Long-Term Stability of Acoustic Benchmarks Deployed on Thick Sediment for GPS/Acoustic Seafloor Positioning - Accretionary Prisms and Convergent Margin Tectonics in the Northwest Pacific Basin - page 264

The diving surveys have settled a basic problem in seafloor geodesy and paved the
way for long-term geodetic monitoring on the seafloor.

Keywords Seafloor benchmark • Long-term stability • GPS/Acoustic • Seafloor
geodesy

1 Introduction

Most of large earthquakes occur in subduction zones under deep oceans. Owing to
progresses in the analysis of geodetic data as well as extensive GPS networks, like
the GEONET over the Japanese Islands, we can now roughly estimate lateral varia-
tion in the coupling and slip-deficit rate on the plate boundaries (e.g. Hashimoto
et al. 2009 ). GPS is not, however, directly available on the seafloor. That is why the
GPS/Acoustic (GPS/A) technique was developed for seafloor geodesy, combining
kinematic GPS sea surface positioning and acoustic ranging between a sea surface
unit and PXPs (precision acoustic transponders) on the seafloor (Spiess 1985 ).
Since a steady plate motion was observed on the ocean bottom for the first time by
Spiess et al. ( 1998 ), many important results have been reported: continuous plate
motion at mid-ocean ridge 25 km from the spreading axis (Chadwell and Spiess
2008 ), strong seismic coupling on the subducting plate boundary in the Peru-Chile
trench (Gagnon et al. 2005 ) and in the Japan trench (Fujita et al. 2006 ), co-seismic
crustal movements in the Nankai trough (Kido et al. 2006 ; Tadokoro et al. 2006 )
and in the Japan trench (Matsumoto et al. 2006 ), and so on.
The PXP is a geodetic benchmark like a triangulation station on land. While
benchmarks on land are firmly set up on the ground, PXPs are deployed from the sea
surface on the ocean bottom covered with thick sediment. Long-term attitude stabil-
ity of them is one of key problems underlying the GPS/A seafloor positioning.
Repeated observations in a long time scale are crucial for geodetic monitoring, and
electrical/mechanical durability of the acoustic benchmarks is another key problem.
Although seismicity is generally low along the Nankai trough off southwestern
Japan, three large earthquakes with magnitude 7.4 or less occurred in September
2004 near Kumano-nada, where every historic Tonankai earthquake initiated the
rupture. Three research groups, Nagoya University, Tohoku University, and JHOD
(Hydrographic and Oceanographic Department, Japan Coast Guard) collaborating
with IIS (Institute of Industrial Science, University of Tokyo), independently had
deployed several GPS/A seafloor stations in the Kumano-nada region. They carried
out geodetic observations before and after the earthquakes, and detected horizontal
displacements larger than 20 cm (Kido et al. 2006 ; Tadokoro et al . 2006 ). It was the
first co-seismic crustal movement that was detected by seafloor geodetic observa-
tions. Since the stations are located close to one of the estimated faults, the observed
results provided good constraint on the fault model (Fig. 1 ).
It was suspected, however, that strong ground motions of the earthquakes
might have tilted the seafloor benchmarks deployed on thick sediment. An acoustic
transducer stands on the top of each PXP, which is about 0.6 m high. If a PXP