are made at longer and longer wavelengths for two reasons: because of the greenhouse effect and because the radio telescope looks deeper into the Jovian atmosphere at the longer wavelengths, a consequence of the fact that the atmosphere is made partially opaque by ammonia gas, with the amount of obscuration less at long wavelengths. Indeed, temperatures little different from terrestrial temperatures are found deep in the Jovian atmosphere. A second type of radio emission, called decimeter emission, is synchrotron emission from a vast system of radiation belts that are held within the magnetic field of the planet. These belts are like the Van Allen radiation belts of the Earth but contain many more energetic particles whose origins are not understood. The third type of radio emission, called decameter emission, is a sporadic, very intense radio emission that is observed only at low frequencies of about 20 MHz. It consists of groups of short bursts of radio emission usually lasting only fractions of a second, with a group of bursts lasting sometimes tens of minutes. The probability of such emission is very dependent on the position of the satellite Io, suggesting that Io causes energetic electrons from the radiation belts of Jupiter to travel down the magnetic lines of force to the ionosphere of the planet, where they radiate the decameter emission. Although Saturn is similar to Jupiter it possesses a much weaker radiation belt, because the rings of the planet destroy energetic electrons. Relatively high temperatures, approaching terrestrial temperatures, are observed at long wavelengths. Radio emissions of Uranus and Neptune indicate lower temperatures; the presence of a greenhouse effect and almost terrestrial temperatures low in the atmosphere are indicated here also. The asteroids, comets, and Pluto emit too little radio emission for detection by contemporary radio telescopes, except for faint radio emission from several comet molecules. Supernova Remnants and Pulsars Supernova remnants are the clouds of gas that have been expelled in the violent nuclear explosion of a star known as a \Tsupernova\t. These clouds, which often appear as hollow spherical shells, contain large quantities of electrons of relativistic energy, that is, electrons moving at speeds approaching the speed of light. Radiating through the synchrotron process, the electrons were evidently created in the supernova explosion; in some cases, however, this explosion has left behind a spinning \Tneutron star\t, which is both a pulsar and a continuous producer of new relativistic electrons and cosmic rays. The pulsars are radio sources in the Milky Way that are very unusual because, rather than producing a continuous radio intensity at the Earth, they produce short, regularly spaced bursts of radio emission. These bursts typically last about one-twentieth of a second and occur at intervals of about one second. The time between pulses is extremely regular, except that a slow increase in the period between pulses is detected if a pulsar is studied for many months. Astronomers have deduced from such data that the objects are spinning neutron stars, the products of supernova explosions. They are made of bulk nuclear matter, primarily neutrons, and their density is about 10 to the power of 15 times the density of water. They have very strong magnetic fields, about 10 to the power of 12 times stronger than the Earth's field. This field, spinning with the neutron star, acts as a super powerful electric generator capable of accelerating electrically charged particles to relativistic energies that exceed the highest energies produced in terrestrial nuclear accelerators. The particles accelerated in this way radiate synchrotron radiation in well-defined beams that spin with the pulsar. As these beams sweep across the Earth, the pulse of the pulsar radiation can be observed. The most important example of a pulsar is in the \TCrab nebula\t, the remains of a star that exploded on July 4, 1054, and was so bright that it was visible for three weeks in the daytime and for almost two years at night. This remnant is about 6,000 light-years from the Earth, and is still expanding outward at a speed of about 1,000 km/sec (6 mi/sec). At most frequencies it is the third brightest cosmic radio source and is a shell filled with relativistic electrons that radiate synchrotron radiation at X-ray and optical frequencies as well as radio. The total power radiated is equal to about 100,000 times the total power radiated by the Sun. The source of this power is particles accelerated and expelled from a pulsar that is in the center of the object and that spins about 30 times per second. Ionized Emission Nebulae The other bright sources of radiation in the Milky Way are the ionized emission \Lnebula\le. In these objects, a very hot star has produced ultraviolet radiation that has ionized and heated the interstellar gas around it. This hot, charged gas is an excellent emitter of radio waves produced by thermal processes. These sources of radio emission--one of the brightest is the Great Nebula in Orion--are more prominent at the higher radio frequencies because the spectrum of synchrotron radiation is most intense at low frequencies, even though thermal emission has about the same intensity over a broad frequency range. Stars Several dozen normal stars have been detected as faint radio sources in our galaxy. The detectable radiation appears to come from hot shells or coronae, like that of the Sun, surrounding these stars. Interstellar Atoms and Molecules A very important feature of the galactic radio emission is the strong radiation on specific wavelengths, or so-called spectral lines, that are emitted by atoms and molecules in the interstellar gas. The most prominent of these is the radiation of neutral atomic hydrogen, the most abundant element in the universe, at the wavelength of 21 cm. The atoms in this case and others radiate at a specific frequency but may arrive at Earth at a different frequency due to the Doppler effect associated with the motion of the atoms toward or away from the Earth. By studying this Doppler shift in the frequencies of the observed spectral lines, the astronomer can deduce the temperatures and motions of the gas clouds that the radiation comes from. In this way the velocities and arrangements of gas clouds in our galaxy and others have been determined. In addition to atomic hydrogen, spectral lines have been discovered, surprisingly, from more than 40 different molecules. Those found so far are primarily organic molecules. The largest contain as many as 13 atoms and have molecular weights of more than 100. Carbon monoxide is found to be very abundant in space, as is formaldehyde. The molecules that are most effective in producing molecules important to biology in laboratory experiments are also abundant in space. This discovery suggests that interstellar chemistry perhaps played a role in the development of life on Earth and elsewhere, or else that the process was mimicked here. Extraterrestrial Civilizations At several observatories in the United States, Canada, and the Soviet Union, astronomers are searching for radio signals from other civilizations. Much more activity is called for if there is to be a high probability of success in such searches (see \Tlife, extraterrestrial\t). EXTRAGALACTIC RADIO SOURCES Most of the bright objects in the radio sky are outside our galaxy. At least two distinct classes of objects exist--radio galaxies and quasars--but radio surveys of the sky also reveal a variety of less classifiable objects. Radio Galaxies Although all galaxies, including the Milky Way, emit some radio waves because they contain energetic electrons and magnetic field, some galaxies emit from a thousand to ten million times more radio energy than normal galaxies. These radio galaxies are very often peculiar in their optical appearance. A common type has a bright central region; other cases are elliptical galaxies with dust clouds. In a majority of cases the intense radio emission comes not from the optical body of the galaxy but from two very large regions placed symmetrically about the center of the galaxy, several galactic diameters away from the center. An important example is the radio galaxy NGC 5128, also known as Centaurus A, which has two radio lobes extending 10 deg across the sky, or 20 times the diameter of the full moon. These lobes span a distance of 3 million light-years, or 30 galactic diameters. In other cases a radio galaxy may have a very intense region of radio emission near the center of the galaxy, sometimes accompanied by a halo of bright radio emission extending throughout the galaxy. Many hundreds of such galaxies are known. The radio emission from radio galaxies is very highly polarized, indicating that its origin is the radio radiation of very energetic electrons moving at nearly the speed of light and spiraling in the weak magnetic field of the galaxy. These energetic electrons are the result of some very violent event, still not understood, in which an amount of energy equivalent to the total annihilation of up to ten million stars is released. Calculations indicate that the typical radio galaxy will be a brilliant radio emitter for anywhere from 100 million to 1 billion years. Quasars Quasars are far more extreme examples of radio galaxies and are the brightest objects in the universe. Even if situated near the bounds of the observable universe, they are easily detected by small radio telescopes. Their radio emission is typically one million to one hundred million times greater than that of a normal galaxy, and they are as bright as or brighter than the brightest radio galaxies. In every case a quasar appears to be a galaxy with a very bright small region in its center, as seen optically. Because they create optical images that are indistinguishable from stars, they are called "quasi-stellar radio sources," or "quasars" for short. Spectral analysis of the objects, however, shows that they are indeed distant galaxies, in many cases receding from the Earth with a velocity that is a large fraction of the velocity of light. Their radio and optical emission changes with time, sometimes increasing or decreasing substantially in only a few months. This rapid change indicates that the main source of their energy is at most only a few light-months in size. In about ten cases astronomers have found that small regions of radio emission are moving outwards from the nucleus of these galaxies with enormous speeds; the deduced velocities actually may be as much as ten times the velocity of light. It is believed that the true velocities of the emitting regions do not exceed light velocity; if they do exceed light-speed, some new and remarkable laws of physics are required. Instead, astronomers believe that the superlight velocities are an illusion created by projection effects. Otherwise the radio appearance of quasars is much like that of radio galaxies. Accounting for the source of the energy in quasars is one of the most interesting of astronomical problems. Perhaps the energy is the result of a chain of explosions of supernovas or the result of a tremendous number of collisions between stars in a very dense cluster of stars at the center of the galaxy. However, the favored theory at present is that the energy is released from matter falling into a massive \Tblack hole\t at the center of the galaxies. BL Lacertae Objects A class of objects very similar to quasars are the objects known as \TBL Lacertae objects\t, named after the first of these to be recognized, the variable "star" BL Lacertae. Long known as a starlike object that varied in its light, BL Lacertae is actually a galaxy. Many such objects are now known. They are distinguished rather arbitrarily from quasars on the basis that their light and radio variations occur in much shorter times--sometimes in days--than the variations in quasars, and their optical spectra do not show the atomic emission spectral lines characteristic of quasars. They are probably a special variety of quasars. Background Radiation A profound aspect of the radio sky is the uniform glow of radio emission found in all parts of the sky. Careful study shows that it has the same spectra as a thermally radiating body whose temperature is only 2.7 K. Surprisingly, such a spectrum is the same as the spectrum that has been predicted theoretically (see \Tblackbody radiation\t) to result from the big-bang cosmology (see \Tbig bang theory\t). In this cosmology, the universe in its early phases consisted of a primordial fireball, an expanding sphere of high-energy photons that eventually was largely converted into matter. However, as astronomers look back in time with a radio telescope, they observe this fireball receding at nearly the speed of light. This recession, with its accompanying red shift, causes the apparent temperature of the fireball, 2.7 K, to be far less than what would have been measured at the fireball. All tests of this blackbody radiation so far have been consistent with the idea that in this radiation astronomers are observing the earliest periods in the development of the universe. This interpretation is supported by studies of the numbers of radio sources of different brightness, which suggest that the density of objects was much higher long ago, consistent with the big bang theory. RADIO TELESCOPES The first radio telescope, built by Karl Jansky in 1929, was originally intended for the purpose of studying the causes of short-wave interference. It was an ensemble of simple dipole antennas that could be rotated on a track. Most modern instruments are large parabolic reflector radio antennas that focus the radio emission from a small region of the sky to a focal point, where a small antenna captures the radio energy and delivers it to a very sensitive radio receiver. Other types of radio telescopes include the Mills Cross and the Very Large Array (see \TNational Radio Astronomy Observatory\t). Resolution Sensitivity The resolution of a radio telescope is the size of the region in the sky from which the telescope collects radiation. Sometimes called the beamwidth, the area is usually a circular region whose angular size is approximately 57 deg times the ratio of the observed wavelength to the diameter of the telescope. The largest single paraboloidal telescopes have a beamwidth, or resolution, of about one arc-minute--about the same as that of the human eye. Much greater resolution can be obtained with interferometers (see \Tinterferometer\t), in which two or more radio telescopes are connected together to simulate the performance of a much larger telescope. In this case the resolution is 57 deg times the ratio of the wavelength to the maximum separation of the telescopes used in the interferometer. This separation may be thousands of meters or even an intercontinental distance, leading to resolutions as small as a few ten-thousandths of an arc-second. The sensitivity of a radio telescope refers to the faintness of the signals that can be detected. It depends on both the energy-collecting area of the telescope and on the radio noise added to the incoming radio signals, primarily by the radio receiver but also to a lesser extent by the antenna itself. Special circuits have been constructed to minimize the effects of this noise, but physical laws prevent its complete elimination. In a high-quality receiver the minimum detectable signal becomes fainter in proportion to the reciprocal of the square root of the bandwidth used and in proportion to the reciprocal of the square root of the time over which the received signal is averaged. Both these qualities are utilized in modern radio telescopes to improve sensitivity. In some cases bandwidths of hundreds of Megahertz are used. In such cases, signal averaging times may be as much as 30 hours. Astronomers prefer to build their radio telescopes on as large an area as possible, in order to capture the maximum amount of energy, and with energy-collecting elements that are constructed to precise geometric configurations that allow operation at higher radio frequencies and provide better resolution. They have found that it is particularly important to design structures that preserve their precise geometry in the presence of the changing force of gravity as the structure moves, the force of the wind, and the thermal deflections due to uneven heating of the telescope structure. Steerable Paraboloid In 1937, Grote Reber built the first steerable paraboloid antenna as an amateur project at Wheaton, Ill. This antenna became the prototype of most modern large radio telescopes. During and immediately after World War II the large radar instruments built for military purposes were used as radio telescopes. In the 1950s the main thrust of radio telescope construction was to build ever larger steerable paraboloids, a process that culminated in the 91-m (300-ft) paraboloid at Green Bank, W. Va., built in 1963, and the 100-m (328-ft) telescope near Bonn, Germany, constructed in the late 1960s. The development of more precise antennas allowed operation at higher radio frequencies, which is important to the study of interstellar molecules. The most powerful for the study of such molecules is the 11-m (36-ft) radio telescope operated by the National Radio Astronomy Observatory at Kitt Peak, Ariz. An important variation of the paraboloidal antenna is the use of a fixed spherical antenna, as in the world's largest radio telescope, 305 m (1,000 ft) in diameter, constructed in 1963 near Arecibo, Puerto Rico. Another such antenna is the Soviet RATAN-600--an acronym for Radio Astronomy Telescope of the Academy of Sciences (Nauk)--in the northwestern Caucasus; it has about one-fourth the reflecting area of the Arecibo antenna. The operation of the paraboloid radio telescope is identical in concept to that of large optical telescopes. A reflector consisting of a paraboloid is oriented so that its axis is pointed at the place in the sky whose radio emission is to be measured. An excellent reflector surface at radio wavelengths need be smooth only to an accuracy of about one-fortieth wavelength or better, or typically about 1 millimeter. Thus ordinary sheet metal is usually used. The paraboloid reflects all rays coming to it from the place of interest to the focus of the paraboloid. At that point a radio antenna is located to capture the focused radio waves and convert them into an electrical signal of the same frequency as the incoming waves. This antenna is often a microwave horn or a simple dipole antenna. In the Arecibo telescope the reflector is spherical and focuses the radio emission to a long line, some 29 m (96 ft) long, rather than to a point. An ensemble of antennas must be placed all along this line to capture the focused radiation. The electrical signal is carried from the antenna through a waveguide or wires to a high-sensitivity radio receiver. This receiver has electronic or waveguide filters that allow only those radio frequencies to pass that the astronomer wishes to observe. The signals, which may be at a very low power level such as (10 to the power of -20) watt, are then amplified in a special amplifier. The amplifiers used are low noise, meaning that they themselves add a minimum of radio noise to the weak incoming signals. The most commonly used amplifiers are the parametric amplifier, often cooled to a very low temperature with liquid air to give low noise performance, or the maser (see \Tlaser\t AND \Tmaser\t), a special form of amplifier using atomic processes to give exceptionally low noise performance. These must usually be cooled to a temperature of 4 C degrees above absolute zero or lower, usually by immersing them in liquid helium, a procedure that leads to practical difficulties and high expense. Following amplification, a circuit detects the signal, meaning that it establishes the average power of the rapidly oscillating signal. This average radio power, which often changes many times in a second and may even change in less than a millionth of a second in some pulsar radiation, is then recorded. At present the most common recording method is an electronic digital voltmeter that provides a digital value of the radio emission; a computer is then used to record these digital values on magnetic tape. Later, the scientist involved will take these raw values and subject them to further mathematical manipulations to take into account such things as the amplifier gain and telescope size to lead to a value that is relevant to the physics taking place in the radio source observed. A common application of radio telescopes is in the observation of radio spectral lines of such things as atomic hydrogen, carbon monoxide, or formaldehyde. To do this expeditiously a radio receiver is preferred that receives a large number of adjacent frequency bands, or channels, simultaneously. This reception can be achieved by incorporating the appropriate number of selective filters in the radio receiver and providing each one with its own detector and data recorder; the latter may be a single computer time-shared by all the channels. This procedure is technically demanding and inflexible, however, because it calls for an entire new set of filters whenever channel bandwidth must be changed. One way to overcome these difficulties is to use an electronic device called an autocorrelator to analyze the radio signals amplified by the basic receiver amplifier. This device utilizes digital electronic circuitry to calculate the mathematical "autocorrelation function" describing the waveform of the amplified signal developed in the receiver. A simple mathematical treatment of this function in a computer, called a Fourier transformation, can recover the spectrum of the radio emission. Most radio spectrum analyzers in radio astronomy now use this approach; in some systems more than 4,000 channels are observed simultaneously. The computer has come to play a wide range of roles in the radio telescope. It converts the signals to digital values for recording, records the signals, and performs mathematical manipulations on the recorded signals to convert them into physically meaningful data. In some cases computers automatically steer telescopes, control receiver frequencies, bandwidths, and observing times, and perform other functions. Mills Cross The quest for good resolution and large energy-collecting area has led to the development of several ingenious radio telescope systems. One of the earliest was the Mills Cross, invented in 1953 by Bernard Mills of the University of Sydney, Australia. In the Mills Cross two antennas are used, each long and thin and fixed to the ground. Because of their shapes each antenna has good resolution in one direction, the direction parallel to its long dimension, and poor resolution in the other direction. In a Mills Cross, the two antennas are arranged at right angles to one another, usually so that they look like a cross from above (hence the name), and the signals from the two antennas are multiplied together. The result is an overall response that gives good resolution in both directions. The largest Mills Cross is at Molonglo, Australia, where the antennas are each one mile long. Other large Mills Crosses are at Bologna, Italy and Penticton, British Columbia. The configuration of a Mills Cross makes pointing the telescope's response pattern difficult and following objects as the Earth moves impossible; the telescope is also difficult to operate at different radio frequencies. As a result new Mills Crosses are not being built. Radio Interferometers A very important development of the early 1960s and subsequent years was the application of groups of radio antennas as radio interferometers. In this approach, several antennas are connected together simultaneously, usually by an ordinary electrical cable connection but sometimes via radio links over distances of more than 80 km (50 mi). Alternatively, the signals received at various antennas can be tape-recorded and subsequently played into a common radio receiver simultaneously. The radio interferometer permits extremely high resolution and a very large equivalent antenna collecting area at much less cost than would be called for if these were achieved with a single large antenna. This procedure is the basis of most of the major instruments recently constructed or planned. One of the most important developments to grow out of the successful application of interferometers was the process of aperture synthesis, pioneered by Sir Martin \TRyle\t. When two parabolic antennas are connected together as an interferometer, the pair gives the same information to the radio receiver as two points on a much larger paraboloid. Research showed mathematically that if a pair of antennas was moved so that the many lengths and orientations of the line connecting them duplicated all the lines that occur in a paraboloid, then the information from the antennas could be combined to give exactly the same performance as the larger paraboloid would have achieved. This method made it feasible to duplicate the performance of a large paraboloidal radio telescope, for example, 1.6 km (1.0 mi) in diameter, by moving two antennas to a large number of positions within a circle on the Earth 1.6 km (1.0 mi) in diameter. In aperture synthesis the signals received at the antennas must be recorded precisely. When all the moves of the antennas have been achieved, the recorded signals can be combined to construct a picture of a small portion of the sky that would have the same clarity as though a 1.6-km (1.0-mi) telescope, in this example, had been built. This approach has proven effective. In existing aperture synthesis systems, large numbers of antennas are used, with each antenna connected to every other antenna; this procedure allows observance of a large number of antenna spacings and orientations at a given instant. The antennas themselves are physically moved to change the length of the lines connecting the antennas, and the rotation of the Earth is utilized to achieve an effective rotation with respect to the sky. A large operating aperture synthesis is the one at Westerbork, the Netherlands, where antennas are spaced along a line one kilometer long. Another powerful system is operated by Cambridge University. The largest system now completed is the Very Large Array, which the National Radio Astronomy Observatory constructed on the Plains of San Augustin near Socorro, N.Mex. This system became operational in 1981, and possesses 27 steerable paraboloids, each 25 m (82 ft) in diameter, arranged along three railroad tracks. The three rail lines form an equiangular "Y" shape, and each is 20 km (12 mi) long. All the antennas are connected to the other antennas, giving 351 interferometer pairs at any given time. The resolution achieved by this instrument is a few tenths of an arc-second, about the same as the largest optical telescopes under the best atmospheric conditions. The entire array provides astronomers with the equivalent performance of a fully steerable radio dish 27 km (17 mi) in diameter. While it was under construction, the array already produced magnificent results. Another powerful application of interferometry is called Very Long Baseline Interferometry (VLBI). In this technique, two or more radio telescopes at different locations observe the same region simultaneously. Atomic clocks are used at each telescope to control the radio telescope electronics and to synchronize the observations to accuracies that are better than one-thousandth of a second. At each station video tape recorders are used, with atomic clock synchronization, to record the signals received. The tapes are then brought to a processing location where they are played simultaneously into a device that mimics the electronic operation of the interferometer electronics that would have been used if the telescopes had been connected together in the conventional way. The output is a normal output from an interferometer but with the telescopes widely separated. This procedure is commonly used with a group of four or more telescopes spread all across the United States and a Very Long Baseline Array (VLBA) of 10 telescopes has been proposed. The procedure is also sometimes used with intercontinental baselines utilizing telescopes in the United States and in Australia, Britain, or the Soviet Union. Resolutions far better than optical resolutions are obtained which are comparable to those that could be obtained by a fully steerable radio dish nearly the size of the Earth. FUTURE OF RADIO ASTRONOMY Radio astronomy is pursued at a large number of institutions, including many universities and at two U.S. centers sponsored by the National Science Foundation: the National Astronomy and Ionosphere Center, which operates the Arecibo Observatory, and the National Radio Astronomy Observatory, which operates the Very Large Array near Socorro, N. Mex., and telescopes at Kitt Peak, Ariz., and Green Bank, W. Va. (where the telescope collapsed, however, in 1988). Work began in the mid-1980s on the construction of a radio installation that is to stretch across the United States and its territories. Called the Very Long Baseline Array, it will consist of ten parabolic antennas, each 25 m (82 ft) in diameter. The dishes are to be located in the states of Hawaii, Washington, California, Arizona, New Mexico (two antennas), Texas, Iowa, and Massachusetts, and on St. Croix in the Virgin Islands. Its operations center will be in Socorro, N. Mex. Scheduled to be completed in the early 1990s, the Very Long Baseline Array will have a resolution 1,000 times that of any currently existing optical or radio telescope. Canada is planning a similar array, called the Canadian Long-Baseline Array, which will consist of eight telescopes, each 32 m (105 ft) in diameter, arranged in a line across the southern part of the country. In the Southern Hemisphere, Australia is also planning a continent-wide array of radio telescopes. Another idea being explored is the development of an interferometer using a telescope aboard a spacecraft to observe simultaneously with another telescope on Earth or also in orbit. Radio telescopes are also being constructed to explore the last remaining untapped region of the electromagnetic spectrum: the submillimeter region that lies between the very shortest radio wavelengths and the very longest wavelengths of infrared radiation. Because waves in this region of the spectrum are strongly absorbed by atmospheric water vapor, the submillimeter radio antennas to receive them must be built in arid regions, preferably at high altitudes. Such installations exist or are being planned by British, Dutch, German, U.S., and other radio astronomers at sites such as Mauna Kea in Hawaii. Frank D. Drake Bibliography: Christiansen, W.N., and Hogbom, J.A., Radiotelescopes, 2d ed. (1988); Field, George B., and Chaisson, Eric J., The Invisible Universe (1985); Goldsmith, P.F., Instrumentation and Techniques for Radio Astronomy (1988); Gordon, M.A., "VLBA--A Continent-Size Radio Telescope," Sky & Telescope, June 1985; Hey, J. S., The Radio Universe, 3d ed. (1984); Hjellming, R. M., and Bignell, R. C., "Radio Astronomy with the Very Large Array," Science, June 18, 1982; Learner, Richard, Astronomy through the Telescope (1981); Morris, Mark, ed., The Center of the Galaxy (1989); Rohlfs, Kirsten, Tools of Radio Astronomy (1986); Shields, J.P., The Amateur Radio Astronomer's Handbook (1986); Smith, D.H., "The Submillimeter Giants," Sky & Telescope, August 1985; Spitzer, Lyman, Jr., Searching Between the Stars (1982); Sullivan, W.T., ed., The Early Years of Radio Astronomy (1984); Thompson, A.R., and Moran, J.W., Interferometry and Synthesis in Radio Astronomy (1986). See also: \Tobservatory, astronomical\t.