5. CONCLUSIONS AND QUESTIONS
The spectral energy distributions (SEDs) of low-redshift (z) AGN are now
fairly well-observed and understood throughout much of the
electromagnetic spectrum. The dispersion around the median SED is large
and needs to be addressed by models. Key regions which remain under-studied
are the far-IR-submm, the EUV and the
-ray.
At higher redshift, samples remain small but are growing.
The main conclusions are:
- an optically thick, geometrically thin AD plus surrounding corona
is the most viable
model for the OUV BBB and ultra-soft X-ray excess components of AGN SEDs.
- cool dust dominates the far-IR continuum of RQQs and LDRLQs while
CDRLQs are dominated by non-thermal, synchrotron emission.
- the X-ray spectrum is a combination of an underlying power law and
reflection of this emission in the cool material at the surface of an AD.
The amount of reflection occurring decreases with the luminosity of the
source.
- there is significant evidence for an increase in the relative
strength of the OUV BBB with luminosity.
Key open questions in the interpretation of AGN SEDs are:
- Are the radio-IR SEDs of CDRLQs and LDRLQs different? This
is expected if dust dominates the IR emission in LDRLQs and RQQs while
the CDRLQs are non-thermal.
- What is the true range of SEDs in the AGN population?
The relatively unbiassed, hard-X-ray selected samples facilitated by the
X-ray satellites AXAF (NASA) and XMM (ESA), both due
for launch in 1999, provide the best opportunity for a relatively unbiassed
sample. Radio samples are also relatively unbiassed
but are largely confined to the radio-loud
10% of the population, which may introduce its own biasses.
These samples would allow us to address the following questions:
- How important is dust attenuation to the observational
properties of AGN?
- How important is orientation to the observed properties of
RQQs? This question is probably linked to the last one since
the distribution of attenuating material is likely to depend on
the source orientation. The question may perhaps be addressed by further
studying the NLSy1 objects and the possible role of orientation
in explaining their properties.
- What is the interplay between thermal and non-thermal
emission in the near and far-IR? ISO may begin to address
this question by observing the IR SEDs of
the brightest sources but we will not be able
to observe representative samples until SIRTF begins its operation
~ 2002. IR variability studies also address this question
and are becoming more possible as IR array technology advances.
- Does the SED evolve? More SEDs for high redshift quasars
are needed to address this question.
Acknowledgments.
I would like to thank the conference organisers for inviting me to
present this review and so forcing me to catch up with at least some
of the recent literature. I am very grateful to my colleagues at
SAO, in particular Martin Elvis, Paul Green, Eric Hooper,
Joasia Kuraskewicz, Smita Mathur, Harvey Tananbaum and Marianne Vestergaard
for many fruitful discussions. I acknowledge the financial support
of NASA contract NAS8-39073 (AXAF Science Center).