12.4.4. Intensity Variations
Many of the compact sources show pronounced intensity variations on time scales from less than a week to a few years. There is no simple pattern to the observed variations; in particular, there is no evidence for any periodic phenomena. Rather the variations appear as bursts, first at short wavelengths, and then at longer wavelengths with reduced amplitudes, with the duration of each burst being longer at the longer wavelengths. Below some critical wavelength, the amplitude, shape, and time of occurrence of the burst are independent of wavelength. Often the duration of a single outburst is comparable to the time between outbursts so that the individual events are not resolved. Figure 12.7 shows the relatively rapid intensity variations observed at three wavelengths in the source BL Lac (VRO 42.22.01). Slower, more typical variations are found in the radio galaxy 3C 120, shown in Figure 12.8.
Quasars and the nuclei of some galaxies also vary at optical wavelengths. Although there appears to be no detailed relation between the intensity variations seen at radio and optical wavelengths, those sources which are most active at radio frequencies are also usually prominent optical variables.
Figure 12.8 Variations in the intensity of the radio galaxy 3C 120 observed at 2, 6, 11, 22, and 40 cm. Three distinct outbursts are apparent in the 2-cm data. These bursts occur at later times and with reduced amplitude at the longer wavelengths.
There is no obvious difference in the pattern of the radio intensity variations seen in radio galaxies and quasars. The change in radiated power observed in the radio galaxies is typically of the order of 1042 ergs/sec/yr. For the quasars, if they are at cosmological distances, the change is very much greater and may be as much as 1045 ergs/sec/yr, e.g., comparable with the total radio luminosity of the strongest radio galaxies such as Cygnus A.
Variations are also observed in the polarization of the compact extra-galactic sources. These measurements are very difficult since the observed polarization is typically only a few percent, and so the experimental uncertainties are large. The limited data indicate that the most rapid changes in polarization occur when the total flux is increasing or is near a maximum (e.g., Aller, 1970).