Bibcode
Deeg, H.-J.
Bibliographical reference
Ph.D. thesis, University of New Mexico, (1993)
Advertised on:
7
1993
Citations
4
Refereed citations
4
Description
A sample of actively star forming H II galaxies has been observed in the
radio continuum and several optical bands. These galaxies are currently
undergoing very active star formation and have been selected on the
basis of strong radio continuum emission. Most of the galaxies are
irregular and have been associated with merger or interaction events,
which may have triggered the star formation and the radio emission.
Radio continuum observations at 0.32 GHz, 1.5 GHz, and in the range of
8-15 GHz were obtained at the NRAO-VLA, to determine their radio
spectra. Several of the spectra were found to flatten towards lower
frequencies, which is unusual. Surface and aperture photometry was
obtained in the B,R, and I bands and in the H-alpha emission line. Radio
emission, absorption, and relativistic electron loss mechanisms are
reviewed and their suitability to account for the observed spectral
shapes is discussed. Energy equipartition calculations led to galaxy
magnetic fields of 10-30 mu-G; the radiation density inside the H II
regions has been derived by a new method and was found to be in the
range of 2 to 15 eVJcm-3. Mechanisms which may account for
the observed radio spectra were fitted to the radio continuum data. The
spectra resulting from a time variable relativistic electron injection
('synchrotron aging') have been calculated over a greater range of
parameters than previously published. Fits of these spectra show
variations of the injection rate with time scales of the order of a few
Myrs. A fit based on free-free absorption uses the emission measure to
balance both free-free absorption and thermal emission, thereby
constraining the maximum size of the thermally emitting region. This
permits a direct comparison with the optical H-alpha observations which
typically show H II regions with sizes between 0.5 and 1 kpc. The fits
allowed the derivation of a variety of physical parameters, such as the
electron density, which is typically between 10 and 60 cm-3,
and the emission measure, which is of the order of a few 105
pc cm-6. Fits of different mechanisms may describe the same
spectrum well, thus, in some cases, a unique determination of a
particular mechanism is not possible, although an estimate for its
likelihood can be made. Parts of this work have been published by Deeg
et al. (1993, ApJ 410, 626). The fraction of thermal emission in the
galaxies has been determined with a higher reliability than usual by
combining radio, H-alpha, and UV data. A correlation between the
fraction of thermal emission and the size, as well as the total
luminosity of the galaxies has been found. A simple model of
relativistic electron diffusion losses-dependent on a galaxies' size-can
reproduce the observed correlation well; its suitability and limits are
discussed. This correlation may lead to a better understanding between
the supernova rate in a galaxy and the expected non-thermal emission, a
relation which is currently only very poorly known. Star formation rates
based on thermal and nonthermal radio and FIR emission all indicate star
formation which is enhanced during the last 10^6-7 years as compared to
the long term (1 Gyr) rates based on B band photometry. `Synchrotron
aging', optical colors and the thermal to FIR ratio were used to derive
typical ages of star formation. Based on the star formation rates and
the age indicators, the galaxies were sorted into an sequence of their
starbursts ages. The physical picture of a region in which star
formation, subsequent SN explosions, and the resulting nonthermal radio
emission takes place, can be accounted for well, by comparing the
different star formation estimators which are based on a variety of
radiative processes and across three regions of the electromagnetic
spectrum.