WDS'08 Proceedings of Contributed Papers, Part III, 213–219, 2008.
ISBN 978-80-7378-067-8 © MATFYZPRESS
Study of Blazars Across Electromagnetic Spectrum
I. Sujová
Charles University, Faculty of Mathematics and Physics, Prague, Czech Republic.
R. Hudec, F. Münz, and A. Kutka
Astronomical institute of the Academy of Sciences of the Czech Republic, Ondřejov, Czech
Republic.
Abstract. The X-ray sky is populated with various fascinating objects, many of which
are variable or transient. Blazars are the most variable class of the Active Galactic Nuclei
(AGN). Study of this kind of objects with standard instrumentation is difficult. We will
introduce the Lobster Eye X-ray telescope (LET), an ideal tool for studying variable,
transient, and rare events. We will discuss the possibilities of AGN/blazar studies with this
new class of X-ray instruments. Additionally, we demonstrate a possible future application
of LET data to study the unique variability of blazars by the means of variogram analysis.
We present preliminary results of optical light curve analysis of object 1ES 2344+514 and
discuss our future plans.
The Unified Model of AGN
Our current understanding of Active Galactic Nuclei (AGN) is summarized by the Unified Model
of AGN (on the Fig. 1). The basic idea behind it is, that all of the different AGN types which we observe
have the same internal structure of their nucleus. The observed differences arise from the orientation
dependence of the system.
• When we are looking directly into the jet, we call this class of AGN blazars. Broad and narrow
lines are often visible and emission from the jet itself can be strong.
• Objects known as Seyfert galaxies type II and Radio galaxies are observed in the perpendicular
direction to the jet. Black hole, accretion disc and broad line region are obscured by dusty torus.
• Seyfert galaxies I and Quasars are observed, if the line of sight is in the direction of about 45◦
from the axis of AGN.
The basic model of AGN is consists of a supermassive black hole (> 106 M ) which accretes matter
into an accretion disk (AD). Material in the AD is strongly heated and matter together with radiation can
be ejected along the rotation axis in jets.
Blazars
Blazars are the most extreme class of Active Galactic Nuclei with the angle between the line of sight
and jet ≤ 10◦ . They are very compact, highly variable energy sources and they usually can be classified
as core-dominated radio objects. Strong variability and polarization are the most important indicators of
blazars. Polarization is significant from radio to optical wavelengths and variability at all wavelengths.
Blazars can be divided into two basic subgroups:
• BL Lac objects – show featureless optical continuum and very weak or no emission lines
• Flat-spectrum radio quasars (FSRQs) – exhibit broad optical emission lines
The electromagnetic continuum spectra of blazars are dominated by non-thermal emission at all wavelengths. This non-thermal spectrum is characterized by two spectral components:
• low-energy component – is located in radio to UV/X-rays and arises from synchrotron emission
of ultrarelativistic electrons
213
SUJOVA ET AL.: STUDY OF BLAZARS ACROSS ELECTROMAGNETIC SPECTRUM
Figure 1. Unified model of AGN is shown on the left panel. The different types of radiation are emitted
from individual parts of AGN (panel on the right), therefore the observed spectrum depends on the
orientation of the AGN. (Image credit: Brooks/Cole Thomson Learning)
• high-energy component – is created by inverse Compton upscattering of photons by electrons and
is visible from X-ray to γ-rays
BL Lac objects are categorized according to the location of the peak of their low-energy component
as low-frequency peaked (LBL) and and high-frequency peaked (HBL) with maximum in the nearIR/optical and in the UV/X-ray bands respectively.
Multi-band cross-correlation studies
The relations between the main spectral components are very important for the understanding
physics of blazars, therefore we are searching for correlation between variations of fluxes in different
parts of the electromagnetic spectrum. We are able to explain correlations between some spectral parts,
but we don’t find any link between others. Moreover, sometimes we can observe time lags between
variability in different bands.
Correlated variability between the γ-rays and X-rays was observed during strong flares in the Mkn
421 and Mkn 501 systems over periods of weeks to months (Krawczynski, 2000). Similarly, correlations
in optical, X-rays, and γ-rays were found after strong flares in 1ES 1959+650 in 2002 (Krawczynski,
2004). For more details on correlations of variability in different spectral bands see Wagner, 2008.
Correlation between X-ray and γ-ray emission can be explained by the synchrotron self-Compton
emission model (SSC) (Blazejowski et al., 2005). Synchrotron self-Compton emission arises when γray photons are produced from Compton-upscattering of X-ray photons. X-ray variability leading the
variability in γ-rays is naturally expected in this model.
Variability in X-rays following optical variability can be due to inverse Compton scattering of radio
or IR photons. To explain variability occurring first at X-rays, a model was proposed with synchrotron
emission emitted from perturbations propagating along the jet (Abraham et al., 2004).
We encounter various problems during correlation studies. Most importantly, data in different parts
of spectrum has to be obtained at the same time, otherwise cross-correlation analysis can not be carried
out. Especially the X-ray coverage with current instrumentation is insufficient. One of the projects that
could help to resolve these problems and provide a significantly better X-ray coverage is the X-ray
Lobster Eye Telescope (LET).
Lobster Eye Telescope
Some cameras are based on the principle of refraction of light, analogously to eyes of vertebrates.
The Lobster Eye telescope is inspired by the structure of the eye of lobsters, which works on the principle
214
SUJOVA ET AL.: STUDY OF BLAZARS ACROSS ELECTROMAGNETIC SPECTRUM
of reflection. Lobster eye is composed of numerous square segments and reflected light is focused through
them onto the retina (in our conception onto a detector).
Figure 2. Lobster Eye telescope with Schmidt and Angel X-ray optics.
The LET has X-ray optics which were proposed by Schmidt (1975) and Angel (1979) (see Fig. 2).
While the widely used classical Wolter grazing incidence mirrors are limited to roughly 1 deg field of
view (FOV) (Priedhorsky et al., 1996; Inneman et al., 2000), the FOV of the LET is very wide (= 1000
square degrees). The angular resolution is better than 4 arcmin.
The energy range of Lobster Eye Telescope is 0.1 − 6.0 keV and it is expected to achieve daily
limiting flux ∼ 10−12 erg cm−2 s−1 (Hudec et al., 2004). Several projects with LET were proposed, but
by this time, no proposals has been approved yet (Hudec et al., 2006).
Blazar counts prediction for LET
To show how LET can help us to study blazars, we made a prediction of how many known blazars
could be observed with it. We used the data from the ROSAT all-sky survey bright source catalogue
(Voges et al., 1999) with total ∼ 18800 sources. This catalogue is publicly available on the web page
ftp://ftp.xray.mpe.mpg.de/rosat/catalogues/rass-bsc/. We compare blazar fluxes with the daily limiting
flux of LET ∼ 10−12 erg cm−2 s−1 . As we see on Fig. 3, LET, owing to its wide field of view and good
sensitivity, will be able to observe unprecedentedly many objects for long time periods. Providing long
term X-ray light curves for many blazars would be one of the major contributions of LET and can not
be matched by any of the currently existing or near-future X-ray missions.
The number of detectable blazars to all ROSAT blazars in percentage for individual energy ranges
are:
• 0.1 − 0.5 keV = 90.2%
• 0.5 − 0.9 keV = 23.2%
• 0.9 − 2.0 keV = 45.1%
Additionally, fluxes in this prediction are used as given in the ROSAT bright source catalogue and
thus some blazars that can’t be detected in their quiescent state, will still be detectable during their flaring
periods.
215
SUJOVA ET AL.: STUDY OF BLAZARS ACROSS ELECTROMAGNETIC SPECTRUM
Figure 3. Graphs show fluxes of blazars for individual energy ranges compared to limiting flux for daily
scanning observation of LET.
Study of variability
Light curves with good time coverage of light curves (i.e. more frequent and regular sampling)
obtained by LET could be used not only for cross-correlation analyses, but also for variability studies.
We distinguish between different modes of the blazar variability:
• micro-variability - few tenths of magnitude on time scales of a day or less
• short-term outbursts - few weeks to several months
• long-term trends - few months to years
The main objective of variability studies is to establish whether there are periodic (or semi-periodic)
variations in the observed light curves, quantify them and eventually describe by a physical model.
There is a great variety of different methods for studying periodicity in time series. The best know
methods like the Fourier analysis (Deeming, 1975), analysis of variances (Schwarzenberg - Czerny,
1989) etc. are most suitable for description of (nearly-) strictly periodic signals, where there is a full
conservation of the phase and amplitude of the signal.
However, the available datasets for blazars are marked by strong non-equidistance of observations,
non-negligible noise and by the intrinsic complex variability patterns in these systems. Standard methods
of period analysis are therefore not very suitable and we have to resort to robust methods like the
autocorrelation or variogram methods (see e.g. Eyer and Genton, 1999, also including a more detailed
discussion about the comparison with standard methods). We will discuss and demonstrate the last
mentioned method - the variogram analysis.
Variogram
Variogram or the ”Structure Function” is a statistical function that describes the increasing differences or decreasing correlation, or continuity, between sample values as separation between them
increases.
216
SUJOVA ET AL.: STUDY OF BLAZARS ACROSS ELECTROMAGNETIC SPECTRUM
Mathematical form of the variogram is:
Nh
1 X
[I(ti ) − I(ti + h)]2 ,
2γ(h) =
Nh
h ∈ R,
i=1
where Nh is number of experimental pairs [I(ti ), I(ti + h)] of data with distance h. They are computed
for all pairs with time differences hij = JDj − JDi (JD is the Julian Date).
We can see from this formula that the presence of a periodicity in the variability would create a
minimum in the variogram at the time lag corresponding to the period.
The term variogram is sometimes used incorrectly in place of ”semivariogram”. The difference is
only that the semivariogram uses each pair of data elements only once, whereas the variogram uses all
possible data pairs. For more details about variogram see Eyer and Genton, 1999.
We will demonstrate the use of this method to study variability of the so called Very High Energy
(VHE) blazars.
Very High Energy objects
One of the most interesting subclasses of blazars are the Very High Energy blazars (sources of VHE
gamma emission). Very High Energy objects are the objects observable in energies E > 100 GeV (Horns,
2008). Studying the VHE gamma-ray production mechanisms is crucial to understand the association to
the massive black hole in the center of the AGN. The Major Atmospheric Gamma Imaging Cerenkov
(MAGIC) Telescope program (Wagner, 2003) is following up known VHE objects and searching for
new low and medium redshift blazars emitting at VHE gamma emission.
The object 1ES 2344+514
1ES 2344+514 is one of the best known VHE sources. It was identified as a BL Lac object by
(Perlman et al., 1996) and later specified as a HBL blazar. The basic characteristics of the object are:
• redshift = 0.044
• black hole mass = 10(8.8±0.16) M
• energy threshold = 350GeV
For more see Schroedter et al., 2005.
In collaboration with Gino Tosti and Stefano Ciprini from Perugia University Observatory, Italy we
obtained multi-band photometry of 1ES 2344+514. The observations were carried out in V, R, I bands
during the period (21.06.2000–23.08.2006).
Light curves for this object in V, R, I bands are shown on the Fig. 4. As an example, we have
constructed a variogram of the V band light curve (right panel of Fig. 4). It doesn’t exhibit a distinct
minimum, but a minimum at the time lag about 700 days might hint to a possible (quasi-) period on this
time scale. We will present more detailed analysis the light curves of this object and several other VHE
blazars in the near future.
Period searches in optical bands are possible on large range of time scales, but the available light
curves suffer with highly non-equidistant sampling. This makes the the period analysis with standard
tools very hard, because of the possible presence of false periods and obfuscation of real periodicity. The
demonstrated robustness is a great asset of the variogram method.
Search for periodicity especially in X-ray band is strongly restricted only to short time scales and has
very sparse time sampling on the longer scales. This is due the the currently available X-ray instruments
allowing only pointed observations on relatively short time scales. The monitoring mode of LET will
be able to provide well sampled light curves for large number of blazars and extend variability studies
to much longer time scales. Variograms will still be one of the most promising methods to look for
periodicity in future LET data.
217
SUJOVA ET AL.: STUDY OF BLAZARS ACROSS ELECTROMAGNETIC SPECTRUM
Figure 4. Light curves of blazar 1ES 2344+514 in V, R, I bands on the left panel. Right: Variogram of
the V band.
Conclusion
Blazars are very a compact, highly variable class of Active Galactic Nuclei. Searching for correlation
between variation of fluxes in different parts of electromagnetic spectrum and studying of their variability
can tell us lot of about the underlying physical mechanisms.
To improve the X-ray coverage of blazars the Lobster Eye X-ray telescope was proposed. We
showed that a future mission with a LET will be able to detect and monitor many blazars providing us
with an unprecedented wealth of information. A LET mission will provide light curves with good time
sampling and sensitivity for a large number of systems, making them an ideal input for crosscorrelation
and period analyses. These in turn might help us advance in our understanding of these fascinating
objects.
Additionally concerning the period analysis, we reviewed the method of variograms and applied it
to the optical light curves of the blazar 1ES 2344+514. Preliminary analysis reveals a possible ∼ 700
days (quasi-) period. In the near future, we will carry out cross-correlation and variogram analyses of
the optical data from Perugia Observatory for several blazars.
Acknowledgments.
Czech Republic.
This research was partially supported by a grant 205/08/H005 of Grant agency of
References
Abraham, Z., and Dominici, T. P., Raio to X-rays properties of TeV blazars, Proc. of IAU Symposium, p. 71-72,
2004.
Angel, J.R.P., Lobster eyes as X-ray telescopes, The Astroph. Journal, vol. 233, p. 364-373, 1979.
Blazejowski, M., Blaylock, G., et al., A multiwavelength view of the TeV Blazar Markarian 421: Correlated
variability, flaring, and spectral evolution, The Astroph. Journal, vol. 630, p. 130-141, 2005.
Deeming, T. J., Fourier analysis with unequally-spaced data, Astrophysics and Space Science, vol. 36, p. 137-158,
1975.
Eyer, L., Genton, M. G., Characterization of variable stars by robust wave variograms: An application to Hipparcos
mission, A&A, vol. 136, p. 421-428, 1999.
Horns, D., High-(Energy)-Lights. The Very High Energy Gamma-Ray Sky, Arxiv, 2008.
Hudec, R., Pina, L., Inneman, A., Sveda, L., Lobster - Eye X-Ray Telescope, Proceedings of the 5th INTEGRAL
Workshop on the INTEGRAL Universe, p. 785, 2004.
Hudec, R., Simon, V., Sveda, L., Pina, L., Inneman, A., Astrophysics with LOBSTER, Chinese Journal of Astr.and
Astroph., vol. 6, p. 383-387, 2006.
Inneman, A., Hudec, R., and Pina, L., Progress in lobster-eye X-ray optics development, Proc. SPIE, vol. 4138, p.
94-104, 2000.
Krawczynski, H., Aharonian, F., et al., TeV γ-ray / X-Ray Multiwavelength Observations of the BL Lac Objects
Mkn 501 and Mkn 421, American Astronomical Society, vol. 32, p. 1191, 2000.
218
SUJOVA ET AL.: STUDY OF BLAZARS ACROSS ELECTROMAGNETIC SPECTRUM
Krawczynski, H., Hughes, S. B., et al., Multiwavelength Observations of Strong Flares from the TeV Blazar 1ES
1959+650, The Astroph. Journal, vol. 601, p. 151-164, 2004.
Perlman, E. S., Stocke, J. T., et al., The Einstein Slew Survey Sample of BL Lacertae Objects, Astroph. Journal
Supplement, vol. 104, p. 251, 1996.
Priedhorsky, W. C., Peele, A. G., and Nugent, K. A., An X-ray all-sky monitor with extraordinary sensitivity,
MNRAS, vol. 279, p. 733-750, 1996.
Schmidt, W.H.K., A proposed X-ray focusing device with wide field of view for use in X-ray astronomy, NucIM,
vol. 127, p. 285-292, 1975.
Schroedter, M., Badran, H. M., et al., A very high energy γ-ray spectrum of 1ES 2344+514, The Astroph. Journal,
vol. 634, p. 947-954, 2005.
Schwarzenberg - Czerny, A., On the advantage of using analysis of variance for period search, Royal Astronomical
Society, vol. 241, p. 153-165, 1989.
Voges, W., Aschenbach, B., et al., The ROSAT all-sky survey bright source catalogue, A&A, vol. 349, p. 389-405,
1999.
Wagner, R., Correlated Variability in Blazars, ARXIV, 15 pages, 2008.
Wagner, R., MAGIC Collaboration, An AGN Observation Catalogue for the MAGIC Cherenkov Telescope,
Proceedings of the 28th International Cosmic Ray Conference, p. 2555, 2003.
219