Techniques for Reducing Spatially-Correlated Errors in High-Rate GPS Positioning

Even with continuing advances in GPS data analysis techniques and receiver technology, sub-daily GPS position time series exhibit a high noise level relative to the desired geophysical signal (e.g. volcanic uplift, seismic waves). Several sources of positioning noise are spatially-correlated, meaning that these errors depend on the position of a GPS satellite relative to the location of the receiving GPS antenna. Examples of spatially-dependent errors include multipath, residual phase center variations, and positioning precision based on satellite constellation geometry. We present here two different but complementary methods for removing spatially-correlated errors, sidereal filtering and signal-to-noise ratio (SNR)-based multipath corrections. By applying these methods to both high-rate (1 Hz) and low-rate (30-second) GPS data, we analyze and discuss the respective differences in position noise reduction.