September 1, 1999
Japan Marine Science and Technology Center (JAMSTEC)

JAMSTEC to Launch World's First Long-term Project for
Seismic and Geodetic Observation in a Deep-sea Floor
-For Understanding the Mechanism of Earthquake Generation off the Sanriku Coast-

On September 2, JAMSTEC (President, Takuya Hirano) will begin operating two deep-sea floor seismic and tectonic movement observation devices off the Sanriku coast (i.e., the Pacific coast in Aomori, Iwate, and Miyagi prefectures) in an area located near a seismically active zone. These devices have been installed in holes drilled approximately 1 km beneath the ocean floor (depth of ocean floor drilling sites: approximately 2500m and 2700m, respectively).

To put the observation systems into operation, we will use our own Dophin-3K deep-sea exploration vehicle to turn on the power and confirm that the observation devices have begun operating. If the devices operate successfully, we expect to carry out the longest-running earthquake observation project ever (the current record is a four-month operation carried out by the University of Hawaii in 1981). In addition, this will be the first project ever to deploy geodetic observation system in the deep-sea floor.

The observation systems consist of a volumetric strain gauge, a tiltmeter, and two wide-band seismographs installed inside the bored hole, and an system control and data recording device in the oceanic crust (Fig. 1). One of the sites where we have placed our observation system is located in an area where minor earthquakes occur very frequently, while the seismic activity very low at the other site (Fig. 2). Both of these observation sites will serve as the stations of the Ocean Hemisphere Network [Note 1]. We expect that the observation results obtained from these two sites would bring a clearer understanding of the following: accumulation of the internal stresses within the crust due to the movement of the Pacific Plate; the relationship between the earthquake occurrence and the internal stress drop in the crust; and a cause governing the difference in the seismic activity between these two sites close geometrically to each other in the crust.

The area off the Sanriku coast is well known for its frequent severe earthquakes. By placing high-sensitivity observation systems in the holes bored into the sea floor there, we hope to gain a clearer understanding of the mechanism behind the occurrence of trench-type earthquakes [Note 2]. One of the most recent and famous such earthquakes was the Sanriku Haruka earthquake in 1994.

After the fabrication of the observation devices, JAMSTEC jointly worked with the other four organizations in July and August aboard the Joides Resolution to install the systems in the sea floor. Our collaborators were: Joint Oceanographic Institutions Incorporated (ODP)[Note 3]; the Carnegie Institution of Washington, the United States; the Ocean Research Institute, University of Tokyo; and the Earthquake Research Institute, University of Tokyo.


1. The Ocean Hemisphere Network is a network that coordinates the undersea activities of the Geophysical Observation Network, an organization run by various Japanese universities and funded by the Ministry of Education. The Geophysical Observation Network comprises four divisions that are responsible for, respectively: installation of devices on islands; installation of devices on the sea floor; development of these devices; and development of methods of analysis.

2. This refers to a reversed-faulted earthquake near trenches, i.e., a type of earthquake caused by the elastic rebound of an upper plate which takes place at the interface with the other lower plate to release the tectonic stress accumulated by the movement of the two plates.

3. The name of Joint Oceanographic Institutions Incorporated is abbreviated as JOI. The "ODP" stands for "Ocean Drilling Program".


For further information, contact the Japan Marine Science and Technology Center.

If you wish to talk with someone in the Deep Sea Research Department, ask for Dr. Mikada or Dr. Suyehiro. Tel: +81-468-67-3983

If you wish to talk with someone in the Public Relations, Training and Education Division, ask for Mr. Taya or Mr. Noguchi.
Tel: +81-468-67-3806


1. Methods

In order to study earthquakes that occur in ocean trenches, JAMSTEC (in cooperation with various other organizations) has previously used the reflection method and the earthquake wave method to carry out structural surveys in the area of the Japan Trench.

In the Japan Trench, the Pacific Ocean plate subducts under Japan at the rate of about 9-10 cm per year, and this movement is thought as the cause of many major earthquakes like the Sanriku Haruka earthquake. However, these major earthquakes can only account for about 1/3 of the deformation that occurs due to this plate subduction. It is believed that about 2/3 of the energy is released in some other form besides earthquakes, most likely a very gradual slippage of the plates and other events that do not constitute earthquakes. For understanding the relationship between the frequent earthquakes off the Sanriku coast and the movement of the Pacific Ocean plate, we spent July and August installing a wide-area seismograph, a tiltmeter, and a volumetric strain gauge in each of two holes bored into the ocean floor, and we will soon carry out an inspection to be sure that the systems are properly installed, after which we will put them into operation (Fig. 1).

2. Preparations prior to the beginning of operations

The following two research objectives are indispensable for studying earthquakes that occur in ocean trenches: understanding of (1) the degree of seismic activity, and (2) the relationship of such seismic activity with the distribution of tectonic stress, i.e., how it changes in time and space. For these objectives, one must: (1) carry out a detailed survey of subsurface structure, and (2) observe earthquakes and earth crust movements over an extended period of time in seismogenic zones close to trenches. Although the importance of such long-term observation has been well recognized, we still have difficulties in the hardware and installation location for the above observation. The development of observation devices appropriate for use in the deep sea floor near the trenches and the location to install such devices so as to keep measurements from being affected by thick sedimentary layers must be well considered. JAMSTEC has developed a volumetric strain gauge in cooperation with the Ocean Research Institute, The University of Tokyo, and the Carnegie Institution of Washington, the United States, to solve the hardware problem. For the installation location of the devices, JAMSTEC worked jointly with several other organizations to drill holes more than one kilometer into the sea floor near the Japan Trench. The installation of the systems was successfully completed in these holes in July and August 1999. Our collaborators were: Joint Oceanographic Institutions Incorporated; the Ocean Research Institute, University of Tokyo; and the Earthquake Research Institute, University of Tokyo. The only thing that remains to be done before the observation can begin is to turn on the power and inspect the devices to make that they are operating properly. This work will get under way very soon.

3. Brief description of the project

The sites of the holes where the systems were installed were located next to each other near the 39th parallel (Fig. 2). There are mainly two different and characteristic zones where the instruments were deployed. One site is in the middle of seismically active zone and the other in a zone characterized by quite low seismic activity. A study of the difference in the seismic activity in these two zones will help us obtain a clearer understanding of the relationship among the plate subduction at the Japan Trench, the accumulation of crustal stress and the earthquake occurrence. All of the observation devices are super-sensitive. The volumetric strain gauge has a sensitivity of 10-11 (i.e., the equivalent of a 1 cc change in a mass of 10,000 kl; the earth's tides, for example, represent a change of about 10-7. The tiltmeter is sensitive to a range of 10 nanoradians (i.e., a height change of 1 mm at a distance of 100 km from the tiltmeter). The seismograph's wide-area (DC 10Hz) acceleration gauge is sensitive to change of 10-9 m/sec/sec (i.e., the equivalent of 1/10-billionth the force of the gravity). We expect that system with this degree of sensitivity will be able to pick up subtle change in the crustal deformation associated with microseismic activity near the ocean trench.

This is not the first time that seismic observation systems have been installed in holes bored in the sea floor. The University of Hawaii carried out a study using short-frequency/short-cycle seismographs for four months in 1981, and the Scripps Oceanographic Institute carried out a three-month study in 1998 using high-sensitivity wide-area seismographs. However, this is the first such study to make use of geodetic observation such as a volmetric strain gauge or a tiltmeter in the deep sea floor.

4. Expected results

One of the sites where we have placed our observation system is located in an area of high seismic activity, while the other site is one of low seismic activity. By comparing the data from these two sites, we expect to obtain the following understandings:

(1) The crustal stress changes in the seismically active and inactive areas to account for the difference in the seismic activity in the same tectonic environment.

(2) The variation of crustal deformation in time and space to make clear the relationship of the microseismic activity with the crustal deformation.

(3) The relationship of the plate subduction with earthquakes.