Broadband Properties of Markarian 501
Markarian 501 (RA:16 52 11.75, DEC: +39 50 24.6 (1950)) was discovered
on the basis of its strong UV continuum and unusually blue spectrum
(Markarian and Lipovetskii, 1972). It has been identified with the
radio source B2 1652+39 (Colla et al., 1975). The properties of the
object include large optical polarization and variability, and
spectroscopic studies (e.g. Ulrich et al., 1975) show weak absorption
lines characteristic of an elliptical host galaxy in an otherwise
featureless spectrum. Thus, Markarian 501 is classified as a BL Lac
object. The derived redshift of the galaxy is 0.0337. Because of its
strong radio emission Markarian 501 is often classified as an
RBL. However, it is also a strong X-ray source and may be classified
as an XBL. Examination of its broadband multi-wavelength spectrum
clearly reveals that it is an HBL. In this section the properties of
Markarian 501 at radio, IR, optical, UV and X-ray energies prior to
its discovery at gamma-ray energies are reviewed.
(Thanks to John Quinn for most of the text)
Host Galaxy
Photometric profiles from CCD photometry studies of Markarian 501 by
Hickson et al., (1982) are found to be consistent with composite
profiles from a point source (the unresolved nuclear component) and
from an elliptical galaxy that obeys the de Vacouleurs surface
brightness law. The photometric parameters derived for the underlying
galaxy show it to be consistent with a normal elliptical galaxy.
The redshift of Markarian 501 has been derived from the Ca II lines,
the G band, the Mg I b band and the Na I D line (Ulrich et al.,
1975; Moles et al., 1987) seen in the outer parts of the
galaxy. Spectroscopic studies of the inner region apparently reveal
some emission lines which, at the previously quoted (absorption)
redshift, do not match the position of the expected lines (Moles et
al., 1987). the emission features as a whole appear at a larger
redshift. The best guess of Moles et al. is that the nucleus of
Markarian 501 shows Seyfert-line characteristics with a narrow Halpha;
line at a larger redshift than the other emission lines. Moles also
found that the main difference (apart from the active galactic
nucleus) between the host galaxy and normal elliptical galaxies is the
high metallicity with the prescence of Fe and Mg lines.
|
|
 |
Radio Band
Markarian 501 is a moderately luminous flat spectrum radio source,
with a flux of ~1.3 Jy at 5 GHz (e.g. Mufson et al.,
1984). Observations made over the last 20 years have produced no
evidence for short term variability and only marginal variability on
longer time scales. The largest variations seen involve flux changeds
of less than 20% (at 5 GHz) over several years, with fluxes ranging
from 1.20 Jy (Cruz-Gonzales and Huchra, 1984) to 1.42 Jy (Kuhr et al.,
1981). Most of the radio emission comes from the core; at 1.5 GHz the
core flux is 1.35 Jy while the extended flux is 0.067 Jy
(Laurent-Muehleisen et al., 1993. For flux measurements spanning the
radio regime see, for example, Owen et al. (1980) and Gear et
al. (1994).
The University of Michigan 26m Radio telescope has taken data on
Markarian 501 since 1977. Right is a plot of data taken at (from top)
4.5, 8 and 14.5 GHz which show up the small variations of Markarian
501 over the last 25 years.
|
|
 |
|
The radio structure of Markarian 501 has been the focus of many
studies due to the good linear resolution obtainable because of its
proximity. VLBA and VLA measurements have produced evidence for
grossly misaligned small-scale and large scale structure (van Breugal
and Schilizzi, 1986). The distribution of alignments of radio and jet
position angles (PA) of extra-galactic radio sources on parsec and
kilo-parsec scales shows an unusual and bimodal distribution, with one
population having well aligned jets and the second have orthogonal
misalignments. The results of van Breugel and Schilizzi indicate that
a one-sided jet emerges from the nucleus of Markarian 501 with PA
133o and curves towards 104o within a projected
distance of 14pc. This small-scale jet is misaligned by
~80o with the elongation seen in the VLA map (>40 pc) of
the nucleus and the more extended diffuse emission. The small-scale
jet is approximately aligned with the optical polarization position
angle of 141o pm 7o (Angel and Stockman,
1980). No evidence for variability on scales as small as 1 pc were
seen.
Possible explainations for the missalignments given by Breugel and
Schilizzi are projection effects enhanced by relativistic beaming,
collision with circumnuclear gas or precession. However, given the
lack of variability seen in Markarian 501 and consequently the
abscence of superluminal motion they claim that there is little
evidence for bulk relativistic motion, in agreement with the SSC
interpretation by Mufson et al. (1984) that the broad-band spectrum of
Markarian 501 is caused by a jet with a Doppler factor of 1.1,
orientated at 23o w.r.t. the line of sight. The theory of
collision with circumnuclear material may explain the absence of
bright radio lobes. If precession is the source of the misalignments
then van Breugel and Schilizzi have estimated, from the absence of
radio variability, that the precession time tprecess > 8 x
103 yr.
An alternative explanation for the orthogonal misalignments, suggested
by Conway and Murphy (1993), is that the effect is due to Doppler
boosting of the jet with a helically distorted path. Such a geometry
should develop naturally in a hydrodynamically driven jet due to the
effects of Kelvin-Helmholtz instabilities (for a review see
Birkinshaw, 1991) caused by the orbital motion of a binary black hole
system. In fact, Conway and Murphy (1993) argue that all sources with
orthogonally mis-aligned jets contain binary black-hole nuclei while
aligned jet sources contain single black-holes. Six years of Markarian
501 observations with VLBA (Conway and Wrobel, 1995) shows the jet
morphology unchanged, implying pattern speeds much less than 1c. The
observations reveal a twist of about 79o in the jet PA
which occurs between the scales of 2 mas and 100 mas.
|
|
 |
Infra-red, Optical and Ultra-violet
Markarian 501 has been extensively monitored at IR, optical and UV
wavelengths. It is not a very active source in these wave-bands
(Mufson et al., 1984), possibly due to the very strong contribution
from the galaxy which swamps the emission from the nucleus and the
jet. The ratio of non-thermal to galactic flux in the V band was found
to be 0.33 by Mufson (1984) while in the K band it is estimated to be
0.1 (Kidger and de Diego, 1992), implying that, in this band, the
galaxy totally dominates the emission. Nevertheless, variability in
these wave-bands has been reported.
Kidger and de Diego (1992) have reported evidence for co-ordinated
variability on time scales of hours in the visible and IR. On one nigt
(890724) there is evidence for a 0.4 magnitude flare in B with a
simultaneous but smaller amplitude in K. The next night thre is a
steady decrease of 0.035 magnitude oer hour in B with no change in
K. Further observations in the IR are reported in Kidger et al.,
(1992). During these observations, Markarian 501 was in a low state,
with K-band magnitude of ~ 11.5 whereas its typical magnitude is 10.5
(Impey, 1983; Kidger and de Diego, 1992). Kidger et al., (1992) report
a large anti-flare which involved a change in the J-band magnitude of
1.11 in 2 days, the largest day-scale IR variation ever seen in
Markarian 501. Other rapid fluctuations observed include a change in
magnitude off 0.2 in half and hour. Assuming a synchrotron source for
the IR, the authors used the rapid variations to derive a value for
the magnetic field of 13 G.
Pictured right is
the Mount Lemmon Observatory 60-inch Cassegrain reflector used by Mufson et al.
|
|
 |
| |
|
In the optical wave-bands Markarian is usually steady (Mufson et al.,
1984) but there have been some reports of small amplitude
variations. Heidt and Wagner (1996) included Markarian 501 in their
sample of 34 radio selected BL Lac objects which were examined for
optical intra-day variability. Markarian 501 was one of the 28 which
exhibited variability at the 99.5% confidence level using a
chi-squared test. However, they could not determine a pronounced
time-scale for variability in Markarian 501 using a structure function
analysis. Xie et al. (1996) found that Markarian 501's
flickering amplitude is smaller than 0.2 magnitude (20%) in B
and V bands and the largest variability they witnessed was dV =
0.8 mag. between observations separated by 345 days. They also saw
short term variations with a brightening of 0.3 magnitude in B in 30
minutes and then a decrease of 0.2 magnitude. Pictured right is
the Mount Lemmon Observatory 60-inch Cassegrain reflector used by Mufson et al.
|
|
 |
| |
|
In a study of optical polarization properties of X-ray selected BL Lac
objects, Januzzi et al. (1994) found that the optical emission from
Markarian 501 is polarized with the maximum polarization angle
(Pmhaving a value of 41.8% (white) and 7.77%
(filtered). Both the percentage polarization and the polarization
angle were found to be variable. The polarization angle showed a
preffered position angle of 115o but varied over a range
91o to 130o.
|
|
|
| |
| The trends seen in Markarian 501 in the IR
and optial are also seen in the UV. Markarian 501 has a mean UV flux
of 1.64 mJy (Edelson, 1992) and shows small amplitude variability on
time-scales of months to years. In a sample of 14 IUE (International Ultraviolet
Explorer) blazars examined by Edelson, Markarian 501 was found to
exhibit the smallest amplitude fluctuations with a fractional
r.m.s. variability of 7.8%. Snijders et al. (1979) found that
Markarian 501's UV spectrum could not be fitted by a simple power law,
it is steeper from 3000 A to 2000 A and flattens from 2000 A to 1250
A.
|
|
 |
| |
| |
X-ray
Markarian 501 is among the brightest of BL Lacs at X-ray energies,
with a typical flux of ~ 10 microJy at 1 keV. Markarian 501 and
Markarian 421 were the first two BL Lacs to be indentified as X-ray
sources (Rickets et al., 1976). Markarian 501 appears to be more
variable, both in flux and spectrum, at X-rays than at lower
energies. However, the variability seen is less extreme than in other
BL Lacs.
Consistent with observed properties at other wavelengths Markarian 501
exhibits flux variability preferentially on long time-scales,
typically months to years. Reports of rapid variability are relatively
rare. Singh and Garmine (1985) have reported seeing a decrease in flux
in 0.15-0.50 keV band by a factor of two over a period of two days in
observations made with HEAO1 A-2 experiment in August 1977. In an
analysis of the EXOSAT archive, Giommi et al. (1990) have found
evidence for flux variability of amplitude 20% to 40% within a single
nights observations. The variations appear random. Many other
observations have found no evidence for short-term variability
(e.g. Comastri et al., 1997, Urry et al., 1996, Sambruna et al.,
1994a, Fink et al.,1991, Staubert et al., 1986, George et al., 1985,
Snijders et al., 1979).
Long term variability (months to years) has been reported by many
authors (e.g. Urry et al., 1996, Ciliegi et al., 1995, Sambruna et
al., 1994a, Sambruna et al., 1994b, George et al., 1985). Giommi et
al. (1990) have produced a chi-sqaured test for medium to long term
variability in the EXOSAT archival Markarian 501 data and fin that the
probability for constant emission is ~ 10-7 for the flux in
the 0.05-2.0 keV band, and ~ 0 for the 0.7-50 keV range. Ciliegi et
al., (1993) have produced a complete catalogue of X-ray properties of
42 BL Lacs, spanning 20 years and several satellite missions including
ARIEL V, HEAO1, HEAO2, GINGA, EXOSAT and ROSAT. In an analysis of this
archive they Ciliegi et al., (1995) find that Markarian 501 is
variable but the amplitude of variability is less than for some other
sources in the archive. The flux from Markarian 501 is found to vary
by a maximum factor of ~ 5 whereas other objects, such as Markarian
421, vary in flux by factors of 100 or more. This, they classify
Markarian 501 as weakly variable.
The X-ray spectrum of Markarian 501 is generally not well fit by a
thermal bremsstrahlung model. The best fit is usually obtained with a
broken power-law with the higher energies having a steeper spectrum
resulting in a convex spectrum (Comastri et al., 1995; Sambruna et
al., 1994b). The break energy appears to be in the range 1-4 keV
(Sambruna et al., 1994a). A convex sprectrum appears to have been
observed on one occasion by simultaneous observations by singh and
Garmire (1985) and Mushotzky et al. (1978) but this result us disputed
by George et al. (1985).
The X-ray spectrum exhibits variability which appears to be correlated
with the flux. In early observations it appeared that the spectrum
featured a fluctuating hard component which was present about 20% of
the time (Mufson et al., 1984). Sambruna et al. (1994b) and Giommu et
al., (1990) have found that the spectrum hardens with increasing
flux. Giommu et al. (1990) also find that the flux is more variable in
the 0.7-50 keV band than in the 0.05-2 keV one. George et al. (1985)
attributed the flux variability to pivoting of the high energy
continuum about a break point. In the analysis of six BL Lac objetcs
by Ciliegi et al. (1995) for which low and medium energy spectra were
available, Markarian 501 is the only source found to show a
correlation between flux and spectral index. The effect was
significant at soft energies (0.2-4 keV) and marginal but still
present at higher energies (2-10 keV). The low energy spectrum was
found to soften as the source brightens while the medium energy
spectrum hardened.
|
|
 |
| |
|