Filling in the Gaps in the 4.85 GHz Sky

We describe a 4.85 GHz survey of bright, flat-spectrum radio sources conducted with the Effelsberg 100 m telescope in an attempt to improve the completeness of existing surveys, such as CRATES. We report the results of these observations and of follow-up 8.4 GHz observations with the VLA of a subset of the sample. We comment on the connection to the WMAP point source catalog and on the survey’s effectiveness at supplementing the CRATES sky coverage.

💡 Research Summary

The paper presents a targeted 4.85 GHz survey aimed at filling the gaps left by existing high‑frequency radio catalogs (GB6, PMN, S5) that together form the backbone of the CRATES all‑sky flat‑spectrum source list. The authors focus on two poorly covered regions: the North Polar Cap (declination +75° to +90°) and two “PMN holes” just south of the celestial equator (≈ 300 deg² total). Using the Effelsberg 100‑m telescope, they selected candidate flat‑spectrum sources by cross‑matching NVSS (1.4 GHz) with low‑frequency surveys—WENSS (325 MHz) for the polar cap and the Texas survey (365 MHz) for the equatorial holes. Spectral indices were computed assuming a power‑law (S ∝ ν^α) and a predicted 4.85 GHz flux density (S_pred) was derived for each source.

Selection criteria differed slightly between the regions: for the polar cap they required S_pred > 65 mJy and α > ‑0.6, excluding any source already present in the S5 catalog; for the PMN holes they required S_pred > 75 mJy and α > ‑1.0. This yielded 221 candidates in the polar cap and 174 in the holes, for a total of 395 targets.

Observations were carried out in June–July 2008 over roughly 24 hours of telescope time. The 4.85 GHz multi‑horn receiver was used in a cross‑scan mode (four scans for bright sources, six for fainter ones) to obtain simultaneous on‑source and off‑source measurements, allowing real‑time correction for atmospheric contributions and telescope pointing errors. Primary flux calibrators (3C 286, 3C 295, NGC 7027) were observed frequently to tie the measurements to the Baars et al. (1977) absolute scale. After baseline subtraction, Gaussian fitting, pointing correction, opacity correction, and gain‑elevation correction, the authors derived flux densities with a mean fractional uncertainty of 2.3 % (≈ 1.2 mJy absolute error for sources near 100 mJy).

Out of the 395 targets, 368 (93 %) yielded reliable flux densities. The mean measured flux was systematically lower than the predicted values; the ratio S_obs/S_pred averaged 0.75, indicating that a simple extrapolation from low‑frequency data overestimates the true 4.85 GHz flux for about 80 % of the sample. Moreover, a substantial fraction (≈ 75 sources) turned out to be GHz‑peaked spectrum (GPS) objects whose spectra rise from ~0.35 GHz to 1.4 GHz but fall off toward 4.85 GHz, meaning they do not satisfy the CRATES flat‑spectrum criteria (α > ‑0.5, S_4.85 > 65 mJy).

To bring the newly measured sources onto the same footing as the original CRATES catalog, the authors performed follow‑up interferometric observations with the VLA in A‑configuration at 8.44 GHz (two 50 MHz IFs). A subset of 109 sources (72 from the polar cap, 37 from the holes) with S_4.85 ≥ 65 mJy and α > ‑0.6 (NVSS–Effelsberg) were observed on 1 Nov 2008 (program AH0976). Each source received a 60 s on‑source integration. Standard AIPS calibration, followed by DIFMAP imaging and Gaussian component fitting, yielded integrated 8.4 GHz fluxes with ≈ 3 % radiometric errors and positional accuracies of ≈ 0.06″. Four sources displayed resolved structure, suggesting they are not compact blazar cores.

The impact on sky coverage is significant. In the north polar cap, CRATES originally missed ≈ 160 flat‑spectrum sources; this effort identified 57 of them, raising the coverage from ~33 % to ~57 %. In the PMN holes, coverage rose from 0 % to ~24 % (24 of an estimated 100 missing sources). Overall, the fraction of the |b| > 10° sky with CRATES‑type sources increased from 97.6 % to 98.3 %.

Cross‑matching with the five‑year WMAP point‑source catalog (390 sources detected in at least one WMAP band) shows that 84 % of WMAP sources now have a counterpart in the expanded CRATES‑like list, including three WMAP sources that lie within the PMN holes and were not previously cataloged. This demonstrates the utility of the new measurements for high‑frequency (23–94 GHz) surveys, especially for the forthcoming Planck mission, whose sensitivity is ~30× that of WMAP and will produce thousands of point sources across the sky. Accurate low‑frequency counterparts will be essential for source identification, spectral modeling, and calibration of radio interferometer arrays.

In summary, the authors have shown that a modest amount of dedicated single‑dish observing time can substantially improve the completeness of high‑frequency flat‑spectrum source catalogs in regions that were previously neglected. The resulting data set provides precise flux densities, sub‑arcsecond positions, and complementary 8.4 GHz interferometric measurements, thereby enhancing the utility of the CRATES catalog for γ‑ray blazar studies, CMB foreground cleaning, and radio interferometer calibration. Future work could extend the survey to the entire northern cap and the remaining PMN gaps down to the CRATES flux limit (~65 mJy), which would require only about a week of additional telescope time but would deliver a truly all‑sky, uniform 4.85 GHz catalog.