Aller au contenu principal
OceanData Consulting
Offering

Cable corridor atlas

Where the seabed moves, and where to route. Quantified screening before committing to a geotechnical survey, to arbitrate a route and a burial depth on physical grounds.

The principle

Surveys are expensive. Screening tells you where to run them.

A geophysical campaign costs hundreds of thousands of euros and weeks of vessel time. This atlas does not replace it: it targets it. You know where the seabed is mobile, where scour threatens, and which corridor minimizes risk before chartering anything.

500 mGrid resolution
30 yearsHindcast depth
5285Classified cells (proof tile)
51 ptsSHOM validation points
The method

From bed shear stress to the route

Nothing empirical: the chain starts at bottom-boundary-layer physics and works up to the corridor, each step traceable and reproducible.

01

Bed shear stress

Combined wave-current shear stress following Soulsby (1997), from wave orbital velocities and hourly tidal currents.

02

Motion threshold

Critical Shields threshold via the Soulsby & Whitehouse formulation, calibrated on grain size (d50) derived from Folk geology and SHOM seabed types.

03

Extreme value statistics

POT-GPD extreme value analysis over 30 years, with bootstrap confidence intervals. The output is return levels, not an average.

04

Corridor and burial

The cost-risk field feeds an A* search that traces the corridor between two landing points, followed by per-segment depth-of-lowering guidance.

Measured results

What the pipeline has already produced

Two proof tiles, North Sea and Dover Strait. The figures below come from the golden tests, not from orders of magnitude.

Validation against SHOM

Across 51 comparison points, the ratio of peak velocities falls between 0.98 and 1.03, with a phase lag under 25 minutes. The current model is calibrated.

Validation against observed dunes

Across 908 mapped dune crests, the correlation with predicted mobility zones is significant (rank-biserial +0.33, p ≈ 8e-9). Where the model says the seabed moves, it does carry sedimentary bedforms.

Robustness to grain-size uncertainty

Shifting the Folk class by one step (±1), the mobility classification stays stable across 100% of nodes in the North Sea and 86% in the Channel. The corridor itself shifts by at most 707 m, at an unchanged cost within 0.1%.

Uncertainty, stated

Return levels carry their 95% confidence interval: median half-width of roughly 12%, p90 at 33%. A figure without its uncertainty is not a result.

Packages

From the area to an arbitrated route

Choose the depth of analysis that matches your project phase. Firm quote after scoping.

Area atlas
On request
3 weeks

The sediment mobility and scour atlas over a given area, without route optimization.

  • Combined wave-current bed shear stress (Soulsby)
  • Mobility and scour maps, 500 m grid
  • POT-GPD return levels with 95% CI
  • Percentage of time mobile, per cell
Deliverables : PDF report, GeoTIFF rasters, NetCDF data, PNG figures
SoulsbyShieldsPOT-GPD
Atlas and corridorMost requested
On request
4 to 5 weeks

The atlas, plus the optimized corridor between two landing points and the Pareto alternatives.

  • Everything in "Area atlas"
  • A* optimized corridor over the cost-risk field
  • Pareto alternatives (length versus risk)
  • Per-segment depth-of-lowering guidance
Deliverables : Extended PDF report, GeoJSON/KML corridor, rasters, NetCDF
A*ParetoBurial depth
Enhanced study
On request
6 to 8 weeks

The corridor, plus calibration against your own data and a full sensitivity analysis.

  • Everything in "Atlas and corridor"
  • Calibration against your measurements (ADCP, cores, existing survey)
  • Sensitivity analysis on grain size and weights
  • Cost-risk weighting co-designed with your teams
Deliverables : Extended report, calibrated datasets, sensitivity note
CalibrationSensitivityBespoke
Limits

Stated honestly

Screening is not a survey. Here is precisely what this atlas does not do.

It is not a substitute for a survey

The atlas works from public data at 500 m resolution. It sees neither an isolated obstacle, nor a local geotechnical anomaly, nor an object on the seabed. It exists to target the survey, not to replace it.

Grain size is a proxy

The d50 is derived from EMODnet Folk classes, refined by SHOM seabed types where available. The cross-check shows a median factor of 7.9 between the Folk proxy alone and the SHOM data: this is precisely why the sensitivity analysis ships with the atlas.

Audience

Who this is for

Any routing decision taken before, or instead of, an exhaustive survey.

AudienceTriggerContact
Submarine cable layers and installersPrimary
Route and burial-depth arbitration ahead of surveyEngineering office, route engineering lead
Offshore wind developers
Inter-array cables and shore connectionStudy lead, owner's engineer
Telecommunications operators
New intercontinental link, sensitive landing pointsNetwork engineering, project management
Engineering and survey companies
Upstream scoping to size a campaignGeophysical project manager
Frequently asked

What clients ask us

Can a mobility atlas replace a geophysical survey campaign?

No. The atlas works from public data at 500 m resolution: it detects neither isolated obstacles nor local geotechnical anomalies. Its role is to target the survey, by showing where the seabed is mobile and which corridor minimizes risk, before you commit a vessel.

How do you determine whether the seabed is mobile?

Through physics, not from a map. We compute the combined wave-current shear stress following Soulsby (1997), then compare it to the critical Shields threshold (Soulsby & Whitehouse) calibrated on the site grain size. Above the threshold, sediment starts moving.

Where does your data come from?

From public, referenced sources: IFREMER RESOURCECODE for waves, Copernicus Marine (CMEMS) for hourly tidal currents, EMODnet for bathymetry and Folk geology, and SHOM for 1:50,000 seabed types. No proprietary licensed data.

How is the optimal corridor traced?

The mobility, scour and constraint layers are merged into a cost-risk field. An A* algorithm searches it for the least-cost path between the two landing points. We also deliver the Pareto alternatives: on one of our tiles, the shortest route saves 0.8 km but costs 0.6 additional risk points.

How reliable are the results?

Reliability is measured and shipped with the atlas. Validation against 51 SHOM points (velocity ratio 0.98 to 1.03), against 908 observed dune crests (p ≈ 8e-9), and classification robustness at 100% in the North Sea when grain size is shifted by one Folk class. Return levels carry their 95% confidence interval.

Does the pipeline work outside the Channel and the North Sea?

Yes, it is zone-agnostic. The two proof tiles are in the North Sea and the Dover Strait, but the chain only depends on the availability of wave, current and geology data. A new area costs one run and one configuration.

Let us scope your route

Give us two landing points and an area

We come back to you with a firm quote after a first scoping iteration. First atlas within three weeks.