Short Time-Scale Variability in the Faint Sky Variability Survey Short Time-Scale Variability in the Faint Sky Variability Survey
ING Banner
Home > Public Information > Scientific Highlights > 2006 > Short Time-Scale Variability in the Faint Sky Variability Survey

Short Time-Scale Variability in the Faint Sky Variability Survey


There is a wide range of photometrically variable systems in the Universe. The range of time-scales on which these systems vary is as wide as the physical processes that produce their variability. For example, we have intrinsically variable stars, where the variability is caused by changes in their internal structure or atmosphere that vary with time-scales of minutes to years. Other stars show variability because they rotate and their surface is inhomogeneous, for example, because of star spots, or because they form part of a binary or multiple system and their revolution around the centre of mass of the system results in changes on the detected flux due to the changing aspect of a non-isotropically emitting surface or eclipses.

This is also the case for planets orbiting stars. The time-scale of the variability in this case is dictated by the orbital parameters of the system and can range from seconds to years. Near-earth objects (NEOs), such as asteroids, also show variability as they rotate and are non-spherical.

We find photometric variability in extragalactic objects as well, such as quasars, where the variability is probably the result of material being accreted by the central engine, or ‘one of’ systems, such as gamma-ray bursts (GRBs) or supernovae (SNe) where the variability is produced by intrinsic changes in the structure of an astronomical object that take place only once.

The study of variability provides important information about the physical nature of the variable objects, leads to the discovery of new classes of objects, helps to study the physical structure of stars, for example, pulsating stars, allows us to obtain information on Galactic structure through the use of variables such as RR Lyrae as standard candles, and is the key to determining extragalactic distances through the use of standard candles, such as Cepheids and Type Ia SNe.

Most of our knowledge of variability is based on the study of apparently bright sources, which naturally selects members of intrinsically bright populations. At present, little is known about variability of intrinsically fainter populations, because in bright samples they are lacking altogether or are only represented by a few members.

The Faint Sky Variability Survey (FSVS) was one of the programmes of the INT Wide Field Survey and it was designed to account for this deficit by studying two unexplored regions of the variability space: the short time-scale variability region (down to tens of minutes) and the intrinsically faint variable sources (down to V = 24 mag) at mid and high Galactic latitudes. The FSVS also contains colour information for all targets, giving us the option of positioning objects in the colour–colour diagram, as well as finding the variability time-scales and amplitudes that characterize them. The main aims of the FSVS are thus to obtain a map of a region of the Galaxy (~21 deg2) in variability and colour space, to determine the population density of the different variable objects that reside in the Galaxy and to find the photometric signature of up-to-now unknown intrinsically faint variable populations.

The full FSVS data set consists of 78 Wide Field Camera (WFC) fields, covering an area in the sky located at mid and high Galactic latitudes (-40 < b < -21, 15 < b < 50, 89 < b < 90). For each field, they took one set of B-, I- and V-band observations on a given photometric night. Photometric variability observations were taken with the V filter on several consecutive nights. On average, fields were observed 10–20 times within one week in the V band. Exposure times were 10 min with a dead time between observations of 2 min. This observing pattern allows to sample periodicity time-scales from 2× (observing time + dead time) (i.e. 24 minutes) to twice the maximum time-separation of observations (which ranges from 3 to 13 days). All fields were re-observed years later to determine proper motions.

Period and amplitude distribution in the FSVS survey
Histograms showing the cumulative period distribution (left-hand panels), the period distribution (middle panels) and the amplitude distribution (right-hand panels) for the short-term variable point sources in the FSVS. Top panels: the 689 out of 744 sources with accuracies in the periods and amplitudes of the order of 30 per cent or less. Bottom panels: sources where the error in their periods and amplitudes is of the order of 10 per cent or less.(Extracted from Morales-Rueda, 2006, MNRAS, 371, 1681). [ GIF ]

A V-band variability analysis of the point sources in the FSVS on time-scales from 24 min to tens of days showed that about one per cent of the point sources down to V = 24 are variables. Of these, about 50 per cent show variability time-scales shorter than 6 h, 22 per cent show variabilities between 6 h and 1 d, 20 per cent between 1 and 4 d and 8 per cent show periods longer than 4 d. The total number of variables is dominated by main sequence sources. The distribution of variables with spectral type is fairly constant along the main sequence, with 1 per cent of the sources being variable, except at the blue end of the main sequence, between spectral types F0 and F5, where the fraction of variable sources increases to about 2 per cent. For bluer sources, above the main sequence, beyond the blue cut-off at (B-V)<0.38, this percentage increases to about 3.5. About a third of the total number of short time-scale variables found in the survey were not detected in either B or/and I band. These show a similar variability time-scale distribution to that found for the variables detected in all three bands.

The observed variability is likely to be the result of pulsations, binarity and stellar rotation as well as intrinsic flaring activity in the case of main sequence stars. One RR Lyr candidate found is the furthest known in the Galaxy and could allow the determination of the mass of the Milky Way.

The highest space densities of variables found in the FSVS (i.e.17 deg-2) show periods below 12 h. These include CVs, RR Lyr stars, and other short-period pulsators, such as δ Scuti stars. They find a density of four variables per deg2 centred on a 1 d period which includes longer-period CVs, RR Lyr and other pulsators, like γ Doradus stars and Population II Cepheids. A space density of two variables per deg2 at 3.75 d includes some longer-period CVs, γ Doradus stars, Population II Cepheids and longer-period pulsators, such as subdwarf B stars. At 12.75 d, they also find two variables per deg2. These would be mainly binaries with those orbital periods and Population II Cepheids.

The FSVS is complete down to V =22 for CVs in the minimum period (80 min) as long as they show variability amplitudes of the order of 0.4 mag. It is complete down to V=22 for periods between 80 min and 1 d in a 17.82-deg2 area of the survey as long as the amplitude of the variability is at least 0.7 mag. This includes most RR Lyr stars. 1.5 mag.


Top | Back

Contact:  (Public Relations Officer)
Last modified: 13 December 2010