Abstract
The Groningen gas field in the Netherlands is the world's 7th largest onshore gas field and has been producing from 1963.
Since 2013, the year with the highest level of induced seismicity, the reservoir is monitored by two geophone strings at reservoir level at about 3 km depth.
For borehole SDM, 10 15-Hz geophones are positioned from 30 m above the reservoir to 240 m within the reservoir with a geophone spacing of 30 m.
We used seismic interferometry to determine, accurately as possible, the inter-geophone P-wave velocity from the ambient noise.
We used 1-bit normalization and spectral whitening, together with a bandpass filter from 3 to 400 Hz. After that, for each station pair, the normalized cross-correlation was calculated for 6 seconds segments with 2/3 overlap.
These segmented cross-correlations were stacked for every hour. Finally, 24(hours)*33(days) segments were obtained for each station pair.
So, in total, for the 10 geophones, 10! of these cross-correlation pairs were calculated for each of the 3 components.
We found that during the day the noise is dominated by cultural noise from the surface.
The cross-correlations show both day-and-night and weekly variations reflecting these fluctuations in cultural noise.
The apparent P-wave travel time for each geophone pair is measured from the maximum of the vertical component cross-correlation for each of the hourly stacks.
Because the distribution of these (24*33) picked travel times is not Gaussian but skewed, we used Kernel density estimations (KDEs) to obtain probability density functions of the travel times.
The maximum likelihood travel times of all the geophone pairs was subsequently used to determine inter-geophone P-wave velocities.
A good agreement was found between our estimated velocity structure and the P velocity structure obtained from well logging, with errors is less than 5%.
It is concluded that noise interferometry can be used to determine the seismic velocity structure from deep borehole data. It might potentially even be possible to monitor seismic variations in reservoirs if the velocity variations are sufficiently large.
Since 2013, the year with the highest level of induced seismicity, the reservoir is monitored by two geophone strings at reservoir level at about 3 km depth.
For borehole SDM, 10 15-Hz geophones are positioned from 30 m above the reservoir to 240 m within the reservoir with a geophone spacing of 30 m.
We used seismic interferometry to determine, accurately as possible, the inter-geophone P-wave velocity from the ambient noise.
We used 1-bit normalization and spectral whitening, together with a bandpass filter from 3 to 400 Hz. After that, for each station pair, the normalized cross-correlation was calculated for 6 seconds segments with 2/3 overlap.
These segmented cross-correlations were stacked for every hour. Finally, 24(hours)*33(days) segments were obtained for each station pair.
So, in total, for the 10 geophones, 10! of these cross-correlation pairs were calculated for each of the 3 components.
We found that during the day the noise is dominated by cultural noise from the surface.
The cross-correlations show both day-and-night and weekly variations reflecting these fluctuations in cultural noise.
The apparent P-wave travel time for each geophone pair is measured from the maximum of the vertical component cross-correlation for each of the hourly stacks.
Because the distribution of these (24*33) picked travel times is not Gaussian but skewed, we used Kernel density estimations (KDEs) to obtain probability density functions of the travel times.
The maximum likelihood travel times of all the geophone pairs was subsequently used to determine inter-geophone P-wave velocities.
A good agreement was found between our estimated velocity structure and the P velocity structure obtained from well logging, with errors is less than 5%.
It is concluded that noise interferometry can be used to determine the seismic velocity structure from deep borehole data. It might potentially even be possible to monitor seismic variations in reservoirs if the velocity variations are sufficiently large.
| Original language | English |
|---|---|
| Publication status | Published - 2016 |
| Event | COST action TIDES2016: Seismic noise - Sesimbra, Portugal Duration: 18 Sept 2016 → … |
Conference
| Conference | COST action TIDES2016 |
|---|---|
| Country/Territory | Portugal |
| City | Sesimbra |
| Period | 18/09/16 → … |