Selective Availability, which involves destabilizing the satellite oscillators and altering broadcast orbit and clock parameters, should have no effect on precise GPS geodesy. The differential solution removes oscillator errors and either estimates the orbits and clocks directly or obtains them from other sources. This all works as expected so long as care is taken to ensure that the data samples at different receivers are taken within 10 or 20 msec of a common transmit time. Where that is not the case in the original data it can often be attained by later adjustment. Wu et al [1992] describe a simple interpolation scheme that is effective under a wide range of conditions. Feigl et al [1991] describe another, somewhat more laborious approach.
Anti-Spoofing is a different matter. All current geodetic receivers produce a degraded L2 phase observable under AS, which in theory could lead to degraded geodetic solutions. In many applications, however, atmospheric variation, not instrumental precision, is the limiting factor and data corruption under AS has little or no effect. Lindqwister and Meehan [1992] achieved 2--3 mm horizontal repeatability over a 50-km baseline with SNR-8000 receivers in AS mode, or about the same as in code mode. Results with modern receivers in the PGGA show little loss of precision since AS went on full time in early 1994. Although the daily precision of GPS orbits and global site solutions degraded by 30--100% at the onset of AS, by late 1994 performance was back at (or even better than) pre-AS levels, as a result of intensive tuning efforts [ Jefferson et al, 1994]. Like multipath, AS will most directly affect kinematic, rapid static, and other users employing short (<30 min) averaging times. Even then, multipath and atmospheric effects (which are comparable to AS data noise) will tend to cushion the effect of AS error.