HR: 09:40h
AN: G51A-07 [PDF]
TI: Utilizing Site-Specific Antenna Phase Center Calibration Maps for Reducing the Noise in the GPS
Vertical
AU: * Morken, D M
EM: Morken@Colorado.EDU
AF: Colorado Center for Astrodynamics Research, University of Colorado at Boulder, Campus Box 431,
Boulder, CO 80309 United States
AU: Haines, B J
EM: BJH@Cobra.JPL.NASA.GOV
AF: Jet Propulsion Laboratory, 4800 Oak Grove Drive, Pasadena, CA 91109 United States
AU: Nerem, R S
EM: Nerem@Colorado.EDU
AF: Colorado Center for Astrodynamics Research, University of Colorado at Boulder, Campus Box 431,
Boulder, CO 80309 United States
AB:
The Global Positioning System has emerged as a leading tool in the field of geodesy. GPS observations have been shown to be
a useful geodetic tool for such applications as monitoring post-glacial rebound and calibrating satellite altimeters to study
sea level change. We have been analyzing data from 40 GPS sites near tide gauges for the purpose of developing vertical
crustal motion estimates, which are subsequently used in the calibration of satellite altimeter measurements. However,
site-specific errors such as multipath have limited the positioning accuracy, especially in the vertical. To combat these
site-specific errors, antenna phase center calibration maps were developed by averaging the residuals over the time series of
each point-positioned GPS site into bins 2 x 5 degrees in elevation and azimuth. No elevation cut-off was used for the
purpose of exploring the effects of the maps on observations made at very low elevation angles. This was done separately for
pseudorange and carrier phase. In addition, each antenna used at a given site had a unique map. This technique was
iterated upon several times, using every day in the time series, until the variations were sufficiently captured. It will be
shown that the antenna phase center calibration maps dramatically reduced the elevation dependence of the point-positioned
solutions, especially between 0 degree and 10 degree elevation cut-offs. For example, at MacQuarie Island (MAC1), the
difference in estimated trends over the 0 - 10 degree elevation cut-offs varied by as much as 3.5 mm/yr without using the
calibration maps, whereas the maximum difference with the maps was 0.5 mm/yr over the same range. Also, there was a
significant reduction in the vertical repeatibility of 1 - 2 mm in most cases. A list of new estimated trends at
approximately 40 sites, as well as a thorough analysis of the associated errors will be shown.
DE: 1223 Ocean/Earth/atmosphere interactions (3339)
DE: 1241 Satellite orbits
DE: 1709 Geodesy
DE: 1724 Ocean sciences
DE: 4556 Sea level variations
SC: Geodesy [G]
MN: 2003 Fall Meeting