Low frequency measurements of the cosmic microwave background anisotropy are significantly contaminated by the thermal bremsstrahlung and synchrotron radiations emitted by ionised gas in our galaxy. However, we have only limited information on either the frequency spectrum or the spatial distribution of this emission, and nowadays the sensitivity of CMB experiments is high enough that this lack of knowledge of the galactic signal is potentially a serious hindrance. On larger angular scales, greater than a few degrees, the low frequency radio surveys provide some information, as do the COBE DMR skymaps, but on subdegree scales there is indeed almost no information as to what to expect. This problem will become more acute when subdegree scale experiments such as the VSA, MAP, and Planck Surveyor come online. These experiments plan to measure the CMB angular power spectrum, and determine cosmological parameters, with a few percent precision, but of course, this is only possible if the galactic contamination can be modelled with similar precision.
Here we present some results on the spatial distribution of the galactic free-free emission at intermediate and high galactic latitudes. Free-free emission is the dominant foreground at 25--75 GHz, and its frequency spectrum is very well known. On large scales, Bennett et al. (1992) and Bennett et al. (1994) have constructed a map of the free-free emission by linearly combining the four-year COBE DMR skymaps at different frequencies to isolate emission with antenna temperature . Kogut et al. (1996a) and Kogut et al. (1996b) compare this emission with dust emission as mapped by the far infrared channels on COBE DIRBE and find that the free-free and dust distributions are correlated. This correlated distribution has an angular power spectrum of the form for angular scales . The power spectrum scales in the same way as the power spectrum measured for IRAS 100 fields on scales to (Gautier et al. (1992), Gautier & Stewart (1993)) and for neutral hydrogen by Crovisier & Dickey (1983).
Diffuse galactic emission is considered to be a good tracer of the diffuse free-free emission because both are emitted by the same ionised hydrogen (Reynolds (1980)). Both scale with electron density in the same way, . An intensity of 1 Rayleigh (1 Rayleigh at the line at 6563Å) implies a free-free antenna temperature of at 53 GHz, and at 5 GHz. We have used publicly available maps of the north celestial pole to determine the spatial distribution of galactic free-free emission on subdegree scales. Our main conclusion is very pleasing -- galactic bremsstrahlung is not going to be a serious contaminant for subdegree scale CMB measurements.