ULTRALUMINOUS INFRARED GALAXIES AND THE RADIOOPTICAL CORRELATION FOR QUASARS

Through analysis of available optical spectrophotometric data and radio flux density measurements in the literature, it is demonstrated that a good correlation exists between the radio power and bolometric luminosity of the optically-selected QSOs in the Bright Quasar Sample (BQS) of Schmidt and Green (1983). This correlation, noted previously by others as a correlation with absolute B-magnitude, is shown to be robust, and to be independent of a variety of assumptions used in the calculation of the bolometric luminosity. The correlation is present for the entire BQS sample, but is improved when QSOs with high values of radio-to-optical flux density (radio-loud) are excluded. Using this correlation, radio measurements can therefore be used to predict the bolometric luminosity of quasars even if their optical and UV continua are not directly observable. We have recently used VLBI meausurements of a sample of ultraluminous infrared galaxies to infer the likely existence of radio-quiet AGNs deeply enshrouded in dust within their nuclei (Lonsdale, Smith, and Lonsdale 1993). We employ the radio-bolometric luminosity correlation for the BQS quasars to test whether these hypothetical buried AGNs can be energetically responsible for the observed far-infrared luminosities of the ultraluminous infrared galaxies. The ultraluminous infrared galaxies are shown to follow the same relation between radio core power and bolometric luminosity as the radio-quiet QSOs, suggesting that buried AGNs can account for essentially all the observed infrared luminosity, and raising the possibility that any starburst which may be in progress may not be energetically dominant. The broader implications of the radio-optical correlation in quasars for AGNs and luminous infrared galaxy models and the use of radio astronomy as a probe of the central powerhouse in radio quiet AGNs and luminous infrared galaxies are briefly discussed.