Powerful Radio Sources in the Southern Sky. I. Optical Identifications

Since the early sixties, our view of radio galaxies and quasars has been drastically shaped by discoveries made thanks to observations of radio sources listed in the Third Cambridge catalog and its revised version (3CR). However, the largest fraction of data collected to date on 3CR sources was performed with relatively old instruments, rarely repeated and/or updated. Importantly, the 3CR contains only objects located in the Northern Hemisphere thus having limited access to new and innovative astronomical facilities. To mitigate these limitations we present a new catalog of powerful radio sources visible from the Southern Hemisphere, extracted from the GLEAM 4-Jy (G4Jy) catalog and based on equivalent selection criteria as the 3CR. This new catalog, named G4Jy- 3CRE, where the E stands for"equivalent", lists a total of 264 sources at declination below -5 degrees and with 9 Jy limiting sensitivity at ~178 MHz. We explored archival radio maps obtained with different surveys and compared then with optical images available in the Pan-STARRS, DES and DSS databases to search for optical counterparts of their radio cores. We compared mid-infrared counterparts, originally associated in the G4Jy, with the optical ones identified here and we present results of a vast literature search carried out to collect redshift estimates for all G4Jy-3CRE sources resulting in a total of 145 reliable z measurements.

The first revised version of the Third Cambridge catalog of radio sources (3CR; Spinrad et al. 1985) lists the most powerful radio sources detected in the Northern Hemisphere at 178 MHz above a 9 Jy flux density threshold listing 298 extragalactic sources, with only a small fraction (i.e., less than ∼8%) still unidentified.
The legacy value of 3CR follow up surveys can be also highlighted by results achieved thanks to: (i) the Hubble Space Telescope (HST) Snapshot Survey of 3CR Radio Source Counterparts 1 (Privon et al. 2008;Chiaberge et al. 2015;Hilbert et al. 2016), (ii) the 3CR Chandra Snapshot Survey (Massaro et al. , 2012a(Massaro et al. , 2013aStuardi et al. 2018;Jimenez-Gallardo et al. 2020) and (iii) the MUse RAdio Loud Emission line Snapshot survey (Balmaverde et al. 2018b(Balmaverde et al. , 2019Speranza et al. 2021;Balmaverde et al. 2022). The former campaign allowed us to obtain a full overview of optical properties of these powerful radio sources at ∼90% level of completeness (even if not uniform in terms of instruments and filters adopted) while the latter one (i.e., the MURALES campaign), still ongoing, can be performed only on 3CR sources at z < 0.8 and at Declination < 20 • , being visibile from the Very Large Telescope (VLT) in Chile. In 2008 the 3CR Chandra Snapshot survey also began aiming to (i) detect new jets, hot spots and lobes emitting in the X-ray band (ii) investigate nuclear emission of powerful radio sources and (iii) discover new galaxy clusters (see also Hardcastle et al. 2010Hardcastle et al. , 2012Dasadia et al. 2016;Madrid et al. 2018;Ricci et al. 2018;Jimenez-Gallardo et al. 2021, 2022aMissaglia et al. 2023), covering all 3CR sources lacking X-ray observations. Despite a small number of 3CR sources that are still unidentified and unobserved in the X-rays (Maselli et al. 2016;Missaglia et al. 2021), more than 95% of the 3CR catalog has high energy data already available in the Chandra archive (see e.g., Massaro et al. 2015a).
However, the 3CR has, unfortunately, the following drawbacks and limitations. It was created more than six decades ago and the largest fraction of 3CR radio, infrared and optical data collected to date was obtained with relatively old instruments, only rarely repeated and/or updated. It lists radio sources lying in the Northern Hemisphere, with limited access to observations that can be performed with state-of-the-art and upcoming astronomical facilities, such as Multi Unit Spectroscopic Explorer (MUSE; Bacon et al. 2010) mounted at the Very Large Telescope (VLT), the Atacama Large Millimeter/submillimeter Array (ALMA) and in the near future the world's most powerful radio telescope: Square Kilometre Array 2 (SKA; McMullin et al. 2020), the Large Synoptic Survey Telescope (LSST; Ivezić et al. 2019) as well as the Extremely Large Telescope 3 (ELT).
In the eighties the Molonglo Reference Catalog of Radio Sources (MRC; Large et al. 1981) containing nearly 12000 discrete sources with flux densities greater than 0.7 Jy at 408 MHz in the declination range between +18.5 • and -85 • and excluding regions within 3 • of the Galactic equator was created. Several multifrequency campaigns were then dedicated to augment the information of MRC sources, eventually restricted to bright samples (see e.g. Kapahi et al. 1998).
A first attempt to create a complete sample similar to the 3CR, but selected at 408 MHz, was performed by Best et al. (1999) using the Molonglo Reference Catalogue. They selected a sample listing 178 radio sources with flux density S 408 above 5 Jy, in the range of declination between -30 • and 10 • and having Galactic latitudes |b| >= 10 • . The equatorial location of all sources listed therein allowed them to achieve high spectroscopic completeness, and its footprint certainly mitigating one of the previously mentioned limitations of the 3CR: visibility from Southern Hemisphere telescopes.
An additional attempt to build a catalog equivalent to the 3CRR was carried out by Burgess & Hunstead in 2006, starting from the Molonglo Southern 4 Jy sample (MS4; see also Burgess & Hunstead 2006b). The MS4 is a complete sample of 228 southern radio sources detected at 408 MHz with integrated flux densities above 4.0 Jy, Galactic latitude |b| >10 • and declination in the range between -85 • and +30 • , all imaged at 843 MHz with the Molonglo Observatory Synthesis Telescope to obtain positions with an accuracy of ∼1 ′′ . Then radio spectra for the MS4 sources were compiled from the literature to estimate flux densities at 178 MHz and the subset of SMS4, with S 178 >9 Jy was extracted. Some sources listed in the SMS4 were also recently observed in the soft X-rays (Maselli et al. 2022).
The recent GaLactic and Extragalactic All-sky Murchison Widefield Array (MWA) survey (GLEAM; Wayth et al. 2015;Hurley-Walker et al. 2017) offers today a unique opportunity to create a southern sample of powerful radio sources matching the selection criterion of the 3CR catalog. Thanks to the MWA observations available for the whole southern sky at declinations δ < 30 • in the frequency range between 72 and 231 MHz, White et al. (2018White et al. ( , 2020aWhite et al. ( , 2020b built a complete sample of radio sources with flux density above 4 Jy at 151 MHz, namely the GLEAM 4-Jy sample (hereinafter labelled as G4Jy). The majority of radio sources included therein are extragalactic, mainly AGNs with extended structures detected at low radio frequencies (i.e., at hundreds of MHz). This sets the basis for extracting, from the G4Jy catalog, a new sample of radio sources equivalent to the 3CR (hereinafter G4Jy-3CRE) but including only those located in the Southern Hemisphere.
Here we introduce the G4Jy-3CRE catalog, listing the 264 very brightest radio sources, selected from the larger parent G4Jy catalog, above a flux density threshold of 9 Jy at ∼178 MHz, as the nominal value of the 3C, at declinations below -5 • . There are several differences with respect to the previous analysis carried out on the G4Jy catalog as listed below. In the present work we restricted our investigation to a small fraction of sources, 264 with respect to ∼1900 listed in the G4Jy catalog, with the potential advantage of performing deeper analyses with more astronomical facilities and instruments, but with the disadvantage of a limited number of high z sources implying less robust claims from a statistical perspective. On the other hand, radio sources listed in the G4Jy-3CRE sample are potentially primary targets for SKA, being the brightest ones and preparing the sample before its advent allows us to start collecting multifrequency observations that will be crucial to investigate their nature and that of their environments.
This first paper is mainly devoted to the comparison between radio images, at higher resolution than that achievable with previously available radio maps, with mid-IR and optical archival observations to confirm counterparts previously assigned to each radio source and/or determine potential incorrect associations. The analysis presented here is based on archival radio maps with higher angular resolution than those used for the G4Jy associations, such as those retrieved from the Very Large Array (VLA) Sky Survey (VLASS; Lacy et al. 2020) and the National Radio Astronomy Observatory (NRAO) VLA Archive Survey (NVAS) 4 databases, available for at least 60% of the G4Jy-3CRE catalog. This counterpart search is crucial to identify targets for future spectroscopic campaigns that are necessary to obtain source redshifts and their optical classification. We also present results of an extensive literature search carried out to obtain redshift estimates for G4Jy-3CRE sources.
The paper is structured as follows. In § 2 we present the criteria underlying the G4Jy-3CRE sample selection. In § 3 we present the results of our search for optical counterparts and a comparison with the mid-IR sources associated with the G4Jy catalog, while § 4 is dedicated to a brief description of a literature search for redshift estimates. Summary, conclusions and future perspectives are given in § 5. Appendix A is dedicated to a brief description of individual sources while Appendix B is devoted to the cross-identifications obtained comparing the G4Jy-3CRE sample with other radio catalogs based on observations carried out in the Southern Hemisphere. Finally, Appendix C is dedicated to a statistical test for the radio and mid-infrared crossmatches carried out to search for counterparts of G4Jy sources, in comparison with the optical analysis presented here.
We adopt cgs units for numerical results and we assume a flat cosmology with H 0 = 69.6 km s −1 Mpc −1 , Ω M = 0.286 and Ω Λ = 0.714 (Bennett et al. 2014). For optical photometric data we used the catalog obtained from the Panoramic Survey Telescope & Rapid Response System (Pan-STARRS; Flewelling et al. 2020) survey, where magnitudes are reported in the AB system (Oke 1974;Oke & Gunn 1983). The same applies for the Dark Energy Survey (DES; Abbott et al. 2018) for which observations are performed in optical filters similar to those of Pan-STARRS and Sloan Digital Sky Survey (SDSS; Ahn et al. 2012). Limiting sensitivity for both the Pan-STARRS and DES optical survey reaches ∼23 mag in the r band. For optical magnitudes we did not apply the correction for Galactic extinction but we report the total extinction A V , extracted from the Galactic Dust Reddening and Extinction database 5 (Schlegel et al. 1998;Schlafly et al. 2011). Spectral indices, α, are defined by flux density, S ν ∝ ν −α . Finally, given the large number of acronyms used in the paper, these are summarized in Table 1.

SAMPLE SELECTION
The MWA, operating since 2013 and being the SKA precursor at low radio frequencies (Tingay et al. 2013), performed the GLEAM survey 6 (Wayth et al. 2015). The extragalactic GLEAM catalog  covers ∼25000 square degrees, at declinations south of +30 • and Galactic latitudes |b| > 10 • , and excluding some regions such as the Magellanic Clouds. It lists ∼3×10 5 radio sources with 20 separate flux density measurements across the frequency range 72-231 MHz, selected from a time-and frequency-integrated  White et al. (2020a,b) built the G4Jy catalog (see also Wayth et al. 2015), a flux limited sample listing nearly 2000 sources over an area of ∼ 25000 square degrees. The selection criterion on the Galactic latitudes (i.e., Galactic latitudes |b| > 10 • ) was used to lower the contamination of Galactic sources and focus on the extragalactic sky. This sets the basis to extract a southern catalog of extragalactic radio sources: the G4Jy-3CRE, fully equivalent, in terms of radio flux density selection, to the northern 3CR extragalactic sample, that lists powerful radio sources at declinations above -5 • and having a flux density higher than 9 Jy at 178 MHz (Bennett 1962;Spinrad et al. 1985). We remark that the brightest sources at declinations below +30 • and Galactic latitudes |b| > 10 • , including the Orion Nebula, were all masked in the GLEAM extragalactic catalog and thus are not listed in the G4Jy Sample (see White et al. 2020a, and references therein for a list of them and additional details). Sources listed in the G4Jy-3CRE are selected to have (i) Dec. < −5 • and (ii) flux densities at 174 MHz and at 181 MHz, integrated over the GLEAM bands, above the following thresholds: S 174 >8.13 Jy and S 181 >7.85 Jy, respectively 7 . The flux density thresholds adopted here correspond to the 9 Jy limiting sensitivity at ∼178 MHz, the nominal value adopted to prepare the 3CR. Sources close to the Galactic plane (i.e., at Galactic latitudes |b| <10 • ) are excluded since they were not originally listed in the G4Jy catalog (White et al. 2020a). All sources listed in the G4Jy-3CRE catalog are visible from the Southern Hemisphere and lie in a footprint not covered by the 3CR catalog, whereas the handful of targets in common were removed. The final G4Jy-3CRE catalog lists 264 radio sources. At declinations below +20 • there are 106 out of 298 3CR objects thus the new G4Jy-3CRE sample includes about three times the number of powerful radio sources visible with southern telescopes.
After the release of the 3CR, several radio analyses (see e.g., Roger et al. 1973) were carried out to refine the sample, some of them led to the 3CRR release (Laing et al. 1983). The flux density threshold adopted to create the G4Jy-3CRE, even if similar to the nominal value of the 3CR, allowed us to create a comparable statistical sample for radio sources mainly visible from the Southern Hemisphere, with access to the state-of-the-art observing facilities there. For reference the 3CR lists 298 radio sources while the 9 Jy cut used for the G4Jy-3CRE results in a list of 264 objects. However, to make a more precise comparison with the 3CR, we added a flag to all 181 radio sources out of 264 filtered for having radio flux densities above 9.8 Jy 8 at 178 MHz, this threshold has 7 These thresholds are computed assuming a power-law description for the radio spectrum of the G4Jy sources and adopting the spectral index reported in the G4Jy catalog 8 This flux density threshold was computed at 178 MHz extrapolating that between 174 MHz and 181 MHz reported in the G4Jy catalog using the radio spectral index between these two frequencies. It corresponds to that used to create the original 3C catalog taking into account refined intercalibrations.
been computed to account calibration differences with respect to original 3C observations (see e.g., Roger et al. 1973, and references therein).
The analysis of White et al. (2020aWhite et al. ( , 2020b revealed that in the G4Jy catalog 86% of the radio sources appear to be associated with a mid-infrared (mid-IR) counterpart, detected in the all-sky survey performed with the Wide Infrared Survey Explorer (WISE; Wright et al. 2010). In the present analysis we only report WISE images at 3.4µm but we refer to mid-IR counterparts of G4Jy sources considering those listed in the allWISE data release and thus based on the detection in all filters (Cutri et al. 2012(Cutri et al. , 2013. In our selected sample, 225 G4Jy-3CRE sources (i.e., ∼85%) are associated with a mid-IR counterpart detected in the WISE all-sky survey (White et al. 2020b).
Radio positions reported in the G4Jy were then computed using NRAO VLA Sky Survey (NVSS; Condon et al. 1998) and/or Sydney University Molonglo Sky Survey (SUMSS; Mauch et al. 2003) images while data available from the Tata Institute of Fundamental Research (TIFR) Giant Metrewave Radio Telescope (GMRT) Sky Survey (TGSS; Intema et al. 2017) were inspected to obtain a better overview of the radio structure. Thus for sources at declinations above -39.5 • , the position reported in the NVSS catalog was used, while for the remaining ones the one reported in the SUMSS catalog was adopted. Thus, given their relatively low angular resolution, there are sources, having an unusual radio morphology and/or being asymmetric, for which the location of the radio core, and consequently that of their host galaxy, was not clearly identified. For these reasons, as we described below, we expended considerable effort to investigate higher resolution radio observations.
In Figure 1 we show the sky distribution, plotted adopting the Hammer-Aitoff projection, of the 3CR sources in the Northern Hemisphere alongside their equivalent sample of southern celestial objects, G4Jy-3CRE, to highlight their complementary footprints. Sources at Galactic latitudes |b| <10 • are excluded since they were not originally listed in the G4Jy catalog (White et al. 2020a).

Identification flags
We identified optical counterparts of radio nuclei overlaying radio contours on optical images. The same comparison with mid-IR images of the WISE all-sky survey 9 was then crucial to verify both (i) the presence of a mid- IR counterpart and (ii) the robustness of all associations reported in the original G4Jy catalog. As previously stated, we only report in our analysis a comparison between radio maps and mid-IR images at 3.4µm and the mid-IR associated counterpart as in the G4Jy based on detections of the AllWISE survey (Cutri et al. 2012). We initially explored several radio databases to search for radio images of sources listed in the G4Jy-3CRE sample to identify the correct position of their radio cores. At radio frequencies we compared GLEAM images with those available in the databases of the VLA Low- Frequency (Lacy et al. 2020) for sources visible from the Northern Hemisphere while we used mainly the SUMSS 14 for those lying out of footprints of previous surveys. High resolution radio surveys, such as the VLASS, allowed us to obtain a precise measurement of the radio core location.
To achieve our goals we additionally checked the NVAS radio archive, mainly searching for radio images at 1.4 GHz, 5 GHz and 8 GHz, however these were useful only for a handful of sources, allowing us to clearly detect the position of their radio cores. NVAS radio maps, with angular resolution in the range between 0 ′′ .5 and ∼10 ′′ , are available for 124 out of 264 sources listed in the G4Jy-3CRE. Here, host galaxy search and identification are partially biased, being more efficient, towards sources located above declinations of ∼-40 • where JVLA data are available.
For optical images, we retrieved optical images for 178 G4Jy-3CRE sources from the Pan-STARRS 15 and the DES 16 databases, while for the remaining 86 objects, lying outside the footprints of these surveys, we inspected observations/images provided in the red filter of the Digital Sky Survey 17 (DSS). We adopted the following identification flags (IDFs) with a summary of all criteria described below. Association results based on these IDFs are all summarized in Table 2, together with several examples. In Table 3 we report all IDFs assigned to each source listed in the G4Jy-3CRE, together with other parameters. For the IDFs we distinguished the following cases: • IDF=1.0: sources for which the position of the radio core is coincident with the mid-IR source associated in the original G4Jy catalog and it also corresponds to a unique optical source detected in the r-band of the Pan-STARRS/DES catalogs or in the red filter image obtained with the DSS, as shown for G4Jy 312 in Figure 2; • IDF=2.0: sources having an optical counterpart of their radio core. However, on the basis of the mid-IR counterpart, we distinguished: -IDF=2.1: those sources having the radio core associated with a unique optical counterpart, different from the mid-IR one listed in the original G4Jy catalog (IDF=2.1, as reported for G4Jy 524 in Figure 3); -IDF=2.2: those cases for which there is an optical counterpart but lacking an associated mid-IR source (i.e., IDF=2.2, as shown for G4Jy 593 in Figure 3).
According to the analysis performed by White et al. (2020aWhite et al. ( ,2020b) the latter ones are cases for which the identification of the host galaxy was partially or completely limited by the mid-IR data, being either undetected in the AllWISE survey or having its emission affected by a relatively bright nearby object; • IDF=3.0: sources for which the low angular resolution of radio observations did not allow us to uniquely identify the optical counterpart (IDF=3.0, as reported for G4Jy 934 in Figure 4), despite the presence of an associated mid-IR source. These are simply "confused" cases; • IDF=4.0: sources for which there is no optical counterpart, being undetected at the sensitivity limit of the survey data we used to search for it. We also distinguished here several subcategories: -IDF=4.1: those lacking an optical counterpart but with an assigned mid-IR counterpart in the original G4Jy catalog (IDF=4.1, as shown for G4Jy 1846 in Figure 5); -IDF=4.2: those for which there is no optical counterpart but the radio map used allowed us to verify that the previously assigned mid-IR one is incorrect (IDF=4.2, as shown for G4Jy 1057 in Figure 5); -IDF=4.3: those lacking counterparts at both mid-IR and optical frequencies (IDF=4.3, as reported for G4Jy 1587 in Figure 5).
Here we assumed that a source is detected only if it is reported in the catalog corresponding to each survey used in the current analysis, thus having a detection threshold equal to the level of confidence of the survey itself. No associations between radio and mid-IR/optical counterparts were considered reliable if the angular separation between their positions is greater than 5. ′′ 4. This is also supported by the seeing of optical surveys being ∼2 ′′ and by the statistical analysis reported in § 3.
In Figures 2, 3, 4 and 5, we show WISE images collected at 3.4µm with radio contours overlaid. The frequency of the radio map from which radio contours were drawn is reported in the figure together with the intensity of the first level. All radio contours increase in level by a binning factor also reported in the figure. Radio maps obtained through the VLASS, NVSS and SUMSS archives correspond to a nominal frequency of 3 GHz, 1.4 GHz and 843 MHz, respectively. We also show optical images collected from one of the surveys used in our analysis. If the optical image has a label "red filter" it was obtained from the DSS archive, while "r band" marks those retrieved from Pan-STARRS or DES archives. Radio contours are also overlaid on optical images. The red cross, if present, marks the position of the associated mid-IR counterpart according to the G4Jy catalog, while the cyan cross corresponds to the position of brightness-weighted radio centroids reported in the G4Jy (White et al. 2020a,b). The blue dashed circle, whenever present, indicates the position of the optical counterpart identified in our analysis. We remark that the scale of the mid-IR and the optical images are different on purpose. The underlying reasons are: (i) the angular resolution is different and (ii) mid-IR images, reported with a larger field of view, allow us to highlight the large-scale radio structure, while optical images were mainly used to identify counterparts as finding charts.
Sources having IDF=1.0 (namely, those for which the resolution of radio maps allows us to firmly establish the position of the radio core) include also cases for which we do not clearly detect the radio nucleus but the brightness-weighted radio centroid of all radio maps used is coincident with a unique optical counterpart with no nearby companions, within an angular separation of 5 ′′ -10 ′′ . An example is shown in Figure 6, where we report the case of G4Jy 122, a classical FR II radio galaxy at z=0.4, for which the radio morphology and the lack of optical sources within an angular separation of 10 ′′ from the potential optical counterpart allowed us to assigned it an IDF=1.0. We emphasize that for several sources we also used archival radio maps, collected from the NVAS database, that allowed us to unequivocally assign an optical counterpart (i.e., IDF=1.0 rather than 3.0).
Once we identified the optical counterpart we also searched both the latest releases of the Pan-STARRS and the DES catalogs to obtain an estimate of its rband magnitude. In Table 3 we summarize our main results indicating: (i) the G4Jy name; (ii) the name of the mid-IR counterpart associated in the G4Jy catalog using WISE images; both (iii) radio and (iv) optical positions; (v) the counterpart IDF and (vi) the r magnitude of the optical counterpart, as previously mentioned, together with the optical magnitude and the Galactic extinction. Then, we added to the table the estimate of the Galactic extinction 18 A V (Schlegel et al. 1998;Schlafly et al. 2011). The entire Table 3 is reported in the on-line version of the journal while only the first 10 lines are shown here for guidance regarding its form and content. Optical positions are reported from the Pan-STARRS and/or DES catalogs, when available, having all uncertainties lower than ∼0 ′′ .5. We did not measure DSS magnitudes and positional uncertainties since we are already carrying out a photometric survey, mainly using telescopes at Complejo Astronómico El Leoncito (CASLEO), to estimate them for those sources lying out of the Pan-STARRS and DES footprints.
We also inspected other optical images available in the Pan-STARRS/DES searching in the g, i, z and y bands to detect the host galaxy of sources lacking an optical counterpart of their radio core but having an associated mid-IR object. This analysis confirmed previous results with the unique exception of G4Jy 818, for which we identified a marginal detection of an optical counterpart in y-band only.

Optical identifications
We found that for 184 out of a total of 264 sources listed in the G4Jy-3CRE catalog (i.e., 70%) their optical counterpart is spatially coincident with the mid-IR match reported in the G4Jy catalog (White et al. 2020a,b), thus confirming previous results. These are all marked with an IDF=1.0 in our Table 3. In Figure 7 we report the distribution of the angular separation θ ow between the optical and the mid-IR counterparts.
Only 5 out of 264 sources (i.e., ∼2%) show an incorrect association between the optical source we assigned and the mid-IR counterpart reported in the G4Jy catalog, namely: G4Jy 4, G4Jy 524, G4Jy 1197, G4Jy 1302, G4Jy 1854. For these five cases an IDF=2.1 was assigned. Then we identified new optical counterparts for 21 sources out of 264 (i.e., ∼8%, correspond-18 https://irsa.ipac.caltech.edu/applications/DUST/ ing to IDF=2.2) that did not have a mid-IR counterpart assigned in the original G4Jy catalog. This was mainly possible thanks to the use of high resolution radio maps as those retrieved from the VLASS and the NVAS archives. In particular, we found that 4 sources with IDF=2.2, namely: G4Jy 162, G4Jy 730, G4Jy 1301 and G4Jy 1330, have no mid-IR counterparts since they lie close to bright WISE sources and their detection could be contaminated by artifacts. In the case of G4Jy 593, again with IDF=2.2, the detection of its mid-IR counterpart is indeed compromised by the poor angular resolution of WISE images. An additional 12 radio sources (out of 21) with IDF=2.2, marked with a 'u' host flag in the G4Jy catalog due to possible ambiguities related to the complexity of their radio structure and/or the spatial distribution of nearby mid-IR sources (see White et al. 2020a,b, for more details), have mid-IR counterpart, identified thanks to the refined optical analysis presented here. These radio sources are: G4Jy 113, G4Jy 350, G4Jy 530, G4Jy 611, G4Jy 672, G4Jy 680, G4Jy 939, G4Jy 1262, G4Jy 1365, G4Jy 1401, G4Jy 1518, G4Jy 1740, and all their mid-IR associated counterparts are now reported in Table 3. In addition for the two cases of G4Jy 837 and G4Jy 1590, both having IDF=2.2 and both previously labelled with a 'm' host flag in the G4Jy catalog (i.e., identification of their host galaxy limited by the mid-infrared data White et al. 2020b), the optical analysis presented here allowed us to recognize their mid-IR counterpart. In Table 4 we report the WISE name of these 14 mid-IR counterparts identified by refined optical analysis performed here. These are also included in Table 3. We marked the location of these 14 newly associated mid-IR counterparts using a red circle on the WISE image at 3.4µm in the finding charts to distinguish them from those associated in the G4Jy catalog labelled with a red cross. The only two remaining sources: G4Jy 1498 and G4Jy 1532 are those clearly lacking a mid-IR counterpart, being undetected in the WISE images. In Figure 7 we also report the distribution of the angular separation θ ow between the optical and the mid-IR counterparts including those with IDF=2.2 that have been assigned thanks to the optical analysis presented here.
We found that 14 out of the total 264 sources (i.e., ∼6%) visually inspected are flagged as "confused" since we were not able to identify a unique optical counterpart, and thus require a deeper investigation and eventually additional follow up observations. If these sources are high excitation radio galaxies (HERGs; Hine & Longair 1979) or quasars, due to their relatively low source density (i.e., number of sources per square degree in the sky), X-ray or optical spectroscopic observations could Table 3. Full list of the G4Jy-3CRE catalog (2) the name of the assigned mid-IR counterpart, detected in WISE, as in the G4Jy catalog; col. (3,4) right ascension (R.A.) and declination (Dec.), in J2000 Equinox, of the brightness-weighted radio centroid collected from the G4Jy catalog; col. (5,6) same as previous two columns but measured from the centroid of the optical counterpart in the Pan-STARRS, DES and DSS images; col. (7) identification flag (IDF) adopted in our analysis (see § 3 for all details); col. (8) the redshift value reported in the literature, where question marks highlight those with uncertain estimates; col. (9) r-band magnitude from the Pan-STARRS and DES counterparts; col. (10) Galactic extinction; col. (11) The check mark indicates if the source belong to the subsample selected with radio flux density above 9.8 Jy at ∼178 MHz (see § 2 for more details). The entire table is reported in the on-line version of the journal while only the first 10 lines are shown here for guidance regarding its form and content. (2) the name of the mid-IR counterpart, detected in WISE, assigned thanks to the optical analysis presented here; col.
(3) the angular separation θow between the position of the optical and mid-IR counterpart, assigned thanks to the analysis performed here. .4µm image available thanks to the WISE all-sky survey, with radio contours overlaid in black. The frequency of the radio map from which the radio contours were drawn is reported in the figure together with the intensity of the first level and the binning factor. The symbol x3, reported in the image indicates that radio contours, starting at 1 mJy/beam level, increase by a factor of three. Radio maps from VLASS, NVSS and SUMSS archives correspond to a nominal frequency of 3 GHz, 1.4 GHz and 843 MHz, respectively. (Right panel) The optical image collected from one of the optical surveys used in our analysis. If, below the source name, the label "red filter" is reported then the optical image is collected from the DSS archive while when it is written "r band", as in this case, optical images were retrieved from Pan-STARRS or DES databases. Radio contours are also overlaid on the optical image. The red cross, if present, marks the position of the mid-IR counterpart associated in the G4Jy catalog while the cyan cross corresponds to the position of the brightness-weighted radio centroid of the G4Jy catalog (White et al. 2020a,b). The blue dashed circle, if present, indicates the position of the optical counterpart identified from our analysis. Both these images are an example of a source, i.e., G4Jy 312, having an IDF=1.0, for which our analysis revealed that the mid-IR counterpart associated in the G4Jy catalog corresponds to the optical source lying at the same position of the radio nucleus in the high angular resolution radio map used here. The blue open square or the blue X symbol, if present in the left panel, mark the location of the closest radio source belonging to the Parkes radio catalog (PKSCAT90, Bolton et al. 1979) or to the Molonglo Reference Catalog of Radio Sources (MRC, Large et al. 1981), respectively (see also Appendix C for additional details about radio cross identifications). The complete figure set (264 images), showing finding charts for all sources in the G4Jy catalog, is available in the online journal.  Figure 2 but for the case of G4Jy 524, a radio source for which the mid-IR counterpart identified in the G4Jy catalog is different from the optical one that corresponds to the position of the radio core detected in the VLASS radio map at 3 GHz. Sources showing the same behavior were marked with an IDF=2.1 in our analysis. (Bottom panels) Same as Figure 2 but for G4Jy 593. This radio source, having IDF=2.2, is an example of those cases for which there is a clear optical counterpart but they lack an assigned mid-IR source in the G4Jy catalog.  Figure 2 for G4Jy 934. The lack of a high angular resolution radio map, in this case the one from which radio contours are computed was collected from the SUMSS archive, prevented us to clearly identify the host galaxy of the radio source. These cases are flagged as "confused" in our analysis and have IDF=3.0 reveal the position of their counterpart and thus iden-tify their host galaxies. However the precise location of As shown in the right panel, the optical counterpart of the radio nucleus is too faint to be detected in the images available from Pan-STARRS, DES and the DSS databases, however it is clear from the left panel that the radio source has an infrared counterpart. These cases are indicated with IDF=4.1. (Central panels) Same images for G4Jy 1057 but in this case, despite the lack of a plausible optical counterpart the radio map available indicates that the assigned mid-IR counterpart appears to be incorrect, thus marked in our analysis with IDF=4.2. (Bottom panels) Same as above panels but for sources as G4Jy 1587 that lack both a mid-IR and an optical counterpart of its radio core. G4Jy 1587 has IDF=4.3. These are radio sources for which we also visually inspected other optical images available in the Pan-STARRS and in the DES databases in the g, i, z and y bands searching for signatures of their host galaxies.
their radio core, necessary to identify the host galaxy position, can be only achieved using higher resolution radio maps, in particular when they are hosted in elliptical galaxies with weak optical emission lines, as often Figure 6. Same as right panel of Figure 3 for G4Jy 122. The dashed blue circle, having 10 ′′ radius, is centered on the location of the potential optical counterpart identified here since (i) it lies between the two radio lobes of G4Jy 122 showing a classical FR II radio morphology and (ii) there are no other optical sources within this area. Figure 7. The black histogram shows the distribution of the angular separation θow between the mid-IR counterpart, assigned in the G4Jy catalog, and the optical one for all those G4Jy-3CRE radio sources having IDF=1.0, that, according to our analysis, implies that both counterparts are coincident. Two dashed black lines mark the location for θow equal to 0 ′′ .2, and 3 ′′ , respectively, while the red one corresponds to 1 ′′ . The red histogram includes also those radio sources with IDF=2.2 for which our optical analysis helped to identify the mid-IR counterpart (see § 3.2 for more details). As a comparison, we also report here the typical range of positional uncertainties of the NVSS and the SUMSS catalogs (Vollmer et al. 2005) mainly used in the G4Jy catalog to compute brightness-weighted radio centroids.
occurs in low excitation radio galaxies (LERGs; Hine & Longair 1979). For all of them the value of IDF=3.0 is reported in Table 3.
There are 19 out of 264 sources that lack an optical counterpart of the radio core but have at least a mid-IR source associated with it (i.e., ∼7% of the whole G4Jy-3CRE catalog), thus being marked with IDF=4.1, as for G4Jy 1846 shown in Figure 5. For 4 more objects out of 264 (i.e., ∼2%), the associated mid-IR source does not appear to be correct thus having IDF=4.2 (see G4Jy 1057 in Figure 5, as well as G4Jy 183, G4Jy 1551, G4Jy 1782). Lastly, 17 remaining sources (∼6% of the total) have no optical and no mid-IR counterpart associated with their radio core and are indicated with IDF=4.3. Two of those radio sources labelled with IDF=4.3 are: G4Jy 77 and G4Jy 1605, the former is a radio phoenix of the galaxy cluster Abell 85 (see e.g., Bagchi et al. 1998;Kempner et al. 2004;Ichinohe et al. 2015) while the latter is the radio relic of Abell 3667 (see e.g., Johnston-Hollitt et al. 2008;Owers et al. 2009), as also discussed, with more details, in Appendix A.
All sources with IDF=4.1 or IDF=4.2 or IDF=4.3 have no detection of their radio cores in radio maps due to their poor angular resolution, with the only exceptions of four objects, namely: G4Jy 854, G4Jy 1010, G4Jy 1136 and G4Jy 1830. In these four cases the detection of the mid-IR counterpart could be contaminated by artifacts, due to the presence of bright WISE sources, for G4Jy 854 and G4Jy 1136, while for G4Jy 1010 and G4Jy 1830 no clear mid-IR emission is reported in the WISE all-sky catalog.
All numbers reported in this section are also summarized in Table 2.
Finally, there are 203 out of 264 sources listed in the G4Jy-3CRE sample (i.e.,∼77%) (i) having the optical counterpart coincident with the mid-IR one (i.e., IDF=1.0) or (ii) lacking optical counterpart of their radio core but having a mid-IR counterpart (i.e., IDF=4.1). Thus in Appendix C we computed the probability of spurious associations between mid-IR sources listed in the WISE all-sky survey and the full G4Jy radio catalog, as originally performed to assign mid-IR counterparts, and we found a good agreement with the refined analysis presented here.

LITERATURE RESULTS
Once we identified optical counterparts of radio sources we performed a literature search to investigate the availability of redshifts. Results of this literature search, including uncertain z estimates are also reported in each figure when available and in the finding charts. We also compared our final catalog with the several radio catalogs obtained from radio surveys (see Appendix B for a brief overview of all radio catalogs used here). All these identifications derived from these radio cross-matches are reported in Appendix B to simplify searches in astronomical databases as NED and SIMBAD.
We found that for a total of 157 sources out of 264 (i.e., 59% of the whole sample) there is a spectroscopic redshift measurement already available in the literature, twelve of them considered uncertain and thus labelled with a question mark for a total of 145 radio sources with firm z estimates. Our search was carried out also using the NASA Extragalactic Database (NED) 19 and the SIMBAD Astronomical Database 20 . According to previous analyses carried out during past follow up spectroscopic campaigns Peña-Herazo et al. 2020, 2022Kosiba et al. 2022), we adopted the same conservative criteria and we only considered confident redshift measurements reported in the literature those for which we could verify (i) a published image of the optical spectrum, or (ii) there is a description of it with emission and/or absorption lines clearly reported in a table or in the manuscript. In this way, we marked redshifts z we could not verify with a question mark for 12 sources out 157.
Moreover we did not consider photometric redshifts since we are currently carrying out optical spectroscopic observations, based on the analysis presented here and more redshifts will be presented in a forthcoming paper (García-Pérez et al. 2022 in prep.). In some cases, and only for sources belonging to the Molonglo Southern 4 Jy sample (MS4 Burgess & Hunstead 2006a,b), we reported values of photometric redshift estimates in Appendix A.
A large fraction of the spectra we found are available in the Data Release 3 of the Six-degree Field Galaxy Survey (6dFGS) Database 21  in addition to the spectroscopic observations of the equatorial sample of powerful radio galaxies . In a few cases we also used other databases such as the On-Line Inventory of Extragalactic X-ray Jets 22 (XJET; Massaro et al. 2011a) and references reported therein.
According to our literature search we found that 131 sources out of 157 having IDF=1.0 already have an available redshift measurement. For all these cases, with only two exceptions, namely: G4Jy 1158 and G4Jy 1225, we found these estimates reliable (i.e., no question mark reported). The underlying choice of considering these estimates, at a first look correct, is also based on the relatively low sky density (i.e., number of sources per square degree) of radio galaxies and QSOs, thus the probability of having more than a single radio-loud AGN close 19 http://ned.ipac.caltech.edu 20 http://simbad.u-strasbg.fr/simbad/ 21 http://www-wfau.roe.ac.uk/6dFGS/index.html 22 https://hea-www.harvard.edu/XJET/ Figure 8. The redshift distribution obtained from our literature search for those 157 radio sources listed in the G4Jy-3CRE catalog with a z estimate. This red histogram includes also 12 radio sources with uncertain values in comparison with that of radio sources with firm z measurements shown in black.
to the radio position within a circle of a few arcseconds radius is extremely low. This also proves the importance of collecting spectroscopic information that could confirm optical counterparts associated in the present analysis (see e.g., results of spectroscopic campaigns carried out on radio loud AGNs lying within the positional uncertainty regions of gamma-ray sources Massaro et al. 2015c;Landoni et al. 2015;Ricci et al. 2015;. In Figure 8 we show the comparison between the redshift distribution for 157 radio sources, out of 264 listed in the G4Jy-3CRE catalog, having a z estimates, and that excluding those 12 with uncertain z.
For those sources with IDF=2.1 or IDF=2.2 we found a redshift estimate for 17 cases out of 26, namely 3 objects with IDF=2.1 and 14 with IDF=2.2. In 2 out of the 3 cases with IDF=2.1, we report z estimates in Table 3 adding the question mark to indicate that these measurements are uncertain given the lack of positional data about the target, while this situation does not occur for all radio sources with IDF=2.2.
For only one case out of the "confused" sources (i.e., those labelled with IDF=3.0), namely: G4Jy 1626, that could potentially reside in galaxy-rich large-scale environments (i.e., groups or cluster of galaxies), we also report a z estimate with a question mark while the remaining 13 sources all lack a z measurement.
Finally, we found that for 6 sources (out of 19) marked with IDF=4.1, for which we were not able to identify the optical counterpart in the archival images, there is a z estimate in the literature, but since we could not verify which sources were targeted, they are all flagged as uncertain, with the only exception: G4Jy 417 for which we found the finding chart in the literature since it belongs to the 2 Jy sample (Wall & Peacock 1985;Morganti et al. 1997), even if we could not detect it in the DSS archival images used here. Then all 4 radio sources with IDF=4.2 have no z measurement and the same occurs for 15 out of 17 of those labelled with IDF=4.3. Both remaining 2 radio sources with IDF=4.3 have the z estimate marked with a question mark thus considered uncertain.

SUMMARY, CONCLUSIONS AND FUTURE PERSPECTIVES
We present the G4Jy-3CRE catalog extracted from the G4Jy catalog (White et al. 2020a,b) and based on the low radio frequency observations of the MWA as part of the GLEAM survey. The G4Jy-3CRE catalog lists 264 sources with 9 Jy limiting sensitivity at ∼178 MHz, which is the same as the nominal threshold for the 3CR, but including only targets at Dec. < −5 • , having Galactic latitudes |b| > 10 • , all lying in the footprint not covered by the 3CR.
Thanks to a huge amount of effort carried out on the G4Jy catalog, a large fraction of radio sources listed therein (i.e., ∼85%) are associated with a mid-IR counterpart (being their host galaxy White et al. 2020b) detected in the WISE all-sky survey (Wright et al. 2010).
Here we present a refined analysis, restricted only to the G4Jy-3CRE sample, aimed at locating optical counterparts of their host galaxies. Thanks to recent high angular resolution radio observations, such as those available in the VLASS, we can obtain a precise estimate of the position of their radio cores, for a significant fraction of the G4Jy-3CRE sources (i.e., 207 out of 264, nearly 78%). This allowed us to (i) improve the localization of the host galaxy in the optical images of Pan-STARRS, DES and DSS archives and then (ii) search the literature for sources for which a redshift estimate is already available.
Results achieved by our inspection of archival radio, infrared and optical images can be summarized as follows: 1. We found that for 184 out of 264 G4Jy-3CRE sources the optical counterpart associated in the present analysis is coincident with the mid-IR counterpart reported in the G4Jy catalog, confirming the robustness of the previous analysis (White et al. 2020a,b).
2. There are 26 G4Jy-3CRE sources for which the optical counterpart we identified is different from the, previously assigned, mid-IR one. In particular for 21 of them there is not a mid-IR source assigned/associated in the G4Jy catalog. For 14 out of these 21 radio sources we have been able to assign a mid-IR counterpart on the basis of our optical analysis.
3. For 14 sources the poor angular resolution of radio maps available and the presence of several optical sources around the position of their radio cores did not allow us to confirm their host galaxies.
4. There are an additional 40 sources with no optical counterpart of their radio core. For 4 cases the mid-IR counterpart associated in the G4Jy catalog does not appear to be correct (IDF=4.2), while for 17 sources there is no mid-IR counterpart detected at the location of the radio cores, as reported in the G4Jy catalog (i.e., IDF=4.3). The remaining 19 sources are only detected in the mid-IR images (IDF=4.1).
According to our analysis, radio sources having the identification flags: IDF=1.0 or IDF=4.1 are ∼77% (i.e., 203 out of 264) of the whole G4Jy-3CRE sample and correspond to the more reliable mid-IR associations. This is in agreement with the statistical test we used to compute the expected number of spurious associations when matching the G4Jy with the AllWISE catalog.
Given the identified location of the host galaxies for a large fraction of the G4Jy-3CRE sources, we also checked the literature to search for possibile redshift estimates. Adopting a conservative criterion we found a total of 157 spectra, and 145 of them appear to provide a firm z estimate. Moreover, 129 are reliable since their optical counterpart coincides with that associated at mid-IR frequencies and are not labelled as possible sources having an uncertain z measurement.
Finally, we conclude by highlighting future perspectives on the potential use of the G4Jy-3CRE sample. Several proposals were already submitted to collect optical spectroscopic information for all sources listed in the G4Jy-3CRE catalog, and in a forthcoming paper of this series, part of these observations, already acquired, will be presented (García-Pérez et al. 2022 in prep.). A dedicated paper presenting the X-ray analysis, based on Swift data collected for ∼80 G4Jy-3CRE sources, is also in preparation, to highlight the potential use of Xray snapshot observations to refine the search for optical counterparts and host galaxies (Massaro et al. 2023 in prep.).
We thank the anonymous referee for useful and valuable comments that led to improvements in the paper.
We wish to dedicate this paper to D. E. Harris and R. W. Hunstead. Their insight, passion and contributions to radio astronomy are an inspiration.
A This research has made use of the NASA/IPAC Extragalactic Database (NED), which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. This research has made use of the SIMBAD database, operated at CDS, Strasbourg, France (Wenger et al. 2000).
This research has made use of the CIRADA cutout service at URL cutouts.cirada.ca, operated by the Canadian Initiative for Radio Astronomy Data Analysis (CIRADA We acknowledge the efforts of the staff of the Anglo-Australian Observatory, who have undertaken the observations and developed the 6dF instrument. SAOImageDS9 development has been made possible by funding from the Chandra X-ray Science Center (CXC), the High Energy Astrophysics Science Archive Center (HEASARC) and the JWST Mission office at Space Telescope Science Institute (Joye & Mandel 2003). This research has made use of data obtained from the high-energy Astrophysics Science Archive Research Center (HEASARC) provided by NASA's Goddard Space Flight Center. We acknowledge the use of NASA's SkyView facility (http://skyview.gsfc.nasa.gov) located at NASA Goddard Space Flight Center. This dataset or service is made available by the Infrared Science Archive (IRSA) at IPAC, which is operated by the California Institute of Technology under contract with the National Aeronautics and Space Administration. TOPCAT and STILTS astronomical software (Taylor 2005) were used for the preparation and manipulation of the tabular data and the images. The analysis is partially based on the OCCAM computing facility hosted by C3S 24 at UniTO (Aldinucci et al. 2017).

A. NOTES ON INDIVIDUAL SOURCES
Here we provide additional information, in addition to that retrievable from the main table. This list will be then updated in all forthcoming publications. Radio sources not listed below are those for which relevant information were not found in the literature. For the FR I and FR II radio classification of radio galaxies we mainly considered the information reported in the literature as well as the radio morphology observed in high resolution radio maps when available, adopting the same criteria of Capetti et al. (2017a,b). In several sources we also reported additional names provided in the literature, but full information about radio cross-identifications can be retrieved in Appendix B.
G4Jy 4: Hunstead et al (1978) reported the results of optical spectroscopic observations, collected with the 3.9-m Anglo-Australian Telescope, for 22 QSOs and emission-line galaxies associated with southern radio sources detected with Molonglo telescope at 408 MHz, and having flux densities above 0.95 Jy. The first source in their sample is 0000-117, a QSO at z=1.465, having the optical position reported therein (00:03:22.12, -17:27:14.1 in J2000), that is a 3 ′′ .5 angular separation from the location of the counterpart assigned in our analysis with G4Jy 4. Despite the one-side radio structure of G4Jy 4, typical of core dominated QSOs, the source identified by Hunstead et al (1978) appears to be more consistent with the relatively brighter object located in the southern-eastern direction with respect to that coincident with the radio core of G4Jy 4, thus having a z estimate labelled with a question mark. It could be also associated to the radio source PKS 0000-17, but the lack of optical information prevented us to claim this association.
G4Jy 9: a nearby radio source at z=0.2912 ) and correspondent to PKS 0003-56. This source also belongs to the MS4 radio catalog (Burgess & Hunstead 2006a,b) and has a radio counterpart at 20 GHz (a.k.a. AT20G J000558-562828; Murphy et al. 2010), optically identified as a normal galaxy (Mahony et al. 2011) in agreement with the optical image reported in our analysis.
G4Jy 12: also known as PKS 0003-83 and with a photometric redshift estimate of z=0.32 (Burgess & Hunstead 2006b). It is a classical double source with unresolved components. The association provided by our analysis is consistent with that performed by Jones & McAdam (1992) reporting an optical magnitude of 19.0 mag and a nearby fainter companion galaxy of 20.0 mag.
G4Jy 20: (a.k.a. PKS 0008-44) has a photometric redshift estimate of z=1.0 reported in the MS4 optical identification analysis (Burgess & Hunstead 2006b). In this case the lack of high resolution radio maps prevented us to claim that the optical counterpart is the one assigned in the MS4 catalog.
G4Jy 26: a celestial object belonging to several catalogs of southern radio sources and thus known also as PKS 0012-38 and PMN J0015-3804 Gregory et al. 1994). It has a tentative photometric redshift estimate reported in the literature of z=0.57 (Burgess & Hunstead 2006b), however having an IDF=3.0 if the optical counterpart associated in the MS4 sample is the correct one.
G4Jy 33: is also known as 3C 8, PKS 0016-12 and MRC 0016-129. This source belongs to the original 3C catalog (Edge et al. 1959) but not to its revised version . This radio source, classified as a high redshift radio galaxy, lies in a galaxy-rich large-scale environment (Wylezalek et al. 2013) and it has a redshift estimate of z=1.589 ).
G4Jy 43: is a classical lobe dominated radio source, associated with PKS 0020-25 since its large-scale radio structure resembles that visible at 4.8 GHz (Kapahi et al. 1998) and its optical identification  corresponds to the WISE source associated in the G4Jy being a radio galaxy at redshift z=0.35. Morganti et al. 2021a), hosted by an early-type galaxy (Ramos . G4Jy 48 was also observed during the X-ray survey of the 2 Jy sample (see Wall & Peacock 1985), showing an X-ray spectrum with a dominant jet component and a low intrinsic absorption (Mingo et al. 2014). Morganti et al. (2021a) recently observed this radio source using ALMA and discovered a very extended distribution of molecular gas revealing that, already on galaxy scales, the impact of the AGN is not limited to outflows. Given its high star formation rate, G4Jy 48 lies in the region occupied by the star forming galaxies on the on the SFR-M * plane (see e.g., Bernhard et al. 2021).
G4Jy 77: is an extended radio source located at ∼6 ′ from the X-ray position of the cool core galaxy cluster Abell 85 (see e.g., Durret et al. 2005;Ichinohe et al. 2015, for details about its X-ray emission). At the redshift of the galaxy cluster (i.e., z=0.0557 ;Abell 1958;Pislar et al. 1997;Oegerle & Hill 2001) the kpc scale is ∼1.1 kpc/arcsec, thus the distance between G4Jy 77 and Abell 85 is ∼360 kpc. It is a well-known radio phoenix (Kempner et al. 2004), example of fossil plasma present in galaxy clusters due to past AGN activity, with a very steep radio spectrum (Bagchi et al. 1998;Slee et al. 2001;Rahaman et al. 2022(@).
G4Jy 85:, belonging to the 2 Jy sample (a.k.a. PKS J0043-42; see e.g., Wall & Peacock 1985;Morganti et al. 1997), is LERG (Hine & Longair 1979) at z=0.0526 (Whiteoak 1972;Tadhunter et al. 1993) with a very extended radio structure ) and a classical FR II radio morphology. It appears to be located at the centre of a group/cluster of galaxies being also surrounded by a diffuse halo (Ramos . It also shows the presence of a bridge detected at infrared and optical frequencies related to the interaction with a nearby companion galaxy (see also Inskip et al. 2010). Both hotspots are detected in the soft X-rays together with a relative faint extended emission of the ICM .
G4Jy 86: is the nearby star forming galaxy NGC 253 (a.k.a. Sculptor Galaxy; see e.g., Hoopes et al. 1996) with a z=0.00081 (Springob et al. 2005). More details about this association are reported in the G4Jy catalog.
G4Jy 93: (a.k.a. PKS 0049-43) is a radio source with a tentative photometric redshift estimate of z=0.39 (Burgess & Hunstead 2006b) also listed in the catalog of γ-ray blazar candidates  with mid-IR colors similar to FSRQs.
G4Jy 108: is a radio galaxy with a typical FR II radio structure at z=1.019 , also associated with PKS 0056-17.
G4Jy 113: is a "retired" radio source (López-Cobá et al. 2020) with a z=0.0564, and located in the galaxy cluster Abell 133 . McDonald et al. (2010) observed the presence of a thin Hα filament toward the northeast, extending ∼25 kpc, co-spatial with an X-ray filament, and not consistent with a buoyant radio bubble. Radio and X-ray observations of the radio relic revealed that the relic lobe is energized by the central cD galaxy associated with G4Jy 113, rather than by a shock generated in the radio relic (Fujita et al. 2002).
G4Jy 120: is a lobe dominated radio source. The SUMSS radio position marks the location of G4Jy 120 at an angular separation of ∼17 ′′ from PKS 0103-45, a value too large to claim this association.
G4Jy 122: is a classical double radio galaxy at z=0.4 , optically classified as a narrow line radio galaxy (see also Ramos Almeida et al. 2011) and showing a prominent Fe Kα emission line in its XMM-Newton observation (Mingo et al. 2014). This radio source is known as PKS 0105-16 and it is listed in the original 3C catalog as 3C 32 (Edge et al. 1959) but not in the 3CR revised version . Archival optical images seems to indicate that the host galaxy of G4Jy 122 appears to be connected with an early-type galaxy of similar brightness located at ∼70 kpc in the north-western direction (Ramos . G4Jy 129: is potentially associated with the radio source PKS 0110-69 (a.k.a. AT20G J011143-690016; Murphy et al. 2010) due to the proximity to the intensity peak of the SUMSS radio map at 843 MHz.
G4Jy 133: according to our analysis is labelled with an IDF=3.0, however if follow up observations will confirm its WISE association reported in the G4Jy the radio source could be associated with PKS 0114-47 (White et al. 2020b), a giant radio galaxy (Jones & McAdam 1992) with a z=0.146 ).
G4Jy 136: (a.k.a. PKS 0114-21) is a Compact Steep Spectrum (CSS) radio source with two hotspots embedded in a dense gaseous environment (Mantovani et al. 1994) with a z=1.4153 (De Breuck et al. 2010;Seymour et al. 2007), with mid-IR emission dominated by a stellar continuum and several rest-frame optical emission lines detected in its spectrum (Nesvadba et al. 2017).
G4Jy 143: is associated with 3C 38 (a.k.a. PKS 0117-15), again not listed in the revised release 3CR , and it is classified as a radio galaxy at z=0.565 , having a clear FR II morphology and showing high ionization emission lines detected in the optical spectrum. There is a nearby optical source, potentially a companion galaxy, however, due to the lack of spectroscopic information, we cannot claim that this is the case of a galaxy pair.
G4Jy 145: is a QSO also at mid-IR frequencies , originally located at z=0.837 . More recent ultraviolet observations give a redshift measurement of z=0.834 (Monroe et al. 2016).
G4Jy 157: belongs to the sample of Best et al. (1999) and is classified as a radio galaxy at z=0.372 showing a typical FR II radio morphology.
G4Jy 162: has a clear optical counterpart, while the lack of a mid-IR associated source is mainly due to the artifacts present in the WISE image created by a nearby bright star located in the south-western direction. In the literature G4Jy 162 is classified as a radio galaxy with a z=2.34665 Nesvadba et al. 2017) being the most distant source listed in the G4Jy-3CRE sample to date. G4Jy 162 also shows a fairly regular emission line region morphology, albeit an increasing of the [OIII] surface brightness towards the northern side, and roughly aligned with the radio axis in the north-western side (Nesvadba et al. 2017).
G4Jy 168: is a classical double lobed radio source associated with PKS 0129-073 (a.k.a. MRC 0129-073 and PMN J0131-0703). There are three sources detected in the optical image that could be its potential counterpart, but in the WISE image only the brightest one is detected. This prevents us to label this source with an IDF different from 3.0.
G4Jy 171: is the nearby lenticular galaxy NGC 612 (Ekers et al. 1978) located at z=0.02977 (see e.g., Menzies et al. 1989;da Costa et al. 1991), associated with the powerful radio source PKS 0131-36 and showing a wide double radio structure, with the axis perpendicular to the disk of the galaxy. G4Jy 171 also shows an unusual absorption in its optical spectrum mainly due to its host galaxy structure with gas and dust in its disk ).
G4Jy 192: is a double radio galaxy with a redshift z=0.41, evaluated comparing our finding chart with that reported in the literature . Three nearby companion optical galaxies appear to lie close to the host galaxy of G4Jy 192.
G4Jy 213: is a radio source at z=0.67680 Croom et al. 2004) with a mid-IR counterpart showing WISE colors similar to those of FSRQs , and also tentatively classified as a QSO on the basis of its optical properties .
G4Jy 219: is classified as a radio galaxy with a photometric redshift estimate of z=0.45 belonging to the MS4 sample (Burgess & Hunstead 2006b). In our analysis we used the radio map at 14.9 GHz to clearly identify the optical counterpart.
G4Jy 227: comparing the VLASS observation and the optical finding chart used in our analysis to those available in the literature at nominal frequency of 5 GHz (Reid et al. 1999), we confirmed the location of the host galaxy. G4Jy 227 shows mid-IR colors typical of FSRQs (D'Abrusco et al. 2019) but does not have spectroscopic information.
G4Jy 241: shows a flat radio spectrum since it was selected as part of the Combined Radio All-Sky Targeted Eight GHz Survey (a.k.a. CRATES J021645.12-474908.9; Healey et al. 2007). According to the literature this is a giant radio galaxy (∼1 Mpc size; Christiansen et al. 1977) hosted in an elliptical galaxy at z=0.06427 that is harbored in a galaxy-rich large-scale environment Zirbel 1997;Jones et al. 2009) and located in the direction of the Abell S 239 galaxy cluster ) that lies at similar distance, being at z=0.0635 (Abell 1958;. G4Jy 247: with no associated counterparts in the MRC, PKSCAT and PMN samples, has a counterpart at 20 GHz (AT20G J021902-362607; Murphy et al. 2010) and it also appears to be hosted in an elliptical galaxy (see also Mahony et al. 2011) clearly visible in both the mid-IR and in the optical images with a z=0.48881 ).
G4Jy 249: is associated with PKS 0219-706 (a.k.a. MRC 0219-706 and PMN J0220-7022). This source was tentatively associated with an elliptical galaxy having a photometric redshift estimate of z=0.4 belonging to the MS4 sample (Burgess & Hunstead 2006b), but according to our analysis, it has an IDF=3.0 since there are too many sources close to radio intensity peak that prevent us to claim the location of the host galaxy.
G4Jy 257: as the previous source G4Jy 249, is a radio galaxy with several counterparts at radio frequencies and a photometric redshift estimate of z =1.27 belonging to the MS4 sample (Burgess & Hunstead 2006b).
G4Jy 260: is a lobe dominated radio QSO with a z=0.23224 Jones et al. 2009) with mid-IR colors similar to those of γ-ray emitting QSOs (D'Abrusco et al. 2019), and with a hard X-ray counterpart Baumgartner et al. 2013;Koss et al. 2017).
G4Jy 280: is also associated with PKS 0235-19 and is classified as a broad line radio galaxy with a classical FR II radio structure (Ramos  at z=0.62 Best et al. 1999), with mid-IR colors of γ-ray blazars . Despite the fact that the radio core is not detected in the high resolution JVLA radio map at 5 GHz (Reid et al. 1999), its radio morphology confirms our assigned optical counterparts and association.
G4Jy 293: is also associated with the MS4 radio galaxy MRC 0245-558 with a photometric redshift estimate of z=0.82 (Burgess & Hunstead 2006b), however the presence of several nearby companions, coupled with the lack of spectroscopic information, prevents us to positively assign an optical counterpart.
G4Jy 312: is associated with PKS 0254-23 and it is classified as a radio galaxy located at z =0.509  showing a classical FR II radio morphology (see also Kapahi et al. 1998;Best et al. 1999).
G4Jy 326: is a typical FR II radio galaxy with a z=0.268 ).
G4Jy 347: is a 2.5 Mpc giant radio galaxy (a.k.a.MRC 0319-454, PMN J0321-4510 and MSH 03-43; see e.g., Burgess & Hunstead 2006b;Malarecki et al. 2015, for a recent analysis) located within a galaxy filament of the Horologium-Reticulum supercluster (Fleenor et al. 2005). The host galaxy (Bryant & Hunstead 2000) is located close to the north-eastern radio lobe, the only part of its radio structure reported in our finding chart, and it is associated with the optical source ESO 248-G-10 having a weak counterpart at 20 GHz (Saripalli et al. 1994). The redshift of the host galaxy is z =0.0622 (Safouris et al. 2009) and its radio structure in the north-eastern side goes through an environment having higher galaxy density than the southern radio structure, indicating the presence of a surrounding group of galaxies. There are also several galaxy clusters in its vicinity (Fleenor et al. 2006), namely: S0345, Abell 3111, Abell 3112 and APMCC369 at z=0.071, 0.078, 0.075 and 0.075, respectively.
G4Jy 350: is a large radio galaxy with the south-eastern radio lobe located at the same position of PKS 0352-88 (a.k.a. MRC 0352-884 and SUMSS J032359-881618), at ∼25 ′′ from the G4Jy position, as marked in the finding chart. A recent MeerKat observation revealed that the position of the mid-IR counterpart is consistent with that of WISE J032259.32-881600.4 (Sejake et al. 2022 subm.).
G4Jy 381: is a FR II radio galaxy with a redshift estimate of z=0.0535 (Scarpa et al. 1996;Drinkwater et al. 2001), in the foreground of the galaxy cluster Abell 3165 (Abell 1958; but not related to it. G4Jy 386: is also known as PKS 0349-27 and PMN J0351-2744 and it was already detected as extended at 408 MHz (Schilizzi & McAdam 1975;Schilizzi 1975) as well as at higher frequencies (Reid et al. 1999). It was then optically identified with the same counterpart associated in both the G4Jy at mid-IR frequencies and in our analysis being (Bolton & Ekers 1965) classified as a radio galaxy at z =0.6569 (Searle & Bolton 1968;Jones et al. 2009). This radio source shows extended emission in the optical band with remarkable features, including an extended narrow line region and bridges connecting G4Jy 386 to two companion galaxies Ramos Almeida et al. 2013). These bridges are interpreted as due to tidal interaction with neighbor galaxies and/or mergers (Danziger et al. 1984;Tadhunter et al. 1989). X-ray emission was also detected between the lobes and, with an offset of a few arcsec, on the location of the northern hotspot .
G4Jy 392: is a QSO (a.k.a. 3C 94) showing a double radio structure and with a z=0.96354 (Lynds 1967;Best et al. 1999;Ahn et al. 2012), recently included in the list of giant radio quasars having the size of the extended radio emission larger than 0.7 Mpc .
G4Jy 415: is one of the most luminous QSO at z < 1 (see e.g., Punsly et al. 2016), optically identified by Hunstead (1971b) with a measured z =0.5731 (Kinman & Burbidge 1967;Bechtold et al. 2002;Decarli et al. 2010;Johnson et al. 2018). G4Jy 415 (a.k.a. PKS 0405-12) shows X-ray emission arising from the northern hotspot (Sambruna et al. 2004), thus being included in the XJET database (Massaro et al. 2011a), and it is also listed in the Roma-BZCAT (Massaro et al. 2015b) classified as blazar of uncertain type (i.e., BZU J0407-1211). Recent MUSE observations revealed the presence of six spatially extended line-emitting nebulae in the galaxy group where it is harbored, suggesting a connection between large-scale gas streams and the nuclear activity (Johnson et al. 2018). It also shows the detection of a narrow filament extending toward the QSO consistent with a cool intragroup medium filament similar to those occurring in cool-core galaxy clusters (see e.g., McDonald et al. 2010).
G4Jy 417: does not have an optical counterpart in the DSS image we retrieved from the archive, but it was identified with the same mid-IR counterpart listed in the G4Jy catalog in the literature (Hunstead et al. 1971b;Alvarez et al. 1993). However, we did not report any optical position and/or magnitude and we assigned it an IDF=4.1. G4Jy 417 (a.k.a. PKS 0409-752) also belongs to the MS4 catalog and it is classified as a narrow line radio galaxy with a typical FR II radio morphology at z =0.694 (Alvarez et al. 1993;Tadhunter et al. 1993;di Serego et al. 1994). It is harbored in a galaxy-rich environment (Ramos  and shows evidence for a young stellar population ) and a far infrared excess (Dicken et al. 2009b).
G4Jy 427: has the same optical counterpart associated in the MS4 catalog (Burgess & Hunstead 2006b) where a photometric redshift estimate of z=0.42 is also reported. In the optical image used in our analysis we found a relatively brighter star in the south-western direction and several nearby companion galaxies of similar intensity suggesting that G4Jy 427 could lie in a galaxy-rich environment.
G4Jy 436: is also known as PKS 0413-21, a core dominated quasar at z=0.808 (Wilkes et al. 1986;Best et al. 1999) showing mid-IR colors similar to those of γ-ray blazars , and having a radio jet detected in the X-rays Massaro et al. 2011a).
G4Jy 446: is optically identified with a galaxy ) and has a photometric redshift estimate obtained thanks to the MS4 analysis that places the radio source at z=0.81 (Burgess & Hunstead 2006b). In our analysis we could not clearly identify the radio core position and its optical counterpart. The mid-IR counterpart was not found in the G4Jy catalog (White et al. 2020b). G4Jy 436 also shows a CSS radio core  G4Jy 453: is a giant radio source already known since the MRC release (Jones & McAdam 1992), but the lack of high resolution image prevented us to verify the position of the host galaxy and if the redshift estimate reported in the literature is correct .
G4Jy 462: is one of the dumb-bell FR I radio galaxies in the southern hemisphere (Ekers et al. 1969;McAdam et al. 1988;Morganti et al. 1993), also known as IC 2082 and PKS 0427-53, with a twin tail (Carter et al. 1981;Lilly & Prestage 1987), and lying in the nearby galaxy cluster Abell S 463. It has a redshift measurement of z=0.03931 (see e.g., Raimann et al. 2005) showing weak emission lines in its optical spectrum. It also belongs to both the sample of local radio galaxies detected at 20 GHz (Sadler et al. 2014). We adopted here the same optical identification provided in the literature (see also Jones & McAdam 1992;Burgess & Hunstead 2006b;White et al. 2020b). However, the lack of high resolution radio maps prevented us to classify this radio source as a wide-angle tail (WAT) radio galaxy (Burns et al. 1981;Owen & Rudnick 1976;O'Donoghue et al. 1990O'Donoghue et al. , 1993Sakelliou & Merrifield 2000;Missaglia et al. 2019).
G4Jy 492: is a FR II radio galaxy at z=0.147 di Serego et al. 1994;Best et al. 1999) with several nearby companion galaxies (Ramos ) and surrounding diffuse X-ray emission. G4Jy 492, also belonging to the 2 Jy sample (see e.g., Wall & Peacock 1985;Morganti et al. 1993), has been recently observed in the X-rays and both its radio core and the northern hotspot were detected . It was also detected in hard X-ray band (see e.g., Cusumano et al. 2010;Oh et al. 2018).
G4Jy 506: was associated with the same optical counterpart in the literature Hunstead et al. 1971b). It shows a classical FR II radio structure. It is listed in the MS4 catalog with a photometric redshift estimate of z=0.22 (Burgess & Hunstead 2006b).
G4Jy 507: is also known as NGC 1692 a radio galaxy at z=0.035364 , optically identified in the literature (Bolton & Ekers 1965;Wills et al. 1973;Best et al. 1999). According to the optical image used in our analysis its entire radio structure lies within the brightness profiles of its host galaxy and it has two nearby companion galaxies, probably harbored in a galaxy-rich environment (Miller et al. 1999).
G4Jy 510: is a nearby QSO at z=0.533 Henriksen et al. 1991;Bechtold et al. 2002) showing MgII absorption system (Tytler et al. 1987) and giant optical nebulae surrounding it (Helton et al. 2021). It shows a lobe dominated radio structure (Reid et al. 1999) also listed in the CRATES catalog .
G4Jy 513: is not optically identified but there is an optical interacting pair of galaxies within the highest radio contour drawn in our finding chart. We found two archival radio maps of G4Jy 513, as reported in Figure 31 in comparison with the archival r-band optical image, showing diffuse radio emission at both 1.4 GHz and 4.9 GHz but lacking a clear detection of its radio core. In this case the nearby radio source known as PKS 0456-301 could be potentially associated with G4Jy 513 thus suggesting the position of its optical counterpart as shown in the finding chart. G4Jy 513 appear also associated with the galaxy cluster Abell 3297 (see e.g., , and references therein), even if spectroscopic confirmation is needed.
G4Jy 524: is a radio galaxy (a.k.a. MRC 0508-187 and TXS 0508-187) with a lobe dominated radio structure. The optical image shows the presence of several nearby companion galaxies. G4Jy 524 is also listed in a sample of ultra steep spectrum radio sources .
G4Jy 530: shows a large double-lobed radio morphology. The lacks of a mid-IR association and the resolution of radio maps used in our analysis did not permit us to locate the host galaxy in the optical image. However we tentatively Figure 31. Left and central panels) Radio maps of G4Jy 513 at 1.4 GHz and 4.9 GHz, respectively, retrieved from the NVAS. The frequency of each radio map from which radio contours were drawn is reported together with the intensity of the first level and the binning factor indicating how they increase. Right panel) The r-band optical image collected from one of the surveys used in our analysis with three levels of radio contours drawn from the radio map at 4.9 GHz starting at 4.9 mJy and increasing by 0.5 mJy. No clear detection of the radio core and of a potential optical counterpart is reported for this radio source showing diffuse radio emission. associated this radio source with the narrow line radio galaxy PKS 0511-48 (05:12:47.22, -48:24:16.4 in J2000) lying in the center of the radio structure and having a redshift estimate of z=0.30638 , 2004.
G4Jy 538: is a compact radio galaxy ) also known as PKS 0519-20, with a z =1.086 (Best et al. 1999) with a peaked spectrum determined thanks to the GLEAM observations .
G4Jy 580: has a mid-IR counterpart associated in the G4Jy that is the same listed in the MS4 sample (Burgess & Hunstead 2006a,b) as discussed in White et al. (2020b). However the presence of several optical sources around the radio intensity peak, marked in the finding chart, coupled with relatively poor angular resolution of the radio map available, prevented us to claim a firm identification.
G4Jy 1513: is a radio galaxy (a.k.a PKS 1859-23) at z=1.430 , and it resides in a region with several nearby companion galaxies. However, the lack of spectroscopic information about the nearby sources coupled with the lack of the counterpart associated with the radio emission at both mid-IR and optical frequencies did not allow us to confirm this association.
G4Jy 1518: also belongs to the equatorial sample of Best et al. (1999) where it is classified as a radio galaxy at z=0.226, as well as in the Molonglo Reference Catalogue of radio sources (MRC 1912-269 Large et al. 1981).
G4Jy 1532: is a radio galaxy (a.k.a MRC 1920-077 andTXS 1920-077) listed in the sample of Best et al. (1999) with the optical counterpart correspondent to our association, but lacking a mid-IR correspondence. It has a redshift estimate of z=0.648.
G4Jy 1558: is a Broad Line Radio Galaxy (BLRG; a.k.a. PKS 1932-464), with a relatively broad and strong Hα emission line, with a z=0.2307 Villar-Martin et al. 1998;Hernán-Caballero et al. 2016), and, a FR II radio morphology . G4Jy 1558 shows a complex gas distribution resulting from the interaction with a nearby companion galaxy, and there is a knotty extended emission line nebula extending beyond the radio structure and the ionization cones, one of the largest ever detected around a radio galaxy at any redshift (Villar-Martin et al. 2005). The origin of the nebula is due to the presence of a star forming halo associated with the debris of the merger that triggered the activity. G4Jy 1558 has sufficient luminosity at mid-to far-IR wavelengths to be classified as a luminous infrared galaxy . The star formation structure can extend on the scale of a galaxy group, beyond the old stellar halo of the host galaxy (Villar-Martin et al. 2005). The gas in the emission line nebula is predominantly ionized by a mixture of AGN photoionization and emission from young stars Tadhunter et al. 2011). G4Jy 1558 is a member of an interacting galaxy group which includes a highly disturbed starburst galaxy at a similar redshift, located at ∼100 kpc in the north-eastern direction , and connected with G4Jy 1558 by a series of arc-like irregular features up to ∼70 kpc distance from the galaxy centre (Ramos . G4Jy 1562: a radio galaxy listed in the MS4 sample with a photometric redshift estimate of z=1.92 (Burgess & Hunstead 2006b). Figure 32. The two radio galaxies G4Jy 1677 and G4Jy 1678 (a.k.a., NGC 7016 & NGC 7018) harbored in the galaxy cluster Abell 374 and lying at an angular separation of ∼190 ′′ (i.e., corresponding to ∼150 kpc at the galaxy cluster redshift). Radio contours from the TGSS and the VLASS archival images at 150 MHz and 3 GHz are overlaid, in blue and in black respectively, on the optical image in r band available in the Pan-STARRS database.
McLure & Dunlop 2001), and detected in the hard X-rays Koss et al. 2017) with extended soft X-ray emission . Optical images reveal a disturbed morphology with a shell in the western side embedded in an amorphous halo, and a faint tidal tail pointing to the south-eastern direction (Ramos .
G4Jy 1709: (a.k.a. PKS 2135-20) is a BLRG at z=0.63634 Holt et al. 2008;Hernán-Caballero et al. 2016) with a CSS radio core . It shows broad fan on the northern side, interpreted as due to a past interaction, and in agreement with the detection of a young stellar population in the nuclear region , and a far infrared excess (Dicken et al. 2009b) being also extremely luminous at mid-IR frequencies (Ramos .
G4Jy 1748 & G4Jy 1749: the former one is a lobe dominated QSO (a.k.a. PKS 2152-69 Jones & McAdam 1992;Morganti et al. 1993) with a z=0.0281 (Marenbach & Appenzeller 1982;Tadhunter et al. 1993) and showing a wide range of features associated with radio-galaxy/gas interactions typical of sources where radio mode feedback processes are occurring (Worrall et al. 2012). High resolution radio observations reveal a radio component at ∼10 arcsec in the north-eastern direction from the core, close to an optical highly ionized cloud. At larger scale, G4Jy 1748 shows a FR II morphology with the northern lobe having a "relaxed" structure, while the southern lobe shows an edge-brightened, arc-like structure (Fosbury et al. 1998). The X-ray surface brightness has two depressions spatially associated with the radio lobes thus suggesting the presence of X-ray cavities inflated with radio plasma . Both radio lobes have their hotspot detected in the X-rays (see also Ly et al. 2005;Massaro et al. 2011a). G4Jy 1748 has been also detected in both the hard X-ray and the γ-ray bands (see e.g., Cusumano et al. 2010;Baumgartner et al. 2013;Abdollahi et al. 2020, respectively). On the other hand, at an angular separation of ∼3.6 ′ from G4Jy 1748, in the eastern direction, there is the nearby radio source G4Jy 1749 as shown in Figure 33. In this case the lack of spectroscopic information on G4Jy 1749 prevent us to claim that they belong to the same galaxy cluster as for G4Jy 1677 & G4Jy 1678. Figure 33. The DSS optical image, in the red filter, of the field including the two radio sources G4Jy 1748 and G4Jy 1749, the former one harbored in a galaxy cluster at z=0.0281 and lying at an angular separation of ∼3.6 ′′ (i.e., corresponding to ∼120 kpc at its redshift) from G4Jy 1749. Radio contours from the SUMSS at 843 MHz are overlaid in black while those obtained from the GLEAM radio maps between 171 MHz and 230 MHz are reported in cyan, the former ones start at level of 0.1 Jy/beam while the latter ones at 0.4 Jy/beam and both increase by a factor of 4.
The first comparison was carried out with latest release (i.e., v1.01) of the Parkes radio catalog (PKSCAT90) 25 ) listing radio and optical data for ∼8000 radio sources and covering essentially all the sky south of declination +27 • but largely excluding the Galactic Plane and the Magellanic Cloud regions as for the G4Jy. The original catalog included observations performed at frequencies of 408 MHz and 2700 MHz.
Then we compared the G4Jy-3CRE sample with the Molonglo Reference Catalog of Radio Sources (MRC, Large et al. 1981). The MRC is one of the largest homogeneous catalogs of radio sources observed at 408 MHz, containing ∼12000 discrete sources with flux densities greater than 0.7 Jy in the declination range between +18.5 • to -85 • (in B1950 equinox) and excluding regions within 3 • of the Galactic equator.
We also used the Texas Survey of 66841 discrete radio sources (TXS; Douglas et al. 1996) detected in the declination range between -35.5 • and +71.5 • (in B1950 equinox), which was performed at 365 MHz. The Survey lists accurate positions with positional uncertainty of the order of arcseconds and flux densities of a few percent. The TXS Survey is 90% complete at 0.4 Jy and 80% complete at 0.25 Jy, being nearly free from spurious sources and has a low level of lobe-shift incidence.
In addition the Australia Telescope 20-GHz Survey (AT20G Murphy et al. 2010) was also compared with the G4Jy-3CRE sample. The AT20G is a blind radio survey carried out at 20 GHz with the Australia Telescope Compact Array (ATCA) from 2004 to 2008, and covers the whole sky south of declination 0 • . The latest release of the AT20G source catalog lists 5890 sources above a 20-GHz flux-density limit of 40 mJy. All AT20G sources have total intensity and polarization measured at 20 GHz, and most sources south of declination -15 • also have near-simultaneous flux-density measurements at 5 and 8 GHz with a completeness level of 91% above 100 mJy/beam.
Then we crossmatched the G4Jy-3CRE sample with the Parkes-MIT-NRAO catalog ( PMN Griffith & Wright 1993;Wright et al. 1994) in several regions of the sky: Southern, Zenith, Tropical and Equatorial surveys. These surveys were made using the Parkes 64-m radio telescope at a frequency of 4850 Hz with the NRAO multibeam receiver mounted at the prime focus. These surveys had a spatial resolution of ∼4.2 arcminutes. This survey covers 2.50 sr listing 23,277 radio sources to a flux limit ranging as a function of declination between ∼20 mJy at the southern survey limit and ∼50 mJy at the northern limit.
We found that 237 sources out of those 264 included in the G4Jy-3CRE catalog have a radio counterpart in the PKSCAT90. In particular 171 of them have IDF=1.0 and this optical identification was augmented by a literature search for G4Jy 538, G4Jy 939, G4Jy 1401, G4Jy 1854 (see e.g., Bolton & Ekers 1965, 1966aBolton et al. 1968;Hunstead et al. 1971a;Peterson et al. 1973Peterson et al. , 1976White et al. 1987). Then 249 out of 264 show a radio counterpart in the MRC catalog being detected at 408 MHz, in particular 84 are also selected in the equatorial sample of Best et al. (1999), while 125 were selected to create the MS4 catalog, both built using MRC observations. There are also 235 G4Jy-3CRE radio sources with a counterpart listed in the PMN catalog while only 126 and 118 in the TXS and in the AT20G catalogs, respectively.
In the following tables, we report the G4Jy name together with those available in several radio catalogs based on surveys carried out at different frequencies and mainly covering the Southern Hemisphere, namely: TXS, MRC, PKS, PMN catalogs. These cross-identifications can be used to retrieve observations out of different databases. These associations were mainly based on the NED and SIMBAD databases. The last two columns are dedicated to common names and to highlight those objects that are associated to Abell galaxy clusters as found in our literature search (see Appendix A).
This search for radio counterparts, as well as that on information regarding classifications and redshifts, was also augmented by the results achieved for the 2 Jy catalog 26 (see e.g., Wall & Peacock 1985), a southern sample of radio galaxies defined as having flux densities above 2 Jy at 2.7 GHz, declination below +10 • and redshifts up to 0.7 and its full subsample (see e.g., Tadhunter et al. 1993;Morganti et al. 1997) for which a large suite of multifrequency observations is already available (see e.g., Morganti et al. 1993Morganti et al. , 1999Tadhunter et al. 2002;Ramos Almeida et al. 2011;Mingo et al. 2014). However, crossmatching the G4Jy-3CRE sample with the 2 Jy catalog we found that only 45 out of 264 radio sources are also listed therein. This analysis will be also updated in all forthcoming papers. Table 5. Radio cross-identifications. Table 5 is published in its entirety in the machine-readable format. A portion is shown here for guidance regarding its form and content. Assuming those radio sources for which (i) the optical counterpart is coincident with the mid-IR one (i.e., IDF=1.0) plus those (ii) lacking an optical counterpart but having a mid-IR one associated with their radio core (i.e., IDF=4.1) as correct mid-IR associations, we count 203 out of 264 objects, being ∼77% of the whole G4Jy-3CRE sample. This fraction can be compared with the expected number of spurious associations that can arise when matching the G4Jy catalog with the AllWISE potential counterparts. We computed the chance probability of associations between mid-IR sources and those listed in the full G4Jy catalog since this was originally used to assign counterparts while the G4Jy-3CRE sample was extracted out of it later. We adopted here the same procedure described in D'Abrusco et al. 2013;Massaro et al. 2014a;D'Abrusco et al. 2014) to compute the probability of having spurious associations between those sources listed in the G4Jy catalog, using their brightness-weighted radio centroids as positions, and their AllWISE potential counterparts. Here we report just a brief overview of the method used. All crossmatches computed in the following analysis are based on their positions reported in the catalogs.
We started counting the total number of mid-IR counterparts N (R) within circular regions of radius R in the range between 0 ′′ and 10 ′′ , for each G4Jy source. Next, we generated 1000 mock catalogs, based on the distribution of the mid-IR sources around the brightness-weighted radio centroid reported in the G4Jy catalog, shifted it by a random value uniformly distributed between 10 ′′ and 20 ′′ in a random direction of the sky. The shifts used to create the mock catalogs were chosen to be not too distant from the original position reported in the G4Jy to guarantee that fake catalogs have the same sky distribution as the original G4Jy. This allowed us to crossmatch mock sources with real G4Jy objects taking into account the local density distribution of mid-IR sources (see Massaro et al. 2014a, for additional information). The total number of G4Jy sources in all mock catalogs is also preserved.
For each mock realization of the G4Jy catalog, we counted the number of fake associations with the AllWISE catalog occurring at angular separations R smaller than 10 ′′ . Then we computed the mean number λ(R) of these fake associations, averaged over the 1000 mock catalogs, verifying that λ(R) has a Poissonian distribution. Increasing the radius by ∆ R =0 ′′ .1, we also calculated the difference ∆ λ(R) between the number of mock sources within a radius of R + ∆(R) and those within R, defined as: ∆ λ(R) = λ(R + ∆ R) − λ(R).
Finally, in Figure 34 we show the comparison between ∆ N (R) (i.e., the difference between the number of real matches within a radius of R + ∆(R) and those within R) and ∆ λ(R). For angular separations larger than R assoc =5 ′′ .4 the ∆ λ(R) curve begins to match that of ∆ N (R). Thus we choose 5 ′′ .4 as the maximum angular separation at which we could consider the mid-IR source a reliable counterpart of the G4Jy radio object. An association between a G4Jy source and its potential mid-IR counterpart, occurring at angular separation above 5 ′′ .4 has almost the same probability of being either correct or random. Positional uncertainties of the NVSS and SUMSS radio surveys typically ranges between 1 ′′ to 5 ′′ and 2 ′′ to 10 ′′ , respectively (see e.g., Vollmer et al. 2005, and references therein for a recent analysis).
These were the radio surveys used in the G4Jy analysis to compute brightness-weighted radio centroids and associate radio sources with their mid-IR counterparts. Thus our statistical result is also in agreement with previous analyses. Figure 34. The values of ∆ λ(R) (red squares) and ∆ N (R) (black circles) as a function of the angular separation R. Our choice of Rassoc is marked by the vertical dashed line. It occurs when ∆ λ(R) ≃ ∆ N (R). Uncertainties on the average λ(R) values obtained by the crossmatches with mock catalogs were computed from their distributions at each R. The correspondent chance probability of having spurious associations at R = Rassoc is then reported in the lower panel. Left panels correspond to the G4Jy catalog, used to extract the G4Jy-3CRE sample while, for comparison, in the right panels we show the results of the same procedure applied to the 3CR sample. In the latter case we used NVSS coordinates similar to those adopted in the G4Jy catalog to estimate the brightness-weighted radio centroids at declinations above -39.5 • . Considering that in the Southern Hemisphere the source sky density (i.e., number of sources per square degree) of mid-IR potential counterparts is ∼10% larger than that correspondent to the 3CR, results on these two different catalogs are in agreement. Figure 34 the chance probability of spurious associations p(R assoc ) was computed as the ratio between the number of real associations N (R assoc ) and the average of those found in the mock realizations λ(R assoc ), corresponding to a value of ∼26% (see also Massaro et al. 2011b;D'Abrusco et al. 2013, for additional details on p(R assoc )). Thus the choice of R assoc is based on the comparison between differential distributions of real and average mock matches while p(R assoc ) on their cumulative ones. Adopting the same statistical procedure used here to search for mid-IR counterparts of blazars listed in the Roma-BZCAT (Massaro et al. 2009(Massaro et al. , 2015b, we found p(R assoc ) below 1% at radius of ∼3 ′′ (see also Massaro et al. 2013bMassaro et al. , 2014bD'Abrusco et al. 2019;de Menezes et al. 2020). This was mainly due to the blazar nature (i) being core-dominated radio sources, thus mostly point-like objects at GHz frequencies, and (ii) having more precise positions reported in the comparison catalog, as well as a combination of these two effects.

As shown in
The probability of having spurious associations (i.e., ∼26%) is certainly in good agreement with our refined analysis on optical counterparts, for which the number of incorrect associations is expected to be ∼23%. Then, angular separations θ ow between the mid-IR counterpart, assigned in the G4Jy catalog, and the optical one with IDF=1.0 are all below 4 ′′ .8 with only one exception having 5 ′′ .6, as shown in Figure 7, thus consistent with being correct. In Figure 34 we also report the same plots but computed for the 3CR catalog . In this case the value of the association radius is 5 ′′ .2, estimated according to the same method previously described. This was computed using the NVSS coordinates of the 3CR radio sources that have a similar precision of those used in the G4Jy catalog to determine the brightness-weighted radio centroids. This association radius corresponds to a chance probability of spurious associations of ∼19%. However the source sky density (i.e., number of sources per square degree) of mid-IR potential counterparts around 3CR sources is ∼10% smaller than the one measured for the G4Jy-3CRE catalog thus, taking into account of this, the chance probability computed for the G4Jy-3CRE catalog is also in agreement with the expectations based on the 3CR catalog.
In the G4Jy-3CRE catalog 225 sources out of 264 have a mid-IR counterpart assigned by the original G4Jy catalog (White et al. 2020a,b), and 136 out of 225 sources lie below the threshold of 5 ′′ .4 angular separation between the brightness-weighted radio centroid and the position of the assigned mid-IR counterpart. A similar situation occurs when comparing the brightness-weighted radio centroid with the optical position of the counterpart assigned thanks to our optical analysis. We have been able to find 211 optical counterparts out of 264 examined sources and only 109 have  Figure 3 for G4Jy 171. The blue cross marks the position of the brightness-weighted radio centroid while the red one that of the mid-IR and optical counterpart, associated with the radio core. The location of the radio core was identified thanks to the NVAS archival radio map at 1.4 GHz used also to draw radio contours overlaid to the mid-IR image. The radio core is not clearly detected in archival images of the TGSS, NVSS and SUMSS. the angular separation between the two crosses is ∼49 ′′ . angular separation between radio and optical position below 5 ′′ .4, in agreement with the expectations of the statistical analysis. The main reason underlying the relatively high probability of getting spurious associations is the use of a brightness-weighted radio centroid since, as previously stated, does not always provide a reliable position of the host galaxy for sources having an extended radio morphology that is unresolved in radio maps used to compute it and/or for those clearly asymmetric. This motivated our analysis based on higher resolution radio images thus allowing us to determine the precise location of the host galaxies for the G4Jy-3CRE catalog presented here. An extreme example on how the radio centroid can provide a misleading information about the position of the host galaxy is shown in Figure 35 where the high resolution radio map at 1.4 GHz available in the NVAS archive allowed us to confirm the position of the host galaxy, being the same assigned at mid-IR frequencies in the G4Jy catalog, but lying at ∼49 ′′ angular separation from the location of the brightness-weighted radio centroid. Diffuse radio emission arising from lobes could bias the location of the brightness-weighted radio centroid as shown in the case of G4Jy 171 ( Figure 35).