Simulations based on a month of radar data from the Patrick Air Force Base radar in Florida give the following results for the estimation of daily rainfall amounts over the 124,000 km2 area covered by the radar: (1) 30-min sampling by a perfect satellite sensor will increase the rainfall measurement error by 7% compared with 5-min radar sampling, (2) if the instrument is only able to determine the raining area and a good climatological rainfall rate is available, then the measurement error is increased to 35%, and (3) exceptionally dense gage networks are needed to estimate daily convective rainfall (for example, a network with 625 km2 per gage would be required to equal a minute rain-area-only technique). For monthly areal mean rainfall, the proposed Tropical Rain Measuring Mission (TRMM) sampling strategy with a perfect satellite sensor give errors of the order of 22%. A 30-min rain-area-only technique combined with a good climatological rainfall rate yields an error of at least 32%. It would seem therefore that the main contribution of TRMM could be to provide good estimates of the mean climatological rainfall rate (given that it is raining), which could then be used with the geostationary weather satellite to provide the required monthly area rainfall estimates. ¿ American Geophysical Union 1990