Offshore platform jacket installation and marine construction monitoring
SAR and very-high-resolution optical imagery provide independent verification of heavy-lift and lay-vessel positioning during offshore jacket and pipeline installation, cross-referenced against AIS to resolve vessel identity at declared coordinates.
Sensors
- Sentinel-1 C-band SAR (ESA): 5 x 20 m resolution in Interferometric Wide Swath mode, 250 km swath, 6-day repeat at the equator (1-3 days at higher latitudes with both satellites). Detects large vessels reliably; struggles to distinguish vessel type at this resolution without AIS cross-reference.
- ICEYE X-band SAR: Spotlight mode delivers approximately 0.5 m azimuth resolution, sufficient to resolve deck geometry and distinguish crane vessels from lay barges by superstructure signature. Revisit is task-dependent but the constellation supports sub-daily access over a fixed point.
- Planet SkySat: 0.5 m panchromatic, 0.8 m multispectral. Tasked optical imagery in clear conditions; confirms vessel colour, markings and deck configuration that SAR cannot provide. Revisit is task-driven, typically same-day or next-day with commercial priority.
- Maxar WorldView Legion: 30 cm panchromatic resolution, the highest commercially available optical. Up to 15 revisits per day over a target in mid-latitudes. Resolves individual deck cranes, mooring lines and jacket skid-beams when cloud permits.
Why independent positioning evidence matters offshore
Offshore jacket installation is a contractual minefield. Installation contractors invoice milestone payments tied to declared vessel positions and completion of discrete installation events: jacket upending, pile driving, hook-up. Project financiers, insurers and government regulators often have no independent means of verifying that a vessel was where the contractor said it was, when they said it was there.
AIS solves part of this problem. The Automatic Identification System broadcasts GPS-derived position, heading and speed for vessels above 300 gross tonnes, and the data is publicly archived. But AIS can be spoofed, switched off, or simply wrong due to GPS error in congested anchorages. Satellite imagery provides a physics-based check: the vessel either appears at those coordinates or it does not.
What a SAR return tells you about a vessel, and what it cannot
SAR backscatter from a steel vessel is dominated by corner-reflector geometry: the junction of vertical hull plating and flat deck surfaces produces very bright returns regardless of weather or time of day. This makes vessel detection reliable in Sentinel-1 data even at 5 x 20 m resolution. A 200-metre semi-submersible crane vessel will appear as a bright elongated blob, clearly distinguishable from sea clutter.
Vessel type identification is a different matter. At Sentinel-1 resolution, a crane vessel, a pipe-lay barge and a heavy-transport vessel can produce similar backscatter envelopes. The geometry is simply too coarse to resolve deck superstructure. ICEYE Spotlight mode changes this: at sub-metre azimuth resolution, the distinctive A-frame or gantry crane of a derrick lay vessel produces a characteristic high-backscatter spike absent from a flat-deck jacket barge. Even so, confident type identification from SAR alone requires analyst experience and should be cross-referenced with AIS vessel identity before being cited in a contract dispute.
Cloud is irrelevant to SAR. For offshore construction in the North Sea, Norwegian Sea or South China Sea, where overcast conditions are frequent, this is not a minor advantage.
AIS cross-referencing: closing the identity gap
The practical workflow pairs a SAR detection with contemporaneous AIS records. If a vessel's MMSI number is broadcasting from coordinates within 200 metres of the SAR centroid at the image acquisition time, identity is effectively confirmed. The position uncertainty of Sentinel-1 geolocation is typically better than 10 metres after terrain correction, so a 200-metre tolerance is generous.
Discrepancies are where the analysis earns its value. A vessel appearing in imagery but absent from AIS at those coordinates may indicate AIS spoofing or equipment failure. A vessel whose AIS track shows it at the declared installation site but which does not appear in a cloud-free optical image, or produces no SAR return, warrants immediate investigation. These are not hypothetical scenarios: AIS dark periods and position manipulation are documented in published maritime domain awareness literature.
For offshore wind-farm construction, where installation windows are narrow and turbine foundation sequencing is contractually fixed, the ability to confirm that a specific jack-up vessel was over a specific monopile location on a specific date has direct financial consequences.
Pipeline lay vessels and the subsea construction problem
Pipe-lay vessels present a particular challenge. The vessel is visible; the pipe going over the stern is not. Subsea pipeline position cannot be confirmed by optical or SAR imagery. What satellite data can confirm is that a lay vessel of the correct type was stationary or moving at lay speed (typically 1 to 4 km per day for S-lay) along the declared route corridor, on the declared dates.
Speed-over-ground derived from sequential SAR acquisitions or AIS provides a plausibility check. A vessel supposedly laying pipe at 2 km per day but whose AIS track shows 8 knots of transit speed was not laying pipe. Conversely, a vessel moving at the correct speed along the correct azimuth, confirmed by two SAR passes 6 hours apart, provides strong corroborating evidence that installation was proceeding as declared.
ICEYE's sub-daily revisit capability is particularly useful here. Two Spotlight acquisitions bracketing a 12-hour lay window can establish start and end positions to within tens of metres, giving a direct measure of the pipe laid in that shift.
Resolution, revisit and the honest limits of the method
No satellite sensor currently resolves individual piles being driven or confirms that a jacket has been set to the correct orientation on the seabed. The installation event itself is below the detection threshold. What imagery confirms is vessel presence, vessel type (at ICEYE or WorldView resolution), and vessel movement consistent or inconsistent with the declared activity.
Cloud cover limits optical sensors. In the southern North Sea, cloud-free conditions occur on roughly 30 to 40 per cent of days in winter months, which means WorldView Legion or SkySat tasking cannot guarantee a clear acquisition on any given day. SAR removes this constraint but introduces the type-ambiguity problem described above. A monitoring programme that combines both sensor types, using SAR as the primary all-weather layer and optical as the periodic confirmation layer, is more reliable than either alone.
Archive depth matters for retrospective claims. Sentinel-1 data is publicly archived from 2014, and commercial providers including ICEYE and Planet retain tasked imagery under client licence. A dispute arising two years after installation can still be investigated if the right tasking was commissioned at the time.
Putting it into practice
A monitoring programme for a major jacket installation campaign typically requires: a defined vessel watch-list with MMSI numbers and vessel dimensions; a tasking schedule aligned to key installation milestones; a SAR baseline covering the full campaign period; and optical tasking on milestone dates when cloud permits. Outputs are timestamped position reports with imagery evidence, AIS correlation tables, and a deviation log flagging any discrepancy between declared and observed vessel state.
Satellize structures this kind of programme as a scheduled analytics service, drawing on open Sentinel-1 data for daily coverage and adding commercial ICEYE or SkySat tasking at milestone events. The methodology is the same class of satellite-derived maritime monitoring applied in the Tonga crop-estimation programme's vessel-tracking component, adapted for the specific geometry of offshore construction. Clients receive a structured evidence package suitable for submission to project financiers or regulatory bodies, not a dashboard that requires interpretation.
If you are considering this for a specific campaign, the most useful first step is a site-geometry review: latitude determines Sentinel-1 revisit, local weather statistics determine optical reliability, and installation schedule determines where commercial tasking budget is best concentrated.
Typical figures
| SAR spatial resolution (Sentinel-1 IW mode) | 5 m range x 20 m azimuth |
| SAR spatial resolution (ICEYE Spotlight) | ~0.5 m azimuth x ~0.5 m range |
| Optical resolution (SkySat) | 0.5 m panchromatic, 0.8 m multispectral |
| Optical resolution (WorldView Legion) | ~0.30 m panchromatic |
| Sentinel-1 revisit (mid-latitudes, dual satellite) | 1 to 3 days |
| ICEYE revisit (tasked) | Sub-daily possible; constellation-dependent |
| Minimum detectable vessel length (Sentinel-1) | ~50 m in calm to moderate sea state |
| Sentinel-1 archive depth | From 2014 (publicly accessible via Copernicus Data Space) |
| Geolocation accuracy (Sentinel-1, terrain corrected) | Better than 10 m CE90 |
| Delivery format | Timestamped position reports, GeoTIFF imagery chips, AIS correlation tables, PDF evidence package |
Analytics Satellize can run
| Vessel presence confirmation | SAR ship detection (CFAR thresholding on backscatter intensity) cross-referenced with AIS MMSI records | Timestamped position report with imagery chip and AIS correlation table, per milestone event |
| Vessel type classification | SAR backscatter morphology analysis (superstructure signature at ICEYE Spotlight resolution) supplemented by optical deck-geometry review | Vessel type confidence assessment (confirmed / probable / ambiguous) with supporting imagery |
| AIS discrepancy alert | Spatial comparison of SAR-detected centroid against contemporaneous AIS broadcast position; flagging of dark periods or position offsets exceeding 500 m | Deviation log with timestamp, offset distance and recommended follow-up action |
| Lay-vessel progress estimation | Sequential SAR position differencing to derive speed-over-ground and heading; comparison against declared lay-rate and route azimuth | Daily lay-progress table with estimated pipe-laid distance and route conformance score |
| Installation milestone evidence package | Multi-sensor composite (SAR + optical where cloud-free) assembled around declared milestone dates; analyst-annotated imagery | PDF evidence dossier formatted for submission to project financiers or regulatory bodies |
| Campaign-period archive review | Retrospective processing of Sentinel-1 archive and any commercially tasked imagery over the installation area for the full campaign duration | GIS layer of all detected vessel positions with timestamps, exportable as GeoJSON or shapefile |
Who does the work
We can get this done for you. Satellize runs its own analyst desk and a strong science team. You do not buy a data feed and work out what it means; our people source the imagery, run the analysis described on this page, and hand you the answer with its confidence limits stated. Discuss this requirement.