We installed two sea bottom tide gauges at the nadir of T/P - Jason-1 and ERS tracks in
the southwest Pacific (off Santo Island, Vanuatu) in November 1999. Salinity and surface
temperature are also recorded. Combination of tide-gauge and altimetry data, cross-correlation
of both tide-gauge data, and geodetic measurements on the nearby islands (GPS network and
a DORIS beacon planned at Wusi) should allow us to circumvent the widespread problem of
sea level versus crustal motion mix in tide-gauge data and provide an alternate calibration
site for the Jason altimeter.
Recent works [Mitchum, 1998] have highlighted the importance of tide gauges to validate
altimetry-derived results of long-term sea level variations [Nerem, 1999]. However,
tectonic rates of vertical motion are difficult to separate from long-term sea level
variations in studies based on tide gauge (TG) data unless the potential vertical motion
to which the TG is subjected is surveyed accurately. Places where GLOSS tide gauges are
maintained in the Pacific area and where crustal deformations have been evidenced are
reported in figure 1. In fact, TG data have even been used as a geodetic marker to
evidence crustal motion in some of the reported studies.
The Pacific ocean is mostly surrounded by tectonically active margins that are likely
to undergo vertical deformations. These deformations are due to the seismic stress
accumulation/relaxation cycle. One example among many is provided by the Ambrym earthquake,
in November 1999 (magnitude 7.5), southwest Pacific [Pelletier et al., 2000]. Co-seismic
(instantaneous) vertical displacements in East Ambrym varied from metric uplift to
decimetric subsidence within five kilometres. Moreover, long-term uplift of four
millimetres per year is recorded by risen coral terraces at the easternmost point,
while the lagoon in the south exhibits slow subsidence in that area. The long-term
trends are likely to be the cumulative effect of opposite millennial and centennial
trends. Within the Pacific Ocean, volcanic islands offer the best opportunity to
set TGs. However, they load the underlying seafloor that regionally deflects by
several kilometres. That Hawaii was thus subsiding until very recent times at
even faster rates (4 to 5 mm/yr) is strongly suggested by deep drowned
reefs [Jones, 1995]. In both places, a TG operating without monitoring of its own
vertical motion is likely to provide erroneous long-term trends of sea level variation.
The MOTEVAS project (Mouvements Océaniques et TEctoniques Verticaux par Altimétrie
Spatiale) aims to collect TG time series whose ground and sea level components can
be separated and which are comparable to altimetric data without spatial interpolation.
The site of Wusi has been instrumented for this purpose.
The Wusi bank lies off the West coast of Santo Island, Vanuatu, in the southwest
Pacific (figure 2). T/P track #238 runs uninterrupted along the coast of Santo.
A sea-bottom TG was immersed in November 1999 (figure 3) right under the T/P track
at Wusi, about five kilometres offshore. Because it is likely to undergo vertical
motions, a backup TG has been immersed at Sabine bank, about 70 km southwest of Santo,
under an ERS crossover point. It is unlikely to undergo significant vertical motions.
The potential vertical motion of the TGs will be monitored by directly adding the TG
and altimetry data, since the result is independent of the sea level. TG-observed sea
level is then derived and compared to altimetry-derived sea level. Results will be
validated by a cross-analysis of TG data series (that must output the same sea level
variations) and geodetic controls (GPS surveys in neighbouring islands; a DORIS beacon
at Wusi is under review by IDS). RMS scatter between altimetry and TG data will provide
an estimate of the altimetry data noise and a level of the sea surface with GPS buoys
is planned to enable altimeter calibration.
|The MOTEVAS TGs also collect data in a remote area of the tropical Pacific ocean. As
such, pressure data will feed global tide models in that poorly constrained area.
Height, temperature and salinity records will be included in ENSO studies.
Jones A.T., 1995: Geochronology of drowned Hawaiian coral reefs. Sed. Geology, 99, 233-242.
Mitchum G.T., 1998: Monitoring the stability of satellite altimeters with tide gauges. J. Atmos. Oceanic Tech., 15, 721-730.
Nerem R.S., 1999: Measuring very-low-frequency sea level variations using satellite altimeter data. Global and Planetary Change, 20, 157-171.
Pelletier B., M. Regnier, S. Calmant, R. Pillet, G. Cabioch, Y. Lagabrielle, J.M. Bore, J.P. Caminade, P. Lebellegard, I. Christofer, S. Temakon, 2000: Le séisme d'Ambrym-Pentecôte (Vanuatu) du 26 novembre 1999 (Mw: 7.5): Données préliminaires sur la séismicité, le tsunami et les déplacements associés. C. R. Acad. Sc. Paris, 331, 21-28.