Many of us have heard of the term "quasar", and maybe even have a vague idea of what that means. But did you know that a quasar is only one type of a much larger class of astronomical objects called "active galaxies"? Active galaxies are those galaxies whose innermost regions are producing tremendous amounts of energy - the most powerful sources of energy in the universe. Because it is the centers of these galaxies that are producing all the fireworks, astronomers focus their attention on these regions, which they call "active galactic nuclei" (abbreviated as AGN).
There are many types of AGN (for example, "quasars", "blazars", "QSOs", "Seyferts", etc.) but they all have one thing in common - they produce enormous amounts of energy in a very small region of space. How much energy are we talking about? Well, an AGN can produce more energy than 10 trillion (10^13) suns ! This is far more than an entire galaxy of stars combined - even more than a thousand galaxies! What makes this even more astounding is the fact that this energy is produced in such a small volume - only a few times larger than our solar system!
Although AGN are bright, they are also very far away (luckily for us!). Through a telescope they appear as points of light, with their host galaxy sometimes seen as faint nebulosity around the bright nucleus. Thus the details inside an AGN are too small to be seen with any telescope, and we are forced to rely on our cleverness to figure out what is happening inside the AGN "monster".
Over the past 30 years astronomers have come to the conclusion that the only way AGN can produce so much power is by a process known as "accretion". Accretion can be the most efficient means of producing energy known, far more than even nuclear fusion (which powers stars like our sun). But accretion is not possible by itself - it needs a central object to provide the necessary gravitational force. (Accretion works by converting gravitational potential energy into mechanical energy, heat and light. The efficiency of the conversion depends on the mass and density of the central object.) To power an AGN, this central object must be extremely massive and dense. "Normal" objects, such as everyday matter, will not work. Even white dwarfs and neutron stars (the super-dense end-products of stellar evolution) will not work. To power an AGN, requires something truly exotic: a super-massive black hole.
This may seem more like science fiction than real astronomy, but in fact we see many objects in our own galaxy that are powered by accretion. The central object in these systems are usually white dwarfs or neutron stars, but in several cases the central object really is a black hole. These black holes contain about 10 times the mass of the sun. But there is no reason why a much more massive black hole could not exist. And to power AGN, black holes with millions or billions of times more mass than the sun are indeed needed.
One of the most useful techniques used by astronomers takes the incoming light from the object being observed and breaks it up into its constituent colors (just as sunlight can be broken up into a rainbow of colors using a prism). This kind of work is known as "spectroscopy", and by studying the very fine details of the spectrum, astronomers can learn a great deal about the different elements and processes taking place inside the object being observed. From spectroscopy, we have learned a lot about AGN. But the internal structure inside an AGN still eludes us.
In addition to spectroscopy, another kind of tool can be used to probe the powerhouse inside an AGN. This other tool makes use of another amazing aspect of AGN: all AGN are variable. Their brightness changes constantly, and by large amounts over short periods. Time variability analysis, which studies these brightness fluctuations, can be used to make a crude estimate of the size of the AGN. This has proved to be extremely valuable - early models for AGN overestimated the size of AGN by a factor of 10. But even more impressive than determining the size of an AGN would be to determine the shape and motions inside the AGN. This can be achieved through a new technique known as "echo mapping".
The idea of echo mapping (also known as "reverberation mapping") was invented in 1982, but it wasn't until about a decade later that the idea was first put into practice. This is because only in the 1990s did the high quality data needed for echo mapping start to become available. Also, progress in computer techniques allowed this difficult problem to be tackled (technically speaking, this sort of problem is known as an "inversion problem", similar in many ways to "deconvolution" and image reconstruction).
While mathematically difficult to solve, the concept of reverberation mapping is simple and can be grasped by anyone. The ideas goes as follows: a bright flash of light is produced at the center of the AGN. This burst of light will travel out away from the center, spreading throughout the cosmos. Some small fraction of this light will be intercepted by astronomers on Earth, and a variation in the AGN's brightness will be noticed. However, if the AGN contains any internal structures, as that initial flash of light travels outwards, it will collide with these structures. These structures will glow for a short while as a result of being illuminated, and they will reflect and scatter the light back in all directions. This reflection of light will also be seen from Earth, and will appear as an "echo" of that initial flash. This echo will look similar to the flash, but will be delayed and blurred (in technical terms, the "echo function" is identical to the "point-spread function" used in optics or the "impulse-response function" in engineering).
If this is a bit unclear, the following analogy may help. Imagine you are blindfolded, sitting at the edge of a great chasm, like the Grand Canyon, with a tape recorder. Somewhere far below your colleague sets off an explosion and you hear the bang. For a few seconds after the bang you will hear the echoes of that bang, reverberating throughout the canyon. In theory, you could take that recording back home and through some sophisticated mathematics, reconstruct a rough model of the shape of the chasm, based solely on the sounds of the echoes. This is because there is a connection between the echo and the topology: the nearer an object is to either you or the source of the explosion, the sooner you will hear its echo; you will hear the echo of a distant object only after a long delay. So it is the geometry (that is, the combined distance of the object from both the source of the bang and the observer (you)) that determines the time it takes for an echo to be heard. *The geometry determines the time delay.*
Thus echo mapping is analogous to radar, sonar, and echo-location used by bats to "see" in the dark. The only difference is that the source of the "bang" is not the astronomer, but the AGN itself. In some sense, echo mapping is better described as "passive sonar", where for example, a submarine will only listen to other marine objects and not send out any pulses of its own (to avoid detection by an enemy ship).
Although the basic idea of echo mapping is relatively simple, the technique is potentially very powerful. There are of course more details and complications. In fact it should be made clear that echo mapping really is only looking at a particular region inside the AGN, the region known as the "broad line region" (BLR). This is because only the BLR part of the AGN can produce an echo (technically speaking, the BLR produces "emission lines", while most of the rest of the AGN produces "continuum" radiation. Thus the flashes are seen in the continuum and the echos are seen in the emission lines). But this alone is enough to provide a wealth of information on the workings of the central engine. Through echo mapping we can deduce shapes and motions of material which are far, far too small to be able to be seen through a telescope. This is the ultimate value of echo mapping.
Echo mapping is still in its infancy, but progress is being made. Improvements to the mathematical techniques and computer programs are continually being made, and early results are extremely encouraging. The Seyfert galaxy NGC 5548 has been one of the prime targets for echo mapping because it is relatively bright (V=13.7 magnitude) and is known to be highly variable. NGC 5548 happens to be nearby, cosmologically speaking: it has a redshift of z=0.017 which translates into "only" about 225 million light years away, just over 100x further away than the Andromeda galaxy!). One major discovery made by using echo mapping is that in NGC 5548 the BLR structures do not emit light equally in all directions - they tend to reflect the light back towards the direction from which it came. This "anisotropy" tells us that the BLR must consist of gas that is much denser than previously thought (astronomers would say the gas clouds are "optically thick").
At the moment, the biggest difficulty for echo mapping is getting enough high quality data. Because AGN are faint, big telescopes are needed. But more importantly, the AGN must be monitored carefully for months and months in order to catch the AGN making a flash and seeing its light echoes. We cannot predict ahead of time when an AGN will suddenly vary. Previous work has shown that these variations are actually quite slow - the "flash" itself will last for perhaps a month, and the echo will be delayed by as much as three weeks! This is due to the large sizes involved - although miniscule by astronomical standards, the BLR is several light-weeks across. Also, these flashes overlap (the next one begins before the previous one has finished) making the analysis more complicated, and again requiring more high quality data to un-tangle the echoes. Telescope time is extremely precious and there is fierce competition to be able to use the larger telescopes; monopolizing a large telescope for months would deny its use to other astronomers investigating other wonders of the Universe. Hence many compromises are made to ensure fairness. To carry out the difficult task of obtaining and calibrating the data, large numbers of astronomers need to work together as a team. And of course, bad weather is always a factor. But the data are slowly coming in. The big questions of whether the material in the BLR is falling in, being blown away, or orbiting inside the AGN have not been answered, nor has the ultimate question of how massive is the black hole monster powering the AGN been answered. But they will be!
Dr. William F. Welsh
Dept. of Physics
Keele University
Keele, Staffordshire ST5 5BG
UK