Subsea cables and pipelines: how condition monitoring actually works

Two very different asset families share the same seabed. The first is telecommunications: according to TeleGeography, more than 1.5 million kilometres of submarine cable were in service worldwide at the start of 2026. The second is energy, meaning the export cables and interconnectors that connect offshore wind farms to shore and grids to each other, plus gas pipelines. In German waters the operating offshore wind fleet alone comprised 31 fully operational wind farms with 1,764 turbines and 10,818 MW as of 30 June 2026, according to Deutsche WindGuard, and every one of those wind farms is attached to shore by cable.

On the pipeline side, the documented offshore lines relevant to German landfalls include Europipe I and Europipe II, Norpipe and Nord Stream 1. A common mistake in asset overviews is to file OPAL and NEL alongside them. Both are onshore pipelines and are not laid in the sea.

What this article deliberately does not give is an aggregate route length or a count of converter platforms in German waters. Those figures circulate widely, but no robust primary source could be established for them. For an operator the aggregate is in any case the wrong number: the planning unit is the individual route, its sections, its crossings and its landfall.

The failure modes are unglamorous and well understood, which is precisely why they are inspectable. They fall into three groups: external mechanical impact, seabed movement, and internal ageing.

• External mechanical impact: anchors and anchor drag, and bottom-contact fishing gear such as trawls. This is the dominant cause of damage in shallow, busy waters, and it is concentrated in shipping lanes, anchorages and approaches.
• Seabed mobility: sandbank migration, scour and sediment transport progressively remove the cover over a buried line. The result is first exposure, then free spans, and a free-spanning section is exposed to fatigue and vortex-induced vibration in a way a buried one is not.
• Thermal and electrical ageing: high-voltage export cables dissipate heat. Where cover is lost or sediment changes, the thermal environment changes with it, and sustained hot spots accelerate insulation degradation.
• Third-party works: dredging, construction and crossings by later projects create risk at points that were quiet when the route was surveyed.

Two things follow for an inspection programme. First, most of the risk is spatially concentrated, so a uniform inspection interval over an entire route wastes effort on sections nothing ever happens to. Second, the parameters worth tracking are cover, exposure and free span length, because those are the leading indicators, while a fault is the lagging one. This is the same logic that governs foundation and scour work described in the article on what ROV and AUV inspection delivers today.

Ask what a cable in the German exclusive economic zone has to be buried to and you will usually be told one to three metres. That number appears in trade press, in tender documents and in supplier presentations. It is worth being blunt about it: the one-to-three-metre range is not a binding requirement of the Federal Maritime and Hydrographic Agency, and there is no uniform, legally fixed metre value for depth of cover.

What actually happens is project-specific. Burial depth is determined in the planning approval procedure for the individual route, and the determination weighs the conditions of that route: seabed mobility and expected sand wave migration, shipping density and anchoring behaviour, fishing gear in use, crossings with other infrastructure, the thermal design of the cable, and the soil itself. Two routes a few dozen kilometres apart can legitimately end up with very different requirements. Quoting a generic figure back to an authority or an insurer is therefore not a compliance argument.

For condition monitoring this has a concrete and rather liberating consequence. The relevant measurement is not whether a nominal design value was met at installation, but what the as-built depth of cover is per section today and how it has moved since the last survey. A route that was laid deep and has lost half a metre of cover in one sand wave field is a different risk from a route that was laid shallower by design in stable clay and has not moved. Depth of burial measurement, repeated on a comparable track, is the data set that answers this. A single nominal number never will.

A route survey is a stack of sensors, each answering a different question, flown either by an autonomous underwater vehicle over long corridors or by a remotely operated vehicle where a section needs close attention. There is no GPS under water, so all of it depends on an acoustic position from a USBL or LBL system combined with an inertial system and a Doppler velocity log. Without that, a measurement cannot be repeated at the same place next time, and comparability is the entire point of condition monitoring.

• Multibeam echosounder: bathymetry of the corridor. Detects seabed change, scour, exposure and the geometry of free spans.
• Side-scan sonar and synthetic aperture sonar: wide acoustic imagery for debris, trawl scars, boulders and evidence of third-party activity along the route.
• Sub-bottom profiler: penetration below the seabed surface, used to establish cover over a buried line rather than the shape of the surface above it.
• Magnetometer and cable tracker: locates the line itself and measures depth of burial from the magnetic or electromagnetic signature, which is the core measurement for cover.
• High-definition optical imaging with dedicated lighting: condition of exposed sections, rock placement, mattresses, crossings and landfall structures, within whatever visibility the water allows.
• Cathodic protection measurement: potential readings against a reference electrode on pipelines and steel structures, confirming that corrosion protection is still active.

Which vehicle carries which payload is a question of geometry, not prestige. Long, uneventful corridors reward an autonomous vehicle: it flies a planned track without a vessel holding station above it, and it can repeat that track years later. Complex ground, crossings, landfall approaches and anything needing contact reward a tethered vehicle with a pilot. Most real programmes use both, with the autonomous pass generating the anomaly list and the tethered visit resolving it. Vehicle categories and typical payload capacities are collected in the robot catalogue.

Power cables normally contain optical fibres, and those fibres can be used as a distributed sensor along the entire route without any vehicle in the water. Two techniques dominate. Distributed temperature sensing (DTS) produces a thermal profile along the cable, which reveals local hot spots caused by lost cover, changed sediment, or sustained electrical loading. Distributed acoustic sensing (DAS) evaluates variations in laser backscatter to detect vibration and acoustic events, which in practice means anchor drag, dredging activity, impacts and construction work near the route.

Both are genuinely useful and both are routinely oversold. What they deliver is an event, a time and a position along the fibre. What they do not deliver is a diagnosis. A DAS alert says something happened at kilometre 42; it does not say whether the armouring is damaged, whether the cover has been displaced, or whether nothing of consequence occurred at all. Answering that still requires someone to go and look, which is why the practical value of permanent sensing lies in targeting survey effort rather than replacing it. An operator with DAS can send a vehicle to a specific kilometre within days instead of scheduling a full route survey and finding the damage eighteen months later.

Two limits are worth naming. Gas pipelines usually have no integral fibre, so the equivalent capability requires a separately laid sensing fibre or falls back to periodic survey. And a permanent system generates a continuous data stream that has to arrive somewhere useful: alongside the maintenance history and the previous survey results, not in an isolated dashboard that nobody reads between incidents. Where landfall sections and coastal approaches are involved, aerial inspection can complement the subsea data, and wedrone offshore covers that side. wedrone does not fly itself and holds no aviation certificates of its own; any operation runs through approved partner operators.

Until recently, recurring condition monitoring of subsea cables was largely an economic decision. It is becoming a documented obligation, and the drivers are civil resilience and continuity of supply rather than anything else. The consequences for an inspection programme in the subsea vertical are practical rather than abstract.

The most direct of them in Germany is the KRITIS umbrella act, in force since 17 March 2026. It explicitly names maritime infrastructure, including installations on and under the sea, and requires operators to perform risk analyses that cover those installations. A risk analysis for an asset whose condition is unknown is not a serious document, which is why the act indirectly makes recurring condition data a practical necessity rather than a nice-to-have.

At European level, the NIS2 Directive (EU) 2022/2555 addresses the resilience of essential entities, and its recital 97 refers specifically to submarine cables. The German transposition, the NIS2 implementation act, has been in force since 6 December 2025. Alongside it, the Commission published the EU Action Plan on Cable Security, JOIN(2025) 9 final, on 21 February 2025, building on Commission Recommendation (EU) 2024/779 of 26 February 2024. One point of housekeeping here, because the error is widespread in secondary sources: there is no Council Recommendation on Submarine Cable Infrastructure Security. Documents citing one are citing something that does not exist.

• Establish a defensible baseline for the route, section by section, rather than an average over its length.
• Derive the inspection interval from exposure and traffic on each section, not from a single figure for the whole asset.
• Keep survey data comparable between campaigns, which means fixing the track, the sensor configuration and the positional reference in advance.
• Record who assessed which finding and on what basis, since the technical evaluation stays with the qualified expert.

The technical part of subsea condition monitoring is largely solved. The part that fails in practice is organisational: surveys commissioned in different formats from different contractors, positional references that do not line up, anomaly lists that cannot be compared with the previous campaign, and evidence that satisfies nobody when it is actually needed.

werob is a manufacturer-independent systems integrator for service robotics and a brand of CITO GmbH in Hamburg, and this is the layer it works on. Not the survey itself, and not the assessment, but the specification and the interfaces between them: translating the obligation into a technical requirement, comparing vehicle classes and sensor payloads across manufacturers rather than inside one product line, checking data formats and positional references before procurement, and making sure the results land in the operator's existing asset and maintenance systems in a form that is still usable three campaigns later.

The economics of that are portfolio-specific. The variables are vessel time, the survey interval per section, the share of the route that genuinely needs annual attention, and the value of finding a lost metre of cover before it becomes a fault rather than after. The ROI calculator is the place to put figures against a concrete asset base. What no integrator or vehicle changes is where responsibility sits: the technical evaluation of a finding remains with the qualified expert, and a sonar record is an indication, not a diagnosis.