4500
4000
3500
3000
- 11:1
::s 2500
8
:;i 2000
U
1500
1000
500
• Area Adjusted GEOSA T
Network
EI Weekly Adusted Seasat
Network
o 1 2 3 4 5 6 7 8 9 10 11 12 13 1415 1617 18 192021 22 23 24252627 28 29
RMS(cm)
Figure 1. Distribution of 1 °xl 0 cell adjusted crossing point statistics before and after the area
adjustment of the GEOSAT GM data to the GEOSAT ERM network.
adjustment, caused by the currents moving back and forth, can be effectively reduced by
meaning the point values. However, the currents also displace the mean sea surface in a
systematic way. A model is required to compensate for this systematic displacement that is
present in the mean surface.
Products on hand at the beginning of the project. DMA had a file of 5'x5' mean
heights computed from the readjusted GEOSAT data. Also on hand was a set of 5'x5'
GEOSAT mean gravity anomalies computed from these GEOSAT mean heights. The
anomalies were computed using an earlier version of the current least squares collocation
program and a set of auto-covariance values for rough, moderate, and smooth areas. The
auto-covariance values were empirically derived from DMA's highest quality ship data.
DA TA PREPARATION
NASA Dynamic Sea Surface Topography model. The density of the GEOSAT point
data, combined with meaning, effectively resolves the variability seen in the crossing points
(Figure 3). However, a systematic difference remains between the mean sea surface and the geoid
that is referred to as the Dynamic Sea Surface Topography. The NASA Dynamic Sea Surface
Topography model(Nerem, 1994) derived from Topex and ERS-l data(see Figure 4) was used to
resolve this remaining systematic regional error. There is a strong north south effect across the
Antarctic circum-polar current. The application of this model made a significant improvement in
the medium wavelength portion of the final me.
Transformation to ITRF 91. The GEOSAT mean geoid heights were transformed
from World Geodetic System 1984(WGS 84) to the International Terrestrial Reference
Frame(ITRF) using:
hITRF91 = hWGS 84 + Lh cos cos A + /J.y cos sin A + /J.z sin + B
73
4000
3500
3000
- 11:1
::s 2500
8
:;i 2000
U
1500
1000
500
• Area Adjusted GEOSA T
Network
EI Weekly Adusted Seasat
Network
o 1 2 3 4 5 6 7 8 9 10 11 12 13 1415 1617 18 192021 22 23 24252627 28 29
RMS(cm)
Figure 1. Distribution of 1 °xl 0 cell adjusted crossing point statistics before and after the area
adjustment of the GEOSAT GM data to the GEOSAT ERM network.
adjustment, caused by the currents moving back and forth, can be effectively reduced by
meaning the point values. However, the currents also displace the mean sea surface in a
systematic way. A model is required to compensate for this systematic displacement that is
present in the mean surface.
Products on hand at the beginning of the project. DMA had a file of 5'x5' mean
heights computed from the readjusted GEOSAT data. Also on hand was a set of 5'x5'
GEOSAT mean gravity anomalies computed from these GEOSAT mean heights. The
anomalies were computed using an earlier version of the current least squares collocation
program and a set of auto-covariance values for rough, moderate, and smooth areas. The
auto-covariance values were empirically derived from DMA's highest quality ship data.
DA TA PREPARATION
NASA Dynamic Sea Surface Topography model. The density of the GEOSAT point
data, combined with meaning, effectively resolves the variability seen in the crossing points
(Figure 3). However, a systematic difference remains between the mean sea surface and the geoid
that is referred to as the Dynamic Sea Surface Topography. The NASA Dynamic Sea Surface
Topography model(Nerem, 1994) derived from Topex and ERS-l data(see Figure 4) was used to
resolve this remaining systematic regional error. There is a strong north south effect across the
Antarctic circum-polar current. The application of this model made a significant improvement in
the medium wavelength portion of the final me.
Transformation to ITRF 91. The GEOSAT mean geoid heights were transformed
from World Geodetic System 1984(WGS 84) to the International Terrestrial Reference
Frame(ITRF) using:
hITRF91 = hWGS 84 + Lh cos cos A + /J.y cos sin A + /J.z sin + B
73