We examined variations of oceanic rain rate parameters derived from microwave brightness temperature (T_b) collected by the Special Sensor Microwave Imager (SSM/I) on board the Defense Meteorological Satellite Program (DMSP) satellites using the technique developed by Wilheit, Chang and Chiu (1991).  Monthly oceanic rain rates are estimated from the monthly histogram by fitting it to a mixed lognormal rain rate distribution via a rain rate – T_b relation derived from radiative transfer calculations based on a model.  To minimize the effect of water vapor on the rain signal, a combination channel of twice the vertical polarization at 19 GHz minus that at 22 GHz is used.  The mode of the T_b histogram, T₀, is interpreted as the background T_b in the absence of rain. This paper examines the spatial and temporal variations of T₀ using 17 years (1998-2004) of data. Analyses of global averages T₀ show distinct changes in the characteristics of the time series with respect to different DMSP satellites.  The time series of global average T₀ also shows a slight linear decreasing trend (~0.8K over 17 years) at the 95% level of significance. An Empirical Orthogonal Function (EOF) analysis is performed on the non-seasonal T₀. The first EOF pattern has the same sign everywhere and the associated time series is similar to that of the global average. The second EOF pattern shows a pattern similar to the ENSO response in precipitation, and the associated time series is correlated to an Southern Oscillation index at -0.66.  Empirical Mode Decomposition (EMD) analyses of the zonal averages show decreasing trends of about 2K and 1K between 40N-50N and between 30S-40S, respectively.  These results are interpreted in terms of the trends of oceanic precipitation.