Very-high-energy flat spectral radio quasar candidates Free

ABSTRACT

1 INTRODUCTION

Very-high-energy (VHE; E> 100 GeV) gamma-ray astronomy has the potential to provide unique insights on open issues related to cosmology and particle physics. Additionally, it serves as an important probe for multiwavelength and multimessenger astronomy. VHE emission from astrophysical sources can be observed from the ground by studying the shower of secondary charged particles initiated through the interaction of the primary gamma-rays with the atmosphere (de Naurois & Mazin 2015). Among various techniques employed, the Imaging Atmospheric Cherenkov Telescopes (IACTs) detect the primary gamma-rays through the image of the Cherenkov pool caused by the secondary shower. The current-generation IACTs, which include the High Energy Stereoscopic System (H.E.S.S.; Aharonian et al. 2006), the Major Atmospheric Gamma-Ray Imaging Cherenkov telescope (MAGIC; Aleksić et al. 2012), and the Very Energetic Radiation Imaging Telescope Array System (VERITAS; Holder et al. 2006), have been contributing extensively to VHE astrophysics for nearly two decades. In tandem with telescopes operating at other energies, such as Fermi-LAT (Atwood et al. 2009), Swift (Burrows et al. 2005), NuSTAR (Harrison et al. 2013), and XMM–Newton (Jansen et al. 2001), IACTs have provided clues to help us understand the non-thermal emission processes in blazars (Knödlseder 2016).

With the advent of new-generation ground-based VHE telescopes, including the proposed Cherenkov Telescope Array Observatory (CTAO; Acharya et al. 2019), gamma-ray astronomy is entering into a new era. Operating from a few tens of GeV to the multi-TeV energy band, the CTAO is designed to be the largest and the most sensitive gamma-ray observatory in this energy range (Gueta 2021). It will be configured as two sets of Cherenkov telescope arrays, one in each of the Earth’s hemispheres, and it is expected to start science operations at full capacity within a few years. The CTAO along with other upcoming VHE experiments, such as the Large High Altitude Air Shower Observatory (LHAASO; Cao 2021), the ASTRI Mini-Array (ASTRI MA; Antonelli 2021), the Southern Wide-field Gamma-ray Observatory (SWGO; Barres de Almeida, Giacinti & Longo 2021), and the Major Atmospheric Cherenkov Experiments (MACE; Bhatt et al. 2021), will be able to explore the gamma-ray sky with unprecedented performance, notably in the multi-TeV energy range.

Blazars dominate the extragalactic sky at VHE energies. These objects are a subclass of radio-loud active galactic nuclei (AGNs) having a relativistic jet pointing close to the line of sight of the observer on Earth (Urry & Padovani 1995). The spectral energy distribution (SED) of blazars is dominated by non-thermal emission from the jet and consists of two broad peaks. The low-energy component extends from radio to X-rays and is attributed to the synchrotron emission, while the high-energy component is commonly interpreted as the inverse Compton (IC) scattering of low-energy photons by the relativistic electron distribution in the jet (Urry & Padovani 1995). Blazars are further classified into BL Lacertae objects (BL Lacs) and flat spectrum radio quasars (FSRQs) based on the presence/absence of broad emission/absorption-line features in their optical spectra. The synchrotron SED of BL Lacs generally peaks at optical to X-ray energies whereas, for FSRQs, this spectral peak falls in the infrared to optical energy range. Besides the variation in peak location, the IC component of blazars is significantly different for BL Lacs and FSRQs. Particularly, the IC luminosity for FSRQs is larger than the synchrotron luminosity, commonly referred to as Compton dominance, while it is of a similar order in the case of BL Lacs (Sikora, Begelman & Rees 1994). The target photon field for the IC process is also different for these two classes of blazars. The IC scattering of synchrotron photons (the synchrotron self-Compton mechanism, SSC) is successful in explaining the Compton spectral component of BL Lacs. However, modelling the high-energy SED of FSRQs through the IC mechanism demands an additional photon field other than synchrotron emission. This photon field can be external to the jet, and plausible sources are the thermal photons from the accretion disc (Dermer & Schlickeiser 1993), broad emission-line photons (Błażejowski et al. 2000) and/or the IR photons from the dusty torus (Sikora et al. 1994).

The interaction of VHE gamma-rays with the extragalactic background light (EBL) results in the formation of electron/positron pairs (Dwek & Krennrich 2013). The amount of attenuation depends on the redshift of the source and the energy of the VHE photons. For distant blazars, this causes the observed VHE spectra to steepen, causing the flux to fall below the telescope’s sensitivity. Consequently, this makes the Universe opaque above a few tens of GeV for objects that have larger redshifts (gamma-ray horizon), which was initially predicted at z > 0.1 (Gould & Schréder 1967). Improved sensitivity of the current-generation Cherenkov telescopes and better estimation of EBL intensity have significantly modified the gamma-ray horizon; nevertheless, detection of the FSRQs 4FGL J0221.1+3556 and 4FGL J1443.9+2501, located at redshifts 0.954 (Ahnen et al. 2016) and 0.939 (Abeysekara et al. 2015; Ahnen et al. 2015), suggests that the EBL intensity may still be less than the prediction.

Knowledge of EBL intensity and the intrinsic source VHE spectrum is crucial for identifying VHE blazar candidates. The intrinsic source VHE spectrum is often estimated by extrapolating the Fermi high-energy gamma-ray spectrum to VHE energies (Paiano et al. 2021; Zhu et al. 2021). The VHE blazar candidates are then obtained by convolving the extrapolated spectrum with EBL-induced opacity predicted through cosmological models. However, such candidates are put forth only for BL Lacs (Massaro et al. 2013; Balmaverde et al. 2019; Costamante 2019; Foffano et al. 2019; Paiano et al. 2021; Zhu et al. 2021) and no such study has been performed for FSRQs. The primary reason for this is that the Fermi spectrum of FSRQs generally deviates from a power law and is often represented by a log-parabolic function, and extending this function to VHE energies rolls off the spectrum significantly. Additionally, the FSRQs are populated at higher redshifts and the current EBL models disfavour them as probable candidates. Besides this, most VHE detections of FSRQs have been during enhanced flux states and hence flux variability plays a crucial role in VHE studies of these sources.

In our earlier study (Malik et al. 2022, hereafter Paper I), using the correlation between the observed VHE spectral index and the redshift, we highlighted the deviation of different cosmological EBL models from the observations. The important assumptions in the earlier study are the following: (i) the average intrinsic VHE spectral index is consistent with the regression line extrapolated to redshift, z = 0; (ii) the spectral index variation of the individual source is much smaller than the steepening introduced by the EBL at large redshifts; (iii) the cosmological evolution of the source does not modify the intrinsic VHE spectral indices. This study suggests that the EBL intensity may be much less intense than that predicted by cosmological models. Consistently, this also suggests that the gamma-ray horizon may fall at much larger redshifts than the one presumed. Moreover, from the X-ray spectral studies of blazars, it is known that the log-parabolic function is successful in reproducing only a narrow energy band (Massaro et al. 2004). Hence, it may not be judicious to expect the Fermi log-parabolic spectral shape to extend up to very high energy. In this work, we predict the plausible VHE FSRQ candidates considering these discrepancies. We perform a linear extrapolation of the high-energy spectrum of FSRQs listed in the 4FGL-DR2 catalogue (Ballet et al. 2020) to very high energy as a prediction for the intrinsic VHE spectra. To account for the reduction in EBL intensity, suggested by VHE observations of FSRQs (Paper I), we add a redshift-dependent correction factor to the EBL opacity provided by Franceschini & Rodighiero (2017), hereafter FM. These modifications are then used to predict the list of VHE FSRQs that can be studied by the CTAO and other operational Cherenkov telescopes. In Section 2, we first introduce a correction factor to the existing EBL estimates (using Paper I) followed by intrinsic VHE flux estimations using Fermi spectral information. The section concludes by overplotting the sensitivity curves of present and upcoming VHE telescopes to look for possible VHE FSRQ candidates. Finally, the results are summarized and discussed in Section 3. In this work, we adopt a cosmology with |$\rm \Omega _M = 0.3$|⁠, |$\rm \Omega _\Lambda = 0.7$|⁠, and |$H_0 = 71\, {\rm km\, s^{-1}\, Mpc}^{-1}$|⁠.

2 VHE FSRQ CANDIDATES

2.1 Modified EBL photon density

Figure 1.

Comparison of observed VHE spectral indices with those estimated using the EBL model by Franceschini & Rodighiero (2017) (FM) for FSRQs. The black solid line represents the best-fitting straight line to seven FSRQs detected in very high energy (Paper I), and the dashed line represents the index values estimated using FM. The grey shaded region shows the deviation of the EBL model from the observational best-fitting line to VHE indices.

Open in new tabDownload slide

2.2 Intrinsic VHE source flux

The opacity of VHE photons, along with the correction factor (equation 1), allows us to derive the observed flux of FSRQs provided the intrinsic flux is known. To estimate the latter, we perform a linear extrapolation of the high-energy spectrum of FSRQs to very high energy. The information regarding the high-energy spectrum is obtained from the 4FGL-DR2 (Ballet et al. 2020). The data used in this catalogue were collected over a period of 10 yr, from 2008 August 4 (15:43 utc) to 2018 August 2 (19:13 utc). It employs the same analysis methods as the 4FGL ( Abdollahi et al. 2020) in the energy range of 50 MeV to 1 TeV. Among the 5788 sources listed in the catalogue, 744 fall into the list of FSRQs. The attenuation due to EBL depends upon the source redshift and this is obtained from the online data bases, NED¹ and SIMBAD.² These data bases provide the redshifts of 586 FSRQs listed in the 4FGL-DR2 and hence only these sources are considered for this study. The information regarding the high-energy spectrum of these FSRQs is extracted from the xml file available online³ and the same information is plotted as a black dashed line in Fig. 2.

Open in new tabDownload slide

Open in new tabDownload slide

Open in new tabDownload slide

Figure 2.

32 VHE FSRQ candidates falling within the detection threshold of the CTAO and other instruments. The black dashed line represents the 4FGL-DR2 spectral fit. The red dashed line shows the linear extrapolation of the Fermi spectrum to very high energy. The dashed green and solid black lines represent the EBL attenuated spectrum using the model of Franceschini & Rodighiero (2017) (FM) and the modified EBL model (MFM), respectively. The differential sensitivities of CTAO-North and South (Omega configuration, 50 h), MAGIC (50 h), VERITAS (50 h), and H.E.S.S. (50 h) are also plotted. Sensitivity curves are given in different colours as depicted in the panel labels.

2.3 Comparison with IACT sensitivity

To identify the FSRQ candidates that can be detected by the operational and upcoming IACTs, we compare the predicted EBL attenuated VHE spectrum with the available sensitivity curves. The sensitivity curves of CTAO-South and North (zenith angle, ZA < 20 deg) are obtained from the CTAO webpage⁴ while the sensitivity curve for VERITAS (ZA < 40 deg) is obtained from the VERITAS webpage.⁵ For other instruments such as MAGIC (ZA < 35 deg) and H.E.S.S. (ZA < 18 deg), the sensitivity curves are obtained from Aleksić et al. (2016) and Holler et al. (2016), respectively. These sensitivities are calculated at 5σ significance for 50 h of exposure time and are shown in Fig. 2 with legends.

Comparing the sensitivity of the IACTs and the predicted VHE spectrum of Fermi-detected FSRQs, we identify the plausible VHE FSRQ candidates and list them in Table 1. Our study suggests that 32 FSRQs would fall within the detection threshold of CTAO’s Omega configuration (full-scope configuration), while the detection status with CTAO’s first construction phase (Alpha configuration) will be 29 using opacity estimates from the MFM. Considering FM, the numbers of sources falling under the detection status of CTAO reduce to 23 and 20 for Omega and Alpha configurations of the CTAO, respectively. With the operational IACTs, the number of FSRQs falling within the detection threshold are five for VERITAS and two sources each for MAGIC and H.E.S.S. using opacity estimates from the MFM, while using the FM the numbers of sources reduce to two, one and one for VERITAS, MAGIC and H.E.S.S., respectively (Table 1).

Because blazars are extremely variable (Abdo et al. 2010; Meyer, Scargle & Blandford 2019; Rajput, Stalin & Rakshit 2020), the intrinsic VHE flux derived from cumulative Fermi observations may portray only the average spectrum. To account for this variability, we assume a scenario where the VHE flux enhances by a factor of 10 above the prediction. Furthermore, to be consistent with the rapid variability, we use the CTAO sensitivity corresponding to a 5-h exposure time only. We refer to the FSRQs that fall within the detection threshold of the CTAO under this assumption as marginally detectable. In Table 2, we list the FSRQs that are marginally detectable and find that an additional 82 sources can be detected under the Alpha configuration of the CTAO for the opacity estimated by the MFM. This number increases to 90 under the Omega configuration of the CTAO. For the opacity estimated by the FM, the numbers of sources falling under the marginal detection list are 43 and 40 with the Omega and Alpha configurations of the CTAO, respectively.

3 SUMMARY AND DISCUSSION

The IACTs are narrow-angle telescopes and require long-duration observations of distant faint sources for significant detection. Hence, the identification of plausible candidate FSRQs can provide a guideline for the upcoming VHE telescopes. Our earlier study, based on the correlation between the observed VHE spectral index with redshift (Paper I) suggests that the Universe may be more transparent to VHE photons than those predicted by cosmological EBL models. Taking our cue from this, we predict the observed VHE fluxes of Fermi-detected FSRQs and compare the same with the sensitivity of operational/upcoming telescopes. We find that a significant number of FSRQs, listed in Tables 1 and 2, can be studied using the CTAO, while a few of them can even be studied using the operational IACTs. The sources that fall within the sensitivity limits of operational IACTs, and are not reported in very high energy yet, are 4FGL J0043.8+3425 (VERITAS) and 4FGL J0957.6+5523 (VERITAS, MAGIC) (see Table 1). Among them, 4FGL J0043.8+3425 is located at z = 0.966, which will be the second most distant FSRQ if detected, next only to the newly announced 4FGL J0348.5−2749 at z = 0.991 (Wagner et al. 2021). However, the 10-GeV spectral index for this source is ∼2.08. Though this satisfies our selection criteria, the hard index suggests the Compton spectral peak may fall close to 10 GeV and hence the extrapolation can be questionable. Interestingly, the VHE emission from FSRQ 4FGL J1159.5+2914, newly announced by VERITAS (Mukherjee et al. 2017) and MAGIC (Mirzoyan 2017), is also predicted to be a candidate source in this work (see Table 1). We also examined the detection status for SWGO, LHAASO and MACE, though we did not find any source falling within the detection threshold of these instruments.

Among the FSRQs predicted for the CTAO, we find 11 sources at z > 1 that fall under the detectable list and 33 that fall under the marginally detectable list (see Tables 1 and 2). If detected, these high-redshift sources may pose challenges to the existing cosmological EBL models. Alternatively, they can also play an important role in understanding cosmology. Limits on the EBL are mainly obtained through numerical models of galaxy formation and/or their evolution with appropriate cosmological initial conditions (Domínguez et al. 2011). The model parameters are fine tuned to reproduce the observed Universe. VHE identification of the sources at large redshifts can therefore be an important element to constrain the EBL, which in turn can provide a better understanding of galaxy formation and evolution. These identifications can also be used to test alternative theories involving the oscillation of photons and axion-like particles proposed by the standard model (Galanti & Roncadelli 2018; Irastorza & Redondo 2018).

Considering the fact that blazars are extremely variable (Abdo et al. 2010; Meyer et al. 2019; Rajput et al. 2020), certain FSRQs may still be detectable by operational/upcoming telescopes even though our study suggests otherwise. For instance, FSRQ 4FGL J1422.3+3223 falls below our criteria for detection though it is detected at very high energy (Acciari et al. 2021). This probably indicates that during the flaring epoch, the flux of this source can enhance by than 10 times its average flux. Consistently, we find that the Fermi flux of 4FGL J1422.3+3223 during the VHE detection is typically more than 100 times larger than the average flux quoted in the 4FGL catalogue (Ciprini & Cheung 2020). Considering a similar factor of flux enhancement in this work would make this source also detectable by the operational IACTs MAGIC and VERITAS. A similar conclusion can be made for the newly announced FSRQ 4FGL J0348.5−2749 where the increase in Fermi flux, contemporaneous to the VHE detection, was ∼200 times compared to the average flux reported in the Fermi 4FGL catalogue (Wagner et al. 2021).

The VHE FSRQ candidates predicted in this work depend on the choice of a(z) and the robustness of the regression line. However, the regression line is obtained by fitting merely seven data points and this may change with future detections. Because the estimation of a(z) assumes that the intrinsic VHE index is the y-intercept of the regression line, any deviation in the fit parameters can modify these predictions considerably. Conversely, a better regression analysis needs more FSRQs to be detected in very high energy, and the prediction based on the available information can facilitate this requirement. Detection of more VHE FSRQs with precise index measurements will also let us fit the redshift-index dependence with non-linear functions. Such a study will also have a major role in constraining the cosmological models.

The VHE spectrum of FSRQs is better explained by a power-law function and hence we have assumed that the intrinsic source spectrum is also a power law. If we consider the EC interpretation for the VHE emission, the Klein–Nishina effects will be substantial and the spectrum will deviate from a simple power law (Dermer & Schlickeiser 1993). In addition, emission at very high energy may involve high-energy electrons that may fall close to the cut-off energy of the underlying electron distribution (Kirk, Rieger & Mastichiadis 1998). Under these conditions, the intrinsic VHE spectrum may deviate significantly from a power law and hence the power-law extrapolation to VHE energies can be an overestimate. Therefore, these predictions on stringent conditions should be treated as an upper limit. The Klein–Nishina effect will depend upon the energy of the target photons, and under extreme limits, the VHE spectrum will be very steep, with the photon spectrum nearly following the electron distribution (Blumenthal & Gould 1970). Hence, a VHE study of FSRQs can potentially also help us to understand the photon field environment of FSRQs.

ACKNOWLEDGEMENTS

The authors thank the anonymous referee for valuable comments and suggestions. MZ, SS, NI and AM acknowledge the financial support provided by Department of Atomic Energy (DAE), Board of Research in Nuclear Sciences (BRNS), Government of India via Sanction Ref No. 58/14/21/2019-BRNS. SZ is supported by the Department of Science and Technology, Government of India, under the INSPIRE Faculty grant (DST/INSPIRE/04/2020/002319). This research has made use of the CTA instrument response functions provided by the Cherenkov Telescope Array Consortium and Observatory; see https://www.cta-observatory.org/science/cta-performance/ (version prod3b-v2, https://doi.org/10.5281/zenodo.5163273, and version prod5 v0.1, https://doi.org/10.5281/zenodo.5499840) for more details.

DATA AVAILABILITY

The codes and model used in this work will be shared on reasonable request to the corresponding author, Zahoor Malik (email: [email protected]).

Footnotes

REFERENCES