The γ-ray sky seen at X-ray energies

Abstract

Context. Nearly 50% of all sources detected by the Fermi Large Area Telescope are classified as blazars or blazar candidates, one of the most elusive classes of active galaxies. Additional blazars can also be hidden within the sample of unidentified or unassociated γ-ray sources (UGSs) that constitute about one-third of all gamma-ray sources detected to date. We recently confirmed that the large majority of Fermi blazars of the BL Lac subclass have an X-ray counterpart. Aims. Using the X-ray properties of a BL Lac training set and combining these with archival multifrequency information, we aim to search for UGSs that could have a BL Lac source within their γ-ray positional uncertainty regions. Methods. We reduced and analyzed the Swift X-ray observations of a selected sample of 327 UGSs. We then compared the X-ray fluxes and hardness ratios of all sources detected in the pointed fields with those of known Fermi BL Lacs. Results. We find at least one X-ray source, lying within the γ-ray positional uncertainty at 95% confidence level, for 223 UGSs and a total of 464 X-ray sources in all fields analyzed. The X-ray properties of a large fraction of them, eventually combined with radio, infrared, and optical information, exhibit BL Lac multi-frequency behavior, thus allowing us to select high-confidence BL Lac candidates; some of them were recently observed during our optical spectroscopic campaign which confirmed their nature. Conclusions. We find that out of 50 X-ray sources that were confirmed as BL Lacs through optical spectroscopy, 12 do not show canonical mid-infrared or radio BL Lac properties. This indicates that the selection of X-ray BL Lac candidates is a strong method to find new counterparts within Fermi UGSs. Finally, we pinpoint a sample of 32 Swift/XRT candidate counterparts to Fermi UGSs that are most likely BL Lac objects.

Optical spectroscopic observations of low-energy counterparts of Fermi-LAT γ-ray sources

“…The new spectra presented here were collected for targets selected from the following lists: 1. BCUs already assigned as counterparts to 4FGL catalog sources, but whose blazar nature were still uncertain; 2. radio and X-ray sources located within the γ-ray positional uncertainty of UGSs (Marchesini et al 2020); 3. BL Lacs whose optical spectra are not available in the literature, or do not have a redshift estimate; 4.…”

“…The results were derived from new observations with the Blanco and OAN-SPM telescopes and from archival data collected from the SDSS. Our targets were selected from the KDEBLLACS and WIBRALS2 catalogs (D'Abrusco et al 2019), a sample of Xray selected sources (Marchesini et al 2020), and BL Lac objects with previously undetermined redshifts that would benefit from new observations. In our sample 7 sources were observed at Blanco (2 UGSs, 5 BCUs), 15 at OAN-SPM (6 UGSs, 9 BCUs), while the remaining 8 (all 5 BL Lacs, 1 UGS, and 2 BCU) were found in the SDSS DR15 archive (Aguado et al 2019).…”

Classifying blazar candidates from the 3FGL unassociated catalog into BL Lacs and FSRQs using Swift and WISE data

“…All except one (3FGL J0427.9-6704 classified as a "Binary" in the 4FGL catalog) of our results match with the 4FGL predictions as seen in Table 2, although three identified BL Lacs in the 4FGL catalog were characterized as "ambiguous" by our RF classifier (note: our classifier did find that these 3 sources were > 81% likely to be BL Lacs). In addition, we compared our classifi-cation results with a similar study conducted by Marchesini et al (2020a), in which the authors searched for BL Lac candidates among the Fermi Unassociated sources. These authors selected their sample using a slightly different criteria such as SNR>3 as compared to ours with SNR>4.…”

“…Finding associations to these sources or classifying these is a multi year task which often requires multiwavelength observations for confirmations. In the past few years, various studies have been conducted on these gamma-ray sources where machine learning methods were employed to classify the unassociated sources; e.g., (Ackermann et al 2012;Saz Parkinson et al 2016;Marchesini et al 2020a). In particular, Marchesini et al (2020b,a) first explored the connection between gamma-rays and X-rays for blazars and later utilized X-ray data from Swift in conjunction with the Fermi gamma-ray data to find BL Lacs among the unassociated sources.. We (Falcone A. D. & et al Stroh 2015;Kaur et al 2019) have conducted an X-ray survey targeting the Fermi unassociated source fields and found various possible X-ray associations 1 .…”

Classifying Blazar Candidates from the 3FGL Unassociated Catalog into BL Lacertae Objects and Flat Spectrum Radio Quasars Using Swift and WISE Data

“…Finding associations to these sources or classifying these is a multiyear task that often requires multiwavelength observations for confirmation. In the past few years, various studies have been conducted on these gamma-ray sources where machine-learning methods were employed to classify the unassociated sources (Ackermann et al 2012;Saz Parkinson et al 2016;Marchesini et al 2020a). In particular, Marchesini et al (2020aMarchesini et al ( , 2020b first explored the connection between gamma rays and X-rays for blazars and later utilized X-ray data from Swift in conjunction with the Fermi gamma-ray data to find BL Lac objects among the unassociated sources.…”