The radio delay of the exceptional 3C 454.3 outburst. Follow-up WEBT observations in 2005-2006

1 MB·English

by  M. Villata

Astronomy&Astrophysicsmanuscriptno.6958 c ESO20081
(cid:13)
February5,2008
LE
The radio delay of the exceptional 3C 454.3 outburst
⋆
Follow-up WEBT observations in 2005–2006
M.Villata1,C.M.Raiteri1,M.F.Aller2,U.Bach1,3,M.A.Ibrahimov4,Y.Y.Kovalev3,5,6,O.M.Kurtanidze7,8,9,
V.M.Larionov10,11,C.-U.Lee12,P.Leto13,A.La¨hteenma¨ki14,K.Nilsson15,T.Pursimo16,J.A.Ros17,N.Sumitomo18,
A.Volvach19,H.D.Aller2,A.Arai18,C.S.Buemi20,J.M.Coloma17,V.T.Doroshenko21,Yu.S.Efimov22,
7 L.Fuhrmann1,23,3,V.A.Hagen-Thorn10,11,M.Kamada18,M.Katsuura18,T.Konstantinova10,E.Kopatskaya10,
0 D.Kotaka18,Yu.A.Kovalev5,M.Kurosaki18,L.Lanteri1,L.Larionova10,M.G.Mingaliev24,S.Mizoguchi18,
0 K.Nakamura18,M.G.Nikolashvili7,S.Nishiyama18,K.Sadakane18,S.G.Sergeev22,L.A.Sigua7,A.Sillanpa¨a¨15,
2
R.L.Smart1,L.O.Takalo15,K.Tanaka18,M.Tornikoski14,C.Trigilio20,andG.Umana20

n
a
J 1 INAF,OsservatorioAstronomicodiTorino,Italy
e-mail:[email protected]
0
2 DepartmentofAstronomy,UniversityofMichigan,MI,USA
1
3 Max-Planck-Institutfu¨rRadioastronomie,Germany

4 UlughBegAstronomicalInstitute,AcademyofSciencesofUzbekistan,Uzbekistan
1
5 AstroSpaceCenterofLebedevPhysicalInstitute,Russia
v
6 NationalRadioAstronomyObservatory,GreenBank,WV,USA
9
7 AbastumaniAstrophysicalObservatory,Georgia
9
8 AstrophysikalischesInstitutPotsdam,Germany
2
9 LandessternwarteHeidelberg-Ko¨nigstuhl,Germany
1
10 AstronomicalInstitute,St.-PetersburgStateUniversity,Russia
0
11 IsaacNewtonInstituteofChile,St.-PetersburgBranch
7
12 KoreaAstronomyandSpaceScienceInstitute,SouthKorea
0
13 INAF,IstitutodiRadioastronomiaSezionediNoto,Italy
/
h 14 Metsa¨hoviRadioObservatory,HelsinkiUniversityofTechnology,Finland
p 15 TuorlaObservatory,Finland
- 16 NordicOpticalTelescope,RoquedelosMuchachosAstronomicalObservatory,TF,Spain
o 17 Agrupacio´ Astrono`micadeSabadell,Spain
tr 18 AstronomicalInstitute,OsakaKyoikuUniversity,Japan
s 19 RadioAstronomyLaboratoryofCrimeanAstrophysicalObservatory,Ukraine
a 20 INAF,OsservatorioAstrofisicodiCatania,Italy
:
v 21 MoscowStateUniversity,Russia
i 22 CrimeanAstrophysicalObservatory,Ukraine
X 23 DipartimentodiFisicaeOsservatorioAstronomico,Universita`diPerugia,Italy
r 24 SpecialAstrophysicalObservatory,Russia
a
ABSTRACT
Context.In spring 2005 the blazar 3C 454.3 was observed in an unprecedented bright state from the near-IR to the hard X-ray
frequencies.AmmoutburstpeakedinJune–July2005,anditwasfollowedbyafluxincreaseathighradiofrequencies.
Aims.InthispaperwereportonmultifrequencymonitoringbytheWEBTaimedatfollowingthefurtherevolutionoftheoutburstin
detail.Inparticular,weinvestigatetheexpectedcorrelationandtimedelaysbetweentheopticalandradioemissionsinordertoderive
informationonthevariabilitymechanismsandjetstructure.
Methods.A comparison among the light curves at different frequencies is performed by means of visual inspection and discrete
correlationfunction,andtheresultsareinterpretedwithasimplemodeltakingintoaccountDopplerfactorvariationsofgeometric
origin.
Results.Thehigh-frequencyradiolightcurvesshowahugeoutburststartingduringthedimmingphaseoftheopticaloneandlasting
morethan1year.Thefirstphaseischaracterizedbyaslowfluxincrease,whileinearly2006amajorflareisobserved.Thelower-
frequencyradiolightcurvesshowaprogressivelydelayedandfainterevent,whichdisappearsbelow8GHz.Wesuggestthattheradio
majorpeakisnotphysicallyconnectedwiththespring2005opticalone,butitisactuallycorrelatedwithaminoropticalflareobserved
inOctober–November2005.Thisinterpretationinvolvesbothanintrinsicandageometricmechanism.Theformerisrepresentedby
disturbancestravellingdowntheemittingjet,thelatterbeingduetothecurved-jetmotion,withtheconsequentdifferentialchanges
ofviewinganglesofthedifferentemittingregions.
Keywords.galaxies:active–galaxies:quasars:general–galaxies:quasars:individual:3C454.3–galaxies:jets
1. Introduction reachingatmost 14Jy,whilethesecondhumpismorepeaked,
∼
withamaximumof 20Jy.Thewholeeventlastedabitmore
In May 2005the quasar-typeblazar 3C 454.3(2251+158)was ∼
than400daysand,atfirstsight,itseemstomimicwelltheob-
observed in an unprecedented luminous state from near-IR to
servedslowriseandfastdropoftheopticaloutburst,withalag
hardX-rayfrequencies.Abigobservingeffortwasspenttofol-
of250–300days.
lowtheoutburst,involvingbothground-basedandspaceinstru-
The37GHzfluxdensitybehaviourissimilar,withthemain
ments (Fuhrmannetal., 2006; Giommietal., 2006; Pianetal.,
differencethatthefirsthumpissubstitutedbyaquasi-monotonic
2006;Villataetal.,2006).Inparticular,alargemultiwavelength
increase,which becomesflatter beforethe steep rise leadingto
campaignwas organizedby the Whole Earth Blazar Telescope
the maximum peak. This is even higher ( 22 Jy) than the 43
(WEBT)1, whose first results, up to September 2005, were re- ∼
GHzone.Thedatesofstart,peak,andendoftheeventseemto
ported and analysed by Villataetal. (2006). A huge mm out-
bepracticallyidenticaltothepreviousones.
burst followed the optical one, peaking in June–July 2005. In
Intheotherradiolightcurves,onecanseesimilarfeatures.
themeantimethehigh-frequency(43–37GHz)radiofluxstarted
However, going towards lower frequencies, the variability am-
to increase. VLBA observationsat 43 GHz during the summer
plitudebecomessmallerandsmaller,whilethestartingpointof
confirmed the brightening of the radio core and showed an in-
the outburst,and to a muchless extentalso the peak,are more
creasingpolarization.
andmoredelayed.Inthe11GHzlightcurve(notshowninthe
Follow-upradio-to-opticalobservationsby the WEBT con-
figure) the outburst is still clearly observed, with a maximum
tinued until the end of the optical observing season, while a
fluxdensityincreaseofmorethan20%,whileat8GHztheevent
new campaign including three pointings by the XMM-Newton
isreducedtoabarelyvisible 10%fluxdensityenhancement,
satellitestartedjustafteranditisstillongoing.Inthispaperwe ∼
whichbeginsabout200dayslaterthan thehighest-frequencies
presentpartofthesenewobservationswiththeaimofstudying
one.The 5 GHz lightcurve(notshown in the figure)is almost
thecorrelationbetweentheradioandopticalemissions.
completelyflat,orevenslightlydecreasing,inthecorresponding
period.
2. Observationsandanalysis
2.2.Cross-correlationbetweenradiolightcurves
2.1.Opticalandradiolightcurves
In the following, we analyse the time lags mentioned
In the top panel of Fig. 1 we show the R-band data taken by
above by means of the discrete correlation function (DCF;
the WEBT from June 2004 to January2006;greydots refer to
Edelson&Krolik 1988; Hufnagel&Bregman 1992). We first
data already published in Villataetal. (2006), while red dots
investigatethe delaysamong the starting pointsof the outburst
represent the new data collected for this paper. The rightmost
atthedifferentradiowavelengths.Todothis,werestrictthecal-
point at JD = 2453888.7(June 2, 2006) indicates the low op-
culation to the period before the rise to the maximum peak, to
tical level at which the source was found after the 2006 solar
avoiditsstrongimprint.
conjunction. These new optical data were taken at the Osaka
The DCF computed on this pre-maximum period (JD <
Kyoiku,Mt.Maidanak,Abastumani,Crimean,Torino,Sabadell,
2453750) indicates a delay of the 37 GHz starting point wit∼h
Roque de los Muchachos (KVA and NOT), and Mt. Lemmon
respect to the 43 GHz one of 5 days3. The DCF between the
Observatories. In the same panel the 43 GHz radio data from
37and22GHz lightcurvesshowstwo equivalentmaxima;the
June 2004 to the end of August 2006 are shown as black dots
firstandbroaderonesuggestsameantimelagof25days,while
together with a cubic spline interpolation through the 15-day
the secondone indicatesa 95 day delay.Finally, the 14.5 GHz
binned light curve (blue line). Analogously,the following four
outburstwouldstart160dayslaterthanthe37GHzone.
panels display data and corresponding splines at 37, 22, 14.5,
Now we consider the whole 815 day period shown in Fig.
and8GHz.ThepointsafterJD=2453644.5(endofSeptember
1;consequently,theDCFresultswillbedominatedbythepres-
2005) represent new data. They were acquired at the Crimean
enceofthepeaks.TheDCFappliedtothesplineinterpolations
(RT-22),Medicina,Metsa¨hovi,Noto,SAORAS(RATAN-600),
leadstothesameresultsaswhenusingtheoriginaldatapoints;
and UMRAO Radio Observatories. Measurements from the
hence, in Fig. 2 we show the less noisy cross-correlations be-
VLA/VLBAPolarizationCalibrationDatabase2arealsoused.
tween splines. As one can see, the radio emissions from 43 to
IntheR-bandflux-densitylightcurveonecanrecognizethe
14.5GHzareallwellcorrelated.Whilenoevidentlagispresent
rather slow rising phase of the outburst started around JD =
between the 43 and 37 GHz peaks (blue filled circles), the 22
2453250, followed by a period of missing data because of the
GHz peak is delayed by 10 days(green empty circles), and 40
2005solar conjuction.Afterthis, an unprecedentedlyhighflux
daysseparatethe14.5GHzeventfromthe37GHzone(redfilled
densitywasobserved,rapidlydecreasingtowardstheendofthe
circles).
outburst,whichcanbelocatedaroundJD = 2453620.Thetotal
The progressive delays of the outburst starting points can
outburstdurationwasthusabout1year.
be interpreted in terms of a disturbance travelling down the
The 43 GHz light curve shows a double-humped outburst
jetthroughmoreandmoretransparentemittingregions.Inthis
starting about 250 days after the optical one, and representing
view, the aboveresults suggestthatin the observer’sframe the
thebrightesteventthusfarrecordedforthisobjectathighradio
jetbecomestransparentquasi-simultaneouslyat43and37GHz.
frequencies(seethehistoricallightcurvesinVillataetal.2006).
Thefirsthump(JD 2453500–2453750)isbroaderandlower, 3 TimelagsaredeterminedbytakingthecentroidoftheDCFpeaks
∼
(seee.g.Peterson,2001;Raiterietal.,2003).Apreciseestimateofthe
Sendoffprintrequeststo:M.Villata
uncertaintyforthetimelagsisnoteasy,dependingonthemethodused
⋆ Forquestionsregardingtheavailabilityofthedatapresentedinthis and parameters adopted. However, through Monte Carlo simulations,
paper,pleasecontactMassimoVillata([email protected]). we found typical values around 5 days for almost all the timedelays
1 http://www.to.astro.it/blazars/webt/ reported in thispaper, but for the more uncertain lags of the outburst
seee.g.Villataetal.(2004a,b);Raiterietal.(2005,2006). starting points at 22 and 14.5 GHz, having uncertainties of 10–20
2 http://www.vla.nrao.edu/astro/calib/polar/ days. ∼
M.Villataetal.:Theradiodelayoftheexceptional3C454.3outburst 3
Fig.1.Opticalandradiolightcurvesof3C454.3fromJune2004totheendofAugust2006.Inthetopandbottompanelsthegrey
dotsrepresentR-bandfluxdensitiesfromVillataetal.(2006),whilereddotsdisplaynewdata.InthelastbutonepaneltheR-band
lightcurve(greyandreddots)hasbeenmodifiedby“rebeaming”andtimeshifting,asexplainedinthetext.Thefirstfivepanels
fromtopshowthe43,37,22,14.5,and8GHzlightcurvesalongwiththeir15-daybinnedcubicsplineinterpolations.The43and
14.5GHzsplinesarealsoreportedinthelasttwopanelsforacomparisonwiththeotherlightcurves.Inthebottompanelasketch
ofthepredicted1mm(230GHz)lightcurveisalsoplottedasadottedline.
Then,sometensofdaysareneededtoreachtheregionwhereit seemsthatsomethingoccurred,makingtheoutburststopquasi-
is barely transparentalso to the 22 GHz emission, which how- simultaneouslyatallradiofrequencies.
everismaximallyradiatedabout100daysafterthehighestfre-
quencies.Acoupleofmonthslateralsothe14.5GHzemission
2.3.Cross-correlationbetweenopticalandradiolightcurves
canescapethejet.
As noticed in Sect. 2.1 from a visual inspection of Fig. 1, the
However,inasimplemodel,onewouldexpectthatalsothe optical and 43 GHz light curves seem to nicely correlate with
peaks and subsequentdimming phases present comparablede- a 250–300day radio lag. Indeed,the cross-correlationanalysis
lays, while here theyare foundto be much shorter.Actually,it yieldsa 275day delay,as shown in Fig. 3, where the DCF be-
4 M.Villataetal.:Theradiodelayoftheexceptional3C454.3outburst
to exclude the optical outburst. Actually, it becomesthe domi-
nantone,evenif faint(DCF 0.3).Itisdueto the correlation
∼
between the minor optical flare peaking at JD 2453670(see
∼
toppanelofFig.1)withthemaximumpeakofthe43GHzlight
curve.Inthefollowingwethusconsiderthepossibilitythatthe
43GHzmaximumpeakisnotphysicallylinkedwiththemajor
opticaloutburstofspring2005,butwiththeminoropticalflare
ofOctober–November2005.
3. Discussionandconclusions
Villataetal. (2006) suggested that the unprecedentedluminos-
ity of the spring 2005opticaloutburstwas mainly dueto a de-
creaseoftheviewingangleoftheopticallyemittingjetregion,
implying an increase of the beaming factor. In the same paper
theauthorsarguedalsothatthejetshouldpresentsomebending,
becauseofthelackofcorrelationbetweenthehistoricaloptical
Fig.2.Discretecorrelationfunctions(DCFs)betweenradiolight
andradiolightcurves.Hence,theradiooutburstwhichfollowed
curves:43versus37GHz(bluefilledcircles),37versus22GHz
after 4–5 months (first hump of the 43 GHz light curve) most
(green empty circles), and 37 versus 14.5 GHz (red filled cir-
likelyoccurredinaslightlymisalignedouterregion,sothatthe
cles).
radio flux was not so enhanced as the optical one. At the time
oftheminoropticalflareobservedin October–November2005
thejetmusthaveturned,sincethebeamingoftheopticalradia-
tionhasdecreased.Ontheotherhand,theradioemittingregion
seems to have reached the minimumviewing angle at the time
ofthemaximumpeak.
This misalignment between the two regions makes the
optical-radiocross-correlationanalysisratherdifficult. We thus
investigate what happens when a correction for the misalign-
mentisattempted,bysimulatinganopticallightcurvesuffering
beamingconditionscomparablewiththeonesaffectingtheradio
emission130dayslater,whenthedisturbanceseventuallyreach
thehigh-frequencyradioemittingregion(seeSect.2.3).
The “true” optical flux densities F are transformed into
R
“rebeamed”, simulated ones, F , under the following assump-
R′
tions: i) the observed flux density is proportional to δ3, where
δ = [γ(1 βcosθ)] 1 is the Doppler (beaming) factor and γ
−
is the Lore−ntz factor5; ii) the minimum viewing angle θ of
min
theopticallyemittingregionisachievedatt = t forthe“true”
Fig.3.Discretecorrelationfunction(DCF)betweentheR-band 0
optical light curve, and at t = t in the case of the simulated
and43GHzlightcurves(bluefilledcircles)comparedwiththe 0′
one; iii) the viewing angle increases both forward and back-
DCFbetweenthe“rebeamed”R-bandfluxdensitiesandthe43
ward in time with the square root of the time elapsed since t ,
GHzlightcurve(redemptycircles;seetextfordetails). 0
θ(t)=θ + √t t /t ,andsimilarlyforthe“rebeamed”case.
min 0 s
| − |
InthelastbutonepanelofFig.1 we showthe“rebeamed”
tweenthe1-daybinnedR-bandlightcurveandthe43GHzspline opticallightcurve(greyandreddots) obtainedwith the model
isplottedasbluefilledcircles4. parameters set to: γ = 10, θmin = 5◦, t0 = 2453490, t0′ =
This long time delay is not expected, especially in view of 2453650,t =14days6.Moreover,thesimulatedlightcurvehas
s
the fact that the millimetric outburst peaked around mid 2005, beenshiftedintimeby130daysinordertomakethecomparison
i.e. shortly after the opticalone (Villataetal., 2006). Indeed,it withthe43GHzsplineeasier.Onecanseethatthegeneraltrend
is difficult to conceive why there should be such a little time ofthetwolightcurvesisquitesimilar,withthemaindifference
delay(2–3months)betweentheopticalandmmvariations,and thatthe 43GHz variationsappearto be smoother,as expected,
suchalargelag(about9months)betweentheopticalandradio sincetheradioemissionisthoughttocomefromalargerregion.
ones. The result of their cross-correlation is shown in Fig. 3 as red
In fact, Villataetal. (2006) claimed that the outburst had emptycircles,indicatingastrongcorrelationwitharadiodelay
fully propagated to the high radio frequencies by the end of of125days.
September2005,correspondingtothefirsthumpofthe43GHz In other words, the radio delays come from the concomi-
lightcurveandimplyingan optical-radiodelayof4–5months. tance of two mechanisms acting on different time scales. The
Indeed,intheaboveoptical-radioDCF(bluefilledcirclesinFig.
3) thereisa littlesignalatabout130days.Thissignalremains 5 We neglect here the minor Doppler effects on time intervals and
alsowhentheDCFiscalculatedonthedataafterJD=2453600 frequencies.
6 Whiletheresultinglightcurveisnotverysensitivetochangesinγ,
4 Wechosethe43GHzlightcurveinsteadofthebetter-sampled37 θ ,andt,itisobviouslystronglyaffectedbythevaluesoft andt ,
min s 0 0′
GHz one because this latter misses an important feature like the first whichhavebeenchosenclosetothemajor(spring)andminor(autumn)
humpoftheoutburstmentionedinSect.2.1. opticalevents,respectively.
M.Villataetal.:Theradiodelayoftheexceptional3C454.3outburst 5
firstmechanismistheperturbationtravellingalongtheinhomo- by the other ones. This train givesrise to the broadand modu-
geneousjet,whichcausesthe 125dayradiolag.Theotherone lated hump lasting about 250 days (JD 2453500–2453750).
∼ ∼
isthe“turning”ofthecurvedjet,producingadifferentialvaria- Thelower-frequencyemittingregionsarealsoprogressivelyper-
tionoftheviewingangleofthedifferentemittingregions.This turbed. A new perturbation crosses the now misaligned opti-
mechanismisresponsibleforthe275dayseparationbetweenthe cal region (flare around JD = 2453670),and producesa mild-
timesoftheopticalandradiomaximumbeaming.Inthisframe- intensity outburst when reaching the less misaligned mm re-
workthedifferencebetweenthelongdelaysoftheradiooutburst gion. Finally, it enters the high-frequency radio regions when
starting points and the shorter delays of the peaksdiscussed in theirviewingangleisminimum,anditisDoppler-enhancedinto
Sect.2.2canbeunderstoodintermsofthedifferentorientation an exceptionallybrightevent.All thishappensinthe VLBI ra-
oftheradioemittingregionatthetimesofthepassageofthetwo diocore,wherelowfrequenciesarestillwellabsorbed,andthe
perturbations,implyinga greater contractionof the time scales outburstradiospectrumisstronglyinverted.Onlyattheendof
whentheviewingangleissmaller. the outburst it gets softer, when the last disturbance leaves the
Afterthemaximumpeak,theradiofluxdropsrapidlyatall highest-frequencyradioregions.
the higher frequencies, and no flux enhancementis seen at the Optical-to-radio monitoring by the WEBT is continuing to
lower frequencies, as one would expect from the transit of the followthesourcepost-outburstphases.
disturbancethroughouterregions.Thereasonmaybethatthese
jetregionsarebentinsuchawaythattheradiationemittedthere Acknowledgements. We thank Alan Marscher for sharing useful information
concerning changes in the VLBA images of 3C 454.3, and Philip Hughes
isbeamedelsewhere.Moreover,fromthe lightcurvesinFig. 1
for useful discussion. This work is partly based on observations made with
one can see that the high-frequencyradio spectrum is strongly the Nordic Optical Telescope, operated on the island of La Palma jointly by
inverted during the outbursts, suggesting that these events oc- Denmark,Finland,Iceland,Norway,andSweden,intheSpanishObservatorio
curred inside the radio core, where low radio frequencies are delRoque delosMuchachos oftheInstituto deAstrof´ısica deCanarias. Itis
partlybasedalsoonobservationswiththeMedicinaandNototelescopesoper-
stillstronglyabsorbed.Inotherwords,aftermid2006thedistur-
atedbyINAF-IstitutodiRadioastronomia.WethankthestaffattheMedicina
banceswouldbetravellinginthemisalignedjetregion,i.e.out
andNotoradioobservatoriesfortheirhelpandsupportduringtheobservations.
ofthecore(seethediscussionbyBachetal.2006ontheVLBA This research has made use of data from the University of Michigan Radio
mapsof BL Lacertae).Eventuallytheymay becomeagain vis- AstronomyObservatory,whichissupportedbytheNationalScienceFoundation
ible when passing through outer regions having again a small andbyfundsfromtheUniversityofMichigan.Thisworkwaspartlysupported
by the Italian Space Agency (ASI)under contract ASI/INAF I/023/05/0. The
viewing angle, and consequently appear as new radio compo-
St. Petersburg team acknowledges support from Russian Federal Program for
nents.Indeed,VLBA radiomapsat43GHz takeninApriland BasicResearchundergrant05-02-17562.RATAN-600observationswerepartly
August 2006 do not show any new component yet, even if a supported by the Russian Foundation for Basic Research grant 05-02-17377.
slight elongation of the core in the jet direction is visible in ThisprojectwasdonewhileYYKwasaJanskyfellow oftheNational Radio
AstronomyObservatoryandaresearchfellowoftheAlexandervonHumboldt
the August map, i.e. towards the end of our observing period
Foundation.
(Marscheretal.,inpreparation).
Finally,theabovemodelcanleadtoapredictionforthebe-
haviour of the millimetric emission in the same period. A rea- References
sonablehypothesisisthatatthesewavelengthstheoutburstde-
Bach,U.,Villata,M.,Raiteri,C.M.,etal.2006,A&A,456,105
velopment is intermediate between those observed in the opti-
Edelson,R.A.&Krolik,J.H.1988,ApJ,333,646
cal and radio bands, with a first very strong mm outburst (as Fuhrmann,L.,Cucchiara,A.,Marchili,N.,etal.2006,A&A,445,L1
already reported by Villataetal. 2006), followed by a second, Giommi,P.,Blustin,A.J.,Capalbi,M.,etal.2006,A&A,456,911
lessprominentone.InthelastpanelofFig.1thedottedlinerep- Hufnagel,B.R.&Bregman,J.N.1992,ApJ,386,473
Peterson,B.M.2001,inAdvancedLecturesontheStarburst-AGNConnection,
resents a sketch of the predicted 1 mm (230 GHz) light curve.
ed.I.Aretxaga,D.Kunth,&R.Mu´jica(Singapore:WorldScientific),3
Ithasbeenobtainedasthe averagebetweenthe 43GHz spline
Pian,E.,Foschini,L.,Beckmann,V.,etal.2006,A&A,449,L21
and a curve representing the optical data, shifted by 125 days. Raiteri,C.M.,Villata,M.,Ibrahimov,M.A.,etal.2005,A&A,438,39
Thislatter curveis a cubic spline interpolationthroughthe 15- Raiteri,C.M.,Villata,M.,Kadler,M.,etal.2006,A&A,459,731
day binned R-bandflux densities, scaled as shown in the same Raiteri,C.M.,Villata,M.,Tosti,G.,etal.2003,A&A,402,151
Villata,M.,Raiteri,C.M.,Aller,H.D.,etal.2004a,A&A,424,497
panelto make them comparablewith the radioones. Then, the
Villata,M.,Raiteri,C.M.,Balonek,T.J.,etal.2006,A&A,453,817
so obtained 1 mm curve has been normalized to have a maxi- Villata,M.,Raiteri,C.M.,Kurtanidze,O.M.,etal.2004b,A&A,421,103
mum peak of about 40 Jy, and it has been shifted by 30 days
backintimeto makethepeakoccurinmid2005,tomatchthe
observationsquotedbyVillataetal.(2006).
ListofObjects
In summary, we are envisaging a scenario where the opti-
callyemittingregionofacurvedjetacquiresitsminimumview-
‘3C454.3’onpage1
ing angle in spring 2005(JD 2453490),so thatsome distur-
‘3C454.3’onpage1
∼
bance(s)travellinginthatregionproducesamaximallyDoppler- ‘3C454.3’onpage1
enhancedoutburst.Soonafteralsothemmregionisaffectedby
‘3C454.3’onpage2
similar intrinsic (perturbation) and geometric (minimum view-
ing angle) conditions, even if this latter condition is probably
reached with some delay. In the meanwhile some other pertur-
bations (minor flares up to JD 2453575 in the optical light
∼
curve) are crossing the optical region, which is becoming less
and less well aligned with the line of sight. Most likely, some
of these disturbances enter the mm region when it is still well
oriented,thus producinga mm outburstmore extendedin time
(probablymoreextendedthaninourtoy-modelprediction).The
firstdisturbanceisnowcrossingthe43–37GHzregion,followed