Identifying the 3FHL Catalog. III. Results of the CTIO-COSMOS Optical Spectroscopy Campaign

Blazars are a peculiar class of active galactic nuclei (AGNs) that dominate the observable γ-ray universe because of their extreme properties and abundant population. The blazar properties are a result of nonthermal emitting plasma traveling toward the observer, causing relativistic amplification of the flux. This leads to an amplification of low-energy photons in the medium to intense levels via the inverse Compton process, making blazars valuable sources to understand the physics of an AGN. The Third Fermi-LAT Catalog of High-Energy Sources (3FHL; Ajello et al. 2017), which encompasses seven years of observations made by the Large Area Telescope (LAT) on board the Fermi Gamma-ray Space Telescope (Atwood et al. 2009), contains more than 1500 sources detected at >10 GeV, the vast majority of which (≈1160) are blazars (Ajello et al. 2017).

Innovative scientific results can be obtained using the blazar data collected by the Fermi-LAT in the γ-ray regime, provided that the redshift (z) of the observed blazar source is known. These are not only limited to blazar physics, such as understanding their basic emission processes (e.g., Ghisellini et al. 2017) or their evolution with redshift (Ajello et al. 2014), but also to other areas of study, like understanding the extragalactic background light, which encompasses all the radiation emitted by stars and galaxies and reprocessed radiation from interstellar dust, and its evolution with z (Ackermann et al. 2012; Domínguez et al. 2013). Out of the confirmed blazar sources reported in the 3FHL catalog, a redshift measurement of only ≈50% of sources is present (Ajello et al. 2017). To overcome this limitation, extensive optical spectroscopic campaigns, targeting those 3FHL objects still lacking redshift and classification, must be performed.

Besides being used for redshift determination, optical spectroscopy campaigns of blazars are also essential to distinguish between blazar subclasses, namely BL Lacs (BLL) and flat spectrum radio quasars (FSRQs). FSRQs are generally high-redshift objects with an average luminosity larger than that of the BLL (Padovani 1992; Paiano et al. 2017). As a result, the emission lines in the BLL spectra are weak or absent, and the lines in FSRQs are extremely prominent. This is seen by the difference in the equivalent width (EW) of the lines where, generally, FSRQ have lines with EW > 5 Å and BLL have lines with EW < 5 Å (Urry & Padovani 1995; Ghisellini et al. 2017). The blazar sources not classified as FSRQ or BLL are listed as blazar candidates of an uncertain type (BCU) in the 3FHL catalog and constitute ≈25% of the reported blazar sample (Ajello et al. 2017). Obtaining a spectroscopically complete classification of the blazars observed by Fermi-LAT in the γ-ray regime is essential to validate claims of different cosmological evolution of the two classes (Ajello et al. 2012, 2014).

The ground-based telescopes used in the spectroscopy campaigns are generally of the 4, 8, and 10 m class type. While the 10 and 8 m class telescopes are shown to be significantly more effective in obtaining redshift measurements for blazars (60%–80% versus 25%–40% success rate; see, e.g., Paiano et al. 2017; Marchesi et al. 2018), even 4 m class telescopes have proven to be useful for effectively distinguishing between the two different blazar subclasses (see Shaw et al.2013; Massaro et al. 2014; Paggi et al. 2014; Landoni et al. 2015; Ricci et al. 2015; Álvarez Crespo et al. 2016a, 2016b; Marchesini et al. 2016).

This work is part of a larger spectroscopic follow-up campaign to classify the BCUs in the 3FHL catalog and to measure their redshift. The first part of the campaign took place in the second half of 2017, when we observed 28 sources in seven nights of observations with the 4 m telescope at the Kitt Peak National Observatory (KPNO). The results of this work are reported in Marchesi et al. (2018): we classified 27 out of 28 sources as BL Lacs, while the remaining object was found to be an FSRQ. Furthermore, we measured a redshift for three sources and set a lower limit on z for other four objects; the farthest object in our KPNO sample has z > 0.836. The spectroscopic campaign will then continue with seven nights of observations with the 4 m telescope at the Cerro Tololo Interamerican Observatory (CTIO) in Chile and five nights of observations with the 8 m Gemini-N and Gemini-S telescopes (to be performed in 2019). In this work, we report the results of the observations made during the first four nights at CTIO. Our source sample contains 23 BCUs in the 3FHL catalog without a redshift measurement. The paper is organized as follows: Section 2 reports the criteria used in sample selection, Section 3 describes the methodology used for the source observation and spectral extraction procedures, Section 4 lists the results of this work, both for each individual source and also in general terms, while Section 5 reports the conclusions inferred from this spectroscopic campaign.

We selected the 23 objects in our sample among the BCUs in the 3FHL catalog, using the following three criteria.

• 1.  
  The object should have a measured optical magnitude measurement, and it should be V ≤ 19.5. Based on previous works, sources with a magnitude of V > 19.5 require more than two hours of observations to obtain an acceptable signal-to-noise ratio (S/N), therefore, significantly reducing the number of sources that one can observe in a night.
• 2.  
  The 3FHL source should be bright in the hard γ-ray spectral regime (${f}_{50\mbox{--}150\mathrm{GeV}}\gt {10}^{-12}$ erg s⁻¹ cm⁻²). Selecting 3FHL objects bright in the 50–150 GeV band ensures that the completeness of the 3FHL catalog evolves to lower fluxes as more optical observations are performed.
• 3.  
  The target should be observable from Cerro Tololo with an altitude above the horizon δ > 40° (i.e., with an air mass <1.5); this corresponds to a decl. range of −80° < decl. < 20°. The target should also be observable in October, when the observations take place (i.e., it should have R.A. ≥ 09^h0^m00^s and R.A. ≤ 0^h30^m00^s).

A total of 77 3FHL sources satisfy all of these criteria. Our 23 sources were selected among these 77 objects with the goal of covering a wide range of optical magnitudes (V = [16–19.5]) and, consequently, of potential redshifts and luminosities. In Figure 1, we show the normalized V-band magnitude distribution of our sources, compared with the 1 of the overall population of the 173 3FHL BCUs that still lack redshift measurements and have available magnitude information. We also plot the magnitude distribution of the 28 sources studied in Marchesi et al. (2018), where we sampled a larger number of bright sources (V < 16), which all turned out to be featureless BL Lacs. The sources used in our sample and their properties are listed in Table 1.

Zoom In Zoom Out Reset image size

Table 1.  List of Sources and Their Properties Sorted in the Order of Increasing R.A. Values

3FHL Name	Counterpart	R.A.	Decl.	$E(B-V)$	mag	Obs. Date	Exposure	Continuum Slope
(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
3FHL J0002.1−6728	SUMSS J000215−672653	00:02:15.21	−67:26:52.91	0.0253	18.6	2018 Jun 1	5400	−1.44
3FHL J0935.2−1735	NVSS J093514−173658	09:35:14.77	−17:36:58.30	0.0643	17.8	2018 Jun 1	3900	−0.12
3FHL J0936.4−2109	1RXS J093622.9−211031	09:36:23.08	−21:10:39.00	0.0574	18.5	2018 Jun 2, 3	5100	−0.28
3FHL J1030.6−2029	NVSS J103040−203032	10:30:40.46	−20:30:32.70	0.0469	18.4	2018 Jun 3	3300	−1.91
3FHL J1042.8+0055	RBS 895	10:43:03.84	+00:54:20.43	0.0419	19.3	2018 Jun 4	5600	−1.03
3FHL J1130.5−7801	SUMSS J113032−780105	11:30:32.92	−78:01:05.20	0.1921	17.6	2018 Jun 2	3400	−1.07
3FHL J1155.5−3418	NVSS J115520−341718	11:55:20.43	−34:17:18.30	0.0702	16.8	2018 Jun 1	2400	−1.10
3FHL J1212.1−2328	PMN J1212−2327	12:12:04.54	−23:27:42.00	0.0656	18.2	2018 Jun 1	4500	−0.77
3FHL J1223.5−3033	NVSS J122337−303246	12:23:37.32	−30:32:46.10	0.0593	17.2	2018 Jun 2	3400	−2.15
3FHL J1229.7−5304	AT20G J122939−530332	12:29:39.93	−53:03:32.20	0.1293	17.8	2018 Jun 3	2300	−0.44
3FHL J1315.9−0732	WISE J131552.98−073301.9	13:15:53.00	−07:33:02.07	0.0352	18.2	2018 Jun 4	4500	−0.87
3FHL J1433.5−7304	GALEX J143343.0−730437	14:33:42.81	−73:04:36.84	0.1592	17.9	2018 Jun 1	4000	−0.81
3FHL J1439.4−2524	NVSS J143934−252458	14:39:34.66	−25:24:59.10	0.0862	16.2	2018 Jun 3	2800	−0.01
3FHL J1605.0−1140	WISE J160517.53−113926.8	16:05:17.53	−11:39:26.83	0.2584	18.7	2018 Jun 4	5400	−0.35
3FHL J1612.3−3100	NVSS J161219−305937	16:12:19.95	−30:59:37.80	0.2003	18.1	2018 Jun 2	3600	−1.11
3FHL J1640.1+0629	NVSS J164011+062827	16:40:11.06	+06:28:27.70	0.0695	18.6	2018 Jun 2	3800	−1.71
3FHL J1842.4−5841	1RXSJ184230.6−584202	18:42:29.67	−58:41:57.19	0.0848	17.5	2018 Jun 1	3600	−1.67
3FHL J1924.2−1548	NVSS J192411−154902	19:24:11.82	−15:49:02.10	0.1491	17.7	2018 Jun 3	3600	−1.35
3FHL J2034.9−4200	SUMSS J203451−420024	20:34:51.06	−42:00:37.60	0.0360	17.2	2018 Jun 2, 4	3900	−0.62
3FHL J2041.7−7319	SUMSS J204201−731911	20:42:01.85	−73:19:13.01	0.0544	18.2	2018 Jun 4	3400	−4.47
3FHL J2240.3−5240	SUMSS J224017−524111	22:40:17.64	−52:41:13.07	0.0118	16.7	2018 Jun 4	1950	−5.84
3FHL J2321.8−6437	PMN J2321−6438	23:21:42.17	−64:38:06.90	0.02	17.4	2018 Jun 4	2800	−0.06
3FHL J2339.2−7404	1RXS J233919.8−740439	23:39:20.88	−74:04:36.12	0.0262	16.1	2018 Jun 4	1500	−0.65

Note. (1): 3FHL catalog (Ajello et al. 2017) name for the source. (2): Optical, infrared, X-ray, or radio counterpart of the source. (3) R.A. (4) decl. (5) $E(B-V)$ value obtained using the measurements of Schlafly & Finkbeiner (2011) and the NASA/IPAC infrared science archive online tool. (6) V-band magnitude. (7) Date of observation. (8) Exposure time (in seconds). (9) Slope of continuum fit obtained from the observed fits file.

Download table as:  ASCIITypeset image

All of the sources in our sample were observed using the 4 m Blanco telescope located at the CTIO in Chile. The spectra were obtained using the Cerro Tololo Ohio State Multi-Object Spectrograph (COSMOS) with the red grism and the 09 slit. This experimental setup corresponds to a dispersion of ∼4 Å pixel⁻¹ over a wavelength range of λ = [5000–8000] Å and a spectral resolution of R ∼ 2100. The data were taken with the slit aligned along the parallactic angle. The seeing was 13 during the first and third night, 1'' during the second night, and 22 in the last night, respectively; all four nights were photometric.

All spectra reported here are obtained by combining at least three individual observations of the source with varying exposure times. This allows us to reduce both instrumental effects and cosmic-ray contribution. The data reduction is done following a standard procedure: the final spectra are all bias subtracted, flat normalized, and corrected for bad pixels. We normalize the flat field to remove any wavelength-dependent variations that could be present in the flat-field source but not in the observed spectrum. This is done by fitting a cubic spline function on the calibration spectrum and taking a ratio of the flat field to the derived fit (see the response function in Massey 1997). We choose an order of >5 for the cubic spline function fit with an χ² value less than 1 to account for all variable features in the flat field. An additional visual inspection is also done on the combined spectra to remove any artificial features that may still be present. This data reduction and spectral extraction is done using the Image Reduction and Analysis Facility (IRAF) pipeline (Tody 1986).

The wavelength calibration for each source is done using the Hg–Ne lamp; we took a lamp spectrum after each observation of a source to avoid potential shifts in the pixel-λ calibration due to changes in the telescope position during the night. Finally, all spectra were flux-calibrated using a spectroscopic standard, which were observed using the same 09 slit used in the rest of the analysis, and then corrected for the Galactic reddening using the extinction law by Cardelli et al. (1989) and the E(B − V) value based on the Schlafly & Finkbeiner (2011) measurements, as reported in the NASA/IPAC Infrared Science Archive.¹

To visually enhance the spectral features of our sources, in Figure 2, we report the normalized spectra of the objects in our sample. These normalized spectra are obtained by dividing the flux-calibrated spectra using a continuum fit (an approach similar to the one reported in Landoni et al. 2018). The continuum is taken to be a power law unless the optical shape is more complex, in which this case, the preferred fit is described in Section 4.1. The S/N of the normalized spectrum is then measured in a minimum of five individual featureless regions in the spectrum with a width of Δλ ≈ 40 Å. The spectral analysis results for each source, including the computed S/N, are reported in Table 2.

Table 2.  Results Obtained from Spectral Analysis Discussed in Section 4

Source	S/N	Spectral Line	Observed λ (Å)	Line Type	Redshift
		Rest Frame λ (Å)
3FHL J0002.1−6728	41.4
3FHL J0935.2−1735	51.5
3FHL J0936.4−2109	27.2	Mg ii (2797)	6176	Absorption	>1.197*
		Mg ii (2803)	6160	Absorption
3FHL J1030.6−2029	29.3	Mg ii (2797)	5579	Absorption	>0.995
		Mg ii (2803)	5591	Absorption
3FHL J1042.8+0055	46.6
3FHL J1130.5−7801	72.2
3FHL J1155.5−3418	42.7	Mg ii (2797)	5174	Absorption	>0.849
		Mg ii (2803)	5185	Absorption
3FHL J1212.1−2328	102.8	O iii (5007)	8345	Emission	0.666
3FHL J1223.5−3033	46.5	Mg ii (2797)	5245	Absorption	>0.875*
		Mg ii (2803)	5256	Absorption
3FHL J1229.7−5304	78.6
3FHL J1315.9−0732	60.8
3FHL J1433.5−7304	64.9	G band (4304)	5165	Absorption	0.200
		Mg i (5175)	6209	Absorption
		Ca+Fe (5269)	6340	Absorption
		Na (5895)	7074	Absorption
		${{\rm{H}}}_{\alpha }$ (6562)	7876	Absorption
3FHL J1439.4−2524	82.7	Mg i (5175)	6008	Absorption	0.16
		Ca+Fe (5269)	6115	Absorption
		NaD (5892)	6835	Absorption
3FHL J1605.0−1140	17.2	O ii (3727)	6801	Emission	0.358*
		(or) O iii (5007)	6801	Emission	0.824*
3FHL J1612.3−3100	75.4
3FHL J1640.1+0629	83.1
3FHL J1842.4−5841	32.7
3FHL J1924.2−1548	64.4
3FHL J2034.9−4200	33.4
3FHL J2041.7−7319	70.1
3FHL J2240.3−5240	71.2
3FHL J2321.8−6437	33.7
3FHL J2339.2−7404	45.5

Note. The redshift measurement values marked with a * are Tentative z measurements.

Download table as:  ASCIITypeset image

To find a redshift measurement, each spectrum was visually inspected for any absorption or emission feature. Any potential feature that matched with known atmospheric lines² was not taken into consideration. To test the reliability of any potential feature, its existence was verified in each of the individual spectral files used to obtain the final combined spectrum shown in Figure 2 or the zoomed spectrum shown in Figure 3. For example, the broad emission feature seen in the spectrum of 3FHL J0935.2-1735 around 5633 Å is not found in the individual files and is thus considered to be an artifact. The verified features are then matched with common blazar lines, such as the Mg ii doublet lines (2797 and 2803 Å) or O iii line (5007 Å), to compute the redshift.

All the sources in our sample were classified as BLL based on their spectral properties. Out of the 23 sources, we were able to determine a redshift measurement for three sources, a lower limit on the redshift for two of them, and a tentative redshift measurement for three of them. The remaining 15 sources in our sample were found to be featureless. Details for some of the sources for which a spectral feature or redshift is found are given in Section 4.1. These features are also listed in Table 2 with the derived redshift measurement.

4.1. Comments on Individual Sources

In this work, we present the results the optical spectroscopic campaign directed toward rendering the 3FHL spectroscopically complete sample using the COSMOS spectrograph mounted on the 4 m Blanco telescope at CTIO in Chile. We observed 23 extragalactic sources classified as BCU (blazars of uncertain classification) in the 3FHL catalog.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

All of the objects in our source sample are classified as BLL based on their observed optical spectrum. In the 3FHL catalog, out of the already classified 901 blazars, ≈84.1% sources are classified as BLL. Moreover out of the 28 sources observed by Marchesi et al. (2018), 27 are identified as BLL, denoting that our results are not surprising.

Out of the 23 BLLs in our sample, we find a reliable redshift measurement for three sources, a reliable redshift constraint for two sources, a tentative redshift constraint for three sources, and a featureless spectrum with no redshift measurement for the remaining 15 sources. Combining our results with the results of Marchesi et al. (2018), our optical spectroscopic campaign reports a redshift measurement for ≈23.5% of the observed BLL sources using 4 m telescopes. This measurement is in line with the expected consistency of 10%–35%, obtained for redshift determination of pure BLL using 4 m telescopes (Landoni et al. 2015; Ricci et al. 2015; Álvarez Crespo et al. 2016a; Peña-Herazo et al. 2017). Moreover, our work combined, with Marchesi et al. (2018), also classifies 51 blazars of previously uncertain classification as either BLL or FSRQs.

The third and fourth part of our spectroscopic campaign will include observations from the 4 m CTIO telescope and the 8 m Gemini-N and Gemini-S telescopes, respectively.³ Additionally, we also aim to extend the campaign by inducing follow-up observations,⁴ similar to Kaur et al. (2019), using the Swift X-ray telescope. These follow-up observations in the X-ray regime will help us confirm the classification of the blazar sources contributing to the spectral completion of the 3FHL catalog.

A.D. acknowledges funding support from NSF through grant AST-1715256. S.M. acknowledges support from NASA contract 80NSSC17K0503. The authors thank Alberto Alvarez and Sean Points for the help provided during the observing nights at CTIO. This work made use of the TOPCAT software (Taylor 2005) for the analysis of data tables.