This poorer accuracy in these GEOSAT anomalies is reflected in the collocation program
accuracy estimates.
A long boundary and possible discontinuities between GEOSAT and ERS-l anomalies
were avoided north of the former Soviet Union, Alaska, and Canada by using only ERS-l
anomalies. ERS-l values were used above 70° north latitude in the Atlantic where a
boundary could not be avoided. The KMS and NOAA multiple source data are very
consistent with DMA's(Table 1 and 2). Examination of values plots do not reveal the
boundary even in the cases where all three sources come together(see Figures 5 and 10).
Because of problems with systematic differences or tilts with altimetry data in the past, a
number of tilt and correlation studies were done. Comparisons were made with values
DMA computed along two Bundesanstalt fUr Geowissenschaften und Rohstoffe(BGR) ship
tracks provided by Rapp OSU and along lines of constant latitude or longitude with DMA's
most accurate available 30' means. The anomalies agree well with the BGR data(see
Figures 6, 7, 8 and 9) and the ship observation derived 30' values.
SUMMARY
The many comparisons with the marine data indicate no systematic regional errors in the
altimetry data and attest to its qUality. All sources differ little when compared with DMA's
ship observations. On the average, the GEOSAT values are about a milligal better than the
other altimetry anomaly sets, but over trenches and sea mount chains the GEOSAT is often
better by more than 20 milligals. The overall accuracy of the GEOSAT values is 2.3
milligals. The processing of the GEOSAT data which resulted in this quality has been
detailed in this paper. GEOSAT was the major altimetry source, comprising ninety percent
of the altimetry values used. The ERS-l anomalies made an important (;ontribution by
extending and improving the coverage especially in polar areas. NOAA's multiple source
anomalies provided coverage in the Weddell Sea near Antarctica. Figure 5 shows the area
of contribution of each source. By combining the DMA, KMS, and NOAA sets, it was
possible to maximize the coverage and improve the quality of the final set. This 30'x30'
mean gravity anomaly file(see Figure 10), which covers 70% of the earth's surface, was the
most accurate and complete altimetry file that DMA could assemble at the time.
This altimetry file, a major source in computing the DMNNASA Earth Gravity
Model(EGM)of degree and order 360, is one reason the project is expected to reach its
goal. The goal is an EGM with a global 30' geoid with an associated absolute accuracy of
+0.5 meter. The DMNNASA EGM geoid will provide a convenient worldwide reference
for precise GPS positioning. DMA as a mapping agency plans to use this precise geoid as a
reference for a worldwide vertical datum replacing the many mean sea level datums currently
in existence. In fact, this geoid will provide an excellent foundation for the many local
precise geoids under development.
REFERENCES
Anderson, 0., Forsberg, R., Knudsen, P., McAdoo, D., and Laxon, S.(1995)
Comparisons of Altimetric and Shipbome Gravity over Ice-covered and Ice-free Polar
Seas, G3 Global Gravity Field and Its Temporal Variation, IUGG XXI General
Assembly, Boulder, CO, Abstracts Week B July 2-14 1995.
Anderson, 0, Knudsen, P. ,and Tscheming, C (1995) Global Gravity Field Recovery
from the Dense ERS-I Geodetic Mission Altimetry, G3 Global Gravity Field and Its
Temporal Variation, IUGG XXI General Assembly, Boulder, CO, Abstracts Week B
July 2-14 1995.
Cloutier, 1., A technique for reducing low-frequency, time-dependent errors present in
network-type surveys, 1. Geophys. Res., Vol. 88, No. Bl, 659-663, 1983.
80
accuracy estimates.
A long boundary and possible discontinuities between GEOSAT and ERS-l anomalies
were avoided north of the former Soviet Union, Alaska, and Canada by using only ERS-l
anomalies. ERS-l values were used above 70° north latitude in the Atlantic where a
boundary could not be avoided. The KMS and NOAA multiple source data are very
consistent with DMA's(Table 1 and 2). Examination of values plots do not reveal the
boundary even in the cases where all three sources come together(see Figures 5 and 10).
Because of problems with systematic differences or tilts with altimetry data in the past, a
number of tilt and correlation studies were done. Comparisons were made with values
DMA computed along two Bundesanstalt fUr Geowissenschaften und Rohstoffe(BGR) ship
tracks provided by Rapp OSU and along lines of constant latitude or longitude with DMA's
most accurate available 30' means. The anomalies agree well with the BGR data(see
Figures 6, 7, 8 and 9) and the ship observation derived 30' values.
SUMMARY
The many comparisons with the marine data indicate no systematic regional errors in the
altimetry data and attest to its qUality. All sources differ little when compared with DMA's
ship observations. On the average, the GEOSAT values are about a milligal better than the
other altimetry anomaly sets, but over trenches and sea mount chains the GEOSAT is often
better by more than 20 milligals. The overall accuracy of the GEOSAT values is 2.3
milligals. The processing of the GEOSAT data which resulted in this quality has been
detailed in this paper. GEOSAT was the major altimetry source, comprising ninety percent
of the altimetry values used. The ERS-l anomalies made an important (;ontribution by
extending and improving the coverage especially in polar areas. NOAA's multiple source
anomalies provided coverage in the Weddell Sea near Antarctica. Figure 5 shows the area
of contribution of each source. By combining the DMA, KMS, and NOAA sets, it was
possible to maximize the coverage and improve the quality of the final set. This 30'x30'
mean gravity anomaly file(see Figure 10), which covers 70% of the earth's surface, was the
most accurate and complete altimetry file that DMA could assemble at the time.
This altimetry file, a major source in computing the DMNNASA Earth Gravity
Model(EGM)of degree and order 360, is one reason the project is expected to reach its
goal. The goal is an EGM with a global 30' geoid with an associated absolute accuracy of
+0.5 meter. The DMNNASA EGM geoid will provide a convenient worldwide reference
for precise GPS positioning. DMA as a mapping agency plans to use this precise geoid as a
reference for a worldwide vertical datum replacing the many mean sea level datums currently
in existence. In fact, this geoid will provide an excellent foundation for the many local
precise geoids under development.
REFERENCES
Anderson, 0., Forsberg, R., Knudsen, P., McAdoo, D., and Laxon, S.(1995)
Comparisons of Altimetric and Shipbome Gravity over Ice-covered and Ice-free Polar
Seas, G3 Global Gravity Field and Its Temporal Variation, IUGG XXI General
Assembly, Boulder, CO, Abstracts Week B July 2-14 1995.
Anderson, 0, Knudsen, P. ,and Tscheming, C (1995) Global Gravity Field Recovery
from the Dense ERS-I Geodetic Mission Altimetry, G3 Global Gravity Field and Its
Temporal Variation, IUGG XXI General Assembly, Boulder, CO, Abstracts Week B
July 2-14 1995.
Cloutier, 1., A technique for reducing low-frequency, time-dependent errors present in
network-type surveys, 1. Geophys. Res., Vol. 88, No. Bl, 659-663, 1983.
80