© CAMBRIDGE UNIVERSITY PRESS 1997

1.2 Radio Surveys and Quasars

Quasars were originally discovered as a result of the first radio surveys of the sky in the late 1950s. By this time, the angular resolution of radio observations was good enough to identify the strongest radio sources with individual optical objects, often galaxies, but sometimes stellar-appearing sources. Important early surveys included the following:

3C and 3CR: The third Cambridge (3C) catalog (Edge et al. 1959), based on observations at 158 MHz, and its revision, the 3CR catalog (Bennett 1962), at 178 MHz, detected sources down to a limiting flux of 9 Jy ¹. The 3C catalog is not limited by flux sensitivity, but by ``confusion'' of sources at low flux levels; fainter sources become so numerous on the sky that they cannot be unambiguously distinguished from one another with poor angular resolution observations. There are 471 3C sources, and 328 3CR sources, numbered sequentially by right ascension (e.g., 3C 273, which is at epoch 1950.0 coordinates ₁₉₅₀ = 12^h 26^m 33.^s35, ₁₉₅₀ = +02°19'42"), with the numbering between the two catalogs kept as consistent as possible. Sources which appear in the 3CR but not the 3C are kept in proper right ascension order by appending a decimal point and additional digit to the immediately preceding source (e.g., 3C 390.3). All 3C sources are north of -22° declination, but the 3CR excludes sources south of -5°.

PKS: This was an extensive survey (Ekers 1969) of the southern sky (declination < +25°) undertaken at Parkes (PKS), Australia, originally at 408 MHz (detection limit 4 Jy) and later at 1410 MHz (to 1 Jy) and 2650 MHz (to 0.3 Jy). These sources are designated by 1950.0 position, using the format ``HHMM±DDT'', where HHMM refers to the hours and minutes of right ascension (epoch 1950.0), ± is the sign of the declination, and DDT is the declination, in degrees (DD) and tenths (T) of degree ² (e.g., 3C 273 = PKS 1226+023); this is still the most common, and useful, system of naming quasars.

4C: The Fourth Cambridge survey was a more sensitive version (limiting flux 2 Jy) of the 3C, again undertaken at 178 MHz (Pilkington and Scott 1965, Gower, Scott, and Wills 1967). Names of 4C sources are given as ``±DD.NN'', where ±DD is the source declination, and NN is a sequence number within the declination band (e.g., 3C 273 = 4C 02.32).

AO: The Arecibo Occultation (AO) survey is notable for the extremely accurate positions it produced as a result of observing radio sources as they were occulted by the Moon (Hazard, Gulkis, and Bray 1967). As a small angular size radio source is occulted by the Moon, the moment of disappearance identifies its position as being somewhere along the locus of points defined by the preceding limb of the Moon. Subsequent occultations of the same source give additional such loci, all of which intersect at a single point. Thus, the location of the points is limited by timing accuracy and the accuracy to which the position of the limb of the Moon is known, not by the size of the radio beam. Names of these sources are Parkes-style.

Ohio: The Ohio radio survey (e.g., Ehman, Dixon, and Kraus 1970) was made with a unique 79 x 21m transit telescope at 1415 MHz. The unusual geometry of the telescope produces an irregular beam of half-power beam width ~ 10' in right ascension and ~ 40' in declination. Source names are given as ``Ox-NNN'', where x is a letter indicating an hour-wide band of right ascension (with ``A'' and ``O'' excluded <sup>3</sup>, so sources between 0<sup>h</sup> and 1<sup>h</sup> are ``OB'' and sources between 23^h and 0^h are ``OZ'') and NNN is a serial number (e.g., 3C 273 = ON 044). As in many of the early surveys, the positions and fluxes are not reliable (particularly at sub-jansky flux levels), but the Ohio survey is notable in that up through the late 1970s some of the most distant and most luminous known quasars (e.g., OQ 172 = 1442+101 and OH 471 = 0642+449) were Ohio sources.

Most radio sources at high Galactic latitude were identified with resolved galaxies. However, the positions of some of these radio sources were found to be coincident with objects that looked like stars on normal photographs, such as the Palomar Sky Survey. The first strong radio source unambiguously identified with a star-like optical source was 3C 48. On the basis of a radio position obtained with a two-element interferometer, Matthews and Sandage (1963) found that the optical counterpart of this source was a magnitude 16 star. However, the photographic spectra obtained of this source were very confusing, as they showed strong very broad emission lines at unidentified wavelengths. It is worth noting how unsuitable photographic spectra are for work on quasars; it was unclear to the first investigators whether these broad features were emission lines or merely the continuum between broad absorption lines, as in white dwarf spectra. Photometry of these objects revealed that they are anomalously blue (relative to normal stars). Needless to say, the nature of these ``radio stars'' was very uncertain.

Another radio source identified with a star-like object, in this case on the basis of an accurate lunar occultation measurement (Hazard, Mackey, and Shimmins 1963), was 3C 273. The first breakthrough in understanding these extraordinary objects came with Maarten Schmidt's realization (Schmidt 1963) that the emission lines seen in the spectrum of this source were actually the hydrogen Balmer-series emission lines and Mg II 2798 at the uncommonly large redshift z = 0.158, where we recall that the redshift is an observational quantity defined by the observed wavelength of a spectral line relative to its laboratory wavelength ₀,

(1.2)

This redshift was approximately an order of magnitude larger than those of the original Seyfert galaxies and was among the largest ever measured at the time, with only a few very faint rich clusters of galaxies rivaling it. The obvious interpretation of the redshift is that it is of cosmological origin, a consequence of the expansion of the Universe, and the Hubble law thus gives the distance

(1.3)

where h₀ is the Hubble constant in units of 100 km s^-1 Mpc^-1. More disturbing than this vast distance was the enormous luminosity implied; 3C 273 was and remains the brightest known quasar (B = 13.1 mag). Using the formula for the distance modulus

(1.4)

(where cz / H₀ is measured in Mpc ⁴), the absolute magnitude of 3C 273 is M_B = -25.3 + 5log h₀, which is about 100 times as luminous as normal bright spirals like the Milky Way or M31. Once the redshift mystery was unlocked, identification of lines in 3C 48 (Greenstein and Matthews 1963) and other quasar spectra followed quickly.

As the physical nature of these luminous star-like objects was not understood, they became known simply as ``quasi-stellar radio sources'', a term which was subsequently shortened to ``quasars''. ⁵

The probable importance of quasars was recognized immediately. The extremely high luminosities of these objects implied physical extremes that were not found elsewhere in the nearby Universe. The suggestion that massive black holes might be involved appeared early (e.g., Zel'dovich and Novikov 1964). The possible role of active nuclei in galaxy formation and evolution was also seen (e.g., Burbidge, Burbidge, and Sandage 1963). The high luminosities of quasars also imply that they might serve as important cosmological probes, since they could be in principle detected and identified at very large distances. These considerations provided early and continuing strong motivation for finding quasars and studying their properties. As the number of known quasars increased, progressively greater and greater redshifts were identified, and as of the mid-1990s the highest observed redshifts are z 5.

¹ A jansky (Jy), named after the pioneering radio astronomer Karl Jansky, is a unit of specific flux, and in early radio astronomy was simply called a ``flux unit''. It is defined as 1 Jy = 10^-26 watts m^-2 Hz^-1 = 10^-23 ergs s^-1 cm^-2 Hz^-1. Back.

² The tenths of degree designation was added only after the number of radio sources became rather large. Since the 1980s, some catalogs have begun using declination tags of ±DDMM, where MM is in arcminutes. Back.

³ The ``OA'' sources actually constitute a preliminary version of the Ohio catalog which was compiled by Kraus (1966). Back.

⁴ We will see in Chapter 9 that this formula for the distance is in general an approximation that is valid for only small values of z. Back.

⁵ The term ``quasar'' is attributed to H.-Y. Chiu (1964). It was some time, however, before the word was generally accepted into the astronomy lexicon. Indeed, under the editorial direction of S. Chandrasekhar, The Astrophysical Journal resisted use of ``quasar'' for many years, finally conceding with ``regrets'' in 1970 (see Schmidt 1970). An informal vote on the term at the Second Texas Symposium on Relativistic Astrophysics in 1964 yielded 20 ayes and 400 or so abstentious (Robinson, Schild, and Schucking 1965). Back.