THE USE OF GPS DATA
FOR GLOBAL GRAVITY FIELD DETERMINATION
Byron D. Tapley, Hyung-Jin Rim, John C. Ries, Bob E. Schutz, and C. K. Shum
University of Texas Center for Space Research
The University of Texas at Austin, Austin, Texas 78712 USA
INTRODUCTION
GPS tracking data collected by a low-Earth orbiting (LEO) satellite provides global and
continuous coverage of the orbit. Consequently, this data set contains valuable
information for both precise orbit determination (POD) and gravity model improvement.
The GPS experiment on TOPEXIPOSEIDON (TIP) presents the first opportunity to use
GPS data for POD (Bertiger et al., 1994; Schutz et aI., 1994) and global gravity field
development.
The JGM-3 model (Tapley et al., 1994a) is the first gravity model which includes the
GPS data. JGM-3 model was determined by combining the JGM-l information array
(Nerem et aI, 1994) with new information from laser, DORIS, and GPS tracking of TIP,
laser tracking of Lageos-l, Lageos-2, and Stella, and additional DORIS tracking on
SPOT -2. The GPS tracking data used for the JGM-3 model consists of GPS
Demonstration Receiver (GPSIDR) data from Cycles 10, 15, 17, and 19. It has been
demonstrated that JGM-3 yields improved orbital ephemerides, substantially reduced
geographically correlated error, improved geodetic results, and an improved marine
geoid. When the JGM-3 solution was obtained, several subset solutions were also
generated to evaluate the effect of particular data set on the gravity model. JGM-3A did
not include GPS data, while JGM-3B excluded laser and DORIS tracking data of TIP.
Recently, GPSIDR data were reprocessed using JGM-3 and the CSR 3.0 ocean tide
model (Eanes, 1994), which was developed by using TIP altimeter data.. Further, four new
cycles of GPS/DR data were processed, which resulted in the JGM-3D model. Also, the
JGM-3B model was updated using the new GPSIDR data.
The following discussion summarizes the solution strategies used to process GPSIDR
data for gravity tuning for TIP. Since the GPS receiver data is projected to be a significant
contributor to future gravity mission, it is important that the specific contribution of the
GPS data to the geopotential parameter determination be quantified. To achieve this
objective, JGM-3 and three related models (JGM-3A, JGM-3B, and JGM-3D) are
compared to evaluate the relative contribution of each data type.
GPSDATASET
Eight cycles of the GPSIDR data, which were collected by the TIP GPS receiver during
the 1992-95 time period, were used in this investigation. Those were Cycle 10 (Dec. 2131, 1992), Cycle 15 (Feb. 8-18, 1993), Cycle 17 (Feb. 28-Mar. 10, 1993), Cycle 19
(Mar. 20-30, 1993), Cycle 20 (Mar. 30-Apr. 9, 1993), Cycle 24 (May 9-19, 1993), and
42
FOR GLOBAL GRAVITY FIELD DETERMINATION
Byron D. Tapley, Hyung-Jin Rim, John C. Ries, Bob E. Schutz, and C. K. Shum
University of Texas Center for Space Research
The University of Texas at Austin, Austin, Texas 78712 USA
INTRODUCTION
GPS tracking data collected by a low-Earth orbiting (LEO) satellite provides global and
continuous coverage of the orbit. Consequently, this data set contains valuable
information for both precise orbit determination (POD) and gravity model improvement.
The GPS experiment on TOPEXIPOSEIDON (TIP) presents the first opportunity to use
GPS data for POD (Bertiger et al., 1994; Schutz et aI., 1994) and global gravity field
development.
The JGM-3 model (Tapley et al., 1994a) is the first gravity model which includes the
GPS data. JGM-3 model was determined by combining the JGM-l information array
(Nerem et aI, 1994) with new information from laser, DORIS, and GPS tracking of TIP,
laser tracking of Lageos-l, Lageos-2, and Stella, and additional DORIS tracking on
SPOT -2. The GPS tracking data used for the JGM-3 model consists of GPS
Demonstration Receiver (GPSIDR) data from Cycles 10, 15, 17, and 19. It has been
demonstrated that JGM-3 yields improved orbital ephemerides, substantially reduced
geographically correlated error, improved geodetic results, and an improved marine
geoid. When the JGM-3 solution was obtained, several subset solutions were also
generated to evaluate the effect of particular data set on the gravity model. JGM-3A did
not include GPS data, while JGM-3B excluded laser and DORIS tracking data of TIP.
Recently, GPSIDR data were reprocessed using JGM-3 and the CSR 3.0 ocean tide
model (Eanes, 1994), which was developed by using TIP altimeter data.. Further, four new
cycles of GPS/DR data were processed, which resulted in the JGM-3D model. Also, the
JGM-3B model was updated using the new GPSIDR data.
The following discussion summarizes the solution strategies used to process GPSIDR
data for gravity tuning for TIP. Since the GPS receiver data is projected to be a significant
contributor to future gravity mission, it is important that the specific contribution of the
GPS data to the geopotential parameter determination be quantified. To achieve this
objective, JGM-3 and three related models (JGM-3A, JGM-3B, and JGM-3D) are
compared to evaluate the relative contribution of each data type.
GPSDATASET
Eight cycles of the GPSIDR data, which were collected by the TIP GPS receiver during
the 1992-95 time period, were used in this investigation. Those were Cycle 10 (Dec. 2131, 1992), Cycle 15 (Feb. 8-18, 1993), Cycle 17 (Feb. 28-Mar. 10, 1993), Cycle 19
(Mar. 20-30, 1993), Cycle 20 (Mar. 30-Apr. 9, 1993), Cycle 24 (May 9-19, 1993), and
42