9:00 AM - 9:15 AM
[S03-3-03] Land-atmosphere coupling and source of low-frequency seismic noise from the analysis of co-located barometers and seismometers
We studied the nature of coupling between atmosphere and solid Earth by examining co-located seismometers and barometers that are available from many arrays. We summarize a few characteristic features from our analysis of Pinon Flat array in Southern California and Earthscope Transportable Array (TA).
First, we can establish from data that atmospheric pressure at Earth's surface is closely related to winds, basically proportional to square of wind speeds. This pressure in turn causes vertical and horizontal motions in solid Earth. Correlation between pressure and vertical motion is high (correlation coefficient ~ 0.8 or higher), especially for frequencies below 0.05 Hz. This high correlation occurs only above a certain threshold pressure, however (Tanimoto and Valovcin, GRL, 43, 2016). Below this pressure, vertical amplitudes show no correlation with atmospheric pressure. Above this pressure the local atmosphere pressure directly controls vertical motions, showing almost perfect phase-to-phase match in waveforms. Above 0.05 Hz, this correlation starts to become smaller as ocean waves become the dominant source of seismic ground motions.
Horizontal ground motions show quite different behaviors. Horizontal seismic amplitudes change with local atmospheric pressure for the entire range of pressure that barometers can measure. It indicates that ground tilt is the dominant cause of horizontal noise. It is noteworthy that tilt effects on horizontal motions are significant even at low pressure.
Pressure vs. horizontal amplitudes show different (bifurcating) behaviors for many stations between summer and winter. In winter, noise level is always high and is constant in pressure-amplitude plots, meaning local atmospheric pressure is not responsible for seismic motions. But in summer, horizontal amplitudes show good correlation with local pressure changes and show wide variations in amplitudes as pressure changes significantly with time.
First, we can establish from data that atmospheric pressure at Earth's surface is closely related to winds, basically proportional to square of wind speeds. This pressure in turn causes vertical and horizontal motions in solid Earth. Correlation between pressure and vertical motion is high (correlation coefficient ~ 0.8 or higher), especially for frequencies below 0.05 Hz. This high correlation occurs only above a certain threshold pressure, however (Tanimoto and Valovcin, GRL, 43, 2016). Below this pressure, vertical amplitudes show no correlation with atmospheric pressure. Above this pressure the local atmosphere pressure directly controls vertical motions, showing almost perfect phase-to-phase match in waveforms. Above 0.05 Hz, this correlation starts to become smaller as ocean waves become the dominant source of seismic ground motions.
Horizontal ground motions show quite different behaviors. Horizontal seismic amplitudes change with local atmospheric pressure for the entire range of pressure that barometers can measure. It indicates that ground tilt is the dominant cause of horizontal noise. It is noteworthy that tilt effects on horizontal motions are significant even at low pressure.
Pressure vs. horizontal amplitudes show different (bifurcating) behaviors for many stations between summer and winter. In winter, noise level is always high and is constant in pressure-amplitude plots, meaning local atmospheric pressure is not responsible for seismic motions. But in summer, horizontal amplitudes show good correlation with local pressure changes and show wide variations in amplitudes as pressure changes significantly with time.