Abstract
We have examined the impact of the implementation of Selective Availability (SA) on GPS satellites on daily relative position estimates of the Permanent GPS Geodetic Array (PGGA) sites in southern California. The PGGA became operational in February 1990. Data are collected daily at Scripps Institution of Oceanography (SIO) and the Jet Propulsion Laboratory (JPL) from three Rogue SNR-8 receivers deployed at JPL, SIO, and Pinon Flat Observatory, and from three Mini-Mac 2816AT receivers at the CIGNFT tracking stations in California, Massachusetts, and Florida. The CIGNET stations are used as fiducial satellite tracking sites. There are two chief components to selective availability, 1) a pseudo-random clock dithering in the GPS carrier phase that is usually between 1-2 cycles per second. 2) Falsified broadcast ephemerides from the GPS satellites. Our study examined the effects of SA upon interferometric baseline measurement accuracy during the first application of selective availability from April to September of 1990. We found that as long as the simultaneity of observation time tags is maintained between receivers, SA alterations of the GPS signals do not degrade positioning accuracy when using several hours of doubly-differenced phase measurements. In fact, adding observations of the SA affected Block II satellites improved baseline repeatability (even in California where Block I geometrical coverage was quite favorable). However, we have learned that in 1990 SA had very adverse effects on networks with non-simultaneousobservations and made data editing more problematic.
