ABSTRACTThe newer Block II GPS satellites have the capability to control the accuracy of the Standard Positioning Service signals by a combination of dithering the satellite clock and manipulating the ephemeris data. This capability is called Selective Availability (SA). Differential GPS operation greatly reduces the effects of SA, in addition to the effects of ionospheric, tropospheric, and other slowly‐varying errors, for users in a local area around a reference station.A major technical issue in the design of reference stations has been the correction update rate. It has been assumed until recently that new differential GPS corrections are required about once every 10 s to maintain 5–10 m service accuracy. Until the recent launch of the new Block II satellites, there has been no firm evidence that this update rate would be adequate. This paper presents statistical analysis of recently obtained data sets from Block II satellites, specifically position and velocity variations and correlation times. Pseudorange corrections generated by differential reference station equipment using these signals are applied to the pseudoranges from a navigation receiver to obtain estimates of the effect of correction update rate on differential GPS performa