In this presentation, we report on the development and evaluation of sensors for Long-Term Borehole Monitoring System (LTBMS) installed at Site C0002 as the first permanent observatory in the Nankai Trough during IODP Expedition 332. The suite of LTBMS sensors includes a broad-band seismometer, volumetric strainmeter, tiltmeter, geophone, accelerometer and thermistor array. The set of sensors was designed to collect broad-band dynamics with wide dynamic range to understand the mechanism of mega-earthquake occurred along the plate boundary faults repeatedly. As a result of dummy run test during IODP Expedition 319, we found that Vortex Induced Vibration (VIV) in high-current "Kuroshio" caused serious damage to sensors. Therefore, we need sensors which have not only high-sensitivity but also toughness enough to survive against VIV. In our development scheme for sensors, broad-band seismometer, CMG3T (manufactured by Guralp Systems Ltd.), tiltmeter, LILY (AGI.) and thermistor digitizer, SAHF (Kaiyo-denshi) were prepared by customizing general products for our purpose. The volumetric strainmeter was designed by JAMSTEC and assembled by Seismotec Corp. from scratch. We developed some PCBs for telemetry, A/D conversion and calibration. Geophone, GS-11D (OYO geospace) and accelerometer, JA-5H200 (JAE) were integrated to our PCBs. Furthermore, we implemented anti-vibration mechanism to all developed sensors. After development was completed, noise evaluation, vibration and shock tests were conducted using tiltmeter, geophone, accelerometer and thermistor digitizer. The purpose of these tests is to confirm that our anti-vibration mechanism is working well by comparison of sensor response before and after vibration and shock tests. The vibration test was conducted with a sweep vibration from 3 to 15 Hz in frequency, and from 0.25 to 2.0 G in acceleration calculated from accelerometer data measured on IODP Expedition 319. The shock pulse test was conducted with 90 G shock pulse and 2 ms pulse width. After these tests were completed, we installed all sensors to Matsushiro Seismological Observatory and conducted noise evaluation test before and after vibration and shock tests. As for the broad-band seismometer CMG3T, vibration test was conducted by Guralp Systems Ltd. in U.K. with same parameters as our test. After delivered, noise evaluation test was conducted in Matsushiro as other sensors. As a result of the noise evaluation test, Power Spectral Density (PSD) was calculated using background noise for sensor evaluation. We confirmed that PSD plots of all sensors have same response before and after these tests. The peak of microseism around 0.2 Hz can be clearly confirmed in the PSD plot of the CMG3T, geophone and tiltmeter. Finally, we confirmed that the responses of these sensors have good coherence with that of the reference sensor which was general package of land-type CMG3T and this result is consistent with each specification of sensors. After final noise evaluation test was completed, all of sensors were loaded to D/V Chikyu for IODP Expedition 332, and were successfully installed to Site C0002 observatory in December, 2010.