Blazars as beamlights to probe the Extragalactic Background Light in the Fermi and Cherenkov telescopes era

0.15 MB·English

by  Massimo Persic

2009 Fermi Symposium, Washington, D.C., Nov. 2-5 1
Blazars as beamlights to probe the Extragalactic Background Light
in the Fermi and Cherenkov telescopes era
M.Persic
INAF-Trieste and INFN-Trieste, via G.B.Tiepolo 11, I-34143 Trieste TS, Italy
N.Mankuzhiyil
Udine University and INFN-Trieste, via delle Scienze 208, I-33100 Udine UD, Italy
F.Tavecchio
INAF-Brera, via E.Bianchi 46, I-23807 Merate LC, Italy
0
1 The Extragalactic Background Light (EBL) is the integrated light from all the stars that have
0
everformed,andspanstheIR-UVrange. Theinteractionofvery-high-energy(VHE:E >100GeV)
2
γ-rays, emitted by sources located at cosmological distances, with the intervening EBL results in
n e−e+ pair production that leads to energy-dependent attenuation of the observed VHE flux. This
a introduces a fundamental ambiguity in the interpretation of the measured VHE blazar spectra:
J neither the intrinsic spectra, nor the EBL, are separately known – only their combination is. In

4 this paper we propose a method to measure the EBL photon number density. It relies on using
simultaneous observations of blazars in the optical, X-ray, high-energy (HE: E > 100MeV) γ-ray

] (from the Fermi telescope), and VHEγ-ray (from Cherenkov telescopes) bands. For each source,
O
themethodinvolvesbest-fittingthespectralenergydistribution(SED)fromopticalthroughHEγ-
C rays (the latter being largely unaffected by EBL attenuation as long as z∼< 1) with a Synchrotron
Self-Compton (SSC) model. We extrapolate such best-fitting models into the VHE regime, and
.
h assumetheyrepresenttheblazars’intrinsicemission. Contrastingmeasuredversusintrinsicemission
p leads to a determination of the γ-γ opacity to VHE photons – hence, upon assuming a specific
-
cosmology,wederivetheEBLphotonnumberdensity. Using,foreachgivensource,differentstates
o
of emission will only improve the accuracy of the proposed method. We demonstrate this method
r
t usingrecentsimultaneousmulti-frequencyobservationsoftheblazarPKS2155-304anddiscusshow
s
a similar observations can more accurately probe the EBL.
[

2 I. INTRODUCTION cal kind, deduces upper limits on the level of EBL at-
v
tenuation making basic assumptions on the intrinsic
5
VHEγ-ray shape of AGN spectra: assuming, specif-
9 The Extragalactic Background Light (EBL), in
1 both its level and degree of cosmic evolution, reflects ically, that the VHE photon index must be Γ ≥ 1.5;
3 e.g., ([3, 16, 18]); but see ([26]), or that the same-
the time integrated history of light production and
. slope extrapolation of the observed Fermi/LAT HE
2 re-processing in the Universe, hence the history of
spectrum into the VHE domain exceeds the intrin-
1 cosmological star-formation. Roughly speaking, its
9 sic VHE spectrum there ([9]). The only unquestion-
shape must reflect the two humps that characterize
0 able constraints on the EBL are model-independent
the spectral energy distributions (SEDs) of galaxies:
: lower limits based on galaxy counts ([6, 8]). It should
v one arising from starlight and peaking at λ 1µm
i (opticalbackground),andonearisingfromwar∼mdust be noted, however, that the EBL upper limits in the
X 2–80µm obtained by [18] combining results from all
emission and peaking at λ 100µm (infrared back-
r ∼ known TeV blazar spectra (based on the assumption
a ground).
that the intrinsic Γ 1.5) are only a factor 2–2.5
Direct measurements of the EBL are hampered by ≥ ≈
above the absolute lower limits from source counts.
thedominanceofforegroundemission(interplanetary
So it would appear that there is little room for ad-
dust and Galactic emission), hence the level of EBL
ditional components like PopIII stars, unless we miss
emission is uncertain by a factor of several.
some fundamental aspects of blazar emission theory
One approach has been modeling the EBL aris- (which we never observed in local sources, however).
ing from an evolving population of galactic stellar AnattempttomeasuretheEBLusedtherelatively
populations: however, uncertainties in the assumed faraway blazar 3C279 as a background light source
galaxy formation and evolution scenarios, stellar ini- ([24]), assumed that the intrinsic VHE spectrum was
tial mass function, and star formation rate have known from modeling and extrapolating the (histor-
led to significant discrepancy among models (e.g., ical) average broad-band data. However, blazars are
[13, 14, 21, 23, 25]). These models have been used highly variable sources, so it’s almost impossible to
to correct observed VHE spectra and deduce (EBL determinewithconfidencetheintrinsicTeVspectrum
model dependent) ’intrinsic’ VHEγ-ray emissions. – which itself can be variable.
The opposite approach, of a more phenomenologi- In this paper we propose a method to mea-
eConf C091122
2 2009 Fermi Symposium, Washington, D.C., Nov. 2-5
sure the EBL that improves on [24] by making a III. BLAZAR SSC EMISSION
more realistic assumption on the intrinsic TeV spec-
trum. Simultaneous optical/X-ray/HE/VHE (i.e.,
In order to reduce the degrees of freedom, we use a
eV/keV/GeV/TeV) data are crucial to this method,
simple one-zone SSC model (for details see [28, 29]).
considering the strong and rapid variability displayed
This has been shown to adequately describe broad-
by most blazars. After reviewing features of EBL
band SEDs of most blazars ([10, 29]) and, for a given
absorption (sect.2) and of the SSC emission model
blazars, both its ground and excited states ([27]).
(sect.3),insect.4wedescribeourtechnique,insect.5
The reason for the one-zone model to work is that in
we apply it to recent multifrequency observations of
most blazars the temporal variability is clearly dom-
PKS2155-304anddeterminethephoton-photonopti-
inated by one characteristic timescale, which implies
cal depth out to that source’s redshift. In sect.6 we
onedominantcharacteristicsizeoftheemittingregion
discuss our results.
([4]).
The emission zone is supposed to be spherical with
radius R, in motion with bulk Lorentz factor Γ at an
angle θ with respect to the line of sight. Special rela-
II. EBL ABSORPTION
tivisticeffectsaredescribedbytherelativisticDoppler
factor, δ = [Γ(1 βcosθ)]−1. The energy spectrum
The cross section for the reaction γγ e± is ([12]), of the emitting re−lativistic electrons is described by a
→
smoothed broken power-law function of the electron
Lorentz factor γ, with limits γ and γ and break at
3 1 2
σγγ(E,ǫ) = 16σT (1−β2) × γbr. In calculating the SSC emission we use the full
Klein-Nishina cross section.
1+β
2β(β2 2) + (3 β4) ln , (1) As detailed in [29], this simple model can be fully
× (cid:20) − − 1 β (cid:21)
− constrainedbyusingsimultaneousmultifrequencyob-
servations. Indeed, the total number of free param-
where σT is the Thompson cross section and eter of the model is reduced to 9: the 6 parame-
β 1 (mec2)2/Eǫ. Fordemonstrationpurposeslet ter specifying the electron energy distribution, plus
≡ −
us apssume, following [24], that n(ǫ) is the local num- the Doppler factor δ, the size of the emission region
ber density of EBL photons having energy equal to R, and the magnetic field B. On the other hand,
ǫ (no redshift evolution – as befits the relatively low from observations ideally one can derive 9 observa-
redshifts accessible to IACTs), ze is the source red- tional quantities: the slopes of the synchrotron bump
shift, and Ω0=1: the corresponding optical depth due after and above the peak α1,2 (uniquely connected to
to pair creation attenuation between the source and n ),thesynchrotronandSSCpeakfrequencies(ν )
1,2 s,C
the Earth, is (see ([24]) and luminosities L , and the minimum variability
s,C
timescale t which provides an estimate of the size
var
c ze 2 x of the sources through R<ctvarδ.
τ (E,z ) = √1+z dz dx
γγ e Therefore, once the relevant observational quanti-
H Z Z 2 ×
0 0 0
∞ tiesareknown,onecanuniquelyderivethesetofSSC
n(ǫ) σ 2xEǫ(1+z)2 dǫ, (2) parameters.
γγ
× ZE2(xm(1e+c2z))22 (cid:0) (cid:1)
where x (1 cosθ), θ being the angle between the IV. THE METHOD
≡ −
photons, and H is the Hubble constant. We further
0
assume, againfollowing[24], thatthelocalEBLspec-
The method we are proposing stems from the con-
trumhasapower-lawform,n(ǫ) ǫ−2.55. ThenEq.(1)
siderationthatboththeEBLandtheintrinsicVHEγ-
yields τ(E,z) E1.55zη with η 1∝.5.
∝ s ∼ ray spectra of background sources are fundamentally
This calculation, although it refers to an idealized unknown. In order to measure the EBL at different
and somewhat simplified situation, highlights an im- z, one should single out a class of sources that is ho-
portant property of the VHE flux attenuation by the mogeneous,i.e. itcanbedescribedbyonesameemis-
γVHEγEBL e+e− interaction: τγγ depends both on sionmodelatallredshifts. Thisapproachismeantto
→
the distance traveled by the VHE photon (hence on minimize biases that may possibly arise from system-
z) and on its (measured) energy E. So the spectrum atically differentSED modelings adopted fordifferent
measured at Earth is distorted with respect to the classes of sources at different distances. So we choose
emitted spectrum. In detail, the expected VHE γ-ray the class of source that both has the simplest emis-
flux at Earth will be: F(E)= (dI/dE)e−τγγ(E) (dif- sionmodelandhasthepotentialityofbeingseenfrom
ferential) and F(>E)= ∞(dI/dE′)e−τγγ(E′)dE′ (in- largedistances: blazars,i.e. theAGNwhoserelativis-
E
tegral). R ticjetspointtowardtheobserversotheirluminosities
eConf C091122
2009 Fermi Symposium, Washington, D.C., Nov. 2-5 3
areboostedbyalargefactoranddominatethesource V. RESULTS: APPLICATION TO
fluxwiththeirSSCemission. Withinblazars, wepro- PKS2155-304
pose to use the sub-class of ”high-frequency peaked
BL Lacs” (HBL), because their Compton peak can We apply the procedure described in Sect.4 to
be more readily detected by IACTs than other types the simultaneous SED data set of PKS2155-304 de-
of blazar, and because their HE spectrum can be de- scribed in [2]. The data and resulting best-fit SSC
scribed as a single (unbroken) power law in photon model (from optical through HEγ-rays) are shown
energy, unlikely other types of blazar ([1, 15]). in Fig.(1). The extrapolation of the model into the
VHEγ-ray range clearly lies below the observational
For a given blazar, our method relies on using, a
H.E.S.S. data, progressively so with increasing en-
simultaneous broad-band SED that samples the op-
ergy. We attribute this effect to EBL attenuation,
t(ifcroaml, Xt-hreayF,ehrimghi-etneleersgcyop(eH)E, :aEnd>V1H0E0γM-reaVy) (γfr-roamy Fobs(E; z)=Fem(E; z)e−τγγ(E;z). The corresponding
values of τ (E; z) for E=0.23, 0.44, 0.88, 1.70TeV
γγ
Cherenkov telescopes) bands. A given SED will be
and z=0.12 are, respectively, τ = 0.12, 0.48, 0.80,
γγ
best-fitted, from optical through HEγ-rays, with a
and 0.87 .
Synchrotron Self-Compton (SSC) model. [Photons
We note that the SED analysis of [2] was based on
with E< 100GeV are largely unaffected by EBL at-
a slightly different SSC model, that involved a three-
tenuatio∼n (for reasonable EBL models) as long as
slope (as opposed to our two-slope) electron spec-
z< 1.] Extrapolating such best-fitting SED model
trum. This difference may lead to a somewhat differ-
in∼to the VHE regime, we shall assume it represents
ent decreasing wing of the modeled Compton hump,
the blazar’s intrinsic emission. Contrasting measured
and hence to a systematic difference in the derived
versus intrinsic emission yields a determination of
τ (E; z). That said, it’s however interesting to note
e−τγγ(E,z), the energy-dependent absorption of the γγ
that the main parameters describing the plasma blob
VHE emission coming from a source located at red-
(B,δ,n )takeonsimilarvaluesinourbest-fitanalysis
e
shift z due to pair production with intervening EBL
and in [2].
photons. Upon assumption of a specific cosmology,
InFig.(2)wecompareourdeterminationofτ with
γγ
thefinalstepisderivingtheEBLphotonnumberden-
some recent results ([8]) or upper limits ([11, 14, 19]).
sity.
Whereasourvaluesaregenerallycompatiblewithpre-
viously published constraints, we note that our val-
Using,foreachblazar,SEDsfromdifferentstatesof
ues closely agree with the corresponding values of [8],
emission will improve the accuracy of the method by
which are derived from galaxy number counts and
increasing the number of EBL measurements at that
hence represent the light contributed by the stellar
redshift.
populationsofgalaxiespriortotheepochcorrespond-
ing to source redshift z – i.e., the minimum amount
s
(i.e., the guaranteed level) of EBL.
VI. DISCUSSION
A. Best-fit procedure: χ2 minimization
The method for measuring the EBL we have pro-
posed in this paper is admittedly model-dependent.
In order to fit the observed optical, X-ray and HE However, its only requirement is that all the sources
γ-ray flux with the SSC model, a χ2 minimization used as background beamlights should have one same
is used. We vary the SSC model’s 9 parameters by emission model. In the application proposed here, we
small logarithmic steps. If the variability timescale have used a one-zone SSC model where the electron
of the flux, t , is known, one can set R ct δ, spectrumwasa(smoothed)doublepowerlawapplied
var var
∼
so the free parameters are reduced to 8. We assume to the SED of the HBL blazar PKS2155-304. While
here γ =1: for a plasma with n O(10)cm−3 and this choice was encouraged by the current observa-
min e
≈
B O(0.1)G (as generally appropriate for TeV blazar tional evidence fact that seem to HBLs have, with
≈
jets: e.g., [5, 7, 10]), this approximately corresponds no exception, single-slope Fermi-LAT spectra ([15]),
to the energy below which Coulomb losses exceed the we could have as well adopted the choice ([2]) of a
synchrotron losses (e.g., [20, 22]) and hence the elec- triple power law electron spectrum in our search for
tron spectrum bends over and no longer is power- thebest-fitSSCmodelofPKS2155-304’sSED.Should
law. However, in general γ should be left to vary – the latter electron distribution be shown to provide a
min
e.g., cases of a ”narrow” Compton component require better fit to HBL Fermi-LAT spectra, then it would
γ >1 ([30]). In order to reduce the run time of the become our choice. In general, what matters to the
min
code, the steps are adjusted in each run such that, a application of this method, is that allsource SEDs be
larger χ2 is followed by larger steps. fit with one same SSC model.
eConf C091122
4 2009 Fermi Symposium, Washington, D.C., Nov. 2-5
FIG. 1: Data (symbols: from [see 2]) and best-fit SSC model (solid curve) of the SED of PKS2155-304. The best-fit
SSC parameters are: ne =150cm−3, γbr =2.9×104, γmax =8×105, α1 =1.8, α2 =3.8, R=3.87×1016cm, δ=29.2,
B =0.056G. The obtained values of R and δ imply a variability timescale tvar ∼R/(cδ), which is compatible with the
observed value of ≈12hr.
Thisworkwillbethesubjectofaforthcomingpaper ([17]).
[1] A.A.Abdo et al. (LAT collab.), ApJ, in press (2009) ceedings (2009)
[arXiv:0910.4881] [10] G.Ghisellini, A.Celotti, G.Fossati, et al., MNRAS,
[2] F.A.Aharonian, et al. (H.E.S.S. collab.), ApJ, 696, 301, 451-468 (1998)
L150-L155 (2009) [11] R.C.Gilmore, P.Madau, J.R.Primack, et al., MN-
[3] F.A.Aharonian,etal.(H.E.S.S.collab.),Nature,440, RAS, 399, 1694-1708 (2009)
1018-1021 (2006) [12] ”TheQuantumTheoryofRadiation(Oxford: Oxford
[4] H.Anderhub,etal.(MAGICcollab.),ApJ,705,1624- University Press)” (1960)
1631 (2009) [13] T.M.Kneiske, K.Mannheim, & D.H.Hartmann,
[5] G.Ghisellini & L.Costamante, A&A, 384, 56-71 A&A, 386, 1-11 (2002)
(2002) [14] T.M.Kneiske, T.Bretz, K.Mannheim, &
[6] H.Dole, G.Lagache, J.-L.Puget, et al. A&A, 451, D.H.Hartmann, A&A, 413, 807-815 (2004)
417-429 (2006) [15] B.Lott, these proceedings (2009)
[7] J.D.Finke, C.D.Dermer, & M.B¨ottcher, ApJ, 686, [16] D.Mazin, & F.Goebel, ApJ, 655, L13-L16 (2007)
181-194 (2008) [17] N.Mankuzhiyil, M.Persic, & F.Tavecchio, in prepa-
[8] A.Franceschini, G.Rodighiero, & M.Vaccari, A&A, ration (2010)
487, 837-852 (2008) [18] D.Mazin, & M.Raue, A&A, 471, 439-452 (2007)
[9] M.Georganopulos, J.Finke, & L.Reyes, these pro- [19] M.Raue, & D.Mazin, Int.J.Mod.Phys.D17, 1515-
eConf C091122
2009 Fermi Symposium, Washington, D.C., Nov. 2-5 5
FIG. 2: Measured values of τ (E; z) for E=0.23, 0.44, 0.88, 1.70TeV derived from comparing, for simultaneous
γγ
observations of the HBL blazar PKS2155-304 (z=0.12), the (EBL-affected) VHEγ-ray data with the eV-through-GeV
best-fitting SSC model extrapolated into the TeV domain. The curves represent, for redshifts z = 0.05, 0.1, 0.2, 0.3,
0.4,and0.5,theopticaldepthτ (E)accordingto[8](top left),andupperlimitstoitaccordingto[11](top right),[14]
γγ
(bottom left), [19] (bottom right).
1520 (2008) [26] F.W.Stecker, M.G.Baring & E.J.Summerlin, ApJ,
[20] Y.Rephaeli, ApJ, 227, 364-369 (1979) 667, L29-L32 (2007)
[21] M.H.Salamon, & F.W.Stecker, ApJ, 493, 547-554 [27] G.Tagliaferri,etal.(MAGICcollab.),ApJ,679,1029-
(1998) 1039 (2008)
[22] C.L.Sarazin, ApJ, 520, 529-547 (1999) [28] F.Tavecchio&L.Maraschi,ApJ,593,667-675(2003)
[23] F.W.Stecker & O.deJager, A&A, 334, L85-L87 [29] F.Tavecchio, L.Maraschi, & G.Ghisellini, ApJ, 509,
(1998) 608-619 (1998)
[24] F.W.Stecker & O.C.deJager, & M.H.Salamon ApJ, [30] F.Tavecchio, G.Ghisellini, G.Ghirlanda, et al., MN-
390, L49-L52 (1992) RAS, in press (2009) [arXiv: 0909.0651]
[25] F.W.Stecker, M.A.Malkan & S.T.Scully, ApJ, 648,
774-783 (2006)
eConf C091122