Basic properties of Fermi blazars and the “blazar sequence”

# Basic properties of Fermi blazars and the “blazar sequence”

Dingrong Xiong, Xiong Zhang, Jinming Bai and Haojing Zhang
National Astronomical Observatories/Yunnan Observatories, Chinese Academy of Sciences, Kunming 650011, China
Department of Physics, Yunnan Normal University, Kunming 650500, China
E-mail: ynzx@yeah.net
###### Abstract

By statistically analyzing a large sample which includes blazars of Fermi detection (FBs) and non-Fermi detection (NFBs), we find that there are significant differences between FBs and NFBs for redshift, black hole mass, jet kinetic power from “cavity” power, broad-line luminosity, and ratio of core luminosity to absolute V-band magnitude (), but not for ratio of radio core to extended flux () and Eddington ratio. Compared with NFBs, FBs have larger mean jet power, and while smaller mean redshift, black hole mass, broad-line luminosity. These results support that the beaming effect is main reason for differences between FBs and NFBs, and that FBs are likely to have a more powerful jet. For both Fermi and non-Fermi blazars, there are significant correlations between jet power and the accretion rate (traced by the broad-emission-lines luminosity), between jet power and black hole mass; for Fermi blazars, the black hole mass does not have significant influence on jet power while for non-Fermi blazars, both accretion rate and black hole mass have contributions to the jet power. Our results support the “blazar sequence” and show that synchrotron peak frequency () is associated with accretion rate but not with black hole mass.

###### keywords:
radiation mechanisms: nonthermal – galaxies: active – BL Lacertae objects: general – quasars: general – gamma-rays: theory – X-rays: general
pagerange: Basic properties of Fermi blazars and the “blazar sequence”LABEL:lastpagepubyear: 2002

## 1 Introduction

Blazars are often divided into two subclasses of BL Lacertae objects (BL Lacs) and flat spectrum radio quasars (FSRQs). FSRQs have strong emission lines, while BL Lacs have only very weak or non-existent emission lines. The classic division between FSRQs and BL Lacs is mainly based on the equivalent width (EW) of the emission lines. Objects with rest frame EW Å are classified as FSRQs (e.g. Scarpa & Falomo 1997; Urry & Padovani 1995). Many authors have proposed that EW alone is not a good indicator of the distinction between the two classes of blazars (Scarpa & Falomo 1997; Ghisellini et al. 2011; Sbarrato et al. 2012, 2014; Giommi et al. 2012, 2013; XZ14). Ghisellini et al. (2011) introduced a physical distinction between the two classes of blazars, based on the luminosity of the broad line region measured in Eddington units. The dividing line is of the order of . The result also was confirmed by Sbarrato et al. (2012) and XZ14. Giommi et al. (2012, 2013) suggested that blazars should be divided in high and low ionization sources.

In this paper, we constructed a large sample of blazars, including Fermi blazars from XZ14 and non-Fermi blazars, and studied the properties of Fermi blazars and the blazar sequence. The paper is structured as follows: in Sect. 2, we present the samples; the results are presented in Sect. 3; discussions and conclusions are presented in Sect. 4. The cosmological parameters , and have been adopted in this work. The energy spectral index is defined such that .

## 2 The samples

The selection criteria for the sample were that we tried to select the largest group of blazars included in BZCAT (Massaro et al. 2009: the Roma BZCAT) with reliable broad line luminosity (used as a proxy for disk luminosity), redshift, black hole mass and jet kinetic power. The sample of Fermi blazars was directly from XZ14. Due to be classed into non-clean 2LAC, the four fermi blazars (2FGL J0204.0+3045; 2FGL J0656.2-0320; 2FGL J1830.1+0617; 2FGL J2356.3+0432) in XZ14 were not included in our sample. In order to have reliable sample, we did not consider the candidate blazars of unknown type (BZU called in BZCAT) in our sample. Also for the same reason, non-Fermi blazars, which were detected by EGRET or recorded in 1LAC but missed in 2LAC, were not included in our sample. In addition, to reduce the uncertainty, we tried to select the data from a same paper and/or a uniform method. The detailed information and calculating methods for broad line luminosity, black hole mass are seen in XZ14. Cavagnolo et al. (2010) searched for X-ray cavities in different systems including giant elliptical galaxies and cD galaxies and estimated the jet power required to inflate these cavities or bubbles, obtaining a tight correlation between the “cavity” power and the radio luminosity at 200-400 MHz

which is continuous over decades in and and . Making use of the correlation between and from Cavagnolo et al. (2010), Meyer et al. (2011) chose the low-frequency extended luminosity at 300 MHz as an estimator of the jet power for blazars. Their 300 MHz extended luminosity was extrapolated from 1.4 GHz extended radio emission or obtained from spectral decomposition. Following Meyer et al. (2011), Nemmen et al. (2012) estimated the jet kinetic power for a large sample of Fermi blazars. We also used Equation (1) to get jet kinetic power from “cavity” power (this is what we mean when we refer to the jet kinetic power in the rest of the paper). In our sample, in order to reduce uncertainty, the jet kinetic power only is gained from extended 1.4 GHz radio data but not from spectral decomposition. The extended 1.4 GHz radio data can not be obtained for all blazars. The proportion of blazars of jet kinetic power estimated from extended radio luminosity in all sources was 28%. From Wang et al. (2004), the uncertainty in the relation was dex; the uncertainty on the zero point of the line width-luminosity-mass relation was approximately 0.5 dex; the correlation for quasar host galaxies had an uncertainty of 0.6 dex. So given the intrinsic uncertainty of the different black hole mass estimators and the heterogeneity of sample, we estimated that the individual BH masses may have an uncertainty as large as 1 dex. The uncertainty in was dominated by the scatter in the correlation of Cavagnolo et al. (2010) and corresponded to 0.7 dex. The uncertainties on broad line luminosity were based on the standard deviation of Mg II/Ly, H/Ly and C IV/Ly in the composite quasar spectrum of Francis et al. (1991) (Wang et al. 2004). We assumed that the uncertainties on broad line luminosity were to be 0.5 dex for all sources.

For the sake of exploring the blazar sequence, we also collected and/or calculated the synchrotron peak frequency and the peak luminosity of the synchrotron component . The and of our Fermi blazars were collected from Finke (2013) and Meyer et al. (2011), and the and of non-Fermi blazar from Nieppola et al. (2006, 2008), Meyer et al. (2011), Wu et al. (2009), Aatrokoski et al. (2011). Generally, the SED was fitted by using a simple third-degree polynomial function. However, many blazars were lack of observed SED. Abdo et al. (2010c) have conducted a detailed investigation of the broadband spectral properties of the -ray selected blazars of the Fermi LAT Bright AGN Sample (LBAS). They assembled high-quality and quasi-simultaneous SED for 48 LBAS blazars, and their results have been used to derive empirical relationships that estimate the position of the two peaks from the broadband colors (i.e. the radio to optical, , and optical to X-ray, , spectral slopes) and from the -ray spectral index. Ackermann et al. (2011) used the empirical relationships for finding the peak frequency of synchrotron component of 2LAC clean blazars from the slopes between the 5 GHz and 5500 Å flux, and between the 5500 Å and 1 KeV flux. Finke (2013) used their results for finding . The rest of authors fitted SED to obtain . When blazars were missed in the above literatures, we used the empirical relationships of Abdo et al. (2010c) to find . Firstly, we collected the fluxes of the blazars at 5 GHz, 5500 Å and 1 KeV from BZCAT, NASA/IPAC Extragalactic Database: NED, Veron-Cetty & Veron (2010) and Ackermann et al. (2011). When more than one flux or magnitude was found, we took the most recent one. Apparent magnitude of optical V band can be converted into flux as with flux KJy and in units of KJy (Mead et al. 1990). All flux densities were K-corrected according to , where was the spectral index and , , and for BL Lacs, and for FSRQs (Cheng et al. 2000). The luminosity was calculated from the relation , and was the luminosity distance. The 1 KeV flux density was transferred from the 0.1-2.4 KeV flux density given in the BZCAT or NED using . After then we used the Equation (1) of Abdo et al. (2010c) to get and . Finally, we used the empirical relationships of Abdo et al. (2010c) for finding the peak frequency of synchrotron component. For blazars without X-ray flux, we adopted to estimate ( from linear regression analysis: ). The proportion of blazars (that the is estimated by the linear regression equation) in all sources is 20%. When the peak luminosity of synchrotron component was not got from above literatures, we used radio luminosity at 5 GHz to estimate the synchrotron peak luminosity ( from linear regression analysis: ). The proportion of blazars (that the is estimated by the linear regression equation) in all sources is 60%.

The ratio of the beamed radio core flux density to the unbeamed extended radio flux density ( with , ) has routinely been used as a statistical indicator of Doppler beaming and thereby orientation (Orr & Browne 1982; Urry & Padovani 1995; Kharb et al. 2010). In order to compare Fermi blazars with non-Fermi blazars, we also estimated at 1.4 GHz. When more than one was obtained, we took the most recent one.

The relevant data for Fermi blazars can be seen in Table 1 of XZ14. The relevant data for non-Fermi blazars were listed in Table 1 with the following headings: column (1), name of the Roma BZCAT catalog; column (2), other name; column (3) is right ascension (the first entry) and declination (the second entry); column (4), redshift from NED; column (5), logarithm of the synchrotron peak frequency (the first entry) and logarithm of the peak luminosity of the synchrotron component in units of (the second entry); column (6), logarithm of jet kinetic power in units of ; column (7), logarithm of black hole mass in units of and references; column (8), logarithm of broad-line luminosity in units of and references; column (9), logarithm of the ratio of the beamed radio core flux density to the unbeamed extended radio flux density and references. For black hole mass or broad-line luminosity, when more than one value was obtained, we took the mean value.

In total, we have a sample containing 244 clean Fermi blazars (187 FSRQs and 57 BL Lacs) and 469 non-Fermi blazars (370 FSRQs and 99 BL Lacs).

## 3 The results

### 3.1 The distributions

The redshift distributions of the various classes are shown in Fig. 1. From Fig. 2 of Roma-BZCAT, the redshift distributions of BL Lacs are much closer than that of FSRQs and their distribution peaks at , whereas FSRQs show a broad maximum between 0.6 and 1.5. There are only very few BL Lacs at redshift higher than 0.8. So the redshift distributions from our sample are consistent with the results of Roma-BZCAT. The redshift distributions for Fermi blazars are and mean value is ; for non-Fermi blazars, the redshift distributions are and mean value is . The mean values for Fermi BL Lacs, Fermi FSRQs, non-Fermi BL Lacs and non-Fermi FSRQs are , , and respectively. Through nonparametric Kolmogorov-Smirnov (KS) test, we get that the redshift distributions between all Fermi blazars and all non-Fermi blazars, between Fermi BL Lacs and non-Fermi BL Lacs, between Fermi FSRQs and non-Fermi FSRQs are significant difference (chance probability ). Based on above results, it is shown that compared with non-Fermi BL Lacs, Fermi BL Lacs have larger mean redshift while compared with non-Fermi FSRQs, Fermi FSRQs have smaller mean redshift.

The black hole mass distributions of the various classes are shown in Fig. 2. The black hole mass distributions between all Fermi blazars and all non-Fermi blazars, between Fermi BL Lacs and non-Fermi BL Lacs, between Fermi FSRQs and non-Fermi FSRQs are significant difference (). So we can get that compared with non-Fermi blazars, Fermi blazars have smaller mean black hole mass.

The jet kinetic power distributions of the various classes are shown in Fig. 3. The jet kinetic power distributions between all Fermi blazars and all non-Fermi blazars, between Fermi FSRQs and non-Fermi FSRQs are significant difference (). So Fermi blazars have larger mean jet kinetic power than non-Fermi blazars.

The broad-line luminosity distributions of the various classes are shown in Fig. 4. Because non-Fermi BL Lacs almost have not broad-line data, we only compare broad-line luminosity distributions between Fermi FSRQs and non-Fermi FSRQs. The mean values for Fermi FSRQs and non-Fermi FSRQs are and respectively. From KS test, the broad-line luminosity distributions between Fermi FSRQs and non-Fermi FSRQs are significant difference (). We also compare the Eddington ratio distributions between Fermi FSRQs and non-Fermi FSRQs (see Fig. 5; from Netzer (1990)). However, the result of KS test shows that they do not have significant difference (). The mean values for Fermi FSRQs and non-Fermi FSRQs are and respectively.

The distributions for ratio of the beamed radio core flux density to the unbeamed extended radio flux density (core prominence parameter ) are shown in Fig. 6. The distributions for Fermi blazars are and mean value is ; for non-Fermi blazars, the distributions are and mean value is . The mean values for Fermi BL Lacs, Fermi FSRQs, non-Fermi BL Lacs and non-Fermi FSRQs are , , and respectively. The distributions between all Fermi blazars and all non-Fermi blazars, between Fermi BL Lacs and non-Fermi BL Lacs, between Fermi FSRQs and non-Fermi FSRQs are not significant difference (). Kharb et al. (2010) found that the ratio of the radio core luminosity to the k-corrected optical luminosity () appears to be a better indicator of orientation than the traditionally used radio core prominence parameter (). Wills & Brotherton (1995) defined as the ratio of the radio core luminosity to the k-corrected absolute V-band magnitude (): , where , and the k-correction is, with the optical spectral index, . Making use of the above Equations, we obtain for our sample. The distributions for are shown in Fig. 7. The mean values of for Fermi blazars and non-Fermi blazars are and respectively. The distribution between all Fermi blazars and all non-Fermi blazars is significant difference (). Therefore, Fermi blazars have larger mean than non-Fermi blazars, which supports that compared with non-Fermi blazars, Fermi blazars are more beamed.

### 3.2 Black hole mass vs jet power

Figure 8 shows black hole mass as a function of jet power. Different symbols correspond to blazars of different classes. Pearson product-moment analysis (hereafter called Pearson analysis) is applied to analyze the correlations between black hole mass and jet power for all blazars. The results show that the correlations between black hole mass and jet power for both Fermi blazars and non-Fermi blazars are significant (Fermi blazars: number of points , significance level , coefficient of correlation ; non-Fermi blazars: , , ).

### 3.3 Broad line luminosity and disk luminosity vs jet power

Figure 9 shows broad line luminosity as a function of jet power. The results of Pearson analysis show that there are significant correlations between broad line luminosity and jet power for Fermi blazars and non-Fermi blazars (, , ; , , ). Linear regression is applied to analyze the correlation between broad line luminosity and jet power. And we obtain for Fermi blazars; for non-Fermi blazars (95% confidence level and ). The Analysis of Variance (ANOVA) is used to test the results of linear regression which shows that it is valid for the results of linear regression (value , probability . Figure 10 shows disk luminosity as a function of jet power. From Fig. 10, it is seen that compared with non-Fermi blazars, Fermi blazars are closer to the ; the jet powers of some BL Lacs are larger than the disk luminosity while some are opposite; the jet power is much smaller than the disk luminosity for most of FSRQs.

We use multiple linear regression analysis to get the relationships between the jet power and both the Eddington luminosity and the broad line region luminosity for Fermi and non-Fermi blazars with 95% confidence level and :

 logPjet=0.49(±0.06)logLBLR−0.01(±0.16)logLEdd+23.6(±7), (2)
 logPjet=0.7(±0.14)logLBLR−0.46(±0.17)logLEdd+34.71(±4.9). (3)

The ANOVA shows that it is valid for the results of multiple linear regression (. From Equations (2) and (3), we see that for Fermi blazars, the black hole mass does not have significant influence on jet power while for non-Fermi blazars, both accretion disk luminosity (referring to accretion rate; Sbarrato et al. 2014; Ghisellini et al. 2014) and black hole mass have contributions to the jet power.

### 3.4 The blazar sequence

The plot of our sample is shown in Fig. 11. The plot of BZCAT catalog is given in Fig. 27 of Abdo et al. (2010c). From Fig. 11 of our sample, it is shown that FSRQs are exclusively located along the top-left/bottom-right band; the of most of BL Lacs are located between 0 and 0.8, and 0.8–2 for . As a comparison, it is found that the space of in Fig. 27 of Abdo et al. (2010c) has some BL Lacs while do not in our Fig. 11; for both our Fig. 11 and Fig. 27 of Abdo et al. (2010c), the top-right space of plot is empty; for the rest of space, our Fig. 11 is consistent with Fig. 27 of Abdo et al. (2010c). Selection effects may cause the deficiency of BL Lacs located in in our Fig. 11 because these BL Lacs are more likely to have nothing data about redshift, black hole mass, jet power. Padovani & Giommi (1995) and Abdo et al. (2010c) presented that the of HBL sources are located between 0.2 and 0.4, and 0.9–1.3 for . In our sample, this region mainly includes non-Fermi BL Lacs. Compared with non-Fermi FSRQs, Fermi FSRQs are more located in top-left of Fig. 11.

Figure 12 shows synchrotron peak luminosity versus synchrotron peak frequency for blazars and Fermi-detected narrow-line Seyfert 1 galaxy. The synchrotron peak luminosity and synchrotron peak frequency for four Fermi-detected narrow-line Seyfert 1 galaxies are collected from Abdo et al. (2009b). Making use of Pearson analysis, we find that there are strong anti-correlations between and for all blazars, Fermi blazars and non-Fermi blazars respectively (; ; ). In Fig. 12, we also find that the four Fermi-detected narrow-line Seyfert 1 galaxies mix in the area of blazars populated, and are in low –low region, which support that -loud narrow-line Seyfert 1 galaxies have similar mechanisms with blazars, and that the low –low blazars are more likely to have low black hole mass (see Chen & Bai 2011).

Ghisellini & Tavecchio (2008) extended the shape of SED from a one-parameter (observed bolometric luminosity) to a two-parameter sequence (black hole mass and accretion rate). So in Fig. 13, we give synchrotron peak frequency () versus black hole mass and Eddington ratio. From results of Pearson analysis, we get that there is not significant correlation between and black hole mass for all blazars () while there is significant anti-correlation between and Eddington ratio for all blazars ().

## 4 Discussions and conclusions

### 4.1 Possible biases in the evaluations of sample and parameters

For our sample, the selection criteria are: (i) we tried to select the largest group of blazars included in BZCAT with reliable broad line luminosity, redshift, black hole mass and jet kinetic power; (ii) in order to have reliable sample, we did not consider the candidate blazars of unknown type; (iii) the non-Fermi blazars, which were detected by EGRET or recorded in 1LAC but missed in 2LAC, were not included in our sample; (iv) the Fermi blazars only focused on clean 2LAC sample. Based on the above criteria, the following objects can be missed: blazars without measured redshift (mainly including BL Lacs), blazars without measured black hole mass, extended 1.4 GHz radio data and broad line data, blazars of uncertain type (BZU), blazars which were detected by EGRET or recorded in 1LAC but missed in 2LAC and blazars classed into non-clean 2LAC. From our selection criteria of sample, it is shown that the number of BL Lacs is missed more in our sample than that of FSRQs because compared with FSRQs, BL Lacs have much less information about redshift, black hole mass and broad line data. Via comparing our redshift distribution to redshift distribution of complete BZCAT sample, we find that the whole distribution and mean value of redshift from our sample agree with the results of complete BZCAT sample. From complete 2LAC sample (Fig. 12 of Ackermann et al. (2011)), it is shown that the redshift distribution peaks around for FSRQs and extends to ; the redshift distribution for BL Lacs extends to and shows a broad maximum between and . So through comparing redshift distributions between our Fermi sample (Fig. 1) and complete 2LAC sample, we find that our redshift distribution of Fermi blazars is similar to redshift distribution of complete 2LAC sample. However, this does not necessarily indicate that the distributions of all other quantities are also similar.

### 4.2 The basic properties of Fermi blazars

The jet formation remains one of the unsolved fundamental problems in astrophysics (e.g. Meier, Koide & Uchida 2001). Many models have been proposed to explain the origin of jets. In current theoretical models of the formation of jet, power is generated via accreting material and extraction of rotational energy of disc/black hole (Blandford & Znajek 1977; Blandford & Payne 1982), and then converted into the kinetic power of the jet. Our results show that the correlations between black hole mass and jet power for both Fermi and non-Fermi blazars are significant. The results of Pearson analysis show that there are significant correlations between broad line luminosity and jet power for both Fermi blazars and non-Fermi blazars, which supports that the jet power has a close link with accretion rate (Sbarrato et al. 2014; Ghisellini et al. 2014). The result is consistent with other authors (e.g. Rawlings & Saunders 1991; Falcke & Biermann 1995; Serjeant, Rawlings & Maddox 1998; Cao & Jiang 1999; Wang, Luo & Ho 2004; Liu, Jiang & Gu 2006; Xie et al. 2007; Ghisellini, Tavecchio & Ghirlanda 2009a; Ghisellini et al. 2009b, 2010, 2011; Gu, Cao & Jiang 2009; Sbarrato et al. 2012). Linear regression is applied to analyze the correlation between broad line luminosity and jet power, and we obtain for Fermi blazars and for non-Fermi blazars. In addition, from Equations (2) and (3), we find that for Fermi blazars, the black hole mass does not have significant influence on jet power while for non-Fermi blazars, both accretion disk luminosity (or accretion rate) and black hole mass have contributions to the jet power. These also can explain our result that the distributions of jet power between Fermi and non-Fermi blazars are significant difference.

### 4.3 The blazar sequence

From Fig. 12 and making use of Pearson analysis, we find that there are strong anti-correlations between and for all blazars, Fermi blazars and non-Fermi blazars respectively. The results support the “blazar sequence” and are not same with results of Meyer et al. (2011). In Fig. 4 of Meyer et al. (2011), an “L”-shape in the plot seems to have emerged which destroys the “blazar sequence”. However, the “L”-shape in Meyer et al. (2011) may be a selection effect of sample. A very possible reason is that in interval of plot of Meyer et al. (2011), there are few blazars. If the interval is full of some blazars, then the “L”-shape will be weaken or disappeared. In our Fig. 12, the interval has many blazars which fill in the interval of Fig. 4 of Meyer et al. (2011). And the result is consistent with others. From Fig. 2 of Nieppola et al. (2008) and Fig. 2, 5 of Finke (2013), it was seen that their the interval still has some blazars. Giommi et al. (2012) used extensive Monte Carlo simulations to get the relation plot of which shows that the interval still has many blazars. The Compton dominance, the ratio of the peak of the Compton to the synchrotron peak luminosities, is essentially a redshift-independent quantity. Finke (2013) used Compton dominance to study the “blazar sequence”. The results of Finke (2013) showed that a correlation exists between Compton dominance and the peak frequency of the synchrotron component for all blazars, including ones with unknown redshift. The sample of Finke (2013) is the 2LAC clean sample. The -ray flares in blazars are likely to occur when the sources are in the high state (Fan et al. 1998). If this is true, the blazars from Finke (2013) are likely to stand for active states of the sources, not their necessarily averaged status. Finke (2013) also presented that SSC emission has the same beaming pattern as synchrotron, and thus Compton dominance does not depend on the viewing angle; however, EC emission does not have the same beaming pattern as synchrotron and SSC, and so Compton dominance is dependent on the viewing angle. Piner & Edwards (2013) have found that beaming factors in the -ray and radio bands are different yet correlated for TeV blazars. An obvious explanation for the ‘bulk Lorentz factor crisis’ is that the radio and -ray emissions are produced in different parts of the jet with different bulk Lorentz factors (Henri & Sauge 2006; Piner, Pant & Edwards 2008, 2010; Piner & Edwards 2013). Therefore, it is possible that the correlation between Compton dominance and peak frequency of the synchrotron component can be resulted from beaming effect. For our Fig. 12, if high high blazars are included, the correlation of will be destroyed. The results of Giommi et al. (2012) also showed that the phenomenological “blazar sequence” is a selection effect. Nieppola et al. (2008) proposed that after being Doppler-corrected, the anti-correlation between and become positive correlation. However, due to limit of Doppler factor, we can not determine the effects of beaming on the blazar sequence in this paper. Ghisellini & Tavecchio (2008) revisited the blazar sequence and proposed that the power of the jet and SED of its emission are linked to the mass of black hole and the accretion rate. From results of Pearson analysis, we get that there is not significant correlation between synchrotron peak frequency and black hole mass for all blazars while significant anti-correlation between synchrotron peak frequency and Eddington ratio for all blazars. The scatter in plot can be resulted from uncertainties of and black hole mass.

## Acknowledgments

We sincerely thank anonymous referee for valuable comments and suggestions. DRX also thanks Minfeng Gu for helpful suggestions. This work is financially supported by the National Nature Science Foundation of China (11163007, U1231203, 11063004, 11133006 and 11361140347) and the Strategic Priority Research Program “The emergence of Cosmological Structures” of the Chinese Academy of Sciences (grant No. XDB09000000). This research has made use of the NASA/IPAC Extragalactic Database (NED), that is operated by Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration.

## References

• Aatrokoski et al. (2011) Aatrokoski J. et al., 2011, A&A, 536, A15
• Abdo et al. (2009) Abdo A.A., Ackermann M., Ajello M. et al., 2009a, ApJ, 700, 597
• Abdo et al. (2009) Abdo A.A. et al., 2009b, 707, L142
• Abdo et al. (2010a) Abdo A.A., Ackermann M., Ajello M. et al., 2010a, ApJS, 188, 405
• Abdo et al. (2010b) Abdo A.A., Ackermann M., Ajello M. et al., 2010b, ApJ, 715, 429
• Abdo et al. (2010c) Abdo A.A., Ackermann M., Agudo I. et al., 2010c, ApJ, 716, 30
• Abdo et al. (2012) Abdo A.A., Ackermann M., Ajello M. et al., 2012, ApJS, 199, 31
• Ackermann et al. (2011a) Ackermann M., Ajello M., Allafort A. et al., 2011, ApJ, 743, 171
• Angel (1980) Angel J.R.P. & Stockman H.S., 1980, ARA&A, 18, 321
• Antonucci (1985) Antonucci R.R.J. & Ulvestad J.S., 1985, ApJ, 294, 158
• Arshakian (2010) Arshakian T.G., Leon-Tavares J., Lobanov A.P., Chavushyan V.H., Shapovalova A.I., Burenkov A.N. & Zensus J.A., 2010, MNRAS, 401, 1231
• Balmaverde et al. (2008) Blandford R.D. & Znajek R.L., 1977, MNRAS, 179, 433
• Balmaverde et al. (2008) Blandford R.D. & Payne D.G., 1982, MNRAS, 199, 883
• Caccianiga (2004) Caccianiga A. & Marcha M.J.M., 2004, MNRAS, 348, 937
• Cassaro (1999) Cassaro P. et al., 1999, A&AS, 139, 601
• Cavagnolo (2010) Cavagnolo K.W. et al., 2010, ApJ, 720, 1066
• Cavaliere & D’Elia (2002) Cavaliere A. & D’Elia V., 2002, ApJ, 571, 226
• Cao & Jiang (1999) Cao X. & Jiang D.R., 1999, MNRAS, 307, 802
• Celotti et al. (1997) Celotti A., Padovani P. & Ghisellini G., 1997, MNRAS, 286, 415
• Chai et al. (2012) Chai B., Cao X. & Gu M., 2012, ApJ, 759, 114
• Chen et al. (2011) Chen L. & Bai J.M., 2011, ApJ, 735, 108
• Cheng & Zhang (2000) Cheng K.S., Zhang X. & Zhang L., 2000, ApJ, 537, 80
• Cooper (2007) Cooper N.J., Lister M.L. & Kochanczyk M.D., 2007, ApJS, 171, 376
• Falcke & Biermann (1995) Falcke H. & Biermann P.L., 1995, A&A, 293, 665
• Falomo (2003) Falomo R., Kotilainen J.K., Carangelo N. & Treves A., 2003, ApJ, 595, 624
• Fan (1998) Fan J.H. et al., 1998, A&A, 338, 27
• Finke (1994) Finke J.D., 2013, ApJ, 763, 134
• Fossati et al. (1998) Fossati G., Maraschi L., Celotti A. et al., 1998, MNRAS, 299, 433
• Francis et al. (1991) Francis P.J., Hewett P.C., Foltz C.B., Chaffee F.H., Weymann R.J. & Morris S.L., 1991, ApJ, 373, 465
• Francis et al. (1991) Georganopoulos M., Kirk J.G. & Mastichiadis A., 2001, in ASP Conf. Ser. 227, Blazar Demographics and Physics, ed. P. Padovani & C. Megan Urry (San Francisco, CA: ASP), 116
• Ghisellini (1998) Ghisellini G., Celotti A., Fossati G. et al., 1998, MNRAS, 301, 451
• Ghisellini & Tavecchio (2008) Ghisellini G. & Tavecchio F., 2008, MNRAS, 387, 1669
• Ghisellini et al. (2009a) Ghisellini G., Maraschi L. & Tavecchio F., 2009a, MNRAS, 396, 105
• Ghisellini et al. (2009b) Ghisellini G., Tavecchio F., Foschini L., Ghirlanda G., Maraschi L. & Celotti A., 2009b, MNRAS, 402, 497
• Ghisellini et al. (2009c) Ghisellini G., Tavecchio F. & Ghirlanda G., 2009c, MNRAS, 399, 2041
• Ghisellini et al. (2010) Ghisellini G., Tavecchio F., Foschini L., Ghirlanda G., Maraschi L. & Celotti A., 2010, MNRAS, 402, 497
• Ghisellini et al. (2011) Ghisellini G., Tavecchio F., Foschini L. & Ghirlanda G., 2011, MNRAS, 414, 2674
• Ghisellini et al. (2014) Ghisellini G., Tavecchio F., Maraschi L., Celotti A. & Sbarrato T., 2014, Nature, 515, 376
• Ghisellini et al. (2015) Ghisellini G. & Tavecchio F., 2015, MNRAS, 448, 1060
• Giommi et al. (1999) Giommi P., Menna M.T. & Padovani P., 1999, MNRAS, 310, 465
• Giommi et al. (2012) Giommi P., Padovani P., Polenta G., Turriziani S., D’Elia V. & Piranomonte S., 2012, MNRAS, 420, 2899
• Giommi et al. (2013) Giommi P., Padovani P. & Polenta G., 2013, MNRAS, 431, 1914
• Gu et al. (2009) Gu M., Cao X. & Jiang D.R., 2009, MNRAS, 396, 984
• Henri (2006) Henri G. & Sauge L., 2006, ApJ, 640, 185
• Jorstad et al. (2001) Jorstad S.G., Marscher A.P., Mattox J.R., Wehrle A.E., Bloom S.D. & Yurchenko A.V., 2001, ApJS, 134, 181
• Kang et al. (2014) Kang S.J., Chen L. & Wu Q.W., 2014, ApJS, 215, 5
• Kharb et al. (2010) Kharb P., Lister M.L. & Cooper N.J., 2010, ApJ, 710, 764
• Kovalev (2009) Kovalev Y.Y. et al., 2009, ApJ, 696, L17
• Landt (2008) Landt H. & Bignall H.E., 2008, MNRAS, 391, 967
• Leon¨CTavares et al. (2011a) Leon-Tavares J., Valtaoja E., Chavushyan V.H. et al., 2011, MNRAS, 411, 1127
• Linford et al. (2011) Linford J.D., Taylor G.B., Romani R. et al., 2011, ApJ, 726, 16
• Linford et al. (2012) Linford J.D., Taylor G.B., Romani R. et al., 2012, ApJ, 744, 177
• Lister et al. (2009a) Lister M.L. et al., 2009a, AJ, 138, 1874
• Lister et al. (2009b) Lister M.L. et al., 2009b, ApJ, 696, 22
• Liu et al. (2006) Liu Y., Jiang D.R. & Gu M.F., 2006, ApJ, 637, 669
• Madau (1987) Madau P., Ghisellini G. & Persic M., 1987, MNRAS, 224, 257
• Maraschi et al. (2008) Maraschi L., Foschini L., Ghisellini G., Tavecchio F. & Sambruna R.M., 2008, MNRAS, 391, 1981
• Maraschi et al. (1992) Maraschi L. & Tavecchio F., 2003, ApJ, 593, 667
• Massaro et al. (2009) Massaro E. et al., 2009, A&A, 495, 691
• Massaro et al. (2012) Massaro F., Paggi A., D’Abrusco R. & Tosti G., 2012, ApJ, 750, L35
• Mead (1990) Mead A.R.G., Ballard K.R., Brand P.W.J.L. et al., 1990, A&AS, 83, 183
• Meier et al. (2001) Meier D.L., Koide S. & Uchida Y., 2001, Science, 291, 84
• Meier et al. (1999) Meier D.L., 1999, ApJ, 522, 753
• Meyer et al. (2011) Meyer E.T., Fossati G., Georganopoulos M. & Lister M.L., 2011, ApJ, 740, 98
• Murphy (1993) Murphy D.W., Browne I.W.A. & Perley R.A., 1993, MNRAS, 264, 298
• Nemmen et al. (2001) Nemmen R.S., Georganopoulos M., Guiriec S., Meyer E.T., Gehrels N., & Sambruna R.M., 2012, Science, 338, 1445
• Netzer et al. (1990) Netzer H., 1990, Active Galactic Nuclei, 57
• Nieppola et al. (2006) Nieppola E., Tornikoski M. & Valtaoja E., 2006, A&A, 445, 441
• Nieppola et al. (2008) Nieppola E., Valtaoja E., Tornikoski M., Hovatta T. & Kotiranta M., 2008, A&A, 488, 867
• Orr (1982) Orr M.J.L. & Browne I.W.A., 1982, MNRAS, 200, 1067
• Padovani (1992) Padovani P. & Giommi P., 1995, ApJ, 444, 567
• Padovani (2003) Padovani P., Perlman E.S., Landt H., Giommi P. & Perri M., 2003, ApJ, 588, 128
• Padovani (2007) Padovani P., Giommi P. & Rau A., 2012, MNRAS, 422, 48
• Piner (2008) Piner B.G., Pant N. & Edwards P.G., 2008, ApJ, 678, 64
• Piner (2010) Piner B. G., Pant N. & Edwards P.G., 2010, ApJ, 723, 1150
• Piner (2012) Piner B.G., Pushkarev A.B., Kovalev Y.Y. et al., 2012, ApJ, 758, 84p
• Piner (2013) Piner B.G. & Edwards P.G., 2013, EPJ Web of Conferences, 6104021P
• Plotkin (2011) Plotkin R.M., Markoff S., Trager S.C. & Anderson S.F., 2011, MNRAS, 413, 805
• Pushkarev (2012) Pushkarev A.B., Kovalev Y.Y., Lister M.L. & Savolainen T., 2012, A&A, 544, 3p
• Rawlings & Saunders (1991) Rawlings S. & Saunders R., 1991, Nature, 349, 138
• Sambruna et al. (2010) Sambruna R.M. et al., 2010, ApJ, 710, 24
• Savolainen et al. (2010) Savolainen T. et al., 2010, A&A, 512, A24
• Sbarrato et al. (2012) Sbarrato T., Ghisellini G., Maraschi L. & Colpi M., 2012, MNRAS, 421, 1764
• Sbarrato et al. (2014) Sbarrato T., Padovani P. & Ghisellini G., 2014, MNRAS, 445, 81
• Scarpa & Falomo (1997) Scarpa R. & Falomo R., 1997, A&A, 325, 109
• Serjeant et al. (1998) Serjeant S., Rawlings S. & Maddox S.J., 1998, MNRAS, 294, 494
• Serjeant et al. (1998) Shen Y., Richards G.T., Strauss M.A. et al., 2011, ApJS, 194, 45
• Taylor (2007) Taylor G.B. et al., 2007, ApJ, 671, 1355
• Tchekhovskoy (2011) Tchekhovskoy A., Narayan R. & McKinney J.C., 2011, MNRAS, 418, L79
• Urry & Padovani (1995) Urry C.M. & Padovani P., 1995, PASP, 107, 803
• Urry & Padovani (1995) Veron-Cetty M.P. & Veron P., 2010, A&A, 518, 10
• Wang et al. (2004) Wang J.M., Luo B. & Ho L.C., 2004, ApJ, 615, L9
• White (2000) White R.L., Becker R.H., Gregg M.D. et al., 2000, ApJS, 126, 133
• Wills (1995) Wills B.J. & Brotherton M.S., 1995, ApJ, 448, L81
• Woo (2005) Woo J.H., Urry C.M., van der Marel R.P. et al., 2005, ApJ, 631, 762
• Woo & Urry (2002) Woo J.H. & Urry C.M., 2002, ApJ, 579, 530
• Wu (2009) Wu Z.Z., Gu M.F. & Jiang D.R., 2009, RAA, 9, 168
• Wu (2014) Wu Z.Z., Jiang D.R. & Gu M.F., 2014, A&A, 562, 64
• Xie et al. (2004) Xie G.Z., Zhou S.B. & Liang E.W., 2004, AJ, 127, 53
• Xie et al. (2007) Xie G.Z., Dai H. & Zhou S.B., 2007, AJ, 134, 1464
• Xiong (2014) Xiong D.R. & Zhang X., 2014, MNRAS, 441, 3375
• Yee (1978) Yee H.K.C. & Oke J.B., 1978, ApJ, 226, 753
You are adding the first comment!
How to quickly get a good reply:
• Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
• Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
• Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters