11:00 〜 13:00
[SSS03-P10] 広帯域地震計センサーの傾斜とその特徴
キーワード：広帯域海底地震計、傾斜ノイズ
Since the development of broadband ocean bottom seismometer (BBOBS), many seafloor observations by BBOBS have been deployed. They are subject to substantial noise from water waves and bottom currents, especially long periods. Crawford and Webb (2000) suggested that vertical-component seismic noise of BBOBS can be reduce by subtracting the coherent signals derived from horizontal seismic observations associated with tilt noise, and pressure measurements related to infragravity waves. While it is already common to apply this noise reduction method in oversea BBOBS observations, it has been recently applied in Japan.
To apply the tilt noise reduction method, we calculate the horizontal-to-vertical transfer function, which enables us to estimate tilt angle of vertical component of seismic sensor: the arcsine of the amplitude of the transfer function.
We applied noise reduction methods to more than 80 BBOBS dataset in Japan and estimated the tilt angle and direction of the vertical component of sensor.
Estimated tilt angles are between 0.02 and 2.4 degrees, and those average and one standard deviation is 0.9 and 0.4 degrees, respectively. The direction of tilt is around the NS output of the seismic sensor. The tilt angle and direction of each BBOBS are stable during its one-year seafloor observation, and those of the BBOBS unit using the same sensor and leveling unit are similar in different observations. These suggest that estimated tilt is considered to be unique to each BBOBS unit.
The BBOBS in Japan contains all components in the titanium alloy sphere housing with 650 mm. CMG-3T of 360 s (Guralp systems) is installed on the leveling unit to keep it in level within ±0.3° during the observation. These tilt and mass position signal output of the seismic sensor are recorded once a day during observation. These outputs showed that tilt angle fluctuation is small. Shiobara et al. (2021) reported that tilt change in 1.5 or 2 years-long seafloor observation is up to several tens of μradian. These suggest that tilt change on the seafloor can be negligible to estimate the tilt of the sensor.
The estimated directions of the tilt are around NS output whereas tilt direction measured by leveling unit are arbitral direction. It is because the vertical component of the CMG-3T is designed to oscillate along NS direction.
Whereas the measured tilts are less than ±0.3°, estimated tilt are larger than that. In some case, we found that horizon of leveling unit was misaligned and up to one degree of tilt was remained. Even when we correct this misalignment, several tenth of degree difference are remained. In some case, differences are more than one degree.
We think that the estimated tilt angle from noise reduction method may be overestimated. The tilt noise reduction method assumes that noise source, such as seafloor current, is completely horizontal. If the noise source is deviate from horizon, tilt angle of source is estimated. One possible explanation of the difference between measured and estimated tilt is that the seafloor current is slightly tilted along with seafloor topography.
To apply the tilt noise reduction method, we calculate the horizontal-to-vertical transfer function, which enables us to estimate tilt angle of vertical component of seismic sensor: the arcsine of the amplitude of the transfer function.
We applied noise reduction methods to more than 80 BBOBS dataset in Japan and estimated the tilt angle and direction of the vertical component of sensor.
Estimated tilt angles are between 0.02 and 2.4 degrees, and those average and one standard deviation is 0.9 and 0.4 degrees, respectively. The direction of tilt is around the NS output of the seismic sensor. The tilt angle and direction of each BBOBS are stable during its one-year seafloor observation, and those of the BBOBS unit using the same sensor and leveling unit are similar in different observations. These suggest that estimated tilt is considered to be unique to each BBOBS unit.
The BBOBS in Japan contains all components in the titanium alloy sphere housing with 650 mm. CMG-3T of 360 s (Guralp systems) is installed on the leveling unit to keep it in level within ±0.3° during the observation. These tilt and mass position signal output of the seismic sensor are recorded once a day during observation. These outputs showed that tilt angle fluctuation is small. Shiobara et al. (2021) reported that tilt change in 1.5 or 2 years-long seafloor observation is up to several tens of μradian. These suggest that tilt change on the seafloor can be negligible to estimate the tilt of the sensor.
The estimated directions of the tilt are around NS output whereas tilt direction measured by leveling unit are arbitral direction. It is because the vertical component of the CMG-3T is designed to oscillate along NS direction.
Whereas the measured tilts are less than ±0.3°, estimated tilt are larger than that. In some case, we found that horizon of leveling unit was misaligned and up to one degree of tilt was remained. Even when we correct this misalignment, several tenth of degree difference are remained. In some case, differences are more than one degree.
We think that the estimated tilt angle from noise reduction method may be overestimated. The tilt noise reduction method assumes that noise source, such as seafloor current, is completely horizontal. If the noise source is deviate from horizon, tilt angle of source is estimated. One possible explanation of the difference between measured and estimated tilt is that the seafloor current is slightly tilted along with seafloor topography.