Over land ice and land, satellite altimeters provide valuable topographic information, in spite of having been designed primarily to operate over the ocean. There is a need, however, for careful data quality assessment and screening as erroneous elevation measurements can be included within the telemetered data, especially when the range tracker encounters complex echoes, or rapidly varying topography. The ERS-1 Fast Delivery (FD) data product provides an excellent source of near real-time data, which has already been used for ice sheet mapping. This is a reduced data set consisting of on-board parameters generated once per second. In this paper we show that by applying thresholds to two parameters, significant amounts of poorly tracked data can be elminated. The effectiveness of this filtering technique is demonstrated by a comparison of filtered and unfiltered altimeter data with a digital elevation model. This filtering technique is applied to the first 35 day repeat cycle of FD data obtained over land to produce the first map of global topography from ERS-1. ¿ American Geophysical Union 1994