latitude, although the 168 day repeat orbit recently flown by ERS-l has already supplied
high quality detailed ocean coverage like that found within the: Geosat GM data
(Sandwell et al., 1994).
OCEANOGRAPHIC APPLICATIONS
Gravity field models affect oceanographic applications using satellite altimeter data in a
variety of ways. First, an accurate gravity model is required to precis(~ly position satellite
altimeters, as was discussed earlier. TIP has achieved much better results that Seasat and
Geosat (Nerem et al., 1994a) due mainly to improvements in the orbit (Nerem et al.,
1994c; Tapley et al., 1994c). In addition, a model of the marine geoid is required to
compute ocean dynamic topography using altimeter data. Current geoid models have
errors that exceed the signal of the dynamic topography for wavelengths shorter than
2000 km. For JGM-2, geoid errors are particularly troublesome in the equatorial Pacific
(Nerem et aI., 1994c; Tapley et aI., 1994c), so significant improvements to the geoid
model are required in the future. On a regional basis, local gravimetric geoids suffer
from similar errors, even in areas with good surface gravity coverage (Rapp and Smith,
1994). The TIP Science Working Team has reiterated the need for an imprOVed marine
geoid as a high priority for future research through a dedicated satellite gravity field
mapping mission.
The determination of tidally-driven temporal variations in the Eruth's gravity field is
another important oceanographic application of gravity field detenmination. Satellite
tracking data have been used with a good deal of success to determine the longwavelength coefficients of the major ocean tides that are resonant with the satellite orbits
(Christodoulidis et al., 1988). However, satellite altimetry has providled the most precise
detailed models of the ocean tides (Ray, 1993). In particular, TIP ha.s provided some of
the best models to date (e.g. Egbert et al., 1994; Ma et al., 1994; Schrama and Ray, 1994)
due to its specially chosen orbit, which by design, reduced tidal aliasing, and through the
small orbit errors which are achieved. Knowing the marine gravity in detail also provides
a way to estimate ocean bathymetry under certain assumptions (Smith and Sandwell,
1994)
VERTICAL DATUM DEFINITION
The detenmination of height or elevation is a classical geodetic problem that requires a
gravity equipotential reference surface. Currently, this reference surface is defined
inconsistently from region to region so that there are a large number of height systems or
vertical datums in the world today. The unification of these datums and the possible
definition of a world height system or datum has been discussed for some time. Recent
references in this area include Xu and Rummel (1991), Rapp and Balasubramania (1992),
and Balasubramania (1994). For the implementation of a World Height System, a single
reference surface must be used. In the past, this surface has been approximated by mean
sea level, which today is an unacceptable and unnecessary approximation which can lead
to errors in excess of±1 m in places. The proposed alternative is to accept the concept of
the geoid as the reference surface and to determine geoid undulations to a sufficient
accuracy and resolution for global applications. These needs can be met through the
estimation of geoid undulations from a degree 360 geopotential modc~1. Although such a
model will not provide the resolution and accuracy for all requirements, it can be a base
model for the determination of high resolution geoid undulations in areas where terrestrial
gravity data are sufficiently dense. The undulations can be used to detenmine the
orthometric height by simply differencing the ellipsoidal height whh the geoid height.
The world height system and the need for the geoid reference surface is applicable for
both land and ocean since reference to bathymetric depths requires a specified surface. A
4
high quality detailed ocean coverage like that found within the: Geosat GM data
(Sandwell et al., 1994).
OCEANOGRAPHIC APPLICATIONS
Gravity field models affect oceanographic applications using satellite altimeter data in a
variety of ways. First, an accurate gravity model is required to precis(~ly position satellite
altimeters, as was discussed earlier. TIP has achieved much better results that Seasat and
Geosat (Nerem et al., 1994a) due mainly to improvements in the orbit (Nerem et al.,
1994c; Tapley et al., 1994c). In addition, a model of the marine geoid is required to
compute ocean dynamic topography using altimeter data. Current geoid models have
errors that exceed the signal of the dynamic topography for wavelengths shorter than
2000 km. For JGM-2, geoid errors are particularly troublesome in the equatorial Pacific
(Nerem et aI., 1994c; Tapley et aI., 1994c), so significant improvements to the geoid
model are required in the future. On a regional basis, local gravimetric geoids suffer
from similar errors, even in areas with good surface gravity coverage (Rapp and Smith,
1994). The TIP Science Working Team has reiterated the need for an imprOVed marine
geoid as a high priority for future research through a dedicated satellite gravity field
mapping mission.
The determination of tidally-driven temporal variations in the Eruth's gravity field is
another important oceanographic application of gravity field detenmination. Satellite
tracking data have been used with a good deal of success to determine the longwavelength coefficients of the major ocean tides that are resonant with the satellite orbits
(Christodoulidis et al., 1988). However, satellite altimetry has providled the most precise
detailed models of the ocean tides (Ray, 1993). In particular, TIP ha.s provided some of
the best models to date (e.g. Egbert et al., 1994; Ma et al., 1994; Schrama and Ray, 1994)
due to its specially chosen orbit, which by design, reduced tidal aliasing, and through the
small orbit errors which are achieved. Knowing the marine gravity in detail also provides
a way to estimate ocean bathymetry under certain assumptions (Smith and Sandwell,
1994)
VERTICAL DATUM DEFINITION
The detenmination of height or elevation is a classical geodetic problem that requires a
gravity equipotential reference surface. Currently, this reference surface is defined
inconsistently from region to region so that there are a large number of height systems or
vertical datums in the world today. The unification of these datums and the possible
definition of a world height system or datum has been discussed for some time. Recent
references in this area include Xu and Rummel (1991), Rapp and Balasubramania (1992),
and Balasubramania (1994). For the implementation of a World Height System, a single
reference surface must be used. In the past, this surface has been approximated by mean
sea level, which today is an unacceptable and unnecessary approximation which can lead
to errors in excess of±1 m in places. The proposed alternative is to accept the concept of
the geoid as the reference surface and to determine geoid undulations to a sufficient
accuracy and resolution for global applications. These needs can be met through the
estimation of geoid undulations from a degree 360 geopotential modc~1. Although such a
model will not provide the resolution and accuracy for all requirements, it can be a base
model for the determination of high resolution geoid undulations in areas where terrestrial
gravity data are sufficiently dense. The undulations can be used to detenmine the
orthometric height by simply differencing the ellipsoidal height whh the geoid height.
The world height system and the need for the geoid reference surface is applicable for
both land and ocean since reference to bathymetric depths requires a specified surface. A
4