Although galactic emission is correlated with the galactic free--free emission, it is not straightforward to estimate the galactic free--free from maps of the sky in . These maps are themselves contaminated with atmospheric emission! The earth's geocorona emits radiation: hydrogen atoms in the earth's exosphere are excited by solar Lyman photons, and about of the excited atoms return to ground state through emission of photons (Shih et al. (1985)). The geocoronal emission varies both diurnally and seasonally with the solar Lyman flux, and the amplitude of this variation ranges from about 2 to 25 Rayleigh. In principle, the geocoronal and galactic emissions are separable, since the centre of velocity of the geocoronal line is generally shifted with respect to the galactic line due to relative velocity of the Earth with respect to the local standard of rest (LSR), but such separation requires spectral measurements (e.g. Reynolds (1980), Reynolds (1992)), not simply imaging. The maps which we use for the analysis below do not have any spectroscopic information.
Gaustad et al. (1996) and Simonetti et al. (1996) have mapped the emission around the North Celestial Pole with resolution of a few arcminutes, and used these maps to predict the level of free--free contamination of the Saskatoon results, i.e. on angular scales above a degree. Gaustad et al. (1996) have generously made their maps publicly available. These maps are a series of images mosaicked to cover the entire NCP region north of declination . Each image has an effective resolution of and field of view of about . This set of images ought to provide a convenient data set with which to measure , and therefore, free-free fluctuations on the few-arcminute angular scale.
Several CMB experiments are being conducted in the NCP region (e.g. Netterfield et al. (1995): Saskatoon, Myers et al. (1993), Leitch et al. (1997): OVRO Ring, Cheng et al. (1994): MSAM) even though this region is not ideal for CMB measurements. Firstly, it is at relatively low galactic latitude (the NCP is at ). Moreover, there is enhanced IRAS 100 emission in this region as compared to the emission in other regions at similar latitude, and this emission has been identified as the north polar cirrus cloud by Gautier et al. (1992). The 100 emission does not appear to correlate well with HI. We expect that, in more quiescent regions of the sky, and in particular at higher galactic latitude, the amplitudes which we calculate for the free--free emission may represent upper limits.