MARINE GRAVITY FROM SATELLITE ALTIMETRY OVER
OCEAN AND SEA ICE
Introduction
David T. Sandwell and Mara M. Yale
Scripps Institution of Oceanography
La Jolla, CA 92093-0225 USA
e-mail sandwell@radar.ucsd.edu
David C. McAdoo and Walter H. F. Smith
NOAA Geosciences Laboratory
N/OESI2 Silver Spring MD 20910
Radar altimeter measurements of the marine geoid collected during the Seasat altimeter
mission gave marine geodesists and geophysicists a hope of uncovering the details in the
gravity field over all the ocean basins. However because of insufficient track density, it
has taken 16 years for the full potential of the satellite altimeter method to be realized. The
high density coverage obtained by ERS-l during its geodetic mapping phase (4/94 - 3/95)
prompted the US Navy to declassify all of the Geosat altimeter data (6/22/95). We are
grateful to ESA for extending the ERS-l mapping phase so that an equatorial ground-track
spacing of 8 km could be completed. The combination of these two high-density data sets
will provide the first detailed view of all the ocean basins. Considering the sparse
shipboard coverage of many ocean areas [Smith, 1993], these new altimeter data are
arguably the most important marine geology and geophysics data set collected over the past
decade.
The focus of this paper is on the efficient recovery of marine gravity anomalies and other
derivatives of the potential using data from satellite altimeters having different orbital
inclinations and different noise characteristics; no attempt is made to recover sea surface
topography (i.e., geoid height plus ocean dynamic topography). After a discussion of data
availability we present a recipe for constructing gridded gravity anomalies from altimeter
profiles. We then briefly show recent results from Laxon and McAdoo [1994] where they
demonstrate that the marine gravity field can be recovered over areas of permanent sea ice
cover by retracking the waveforms of the ERS-l altimeter echo. Finally we assess the
accuracy and resolution of the estimated marine gravity field and discuss future prospects.
This paper does not review other methods for recovery of short-wavelength gravity
information from satellite altimetry [e.g., Andersen et al., 1995] but is instead focused on
our recipe and why certain processes are used. Over the next months and years our recipe
will surely undergo revisions and perhaps others will devise better recipes.
Data A vailbility
The repeat period of the satellite orbit governs the spacing of the altimeter tracks on the
ocean surface. Very long repeat cycles such as 168-day ERS-l geodetic phase (Figure 1,
north of 30
0
S) or the non-repeat (drifting) orbit of the GeosatiGeodetic Mission (Figure 1,
south of 30
0
S) provide the high density coverage needed for complete resolution of the
gravity field. The shorter repeat periods of 10 days for Topex, 17 days for Geosat, and 35
days for ERS-l (Figure 1, thick lines) do not provide dense track coverage. However, the
repeated profiles can be averaged to improve the signal-to-noise ratio ,as well as to assess
the noise properties of the altimeter measurements.
12
OCEAN AND SEA ICE
Introduction
David T. Sandwell and Mara M. Yale
Scripps Institution of Oceanography
La Jolla, CA 92093-0225 USA
e-mail sandwell@radar.ucsd.edu
David C. McAdoo and Walter H. F. Smith
NOAA Geosciences Laboratory
N/OESI2 Silver Spring MD 20910
Radar altimeter measurements of the marine geoid collected during the Seasat altimeter
mission gave marine geodesists and geophysicists a hope of uncovering the details in the
gravity field over all the ocean basins. However because of insufficient track density, it
has taken 16 years for the full potential of the satellite altimeter method to be realized. The
high density coverage obtained by ERS-l during its geodetic mapping phase (4/94 - 3/95)
prompted the US Navy to declassify all of the Geosat altimeter data (6/22/95). We are
grateful to ESA for extending the ERS-l mapping phase so that an equatorial ground-track
spacing of 8 km could be completed. The combination of these two high-density data sets
will provide the first detailed view of all the ocean basins. Considering the sparse
shipboard coverage of many ocean areas [Smith, 1993], these new altimeter data are
arguably the most important marine geology and geophysics data set collected over the past
decade.
The focus of this paper is on the efficient recovery of marine gravity anomalies and other
derivatives of the potential using data from satellite altimeters having different orbital
inclinations and different noise characteristics; no attempt is made to recover sea surface
topography (i.e., geoid height plus ocean dynamic topography). After a discussion of data
availability we present a recipe for constructing gridded gravity anomalies from altimeter
profiles. We then briefly show recent results from Laxon and McAdoo [1994] where they
demonstrate that the marine gravity field can be recovered over areas of permanent sea ice
cover by retracking the waveforms of the ERS-l altimeter echo. Finally we assess the
accuracy and resolution of the estimated marine gravity field and discuss future prospects.
This paper does not review other methods for recovery of short-wavelength gravity
information from satellite altimetry [e.g., Andersen et al., 1995] but is instead focused on
our recipe and why certain processes are used. Over the next months and years our recipe
will surely undergo revisions and perhaps others will devise better recipes.
Data A vailbility
The repeat period of the satellite orbit governs the spacing of the altimeter tracks on the
ocean surface. Very long repeat cycles such as 168-day ERS-l geodetic phase (Figure 1,
north of 30
0
S) or the non-repeat (drifting) orbit of the GeosatiGeodetic Mission (Figure 1,
south of 30
0
S) provide the high density coverage needed for complete resolution of the
gravity field. The shorter repeat periods of 10 days for Topex, 17 days for Geosat, and 35
days for ERS-l (Figure 1, thick lines) do not provide dense track coverage. However, the
repeated profiles can be averaged to improve the signal-to-noise ratio ,as well as to assess
the noise properties of the altimeter measurements.
12