(was 0802-10)
Shimabara Earthquake and Volcano Observatory
Graduate School of Sciences
Kyushu University
Nagasaki-ken 855

Telephone : (81) 957-62-6621
Telefax : (81) 957-63-0225
Director: Dr. Hiroshi Shimizu
Email : This email address is being protected from spambots. You need JavaScript enabled to view it.
Website : www.sevo.kyushu-u.ac.jp/index-e.html

Scientific staff:

Dr. Hiroshi Shimizu - Geophysics (seismology)
Dr. Satoshi Matsumoto
Mr. Norimichi Matsuwo - Geophysics (seismology)
Dr. Takeshi Matsushima - Geophysics (geodesy, seismology)
Dr. Kenji Uehira
Mr. Kodo Umakoshi - Geophysics (seismology)
Dr. Kazuya Ohta - Emeritus - Volcanology (geology, geochemistry)

Technical staff:

Mr. Risaku Fukui - Technical
Mr. Kazunari Uchida - Technical


Shimabara Earthquake and Volcano Observatory (SEVO) is located at the eastern foot of Unzen Volcano in the Shimabara peninsula, western Kyushu. The former name of the observatory is Shimabara Volcano Observatory, which was founded in April 1971 for the purpose of prediction research of volcanic eruptions by geochemical and seismological approaches. The observatory was reorganized to SEVO in April 1984, in order to expand a sphere of research into fundamental studies on earthquake prediction.

Mt. Fugen of Unzen Volcano erupted on November 17, 1990 after 198 years dormancy since the last eruption in 1792. The seismic network of SEVO had detected precursory earthquakes and tremors before the eruption. A dacite lava dome appeared on May 20, 1991 at Jigoku-ato Crater of Mt Fugen. The lava dome has been growing until February 1995, and rock falls from the dome had frequently caused pyroclastic flows. The observation system of SEVO reinforced after the 1990 eruption. Furthermore, intensive geophysical, geochemical and geological investigations were carried out at Unzen Volcano by the Joint University Research Group and national institutes.

Research activity

This observatory is performing fundamental research of volcano phenomenon, and evaluation of the degree of volcanic activity or research of volcanic eruption prediction through observation of the volcanic activity for prediction research of the structural earthquake accompanied by the north-south extension phenomenon of the Beppu-Shimabara graben which is crossing central part of Kyushu, and Amakusa graben which is one segment of the Okinawa Trough which stands in the row in the southwest, or the inland type earthquake by observation of a diastrophism, and the Unzen volcano. Especially, the Unzen volcano is situated in a position unique on the geological structure in southwest Japan called the meeting point of the Beppu-Shimabara graben and the Okinawa Trough, it is thought that the activity has the relation closely with these tectonic movements, and it is the optimum ground for synthetic research covering both fields of earthquakes and volcanoes.


Seismic Observations

Although earthquakes are the destructive phenomenon of the crust, the observation is to the foundations of the earthquake prediction and volcanic eruption prediction research. Always, as observation equipment, this observatory holds 16 networks of seismic observation of the central part and the northern part areas in Kyushu, and 9 networks of seismic observation of the Unzen volcano area, and is performing continuation observation on the real time by telemeter.

As for this telemetry system for observation of earthquakes, four points were first equipped as the part of the first and secondary volcanic eruption prediction plan (1974-1983), and they are the purpose which grasp movement of magma to the highest peak of Mt.Fugen which is the eruption point since the history started, from the Chijiwa caldera of the Unzen volcano western side assumed at the time of the eruption, and they are arranged so that these may be surrounded. After that, it inherits in the 5th earthquake prediction plan (1984-1988), and was expanded on the total observation network of 11 points for the Amakusa graben and the Shimabara-Kumamoto graben, which are earthquakes frequent occurrence zones. Furthermore, in the 6th and the 7th earthquake prediction plan (1988-1998), the extension to the observation network for the whole region of the central part and the northern part of Kyushu is achieved, and it is always observing by 16 observing points of earthquakes now.

The data signal from the observing point of every place is conveyed by the cable telemetry system by NTT dedicated line (D1 standard) in the Shimabara base observatory, and is acting as the monitor by Pen Recorder for a long time. Furthermore, earthquake data automatic processing equipment introduces, it comes to perform automatic focus determination processing of earthquakes, and the grasp of a quick earthquake generating situation came to be made 1992.

1) Seismic Network Around Unzen Volcano

Before the 1990 eruption of Unzen Volcano, 4 permanent and 2 temporary seismic stations had been operated around the volcano by SEVO.

Seismic sensors used in the observation are velocity transducers with the natural period of 1.0 second. The permanent stations had three components of the sensor, while the temporary stations had the vertical component. Seismic signals were telemetered to SEVO and recorded on the magnetic tapes using an event-trigger system.

After the 1990 eruption, 6 permanent stations were installed around Unzen Volcano. Because one of the stations was buried by the pyroclastic flows in march 1992, 9 permanent stations are now in operation. Furthermore, more than 10 temporary stations are operated on the volcanic edifice. Seismic data from the permanent stations and most of the temporary stations have been transmitted to SEVO using radio telemetry or telephone-line telemetry. These data have been continuously recorded on digital audio-tapes (DAT) and 8 mm video-tapes, and event-trigger data have been recorded on CD-ROMs.

2) Seismic Network in the Middle and Northern Kyushu

SEVO has a seismic network for monitoring microearthquake activity in the Kyushu district, especially the Beppu-Shimabara graben.

The Beppu-Shimabara graben is a large depression zone crossing the middle Kyushu from ENE to WSW. The crustal activity around the graben is therefore very active, and characterised by the N-S extension stress-field.

Sixteen permanent stations, except for the stations around Unzen Volcano, are deployed in the middle and northern Kyushu including the Beppu-Shimabara graben. Six-channel data (3 components, high- and low-gain signals) from each station are sent to SEVO using telephone-line telemetry, and are processed by several work-stations. The continuous and event-trigger seismic data are recorded on 8 mm video-tapes and CD-ROMs, respectively.

Geodetic Observations

Four borehole tiltmeters are installed with seismometers and thermometers around Mt. Fugen of Unzen Volcano. Two proton magnetometers are also installed at one of the borehole stations. Both tilt and magnetic data are telemetered to SEVO and registered on the magnetic disk of a work station.

Ground deformation measurements have been carried out at the lava dome and around the volcano using global positioning system (GPS).

Moreover we have attended in precise gravity measurements and precise levelling surveys as a member of the Joint University Research Group.

One of the axes presumed to have arranged like the group of a wild goose in the Beppu-Shimabara graben is the Unzen graben that is crossing the central part of the Shimabara peninsula. Now, sedimentation is continued at the rate of 2 to 3mm per year. Moreover, the western Chijiwa bay is also presumed to be that which sediments doubly and forms the fan in the meeting point with the Amakusa graben, which stands in the row in south-southwest.

In order to grasp the actual condition of these crustal movements, the level point and the reference point for a electro optical distance measurement are installed, cooperation of another universities observation organization is obtained, and the 1st survey was performed in the Shimabara peninsula in 1986. Miraculously, since Mt.Unzen started eruption activity for the first time in 198 years in 1990, the survey is carried out again after that and many results in connection with volcanic eruption prediction research are obtained.

On the one hand, maintenance of crustal-movement observation institution was always accompanied by eruption activity of Mt.Unzen also achieved, and four borehole type tiltmeters were installed on the outskirts of the volcano in 1992. Moreover, crustal-movement observation of Kyushu is also started by introduction of GPS observation equipment.

Geochemical Observations

1) Hot Spring Observations

Chemical composition of hot springs is investigated at Unzen, Aso and Sakurajima volcanoes. The increase of CO2 concentration was observed before eruptive activities of Aso and Sakurajima volcanoes.

Groundwater level and chemical composition of bubble gases in a well (360 m deep) at SEVO are automatically observed every 15 minutes. The groundwater monitoring system and the gaschromatograph are installed in a well.

The gas ingredient currently emitted from magma melts into underground water, and a hot spring is generated by the interaction with the rock in the wall of a passage. Therefore, hot springs are expected as an important point of the direct communication of information about the activity situation in the volcanic underground depths. An unusual change had actually arisen in states that a hot spring springs out, just before the once large eruption in the Sakurajima volcano. Since it is such, continuous observation of the temperature of a hot spring, water pressure (water level), or a chemistry ingredient is very important in grasping a volcanic activity situation.

In this observatory,the continuous observation of the quality of hot springs is carried out about the Unzen, Aso, and Sakurajima, many volcanoes 6 fountainhead. According to the old result, the concentration in which carbonic acid gas melts and remains is increasing gradually before an eruption time and summit activity activated. Moreover, if a summit crater will be in an opening state, changing to gradual decrease immediately is also clear. This is reflecting increase and decrease of the fluid pressure of the gas inside a volcano, and is effective in evaluation of the degree of volcanic activity, or prediction of transition. Moreover, in the hot spring observation well currently installed within the enclosure of this observatory, just before or after earthquake generating in the neighborhood, an unusual rise or unusual descent of the water level is see, and reflecting accumulation and opening of crustal strain sharply is verified.

2) Volcanic Gas Observations

Remote sensing measurements of SO2 by correlation spectrometry have been carried out several times a year at Kuju Volcano, and once or twice a year at Aso, Sakurajima and other active volcanoes. Emission rates of SO2 increased prior to the 1982 eruption of Asama Volcano.

The volcanic gas (smoke) currently emitted from the crater precedes and increases to an eruption, and it is expected that change is seen by the chemistry composition.

In this observatory, telemetry of the amount to which the sulfur dioxide(SO2) is emitted from the summit crater is continuously carried out in Asama, Oshima, Aso, and Sakurajima-many volcanoes as a part of the "development research of the new method of the observation of volcanic activity "based on the first and secondary volcanic eruption prediction plan. Moreover, in Mt.Aso, automatic continuation observation of hydrogen fluoride(SO2), HF, hydrogen chloride, and the four ingredients of H2S in the volcanic gas diffused in the atmosphere near the crater was carried out, and it investigated how change of the ratio of the ingredient would be connected with the volcanic activity. In both the volcanoes in Asama and Aso, it precedes with the eruption and the amount to which SO2 are emitted increases. Moreover, in the Aso volcano, it is recognized that the relative concentration of HF or HC1 also increases, as the volcano becomes active.

In recent years, amount of the sulfur dioxide emitted from the Unzen volcano-Kueyama, which became active, was mainly measured repeatedly.