[ CATS home ] [ Back to CATS list ] [ ftp ]

1980MNRAS.191..607Porcas+
see also Radio sources selected at 966 MHz I.

Radio positions and optical IDs for sources selected at 966 MHz II

R.W. Porcas, C.M. Urry

National Radio Astronomy Observatory, Edgemont Road, Charlottesville, Virginia 22901, USA

I.W.A. Browne, A.M. Cohen, E.J. Daintrce & D. Walsh

University of Manchester, Nuffield Radio Astronomy Laboratories, Jodrell Bank, Macclesfield, Cheshire. SK11 9DL
SUMMARY
In a previous paper, accurate radio position measurements and optical identifications were published for 538 sources selected from a survey made at 966 MHz. The remaining 242 sources in the sample were either heavily resolved or confused in these observations and no accurate radio positions could be obtained. Here we present radio position measurements of those remaining sources, made at 2.7 and 5.0 GHz using the 300-ft telescope at NRAO. The positions obtained have rms errors of about 6" in right ascension and 11" in declination. A few of the sources have also been observed at 2.7 GHz with the NRAO three-element interferometer, giving positions accurate to 1" rms.

These new positions have been used in a search for optical identifications on the Palomar Sky Survey. Optical positions (with rms errors of 0.5") have been measured for candidate identifications. 135 identifications are proposed, and the reliability of these is estimated at about 80 per cent.

1. INTRODUCTION

A study is in progress of radio sources from a 966-MHz survey made at Jodrell Bank. The MKIa radio telescope (diameter 76m) was used to find sources stronger than 0.7 Jy at 966 MHz, between declination +40^o and +71^o . The 780 of these sources which lie more than 10^o from the galactic plane form the basis of this study. In a previous paper (Cohen et al. 1977; hereafter Paper I), we presented radio positions and optical identifications for 538 of the sources, based on measurements made at 966 MHz with the Jodrell Bank MKIA-MKII interferometer. These sources have relatively compact radio structures (less than about 1') and positions accurate to about 2" rms were obtained. However, the observations were insufficient for determining positions of large sources or of those strongly confused by other sources.

In this paper we present positions for the remaining sources in the sample. Most of these positions are derived from new measurements made with the NRAO 300-ft transit telescope at Greenbank, West Virginia, at frequencies of 5.0 and 2.7 GHz. For a few of the sources we have obtained positions from measurements made at 2.7 GHz with the NRAO three-element interferometer, and for some we have taken positions published elsewhere in the literature. In Sections 2 and 3 we describe these new observations, and also discuss the occurrence of multiple source responses.

2. THE OBSERVATIONS

2.1 Pencil-beam measurements at 5.0 GHz
Most of these observations with the 300-ft telescope were carried out during 1975 June and July, and additional observations were made in 1975 November and 1976 March. The feed system consisted of two linearly polarized horns, symmetrically offset from the axis and separated by 7.2' on the sky, with beamwidths (FWHM) of 2.7' (see, e.g. Davis 1971). A beam-switching mode was employed, the switched output from the horns being fed to a single-channel, cooled, parametric amplifier with a nominal 3 dB bandwidth of 150 MHz. Final balancing of the receiver was done using synchronous IF gain modulation. The system temperature was 150 K. A rotatable mount enabled the separation of the feed horns to be orientated at an angle from the east-west line (at which the E vector position angle was 0^o). Thus orientations of 11^o or 22^o produce separations of the two beams in declination of a half or one beamwidth, respectively. Observations of a source at two transits are then sufficient to derive the source position by modelling the data with a twodimensional Gaussian function.

Drift observations were made of most of the 780 survey sources, sampling and recording the data on magnetic tape every 2 s. The scan length was chosen to allow sufficient data before and after transit for the removal of a baseline (linear in most cases) but was longer for those sources without precise positions. In many cases more than two transits were required to obtain sufficient data on these sources. The ratio of the gains of the two feeds was determined by observations of a few sources with the feeds orientated at 0^o. The receiver gain was calibrated by injecting noise from a noise tube in one side of the system every fifteenth sample; in the subsequent data reduction these samples were removed and used to calibrate the data in units of antenna temperature.

>From observations of 250 of the survey sources with accurately known positions, the declination dependences of the telescope pointing offsets and beamshape were determined. The calibration parameters found in this way were then used to correct the positions of the remaining sources. The position errors quoted in Table 1 are those derived from the uncertainties in the calibration procedure or the Gaussian fitting errors if larger. Where the response differed significantly from that expected from a point source, a rough estimate of the source size was determined.

2.2 Pencil-beam measurements at 2.7 GHz
These observations were carried out with the 300-ft telescope during 1975 December, 1976 August and September. For these observations the telescope was equipped with a three-feed system, the central feed being on-axis and straddled by feeds offset by 9.8' on the sky, with beamwidths of 5' (see, e.g. Owen 1975). The outer feeds responded to a single circular-polarization mode and the central feed to both left- and right-hand circular polarizations. All four outputs were switched against a reference load and fed to a four-channel receiver with parametric amplifiers of bandwidths 40 MHz and system temperatures 120 K. IF gain modulation was employed to balance the receiver.

The drift scans and data analysis were similar to those used for the 5.0-GHz observations. A feed box orientation of 14^o from an east-west line was used, and an orientation of 0^o was used to determine the relative gains of the feeds. In many cases a single transit observation was sufficient to model the source response.

A pointing problem was encountered during daytime observations, due to strong sunlight producing differential heating of the telescope support towers and panels. This resulted in systematic pointing excursions of up to 40" in declination and 8" in right ascension. (This effect was not noticeable during the 5.0-GHz observing sessions which were mainly during overcast weather). The effect was removed by calibrating on sources of known position.

Flux densities derived from these 2.7-GHz observations and those at 5.0 GHz described above will be published separately.

2.3 Three-element interferometer measurements at 2.7 GHz
These observations were made for just a few sources which were suspected quasars on the basis of a preliminary search for identifications, and hence had a relatively high probality of being compact. The NRAO three-element interferometer has been described by Hogg et al. (1969). The observations were taken during 1976 October with a baseline configuration of 300-1200-1500m, and during 1976 December with a configuration of 900-1800-2700m. Each source was observed for about 8 min at each of 3 - 5 different hour angles in either one or both of the observing sessions. The instrumental phase was determined every hour by observing compact radio sources of accurately known position, and the calibrated visibility data were Fourier transformed and 'CLEANED' to produce rough source maps. Because of the very sparse nature of the u.v. coverage, these maps were only sufficient to indicate which sources were unresolved. For these unresolved sources only, positions were obtained, using the phase of the visibility data.

3. RESULTS

3.1 The Radio Positions
Radio positions for the sources are listed in Table 1, which is to be found on Microfiche MN191/1. For most sources we have given the most accurate position from our measurements. In five cases, the position from the 966-MHz survey is the best available. A reference number (1 - 4) indicates the origin of each position. For a few sources, previously published positions are quoted and references are given in the footnotes.
3.2 Multiple Sources
The observations made with the NRAO 300-ft telescope have approximately six times the angular resolution of those of the 966-MHz survey, and they show that 38 of the 'sources' listed in the survey consist of more than one peak of radio emission. In 37 cases there were just two peaks and in one there were three. It is probable that the majority of these multiple responses arise from physically unrelated radio sources. We suggest this because the original 966 MHz survey was confusion limited and a number of weaker sources in it will be heavily confused or blends of two (or more) sources of comparable flux density. Such multiple responses near the position of a survey source are designated by A, B ... etc.in Table 1. In a few cases, weak sources were detected in the 300-ft scans, generally at large distances from the survey source position and clearly did not contribute to the original 966-MHz 'source'. Such sources are mentioned in the notes on individual sources.

All the east-west blends should have been found on the drift scans, but most of the north-south ones will have been missed. Using the distributions of separations and position angles of those multiple sources actually found, we estimate the number of multiple sources we are likely to have missed as about 50, giving an estimated total of 90 multiple sources. This is approximately the number expected for such a confusion-limited survey.

Sometimes when only one source is found it has a flux density much lower than expected from other flux density data. In these cases we suspect the existence of another undetected source and mention this possibility in the footnotes.

There also exist very extended sources (e.g. DA240, 3C236) which, whilst being single sources, would nevertheless produce a two-peaked response in the pencil-beam observations. This possibility is considered below in the discussion of the optical identification procedure.

4. OPTICAL IDENTIFICATIONS

Using the best available radio position for each source, a search has been made on the Palomar Sky Survey prints for optical identifications. Survey sources which give rise to widely separated responses were generally treated as two independent sources, and the position of each has been searched. However, the line joining them was also checked for bright galaxies and, for the double sources 0157+405, 0247+467 and 1127+553, both components are now thought to be associated with a bright galaxy lying between them. The remaining 34 pairs appear to consist of unrelated sources, many of them being separately identifiable, and one further source (0730+504) has three distinct peaks of emission. This brings the total number of independent sources in the present list up to 280.

Candidate identifications are classified according to the scheme in Table 2. Their optical positions have been measured to an accuracy of about 0.5" rms, and red and blue apparent magnitudes mr and mb, have been estimated (using the methods described in Paper I). 

Table 2. Identification Classification
BSO,NSO,RSO  Blue, neutral or red stellar object respectively.
QSO          Quasi-stellar object confirmed by optical spectroscopy.
BO, NO, RO   Blue, neutral or red object whose image is too faint to
             class as stellar or extended.
G            Galaxy (extended image on at least one print).
DG           Double galaxy, or two very close galaxies.
GCL          Galaxy in a cluster.
CL           Cluster.
E            Empty field.
NI           Not identified (see text).
...:         A tentative identification only.
4.1 Identification Criteria
In deciding whether to accept or reject any particular object as an identification, the possible radio extent of the present sources has been borne in mind. Although detailed structural information is not available for the majority of these sources, it is known that many of them have angular sizes in excess of 60". The radio position errors are typically 6 x 11" rms in RA and Dec respectively, and thus radio-optical position differences of three or four times these errors are possible for the larger sources. To accept as an identification any object lying within the possible radio extent of the source, or even within three times the position errors, would give too high a number of misidentifications with chance background objects (mostly galactic stars and faint galaxies). To obtain a reasonable level of reliability for the identifications, we have adopted as rough guidelines the following criteria. These are based on the background densities of various types of optical objects given in Paper I.

(i) Galaxies brighter than mr ~ 16 are accepted up to 1' from the radio position. The number of these occurring by chance in 280 random fields of radius 1' is about five. In fact 45 such galaxies have been found near the radio positions, implying a reliability of 90 per cent for these identifications.
(ii) Galaxies with 16.0 < mr < 18.0 are accepted up to four times the typical radio position errors. Again about five such galaxies are expected by chance in 280 search fields, but 27 were actually found, implying a reliability of over 80 per cent for these identifications.
(iii) Stellar objects: at high galactic latitudes (|b| > 20^o) QSOs are accepted out to twice the typical position errors. In the 205 high-latitude fields, about five chance BSOs are expected. Thirty were actually found, again implying a reliability of over 80 per cent. However, at low galactic latitudes the background sky density of blue stars rises sharply and, for sources with 10^o < |b| < 20^o, only two identifications with BSOs have been made, both of which are less than one standard deviation from the radio position.

The density of red and neutral stellar objects is everywhere too high for reliable indentifications to be made with such objects. One tentative identification has been made with an RSO which is possibly a compact galaxy.
(iv) Faint galaxies and other faint objects cannot generally be accepted as identifications as their background density is too high. A total of 16 such objects which lie within only one standard deviation of the radio positions have been proposed as tentative identifications, but their reliability is low.
(v) Few sources can confidently be identified as empty fields since the number of chance objects in the optical fields is so high. Sources for which no reliable identification can be found using the above criteria are classified 'NI' (not identified). Objects which are possible candidates in such fields are mentioned in the notes on individual sources.

4.2 Identification content

Identifications are proposed for 135 of the 280 sources, and are listed in Table l which also gives their optical positions and magnitudes. These identifications are summarized in Table 3. 
            Table 3. Identification content.

         Bright galaxies (mr < 18.0)         72
         Faint galaxies                      10
         QSO                                  7
         BSO                                 23
         RSO                                  1
         Faint objects (RO+NO)                6
         Clusters                            16
                                            ---

         Total positive identifications     135

         E                                   10
         NI                                 135
                                            ---
         Total number of sources            280
The percentage of galaxies brighter than mr ~ 18 amongst these sources (26%) is much higher then the 7% found for the more compact sources in Paper I. This is consistent with the correlation of angular size with apparent magnitude for galaxies first noted by Minkowski (1961).

Since BSOs and bright galaxies can be identified with reasonable confidence, few such identifications should have been missed. However, almost half of the present sources remain unidentified and thus these are probably mostly empty fields and faint galaxies.

4.3 Finding charts
Finding charts for new identifications are given in Plate 1. The reproductions are from the 'E' (red) prints of the Palomar Sky Survey (copyright 1957, National Geographic Society). The fields measure 8.5'; the upper right-hand corner is north preceding.
4.4 Two amendments to the identifications for sources in Paper I
1050+542. The observations with the 300-ft telescope show a lobe error in the MKIA-MK II interferometer position. The 5-GHz radio position (1950) is:
RA: 10h 50m 57s.1 (+-6"). dec +54^o 17' 52'' (+- 11")
The new identification is a BSO at -6.1",-3.9" from this position, mr = 18.2, mb = 18.9.
1636+473. This source was previously listed as an empty field. Unpublished measurements at 2.7 and 8.1 GHz made using the NRAO 3-element interferometer indicate that this source has a double structure with component separation of 19". One component has an inverted spectrum and within 1" there is a QSO, mr = 17.5, mb = 18.0, at
RA: 16h 36m 19s.18. dec: +47^o 23' 28".2 (+- 0.5").
Finding charts for these two new identifications are given in Plate 1.

5. CONCLUDING REMARKS

The new radio positions presented here have enabled us to propose identifications for about half of the extended or confused sources in this sample. Some structural information is necessary to identify the remaining 135 sources.

A paper is in preparation discussing the identifications presented here together with those already published in Paper I.

ACKNOWLEDGMENTS

We are grateful to Bev and Derek Wills and to Pat Moore for communicating unpublished data.

REFERENCES

Bailey, J.A.& Pooley, G.G., 1968. Mon. Not. R. astr. Soc., 138, 51.
Baldwin, J.E.& Pooley, G.G., 1973. Mon. Not. R. astr. Soc., 161, 127.
Bennett, A.S., 1962. Mem. R. astr. Soc., 68, 163.
Berkhuijsen, E.M., 1977. Astr. Astrophys., 57, 9.
Bridle, A.H., Davis,M.M., Fomalont,E.B., & Lequeux, J., 1972. Astr.J., 77, 405.
Brown, R. Hanbury & Hazard,C.,1953. Mon. Not. R. astr. Soc., 113, 123.
Brundage, R.K., Dixon, R.S., Ehman, J.R.& Kraus, J.D.,1971. Astr.J., 76, 777.
Bystedt, J.E.V., 1975. Astr. Astrophys., 39, 155.
Caswell, J.L.& Wills, D.,1967. Mon. Not. R. astr. Soc., 135, 231.
Caswell, J.L.& Crowther, J.H., 1969. Mon. Not. R. astr. Soc., 145, 181.
Cohen, A.M., Porcas, R.W., Browne, I.W.A., Daintrce, E.J.& Walsh, D., 1977. Mem. R. astr. Soc., 84, 1.
Davis, M. M., 1971. Astr.J., 76, 980.
de Bruyn, A.G. & Wilson, A.S., 1976. Astr. Astrophys., 53, 93.
Douglas, J.N., Bash, F.N., Ghigo, F.D., Moseley, G.F. & Torrence, G.W., 1973. Astr.J., 78, 1.
Edge,D.O., Shakeshaft, J.R., MacAdam,W.B., Baldwin,J.E. & Archer,S., 1959. Mem. R. astr. Soc., 68, 37.
Galt, J. A. & Kennedy, J.E.D., 1968. Astr.J., 73, 135.
Gearhart, M.R., Lund, J.M., Frantz, D.J. & Kraus, J.D., 1972. Astr. J., 77, 557.
Gower, J.F.R., Scott, P.F. & Wills, D., 1967. Mem. R. astr. Soc., 71, 49.
Heeschen, D.S. & Wade,C.M.,1964. Astr. J., 69, 277.
Hogbom, J.A. & Carlsson, I.,1974. Astr. Astrophys., 34, 341.
Hogg, D.E., Macdonald, G.H., Conway,R.G. & Wade, C.M., 1969. Astr. J., 74, 1206.
Kuhr,H., 1977. Astr. Astrophys. Suppl., 29, 139.
Kellermann, K. I., Pauliny-Toth, I. I. K. & Williams, P. J. S., 1969. Astrophys. J., 157, 1.
Fruit, P.C. van der l977. Astr. Astrophys., 59, 359.
Macdonald, G.H., Kenderdine, S. & Neville, A. C., 1968. Mon. Not. R. astr. Soc., 138, 259.
Mackay, C.D., 1969. Mon. Not. R. astr. Soc., 145, 31.
MacLeod, J.M., Swenson, Jr., G.W., Yang, K.S. & Dickel, J.R., 1965. Astr. J., 70, 756.
Miley, G.K. & van der Laan, H., 1973. Astr. Astrophys., 28, 359.
Miley, G. K., Wellington, K. J. & van der Laan, H., 1975. Astr. Astrophys., 38, 381.
Minkowski, R., 1961. Proceedings of the Fourth Berkeley Symposium on Mathematical Statistics and Probability p.245, ed. J. Neyman, University of Galifornia Press, Berkeley and Los Angeles.
Nordhover, K.J.E., 1973. Mon. Not. R. astr. Soc., 165, 369.
Owen, F.N., 1975. Astr. J., 80, 263.
Owen, F.N., & Rudnick, L., 1976. Astrophys. J. Lett., 205, L1.
Pauliny-Toth, I.I.K., Preuss, E., Witzel, A., Kuhr, H., Davis, M.M., Fomalont, E.B. & Kellermann, K.I., 1978. Astr. J., 83, 451.
Pauliny-Toth, I.I.K., Wade, C.M. & Heeschen, D.S., 1966. Astrophys. J. Suppl., 13, 65.
Pooley, G.G., 1969. Mon. Not. R. astr. Soc., 144, 101.
Riley, J.M. & Pooley, G.G., 1975. Mem. R. astr. Soc., 80, 105.
Rudnick, L. & Owen, F.N., 1976. Astrophys. J. Lett., 203, L107.
Rudnick, L. & Owen, F.N., 1977.Astr.J., 82, 1.
Seaquist, E.R. & Bignell, R.C., 1976.Astr.Astrophys., 48, 421.
Segalovitz, A., 1976. Astr. Astrophys., 52, 167.
Segalovitz, A., 1977. Astr. Astrophys., 54, 703.
Slingo, A., 1974.Mon.Not.R.astr.Soc., 168, 307.
Sramek, R. A. & Tovmassian, H.M., 1975. Astrophys.J., 196, 339.
Turland, B.D., 1975.Mon.Not.R.astr.Soc., 172, 181.
Vallee, J.P. & Wilson, A.S., 1976.Nature, 259, 451.
van der Kruit, P.C., 1977. Astr. Astrophys., 59, 359.
Veron, M.P., 1971. Astr. Astrophys., 11. 1.
Veron, M.P., 1971. Astr. Astrophys., 11. 1.
Walsh, D., Carswell, R.F. & Weymann, R.J., 1979. Nature, 279, 381.
Walsh, D. & Carswell, R.F., 1979. In preparation.
Williams, P.J.S., Kenderdine, S. & Baldwin, J.E., 1966. Mem. R. astr. Soc., 70, 53.
Willis, A.G., Strom, R.G. & Wilson, A.S., 1974. Nature, 250, 625.
Wills, B.J. & Wills, D., 1974. Astrophys. J. Lett., 190, L97.
Wills, D., 1967. Mon. Not. R. astr. Soc., 135, 339.
Willson, M.A.G., 1972. Mon. Not. R. astr. Soc., 156, 7.
Wilson, A.S & Vallee, J.P., 1977. Astr. Astrophys., 58, 79.
Windram, M.D. & Kenderdine, S., 1969. Mon. Not. R. astr. Soc., 146, 265.
Wyndham, J.D., 1965. Astr.J., 70, 384.
Wyndham, J.D., 1966. Astrophys.J., 144, 459.
Zwicky, F., Herzog, E., Wild, P., Karpowicz, M. & Kowal, C.T., 1961-68. Catalogue of Galaxies and Clusters of Galaxies, California Institute of Technology, Pasadena.

Key to Table 1
Column up Back to CATS manual