Overview
For the purpose of clarifying the mechanism of hydrological precursors to earthquake, we have been monitoring groundwater level on observation wells. Tokai region, a central part of the Honshu (Main Island), is considered to be the area where a devastating earthquake, the Tokai Earthquake, is imminent to occur. Observation of groundwater, especially groundwater level, is regarded as one of useful observations for prediction for the Tokai Earthquake. We have monitored groundwater in Tokai region since 1976 and are responsible for groundwater observation in Tokai region. After the 1995 Hyogo-ken Nanbu Earthquake (Kobe Earthquake), we had made an observation network composed of about 30 wells mainly along active faults in and around the Kansai area for the monitoring of active faults or research for the prediction of inland earthquakes. These monitoring data is transferred through telephone line from observation site to the National Institute of Advanced Industrial Science and Technology (AIST), and offered to Japan Meteorological Agency via on-line system.
1. Tokai area
At 1975, hydrological and geochemical precursors to earthquake prediction seriously started in Japan in response to proposal of the Geodetic council of the Ministry of Education, Science and Culture. AIST has been starting observation in this area (Fig.1). Since 1978 we have been monitoring groundwater level, discharge rate and radon concentration. By the data in these over 20 years, we have been getting the relations between coseismic water-level changes and earthquakes in some observation wells.
2. Kansai area
In this area, AIST newly made these wells after the 1995 Hyogo-ken Nanbu Earthquake. All of the wells have water level sensors and some of them also have borehole strainmeters, GPS's, and seismometers. Most of the wells have single screen or perforated well casing. The depth of them is 130m-1000m and those of the screens are 100m-800m. Continuous observation at all wells started in April 1998 and the results are reported to the Coordinating Committee for Earthquake Prediction. Some preseismic changes in groundwater level and crustal strain, which can be explained by accelerated pre-slip on the earthquake source fault or another nearby fault, have already observed.
3. Observation well
List of observation wells is shown as Table 1. Fig.2 shows schematic diagram of well structure and sensor arrangement, and Photo 1 shows view of the Negoro observation site. The quartz type pressure gauge sensor for groundwater level meter is set about 5m under the water surface. Temperature sensor is set at screen depth. Seismometer and strain sensors are fixed by mortar in the bottom of well. GPS is set on the top of pole in or near the site. These data from each sensor is temporary saved in the computer disk, and transfer via telephone line.
4. Data
Observation data of ground water level is changed by barometric pressure, rainfall and so on. Because these change is larger than the change caused by crustal deformation, monitoring data must be removed those effect. AIST developed correction method for the data of groundwater level. Fig.4 shows groundwater level of Haibara and Kusanagi wells. Each columns of groundwater level include observed data and corrected data. Corrected data is removed effects of barometric pressure, earth tide, and rainfall. It is clearly seen the change caused by earthquakes in corrected data. Changes of water level at Haibara well caused by earthquake are analyzed in Fig.4. The earthquake at May 9th, 1999 in almost on the threshold line in Fig.3. These observing data and developed data correction methods are publishing on our web site. URL is http://gxwell.aist.go.jp/.
References
Kitagawa G. and Matsumoto N.: Detection of coseismic changes of underground water level, Journal of the American Statistical Association, 91, 521-528, 1996.
Koizumi, N., Review of geochemical research for earthquake prediction (In Japanese with English abstract), J. Jap. Soc. Natural Disas. Sci., 16, 41-60, 1997.
Matsumoto, N.: Regression analysis for anomalous changes of ground water level due to earthquakes, Geophysical Research Letters, 19, 1193-1196, 1992
Koizumi, N., Tsukuda, E., Takahashi, M., Sato, T., Matsumoto, N., Ito, H., Kuwahara, Y., Cho, A., Sato, T : Groundwater Level Observation of Geological Survey of Japan in and around the Kinki District, Japan for Earthquake Prediction Research ( In Japanese with English abstract), Journal of the Balneological Society of Japan, 49, 18-33, 1999
Takahashi, M.: Groundwater telemetering system for earthquake prediction(In Japanese with English abstract), Journal of Geography, 102, 241-251, 1993