AGN Variability

Gordon G. Spear

June 2001


Description of the Variability

Variability Time Scales

Longterm

Intraday

Microvariability

What Observations are Needed?

Longterm Goals

Intraday Goals

Microvariability Goals

 


 

Description of the Variability

 

One of the general characteristics of Active Galactic Nuclei (AGNs) such as quasars, blazars, and Seyfert galaxies is that they seem to be variable in brightness. It is often assumed that most, if not all, AGNs are variable at some level. For example, virtually all of the AGNs observed in the Hubble Deep Field were determined to exhibit detectable variability over a period of two years. Furthermore, some categories of AGNs (Blazars, BL Lacs, OVVs) are among the most extremely active variable sources known in the universe. Only supernovas, novas, and some longperiod variables (LPVs) have light ranges greater than the ranges exhibited by the violent AGNs. And, while it is still highly controversial, the most dramatic outbursts yet observed in the universe, the Gamma Ray Bursters (GRBs) may simply be extreme examples of the violent AGN phenomena.

 

Unlike the highly structured supernova outburst which happens only once or a nova outburst which may repeat after decades or centuries, AGN variability is highly unstructured. A nova outburst appears as a rapid increase in brightness followed by a slow decline, ultimately reaching a stable quiescent state which may be below detectable limits. AGN variability shows no such structure or regularity. AGN variability is highly irregular.

 

In general, brightness levels slowly increase and decrease in an irregular, non-periodic fashion. Superimposed on this variation there can be occasional outbursts of varying durations and magnitudes. Outbursts tend to repeat on an irregular basis and each individual outburst tends to have a similar rate of change on the rising and declining branches. The largest outbursts seem to occur over months or years, but this may simply be a selection effect in the available data.

 

This description seems appropriate for all time scales from years and decades to minutes and hours. This description also seems appropriate for all energy ranges from gamma-rays through the visible and into the radio frequency region. Variability seems to be correlated over all energy ranges, but generally increases in amplitude at higher energies. There are some indications that events first appear at lower energies and then propagate with some delay or lag to higher energies. The lags are typically less than one day.

 

There are often attempts made to distinguish between two types of variability. There are the violent variables (VV… blazars, BL Lacs, OVV quasars) and the more quiescent variables (QV… normal quasars, Seyfert galaxies). While it is true that the violent variables have greater ranges in brightness and can vary significantly over shorter time periods, there may simply be a continuous range in this variability intensity. The above general description of AGN variability does indeed apply equally to the violent and to the quiescent variables.

[ top of page ]

 

 

Variability Time Scales

 

Variability studies of AGNs typically focus on one of three time scales. These time scales are related to the nature of the data and to the logistics and practical matters associated with obtaining astronomical observations. These time scales are not necessarily related in any meaningful way to physical process producing the variability. These time scales may be termed longterm, intraday, and microvariability.

 

Time Scale
typical time scale for viewing the data
typical time resolution
longterm
several years or decades
months or years
intraday
several days or weeks
days
microvariability
several hours
minutes

 

 

Longterm

 

Historically, longterm studies have been based on estimates obtained from photographic plate collections. In some cases data can be available for nearly a century, but the coverage is random and can have significant gaps due to economic and political circumstances. Data will normally be based on different photographic materials and processes plus different instrumentation (cameras and telescopes). Seeing conditions and focus will also vary and the limiting magnitude reached can vary substantially from one plate to the next. At best the precision of the data can be a few tenths of a magnitude, but measurement uncertainties of a half magnitude or more are certainly possible. Any detected variability can be expected to be decidedly undersampled.

Longterm observational programs are always difficult to maintain. The difficulties are based on the commitments required in interest, in time, and in funding. Professional research astronomers may have productive careers that extend several decades, but they are expected to produce publishable papers on a yearly basis and to obtain grant funding. Traditional funding agencies rarely award grants that extend for more than a few years at best.

Contemporary observational programs that can produce homogeneous, high precision data that is well sampled are certainly technically possible. However, it is likely that such programs will need to be achieved by dedicated individuals and smaller institutions. Dedicated amateur astronomers could also make a substantial contribution to such studies. A collection or network of such observers would produce the most effective results. Robotic telescopes and observatories can, of course, minimize the physical hardships associated with obtaining the data and can help insure a higher level of uniformity. Such programs could finally produce data that has been adequately sampled.

Several of the large scale survey programs that have been proposed by major observatories and institutions, in conjunction with programs such as the National Virtual Observatory could ultimately provide a uniform homogeneous database for significant longterm studies of AGNs. Until the time that such large scale programs are approved and initiated, a small scale network of dedicated observers can make truly significant contributions to the study of the longterm variability of AGNs.

 

 

Intraday

 

Intraday studies of the variability of AGNs essentially compare the brightness of an AGN from one night to the next for a period of days. Such programs typically involve the intensive observation of a selection of a dozen or so objects for five or ten days. This can be the extent of a typical allocation of observing time on a medium-sized telescope. It is not uncommon for such telescope allocation to be repeated in six months or a year. This can provide a target sample distributed around the sky and can provide follow up observations for targets previously observed.

Unfortunately, observing intraday variability for only a few days at a time will grossly undersample the true extent of such variability. For example, it is well known that AGNs undergo active phases and quiescent phases. Intraday variability would be expected to be different during these different phases and a few isolated observations may not be sufficient to indicate the true activity level.

A few dedicated programs using small telescopes have attempted to monitor a selection of objects at daily or weekly intervals for a few years. Even for such data, gaps remain due to weather systems and the phase of the moon. Such programs have never succeeded in approaching the goal of a time resolution near one day. The effective average time resolution for such limited programs is more likely near 5 or 10 days.

Coordinated observations by a cooperating network of small observatories which include amateur astronomers and robotic telescopes can dramatically improve our knowledge of the true extent of intraday variability for AGNs. Such observations could significantly extend the time spans for intraday data.

 

 

Microvariability

 

Microvariability studies involve sitting on a single AGN for several hours and taking data as fast as possible with the available instrumentation. Using modern CCDs with medium and small telescopes, a large number of AGN are accessible with exposure times less than a few minutes. Thus, the brighter AGNs may be observed with a time resolution of about one minute. Of course, such a program requires the exclusive use of a telescope for a single object. When telescope time is limited or competitive, it is often difficult to justify the allocation of large blocks of time for such programs.

Since microvariability has been confirmed for vitrually all classes of AGNs, such observing programs can be exciting. Flares can occur over a period of hours or less, and rapid qusi-periodic oscillations have also been observed. Variability over periods of hours has been well documented for numerous AGNs. However, variability is not guaranteed for any particular object or on any particular night. One must be prepared to expend several nights of taking data which reveal no significant variations before one "gets lucky." All published descriptions of microvariability typically report that several objects were observed without positive results, and that even objects exhibiting variability on one night may have been monitored on several other nights with no indications of variability.

Microvariability studies can be attractive to professionals since for a limited commitment of time, if one is lucky, one is virtually guaranteed a publication. Do you feel lucky?

Unfortunately, since the probability for success is not high, microvariability proposals tend not to be highly ranked by telescope allocation committees when awarding telescope time at major observatories. Such programs tend to take place at smaller institutions where telescope time is not competitive.

To what extent can we expect microvariability for a particular class of AGN? Under what circumstances is microvariability more likely for a particular object? Blazars tend to be more variable than other classes of AGNs, but over very short time intervals we really do not yet know for sure. Accumulating the necessary data is very expensive in terms of telescope time. Here again, coordinated observations by a cooperating network of small observatories which include amateur astronomers and robotic telescopes can dramatically improve our knowledge of the true extent of microvariability for AGNs.

 

[ top of page ]

 

 

What observations are needed?

 

It is well documented that AGNs are variable. It has been suggested, but certainly not confirmed, that all AGNs are variable at some level. Variability studies have generally concentrated on the longterm, the intraday, and the microvariability levels. However, longterm studies are always undersampled and tend to have substantial gaps in coverage. Intraday and microvariability studies tend to be available only for extremely limited time spans At all levels, objects are generally selected for study because they attract our attention for some reason. For example, 3C 273 has been extensively studied because it was one of the first quasars to have an optical identification. BL Lac has been extensively studied because it was originally identified as a variable star, and because it is subject to extremely large variations. OJ 287 has been extensively studied because it has been systematically recorded photographically for more than 100 years, has undergone many well documented outbursts, and because it appears to be continuously variable at any time over periods extending from several minutes to several months. Mrk 501 has been studied because it was detected as a source of TeV gamma-rays. Objects are often observed extensively after they have inadvertently been detected to undergo an outburst or become more active either in visible light or in the radio region. Systematic studies of a broad selection of objects are rare. Based on the limitations of the existing data and analysis, some systematic observational material would substantially improve our understanding of the nature of the variability of AGNs.

 

In general, the goals for such systematic studies would be to obtain homogeneous data using CCDs with small to moderate telescopes. Standard filters should be adopted along with uniform reduction techniques. A group or network of cooperating observatories and robotic telescopes could be particularly productive. The image data should be archived in an accessible manner. Archival storage of the original images is extremely important to insure that the data remains homogeneous as new data is added. Furthermore, new approaches to data reduction can produce new insights if the original raw data is available for analysis.

 

Broad longterm studies using modern data are rare. Broad intraday and microvariability studies for a range of objects are extremely rare. Intraday studies are normally available only over limited time spans and normally only for objects exhibiting unusual activity. Microvariability studies are only available for very limited time spans for objects expected or known to be active for some reason. Repeat microvariability studies for specific objects are extremely unusual.

 

Longterm Goals

 

The goal should be to achieve a time resolution of 0.08 years for a number of years. Multiple filter observations for a broad range of objects should be obtained at least once a month. A relatively small collection of observatories could produce an impressive collection of data with a time resolution on the order of 0.1 years. Any unusual variations would be reported and thus serve as a trigger for more intensive categories of observation.

 

 

Intraday Goals

 

The goal should be to establish targeted campaigns to achieve a true time resolution of 1 day for a period of several weeks. Objects exhibiting unusual variation in the longterm studies, or objects selected as a result of other observations, or objects selected for programatic reasons, can be placed on special focused campaigns. For such a campaign, cooperating observatories will attempt to obtain an observation once a night for an extended, but finite span of time. A modest distribution of observatories would be expected to produce surveillance data with a time resolution on the order of a day. Such campaigns might extend for 4, 8, or 12 weeks.

 

 

Microvariability Goals

 

The goal should be to establish targeted campaigns to achieve a true time resolution of several minutes for a period of several days. Objects exhibiting unusual night-to-night variations, or objects selected as a result of other observations, or objects selected for programatic reasons, can be placed on special microvariability campaigns. For such a campaign, cooperating observatories with appropriate equipment will attempt to sit on a designated target nearly continuously for several/many hours at a time throughout the night. Such a campaign should extend for an extended but finite span of time. With proper planning and a uniform distribution of observatories in longitude, this should make it possible to keep a target under virtually continuous surveillance with a time resolution on the order of minutes. Such campaigns might extend for 5, 10, or 15 days.

 

[ top of page ]

 

[ AGN Projects at SSU ]


ggs
Version 1.1
20 June 2001