The detection of the blazar S4 0954+65 at very-high-energy with the MAGIC telescopes during an exceptionally high optical state
Key Words.:gamma rays: galaxies / galaxies: active / BL Lacertae objects: individual: S4 0954+65
Aims:The very-high-energy (VHE, GeV) -ray MAGIC observations of the blazar S4 0954+65, were triggered by an exceptionally high flux state of emission in the optical. This blazar has a disputed redshift of =0.368 or 0.45 and an uncertain classification among blazar subclasses. The exceptional source state described here makes for an excellent opportunity to understand physical processes in the jet of S4 0954+65 and thus contribute to its classification.
Methods:We investigate the multiwavelength (MWL) light curve and spectral energy distribution (SED) of the S4 0954+65 blazar during an enhanced state in February 2015 and put it in context with possible emission scenarios. We collect photometric data in radio, optical, X-ray, and ray. We study both the optical polarization and the inner parsec-scale jet behavior with 43 GHz data.
Results:Observations with the MAGIC telescopes led to the first detection of S4 0954+65 at VHE. Simultaneous data with Fermi-LAT at high energy ray (HE, 100 MeV ¡ E ¡ 100 GeV) also show a period of increased activity. Imaging at 43 GHz reveals the emergence of a new feature in the radio jet in coincidence with the VHE flare. Simultaneous monitoring of the optical polarization angle reveals a rotation of approximately 100.
Conclusions:The high emission state during the flare allows us to compile the simultaneous broadband SED and to characterize it in the scope of blazar jet emission models. The broadband spectrum can be modeled with an emission mechanism commonly invoked for flat spectrum radio quasars, i.e. inverse Compton scattering on an external soft photon field from the dust torus, also known as external Compton. The light curve and SED phenomenology is consistent with an interpretation of a blob propagating through a helical structured magnetic field and eventually crossing a standing shock in the jet, a scenario typically applied to flat spectrum radio quasars (FSRQs) and low-frequency peaked BL Lac objects (LBL).
Blazars are a subclass of Active Galactic Nuclei (AGN) in which the relativistic jet presents a small viewing angle towards the observer and thus where relativistic effects on the observed emission are more extreme. Conventionally, blazars are subdivided in BL Lac objects and FSRQs depending on the characteristic of their optical spectrum: while BL Lac objects are dominated by the featureless continuum emission from the jet, FSRQs typically show wide optical emission lines.
The blazar S4 0954+65 hosts a black hole of mass , estimated from the width of the H line (Fan & Cao, 2004). The detection of the H line is not confirmed by Landoni et al. (2015) (see the discussion on the redshift determination) so that the mass estimation cannot be confirmed either.
This blazar presents strong variability in the optical band, already well studied by Wagner et al. (1990) and by Morozova et al. (2014). Intra night variability has been found both in optical and radio wavelengths (Wagner et al., 1993).
The optical high brightness state of February 2015, presented here, is however exceptional for the object, with a brightening of more than 3 magnitudes in the R-band with respect to the average monitored state
The source GRO J0957+65, detected with the EGRET telescope on board the Compton Gamma-Ray Observatory, has been associated through optical and radio observations with S4 0954+65 by Mukherjee et al. (1995). S4 0954+65 has been afterwards always included in the released catalogs of sources detected by the Large Area Telescope (LAT) instrument on board the Fermi satellite (Abdo et al., 2010; Nolan et al., 2012; Ackermann et al., 2013; Acero et al., 2015; Ackermann et al., 2016; Ajello et al., 2017), with the exclusion of the bright source list released after the first 3 months of Fermi-LAT data integration.
The classification of the object, based on the available literature, is still unclear. In most of the ATels mentioned above S4 0954+65 is referenced as a FSRQ, but in most of the literature this is classified as a BL Lac object due to the small equivalent width of the emission lines in its spectrum (see, e.g. Stickel et al., 1991). Sambruna et al. (1996) classified the SED of S4 0954+65 as “FSRQ-like”, in a sample limited to the sources with a detection from EGRET data. It indeed presents a flatter spectral index than most BL Lac objects, in both X-ray and -ray bands (see Raiteri er al., 1999, and references therein). Among BL Lac objects, a further phenomenological subdivision can be made based on the frequency of the synchrotron peak, ranging from optical to X-ray frequency and identifying the classes of low-, intermediate- or high-peaked BL Lac object (LBL, IBL, HBL respectively). Ghisellini et al. (2011) classified this object as a LBL based on the SED. When including the kinematic features from the radio jet in the classification templates, Hervet et al. (2016) classify this as their kinematic class II, mostly composed of FSRQ. S4 0954+65 can thus be interpreted as a transitional object between FSRQ and classical BL Lac objects.
The most numerous extragalactic sources detected at VHE from Imaging Air Cherenkov Telescopes (IACTs), presently, belong to the HBL class. Therefore the VHE detection of an object such as S4 0954+65 provides a rare opportunity to study VHE emission conceivably produced in a different kind of environment. Indeed, while emission in HBL can mostly be satisfactorily modeled taking into account only processes in a compact feature in the jet, for FSRQs the inclusion of the interactions of such a feature with the surrounding ambient becomes of greater importance (see e.g. Tavecchio, 2016). The structure of the broadband SED collected here will also be put in context with other common characteristics of a FSRQ classification, such as intrinsic brightness, peak of the synchrotron component and Compton dominance.
Also the question of S4 0954+65 redshift is still not settled, as claims of line detection in the optical spectrum are not always confirmed. The redshift of the source was first determined at =0.368 by the identification of lines by Lawrence et al. (1986, 1996). Stickel et al. (1993) obtained, from different measurements, the same redshift estimate based on line identification. None of these lines were confirmed by the observations reported in Landoni et al. (2015), who instead pose a lower limit of z. The latter results were obtained with a superior resolution spectra. At the time of the observation the magnitude in R-band of the object was 15.5, while it is known from variability studies that it could be even 2 magnitudes lower. In the following we will adopt the redshift =0.368.
The outline of this paper is as follows. In Section 2, we will present the MAGIC telescopes and the relative data set on S4 0954+65. Section 3 reviews all the MWL data that were collected during this exceptional burst, whereas Section 4 discusses the implication of this burst for the source state and inner jet structure. Additional information on the MAGIC data analysis, the parameters derived from the VLBA data and the full dataset for Swift-XRT X-ray data will be found in Appendix A and B, respectively.
2 MAGIC Observations
The MAGIC telescopes are an array of two IACTs located in the Island of La Palma (Spain) at an altitude of 2200 m asl. The system is sensitive down to an energy threshold of GeV (Aleksić et al., 2016) for low zenith angle observations. This is of particular relevance for the monitoring of variable sources and of those that tend to exhibit a steep spectrum at VHE. The full data have been analyzed using the standard MAGIC analysis chain and the MAGIC Standard Analysis Software (MARS, Zanin et al., 2013; Aleksić et al., 2016).
The MAGIC collaboration supports a program of Targets of Opportunity (ToO), triggered by MWL monitoring. The ToO program was activated for observations of S4 0954+65 at the end of January 2015 after the first hints of enhanced optical state (triggered by the Tuorla monitoring in R-band, see Section 3.3). We observed the source with the MAGIC telescopes for 2 nights (MJD 57049-57050, 2015 January 27 and 28), for a total of 1 hour high-quality dark time data, but obtained no detection. We resumed the ToO observations in February after the Tuorla monitoring revealed a very exceptional flux state, later confirmed by other monitoring programs (see Section 3.3). We obtained a detection at a significance of from observations during 2015 February 14 (MJD 57067, ATel #7080 Mirzoyan et al., 2015). We continued observing S4 0954+65, barring adverse atmospheric conditions, until full moon days when standard MAGIC observations are not possible due to the elevated level of background light (last day of observation, with already large moonlight contamination, on 2015 March 1, MJD 57082). A detailed breakdown of the observation conditions and relative results can be found in Appendix A.
The total excess from the dark-time data is consistent with a point source emission (see Fig. 1). No other significant emission is found in the field of view apart from the one coincident with S4 0954+65 at the center.
The SED points presented in Section 4 below are derived for the day of the flare (MJD 57067, 2015 February 14), using only data taken in dark conditions (that allow for the lowest threshold and lowest systematic uncertainty, Appendix A). We follow the standard MAGIC unfolding procedure (Albert et al., 2007) to obtain the intrinsic spectrum.
The -ray emission from sources at high redshift is absorbed via photon-photon pair production on photons from the extragalactic background light (EBL, see e.g. Domínguez et al., 2011; Finke et al., 2010). S4 0954+65 redshift is assumed to be . The spectral shape of the intrinsic emission, i.e. after the correction for the EBL absorption, can be fitted with a simple power law:
with normalization at TeV and spectral index . The quoted systematic uncertainties are derived from the standard evaluation in MAGIC data presented by Aleksić et al. (2016). Note that the calculated systematic uncertainty on does not contain the uncertainty on the energy scale, that is about 15%. The unfolded MAGIC spectrum is shown in Fig. 2. The unfolded observed spectrum, i.e. without correcting for the EBL absorption, can be described also by a simple power law with at TeV and spectral index .
3 The Multiwavelength coverage
All the data presented in this section are collected to produce the light curves and SED, whose interpretation is later presented in Section 4.
The LAT on board the Fermi satellite scans the entire sky every 3 hours. From the data of the first 4 years of operation, S4 0954+65 was detected with an average significance of in the energy range from 100 MeV to 300 GeV as reported in the 3FGL catalog (Acero et al., 2015). A dedicated analysis from MJD 56952 (2014 October 22) to MJD 57208 (2015 July 05) is presented in this work. We selected Pass 8 source class events within a 10 circular region centered on the position of S4 0954+65, in the energy range 0.1-500 GeV. The spectral analysis was performed through an unbinned likelihood fit, using the ScienceTools software package version v11-05-00 along with the instrument response functions P8R2_SOURCE_V6. The model of the likelihood fit includes a Galactic diffuse emission model and an isotropic component
The light curve was calculated in day timescale bins, modeling the source with a single power-law spectrum (as it is also described in the 3FGL). Both the flux and spectral index of S4 0954+65 were left free during the likelihood fits, while the rest of the point sources were fixed and only the diffuse Galactic and isotropic models were allowed to vary. In case of TS¡4, an upper limit on the flux was calculated fixing the spectral index to 2.38 as given in the 3FGL catalog. The results are shown in Fig. 3. The figure shows also the light curve calculated in 15-days bin as comparison. The light curve was obtained with the same procedure described above for the 1-day binning. During the HE flare in November 2014 (MJD 56976, ATel #6709; Krauss, 2014) the LAT spectral index is compatible with its 3FGL value of , averaged from 4 years of data. Moreover, the visibility of the source by MAGIC was at an unfavorable zenith angle of 60 (implying a high energy threshold). Therefore, no ToO observation was activated with MAGIC for this flare. MAGIC observations were activated later on during the strong flare on February 2015 when the LAT detected a hardening of the spectrum as shown previously by Tanaka et al. (2016) where the LAT analysis using Pass 7 reprocessed data is presented.
The spectral analysis for the MWL SED corresponds to 1-day integration centered in the MAGIC observation (MJD 57067.14, 2015 February 14). From a first likelihood fit we found the best spectral fit was a power-law spectral index of (significantly harder than its average 3FGL value) and was fixed in the model for the spectral points calculation. Moreover, all the sources included in the model except the diffuse Galactic and isotropic models were also fixed. The source was detected during this period with a TS of 379.7. A curved spectral model is not significantly favored in this day (TS for a log parabola fit is TS=380.10 to be compared with a simple power law fit with TS=379.74).
3.2 Swift dataset
The 22 multi epochs event-list obtained by the X-ray Telescope (XRT, Burrows et al., 2004) on board the Swift satellite in the period of 2014 November 17 (MJD 56978.96395) to 2015 March 11 (MJD 57092.26632) with a total exposure time of 11.12 hours were processed using the procedure described by Fallah Ramazani et al. (2017). All these observations had been performed in photon counting (PC) mode, with an average integration time of 1.8 ks each. The equivalent Galactic hydrogen column density is fixed to the value of (Kalberla et al., 2005).
The average integral photon X-ray flux (0.3-10 keV) in this period is . The X-ray flux is peaking at MJD 57070.76523 with which is a factor of about 2 higher than the average flux of the analyzed period. The average flux outside the flare period (2006-2015) is , that we derived from a sample of XRT data comprising 25 X-ray exposures in the XRT database, not including the 22 multi epochs event-list described above. This indicates that the source was clearly in its X-ray high state during the VHE -ray detection. The X-ray spectral index during the analyzed period varies between . It is notable that the softest spectral index was obtained a night prior to the VHE -ray flare while the spectra starts to harden after 2015 February 14 and reach its historical hardest spectra 10 days after the VHE -ray flare. The X-ray spectra on the night before and after the VHE -ray flare can be well described with a power-law with spectral index of (d.o.f.=1.024/41) and (1.025/24 d.o.f.) respectively. The full dataset analysis is given in Appendix C.
The Swift satellite hosts an additional instrument, the Ultraviolet/Optical Telescope (UVOT, Poole et al., 2008). The data taken during the period of interest for this work have already been presented by Tanaka et al. (2016). They follow the behavior of the optical light curve that we will present next. Therefore they are not reproduced again nor shown in Fig. 3. The UVOT bands are however important for the SED modeling presented in Section 4 and will therefore be included there for MJD 57067 (2015 February 14, day of the VHE detection). The dataset presented by Tanaka et al. (2016) suffers from an incorrect exposure calculation by a factor of 2, related to the deadtime correction, and thus a lower reconstructed flux. We therefore have performed a re-analysis here for the two exposures taken with UVOT on MJD 57066.76. Data reduction has been done on all the available filters (), following the standard UVOT data analysis prescriptions
3.3 The optical domain
Optical data were collected with: 35cm KVA telescope (La Palma Island, Spain) used in the Tuorla monitoring program; 1.8 m Perkins telescope of Lowell Observatory (Flagstaff, Arizona); 70 cm telescope AZT-8 at the Crimean Astrophysical Observatory (Nauchny, Russia); 40 cm telescope LX-200 of St. Petersburg State University (St. Petersburg, Russia); IAC80/Camelot at the Teide Observatory (Tenerife, Spain). The data analysis from KVA was performed with the semi-automatic pipeline using the standard analysis procedures (Nilsson et al. in prep). The differential photometry was performed using the comparison star magnitudes from Villata et al. (1997). For the Perkins telescope see Jorstad et al. (2010) and references therein. The details of observations and data reductions with AZT-8 and LX-200 are given by Larionov et al. (2008). IAC80/Camelot data were automatically processed by the pipeline Redcam and calibrated astrometrically using XParallax, both available at the telescope. Instrumental magnitudes for IAC80/Camelot data were extracted using Sextractor (Bertin & Arnouts, 1996) and calibration of the source magnitude was obtained with respect to the reference stars provided by Raiteri er al. (1999).
All the above mentioned telescopes provide R-band photometry. We have applied the calibration of Mead et al. (1990) for all optical measurements to transform magnitudes into flux densities, and dereddened the flux according to the absorption by Schlafly & Finkbeiner (2011). The host galaxy is not detected for this object.
From the Perkins, AZT-8+ST7 and LX-200 telescopes we collect also polarization information. In Fig. 3 we show the optical photometry data and time evolution of the fractional linear polarization and the electrical vector position angle (EVPA) in R-band. The EVPA measurements have been arranged such to minimize the impact of the ambiguity, i.e. adding or subtracting whenever two subsequent measurements differ by more than .
In the same timeframe of the VHE detection and the optical flare, a substantial change in the optical EVPA can be identified (see Fig. 3). The EVPA rotation starts just before the optical and VHE flare and reaches a total change of roughly 100. The optical flare in February 2015 is a factor of about 3 larger in flux than the 2011 flare (see Morozova et al., 2014), that was already exceptional and concurrent with a series of -ray flares evident in Fermi-LAT data. During the most extreme flare in 2011, the EVPA rotated by about 300.
|Knot||Average Flux||Maximum Flux||Average PA||Average Size||Proper motion||Apparent Speed||Time of Ejection|
|mJy||mJy||deg ()||(FWHM) mas||mas/yr||c||MJD|
3.4 The radio and millimeter ranges
S4 0954+65 was monitored at 3.5 mm (86 GHz) and 1.3 mm (229 GHz) wavelengths from the IRAM 30 m Millimeter Radiotelescope under the POLAMI (Polarimetric Monitoring of AGN at Millimeter Wavelengths)
S4 0954+65 is monitored monthly by the Boston University (BU) group with the Very Long Baseline Array (VLBA) at 43 GHz within a sample of bright -ray blazars through the VLBA-BU-BLAZAR program
Of particular interest is knot K15, which is very compact, with a FWHM average size of mas and presents the largest apparent speed of ()c, cf Fig. 5. The zero-epoch separation of this knot is consistent with the VHE flare considering its 18-day uncertainty. The intensity of the core is increasing in the epoch of MJD 57067 observation, but no significant change in the core polarization can be appreciated. The detailed information on the time evolution of the radio knot can be found in Table 4, while the polatization evolution details are shown in Table 5. During November 2014, while the source was high in the HE band as observed by Fermi-LAT but without optical enhancement, no new knot appears. The zero epoch-separation from the core of knots K14a and K14b are not coincident with the high state in Fermi-LAT data of November 2014, but happen months before. We analyzed Fermi-LAT data in the period included within the error band for K14a and K14b zero epoch-separation and found no particular enhancements.
To be noted is also the position angle of K15 with respect to the core, (PA=). This is different than the values reconstructed from previous knots, ranging from roughly PA= to PA= in Morozova et al. (2014), that are in turn consistent with the values for K14a/b. The mean jet direction is at PA. A difference in PA and in apparent speed could be simply related to a small difference in the angle to the observer. However, the highest apparent speed can be used to estimate the Doppler factor, considering the upper limit to largest possible viewing angle and ultimately leading to . Applying this to the above mentioned knots (averaging the apparent speed to c for K14a/b): and ; and .
The 37 GHz observations were made with the 13.7 m diameter telescope at Aalto University Metsähovi Radio Observatory. A detailed description of the data reduction and analysis is given by Teraesranta et al. (1998). The error estimate in the flux density includes contributions from the measurement RMS and the uncertainty of the absolute calibration. The S4 0954+65 observations were done as part of the regular monitoring program and the GASP-WEBT campaign. There are no strictly simultaneous 37GHz data to the MAGIC detection, however an increase in flux can be seen when comparing observation taken one day before (2015 February 13, MJD 57066.15, Jy) and one day after the MAGIC detection (2015 February 15, MJD 57068.15, Jy).
The OVRO 40 m uses off-axis dual-beam optics and a cryogenic pseudo-correlation receiver with a 15.0 GHz center frequency and 3 GHz bandwidth. Calibration is achieved using a temperature-stable diode noise source to remove receiver gain drifts and the flux density scale is derived from observations of 3C 286 assuming the Baars et al. (1977) value of 3.44 Jy at 15.0 GHz. The systematic uncertainty of about 5% in the flux density scale is not included in the error bars. Complete details of the reduction and calibration procedure are found in Richards et al. (2011). The long-term monitoring program at OVRO (Owens Valley Radio Observatory) monitors the variability of this source at 15GHz over a longer time than what shown here. While it is obvious that the source was variable also during February 2015, it is not an exceptionally bright flux state of the source in the radio band. From a decade long monitoring, the source shows brighter levels (highest at Jy) and fainter levels (lowest at Jy).
Both 15 GHz and 37 GHz data seem to be in agreement with the behavior seen from mm wavelength data. Again note the lack of strictly simultaneous data to the MAGIC peak detection (MJD 57067). emission.
The coverage of flaring states at VHE is helpful to understand the jet dynamics. We present a discussion of the SED for the day of the flare (2015 February 14). We do not attempt a SED modeling for other days, for which the MAGIC data would provide only non-constraining upper limits to emission at VHE. The day of the VHE detection is instead put in context with a longer time span behavior in the MWL dataset. However the VHE sampling of the state is too scarce to attempt a numerical correlation study of the light curves.
4.1 Light curve phenomenology
The MWL light curves of the source for all the instruments involved in the present work are reported in Fig. 3, and cover a time range of 7 months, from MJD 56970 (2014 November 19) to MJD 57200 (2015 June 27). The panels of Fig. 3, in order of decreasing energy, show in the top panel the MAGIC detection at VHE, while the radio data collected by OVRO, POLAMI and the other instruments in the radio band are shown in the bottom one. The red region indicates the time window where the knot K15 was ejected in the VLBA analysis, as reported in Table 1: a time range of 36 days centered in MJD 57081 (2015 February 28). The VHE detection and the enhanced activity in the other bands are found inside the K15 ejection time window, making this event important for the understanding of the whole scenario. The spectral index at HE as inferred from the Fermi-LAT data is harder than the average spectral index of from the 3FGL catalog dataspan. In the presented timeframe, the X-ray emission peaks around the observation on MJD 57070.76434 (2015 February 17), with a delay with respect to the detection in VHE. The 3 hours of observations in VHE during the same night did not lead to a detection (see Table 3). However during the period of enhanced MWL activity, there is a clear hardening of the X-ray spectrum. Hardening at both X-ray and -ray energies points toward the emergence of a new component in the non-thermal spectrum.
The optical band is very bright during the VHE detection, reaching peaks of more than 20 mJy of flux density when the average behavior of the source is found around a few mJy (see the optical monitoring from Tuorla observatory). The optical emission is polarized by a fraction of and the polarization angle rotates by during the flare: Blinov et al. (2015) have shown that from a systematic monitoring (Robopol monitoring) of both -ray loud and -ray quiet sources, only the former class of object displays polarization angle rotation similar to the one seen here for S4 0954+65. Blinov et al. (2015) studied the change of EVPA as a function of time for smooth changes of . Requesting the same smoothness requirements, no smooth rotation of can be identified in the dataset presented here, see Fig. 6. A non smooth variation of can however be identified between MJD 57060 and MJD 57075. This variation would imply a change of the EVPA curve slope of , compatible with the bulk of the variations studied by Blinov et al. (2015). The rotations of the polarization angle are often physically linked to high flaring states of the objects in the -ray band. While individual occurrences of -ray flares and rotations cannot be firmly linked to each other, there is a low probability that all the occurrences are due to chance coincidence (from MonteCarlo simulations in Blinov et al., 2015). This hypothesis is still confirmed from 3 years of Robopol monitoring data in Blinov et al. (2018). Kiehlmann et al. (2017) also study whether a simple stochastic variation can account for the observed rotations in the Robopol monitoring. While their model is failing to recover all the observational characteristics in the monitoring, it also highlights a larger discrepancy from the expectations of stochastic model with respect to the occurrence of large variations of EVPA (), however not significant. Smooth variations seem also to be more firmly linked to deterministic processes and not to a random walk effect (Kiehlmann et al., 2016). Robopol monitoring data are also used in Angelakis et al. (2016), to study the difference in the amount of polarization seen on average in -ray loud and -ray quiet sources. The median fraction variability of the S4 0954+65 dataset presented here is . This value can be compared with the average for the -ray loud subset of the Robopol monitoring and a value of for S4 0954+65 computed for the observations on year 2013 and 2014. According to the interpretation by Angelakis et al. (2016), a higher fractional polarization is also expected in LSP/ISP blazars, due to the fact that in such sources the optical synchrotron emission relates to the peak synchrotron emission. Therefore, the particles associated with this emission are the most energetic, with faster cooling and thus probing a small volume of the emission region near the acceleration region, where it is expected to have a stronger ordered (helical) magnetic field, leading to higher polarization fraction.
Images at 43 GHz show the emergence of new knots. In Morozova et al. (2014), a series of optical flares of S4 0954+65 in 2011 are studied, and the emission of knots is found correlated to the simultaneous flaring of the optical and HE bands. The maximum flux in the 2011 state is a factor of 3 lower in optical than the state presented here. The polarization fraction in this 2011 flare was similar to that seen in the present work. In Morozova et al. (2014) the chance coincidence of high optical state and knot emission has very low probability.
The phenomenology of the 2015 flare described here agrees very well with the model put forward by Marscher et al. (2008) and applied to the S4 0954+65 dataset of Morozova et al. (2014). In that model, the flare is due to a newly appearing knot accelerating at the base of the jet and propagating through an helical flow streamline. The helical streamline can be expected due to the anchoring of the accelerating flow to the rotating base of the accretion disk or black hole magnetosphere, depending on modeling. The magnetic field topology in the jet is also helical and ordered. Geometrical effects and the propagation through the helical magnetic field account for the rotation of the EVPA.
In Zhang et al. (2014), a model is proposed where the EVPA rotation is also related to the propagation through an helical magnetic field, but the streamline of propagation is not necessarily helical itself. In this model the magnitude of the swing can depend on the assumptions on the settings for the flare, specifically the magnetic field strength and orientation, the acceleration efficiency and the continuous injection of freshly accelerated particles.
The model described in Marscher et al. (2008) allows the emission at radio wavelengths in a flaring state which is not simultaneous with the VHE flare. In this scenario the radio activity could be delayed several days, even months, with respect to the VHE detection. This is expected if synchrotron self absorption is involved, and hence the emission region is located closer to the central engine than the radio core (A0 in Fig. 4). The peak of radio emission is expected to be lagging behind and appear when the disturbance has propagated further down the jet, where the absorption is not an issue. The X-ray emission peak, then, could also be delayed with respect to the optical outburst. As the X-ray emission is probably due to IC of an external soft photon field by electrons in the jet (see above), the X-ray variability traces both the accelerated particle distribution and a change in the soft photon field. This retraces similar interpretation drawn for flares of other sources where the dataset was however richer and more detailed (Marscher et al., 2008, 2010; Aleksić et al., 2014; Ahnen et al., 2017).
4.2 Emission model for the flare SED
The SED of blazars are dominated by their non-thermal emission and can usually be described by two broad components. The low energy non-thermal emission is explained as synchrotron emission, while the high energy emission is most commonly modeled through inverse Compton (IC) emission, where soft photons are upscattered to -ray energies by electrons within the jet emitting region. The origin of the soft photon field itself can vary for different blazar subclasses. In particular, for most of the classical BL Lac objects, the VHE emission can be reasonably modeled through Synchrotron self-Compton emission (SSC, see e.g. Rees, 1967; Maraschi et al., 1992). Instead, for the case of FSRQs, the modeling of the emission usually requires the inclusion of external soft photon fields from e.g. the infrared dusty torus or the optical-ultraviolet emission from the Broad Line Region (BLR) for the IC process (see e.g. Tavecchio, 2016).
A broadband SED is compiled for 2015 February 14 (MJD 57067). We collect, from the MWL sample described in Section 3, the data closest in time to the MAGIC observation. Fermi-LAT data points are obtained from a 1-day integration centered on the MAGIC observation. The specific dates of other wavelength observations are given in the caption of Fig. 7.
Tanaka et al. (2016) model the SED of S4 0954+65 during a similar integration time as the 2015 flare studied in this work. The data shown in Fig. 7 include, in addition to what is shown by Tanaka et al. (2016), the VHE data from the MAGIC observation, the AZT-8+ST7 and POLAMI data. Moreover, the Fermi-LAT data are reanalyzed as described in Section 3 to be centered at the MAGIC observation time and benefit from the latest Fermi-LAT PASS8. The Swift-XRT and Swift-UVOT data are also reanalyzed for this work.
Tanaka et al. (2016) report that a SSC modeling of the data is challenging, requiring very low magnetic field (G in contrast to the expected in blazar jet components). Alternatively, an External Compton (EC) modeling was able to reproduce the data. In their model, the soft photon field for the EC model was the dusty torus from the source. In Fig. 7, we plot the model from Tanaka et al. (2016). This model reproduces the Fermi-LAT and MAGIC data, although their paper did not include any MAGIC data. However, the model fails to reproduce properly the optical observations. Such underestimation at optical frequencies in the model of Tanaka et al. (2016) is driven by a misreconstruction of the UVOT fluxes, explained in Section 3. With the reanalyzed UVOT dataset presented here, we use a new model, using the same code and most of the same assumptions as in Tanaka et al. (2016), including a redshift of . The code is explained in detail in Finke et al. (2008) and Dermer et al. (2009). Note that the presented SED model curves already include the effect of EBL absorption, i.e the intrinsic emission is absorbed according to the EBL model by Finke et al. (2010). The new EC model provides a good description of the MWL data and is shown in Fig. 7. The parameters of both models are reported in Table 2. The break in the underlying electron population is similar to what expected by classical cooling, with the slope of the electron distribution before of the break () and after the break () differing by . The use of VHE spectral information is crucial to model the falling part of the high energy peak of the blazars SEDs, which is crucial to constrain the most energetic electrons within the leptonic framework scenario (SSC and EC models).
|Parameter||Symbol||Model A||Model B|
|Tanaka et al. (2016)||this work|
|Bulk Lorentz Factor||30||35|
|Variability Timescale [s]|
|Comoving radius of blob [cm]||6.610||3.010|
|Magnetic Field [G]||0.6||0.4|
|Low-Energy Electron Spectral Index||2.4||2.4|
|High-Energy Electron Spectral Index||4.5||3.6|
|Minimum Electron Lorentz Factor|
|Break Electron Lorentz Factor|
|Maximum Electron Lorentz Factor|
|Black hole Mass [|
|Disk luminosity |
|Inner disk radius |
|Seed photon source energy density |
|Seed photon source photon energy [ units]|
|Dust Torus luminosity |
|Dust Torus radius [cm]|
|Dust temperature [K]|
|Jet Power in Magnetic Field |
|Jet Power in Electrons |
As mentioned in the introduction, the classification of a blazar can be aided by the study of its SED characteristics. According to the SED model presented above, the peak of the synchrotron emission is at Hz, making it an intermediate synchrotron peaked BL Lac object (Ackermann et al., 2015)
The census of extragalactic objects that present VHE emission is still limited. We present here the first detection at VHE of the blazar S4 0954+65 obtained through observations with the MAGIC Telescopes. The observations were conducted during an exceptional flare of the source in February 2015, originally identified in the optical band. We collected MWL simultaneous data to better characterize the state of the source.
The HE emission is also found in elevated state from the analysis of Fermi-LAT data, which reveal the hardest state of the HE emission to be concurrent with the detection at VHE. The X-ray emission peak is delayed by a few days with respect to the VHE detection and shows a trend of spectral hardening during the period presented here. The radio and mm wavelength emission reveal a moderate elevation of the flux, that is however not exceptional in the long term behavior of the source.
The source is classified in the literature as a BL Lac, but we have shown here that it presents similarities with the FSRQ class. Results from the monitoring of optical polarization and 43 GHz jet component analysis were compared to archival observation of S4 0954+65 and of statistical behaviour of other sources. Three main measurements were considered: the day of the VHE detection of S4 0954+65 is included in the error box for the zero epoch separation of knot K15; the optical polarization fraction is increasing in the same period; a rotation of optical EVPA of can be identified, also in the same period, possibly related to the helical structure of the magnetic field in the acceleration region. We have discussed how these measurements point to a common behaviour with ISP/LSP sources. Both the best emission model (EC on dust torus) and the MWL light curve behavior show points of contact with other sources that are either clear FSRQ (like PKS 1510-089) or are transitional objects (like BL Lac itself). This is also supported from the moderate Compton dominance in the SED model and the fact that the synchtrotron peak show that the source can be classified as ISP source.
The work presented here reiterates the importance of VHE ray and detailed MWL studies of blazars during different flux states to test their intrinsic characteristics and shed light on the physical processes taking place within their jets.
Acknowledgements.We would like to thank the Instituto de Astrofísica de Canarias for the excellent working conditions at the Observatorio del Roque de los Muchachos in La Palma. The financial support of the German BMBF and MPG, the Italian INFN and INAF, the Swiss National Fund SNF, the ERDF under the Spanish MINECO (FPA2015-69818-P, FPA2012-36668, FPA2015-68378-P, FPA2015-69210-C6-2-R, FPA2015-69210-C6-4-R, FPA2015-69210-C6-6-R, AYA2015-71042-P, AYA2016-76012-C3-1-P, ESP2015-71662-C2-2-P, CSD2009-00064), and the Japanese JSPS and MEXT is gratefully acknowledged. This work was also supported by the Spanish Centro de Excelencia “Severo Ochoa” SEV-2012-0234 and SEV-2015-0548, and Unidad de Excelencia “María de Maeztu” MDM-2014-0369, by the Croatian Science Foundation (HrZZ) Project IP-2016-06-9782 and the University of Rijeka Project 188.8.131.52.02, by the DFG Collaborative Research Centers SFB823/C4 and SFB876/C3, the Polish National Research Centre grant UMO-2016/22/M/ST9/00382 and by the Brazilian MCTIC, CNPq and FAPERJ. The Fermi LAT Collaboration acknowledges generous ongoing support from a number of agencies and institutes that have supported both the development and the operation of the LAT as well as scientific data analysis. These include the National Aeronautics and Space Administration and the Department of Energy in the United States, the Commissariat à l’Energie Atomique and the Centre National de la Recherche Scientifique / Institut National de Physique Nucléaire et de Physique des Particules in France, the Agenzia Spaziale Italiana and the Istituto Nazionale di Fisica Nucleare in Italy, the Ministry of Education, Culture, Sports, Science and Technology (MEXT), High Energy Accelerator Research Organization (KEK) and Japan Aerospace Exploration Agency (JAXA) in Japan, and the K. A. Wallenberg Foundation, the Swedish Research Council and the Swedish National Space Board in Sweden. This work performed in part under DOE Contract DE-AC02-76SF00515 This research has made use of the NASA/IPAC Extragalactic Database (NED), which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. Part of this work is based on archival data, software or online services provided by the Space Science Data Center - ASI. The OVRO 40-m monitoring program is supported in part by NASA grants NNX08AW31G, NNX11A043G and NNX14AQ89G, and NSF grants AST-0808050 and AST-1109911 The research at Boston University was supported by NASA Fermi Guest Investigator program grant 80NSSC17K0694 and US National Science Foundation grant AST-1615796. The VLBA is an instrument of the Long Baseline Observatory. The Long Baseline Observatory is a facility of the National Science Foundation operated by Associated Universities, Inc. This paper is partly based on observations carried out with the IRAM 30 m Telescope. IRAM is supported by INSU/CNRS (France), MPG (Germany) and IGN (Spain). IA acknowledges support by a Ramón y Cajal grant of the Ministerio de Economía, Industria y Competitividad (MINECO) of Spain. The research at the IAA–CSIC was supported in part by the MINECO through grants AYA2016–80889–P, AYA2013–40825–P, and AYA2010–14844, and by the regional government of Andalucía through grant P09–FQM–4784. St. Petersburg University team acknowledges support from Russian Science Foundation grant 17-12-01029. The Submillimeter Array is a joint project between the Smithsonian Astrophysical Observatory and the Academia Sinica Institute of Astronomy and Astrophysics and is funded by the Smithsonian Institution and the Academia Sinica.
Appendix A Additional information on MAGIC data reduction
|MJD||Observation Time||Significance||Significance FR||F(¿150GeV)|
|obs||LE: hadr¡0.28||FR: hadr¡0.16|
The MAGIC telescopes are supported by an extensive weather monitor program. Atmospheric transmission at different heights within the MAGIC field of view is obtained with the use of a LIDAR (for details on this see Fruck & Gaug, 2015). For data quality selection we consider the transmission measured at a height of 9 km, with representing a perfectly clear sky and a complete opacity. MAGIC can carry out observations also during partial moonlight, with the drawback of having a higher energy threshold and larger systematic errors due to a higher contamination from the elevated night sky background (NSB), see Ahnen et al. (2017). The brightness of the NSB can be monitored from the average current in the camera (DC). S4 0954+65 was observed in a zenith range ranging from 35 to 50, for a total of 12.5 hours of data, of which 1 hour was lost due to bad weather. In the following we will refer to the different observation conditions of our data set as follows:
good dark data: data taken with dark sky (DC ¡ A) and good atmospheric condition (), used for detection and spectral reconstruction;
dark data needing atmospheric correction: data taken with dark sky (DC ¡ A) but under non optimal weather conditions (), used for detection and spectral reconstruction after atmospheric correction;
good low moon data: data taken with elevated NSB due to moonlight ( ¡ DC ¡ A) and good atmospheric condition (), used only for detection in this particular dataset;
good moon data: data taken during high NSB due to moonlight (DC¿A) and good atmospheric condition (), used only for detection.
The subsample of dataset selected with condition (1) (9.48 h of good quality data) has been analyzed with the standard MAGIC analysis chain (Zanin et al., 2013).
The subsample of dataset selected with condition (2) (0.32 h of data) follows the same analysis chain until the estimation of the energy for the events and evaluation of the flux. For this last step, the estimated energy and the effective area are corrected taking into account the enhanced atmospheric absorption (for validation of the procedure see Fruck & Gaug, 2015).
The subsample of dataset selected with condition (3) is applicable only at the day of 14th February, with the first VHE detection. The detection can be claimed from dark data alone (i.e. selected with condition 1), but an extra 0.81 h of data were taken under low moonlight. Those data are presented here for completeness, but are not used for spectral reconstruction so not to increase the systematic error and energy threshold.
The subsample of dataset selected with condition (4) (0.94 h of data) requires a special analysis that takes care of the effect of moonlight on data taking, reconstruction and analysis. Details of the procedure can be found in Ahnen et al. (2017).
The detailed breakdown of significances and estimated VHE fluxes is given in Table 3. Numbers are presented for the so-called low energy (LE) and full range (FR) cuts. The LE cuts are optimized for an energy range of GeV and are particularly appropriate for steep spectrum sources, while FR cuts are optimized for an energy range of GeV. The cuts are applied on 2 parameters: the ”size“ parameter, integrated charge (in photoelectrons) in the cleaned shower image; the ”hadronness“ parameter, computed from the gamma-hadron separation Random Forest (RF), with a value ranging from 0 for the most gamma-like images to 1 for the most hadron-like images. Indeed the standard MAGIC analysis chain relies on RF techniques to discriminate among gamma and hadronic shower and to better reconstruct the event directions. Lookup tables are used for energy estimation. This is achieved starting from a parametrization of the shower images in the detector. The significance of signal is then calculated with Eq. 17 from Li & Ma (1983) and using 5 regions of equal size and distance to the center of camera as the signal region for background estimation. Fluxes are calculated above an energy threshold of 150 GeV, which corresponds to the peak of the differential energy distribution of the excess events as a function of estimated event energy. The high energy threshold is due to the high zenith angle of the observation. Please note that for data of condition 4, strong moon, we apply an additional minimum cut in the ”size“ parameter (”size“ 150 phe) of the reconstructed Cherenkov image as prescribed by the moonlight-adapted analysis. This increases the energy threshold to a value of 250 GeV. In case of non-detection, we provide 95% confidence level upper limits to the flux, calculated following Rolke et al. (2005), considering a systematic error on flux estimation of 30% (Aleksić et al., 2016).
Appendix B Additional VLBA derived parameters
|23 Sep 2014|
|15 Nov 2014|
|5 Dec 2014|
|29 Dec 2014|
|14 Feb 2015|
|11 Apr 2015|
|11 May 2015|
|9 Jun 2015|
|2 Jul 2015|
|56924||5.22 0.77||5.25 4.23|
Appendix C Swift-XRT full dataset
Table 6 collects all the analyzed exposures for the Swift-XRT dataset described in Section 3. Fluxes have been extracted from a 20 pixel circular aperture. A different aperture was used on 2015 February 17 (MJD 57070.76), due to pile-up effects.
|DATE-TIME||MJD||EXP||F(2-10 keV)||F(0.3-10 keV)||INDEX||DOF||OBSID|
|[s]||[erg cm s]||[erg cm s]|
- Model available at https://fermi.gsfc.nasa.gov/ssc/data/access/lat/BackgroundModels.html.
- intermediate-synchrotron-peaked blazar (ISP) are defined with rest-frame synchrotron peak frequencies of
- Abdo, A. A., Ackermann, M., Ajello, M. et al. 2010, 188, 405
- Acero, F., Ackermann, M., Ajello, M. et al. 2015, ApJS, 218, 23A
- Ackermann, M.; Ajello, M.; Allafort, A. et al. 2013, ApJS, 209, 34
- Ackermann, M.; Ajello, M.; Atwood, W. B. et al. 2015, ApJ, 810, 14A
- Ackermann, M.; Ajello, M.; Atwood, W. B. et al. 2016, ApJS, 222, 5
- Agudo, I., Thum, C., Wiesemeyer, H. & Krichbaum, T. P., 2010, ApJS, 189, 1A
- Agudo, I., Thum, C., GÃ³mez, J. L. & Wiesemeyer, H., 2014, A&A, 566A, 59A
- Agudo, I., Thum, C., Molina, S. N. et al., 2018, MNRAS, 474, 1427
- Agudo, I., Thum, C.; Ramakrishnan, V. et al., 2018, MNRAS, 473, 1850A
- Ahnen, M. L., Ansoldi, S., Antonelli, A. et al. 2017, APhys, 94, 29
- Ahnen, M. L., Ansoldi, S., Antonelli, A. et al. 2017, A&A, 603, 29
- Ajello, M., Atwood, W. B., Baldini, L. et al. 2017, ApJS, 232, 18
- Albert, J., Aliu, E., Anderhub, H. et al. 2007, NIMPA, 583, 494A
- Aleksić, J., Ansoldi, S., Antonelli, L. A. et al., 2016, APh, 72, 76A
- Aleksić, J., Ansoldi, S., Antonelli, L. A. et al. 2014, A&A, 569A, 46A
- Angelakis, E.; Hovatta, T.; Blinov, D. et al. 2016, MNRAS, 463, 3365A
- Baars J. W. M., Genzel, R., Pauliny-Toth, I. I. K. & Witzel, A., 1977, A&A, 61, 99
- Bachev, R., 2015, The Astronomer’s Telegram #7083
- Bertin, E., & Arnouts, S. 1996, A&AS, 117, 393
- Blinov, D., Pavlidou, V., Papadakis, I. et al. 2015, MNRAS, 453, 1669B
- Blinov, D., Pavlidou, V.; Papadakis, I. et al. 2018, MNRAS, 474, 1296B
- Burrows, D. N., Hill, J. E., Nousek, J. A. et al. 2004, SPIE, 5165, 201B
- Dermer, C. D., Finke, J. D., Krug, H. & Böttcher, M., 2009, ApJ, 692, 32D
- Carrasco, L., Miramon, J., Porras, A. et al. 2015, The Astronomerâs Telegram #6996
- Domínguez, A., Primack, J. R., Rosario, D. J. et al., 2011, MNRAS, 410, 2556
- Fallah Ramazani, V., Lindfors, E. & Nilsson, K. 2017, A&A 608, 68
- Fan Z-H & Cao X 2004, ApJ, 602, 103
- Finke, J. D., Dermer, C. D., Böttcher, M., 2008ApJ, 686, 181F
- Finke, J. D., Razzaque, S., Dermer, C. D., 2010, ApJ, 712, 238F
- Finke, J. D., 2013, ApJ, 763, 134
- Fruck, C., & Gaug, M. 2015, European Physical Journal Web of Conferences, 89, 02003
- Ghisellini, G., Tavecchio, F., Foschini, L. & Ghirlanda, G. 2011, MNRAS, 414, 2674
- Gurwell, M. A., Peck, A. B., Hostler, S. R., Darrah, M. R., & Katz, C. A. (2007), in From Z-Machines to ALMA: (Sub)Millimeter Spectroscopy of Galaxies, Astronomical Society of the Pacific Conference Series, 375, 234.
- Healey, S. E., Romani, R. W., Taylor, G. B. et al. 2007, ApJS, 171, 61H
- Hervet, O., Boisson, C. & Sol, H. 2016, A&A, 592A, 22H
- Impey, C. D., Lawrence, C. R. & Tapia, S. 1991 ApJ, 375, 46I
- Jorstad, S. G., Marscher, A. P., Lister, M. L. et al. 2005, AJ, 130, 1418
- Jorstad, S. G., Marscher, A. P., Larionov, V. M. et al. 2010, ApJ, 715, 362
- Jorstad, S. G.; Marscher, A. P.; Morozova, D. A. et al. 2017, ApJ, 846, 98
- Kalberla, P. M. W.; Burton, W. B.; Hartmann, Dap et al. 2005, A&A, 440, 775K
- Kiehlmann, S.; Savolainen, T.; Jorstad, S. G. et al. 2016, A&A, 590A, 10K
- Kiehlmann, S.; Blinov, D.; Pearson, T. J.; Liodakis, I., 2017arXiv170806777K
- Krauss, F. for the Fermi-LAT Collaboration, 2014, The Astronomer’s Telegram #6709
- Landoni, M., Falomo, R., Treves, A., Scarpa, R. & Reverte PayÃ¡, D. 2015, AJ, 150, 181L
- Larionov, V. M., Jorstad, S. G., Marscher, A. P., et al. 2008, A&A, 492, 389
- Lawrence, C. R., Pearson, T. J., Readhead, A. C. S. & Unwin, S. C. 1986, AJ, 91, 494L
- Lawrence, C. R., Zucker, J. R.; Readhead, A. C. S. et al. 1996, ApJS, 107, 541L
- Li, T.-P.; Ma, Y.-Q., 1983, ApJ, 272, 317L
- Maraschi, L. 1992, ApJ, 397L, 5M
- Marscher, A. P., Jorstad, S. G., D’Arcangelo, F. D. et al. 2008, Nature, 452, 966
- Marscher, A. P.; Jorstad, S. G.; Larionov, V. M. et al. 2010 ApJ, 710L, 126M
- Mattox, J. R., Bertsch, D. L., Chiang, J. et al., 1996, ApJ, 461, 396
- Mazin, D. & Göbel, F. 2007, ApJ, 655L, 13M
- Mead, A. R. G., Ballard, K. R., Brand, P. W. J. L. et al. 1990, A&AS, 83, 183
- Mirzoyan, R. for the MAGIC Collaboration 2015, The Astronomerâs Telegram, #7080
- Morozova, D. A., Larionov, V. M., Troitsky, I. S. et al. 2014AJ, 148, 42M
- Mukherjee, R.; Aller, H. D.; Aller, M. F.; Bertsch, D. L. et al. 1995, ApJ, 445, 189M
- Nolan, P. L., Abdo, A. A., Ackermann, M. et al. 2012, ApJS, 199, 31
- Ojha, R., Carpenter, B. & Tanaka, Y. for the Fermi-LAT Collaboration, 2015, Astronomer’s Telegram #7093
- Poole, T. S., Breeveld, A. A., Page, M. J., et al. 2008, MNRAS,383, 627
- Prandini, E., Bonnoli, G, Maraschi L. et al. 2010, MNRAS, 405L, 76P
- Raiteri, C. M., Villata, M., Tosti, G. et al. 1999, A&A, 352, 19
- Rees, M.J 1967, MNRAS, 137, 429R
- Richards, J. L., Max-Moerbeck, Walter, Pavlidou, V. et al., 2011, ApJS, 194, 29
- Rolke, W. A., López, A. M. and Conrad, J. 2005, NIMPR A, 551, 493
- Sambruna, Rita M., Maraschi, Laura & Urry, C. Megan 1996, ApJ, 463, 444S
- Schlafly, Edward F. & Finkbeiner, Douglas P. 2011, ApJ, 737, 103S
- Spiridonova, O. I., Vlasyuk, V. V., Moskvitin, A. S. et al., 2015, The Astronomer’s Telegram #7057
- Stanek, K. Z., Danilet, A. B., Holoien, T. W.-S. et al., 2015, The Astronomer’s Telegram #7001
- Stickel, M., Padovani, P., Urry, C. M., Fried, J. W. & Kuehr, H. 1991, ApJ, 374, 431S
- Stickel M., Fried J.W. & Kuhr H. 1993, A&AS 98, 393
- Tanaka, Y. T., Becerra Gonzalez, J., Itoh, R. et al 2016, PASJ, 68, 51T
- Tavecchio, F. AIP Conf.Proc. 1792 (2017) no.1, 020007
- Teraesranta, H., Tornikoski, M., Mujunen, A. et al. 1998, A&AS, 132, 305
- Thum, C., Agudo, I., Molina, S. N. et al. 2018, MNRAS, 473, 2506T
- Villata M., Raiteri, C. M., Ghisellini, G. et al. 1997, A&AS 121, 119
- Wagner, S., Sanchez-Pons, F., Quirrenbach, A. & Witzel, A. 1990, A&A, 235, 1W
- Wagner, S. J., Witzel, A., Krichbaum, T. P. et al, 1993, A&A, 271, 344W
- Zanin, R., Carmona, E., Sitarek, J. et al., Proc of 33rd ICRC, Rio de Janeiro, Brazil, Id. 773, 2013
- Zhang, H., Cheng, X. & Böttcher, M. 2014, ApJ, 789, 66Z