COBE DIFFERENTIAL MICROWAVE RADIOMETERS - PRELIMINARY SYSTEMATIC-ERROR ANALYSIS

The Differential Microwave Radiometers (DMR) instrument aboard the Cosmic Background Explorer (COBE) maps the full microwave sky in order to measure the large-angular-scale anisotropy of the cosmic microwave background radiation. Solar system foreground sources, instrumental effects, as well as data recovery and processing, can combine to create statistically significant artifacts in the analyzed data. We discuss the techniques available for the identification and subtraction of these effects from the DMR data and present preliminary limits on their magnitude in the DMR 1 yr maps (Smoot et al. 1992). The largest effect is the instrument response to the Earth's magnetic field, which contributes up to 375 muK to the raw data in the worst channel. Emission from the Earth is weak (less than 47 muK in the raw data at 95% confidence). Residual uncertainties in the best DMR sky maps, after correcting-the raw data for systematic effects, are less than 6 muK for the pixel rms variation, less than 3 muK for the rms quadrupole amplitude of a spherical harmonic expansion, and less than 30 muK2 for the correlation function (all limits 95% confidence level). These limits are a factor of 5-40 lower than the level of anisotropy in the microwave background detected in the 1 yr DMR sky maps: 30 +/- 5 muK rms, 13 +/- 4 muK quadrupole, and 1194 +/- 499 muK2 for the correlation function (Smoot et al. 1992; Bennett et al. 1992b; Wright et al. 1992).