Mapping the cosmic gamma-ray horizon: the 1CGH catalogue of Fermi-LAT detections above 10 GeV

ABSTRACT

We present the First Cosmic Gamma-ray Horizon (1CGH) catalogue, featuring |$\gamma$|-ray detections above 10 GeV based on 16 yr of observations with the Fermi Large Area Telescope (Fermi-LAT) satellite. After carefully selecting a sample of blazars and blazar candidates from catalogues in the literature, we performed a binned likelihood analysis and identified 2791 |$\gamma$|-ray emitters above 10 GeV, at >3|$\sigma$| level, including 62 that are new |$\gamma$|-ray detections. For each source, we estimated the mean energy of the highest energy bin and analysed them in the context of the cosmic gamma-ray horizon. By adopting a reference model for the extragalactic background light (EBL), we identified a subsample of 525 sources where moderate to severe |$\gamma$|-ray absorption could be detected across the redshift range of 0–3. This work provides the most up-to-date compilation of detections above 10 GeV, along with their redshift information. We condense extensive results from the literature, including reports on observational campaigns dedicated to blazars and |$\gamma$|-ray sources, thereby delivering an unprecedented review of the redshift information for sources detected above 10 GeV. Additionally, we highlight key 1CGH sources where redshift information remains incomplete, offering guidance for future optical observation campaigns. The 1CGH catalogue aims to track the most significant sources for studying the |$\gamma$|-ray transparency of the Universe. Furthermore, it provides a targeted subsample where the EBL optical depth, |$\tau _{(E,z)}$|⁠, can be robustly measured using Fermi-LAT data.

gamma-rays: galaxies, galaxies: active, catalogues, radiation mechanisme: non-thermal

1 INTRODUCTION

Blazars are a rare class of active galactic nuclei (AGNs) characterized by relativistic jets pointing towards us, producing non-thermal emission that spans from radio to TeV |$\gamma$|-rays (Urry & Padovani 1995; Blandford, Meier & Readhead 2019). Ranking among the most luminous and variable sources, blazars provide an unparalleled view into high-energy astrophysical processes, serving as cosmic beacons for studying the transparency of the Universe to |$\gamma$|-rays.

The Fermi Large Area Telescope (Fermi-LAT; Atwood et al. 2009) has revolutionized our understanding of blazars, enabling us to collect data from tens of MeV to hundreds of GeV. Its vast energy coverage allows us to study the interaction of |$\gamma$|-rays with the extragalactic background light (EBL), which leads to attenuation of the very high energy (VHE) spectrum from distant sources (e.g. Ackermann et al. 2013; Fermi-LAT Collaboration 2018). This process is driven by photon–photon annihilation via electron–positron pair production, |$\rm \gamma _{VHE} + \gamma _{EBL} \rightarrow e^{+} + e^{-}$| (e.g. Nikishov 1961; Gould & Schréder 1967), turning |$\gamma$|-rays blazars into unique probes of the EBL density and evolution over cosmic time (e.g. Domínguez & Prada 2013).

Direct EBL measurements are hindered by the presence of foregrounds, including zodiacal light (e.g. Finke et al. 2022). However, distant extragalactic |$\gamma$|-ray sources, such as blazars, provide indirect probes to map the EBL density and distribution over cosmic time (e.g. Stecker, de Jager & Salamon 1992). The EBL density in the far-ultraviolet (far-UV) to near-infrared (NIR) is largely driven by cumulative stellar activity along cosmic time (e.g. Hauser & Dwek 2001; Stecker, Scully & Malkan 2016). Thus, measuring the EBL at a given redshift provides an alternative diagnostic to recover the star formation rate (SFR) from earlier epochs, reaching as far back as the epoch of reionization (EoR; e.g. Robertson et al. 2010) at z |$\sim$| 6.

Given the sensitivity of Fermi-LAT, observations are particularly well suited to measuring the EBL in the UV to optical bands, whereas the NIR can still be constrained at lower redshift. For example, Fermi-LAT Collaboration (2018) demonstrated how distant |$\gamma$|-ray sources can trace the EBL density to z |$\sim$| 3.0, providing unique insight into SFR close to the EoR.

The attenuation of |$\gamma$|-rays can be quantified by the EBL optical depth |$\tau (E, z)$|⁠, which describes the cumulative probability of pair production and provides the relationship between observed and intrinsic flux:

$$\begin{eqnarray} F_{\mathrm{ observed}} = F_{\mathrm{ intrinsic}} \times \exp ^{-\tau _{(E,z)}}, \end{eqnarray}$$

(1)

where a value of |$\tau _{(E,z)} = 1$| implies that approximately 63 per cent of the |$\gamma$|-ray flux at energy E – from a source at redshift z – is absorbed. Measuring |$\tau _{(E, z)}$| is essential to understand the transparency of the Universe to γ-rays, and provides a framework for testing and validation of EBL models.

The attenuation of |$\gamma$|-rays by the EBL defines a statistical boundary for the farthest distance at which VHE photons of a given energy are likely to be detected, known as the cosmic gamma-ray horizon (CGH). The CGH concept was initially explored by Nikishov (1961) and Gould & Schréder (1967), while Fazio & Stecker (1970) – with the Fazio–Stecker relation – advanced the discussion of |$\gamma$|-ray attenuation in context of a cosmological origin for the isotropic |$\gamma$|-ray background.

Interestingly, the VHE spectra of distant blazars can exhibit an unexpected hardening at the highest energies (e.g. Essey & Kusenko 2010; Furniss et al. 2015). This phenomenon has led to alternative hypotheses regarding the propagation of VHE photons, including photon oscillations into axion-like particles (e.g. Galanti et al. 2019; Buehler et al. 2020; Abe et al. 2024) and potential Lorentz invariance violation (e.g. Galanti, Tavecchio & Landoni 2020; Abdalla et al. 2024). These mechanisms offer opportunities to probe fundamental physics and represent open questions in high-energy astrophysics. Therefore, a well-characterized sample of VHE emitters across a broad redshift range is essential, not only to probe EBL and SFR evolution more effectively, but also to investigate alternative photon propagation scenarios.

In this work, we use Fermi-LAT observations spanning the first 16 yr of the mission (from August 2008 to August 2024) to search for |$\gamma$|-ray signatures associated with a sample of blazars and blazar candidates. By focusing on sources detectable above 10 GeV, we aim to provide a catalogue where the EBL absorption is most pronounced and could be effectively measured in the highest energy bins.

To create a robust |$\gamma$|-ray sample suitable for studying EBL attenuation, we calculate the mean energy of the four highest energy photons detected for each source. This approach offers a more stable and representative measure of the highest energy bin (⁠|$E^{\mathrm{ bin}}_{\mathrm{ max}}$|⁠) detectable by Fermi-LAT, particularly when ranking sources that experience moderate to severe absorption, compared to using only the highest energy photon (⁠|$E_{\mathrm{ max}}$|⁠).

The First Cosmic Gamma-ray Horizon (1CGH) catalogue represents a significant step forward in describing the |$\gamma$|-ray horizon, and will contribute to our measurements of the EBL content and SFR across cosmic history.

This work builds upon and complements existing efforts, such as the Third Fermi-LAT Catalog of High-Energy Sources (3FHL; Ajello et al. 2017), by adopting a strategy with the following key aspects: an extended observational period spanning 16 yr, the targeted use of blazar positions as seeds for |$\gamma$|-ray analysis, the inclusion of sources down to a lower detection threshold (i.e. >3|$\sigma$|⁠), and an extensive redshift review. These choices aim to highlight blazars that can serve as valuable targets for investigating the transparency of the Universe to |$\gamma$|-rays.

In addition, we propose a selection criterion that focuses on sources above the |$\tau_{(E, z)} > 0.1$| threshold, and build a subsample specifically targeted to the regime where at least 10 per cent of the |$\gamma$|-ray flux is expected to be absorbed in the highest energy bin.

An essential component of this work is the extensive redshift review of the detected sources. Initially, nearly half of them lacked redshift estimates from the reference catalogues (5BZcat, 3HSP, 3FHL, and 4LAC-DR3). To address this gap, we conducted a comprehensive literature search, covering nearly 70 publications. We gathered information on spectroscopic redshift, lower limit estimates, and uncertain redshift (including photometric estimates reported in 4LAC-DR3) and introduce a redshift quality flag to differentiate between those cases. This effort significantly improved the reliability of our redshift data, providing a stronger foundation for subsequent studies.

By providing a carefully curated sample of blazars with measurable EBL absorption, the 1CGH catalogue complements ongoing studies of TeV-detected sources (e.g. Gréaux et al. 2023) and aims to deepen the scientific discussion on cosmic |$\gamma$|-ray propagation.

2 DATA ANALYSIS AND METHODS

In this section, we discuss the selection of |$\gamma$|-ray candidates, the data analysis set-up using the Fermi Science Tools, and the methods employed for cleaning and associating high-energy photons with sources detected above 10 GeV. We also outline the overall strategy used to ensure robust detections and photon associations, as reported in the 1CGH catalogue.

2.1 Selection of γ-ray candidate sources

To create a sample of sky positions and define our |$\gamma$|-ray targets, we combined several major blazar and AGN catalogues. We began by merging the ‘5th edition of the Roma-BZCAT Multi-Frequency Blazar Catalog’ (5BZcat; Massaro et al. 2015a) with the ‘3rd edition of the High Synchrotron Peak sample’ (3HSP; Chang et al. 2019). These catalogues were chosen for their extensive blazar coverage and reliability.

Both 5BZcat and 3HSP are advanced versions of multifrequency catalogues containing thousands of blazars and blazar candidates, and represent significant repositories for the VHE community. These catalogues have been iteratively refined over time, building upon the initial BZcat (Massaro et al. 2009), and the First and Second Catalogues of High Synchrotron Peaked Blazars and Candidates, 1WHSP and 2WHSP (Arsioli et al. 2015; Chang et al. 2017).

We also included the latest version (V3.4) of TeVcat (Wakely & Horan 2008), which contains a follow-up list of VHE-detected sources (http://tevcat.uchicago.edu). Additionally, to improve the list of seed positions, we incorporate newly confirmed blazars and blazar candidates from recent literature, including Maselli et al. (2013), Paliya et al. (2020), and Ighina et al. (2023, 2024).

We further include all sources classified as AGNs, blazars, and blazar candidates in the ‘Fourth Catalog of Active Galactic Nuclei Detected by the Fermi Large Area Telescope: Data Release 3’ (4LAC-DR3; Ajello et al. 2022), since many of those candidates are not covered by the 5BZcat and 3HSP catalogues.

To remove duplicate entries, we use topcat (Taylor 2005) for internal matching. For the 5BZcat and 3HSP samples, the right ascension (RA) and declination (Dec.) positions, corresponding to the optical and infrared (IR) counterparts, respectively, were used. For TeVcat and 4LAC-DR3 sources, we used the associated counterpart position whenever available (e.g. radio, IR, or optical), as these typically offer better astrometric precision compared to the |$\gamma$|-ray position. With this sample of unique sources, we searched for |$\gamma$|-ray associations in the 'Fermi LAT 14-Year Point Source Catalogue' (4FGL-DR4; Ballet et al. 2023) and the 3FHL catalogue (Ajello et al. 2017), considering the 95 per cent containment region reported for each source.

At this point, we have successfully listed unique seed positions recovered from the 5BZcat, 3HSP, TeVcat, and 4LAC-DR3 catalogues. However, still, many 4FGL-DR4 high-latitude sources (⁠|$|b|>10^\circ$|⁠) were not included in our initial sample of 10 GeV candidates. This is partly due to new |$\gamma$|-ray detections from DR4 that had not yet been incorporated into the latest 4LAC release. However, the main reason is that many of these missing sources are actually unassociated in the 4FGL-DR4 – lacking classification.

Therefore, in this work, we consider all unassociated high-latitude 4FGL-DR4 sources as valid 10 GeV candidates. Tracking detections above 10 GeV is essential, as it may motivate observational campaigns for their identification.

In the final sample of seed positions, we removed all sources flagged as extended and as |$\gamma$|-ray bursts (GRBs). Finally, we used the CLASS1 flag from the 4FGL-DR4 catalogue to exclude cases associated with Galactic sources (e.g. pwn, lmb, hmb, msp, glc, bin, nov, spp, and snr; Ballet et al. 2023). The resulting sample contains nearly 8200 seed positions, with 5023 flagged as blazars or blazar candidates. The remainder are unknown types of sources from 4FGL-DR4 and TeVcat. Given how the positional seeds were selected, the 1CGH catalogue includes unassociated |$\gamma$|-ray sources that require further observational efforts to determine their nature, i.e. to confirm whether they are extragalactic in origin.

2.2 Broad-band likelihood analysis with Fermi-LAT science tools

To search for |$\gamma$|-ray signatures associated with the pre-selected seed positions, we performed a binned likelihood analysis in the 10–800 GeV band, covering the first 16 yr of Fermi-LAT observations (from August 2008 to August 2024). For the broad-band analysis, we used the latest version of the Fermi Science Tools (V2.2.0),¹ with Pass 8 event selection (P8R3; Atwood et al. 2013; Bruel et al. 2018), following the Fermi-LAT’s team recommendation for the identification of point-like sources.

We worked with FRONT + BACK source-class events (evtype = 3 and evclass = 128) and the instrument response function P8R3-SOURCE-V3. For each analysis, we considered a region of interest (ROI) of 15|$^\circ$| radius centred on the |$\gamma$|-ray seed positions. For modelling the background of point and extended sources, we adopted the 4FGL-DR4 v34 catalogue (gll-psc-v34.fit), along with isotropic and Galactic-diffuse templates (iso-P8R3-SOURCE-V3-v1.txt and gll-iem-uw1216-v13.fits, respectively). The spectral parameters – normalization and photon index – were set free for all sources within 5|$^\circ$| of the seed position.

A zenith angle cut of 105|$^\circ$| was applied to avoid contamination from Earth’s limb |$\gamma$|-ray photons, which are produced by cosmic ray interactions with the atmosphere. Using the gtmktime routine, we selected good time intervals when Fermi-LAT was operating in ‘science data-taking mode’, by setting the flags DATA-QUAL |$>$| 0 and LAT-CONFIG == 1. Using the gtbin routine, we generated counts maps (CMAP) and counts cubes (CCUBE) with dimensions of |$300 \times 300$| and |$200 \times 200$| pixels at |$0.1^\circ$| pixel⁻¹, respectively. For the CCUBE, we used 20 equally spaced logarithmic energy bins in the 10–800 GeV range. Each |$\gamma$|-ray candidate was modelled as a point-like source characterized by a power-law spectrum:

$$\begin{eqnarray} \frac{\mathrm{ d}N}{\mathrm{ d}E} \, \text{=} \, N_0 \,\, \left(\frac{E}{E_0} \right)^{-\Gamma } \,\,. \end{eqnarray}$$

(2)

In this equation, |$N_0$| is the normalization constant (prefactor), given in units of [photons cm⁻² ⁻¹ MeV⁻¹], representing the flux density at the pivot energy |$E_{0}$|⁠. The parameter |$\rm \Gamma$| represents the photon spectral index, indicating the slope of the spectrum. To restrict the parameter space probed during the fitting process, we set a limit of 6.5 for the maximum |$\rm \Gamma$| value. This limit is compatible with the largest spectral indices observed in the 3FHL catalogue, including all source types (i.e. Galactic, extragalactic, and unassociated/unknown sources; Ajello et al. 2017).

In our analysis, we adopt a low-energy threshold of 10 GeV – similar to the 3FHL catalogue – due to the improved angular resolution and lower background contamination compared to lower thresholds (Ajello et al. 2017). For the upper energy limit, we set a conservative threshold of 800 GeV, whereas 3FHL extended the analysis to 2 TeV. This decision follows from the discussion in the 4FGL paper (section 3.2 of Abdollahi et al. 2020), which highlights that a broad-band analysis reaching 1 TeV can introduce uncertainties in the energy flux estimate, particularly for hard-spectrum sources. Therefore, our choice for the high-energy threshold is meant to mitigate uncertainties regarding the spectral fitting of all sources in the ROI region.

2.3 Detection threshold for multifrequency-selected targets

The use of multifrequency seed positions to detect |$\gamma$|-ray sources has been successfully applied in numerous works (Arsioli & Chang 2017; Arsioli & Polenta 2018; Arsioli et al. 2018; Arsioli, Chang & Musiimenta 2020) and has proven effective in uncovering extreme blazars close to the Fermi-LAT detection threshold. Using this approach, the analysis for each sky position is independent, investigating a single |$\gamma$|-ray candidate at a time. To handle all 8200 candidates efficiently, we relied on parallel High Performance Computing (HPC) resources provided by ‘National Distributed Computing Infrastructure’ (INCD), Portugal.

Considering our likelihood analysis involves only 2 degrees of freedom (⁠|$\rm \Gamma$| and |$N_{0}$|⁠), we adopted a pre-selection threshold² of TS > 12 (equivalent to a 3.03|$\sigma$| detection³) to include faint sources for future follow-up observations. As suggested in Mattox et al. (1996), the use of multifrequency seeds improves the |$\gamma$|-ray detectability of point-like sources near the detection threshold of high-energy observatories, making it a robust strategy for data exploration, especially regarding blazars, which are known to be the dominant extragalactic population of GeV emitters.

In addition, this relatively low selection threshold aims not only to highlight faint sources but also to capture those that undergo short-lived flare episodes, which might otherwise be missed over a long integration period due to signal dilution (Arsioli & Polenta 2018). As the true signal (i.e. the photon counts from a transient flare) remains constant while the background noise accumulates, a lower TS threshold allows us to include those valuable cases. In short time windows, the significance of these flaring events would be higher, enabling the study of the effects of the EBL absorption. Although we did not perform light-curve analyses in this work, the 1CGH catalogue can serve as a basis for future time-resolved studies.

To assess the robustness of our detection threshold, we estimated the spurious detection rate in our analysis set-up. In a chi-squared (⁠|$\chi ^2$|⁠) distribution with 2 degrees of freedom (d.o.f.), the cumulative probability |$P(\mathrm{ TS}>\mathrm{ TS}_{\mathrm{ threshold}})$| represents the likelihood of obtaining a test statistic above the threshold purely by chance. Using TS = 12, the spurious detection probability is calculated as |$\rm p\_value = 1-\chi ^2 \times cdf(TS=12, \ d.o.f=2) \approx 0.0025$|⁠, where |$\mathrm{ cdf}()$| is the cumulative distribution function. Given that 8200 seed positions were analysed, the expected number of spurious detections in the pre-selection phase is approximately |$\rm 8200\times 0.0025 = 20.5$|⁠. This value represents an upper limit of contamination.

In next section, we discuss additional selection criteria – beyond the selection based on TS level alone – to avoid spurious detections and improve reliability of the final catalogue.

2.4 Additional selection criteria: refining the 1CGH sample

As a result of the likelihood analysis, we pre-selected 3004 sources with TS > 12, which are used as a base for building the final sample. Those sources will still go through an additional selection step that involves looking at the |$\gamma$|-ray events associated with each source. In this stage, it is essential to consider the different event types in the data set and also to identify and remove potential sources of contamination from the event sample.

In the Fermi-LAT Pass 8 release (Atwood et al. 2013), each event is assigned a point spread function (PSF) type, which indicates the quality of the reconstructed direction. Events are categorized into four quartiles: PSF0 (lowest quality, evtype = 4), PSF1 (evtype = 8), PSF2 (evtype = 16), and PSF3 (highest quality, evtype = 32). The PSF 68 per cent containment radius varies by energy and event type. For example, at 30 GeV, the containment radius for PSF1 events is approximately 0.12|$^\circ$|⁠, as described in the Pass 8 documentation.⁴

Using the gtselect routine, we created a subsample of the all-sky Fermi-LAT data, selecting only events with energy greater than 10 GeV. We removed low-quality events by excluding PSF0 events, which represent the quartile with the poorest reconstruction accuracy. Next, we used the gtmktime routine to select events during good time intervals (see Section 2.2). After this step, we were left with a photon sample containing PSF1-2-3 source-type events (i.e. evclass = 128 and evtype = 56), with a maximum zenith angle of 105|$^\circ$|⁠, all recorded under ‘science data-taking mode’.

Before recovering the |$\gamma$|-ray events associated with the pre-selected sources, we applied an additional layer of data cleaning to improve the quality of the photon sample. Specifically, we removed events associated with solar emissions and GRBs. Indeed, the solar disc has recently been confirmed as a source of VHE |$\gamma$|-rays (Linden et al. 2022; Albert et al. 2023; Arsioli & Orlando 2024). Therefore, we removed all events that might be linked to solar emissions by assuming an association radius of 0.8|$^\circ$|⁠.⁵

To identify and remove events related to GRBs, we used the Fermi GBM Burst Catalogue⁶ (von Kienlin et al. 2020). We removed events recorded within 0.8|$^\circ$| of GRB locations, also considering a time window of 30 min before and 10 h after each GRB.

The characteristic position uncertainty of PSF1 events at 30 GeV was used as a basis for defining a cross-matching radius of 0.12|$^\circ$| between the pre-selected candidates and the clean photon sample. This conservative radius ensures robustness when associating higher energy events, particularly given the larger PSF1 size at the lowest energy (approximately 0.18|$^\circ$| at 10 GeV).

As an intermediary selection layer of the 1CGH catalogue, we retained only pre-selected sources (TS > 12) that were associated with at least four high-energy events from the clean photon sample, matching within 0.12|$^\circ$|⁠. This approach follows similar criteria used in building the 3FHL sample (Ajello et al. 2017), which required a minimum of four associated photons for source acceptance (as predicted by the adjusted model, i.e. |$N_{\mathrm{ pred}} \ge 4$|⁠). This procedure removed 213 sources (⁠|$\rm {\sim} 7{{\ \rm per\, cent}}$|⁠) from the 3004 pre-selected ones. We are left with a final sample of 2791 sources, and – given the selection criteria described above – the expected level of spurious detections should be lower than the 0.68 per cent figure (i.e. 20.5/3004) presented in Section 2.3.

3 THE FIRST COSMIC GAMMA-RAY HORIZON CATALOGUE

We present the 1CGH catalogue, which lists 2791 blazars and blazar candidates detected with a TS greater than 12 in the 10–800 GeV energy range, integrated over 16 yr of Fermi-LAT observations. For detailed information on the catalogue’s metadata, refer to Table A1.

The 1CGH catalogue includes 62 |$\gamma$|-ray detections never reported in earlier Fermi-LAT catalogue releases (1FGL, 2FGL, 3FGL, 4FGL, 2FHL, and 3FHL), representing new |$\gamma$|-ray sources. These newly identified sources are mostly high synchrotron peak blazars from the 3HSP catalogue (Chang et al. 2019), and could be especially significant as candidates for VHE observations with the upcoming Cherenkov Telescope Array Observatory (CTAO; Acharya et al. 2013; Cherenkov Telescope Array Consortium et al. 2019). The 3HSP catalogue was designed to identify VHE targets for CTAO, and here we confirm its potential.

Table 1 presents a selection of newly detected sources with TS above 16, including the seed-source name, RA, Dec., redshift, and power-law model parameters associated with each new |$\gamma$|-ray detection.⁷

Table 1.

Open in new tab

Power-law model parameters for the new |$\gamma$|-ray sources, with TS above 16, discovered in the 1CGH catalogue. A complete table with all 2791 1CGH sources is available in the online version of this paper (Note: A preliminary version is available in the Authors’ GitHub repository). The first three columns show the original source names, RA, and Dec. in degrees (J2000), respectively. The fourth column lists redshifts from literature, marked with a ‘(?)’ flag for uncertain or photometric values, and with a ‘>’ symbol for lower limits. The power-law parameters regarding the 1CGH analysis are reported in the following columns, corresponding to the fit (equation 2): the normalization ‘|$N_0$|’ in units of ph cm|$^{-2}$| s⁻¹ MeV⁻¹, the photon spectral index ‘|$\rm \Gamma$|’, and the flux integrated over 10–800 GeV, given in units of ph cm|$^{-2}$| s⁻¹. The pivot energy, ‘|$E_0$|’, is fixed at 15 GeV for all sources. The ‘TS’ column indicates the test statistic value. The final column, |$E_{\mathrm{ max}}^{[\mathrm{ GeV}]}$|⁠, provides the highest energy photon associated with each 1CGH source, based on Pass 8 UltraClean event-class, with PSF = 0 events removed, i.e. evclass = 512 and evtype = 56 (see Section 2.4).

Name	RA (deg)	Dec. (deg)	z	b (deg)	N\|$_0$\| (⁠\|$\times 10^{-16}$\|⁠)	\|$\rm \Gamma$\|	Flux\|$^{10-800\, \rm GeV}_{(\times 10^{-12})}$\|	TS	E\|$^{[\rm GeV]}_{\rm max}$\|
3HSPJ032356.5–010833	50.9855	–1.142 72	0.391	–45.15	37.30 \|$\pm$\| 6.90	2.43 \|$\pm$\| 0.25	69.7 \|$\pm$\| 11.9	167	50.5
3HSPJ120711.5–174605	181.797 87	–17.768 30	>0.7	43.83	15.59 \|$\pm$\| 4.69	2.90 \|$\pm$\| 0.52	26.54 \|$\pm$\| 7.82	44.7	37.2
3HSPJ220451.0–181536	331.213 79	–18.260 11	0.26(?)	–50.75	7.37 \|$\pm$\| 3.21	1.91 \|$\pm$\| 0.36	17.19 \|$\pm$\| 6.28	30.6	51.4
3HSPJ030103.7+344101	45.265 58	34.683 66	0.246	–21.00	8.95 \|$\pm$\| 3.37	2.29 \|$\pm$\| 0.51	17.46 \|$\pm$\| 5.96	30.5	32.1
3HSPJ132635.9+254958	201.649 70	25.833 00	0.698	82.02	4.46 \|$\pm$\| 2.33	1.62 \|$\pm$\| 0.35	12.88 \|$\pm$\| 4.97	28.9	116.8
3HSPJ090802.2–095937	137.009 21	–9.993 69	0.054	24.38	5.41 \|$\pm$\| 1.04	1.80 \|$\pm$\| 0.14	13.50 \|$\pm$\| 2.70	26.7	70.6
3HSPJ231023.4+311949	347.597 37	31.330 33	0.48(?)	–26.77	5.30 \|$\pm$\| 0.31	1.85 \|$\pm$\| 0.11	12.88 \|$\pm$\| 1.15	26.1	51.8
3HSPJ155424.1+201125	238.600 54	20.190 38	0.273	47.76	10.22 \|$\pm$\| 4.11	2.09 \|$\pm$\| 0.47	21.69 \|$\pm$\| 8.78	25.9	138.6
3HSPJ101616.8+410812	154.070 08	41.136 80	0.27	55.32	4.04 \|$\pm$\| 2.22	1.77 \|$\pm$\| 0.39	10.39 \|$\pm$\| 4.41	24.1	164.2
3HSPJ231041.8–434734	347.674 00	–43.792 80	0.089	–63.75	3.86 \|$\pm$\| 0.30	1.71 \|$\pm$\| 0.10	10.37 \|$\pm$\| 0.94	23.6	48.7
3HSPJ094606.1+215138	146.525 54	21.860 66	0.489	47.73	7.27 \|$\pm$\| 2.83	2.94 \|$\pm$\| 0.77	12.34 \|$\pm$\| 4.89	22.0	20.3
3HSPJ091222.9–251825	138.095 45	–25.307 02	0.33(?)	15.62	7.76 \|$\pm$\| 3.04	2.45 \|$\pm$\| 0.50	14.41 \|$\pm$\| 6.00	21.8	52.1
3HSPJ141003.9+051557	212.516 33	5.266 05	0.544	61.21	6.99 \|$\pm$\| 3.33	2.26 \|$\pm$\| 0.51	13.79 \|$\pm$\| 5.65	21.6	68.0
3HSPJ061104.1+682956	92.767 16	68.499 05	0.5(?)	21.53	5.38 \|$\pm$\| 0.76	2.29 \|$\pm$\| 0.16	10.51 \|$\pm$\| 1.57	21.5	58.2
3HSPJ024115.5–304140	40.314 54	–30.694 47	0.3(?)	–65.76	6.73 \|$\pm$\| 1.06	2.25 \|$\pm$\| 0.23	13.32 \|$\pm$\| 2.33	21.3	59.9
3HSPJ102523.0+040229	156.345 95	4.041 50	0.208	48.21	6.34 \|$\pm$\| 0.38	2.51 \|$\pm$\| 0.15	11.57 \|$\pm$\| 1.04	20.5	28.7
3HSPJ121038.3–252713	182.659 70	–25.453 83	0.47(?)	36.50	8.60 \|$\pm$\| 1.95	3.22 \|$\pm$\| 0.47	14.31 \|$\pm$\| 3.25	19.7	26.3
3HSPJ213004.8–563222	322.519 87	–56.539 47	–	–43.91	5.26 \|$\pm$\| 2.47	1.94 \|$\pm$\| 0.39	12.06 \|$\pm$\| 4.80	18.4	57.2
5BZQJ1353+0151	208.464 92	1.864 97	1.608	60.64	2.27 \|$\pm$\| 0.13	1.41 \|$\pm$\| 0.08	8.166 \|$\pm$\| 0.73	18.3	55.0
3HSPJ094502.0–044833	146.258 45	–4.809 38	0.43(?)	34.82	6.95 \|$\pm$\| 2.61	2.78 \|$\pm$\| 0.65	12.04 \|$\pm$\| 4.49	17.8	50.1
3HSPJ120543.3+582933	181.430 29	58.492 66	0.4(?)	57.63	5.15 \|$\pm$\| 2.22	3.44 \|$\pm$\| 0.98	8.539 \|$\pm$\| 3.68	17.8	24.5
3HSPJ213448.2–164205	323.700 91	–16.701 55	0.8(?)	–43.51	7.97 \|$\pm$\| 3.48	3.63 \|$\pm$\| 1.19	13.23 \|$\pm$\| 5.74	17.3	29.5
3HSPJ054903.0–215001	87.264 16	–21.833 69	0.35(?)	–23.05	5.12 \|$\pm$\| 2.45	2.46 \|$\pm$\| 0.66	9.495 \|$\pm$\| 4.17	17.2	66.5
5BZGJ1840–7709	280.160 83	–77.157 97	0.018	–25.80	5.97 \|$\pm$\| 2.57	2.44 \|$\pm$\| 0.52	11.10 \|$\pm$\| 4.54	16.8	29.8
3HSPJ064326.7+421418	100.861 33	42.238 52	0.089	16.47	7.31 \|$\pm$\| 3.28	2.81 \|$\pm$\| 0.72	12.62 \|$\pm$\| 5.34	16.5	68.0
3HSPJ100444.8+375212	151.186 54	37.870 00	0.44	53.58	4.52 \|$\pm$\| 0.65	3.86 \|$\pm$\| 0.47	7.580 \|$\pm$\| 1.11	16.3	13.7
3HSPJ151136.9–165326	227.903 75	–16.890 77	0.36(?)	34.38	8.68 \|$\pm$\| 3.59	3.02 \|$\pm$\| 0.82	14.63 \|$\pm$\| 5.93	16.3	30.8
3HSPJ104745.8+543741	161.940 87	54.628 13	0.54(?)	54.46	4.56 \|$\pm$\| 2.07	2.61 \|$\pm$\| 0.71	8.165 \|$\pm$\| 3.64	16.3	31.7
3HSPJ081941.8+053023	124.924 33	5.506 38	0.37(?)	22.11	7.02 \|$\pm$\| 3.26	2.80 \|$\pm$\| 0.75	12.14 \|$\pm$\| 5.52	16.2	17.0
3HSPJ062753.4–151957	96.972 37	–15.332 52	0.3102	–12.04	5.32 \|$\pm$\| 2.89	1.85 \|$\pm$\| 0.41	12.89 \|$\pm$\| 5.63	16.2	38.8
3HSPJ133326.0+623541	203.358 20	62.594 94	0.48(?)	53.86	1.69 \|$\pm$\| 0.10	1.43 \|$\pm$\| 0.08	5.954 \|$\pm$\| 0.35	16.0	93.2

Name	RA (deg)	Dec. (deg)	z	b (deg)	N\|$_0$\| (⁠\|$\times 10^{-16}$\|⁠)	\|$\rm \Gamma$\|	Flux\|$^{10-800\, \rm GeV}_{(\times 10^{-12})}$\|	TS	E\|$^{[\rm GeV]}_{\rm max}$\|
3HSPJ032356.5–010833	50.9855	–1.142 72	0.391	–45.15	37.30 \|$\pm$\| 6.90	2.43 \|$\pm$\| 0.25	69.7 \|$\pm$\| 11.9	167	50.5
3HSPJ120711.5–174605	181.797 87	–17.768 30	>0.7	43.83	15.59 \|$\pm$\| 4.69	2.90 \|$\pm$\| 0.52	26.54 \|$\pm$\| 7.82	44.7	37.2
3HSPJ220451.0–181536	331.213 79	–18.260 11	0.26(?)	–50.75	7.37 \|$\pm$\| 3.21	1.91 \|$\pm$\| 0.36	17.19 \|$\pm$\| 6.28	30.6	51.4
3HSPJ030103.7+344101	45.265 58	34.683 66	0.246	–21.00	8.95 \|$\pm$\| 3.37	2.29 \|$\pm$\| 0.51	17.46 \|$\pm$\| 5.96	30.5	32.1
3HSPJ132635.9+254958	201.649 70	25.833 00	0.698	82.02	4.46 \|$\pm$\| 2.33	1.62 \|$\pm$\| 0.35	12.88 \|$\pm$\| 4.97	28.9	116.8
3HSPJ090802.2–095937	137.009 21	–9.993 69	0.054	24.38	5.41 \|$\pm$\| 1.04	1.80 \|$\pm$\| 0.14	13.50 \|$\pm$\| 2.70	26.7	70.6
3HSPJ231023.4+311949	347.597 37	31.330 33	0.48(?)	–26.77	5.30 \|$\pm$\| 0.31	1.85 \|$\pm$\| 0.11	12.88 \|$\pm$\| 1.15	26.1	51.8
3HSPJ155424.1+201125	238.600 54	20.190 38	0.273	47.76	10.22 \|$\pm$\| 4.11	2.09 \|$\pm$\| 0.47	21.69 \|$\pm$\| 8.78	25.9	138.6
3HSPJ101616.8+410812	154.070 08	41.136 80	0.27	55.32	4.04 \|$\pm$\| 2.22	1.77 \|$\pm$\| 0.39	10.39 \|$\pm$\| 4.41	24.1	164.2
3HSPJ231041.8–434734	347.674 00	–43.792 80	0.089	–63.75	3.86 \|$\pm$\| 0.30	1.71 \|$\pm$\| 0.10	10.37 \|$\pm$\| 0.94	23.6	48.7
3HSPJ094606.1+215138	146.525 54	21.860 66	0.489	47.73	7.27 \|$\pm$\| 2.83	2.94 \|$\pm$\| 0.77	12.34 \|$\pm$\| 4.89	22.0	20.3
3HSPJ091222.9–251825	138.095 45	–25.307 02	0.33(?)	15.62	7.76 \|$\pm$\| 3.04	2.45 \|$\pm$\| 0.50	14.41 \|$\pm$\| 6.00	21.8	52.1
3HSPJ141003.9+051557	212.516 33	5.266 05	0.544	61.21	6.99 \|$\pm$\| 3.33	2.26 \|$\pm$\| 0.51	13.79 \|$\pm$\| 5.65	21.6	68.0
3HSPJ061104.1+682956	92.767 16	68.499 05	0.5(?)	21.53	5.38 \|$\pm$\| 0.76	2.29 \|$\pm$\| 0.16	10.51 \|$\pm$\| 1.57	21.5	58.2
3HSPJ024115.5–304140	40.314 54	–30.694 47	0.3(?)	–65.76	6.73 \|$\pm$\| 1.06	2.25 \|$\pm$\| 0.23	13.32 \|$\pm$\| 2.33	21.3	59.9
3HSPJ102523.0+040229	156.345 95	4.041 50	0.208	48.21	6.34 \|$\pm$\| 0.38	2.51 \|$\pm$\| 0.15	11.57 \|$\pm$\| 1.04	20.5	28.7
3HSPJ121038.3–252713	182.659 70	–25.453 83	0.47(?)	36.50	8.60 \|$\pm$\| 1.95	3.22 \|$\pm$\| 0.47	14.31 \|$\pm$\| 3.25	19.7	26.3
3HSPJ213004.8–563222	322.519 87	–56.539 47	–	–43.91	5.26 \|$\pm$\| 2.47	1.94 \|$\pm$\| 0.39	12.06 \|$\pm$\| 4.80	18.4	57.2
5BZQJ1353+0151	208.464 92	1.864 97	1.608	60.64	2.27 \|$\pm$\| 0.13	1.41 \|$\pm$\| 0.08	8.166 \|$\pm$\| 0.73	18.3	55.0
3HSPJ094502.0–044833	146.258 45	–4.809 38	0.43(?)	34.82	6.95 \|$\pm$\| 2.61	2.78 \|$\pm$\| 0.65	12.04 \|$\pm$\| 4.49	17.8	50.1
3HSPJ120543.3+582933	181.430 29	58.492 66	0.4(?)	57.63	5.15 \|$\pm$\| 2.22	3.44 \|$\pm$\| 0.98	8.539 \|$\pm$\| 3.68	17.8	24.5
3HSPJ213448.2–164205	323.700 91	–16.701 55	0.8(?)	–43.51	7.97 \|$\pm$\| 3.48	3.63 \|$\pm$\| 1.19	13.23 \|$\pm$\| 5.74	17.3	29.5
3HSPJ054903.0–215001	87.264 16	–21.833 69	0.35(?)	–23.05	5.12 \|$\pm$\| 2.45	2.46 \|$\pm$\| 0.66	9.495 \|$\pm$\| 4.17	17.2	66.5
5BZGJ1840–7709	280.160 83	–77.157 97	0.018	–25.80	5.97 \|$\pm$\| 2.57	2.44 \|$\pm$\| 0.52	11.10 \|$\pm$\| 4.54	16.8	29.8
3HSPJ064326.7+421418	100.861 33	42.238 52	0.089	16.47	7.31 \|$\pm$\| 3.28	2.81 \|$\pm$\| 0.72	12.62 \|$\pm$\| 5.34	16.5	68.0
3HSPJ100444.8+375212	151.186 54	37.870 00	0.44	53.58	4.52 \|$\pm$\| 0.65	3.86 \|$\pm$\| 0.47	7.580 \|$\pm$\| 1.11	16.3	13.7
3HSPJ151136.9–165326	227.903 75	–16.890 77	0.36(?)	34.38	8.68 \|$\pm$\| 3.59	3.02 \|$\pm$\| 0.82	14.63 \|$\pm$\| 5.93	16.3	30.8
3HSPJ104745.8+543741	161.940 87	54.628 13	0.54(?)	54.46	4.56 \|$\pm$\| 2.07	2.61 \|$\pm$\| 0.71	8.165 \|$\pm$\| 3.64	16.3	31.7
3HSPJ081941.8+053023	124.924 33	5.506 38	0.37(?)	22.11	7.02 \|$\pm$\| 3.26	2.80 \|$\pm$\| 0.75	12.14 \|$\pm$\| 5.52	16.2	17.0
3HSPJ062753.4–151957	96.972 37	–15.332 52	0.3102	–12.04	5.32 \|$\pm$\| 2.89	1.85 \|$\pm$\| 0.41	12.89 \|$\pm$\| 5.63	16.2	38.8
3HSPJ133326.0+623541	203.358 20	62.594 94	0.48(?)	53.86	1.69 \|$\pm$\| 0.10	1.43 \|$\pm$\| 0.08	5.954 \|$\pm$\| 0.35	16.0	93.2

Table 1.

Open in new tab

Name	RA (deg)	Dec. (deg)	z	b (deg)	N\|$_0$\| (⁠\|$\times 10^{-16}$\|⁠)	\|$\rm \Gamma$\|	Flux\|$^{10-800\, \rm GeV}_{(\times 10^{-12})}$\|	TS	E\|$^{[\rm GeV]}_{\rm max}$\|
3HSPJ032356.5–010833	50.9855	–1.142 72	0.391	–45.15	37.30 \|$\pm$\| 6.90	2.43 \|$\pm$\| 0.25	69.7 \|$\pm$\| 11.9	167	50.5
3HSPJ120711.5–174605	181.797 87	–17.768 30	>0.7	43.83	15.59 \|$\pm$\| 4.69	2.90 \|$\pm$\| 0.52	26.54 \|$\pm$\| 7.82	44.7	37.2
3HSPJ220451.0–181536	331.213 79	–18.260 11	0.26(?)	–50.75	7.37 \|$\pm$\| 3.21	1.91 \|$\pm$\| 0.36	17.19 \|$\pm$\| 6.28	30.6	51.4
3HSPJ030103.7+344101	45.265 58	34.683 66	0.246	–21.00	8.95 \|$\pm$\| 3.37	2.29 \|$\pm$\| 0.51	17.46 \|$\pm$\| 5.96	30.5	32.1
3HSPJ132635.9+254958	201.649 70	25.833 00	0.698	82.02	4.46 \|$\pm$\| 2.33	1.62 \|$\pm$\| 0.35	12.88 \|$\pm$\| 4.97	28.9	116.8
3HSPJ090802.2–095937	137.009 21	–9.993 69	0.054	24.38	5.41 \|$\pm$\| 1.04	1.80 \|$\pm$\| 0.14	13.50 \|$\pm$\| 2.70	26.7	70.6
3HSPJ231023.4+311949	347.597 37	31.330 33	0.48(?)	–26.77	5.30 \|$\pm$\| 0.31	1.85 \|$\pm$\| 0.11	12.88 \|$\pm$\| 1.15	26.1	51.8
3HSPJ155424.1+201125	238.600 54	20.190 38	0.273	47.76	10.22 \|$\pm$\| 4.11	2.09 \|$\pm$\| 0.47	21.69 \|$\pm$\| 8.78	25.9	138.6
3HSPJ101616.8+410812	154.070 08	41.136 80	0.27	55.32	4.04 \|$\pm$\| 2.22	1.77 \|$\pm$\| 0.39	10.39 \|$\pm$\| 4.41	24.1	164.2
3HSPJ231041.8–434734	347.674 00	–43.792 80	0.089	–63.75	3.86 \|$\pm$\| 0.30	1.71 \|$\pm$\| 0.10	10.37 \|$\pm$\| 0.94	23.6	48.7
3HSPJ094606.1+215138	146.525 54	21.860 66	0.489	47.73	7.27 \|$\pm$\| 2.83	2.94 \|$\pm$\| 0.77	12.34 \|$\pm$\| 4.89	22.0	20.3
3HSPJ091222.9–251825	138.095 45	–25.307 02	0.33(?)	15.62	7.76 \|$\pm$\| 3.04	2.45 \|$\pm$\| 0.50	14.41 \|$\pm$\| 6.00	21.8	52.1
3HSPJ141003.9+051557	212.516 33	5.266 05	0.544	61.21	6.99 \|$\pm$\| 3.33	2.26 \|$\pm$\| 0.51	13.79 \|$\pm$\| 5.65	21.6	68.0
3HSPJ061104.1+682956	92.767 16	68.499 05	0.5(?)	21.53	5.38 \|$\pm$\| 0.76	2.29 \|$\pm$\| 0.16	10.51 \|$\pm$\| 1.57	21.5	58.2
3HSPJ024115.5–304140	40.314 54	–30.694 47	0.3(?)	–65.76	6.73 \|$\pm$\| 1.06	2.25 \|$\pm$\| 0.23	13.32 \|$\pm$\| 2.33	21.3	59.9
3HSPJ102523.0+040229	156.345 95	4.041 50	0.208	48.21	6.34 \|$\pm$\| 0.38	2.51 \|$\pm$\| 0.15	11.57 \|$\pm$\| 1.04	20.5	28.7
3HSPJ121038.3–252713	182.659 70	–25.453 83	0.47(?)	36.50	8.60 \|$\pm$\| 1.95	3.22 \|$\pm$\| 0.47	14.31 \|$\pm$\| 3.25	19.7	26.3
3HSPJ213004.8–563222	322.519 87	–56.539 47	–	–43.91	5.26 \|$\pm$\| 2.47	1.94 \|$\pm$\| 0.39	12.06 \|$\pm$\| 4.80	18.4	57.2
5BZQJ1353+0151	208.464 92	1.864 97	1.608	60.64	2.27 \|$\pm$\| 0.13	1.41 \|$\pm$\| 0.08	8.166 \|$\pm$\| 0.73	18.3	55.0
3HSPJ094502.0–044833	146.258 45	–4.809 38	0.43(?)	34.82	6.95 \|$\pm$\| 2.61	2.78 \|$\pm$\| 0.65	12.04 \|$\pm$\| 4.49	17.8	50.1
3HSPJ120543.3+582933	181.430 29	58.492 66	0.4(?)	57.63	5.15 \|$\pm$\| 2.22	3.44 \|$\pm$\| 0.98	8.539 \|$\pm$\| 3.68	17.8	24.5
3HSPJ213448.2–164205	323.700 91	–16.701 55	0.8(?)	–43.51	7.97 \|$\pm$\| 3.48	3.63 \|$\pm$\| 1.19	13.23 \|$\pm$\| 5.74	17.3	29.5
3HSPJ054903.0–215001	87.264 16	–21.833 69	0.35(?)	–23.05	5.12 \|$\pm$\| 2.45	2.46 \|$\pm$\| 0.66	9.495 \|$\pm$\| 4.17	17.2	66.5
5BZGJ1840–7709	280.160 83	–77.157 97	0.018	–25.80	5.97 \|$\pm$\| 2.57	2.44 \|$\pm$\| 0.52	11.10 \|$\pm$\| 4.54	16.8	29.8
3HSPJ064326.7+421418	100.861 33	42.238 52	0.089	16.47	7.31 \|$\pm$\| 3.28	2.81 \|$\pm$\| 0.72	12.62 \|$\pm$\| 5.34	16.5	68.0
3HSPJ100444.8+375212	151.186 54	37.870 00	0.44	53.58	4.52 \|$\pm$\| 0.65	3.86 \|$\pm$\| 0.47	7.580 \|$\pm$\| 1.11	16.3	13.7
3HSPJ151136.9–165326	227.903 75	–16.890 77	0.36(?)	34.38	8.68 \|$\pm$\| 3.59	3.02 \|$\pm$\| 0.82	14.63 \|$\pm$\| 5.93	16.3	30.8
3HSPJ104745.8+543741	161.940 87	54.628 13	0.54(?)	54.46	4.56 \|$\pm$\| 2.07	2.61 \|$\pm$\| 0.71	8.165 \|$\pm$\| 3.64	16.3	31.7
3HSPJ081941.8+053023	124.924 33	5.506 38	0.37(?)	22.11	7.02 \|$\pm$\| 3.26	2.80 \|$\pm$\| 0.75	12.14 \|$\pm$\| 5.52	16.2	17.0
3HSPJ062753.4–151957	96.972 37	–15.332 52	0.3102	–12.04	5.32 \|$\pm$\| 2.89	1.85 \|$\pm$\| 0.41	12.89 \|$\pm$\| 5.63	16.2	38.8
3HSPJ133326.0+623541	203.358 20	62.594 94	0.48(?)	53.86	1.69 \|$\pm$\| 0.10	1.43 \|$\pm$\| 0.08	5.954 \|$\pm$\| 0.35	16.0	93.2

Name	RA (deg)	Dec. (deg)	z	b (deg)	N\|$_0$\| (⁠\|$\times 10^{-16}$\|⁠)	\|$\rm \Gamma$\|	Flux\|$^{10-800\, \rm GeV}_{(\times 10^{-12})}$\|	TS	E\|$^{[\rm GeV]}_{\rm max}$\|
3HSPJ032356.5–010833	50.9855	–1.142 72	0.391	–45.15	37.30 \|$\pm$\| 6.90	2.43 \|$\pm$\| 0.25	69.7 \|$\pm$\| 11.9	167	50.5
3HSPJ120711.5–174605	181.797 87	–17.768 30	>0.7	43.83	15.59 \|$\pm$\| 4.69	2.90 \|$\pm$\| 0.52	26.54 \|$\pm$\| 7.82	44.7	37.2
3HSPJ220451.0–181536	331.213 79	–18.260 11	0.26(?)	–50.75	7.37 \|$\pm$\| 3.21	1.91 \|$\pm$\| 0.36	17.19 \|$\pm$\| 6.28	30.6	51.4
3HSPJ030103.7+344101	45.265 58	34.683 66	0.246	–21.00	8.95 \|$\pm$\| 3.37	2.29 \|$\pm$\| 0.51	17.46 \|$\pm$\| 5.96	30.5	32.1
3HSPJ132635.9+254958	201.649 70	25.833 00	0.698	82.02	4.46 \|$\pm$\| 2.33	1.62 \|$\pm$\| 0.35	12.88 \|$\pm$\| 4.97	28.9	116.8
3HSPJ090802.2–095937	137.009 21	–9.993 69	0.054	24.38	5.41 \|$\pm$\| 1.04	1.80 \|$\pm$\| 0.14	13.50 \|$\pm$\| 2.70	26.7	70.6
3HSPJ231023.4+311949	347.597 37	31.330 33	0.48(?)	–26.77	5.30 \|$\pm$\| 0.31	1.85 \|$\pm$\| 0.11	12.88 \|$\pm$\| 1.15	26.1	51.8
3HSPJ155424.1+201125	238.600 54	20.190 38	0.273	47.76	10.22 \|$\pm$\| 4.11	2.09 \|$\pm$\| 0.47	21.69 \|$\pm$\| 8.78	25.9	138.6
3HSPJ101616.8+410812	154.070 08	41.136 80	0.27	55.32	4.04 \|$\pm$\| 2.22	1.77 \|$\pm$\| 0.39	10.39 \|$\pm$\| 4.41	24.1	164.2
3HSPJ231041.8–434734	347.674 00	–43.792 80	0.089	–63.75	3.86 \|$\pm$\| 0.30	1.71 \|$\pm$\| 0.10	10.37 \|$\pm$\| 0.94	23.6	48.7
3HSPJ094606.1+215138	146.525 54	21.860 66	0.489	47.73	7.27 \|$\pm$\| 2.83	2.94 \|$\pm$\| 0.77	12.34 \|$\pm$\| 4.89	22.0	20.3
3HSPJ091222.9–251825	138.095 45	–25.307 02	0.33(?)	15.62	7.76 \|$\pm$\| 3.04	2.45 \|$\pm$\| 0.50	14.41 \|$\pm$\| 6.00	21.8	52.1
3HSPJ141003.9+051557	212.516 33	5.266 05	0.544	61.21	6.99 \|$\pm$\| 3.33	2.26 \|$\pm$\| 0.51	13.79 \|$\pm$\| 5.65	21.6	68.0
3HSPJ061104.1+682956	92.767 16	68.499 05	0.5(?)	21.53	5.38 \|$\pm$\| 0.76	2.29 \|$\pm$\| 0.16	10.51 \|$\pm$\| 1.57	21.5	58.2
3HSPJ024115.5–304140	40.314 54	–30.694 47	0.3(?)	–65.76	6.73 \|$\pm$\| 1.06	2.25 \|$\pm$\| 0.23	13.32 \|$\pm$\| 2.33	21.3	59.9
3HSPJ102523.0+040229	156.345 95	4.041 50	0.208	48.21	6.34 \|$\pm$\| 0.38	2.51 \|$\pm$\| 0.15	11.57 \|$\pm$\| 1.04	20.5	28.7
3HSPJ121038.3–252713	182.659 70	–25.453 83	0.47(?)	36.50	8.60 \|$\pm$\| 1.95	3.22 \|$\pm$\| 0.47	14.31 \|$\pm$\| 3.25	19.7	26.3
3HSPJ213004.8–563222	322.519 87	–56.539 47	–	–43.91	5.26 \|$\pm$\| 2.47	1.94 \|$\pm$\| 0.39	12.06 \|$\pm$\| 4.80	18.4	57.2
5BZQJ1353+0151	208.464 92	1.864 97	1.608	60.64	2.27 \|$\pm$\| 0.13	1.41 \|$\pm$\| 0.08	8.166 \|$\pm$\| 0.73	18.3	55.0
3HSPJ094502.0–044833	146.258 45	–4.809 38	0.43(?)	34.82	6.95 \|$\pm$\| 2.61	2.78 \|$\pm$\| 0.65	12.04 \|$\pm$\| 4.49	17.8	50.1
3HSPJ120543.3+582933	181.430 29	58.492 66	0.4(?)	57.63	5.15 \|$\pm$\| 2.22	3.44 \|$\pm$\| 0.98	8.539 \|$\pm$\| 3.68	17.8	24.5
3HSPJ213448.2–164205	323.700 91	–16.701 55	0.8(?)	–43.51	7.97 \|$\pm$\| 3.48	3.63 \|$\pm$\| 1.19	13.23 \|$\pm$\| 5.74	17.3	29.5
3HSPJ054903.0–215001	87.264 16	–21.833 69	0.35(?)	–23.05	5.12 \|$\pm$\| 2.45	2.46 \|$\pm$\| 0.66	9.495 \|$\pm$\| 4.17	17.2	66.5
5BZGJ1840–7709	280.160 83	–77.157 97	0.018	–25.80	5.97 \|$\pm$\| 2.57	2.44 \|$\pm$\| 0.52	11.10 \|$\pm$\| 4.54	16.8	29.8
3HSPJ064326.7+421418	100.861 33	42.238 52	0.089	16.47	7.31 \|$\pm$\| 3.28	2.81 \|$\pm$\| 0.72	12.62 \|$\pm$\| 5.34	16.5	68.0
3HSPJ100444.8+375212	151.186 54	37.870 00	0.44	53.58	4.52 \|$\pm$\| 0.65	3.86 \|$\pm$\| 0.47	7.580 \|$\pm$\| 1.11	16.3	13.7
3HSPJ151136.9–165326	227.903 75	–16.890 77	0.36(?)	34.38	8.68 \|$\pm$\| 3.59	3.02 \|$\pm$\| 0.82	14.63 \|$\pm$\| 5.93	16.3	30.8
3HSPJ104745.8+543741	161.940 87	54.628 13	0.54(?)	54.46	4.56 \|$\pm$\| 2.07	2.61 \|$\pm$\| 0.71	8.165 \|$\pm$\| 3.64	16.3	31.7
3HSPJ081941.8+053023	124.924 33	5.506 38	0.37(?)	22.11	7.02 \|$\pm$\| 3.26	2.80 \|$\pm$\| 0.75	12.14 \|$\pm$\| 5.52	16.2	17.0
3HSPJ062753.4–151957	96.972 37	–15.332 52	0.3102	–12.04	5.32 \|$\pm$\| 2.89	1.85 \|$\pm$\| 0.41	12.89 \|$\pm$\| 5.63	16.2	38.8
3HSPJ133326.0+623541	203.358 20	62.594 94	0.48(?)	53.86	1.69 \|$\pm$\| 0.10	1.43 \|$\pm$\| 0.08	5.954 \|$\pm$\| 0.35	16.0	93.2

In Fig. 1, we show the distribution of the detection significance for the entire 1CGH and 3FHL catalogues. This representation is meant as a qualitative overview on both samples, keeping in mind that the 1CGH focuses on seed positions of blazars and blazar candidates, while the 3FHL includes detections of all source types. The number of detections in 1CGH shows consistent growth along the significance bins (i.e. following a trend similar to 3FHL), and includes an extra bin covering the 3 < |$\sigma$| < 4 range.

$The distribution of detection significance ($\sigma$) for the 1CGH and 3FHL samples (respectively, blue line and red dashed line).$

Figure 1.

The distribution of detection significance (⁠|$\sigma$|⁠) for the 1CGH and 3FHL samples (respectively, blue line and red dashed line).

Open in new tab Download slide

We should note that in the full 1CGH catalogue, 26 sources⁸ reached the spectral index limit of |$\Gamma = 6.5$| constrained by our likelihood analysis set-up (see Section 2.2). Moreover, 27 1CGH sources have counterparts in 3FHL where spectral curvature has been identified.⁹ For these 53 (26 + 27) cases, the derived fluxes should be interpreted as upper limits. While this affects only a small fraction of the sample, the vast majority of sources exhibit well-constrained spectral fits, reinforcing the robustness of the 1CGH catalogue for high-energy studies.

In Fig. 2, we present the photon index (⁠|$\Gamma$|⁠) versus the integrated flux derived from a power-law fit over the 10–800 GeV energy range. The plot shows that the 1CGH sample extends the 3FHL coverage towards fainter fluxes, effectively lowering the sensitivity limit for detections above 10 GeV. As in the 3FHL catalogue, the detection threshold in the 1CGH sample shows little dependence on photon index, with the new identifications clustering at the lowest fluxes. Moreover, the overlap between the 1CGH and 4FGL sources highlights a robust improvement in completeness above 10 GeV, which is complemented by the newly identified sources.

$The photon index ($\Gamma$) versus the integral flux in the 10–800 GeV energy band. The green points represent 1CGH sources with counterparts in both 3FHL and 4FGL-DR4 catalogues; the blue open diamonds represent 1CGH sources with counterparts in 4FGL-DR4 only; and the magenta circles represent new $\gamma$-ray identifications.$

Figure 2.

The photon index (⁠|$\Gamma$|⁠) versus the integral flux in the 10–800 GeV energy band. The green points represent 1CGH sources with counterparts in both 3FHL and 4FGL-DR4 catalogues; the blue open diamonds represent 1CGH sources with counterparts in 4FGL-DR4 only; and the magenta circles represent new |$\gamma$|-ray identifications.

Open in new tab Download slide

3.1 Extensive redshift review

In addition to listing detections above 10 GeV, we conducted an extensive literature review to improve the redshift characterization of our sample, which is crucial for investigating the cosmic |$\gamma$|-ray horizon. Based on the latest studies, we revised and updated redshift information, tracked the origin and quality of the data, and incorporated new redshifts from multiple optical campaigns dedicated to characterizing blazars, 3FHL, and 4FGL sources. To track redshift quality, we added a column named ‘z-flag’ to the catalogue, with values (1) for robust spectroscopic redshift; (2) for photometric estimates or uncertain values reported in literature; and (3) for lower limit redshift values.

In the 1CGH catalogue, we added a column named ‘z-origin’ to track the origin of the redshift values reported in the ‘z’ column. Redshift information was gathered from several major blazar and |$\gamma$|-ray catalogues, including 5BZcat (Massaro et al. 2015a), 3HSPcat (Chang et al. 2019), 4LACdr3 (Ajello et al. 2022), and TeVcat (Wakely & Horan 2008).

Additionally, redshift values were extracted from dedicated observational campaigns, such as Boyle90 (Boyle et al. 1990); Appenzeller98 (Appenzeller et al. 1998); Sbarufatti05 (Sbarufatti et al. 2005); Sbarufatti06 (Sbarufatti et al. 2006); Sbarufatti09 (Sbarufatti et al. 2009); Maisner10 (Meisner & Romani 2010); Pulido10 (Acosta-Pulido et al. 2010); Shaw12 (Shaw et al. 2012); Furniss13 (Furniss et al. 2013); Landoni13 (Landoni et al. 2013); Maselli13 (Maselli et al. 2013); Rovero13 (Rovero et al. 2013); Sadrinelli13 (Sandrinelli et al. 2013); Shaw13 (Shaw et al. 2013); Paggi14 (Paggi et al. 2014); Ricci15 (Ricci et al. 2015); AlvarezCrespo16a-b-c (Álvarez Crespo et al. 2016a, b, c); Kaur16 (Kaur et al. 2017); Rovero16 (Rovero et al. 2015, 2016); Juanita17 (Torres-Zafra et al. 2018); Paiano17 (Paiano et al. 2017a, b, c); Gabanyi18 (Gabányi, Frey & An 2018); Landoni18 (Landoni et al. 2015, 2018); Massaro18 (Massaro et al. 2014, 2015b, 2016); Mishra18 (Mishra et al. 2018); Balmaverde19 (Balmaverde et al. 2020); Caccianiga19 (Caccianiga et al. 2019); Johnson19 (Johnson et al. 2019); Desai19 (Desai et al. 2019); Menezes19 (de Menezes et al. 2019, 2020); Marchesini19 (Marchesini et al. 2019); Paiano19 (Paiano et al. 2019); Belladitta20 (Belladitta et al. 2020); Goldoni20 (Goldoni et al. 2015, 2021); Landoni20¹⁰ (Landoni et al. 2020); Pena-Herazo20 (Peña-Herazo et al. 2017, 2019, 2020); Paiano20 (Paiano et al. 2019, 2020); Raiteri20 (Raiteri et al. 2020); B.Gonzalez21 (Becerra González et al. 2021); Paliya21 (Paliya et al. 2021); Pena-Herazo21 (Peña-Herazo et al. 2021a, b); Rajagopal21 (Rajagopal et al. 2021); Foschini22 (Foschini et al. 2022); Kasai22 (Kasai et al. 2023a, b); OlmoGarcia22 (Olmo-García et al. 2022); Rajagopal22 (Rajagopal et al. 2023); Paiano23 (Paiano et al. 2023); dAmmando24 (D’Ammando et al. 2012, 2024); Sarira24 (Sahu et al. 2024); DESI-EDR (DESI Collaboration 2024); and AlvarezCrespo25 (Álvarez Crespo et al. 2025). The numerous dedicated optical observation campaigns highlight the ongoing high demand for the characterization of blazars and blazar candidates, particularly in connection with their |$\gamma$|-ray counterparts.

Starting from the redshift information available in the reference catalogues (i.e. 5BZcat, 3HSP, 4LAC-DR3, and TeVcat), our literature review allowed us to assign or update redshift values for 967 1CGH sources – including 377, 437, and 153 sources with z-flags 1, 2, and 3, respectively. Of the total 2791 1CGH sources, 1062 (38.2 per cent) have spectroscopically measured redshift values, 855 (30.6 per cent) have uncertain/photometric values, 210 (7.5 per cent) have lower limit values, and 664 (23.7 per cent) have no available redshift information.

In Fig. 3, we show the redshift distribution for the 1CGH subsample with z-flag = 1 (i.e. robust spectroscopic redshift) and for the entire 3FHL sample. The comparison reveals that the number of sources detected above 10 GeV has improved along redshift relative to 3FHL. In addition, our redshift characterization aims to complement and refine previous efforts (e.g. 3FHL and 4LAC-DR3) by incorporating redshift quality flags and systematically tracking the reference from which each redshift value was obtained.¹¹

Figure 3.

The redshift distribution for the 1CGH and 3FHL samples. For the 1CGH, the histogram only includes sources with spectroscopic redshift determination (z-flag = 1); for the 3FHL catalogue, we include the entire sample.

Open in new tab Download slide

Extensive redshift characterization is key for samples used to study the CGH, as it helps to mitigate possible biases caused by missing or uncertain redshift values. The best way to reduce uncertainties when measuring the EBL density evolution is to increase the number of sources with robust redshift determinations. In particular, identifying absorbed sources above redshift of 2 would significantly improve EBL measurement at a regime that impacts SFR density estimates near and within the EoR, at z |$\sim$| 6–7 (Fermi-LAT Collaboration 2018). Among the most distant 1CGH sources (at |$2.4\le z \le 2.9$|⁠) are 5BZQJ0601−7036, 5BZQJ0228−5546, 5BZQJ1344−1723, 5BZQJ1748+3404, 5BZQJ1441−1523, 5BZQJ0912+4126, 5BZQJ1345+4452, and 5BZBJ0022+0608, which could contribute to studies of the EoR.

With improved precision in EBL density measurements, we can potentially disentangle the contributions of star formation and AGN to the EBL build-up at high redshift, turning blazars into an effective tool to understand the role played by AGNs during the EoR (e.g. D’Silva et al. 2023).

3.2 The cosmic gamma-ray horizon plot

Plotting the highest energy photons from the 1CGH catalogue against redshift reveals that the Universe is indeed opaque to |$\gamma$|-rays, as predicted by Nikishov (1961) and Gould & Schréder (1967). As previously discussed, the energy versus redshift relationship, where the opacity |$\tau (E,z) = 1.0$|⁠, defines what is known as the ‘CGH’. Beyond this horizon, high-energy photons are severely attenuated by the EBL, rendering the Universe effectively opaque to |$\gamma$|-rays (e.g. Domínguez et al. 2013).

Fig. 4 shows the highest energy photon versus redshift for the 1CGH sources, along with the predicted γ-ray horizon according to the main EBL models (e.g. Finke, Razzaque & Dermer 2010; Domínguez et al. 2011; Franceschini & Rodighiero 2017; Saldana-Lopez et al. 2021). Different markers are used to represent robust (spectroscopic), photometric/uncertain, and lower limit redshift values to emphasize the importance of accurate redshift determination.

$The CGH, showing the highest energy photon versus redshift, based on Fermi-LAT ultra-clean events with PSF0 events removed (evclass = 512 and evtype = 56). Sources detected above 10 GeV are represented, with robust (spectroscopic) redshifts marked as filled circles; lower limit redshifts are represented by right arrows, with the reported redshift value corresponding to the centre of the arrow, and uncertain redshifts (photometric or reported as doubtful) indicated by blue crosses. The EBL optical depth $\tau _{(E,z)} = 1.0$ is depicted for different models: Finke et al. (2010) (blue dashed line), Domínguez et al. (2011) (black dotted line), Franceschini & Rodighiero (2017) (grey line), and Saldana-Lopez et al. (2021) (magenta dot–dashed line). For better visualization, the photon energy of sources with robust redshift is colour-coded. Note that approximately one-third of the 1CGH sources lack assigned redshift and are therefore not included in this plot.$

Figure 4.

The CGH, showing the highest energy photon versus redshift, based on Fermi-LAT ultra-clean events with PSF0 events removed (evclass = 512 and evtype = 56). Sources detected above 10 GeV are represented, with robust (spectroscopic) redshifts marked as filled circles; lower limit redshifts are represented by right arrows, with the reported redshift value corresponding to the centre of the arrow, and uncertain redshifts (photometric or reported as doubtful) indicated by blue crosses. The EBL optical depth |$\tau _{(E,z)} = 1.0$| is depicted for different models: Finke et al. (2010) (blue dashed line), Domínguez et al. (2011) (black dotted line), Franceschini & Rodighiero (2017) (grey line), and Saldana-Lopez et al. (2021) (magenta dot–dashed line). For better visualization, the photon energy of sources with robust redshift is colour-coded. Note that approximately one-third of the 1CGH sources lack assigned redshift and are therefore not included in this plot.

Open in new tab Download slide

Fig. 4 provides an unprecedented view of the redshift characterization of sources used to study the CGH. It highlights the significance of reaching an extensive redshift description of the entire 1CGH sample and aims to further motivate the community’s ongoing efforts in the optical characterization of these rare sources (see Section 3.1).

Here, we draw attention to the number of 1CGH sources that lack robust redshift characterization. As mentioned in Section 3.1, a total of 1729 1CGH sources (⁠|$\sim$|72 per cent) are assigned redshift values that are either uncertain, lower limits, or are absent. All of those cases would benefit from dedicated observational campaigns to improve the overall redshift description, therefore representing a significant challenge for CGH studies and the VHE community as a whole.

3.3 Identification of sources with detectable EBL absorption

To measure the EBL optical depth, one needs to build the |$\gamma$|-ray spectrum and compare the observed flux with the expected flux as a function of energy. For sources observed with Fermi-LAT, the |$\gamma$|-ray spectrum typically spans from tens of MeV to hundreds of GeV, and the flux can only be resolved for a few energy bins (⁠|$E^{\mathrm{ bin}}$|⁠) (i.e. due to sensitivity limitations, Fermi-LAT usually cannot achieve high spectral resolution).

If we look for the most interesting sources for studying the CGH, we should focus on estimating the largest energy bin (⁠|$E^{\mathrm{ bin}}_{\mathrm{ max}}$|⁠) where a source can still be detected. Any source with at least one energy bin that experiences a detectable degree of EBL attenuation is relevant to measure |$\tau _{(E,z)}$| values.

To emphasize sources that could be used to measure |$\tau _{(E,z)}$|⁠, we adopt a selection method based on the (z, |$E^{\mathrm{ bin}}_{\mathrm{ max}}$|⁠) position in the energy versus redshift plane. We calculate |$E^{\rm{bin}}_{\rm{max}}$| as the mean energy of the four highest energy photons associated with a source, to create a robust estimate of the highest energy bin. A similar requirement for the minimum photon counts was applied in the 3FHL catalogue (Ajello et al. 2017) to set a lower acceptance level for robust source detections. When calculating |$E^{\rm{bin}}_{\rm{max}}$|⁠, we use the clean photon sample (as described in Section 2.4), considering source-type events and excluding PSF0 events (i.e. evclass = 128 and evtype = 56).

In Fig. 5, we illustrate the energy versus redshift relation and use the Saldana-Lopez EBL model (Saldana-Lopez et al. 2021) as a reference framework to identify sources where γ-ray absorption might be detectable. The Saldana-Lopez model offers an empirical determination of the evolving EBL spectrum and its uncertainties up to z |$\sim$| 6; it is based on galaxy counts and multifrequency data (from UV to far-IR) covering the Hubble Space Telescope’s ‘Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey’ (CANDELS), and was designed to minimize uncertainties, especially at higher redshift (Grogin et al. 2011; Koekemoer et al. 2011).

$The CGH, showing the ‘largest energy bin detectable with Fermi-LAT’ versus redshift. The ‘largest energy bin’ is calculated as the mean energy of the four highest energy source-type events associated with each source, excluding PSF0 events (i.e. evclass = 128 and evtype = 56). The opacity regimes for EBL optical depths $\tau _{(E,z)} = 1.0$ and $\tau _{(E,z)} = 0.1$ are represented by grey dashed and magenta dot–dashed lines, respectively, corresponding to the Saldana-Lopez model (Saldana-Lopez et al. 2021). This plot highlights sources in the $\tau _{(E,z)} > 0.1$ regime where absorption may be detectable at the highest energy bin observed with Fermi-LAT, while sources in the $\tau < 0.1$ regime are represented by grey markers. Different markers indicate the quality of the redshift determination: robust (spectroscopic) redshifts are shown as filled circles; lower limit redshifts are represented by right arrows, with the reported redshift value corresponding to the centre of the arrow; and uncertain redshifts (photometric or unknown quality) are represented by blue crosses. Approximately one-third of the 1CGH sources lack redshift assignments and are therefore not included in this plot. Note: The $\tau _{(E,z)}$ values regarding the largest energy bin are listed in the 1CGH catalogue to help identify interesting targets.$

Figure 5.

The CGH, showing the ‘largest energy bin detectable with Fermi-LAT’ versus redshift. The ‘largest energy bin’ is calculated as the mean energy of the four highest energy source-type events associated with each source, excluding PSF0 events (i.e. evclass = 128 and evtype = 56). The opacity regimes for EBL optical depths |$\tau _{(E,z)} = 1.0$| and |$\tau _{(E,z)} = 0.1$| are represented by grey dashed and magenta dot–dashed lines, respectively, corresponding to the Saldana-Lopez model (Saldana-Lopez et al. 2021). This plot highlights sources in the |$\tau _{(E,z)} > 0.1$| regime where absorption may be detectable at the highest energy bin observed with Fermi-LAT, while sources in the |$\tau < 0.1$| regime are represented by grey markers. Different markers indicate the quality of the redshift determination: robust (spectroscopic) redshifts are shown as filled circles; lower limit redshifts are represented by right arrows, with the reported redshift value corresponding to the centre of the arrow; and uncertain redshifts (photometric or unknown quality) are represented by blue crosses. Approximately one-third of the 1CGH sources lack redshift assignments and are therefore not included in this plot. Note: The |$\tau _{(E,z)}$| values regarding the largest energy bin are listed in the 1CGH catalogue to help identify interesting targets.

Open in new tab Download slide

However, our ultimate goal is to highlight a sample in which the |$\tau _{(E,z)}$| values can be derived from the data, independently of any specific EBL model. The estimates of Saldana-Lopez et al. (2021) are similar to those by Finke et al. (2010) and both are representative of the most severe attenuation models among the ones considered.

We used an EBL optical depth of |$\tau _{(E,z)} = 0.1$| as the lower limit to introduce an absorption flag (ABS-flag), indicating whether spectral absorption might be detectable at the highest energy bins. For cases where |$E^{\rm{bin}}_{\rm{max}}$| lies in the |$\tau _{(E,z)} > 0.1$| regime, the largest energy bin observed with Fermi-LAT likely experiences a detectable absorption greater than 10 per cent [i.e. >(⁠|$\rm 1-e^{-0.1}$|⁠)]. These sources can be used to investigate the opacity along redshift, and have ABS-flag set to ‘1’.

For sources with |$E^{\rm{bin}}_{\rm{max}}$| in the |$\tau _{(E,z)} < 0.1$| regime, the Fermi-LAT spectrum is likely unaffected by EBL attenuation, and these sources have an ABS-flag set to ‘0’ in the 1CGH catalogue. These sources can be used to study the intrinsic |$\gamma$|-ray spectrum from blazars, helping to refine assumptions regarding their intrinsic flux.

The |$\tau _{(E,z)}$| values – calculated using the Saldana-Lopez EBL model at |$E^{\rm{bin}}_{\rm{max}}$| (Saldana-Lopez et al. 2021) – are listed in the column ‘|$\rm \tau _{Ebin}$|’. Furthermore, the absorption fraction [|$1-\mathrm{ e}^{-\tau _{(E,z)}}$|] is recorded in the ‘|$\rm Abs_{Ebin}$|’ column, which can be used to create subsamples based on different absorption thresholds. In particular, for sources with multiple energy bins that experience detectable absorption, a single source can provide |$\tau _{(E,z)}$| estimates for multiple energy levels at a given redshift.

3.4 Best candidates for optical observation campaigns

Following our extensive review of redshift information (Section 3.1), Appendix B highlights the high galactic latitude sources (|b| > 10|$^{\circ }$|⁠) that are currently the best candidates for optical observations. We have identified two main types of sources to prioritize, which will further improve the catalogue and the science described above:

1CGH sources that currently lack redshift information (z-flag = 0) but already have an optical or radio association are prioritized based on |$E^{\rm{bin}}_{\rm{max}}$|⁠, the highest energy bin detected with Fermi-LAT. These sources are particularly significant due to the presence of a clear counterpart for optical follow-up and their potential proximity to the |$\rm \tau \sim 1$| horizon. Table B1 highlights 42 of these cases.
1CGH sources with lower limit redshifts or photometric/uncertain redshifts (z-flag = 2 or 3) are prioritized based on their calculated opacity (⁠|$\rm \tau _{Ebin}$|⁠), using |$E^{\rm{bin}}_{\rm{max}}$| values. These sources could be experiencing significant |$\gamma$|-ray absorption, potentially detectable across multiple energy bins, and thus could provide valuable |$\tau {_{(E)}}$| measurements across different energies at a given redshift. Table B2 highlights 43 of these cases.

Any candidate for follow-up could also be considered in light of the redshift upper limits proposed by Domínguez et al. (2024), which use EBL attenuation to constrain the distance of |$\gamma$|-ray blazars.

4 SUMMARY AND CONCLUSIONS

In this work, we presented the 1CGH catalogue, which lists all-sky Fermi-LAT detections above 10 GeV, compiled from 16 yr of observations. The catalogue includes 2791 sources, with 62 representing new |$\gamma$|-ray detections, and to date, it represents a valuable sample to study the transparency of the Universe to VHE photons.

We significantly improved the redshift characterization of our catalogue by incorporating new spectroscopic redshift estimates whenever possible. By meticulously reviewing nearly 70 dedicated observational campaigns, we significantly reduced the redshift knowledge gap, introducing a z-flag system to categorize redshift reliability.

The 1CGH catalogue highlights sources experiencing varying degrees of |$\gamma$|-ray attenuation and will contribute to a better determination of the EBL’s density and its evolution over cosmic time. Using the Saldana-Lopez EBL model as a reference framework, we identified sources with |$\gamma$|-ray absorption above the |$\tau _{(E,z)} > 0.1$| limit, creating a targeted sample where absorption is likely measurable.

In addition, we introduced a complementary data exploration method by calculating the mean energy of the four highest energy photons |$E^{\rm{bin}}_{\rm{max}}$|⁠, rather than relying solely on the highest energy event. This provides a more stable and statistically reliable measure of the characteristic ‘largest energy level’ observed for each source.

We identified sources that should be prioritized for follow-up optical observation campaigns. These include (i) sources that lack redshift information but have a clear optical or radio association, prioritized based on the highest |$E^{\rm{bin}}_{\rm{max}}$| values, and (ii) sources with lower limit or uncertain redshift estimates that could benefit from further spectroscopic exploration, prioritized according to their |$\tau$| values measured at |$E^{\rm{bin}}_{\rm{max}}$|⁠. These targets will not only help refine the redshift description of the catalogue but also improve the accuracy of EBL measurements at higher redshift, thus reducing uncertainties in SFR density determination during the EoR, at z |$\sim$| 6–7.

Following our redshift review (Section 3.1), approximately 72 per cent of the 1CGH sources still lack robust redshift determination. This significant gap poses challenges for the VHE community as a whole, and calls attention to the need for extensive follow-up optical observation campaigns targeting |$\gamma$|-ray blazars.

ACKNOWLEDGEMENTS

The authors thank the anonymous referee for the careful review and helpful comments that improved the quality – and impact – of our work. BA thanks the Institute of Astrophysics and Space Sciences (IA) at the University of Lisbon for their ongoing support. BA is currently a Marie Skłodowska-Curie Postdoctoral Fellow at IA, funded by the European Union’s Horizon 2020 research and innovation programme under the MSCA agreement no. 101066981. BA also acknowledges the support from ‘Fundação para a Ciência e a Tecnologia’ (FCT) through the research grant UIDP/04434/2020 (DOI: 10.54499/UIDP/04434/2020) in support of general IA activities. This work was produced with the support of the INCD, Portugal (https://www.incd.pt/), funded by FCT and FEDER under project 01/SAICT/2016 no. 022153. We acknowledge FCT’s support through the A1 Computation Call (project’s doi: 10.54499/2023.09549.CPCA.A1) for the allocation of HPC resources at INCD.Support for this work was provided by the National Aeronautics and Space Administration through Chandra Award Number GO3-24069X issued by the Chandra X-ray Center, which is operated by the Smithsonian Astrophysical Observatory for and on behalf of the National Aeronautics Space Administration under contract NAS8-03060. We are grateful to the entire Fermi-LAT collaboration for maintaining a publicly accessible mission data base, enabling discoveries across the entire |$\gamma$|-ray community. We also used archival data and bibliographic information from the NASA-IPAC Extragalactic Database (NED), and data/software facilities maintained by the Space Science Data Center (SSDC) from the Italian Space Agency. We extend our thanks to the TeVCat team, especially Deirdre Horan and Scott Wakely, for their continuous work on VHE detections, as well as NASA and NSF for their support of these follow-up efforts.

DATA AVAILABILITY

All catalogues used to build the sample of |$\gamma$|-ray candidates are publicly available and cited within this work. The 1CGH catalogue will be made publicly available through VizieR at https://vizier.cds.unistra.fr/, and GitHub. The Fermi-LAT data base and the Science Tools used to build the 1CGH catalogue are also publicly accessible at https://fermi.gsfc.nasa.gov/ssc/data/access/.

Footnotes

The Fermitools-Conda repository is available at https://github.com/fermi-lat/Fermitools-conda. The Fermi-LAT data analysis recommendations are available at https://fermi.gsfc.nasa.gov/ssc/data/analysis/.

The TS parameter is defined as |$-2 \ln \left(L_{\left(\mathrm{ no \hbox{-} source} \right)} \div L_{\left(\mathrm{ source} \right)} \right)$|⁠, where |$L_{ \left(\mathrm{ no \hbox{-} source} \right)}$| represents the likelihood of observing a given photon count due to background alone (the null hypothesis), and |$L_{\left(\mathrm{ source} \right)}$| represents the likelihood of observing the photon count assuming that a source exists at a particular position (Mattox et al. 1996).

See the repository TS-DegFreedom-Sigma-relation-FermiLAT at github.com/BrunoArsioli with a python implementation for the TS to sigma relation, while accounting for the number of degrees of freedom in the analysis.

Fermi-LAT Pass 8 PSF: https://www.slac.stanford.edu/exp/glast/groups/canda/lat_Performance.htm.

Considering the solar disc has an angular radius of |${\sim} 0.26^\circ$|⁠, and the characteristic position uncertainty for PSF1 events at 10 GeV is around 0.18|$^\circ$|⁠, a 0.8|$^\circ$| radius cut safely encompasses and removes solar disc contamination.

The continuously updated GBM Burst Catalogue, also referred to as ‘The FERMIGBRST data base’, is available at https://heasarc.gsfc.nasa.gov/W3Browse/fermi/fermigbrst.html.

The full version of the 1CGH catalogue will be available in public data repositories such as Vizier (https://cdsarc.cds.unistra.fr/viz-bin/cat/J/MNRAS/539/1458) and Author’s GitHub.

Comprising 3 unassociated sources, 4 BL Lacertae, 8 flat spectrum radio quasars, and 11 blazars of unknown type.

To help identify such cases, we include the Curvature_3FHL column to our final catalogue, which tracks whether spectral curvature was identified in 3FHL, and might affect broad-band flux estimates.

ZBLLac Database: ‘A spectroscopic library of BL Lac objects’ at https://web.oapd.inaf.it/zbllac/.

Please refer to section 5.2 of Ajello et al. (2020) regarding redshift contamination in 4LAC, given the lack of flags to differentiate between spectroscopic and photometric redshift values.

For 4FGLJ0351.2−6103, the AllWISE IR counterpart is tentative; and for 4FGLJ2127.6−5959, a tentative IR counterpart is close by NGC7059.

REFERENCES

Abdalla

Cotter

Backes

Kasai

Böttcher

2024

Class. Quantum Gravity

015022

Column name	Description
1CGH_name	Source identifier, formatted as 1CGHJ0123+0123, based on J2000 coordinates. It includes four digits each for RA and Dec. For associated sources, the coordinates are from the counterpart; for unassociated sources, we follow 4FGL-DR4 astrometry.
RA_1CGH	RA (J2000) in degrees.
DEC_1CGH	Dec. (J2000) in degrees.
Counterpart_name	Name of the source from catalogues such as 5BZcat, 3HSP, 4FGL-DR4, or others.
N0 (1E−16)	Normalization constant of the power-law fit in units of photons cm\|$^{-2}$\| s⁻¹ MeV⁻¹ at the pivot energy.
N0_err (1E−16)	Uncertainty in the normalization constant.
Index	Photon spectral index of the power-law model.
Index_err	Uncertainty in the photon spectral index.
Flux (1E−12)	Integrated photon flux (10–800 GeV) in units of photons cm\|$^{-2}$\| s⁻¹, estimated with a power-law fit.
Flux_err (1E−12)	Uncertainty in the integrated photon flux.
TS	Test statistic value for the source detection.
sigma	The significance for the source detection.
z	Redshift of the source.
zflag	Redshift reliability flag: (1) spectroscopic, (2) photometric/uncertain, and (3) lower limit.
z_origin	Reference for the redshift, as described in the text. For cases with no corresponding redshift value, it tracks sources where spectroscopic observation has been attempted.
Nph_128	Number of photons associated (Source_type events, evclass = 128).
Emax_128 (GeV)	Maximum photon energy detected (evclass = 128).
Nph_512	Number of photons associated (UltraClean_type events, evclass = 512).
Emax_512 (GeV)	Maximum photon energy detected (evclass = 512).
Ebin_max (GeV)	Mean energy of the four highest energy photons (evclass = 128).
TransmissionEbin	Flux transmission factor at \|$E^{\rm{bin}}_{\rm{max}}$\|⁠.
AbsEbin	Flux absorption factor at \|$E^{\rm{bin}}_{\rm{max}}$\|⁠, calculated as (⁠\|$1 -$\| Transmission).
tauEbin	EBL optical depth at \|$E^{\rm{bin}}_{\rm{max}}$\|⁠.
Abs_flag	Absorption flag: (1) \|$\tau > 0.1$\| (moderate to severe absorption) and (0) \|$\tau \le 0.1$\|⁠.
GAL_LAT	Galactic latitude in degrees.
BZcat	Identifier in the 5BZcat catalogue, if available.
z_3HSP	Redshift from the 3HSP catalogue, if available.
zflag_3HSP	Redshift flag from the 3HSP catalogue: (1) spectroscopic, (2) uncertain, (3) lower limit redshift, (4) photometric – with featureless optical spectrum, and (5) photometric – without optical spectrum.
4FGL_counter_name	Counterpart name in 4FGL-DR4.
TEVCAT_FLAG_4FGLdr4	Flag for TeVCat association, as per 4FGL-DR4.
ASSOC_TEV_4FGLdr4	TeV association in 4FGL-DR4.
CLASS1_4FGLdr4	Classification in 4FGL-DR4.
ASSOC1_4FGLdr4	Counterpart name in 4FGL-DR4.
Name_4LACdr3	Counterpart name in 4LAC-DR3.
z_4LACdr3_1	Redshift as reported in 4LAC-DR3.
Name_3FHL	Counterpart name in 3FHL.
z_3FHL	Redshift as reported in 3FHL.
Curvature_3FHL	Detectable spectral curvature in 3FHL: (1) yes, (2) no, and (–) unknown. Sources marked as ‘unknown’ are out of the 3FHL catalogue.

Name	RA (deg)	Dec. (deg)	4FGL-DR4	ASSOC1-4FGL	TS	\|$E^{\rm{bin}}_{\rm{max}}$\|	Rmag	AllWISE
4FGLJ1634.9+1222	248.760 49	12.3619	J1634.9+1222	NVSSJ163502+122142	15	233	20.04	J163502.41+122142.1
3HSPJ013632.6+390559	24.135 79	39.0997	J0136.5+3906	B30133+388	4708	183	15.86	J013632.59+390559.1
3HSPJ130738.0–425938	196.908 25	–42.9941	J1307.6–4259	1RXSJ130737.8–425940	732	150	16.26	J130737.98–425938.9
4FGLJ0840.1–0225	130.061 21	–2.4531	J0840.1–0225	PMNJ0840–0227	37	149	14.57	J084014.69–022711.3
4FGLJ1334.1–3521	203.550 10	–35.3372	J1334.1–3521	PKS1331–350	21	138	12.25	J133412.03–352014.2
3HSPJ074642.0–475455	116.676 29	–47.9152	J0746.6–4754	PMNJ0746–4755	1223	134	18.18	J074642.31–475455.0
3HSPJ230436.7+370507	346.152 95	37.0854	J2304.6+3704	1RXSJ230437.1+370506	297	98	20.2	J230436.71+370507.4
3HSPJ102634.4–854314	156.643 16	–85.7206	J1027.0–8542	PKS1029–85	528	95	17.27	J102634.36–854314.2
5BZBJ0700–6610	105.130 17	–66.1792	J0700.5–6610	PKS0700–661	986	94	15.29	J070031.25–661045.2
3HSPJ230940.8–363248	347.420 12	–36.5468	J2309.7–3632	WISEJ230940.84–363248	122	87	18.62	J230940.84–363248.7
3HSPJ033349.0+291631	53.454 16	29.2754	J0333.7+2916	TXS0330+291	591	85	19.1	J033349.00+291631.5
5BZBJ1849–4314	282.358	–43.2369	J1849.4–4313	PMNJ1849–4314	121	82	17.8	J184925.92–431413.3
4FGLJ0703.2–3914	105.802 68	–39.2385	J0703.2–3914	1RXSJ070312.7–391417	52	81	17.6	J070312.65–391418.8
4FGLJ1608.0–2038	241.987 19	–20.6617	J1608.0–2038	NVSSJ160756–203942	50	80	18.22	J160756.90–203943.5
3HSPJ130421.0–435310	196.0875	–43.8861	J1304.3–4353	1RXSJ130421.2–435308	1049	77	17.48	J130421.00–435310.2
4FGLJ2345.9+3413	356.479 85	34.2342	J2345.9+3413	1RXSJ234554.2+341419	21	74	17.8	J234555.21+341404.5
4FGLJ2143.9+3337	325.958 89	33.6196	J2143.9+3337	MG3J214351+3337	74	73	20.6	*J214350.20+333711.7
4FGLJ1345.6–3356	206.429 35	–33.9453	J1345.6–3356	NVSSJ134543–335643	76	73	19.58	J134543.04–335643.3
5BZBJ2108–6637	317.215 92	–66.6229	J2108.9–6638	PKS2104–668	198	73	18.22	J210851.80–663722.7
4FGLJ0836.0–8015	128.954 66	–80.2696	J0836.0–8015	2MASSJ08354940–8016114	29	69	20.74	J083549.36–801610.8
4FGLJ2300.8–0736	345.227 98	–7.5954	J2300.8–0736	2MASSJ23005469–0735438	34	69	19.51	J230054.70–073543.5
3HSPJ081012.0–703047	122.550 16	–70.5130	J0809.9–7028	SUMSSJ081011–703048	12	68	19.44	J081012.04–703047.1
3HSPJ104756.9–373730	161.987 25	–37.6252	J1047.9–3738	GALEXASCJ104756–373730	45	65	19.7	J104756.94–373730.8
4FGLJ1804.4+5249	271.094 84	52.8205	J1804.4+5249	6CB180317.2+524912	14	65	15.87	J180422.64+524915.2
3HSPJ195547.9+021512	298.949 41	2.2535	J1955.7+0214	NVSSJ195547+021514	92	63	20.95	J195547.95+021517.9
4FGLJ1836.5+1948	279.133 80	19.8461	J1836.5+1948	NVSSJ183632+195047	54	62	18.83	J183632.11+195046.3
5BZBJ0718–4319	109.681 83	–43.3304	J0718.6–4319	PMNJ0718–4319	760	62	17.73	J071843.63–431949.7
4FGLJ2025.3–2231	306.313 22	–22.5051	J2025.3–2231	NVSSJ202515–223016	55	62	18.15	J202515.17–223018.4
4FGLJ2127.6–5959	321.870 39	–60.0137	J2127.6–5959	NGC7059(?)	16	60	17.79	*J212729.34–600055.8
4FGLJ0624.2–2943	96.092 91	–29.7469	J0624.2–2943	1RXSJ062422.3–294449	19	59	21.46	J062422.49–294446.7
4FGLJ2049.0–4020	312.291 71	–40.3419	J2049.0–4020	PKS2045–405	19	59	19.53	J204910.02–402030.9
3HSPJ151444.0–772254	228.683 41	–77.3817	J1514.4–7719	1RXSJ151448.8–772249	30	58	19.74	J151444.02–772254.2
4FGLJ0351.2–6103	57.752 49	–61.0467	J0351.2–6103	SUMSSJ035100–610248	15	57	16.61	*J035059.83–610241
4FGLJ0928.7–3529	142.207 62	–35.4969	J0928.7–3529	NVSSJ092849–352947	15	56	19.64	J092849.82–352948.8
3HSPJ145543.7–760052	223.932 04	–76.0145	J1455.8–7601	SUMSSJ145543–760054	45	55	18.41	J145543.69–760052.2
4FGLJ0942.7–2028	145.657 61	–20.4536	J0942.7–2028	1RXSJ094237.9–202720	75	55	19.96	J094237.81–202713.2
3HSPJ094709.5–254100	146.789 66	–25.6833	J0947.1–2541	1RXSJ094709.2–254056	397	55	18.16	J094709.52–254059.9
4FGLJ1513.4–3721	228.327 75	–37.3365	J1513.4–3721	2MASSJ15131867–3720114	56	54	17.83	J151318.66–372011.5
3HSPJ080215.9–094210	120.566 25	–9.7030	J0802.3–0942	WISEJ080215.63–0942(?)	235	54	19.04	J080215.90–094210.9
4FGLJ1828.0+2634	276.985 55	26.5535	J1828.0+2634	NVSSJ182756+263313	40	50	19.4	J182756.54+263313.2
3HSPJ213004.8–563222	322.519 87	–56.5394	–	–	18	50	20.48	J182756.54+263313.2
3HSPJ134706.9–295842	206.778 66	–29.9784	J1347.1–2959	NVSSJ134706–295840	62	50	18.8	J134706.88–295842.4

Name	RA (deg)	Dec. (deg)	z	z-origin	b (deg)	4FGL-DR4	TS	\|${{E}}^{{\rm {bin}}}_{{\rm {max}}}$\|	\|$\tau _{{\rm {Ebin}}}$\|
3HSPJ224753.2+441315	341.971 75	44.220 97	1.9(?)	Foschini22	–13.2	J2247.8+4413	375	179	5.08
3HSPJ015307.4+751742	28.280 70	75.295 22	2.35(?)	Foschini22	12.9	J0153.0+7517	168	149	4.94
4FGLJ0400.7+3920	60.189 11	39.352 71	1.1(?)	Foschini22	–10.2	J0400.7+3920	18	209	3.34
3HSPJ200506.0+700439	301.274 87	70.077 63	2.32(?)	Foschini22	19.4	J2005.1+7003	311	66	1.60
3HSPJ052542.4–601340	81.426 75	–60.227 83	1.78(?)	Kaur16	–33.8	J0525.6–6013	249	70	1.21
4FGLJ1937.0+8354	294.415 22	83.941 39	1.94(?)	Foschini22	25.7	J1937.0+8354	39	61	1.12
4FGLJ1858.1+7318	284.584 86	73.287 00	0.471(?)	Foschini22	25.4	J1858.1+7318	43	221	1.10
3HSPJ031612.7+090443	49.053 04	9.078 66	0.372(?)	3HSPcat	–39.5	J0316.2+0905	621	255	0.99
3HSPJ142829.9+743002	217.124 54	74.500 61	0.31(?)	3HSPcat	41.0	J1428.8+7429	18	294	0.95
3HSPJ144800.6+360831	222.002 45	36.142	>0.738	Paiano20	63.7	J1448.0+3608	1085	137	0.94
5BZBJ1248+5820	192.078 25	58.341 31	0.508(?)	Foschini22	58.7	J1248.3+5820	4149	189	0.94
5BZBJ0007+4712	1.999 88	47.202 14	>1.659	Shaw13	–15.0	J0008.0+4711	709	61	0.87
3HSPJ122337.0–303250	185.904 20	–30.547 25	>0.875	Desai19	31.9	J1223.6–3032	76	113	0.87
4FGLJ0318.7+2135	49.690 28	21.576 85	1.83(?)	Foschini22	–29.6	J0318.7+2135	171	55	0.85
3HSPJ063059.5–240646	97.748	–24.112 80	>1.239	3HSPcat	–14.9	J0630.9–2406	3589	78	0.80
5BZBJ1918–4111	289.566 87	–41.191 89	>1.591	Shaw13	–22.2	J1918.2–4111	555	60	0.79
5BZBJ1925–2219	291.415 79	–22.326 42	1.35(?)	Foschini22	–17.1	J1925.8–2220	37	69	0.76
5BZBJ2300+3137	345.095 12	31.617 89	>1.498	Shaw13	–25.5	J2300.3+3136	283	62	0.76
5BZBJ1314+2348	198.682 54	23.807 44	2.053(?)	SDSSdr18	83.7	J1314.7+2348	387	46	0.76
4FGLJ0453.1+6322	73.301 89	63.354 96	2.1(?)	Foschini22	12.1	J0453.1+6322	39	45	0.75
3HSPJ154015.1+815505	235.066 25	81.918 22	>0.67	Paiano17	32.9	J1540.1+8155	938	129	0.72
3C 66A	35.673 34	43.043 19	>0.3347	Furniss13	–16.7	J0222.6+4302	7060	230	0.70
5BZBJ0856–1105	134.174 21	–11.087 33	>1.397	Shaw13	21.4	J0856.6–1105	517	64	0.70
5BZBJ0612+4122	93.213 25	41.377 06	>1.107	Paiano20	10.9	J0612.8+4122	1383	80	0.68
3HSPJ181118.0+034113	272.825 08	3.687 11	0.717(?)	Foschini22	10.6	J1811.3+0340	473	117	0.67
3HSPJ043145.1+740326	67.937 75	74.057 38	1.35(?)	Foschini22	17.3	J0431.8+7403	226	65	0.67
3HSPJ103744.3+571155	159.434 58	57.198 80	>0.62	Meisner10	51.7	J1037.7+5711	3071	131	0.65
3HSPJ003514.1+151504	8.811 29	15.251 16	>0.64	3HSPcat	–47.4	J0035.2+1514	485	122	0.60
3HSPJ112048.1+421212	170.200 25	42.2035	>0.35	3HSPcat	66.1	J1120.8+4212	1811	199	0.58
S5 0716+714	110.472 50	71.343 32	0.31(?)	Foschini22	28.0	J0721.9+7120	11 282	218	0.57
3HSPJ170433.8–052840	256.140 95	–5.477 97	>0.7	Paiano17	20.7	J1704.5–0527	244	109	0.55
3HSPJ152048.9–034851	230.203 70	–3.814 33	>0.868	Goldini20	42.5	J1520.8–0348	421	88	0.53
3HSPJ193419.6+600139	293.581 75	60.027 63	1.38(?)	Foschini22	18.1	J1934.2+6002	44	56	0.52
4FGLJ1304.2–2412	196.069 55	–24.204 65	1.26(?)	Foschini22	38.5	J1304.2–2412	155	60	0.51
5BZBJ1312–2156	198.1315	–21.939 83	>1.485	Shaw13	40.6	J1312.4–2156	358	50	0.50
4FGLJ0028.5+2001	7.124 24	20.007 42	1.552(?)	4LACdr3	–42.5	J0028.5+2001	489	48	0.50
3HSPJ175615.1+552218	269.066 25	55.371 66	>0.657	3HSPcat	29.7	J1756.3+5522	264	106	0.47
4FGLJ0355.3+3909	58.819 12	39.152 72	0.846(?)	Foschini22	–11.0	J0355.3+3909	18	85	0.47
3HSPJ210421.9–021238	316.091 37	–2.210 80	>0.45	3HSPcat	–30.3	J2104.3–0212	88	140	0.43
3HSPJ222129.3–522527	335.372 08	–52.4243	>0.34	3HSPcat	–52.3	J2221.5–5225	426	174	0.42
3HSPJ125359.3+624257	193.497 25	62.716	>0.867	Shaw13	54.4	J1253.8+6242	179	79	0.42
4FGLJ1537.9–1344	234.487 93	–13.7262	0.984(?)	Foschini22	32.5	J1537.9–1344	31	70	0.42
5BZBJ2016–0903	304.100 13	–9.0592	>0.605	Shaw13	–22.9	J2016.3–0903	286	108	0.42

Article Contents

Mapping the cosmic gamma-ray horizon: the 1CGH catalogue of Fermi-LAT detections above 10 GeV

ABSTRACT

1 INTRODUCTION

2 DATA ANALYSIS AND METHODS

2.1 Selection of γ-ray candidate sources

2.2 Broad-band likelihood analysis with Fermi-LAT science tools

2.3 Detection threshold for multifrequency-selected targets

2.4 Additional selection criteria: refining the 1CGH sample

3 THE FIRST COSMIC GAMMA-RAY HORIZON CATALOGUE

3.1 Extensive redshift review

3.2 The cosmic gamma-ray horizon plot

3.3 Identification of sources with detectable EBL absorption

3.4 Best candidates for optical observation campaigns

4 SUMMARY AND CONCLUSIONS

ACKNOWLEDGEMENTS

DATA AVAILABILITY

Footnotes

REFERENCES

APPENDIX A: COLUMN DESCRIPTIONS OF THE 1CGH CATALOGUE

APPENDIX B: LONG TABLES: BEST CANDIDATES FOR OPTICAL OBSERVATIONS

Supplementary data

Citations

Views

Altmetric

Email alerts

Astrophysics Data System

Citing articles via

Latest

Most Read

Most Cited

This Feature Is Available To Subscribers Only