Submarine telecom / 2025
A cable route across an active fault, surveyed twice at nine months to see what had moved.
- Client
- Nordband Cable Consortium
- Vessel
- HALVARD-9
- Duration
- 34 days
- Position
- 64.8801 N / 4.2216 E
0
Displacement detected
Between campaigns, nine months apart
0
Route section rejected
Active slump, downslope edge
0
Line surveyed
Across both campaigns
The brief
Nordband had a preferred route for a repeater-bearing segment and a geotechnical report that described the slope as stable. The report was eleven years old and predated two moderate seismic events in the region.
The consortium did not want a second opinion on the report. They wanted the seafloor measured, and they wanted the same measurement repeated far enough apart in time to show whether anything was in motion.
The approach
Two survey campaigns nine months apart, flown to the same line plan by the same hull, so that differences in the returns could be attributed to the seabed rather than to the survey.
HALVARD-9 ran 412 line kilometres per campaign at 2.1 knots, holding altitude at 45 metres above the bed. Sub-bottom profiling ran continuously to resolve sediment layering above the fault trace.
The second campaign was differenced against the first at full resolution. Nothing was smoothed, averaged or resampled before the comparison, because both of those operations hide exactly the signal being looked for.
What the returns showed
A 2.3 kilometre section of the preferred route crosses a slump scar that is still moving. Vertical displacement between campaigns reached 1.4 metres at the scar's downslope edge.
Sediment layering shows three prior failure surfaces within nine metres of the bed, which is a recurrence interval, not a one-off event.
A parallel corridor 900 metres east crosses the same fault on exposed bedrock with no measurable displacement between campaigns.
We asked for a survey and got an argument we could not talk our way around. The route moved 900 metres east and nobody had to be persuaded.
Similar water
Trench Line 4 took 34 days and produced findings the client would rather not have had. That is the point of paying for a survey.