[J04-P] Poster

[J04-P-11] Accuracy of a continuous/on-demand GPS/Acoustic seafloor positioning using a slackly moored buoy in the Kuroshio region

^○Misae Imano¹, Motoyuki Kido², Yusaku Ohta³, Narumi Takahashi^4,5, Tatsuya Fukuda⁵, Hiroshi Ochi⁵, Chie Honsho², Ryota Hino³ (1.Graduate school of Science, Tohoku University, Miyagi, Japan, 2.IRiDES, Tohoku University, Miyagi, Japan, 3.RCPEV, Tohoku University, Miyagi, Japan, 4.NIED, Ibaraki, Japan, 5.JAMSTEC, Kanagawa, Japan)

For the real-time detection of seafloor crustal movement associated with large earthquakes in a subduction zone, we have developed a continuous and on-demand GPS/Acoustic (GPS/A) seafloor positioning system using a slackly moored buoy, and tested the system over a year in Kumano-nada, Nankai Trough. In the system, the mooring cable length was designed to be 1.5 times the water depth (~3 km) for safety reasons against the strong Kuroshio current. The moored buoy usually drifts by wind and current within a circle of ~4-km radius around the array, which consists of six transponders forming a ~3 km triangle on the seafloor. Apart from the array center, large systematic error (＆#62; 1 m) arises in the positioning because of significant error propagation of the observation data (e.g. the buoy position, travel time) due to uncertainty of array geometry. To reduce the error, we redetermined the array geometry with 20-30 cm accuracy using new algorithm applied to campaign data. Using the revised array geometry, we compare final accuracies with automatic processed data on the buoy (realtime PPP and traveltime detection) and post-processed data (with precise almanac and manual waveform readout) for further improvement in the automatic algorithm. The former accuracy was 0.7/0.9 m in EW/NS component. In this presentation, we introduce the latter accuracy comparing the former one. Then, for further reduction of these error especially in the automatic procedure, we introduce our effort in the presentation to (1) determine again the array geometry using a new set of intense campaign data, (2) carefully evaluate the precision in real-time PPP using a precise moving table, (3) refine the traveltime detection algorithm against reflection from the surface.