Brightness temperature simulations were used to assess the information loss due to the coarse spatial sampling of the Tropical Rainfall Measuring Mission (TRMM) microwave imager (TMI). While the high-resolution simulation of a three-dimensional rain cloud clearly identifies the 10.7-GHz channel to be superior for surface rainfall measurements, its comparably poor spatial resolution reduces the dynamic range of observable brightness temperatures and associated area-averaged rain rates considerably. This problem can be partly overcome by applying deconvolution techniques as demonstrated by Robinson et al. <1992>. When the sensitivity of resolution enhancement to noise amplification is investigated, an optimum configuration of key parameters was identified. Using an 11¿11 matrix of neighboring pixels and noise reduction parameter of &ggr;=0.5¿ provides a 42¿26 km resolution, compared with the original resolution of 63¿38 km. Thus a noise amplification by a factor of 3 is introduced. Although this method does not match the resolution of the 19.35-GHz channel, the dynamic range of brightness temperatures and area-averaged rain rates could be increased by a factor of 2.7 and 2.2, respectively. Further resolution enhancement leads to noise levels which seem unacceptable with respect to the retrieval of geophysical parameters. ¿ 1998 American Geophysical Union