Long-term Monitoring in Deep Boreholes in the Nankai Subduction Zone

IODP NanTroSEIZE is planning to construct observatories in deep boreholes in the Nankai subduction zone to monitor interseismic behavior at and above the updip limit of seismogenic zone. The deep observatories will be at NT2-03 and NT3-01; NT2-03 will be 3.5-km deep below sea floor to penetrate several splay faults, and NT3-01 will be 6-km deep to penetrate the splay faults and plate boundary fault. Scientific needs for long term in situ monitoring at and near splay faults and plate boundary fault include detailed analysis of low frequency/slow events, strain partioning during interseismic period, and near-source hydrological/seismic/ geodetic observations.


Introduction
IODP NanTroSEIZE is planning to construct observatories in deep boreholes in the Nankai subduction zone to monitor interseismic behavior at and above the updip limit of seismogenic zone.The deep observatories will be at NT2-03 and NT3-01; NT2-03 will be 3.5-km deep below sea floor to penetrate several splay faults, and NT3-01 will be 6-km deep to penetrate the splay faults and plate boundary fault.Scientific needs for long term in situ monitoring at and near splay faults and plate boundary fault include detailed analysis of low frequency/slow events, strain partioning during interseismic period, and near-source hydrological/seismic/ geodetic observations.

Scientific Needs
A deep borehole observatory has advantages in surface noise reduction (Araki et al., 2004), high frequency recording with little attenuation, observation in the close vicinity of active processes, and installing a vertical sensor array along the borehole to improve the depth resolution (Chavarria et al., 2003).
Drilling through the updip limit of the fault and establishing borehole observatories is important to understand the following issues: Answers to the following questions may also be provided from a borehole observatory: Can both (thin slip zone and thick fault zone) occur within "asperity"?How thin or thick are the zones?are the zones? the zones?How do they rupture coseismically ?do they rupture coseismically ?they rupture coseismically ?How do they deform in an interseismic period?do they deform in an interseismic period?they deform in an interseismic period?an interseismic period?interseismic period?How are they deforming now ?are they deforming now ?they deforming now ?How different are they in plate boundary / splay fault?are they in plate boundary / splay fault? in plate boundary / splay fault?
NanTroSEIZE scientists proposed an observatory plan (Harold and Kinoshita, 2006;Shinohara et al., 2003).One of the most essential parts of the basic idea is a distributed, multi-level multi-sensor system (Fig. 1), with seismometer, tiltmeter, strainmeter, and pressure sensors.
, and pressure sensors.and pressure sensors.

Possible Plan
Based on this basic concept, we have been working on a is basic concept, we have been working on a basic concept, we have been working on a a possible plan for sensors, a downhole telemetry system, a s, a downhole telemetry system, a , a downhole telemetry system, a a downhole telemetry system, a downhole telemetry system, a a sensor/downhole telemetry system interface, and system /downhole telemetry system interface, and system downhole telemetry system interface, and system , and system and system installation.We understand that distributed sensors are We understand that distributed sensors are We understand that distributed sensors are essential.Currently, strain and pressure sensors need to be (Tobin and Kinoshita, 2006).

Long-term Monitoring in Deep Boreholes in the Nankai Subduction Zone
by Hisao Ito doi:10.0/iodp.sd.s01..007 and seismometers might be cemented, however for multi-s might be cemented, however for multi-might be cemented, however for multilevel installations we need to consider installing them by s we need to consider installing them by we need to consider installing them by ing them by them by other methods (e.g., by locking arm or bow spring).
s (e.g., by locking arm or bow spring).(e.g., by locking arm or bow spring).e.g., by locking arm or bow spring).by locking arm or bow spring).
Pore pressure monitoring is useful as a proxy of strain and pore pressure variation along the fault (Davis et al., 2006;Kano and Yanagidani, 2006).To be useful as a proxy for strain, the compliance of the pressure measurement system needs to be extremely small.For monitoring pore pressure of the fault, the gault must be isolated from other sections of the borehole either by packer or cementing.

Links to Shallow Boreholes/ Land Stations
The NanTroSEIZE deep borehole observatories, which will deploy vertical array sensor systems, will be linked to a s, will be linked to a , will be linked to a a sea floor cable network and NanTroSEIZE shallow borehole observatories to be integrated to three dimensional ocean observatories.The IODP NanTroSEIZE observatories will be situated at/around the updip limit.Ideally, the IODP Ideally, the IODP the IODP NanTroSEIZE observatories would also be integrated with would also be integrated with e integrated with on-land observatories, which would be above the asperities.
-land observatories, which would be above the asperities.land observatories, which would be above the asperities.ould be above the asperities.be above the asperities.

Possible Application to Other Projects
The telemetry system and sensors that have been developed for NanTroSEIZE will be applied in other deep boreholes, such as Costa Rica Seismogenesis Project CRISP (Ranero et al., this issue), and also The Kanto Asperity set at the bottom of the hole, and tilt/seismic/temperature of the hole, and tilt/seismic/temperature hole, and tilt/seismic/temperature , and tilt/seismic/temperature and tilt/seismic/temperature monitors may be set at bottom-and mid-hole sections (Fig. 2).

Major Technical Challenges
The following major technical issues for developing a deep borehole observatory in IODP are recognized: Great depths telemetry, sensor, etc. High temperatures (80°C-100°C for 3.5 km, 170°C-180°C for 6 km) Long-term reliability and stability Coupling of the sensors to the for mation/casing Geodetic/seismic and pore pressure measurements at multiple intervals (multi-packer vs. multi-hole?)Vertical drilling and core sampling at the fault interval Broadband high dynamic range and high resolution recording Deployment Telemetry system through Christmas tree/wellhead system of riser drill holes Real time monitoring through seafloor cables

Borehole Telemetry System
The system should be reliable and work for many years at high temperatures.At the same time, the system should be able to deliver data that scientists need.The system should be able to be deployed practically and maintained with reasonable efforts.The fundamental requirements may be summarized as follows: reliability, long mean-time-before-follows: reliability, long mean-time-before-eliability, long mean-time-before-, long mean-time-before-ong mean-time-beforefailure (MTBF), redundancy, protection from failure, high , redundancy, protection from failure, high edundancy, protection from failure, high , protection from failure, high rotection from failure, high rom failure, high failure, high , high igh dynamic range, continuous recording, low power , continuous recording, low power ontinuous recording, low power , low power ow power consumption, and limited number of connections through , and limited number of connections through imited number of connections through pressure controlled well head/Christmas tree. .The system consists of three sections: 1) a downhole : 1) a downhole 1) a downhole a downhole downhole module array that digitizes seismic signals continuously and transmits the data to a recorder on sea floor, 2) a subsea recorder that receives data from downhole and stores data, and 3) a communications unit that sends commands from sea s unit that sends commands from sea unit that sends commands from sea surface to the subsea recorder to check the status of the downhole systems and receives QC data (Fig. 3). .

3). 3).
The downhole module can also send data from other wnhole module can also send data from other sensors (e.g., pressure, temperature, and strain monitors) e.g., pressure, temperature, and strain monitors) pressure, temperature, and strain monitors) , and strain monitors) and strain monitors) monitors) ) interfaced to auxiliary channels at a lower sampling rate.
a lower sampling rate.lower sampling rate.Project (KAP): KAP is a linked offshore/onshore regional geodetic and seismic network covering the asperity and nonasperity areas of the Kanto Region (Kobayashi et al., 2007).

Sensor Development
In the KAP, it is proposed to monitor strain, tilt, uplift and seismicity in the area underlain by asperities.This will show the relationship between plate motion, strain accumulation, and earthquake mode.
Crustal deformation throughout the hanging wall to the fault zone How the strain caused by the backstop and subducting plate affect the fault plane strain Existence of any slip across the fault ny slip across the fault Is the fault locked?Is the fault weak?Correlation with asperity inferred from 3-D seismics Importance of mesoscopic structure for dynamic rupture Slow deformation in fault zone before dynamic rupture Structure of plate boundary/splay fault Occurrence of dynamic rupture in thin slip zones (main slip) and slow ductile deformation in thick fault zones.