Passive seismic recording

Passive recordings were made available from 30 hours February 15, 2004, and 14 hours January 19, 2005. Of the former, some 7 hours was lost due to a failed disk-drive. Of the 2500 4C stations, this work uses the hydrophone measurement from each location. The data are sampled at 0.004 ms, and recorded as contiguous files approximately 15 seconds long. Figure  shows an example of the character of much of the data. The traces were balanced against each other, but are otherwise raw recordings.

 

raw1
Figure 3 Twelve seconds of representative passive hydrophone data from February 16, 2004. Trace number increases from the top left to bottom right corners of the array shown in Figure .

Figure  shows a conspicuous noise-train in a section of raw data. While there are no coherent events, a noise-train about 7 seconds long within the background chatter is reminiscent of the envelope of a shot-gather. The minimum travel time of the noise-train is at trace 650. This trace is located at the end of the top half of the fourth receiver line. The various jogs and cable terminations of this region are associated with the surface facilities on location. Plotted over the data are time picks modeled as a direct arrival from the location of the platforms in the plane of the receivers. These were shifted to align with the top and bottom of the noise train. The velocity used to model this event is 1450 m/s. The fact that there is no velocity increase for the bottom of the envelope suggests that this feature does not contain reflections from the subsurface. Similar examples of this type of noise-train are periodically recognizable throughout data recordings. The source of the energy has a complex coda and a finite duration.

 

raw2
Figure 4 Raw data with a powerful noise train envelope. The 13 lines of the array show as the event is repeated and shifted over cross-line offset. The picks are co-planar direct arrivals from the center of the array traveling at 1450 m/s.

Figure  shows a short time window when a crisp series of hyperbolic events is captured by the array. The top panel is the Eastern half of the array and the bottom is the Western half. The minimum travel-time of the hyperbolas decreases to the West of the top panel in a similar fashion to the onset of the noise section in the previous figure. This suggests that the production facilities are again the source of the energy. The time picks overlaying the data in Figure  are the kinematics of a a co-planar direct arrival from the 'O' in Figure  traveling at 1450 m/s. Every third pick was plotted to avoid clutter. However, even when plotting all picks in high resolution, the forward modeling clearly shows that the data are aliased at this slowness value over the center of the array where the event is not visible. Similar events are recognizable throughout the records, though this is a particularly clear example.

 

raw3ev
Figure 5 Raw data with a clear hyperbolic event and time picks calculated for a coplanar direct arrival from the array center traveling at 1450 m/s are.

Figure  is a map of the array rotated to convenient field coordinates. The 'X' symbols mark the locations of the traces with minimum hyperbolic travel-times that were obvious in the previous figures. The line containing the minimum travel-time picks shows no inline offset variation and includes the area around y=6250 m, which is the location of the platforms (presumably near the 'O' symbol). This figure provided the inline location for the picks plotted in Figures  & . The crossline coordinate was deduced by matching the picks to the various observations.

 

minloc Figure 6 Receiver array in field coordinates. Marked with 'X' are the locations of traces with minimum travel-times picked from the hyperbolas in Figure . Even though the event is not clear on the central lines of the array, the line connecting the traces where it is visible intersects the location of the surface facilities.