Spectral Analysis of the B3 VLA Survey

Abstract

The B3VLA sample (Vigotti et al. 1989) is a subset of 1050 sources selected from the B3 survey (Ficarra et al. 1985) in 5 flux bins by narrowing the survey strip in declination and restricting it to high galactic latitudes so that the completeness is very carefully controlled. At present, the identifications of about 550 B3VLA sources are known, some 250 have measured redshifts. Radio continuum flux densities of B3VLA sources have been measured over a wide frequency range spanned from 151 MHz (6C) over 408 MHz (B3), 1.4 GHz (B3-VLA), 4.75 GHz (87GB and Effelsberg) up to 10.5 GHz (Effelsberg). The data analysis of the latter two frequencies is still in progress. In the following, we present preliminary results on the majority of the observed sources. Since high-frequency radio continuum flux densities are most sensitive to synchrotron losses we have used these data not only to get information on the general spectral behaviour but also to adjust synchrotron loss models to the spectra. The spectral index analysis has been performed for 403 sources according to the present state of the data reduction. Only sources with a spectral index have been considered because for these sources we are expecting reasonable fit results. Based on extensive Monte-Carlo tests of the fitting algorithm we consider break frequencies up to 20 GHz to be reliable, although the highest observed frequency is 10.5 GHz. Spectra with fitted break frequencies beyond 20 GHz are considered as pure power laws without a significant cut. A priori, one would expect the continuous injection (CI) model (Pacholczyk 1970) to be the best suitable model in case of integrated flux densities since typically particle accelerating and energy consuming regions will be present at the same time. In fact, this seems to be the case for the majority of the sources in the sample although tests with pure synchrotron loss models (Kardashev 1962, Pacholczyk 1970, Jaffe & Perola 1973) disclose better results for a part of the sources (some 30%). A small fraction of these (the one with the steepest injection spectra) could be relic sources where all particle accelerating features like cores or hot spots have been “switched off”. This point will be the subject of further investigations in near future. We have cross-correlated the values of both the injection spectral index and the break frequency in dependence on the sources’ redshifts, their identifications, and their monochromatic power. The results are as follows: