next up previous print clean
Next: Other OBS models and Up: Vlad: Ocean-bottom seismometers Previous: The Japanese OBS

OBS logistics aboard R/V Kaiyo

OBS deployment is quite simple: the OBS clock drift is measured, the whole assembly goes through a last quality check, the retrieval system switches are set on automatic, and the OBS is lowered into the water (Fig. [*]).

 
launch
Figure 5
The OBS is gently lowered into the water using a small crane. The items closest to the viewer are the sonar transceiver and transducer. The strobe is also visible as a horizontal appendage on the upper part of the OBS. Withought the sinker weights, the center of gravity of the OBS shifts so that the radio beacon and the strobe are standing upright.
launch
view

The clock drift is measured along a period of several hours before the launch of the OBS, and just minutes before the launch the clock is re-synched with the GPS time signal. The instrument is lowered into the water with a small crane. This operation is not likely to be hampered by bad weather because of its simplicity. Sometimes sea currents may cause the OBS to land at a different position than planned. The average drift was 6.5$\%$ of the water depth for the U.S. 2-C OBSs Gunther et al. (2002). The position is found with a precision of 2$\%$ of sea depth by watching the OBS at descent and at retrieval with the sonar array of the ship, or with a precision of 5 to 25 m by triangulating water wave arrivals Gunther et al. (2002). Currently there are no provisions for guiding the OBS during descent. Japanese OBS descent rates are between 79 and 85 m/min. These rates are not constant Ito et al. (2002b), which may indicate turbulence caused by the lack of hydrodynamicity of the base frame, especially since the ascent rates, when the base frame is absent, are constant. Brainstorming for potential future measures to improve the accuracy of the drop reveals two main methods: guiding fins remotely controlled from the surface by acoustic modem and increasing the descent speed. Descent can be accelerated by increasing sinker weight and improving the hydrodynamic properties of the OBS ensemble, by including a simple compressed-air propulsion system or even by reducing the water resistance through super cavitation. Care should be exercised so as not to damage the instrument by a hard impact with the seafloor.

After the OBSs are deployed, the survey is shot. The ship may tow streamers as well. For retrieval, the ship sends to each individual OBS a particular signal by sonar. The electrolytic corrosion process starts, lasting on average 13 minutes. The OBS then rotates 90 degrees to adjust to its new gravity center, which will bring the strobe and the radio beacon into an upright position. It then lifts towards the surface at a constant rate of 66.5 m/min. As at descent, the ship uses the sonar array to monitor the position of the OBS, so that it will be very close to the OBS when it surfaces. In most instances of OBS retrieval during the cruise, the ship was able to maneuver so that the OBS would surface within 50 m in front of the ship.[*] The OBS is then simply fished out with a net (Fig. [*]), and samples of sediment on it are collected if there is any such scientific interest.

 
recovery
Figure 6
The OBS is lifted aboard with a small net.
recovery
view

The OBS is then washed, the clock is checked again with respect to the GPS time reference, then it is dismantled. The current data collection procedure involves taking apart the glass sphere, but that may be made unnecessary with future improvements. Even before improving the instruments, the recovery rate of Japanese OBSs was 97.2$\%$, computed on 472 total individual OBS deployments. Instruments that did not surface were recovered by a Remotely Operated Vessel (ROV) in order to analyze the reasons of failure Ito et al. (2002b). The three main causes of failure were: 1. Water leaks in the glass sphere due to improper sealing and/or sphere shaping defects. This was addressed by placing a pressure gauge inside the sphere to check the integrity of the seal before launch. 2. Leaking into the transponder pressure housing due to a structural defect of the O-ring seal at the pressure housing end cap. Mandatory checks of that component before launch were instituted. 3. Release devices failing to operate. Other improvements are currently under study Ito et al. (2002a). They may include increasing transponder battery life above the current level of three months, raising transmission rate above the current 180 dB, and retrieving data without opening the glass sphere. After improvements, the recovery rate is expected to be above 99$\%$.

The retrieval of about 100 OBSs (Fig. [*]) placed along a 2-D line at a distance of 5 km and 10 km apart from each other took nine days, which is three times longer than their deployment and two times longer than shooting the 2-D line with the airguns. Much of the time was spent by the ship waiting for the electrolytic corrosion to finish and for the OBS to surface from depths sometimes greater than 4000 m (approx. 1 hour ascent time). The preparation time of an OBS for deployment was about 10 minutes, using a team of four technicians.

 
many_stacked
Figure 7
Stacks of OBSs on the deck of the ship. After the base frame is left on the seafloor, they will take only 1/4 of the previous space. The ergonomic arrangements on the ship permit operations with hundreds of instruments.
many_stacked
view


next up previous print clean
Next: Other OBS models and Up: Vlad: Ocean-bottom seismometers Previous: The Japanese OBS
Stanford Exploration Project
7/8/2003