Optical fibre data registration

Kevin W. Hall and Don C. Lawton

ABSTRACT

The Containment and Monitoring Institute (CaMI) Field Research Site (FRS) has three wells on the lease, referred to here (from SW-NE) as the geophysics, injection and geochemistry wells. Borehole and trenched optical fibres are connected in a continuous loop of ~5 km length in the following order: 1) helical in geophysics well, 2) straight fibre in geophysics well, 3) straight fibre in geochemistry well, 4) straight fibre from geophysics well to south end of trench, 5) helical fibre for entire length of trench and 6) straight fibre from the north end of the trench back to the geophysics well. Since we know the trace spacing for each survey, we can assign coordinates to traces once we know the position of any given trace. Tap tests in above ground junction boxes can be spread over as many as 100 traces due to gauge length effects and are not precise enough for this purpose. High-amplitude noise observed at above ground junction boxes also spreads across variable numbers of data traces, depending upon source distance from the junction box. Separation of continuous loop data into discrete datasets has been performed by running a modified STA/LTA algorithm on the fourth power of the sum of the absolute value of uncorrelated trace amplitudes to locate the edges of junction box noise, which also gives us a starting point for determining trace location. Figure 1 shows a correlated record with 0.75 m trace spacing muted using STA/LTA results for all source gathers in this example.

For borehole registration we have calculated least-squares hyperbolic fits to first-break picks to determine which is the deepest trace in each well. Tests of this process give a result with a standard deviation of one trace for straight fibre and three traces for helical fibre for a survey acquired with 0.25 m trace spacing and 10 m gauge length.

View full article as PDF (0.85 Mb)