A&A 453, 817-822 (2006)
M. Villata1 - C. M. Raiteri1 - T. J. Balonek2 - M. F. Aller3 - S. G. Jorstad4 - O. M. Kurtanidze5,6,7 - F. Nicastro8,9,10 - K. Nilsson11 - H. D. Aller3 - A. Arai12 - A. Arkharov13 - U. Bach1 - E. Benítez9 - A. Berdyugin11 - C. S. Buemi14 - M. Böttcher15 - D. Carosati16 - R. Casas17 - A. Caulet9 - W. P. Chen18 - P.-S. Chiang18 - Y. Chou18 - S. Ciprini11,19 - J. M. Coloma17 - G. Di Rico20 - C. Díaz21 - N. V. Efimova13,22 - C. Forsyth2 - A. Frasca14 - L. Fuhrmann1,19 - B. Gadway2 - S. Gupta15 - V. A. Hagen-Thorn22,23 - J. Harvey15 - J. Heidt7 - H. Hernandez-Toledo9 - F. Hroch24 - C.-P. Hu18 - R. Hudec25 - M. A. Ibrahimov26 - A. Imada27 - M. Kamata12 - T. Kato27 - M. Katsuura12 - T. Konstantinova22 - E. Kopatskaya22 - D. Kotaka12 - Y. Y. Kovalev28,29 - Yu. A. Kovalev29 - T. P. Krichbaum30 - K. Kubota27 - M. Kurosaki12 - L. Lanteri1 - V. M. Larionov22,23 - L. Larionova22 - E. Laurikainen31 - C.-U. Lee32 - P. Leto33 - A. Lähteenmäki34 - O. López-Cruz35 - E. Marilli14 - A. P. Marscher4 - I. M. McHardy36 - S. Mondal18 - B. Mullan2 - N. Napoleone10 - M. G. Nikolashvili5 - J. M. Ohlert37 - S. Postnikov15 - T. Pursimo38 - M. Ragni20 - J. A. Ros17 - K. Sadakane12 - A. C. Sadun39 - T. Savolainen11 - E. A. Sergeeva40 - L. A. Sigua5 - A. Sillanpää11 - L. Sixtova24 - N. Sumitomo12 - L. O. Takalo11 - H. Teräsranta34 - M. Tornikoski34 - C. Trigilio14 - G. Umana14 - A. Volvach41 - B. Voss42 - S. Wortel2
1 - INAF, Osservatorio Astronomico di Torino, Italy
2 - Foggy Bottom Observatory, Colgate University, NY, USA
3 - Department of Astronomy, University of Michigan, MI, USA
4 - Institute for Astrophysical Research, Boston University, MA, USA
5 - Abastumani Astrophysical Observatory, Georgia
6 - Astrophysikalisches Institut Potsdam, Germany
7 - Landessternwarte Heidelberg-Königstuhl, Germany
8 - Harvard-Smithsonian Center for Astrophysics, MA, USA
9 - Instituto de Astronomía, UNAM, Mexico
10 - INAF, Osservatorio Astronomico di Roma, Italy
11 - Tuorla Observatory, Finland
12 - Astronomical Institute, Osaka Kyoiku University, Japan
13 - Main (Pulkovo) Astronomical Observatory of the Russian Academy of Sciences, Russia
14 - INAF, Osservatorio Astrofisico di Catania, Italy
15 - Astrophysical Institute, Department of Physics and Astronomy, Ohio University, OH, USA
16 - Armenzano Astronomical Observatory, Italy
17 - Agrupació Astronòmica de Sabadell, Spain
18 - Institute of Astronomy, National Central University, Taiwan
19 - Dipartimento di Fisica e Osservatorio Astronomico, Università di Perugia, Italy
20 - INAF, Osservatorio Astronomico di Teramo, Italy
21 - Departamento de Astrofísica, Universidad Complutense, Spain
22 - Astronomical Institute, St.-Petersburg State University, Russia
23 - Isaac Newton Institute of Chile, St.-Petersburg Branch, Russia
24 - Institute of Theoretical Physics and Astrophysics, Masaryk University, Czech Republic
25 - Astronomical Institute, Academy of Sciences of the Czech Republic, Czech Republic
26 - Ulugh Beg Astronomical Institute, Academy of Sciences of Uzbekistan, Uzbekistan
27 - Department of Astronomy, Kyoto University, Japan
28 - Jansky fellow, National Radio Astronomy Observatory, WV, USA
29 - Astro Space Center of Lebedev Physical Institute, Russia
30 - Max-Planck-Institut für Radioastronomie, Germany
31 - Division of Astronomy, University of Oulu, Finland
32 - Korea Astronomy and Space Science Institute, South Korea
33 - INAF, Istituto di Radioastronomia Sezione di Noto, Italy
34 - Metsähovi Radio Observatory, Helsinki University of Technology, Finland
35 - Instituto Nacional de Astrofísica, Optica y Electrónica (INAOE), Mexico
36 - School of Physics and Astronomy, The University, UK
37 - Michael Adrian Observatory, Germany
38 - Nordic Optical Telescope, Roque de los Muchachos Astronomical Observatory, TF, Spain
39 - Department of Physics, University of Colorado at Denver and Health Sciences Center, CO, USA
40 - Crimean Astrophysical Observatory, Ukraine
41 - Radio Astronomy Laboratory of Crimean Astrophysical Observatory, Ukraine
42 - Institut für Theoretische Physik und Astrophysik der Universität Kiel, Germany
Received 31 January 2006 / Accepted 2 March 2006
Context. The radio quasar 3C 454.3 underwent an exceptional optical outburst lasting more than 1 year and culminating in spring 2005. The maximum brightness detected was R=12.0, which represents the most luminous quasar state thus far observed ( ).
Aims. In order to follow the emission behaviour of the source in detail, a large multiwavelength campaign was organized by the Whole Earth Blazar Telescope (WEBT).
Methods. Continuous optical, near-IR and radio monitoring was performed in several bands. ToO pointings by the Chandra and INTEGRAL satellites provided additional information at high energies in May 2005.
Results. The historical radio and optical light curves show different behaviours. Until about 2001.0 only moderate variability was present in the optical regime, while prominent and long-lasting radio outbursts were visible at the various radio frequencies, with higher-frequency variations preceding the lower-frequency ones. After that date, the optical activity increased and the radio flux is less variable. This suggests that the optical and radio emissions come from two separate and misaligned jet regions, with the inner optical one acquiring a smaller viewing angle during the 2004-2005 outburst. Moreover, the colour-index behaviour (generally redder-when-brighter) during the outburst suggests the presence of a luminous accretion disc. A huge mm outburst followed the optical one, peaking in June-July 2005. The high-frequency (37-43 GHz) radio flux started to increase in early 2005 and reached a maximum at the end of our observing period (end of September 2005). VLBA observations at 43 GHz during the summer confirm the brightening of the radio core and show an increasing polarization. An exceptionally bright X-ray state was detected in May 2005, corresponding to the rising mm flux and suggesting an inverse-Compton nature of the hard X-ray spectrum.
Conclusions. A further multifrequency monitoring effort is needed to follow the next phases of this unprecedented event.
Key words: galaxies: active - galaxies: quasars: general - galaxies: quasars: individual: 3C 454.3 - galaxies: jets
|Figure 1: Optical ( top) and radio ( bottom) light curves from 1966 to the end of September 2005; the yellow strip indicates the period of the 2004-2005 WEBT campaign, while grey vertical lines show the times of the satellite pointings (see Fig. 4). Bottom panel: black points indicate the 8 GHz light curve (1489 data points); red, green, and blue curves represent the cubic spline interpolations through the 30-day binned light curves at 5, 14.5, and 37 GHz, respectively; blue points are sparse 37 GHz data. Top panel: besides the R-band light curve (2709 data points, 1502 of which are from the WEBT campaign, blue points), the radio hardness ratios H37/8 (blue line) and H14.5/8 (green line) are plotted as 18-H; the light blue shading highlights where H37/8 is harder than average.|
|Open with DEXTER|
In addition to these two components, quasars can also show a "big blue bump'' in the UV band, which is interpreted as the signature of the thermal disc feeding the supermassive black hole. This component is mostly visible when the synchrotron emission is faint (see e.g. Pian et al. 1999; Hagen-Thorn & Yakovleva 1994).
One of the main features of blazars is their strong emission variability at all wavelengths, from the radio band to -ray energies, on time scales ranging from a few minutes to several years.
The optical light curve shows two very different phases. Until 2001.0 only moderate variability is observed in the range (an exception being the higher activity shown in 1940-1955, when R reached 14; see Angione 1968). From 2001 the amplitude of the variations starts to increase.
On May 9.36, 2005 (at the start of the observing season) the source was observed at R=12.0,
thus triggering a multifrequency campaign by the Whole Earth Blazar Telescope
In order to have information on both the rising and decreasing phases of the outburst,
data were collected from June 2004 to the end of September 2005.
In total, 5584
observations from 18 telescopes
were performed in this period;
data were taken at Campo Imperatore, Calar Alto, and Roque de los Muchachos (NOT).
Radio data from 1 to 43 GHz were acquired at several telescopes, mentioned above.
In the top panel of Fig. 1 the R-band light curve from the WEBT campaign
is displayed as blue dots (see also Fig. 2): it is composed of 1502 points. In particular,
1146 data points were taken in 143.59 days,
from May 9.36 to September 29.95, 2005, during the outburst decreasing phase,
with a mean time separation of 3.0 h, and
only 4 gaps longer than 36 h (2-4 days).
Previous data are shown as red points; they come from the literature (including the recent
paper by Fuhrmann et al. 2006) as well as from some WEBT observatories.
Here we present some results of this campaign;
a more detailed study is deferred to a forthcoming paper.
|Figure 2: Time evolution of the B-R colour index ( top) during the WEBT campaign compared with the R-band light curve ( bottom); orange points in the light curve highlight the R data used to derive the B-R indices; the red curve is a cubic spline interpolation through the 43 GHz light curve, arbitrarily scaled to fit the figure; the light blue strips indicate the epochs of the INTEGRAL ( left) and Chandra ( right) pointings, while the light green and red ones refer to spectra shown in Fig. 4; the red triangles mark the VLBA epochs of Fig. 3.|
|Open with DEXTER|
By looking at Fig. 1 no evident correlation appears between the optical and radio events. In particular, a low radio state characterizes the 2004-2005 optical outburst as well as the previous optical activity, while the past strong radio outbursts do not show any optical counterpart. Even the hardness ratio between the 37 GHz and 8 GHz splines ( H37/8=F37/F8, plotted in the top panel as 18-H37/8, blue line) does not show correlation with the optical, as is found by Villata et al. (2004a) for BL Lac. Only during the WEBT campaign can one see a relation: a fast rise of H37/8 to its maximum value (1.43), delayed by several months with respect to the rise of the optical outburst. This growth of H37/8 is essentially due to the increase of the 37 GHz flux visible in the bottom panel. The 43 GHz light curve, not plotted there (but in Fig. 2, see below), confirms this trend. In the top panel H14.5/8 is also plotted (green line) for comparison. In general, its trend mimics fairly well that of H37/8, with some delay and reduced variability amplitude. In particular, just a slight increase can be seen in 2005. The 22 GHz flux and the corresponding H22/8 display an intermediate trend, as expected.
The general lack of correlation between the radio and optical behaviour suggests that the optically emitting region is not transparent to the radio frequencies, so that these must come from another, outer part of the jet, possibly misaligned with respect to the former. It seems that before 2001 the radio part was more aligned with the line of sight and the radio variability was thus enhanced by Doppler effects. On the contrary, in the last 5 years the optical radiation would dominate the scenario due to a smaller viewing angle of the corresponding emitting region. Thus the flux increase we are seeing at the high radio frequencies should be produced close to the optical region. The outburst will probably propagate further out, towards lower-frequency emitting regions. However, we cannot foresee whether it will be visible, since this depends on the jet curvature.
A confirmation of the above general picture comes from the source behaviour in the mm bands, where a strong outburst was observed to peak in June-July 2005, reaching 45 Jy at 1 mm and 26 Jy at 3 mm (data from the IRAM 30-m telescope), while previous observations in 1985-2004 did not show any comparable activity, the flux density ranging between 2 and 15 Jy in both bands (data from Tornikoski et al. 1996; Teräsranta et al. 2004,1992,1998; Steppe et al. 1993; Reuter et al. 1997, and unpublished data from IRAM and SEST).
The time evolution of the B-R colour index during the optical outburst is shown in Fig. 2 (top panel), compared with the R-band light curve (bottom panel). Orange dots in the latter panel highlight the R-band points used to derive the B-R colour indices. In general, a "redder when brighter'' trend can be recognized. Since 3C 454.3 is a quasar, this can be interpreted as due to the contribution given by the thermal emission from the accretion disc, which mainly affects the bluer region of the optical spectrum when the jet emission is faint. In other words, there would be two optical components: a variable one from the jet and the other from the disc, not necessarily variable. When the jet is brighter, its redder spectrum dominates, and vice versa. However, there seems to be a kind of "saturation'' effect in the B-R trend: in the brightest part of the outburst () the trend is not visible any longer. Most likely, this is due to the dominance of the non-thermal jet emission, which would display its usual "bluer when brighter'' behaviour, thus balancing the opposite trend.
|Figure 3: VLBA maps at 43 GHz; the total intensity is shown by contours from 0.15% up to 76.8% of the August 2005 peak value, increasing by a factor of 2; the straight lines show the electric vector and their length is proportional to the local polarized intensity.|
|Open with DEXTER|
In the bottom panel of Fig. 2 the 43 GHz flux-density spline is also shown (arbitrarily scaled to fit the figure, the range is 4-13 Jy) for comparison with the optical light curve. It peaks in September 2005, i.e. 2-3 months after the mm outburst, which seems to have a similar delay with respect to the optical event.
Figure 3 shows three recent VLBA maps at 43 GHz compared with an older map dated April 2001. The plot is constructed with both total and polarized intensity, normalized to the peak values achieved at the epoch of August 2005 ( Jy/beam, Jy/beam). The total intensity is shown by contours from 0.15% up to 76.8% of the August 2005 peak value, increasing by a factor of 2. The straight lines show the electric vector and their length is proportional to the local polarized intensity.
There is a significant increase in both total and polarized intensity of the VLBI core compared with the values in April 2001 (see also Table 1), thus confirming the expectation that the 43 GHz rising flux in 2005 seen in Fig. 2 comes from the VLBI core. The three VLBA epochs are indicated in Fig. 2 as red triangles: there is a strict agreement between the trend of the 43 GHz light curve and the total intensity values reported in Table 1. Since the VLBI core of 3C 454.3 at 43 GHz used to have very low polarization compared to the western feature at 0.7 mas from the core, a strong increase of the core polarization along with the total flux suggests a new component emerging from the core. However, as discussed in the previous section, the emerging feature will be more or less visible depending on the jet bending.
|Figure 4: Spectral energy distribution of 3C 454.3 showing contemporaneous radio, near-IR, optical, and X-ray (Chandra and INTEGRAL) data during May 15-20, 2005. Previous data are also plotted for comparison, together with two other spectra from the WEBT campaign (see text for further details).|
|Open with DEXTER|
Table 1: Results of VLBA observations of 3C 454.3 at 43 GHz.
3C 454.3 was observed by Chandra on May 19.7-21.0, 2005 with the HRC-LETG instrumental configuration. The Chandra pointing was performed as part of a ToO program on blazars in outburst, and lasted 112 ks. We used version 3.3 of the CIAO software to reduce and analyse the data. We extracted source and background spectra from the standard HRC-LETG bow-tie regions, and then grouped the source spectrum to contain at least 20 counts per channel. We used the fitting package Sherpa to fit the background-subtracted source spectrum in the range 0.2-8 keV. Here we report the results of the continuum spectral analysis, while the study of narrow absorption features due to possible intervening Warm-Hot Intergalactic Medium filaments is deferred to a forthcoming paper (Nicastro et al., in preparation).
The 0.2-8 keV spectrum is well described ( ) by a rather flat power law, with photon index , absorbed by a large column density , exceeding by more than a factor of 2 the Galactic value. We measured de-absorbed fluxes of erg and erg .
In Fig. 4 we show the broad-band SED of the source. The small black crosses in the radio-optical range indicate archival data taken from NED. In the X-ray band we plotted data from old observations by ROSAT in November 1991 (Sambruna 1997) and in May 1992 (Prieto 1996), as well as data from a BeppoSAX observation in June 2000, and data from a pile-up affected Chandra observation in November 2002 (Marshall et al. 2005).
During the brightest phases of the optical outburst, in May 2005, there were other observations by high-energy satellites (RXTE, Swift, INTEGRAL) besides the Chandra one presented in the previous section. The ToO INTEGRAL observation was performed on May 15.8-18.4. In Fig. 4 we show the corresponding 3-200 keV spectrum (Pian et al. 2006). The Chandra spectrum of May 19.7-21.0 is also plotted, together with radio, near-IR and optical data from the WEBT in the period of the two pointings (blue symbols). Since in this period the optical brightness varied by , we plot two spectra (each composed of simultaneous data) corresponding to a high and a low state. No significant variation in the spectral slope is seen between the two states, despite their different brightness levels, in agreement with the rather stable colour index found in this period (see Fig. 2, light blue strips). The contemporaneous radio data follow a power law from 1 to 22 GHz, which then breaks due to the 37 and 43 GHz fluxes that begin to rise (see Figs. 1 and 2). The grey vertical strips in Fig. 4 show the historical variation range of the radio, mm and R-band data collected for this paper (and partly displayed in Figs. 1 and 2).
One month after the May satellite pointings, around June 21.0, the reddest observed optical state is achieved (light green line in Fig. 2). The corresponding near-IR-optical spectrum is displayed in Fig. 4 with green symbols: it is indeed very steep, crossing the lower May spectrum around the H band, and most likely connecting with the highest parts of the mm strips, since we should be close to the mm outburst peak. The 37-43 GHz flux has also increased (see Figs. 1 and 2), while no appreciable variation is seen at lower frequencies (where different-colour symbols often overlap in the figure, so that only the blue circles are visible).
In late September, at the foot of the outburst, the bluest optical state of 2005 is found (light red strip in Fig. 2). The corresponding (red) spectrum in Fig. 4 is the average over 5 days (September 23.0-28.0) of intensive monitoring: the thermal component seems to strongly affect the spectrum. At this time the 22 GHz flux starts to increase and the radio spectral break shifts downwards, while a peak of the outburst is seen at 37-43 GHz (see also Figs. 1 and 2).
The main feature of the plotted SEDs is the strong variability of the optical-IR, mm and X-ray fluxes. In Fig. 1 vertical grey lines indicate the times of the X-ray pointings. The X-ray flux seems to correlate neither with radio flux (always rather low), nor with the optical level (in 2002 it was brighter than during the previous pointings, but the X-ray flux was lower), nor with the radio hardness (higher during the ROSAT pointings and intermediate during the others). It seems instead to correlate with the mm flux, since a strong mm outburst was close to its peak during the May 2005 X-ray observations. This is expected due to the probable inverse-Compton nature of the hard X-ray spectrum.
Thus, this unprecedented quasar optical outburst preceded an equally unprecedented mm outburst (accompanied by an exceptionally bright X-ray state) by 2-3 months, and, 2-3 months later, it had fully propagated to the high radio frequencies. This occurred in the VLBI radio core. We cannot foresee whether and when further propagation will be visible at lower frequencies and outside the radio core, due to probable jet bending. Further multifrequency monitoring is needed to follow the next phases of this exceptional event.
The mm data from the IRAM 30-m telescope were provided by H. Ungerechts based on preliminary results from regular flux monitoring observations by the IRAM Granada staff. This work is partly based on observations made with the Nordic Optical Telescope, operated on the island of La Palma jointly by Denmark, Finland, Iceland, Norway, and Sweden, in the Spanish Observatorio del Roque de los Muchachos of the Instituto de Astrofísica de Canarias. It is partly based also on observations collected at the German-Spanish Astronomical Center (DSAZ), Calar Alto, operated by the Max-Planck-Institut für Astronomie Heidelberg jointly with the Spanish National Commission for Astronomy. We thank Calar Alto for allocation director's discretionary time to this programme. This research has made use of data from the University of Michigan Radio Astronomy Observatory, which is supported by the National Science Foundation and by funds from the University of Michigan, and 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. This work was partly supported by the European Community's Human Potential Programme under contract HPRN-CT-2002-00321 (ENIGMA). The St. Petersburg team acknowledges support from Russian Federal Program for Basic Research under grant 05-02-17562. RATAN-600 observations were partly supported by the Russian Foundation for Basic Research grant 05-02-17377. C. Díaz acknowledges support from the Spanish Programa Nacional de Astronomía y Astrofísica under grant AYA2003-1676.