Offshore Platform Supply Vessel Activity Monitoring
SAR imagery and spaceborne AIS together reveal how often platform supply vessels call at offshore installations, exposing production tempo, maintenance cycles and potential sanctions violations where AIS goes dark.
Sensors
- Sentinel-1 SAR (C-band, ESA): 5 x 20 m ground resolution in Interferometric Wide Swath mode, 250 km swath, 6-day repeat at the equator and 1-3 days at higher latitudes. Detects platform structures and vessels down to roughly 20-30 m length in calm-to-moderate sea states. Free archive from 2014; all-weather, day-night.
- ICEYE SAR (X-band, commercial): Spotlight mode delivers approximately 0.5 m azimuth resolution, enabling vessel length and beam estimation. Tasked collection on demand; latency to first image can be under two hours for priority orders. Particularly useful for resolving vessel-platform proximity in congested fields.
- Spire spaceborne AIS: Global AIS message collection from a constellation of over 100 LEO satellites. Provides vessel MMSI, name, type, speed and position when transponders are active. Revisit is effectively continuous at busy offshore locations; latency to delivery is typically minutes to low tens of minutes.
- Planet SuperDove (optical, 3-5 m): Eight spectral bands, daily revisit over most offshore fields. Useful for confirming vessel hull colour and type under clear skies, and for detecting hydrocarbon sheen around platforms as a secondary indicator of operational status. Cloud cover is a genuine constraint at high-latitude fields.
Why supply vessel visits are a production proxy
Offshore oil and gas platforms do not operate in isolation. They require a steady rhythm of platform supply vessels delivering fuel, chemicals, drilling mud, food and crew, and returning with waste, produced water and equipment. That rhythm is tightly coupled to production status. A platform running at full capacity typically receives PSV calls every two to four days. A platform in planned maintenance may see a burst of visits followed by a quiet period. A platform that has been quietly shut in, for whatever reason, shows a different pattern: infrequent visits, often by a single support vessel rather than the usual rotation.
This makes PSV visit frequency a surprisingly legible production signal, one that can be read from orbit without any cooperation from the operator. The signal does not tell you the precise production rate in barrels per day. It does tell you whether a platform is actively worked, in a maintenance cycle or essentially idle, and it does so with a time resolution that financial disclosures and shipping manifests cannot match.
What SAR actually sees at an offshore installation
Synthetic aperture radar works by measuring the intensity of microwave backscatter from the sea surface and any objects on it. Metal structures return very strong signals. A typical semi-submersible or jacket platform appears as a bright cluster of point scatterers; a PSV alongside appears as an elongated bright return adjacent to it. In Sentinel-1 IW mode at 5 x 20 m resolution, a 70-metre PSV is detectable as a distinct object, though its identity cannot be confirmed from geometry alone. ICEYE Spotlight imagery at sub-metre azimuth resolution can resolve vessel superstructure detail well enough to estimate length and, in some cases, distinguish vessel classes.
SAR has two genuine limits here. First, very small vessels, workboats under roughly 20 metres, can fall below the detection threshold in high sea states when wave clutter raises the noise floor. Second, when a PSV is moored directly alongside a large platform, its radar return can merge with the platform's own strong return, making it hard to count vessels confidently. Analysts typically use the shadow region and the geometry of the return to separate platform from attendant vessel, but this requires care and is not fully automated at scale.
AIS tells you who is there; SAR tells you when AIS lies
Spire's spaceborne AIS constellation collects VHF transponder messages globally. When a PSV transmits normally, you get its MMSI, registered name, flag, vessel type and position, updated every few seconds while underway. Cross-referencing AIS tracks against SAR detections at a platform gives you a confirmed vessel identity, a departure time and, by extension, a visit duration.
The problem is that AIS can be switched off. This is legal in certain circumstances and illegal in others, but it happens. A vessel operating at a platform subject to sanctions may disable its transponder before approaching. SAR detects the vessel regardless. The combination of a SAR-confirmed vessel presence with a gap in AIS coverage at that location and time is itself an intelligence finding. You cannot confirm the vessel's identity from SAR geometry alone in most cases, but you can confirm that something of PSV size was alongside a restricted platform during an AIS blackout, and that is frequently enough to trigger further investigation.
Building a visit-frequency time series
The analytic product is straightforward in concept. For a given platform or field, you accumulate SAR detections over time, flag each visit event (vessel present alongside or within 500 metres of the installation), and build a time series of visit frequency. Sentinel-1's free archive back to 2014 allows baseline characterisation: what is the normal visit rate for this platform in this season? Deviations from that baseline are the signal of interest.
In practice, the Sentinel-1 revisit of 1-3 days at North Sea or Norwegian Sea latitudes gives reasonable temporal resolution. At equatorial fields, the 6-day repeat is coarser and misses short visits. Commercial SAR tasking from ICEYE or similar systems can fill gaps when a specific platform warrants closer attention. The honest position is that a 6-day revisit will undercount visits; you are measuring visit frequency with a sampling rate, not a continuous watch, and your reported figures should reflect that uncertainty.
Satellize applies this time-series approach operationally, drawing on the same open-constellation methodology used in its Tonga crop-estimation programme, adapted for offshore detection rather than agricultural reflectance.
Sanctions monitoring: the specific case for dark-vessel detection at restricted platforms
Several offshore installations are subject to international sanctions that prohibit servicing by vessels flagged to, or owned by, entities in certain jurisdictions. Monitoring these platforms for PSV activity is a direct sanctions-compliance application. The workflow is: identify the restricted platform's coordinates, task SAR collections at the highest available revisit, compare any detected vessel returns against AIS records for the same time window, and flag cases where a vessel is present but unidentified.
This is not a definitive sanctions-violation finding. SAR geometry can confirm approximate vessel size and heading but rarely confirms identity without AIS or optical corroboration. A Planet SuperDove pass on the same day can sometimes resolve hull colour or markings that narrow the candidate list. The output is an anomaly report, not a legal determination, and analysts should be explicit about that distinction when briefing clients.
Honest limits of the method
Cloud cover does not affect SAR but does affect optical confirmation passes. High sea states above roughly 4-5 metres significant wave height raise the SAR noise floor and reduce small-vessel detection probability. Very short visits, a vessel alongside for less than four hours, may fall entirely between two Sentinel-1 passes and go unrecorded unless commercial SAR is tasked. Platform-mounted cranes and flare booms create radar shadow patterns that can occasionally mask a moored vessel on the lee side. Finally, some remote fields at low latitudes have Sentinel-1 coverage gaps where no ascending or descending track provides adequate geometry; checking coverage before committing to a monitoring programme is essential.
Typical figures
| SAR spatial resolution (Sentinel-1 IW) | 5 m range × 20 m azimuth |
| SAR spatial resolution (ICEYE Spotlight) | ~0.5 m azimuth, ~1 m range |
| Sentinel-1 revisit (mid-to-high latitudes) | 1-3 days (two-satellite constellation) |
| Sentinel-1 revisit (equatorial) | ~6 days |
| ICEYE tasking latency | Under 2 hours for priority collections |
| Minimum detectable vessel length (SAR, moderate sea state) | ~20-30 m |
| Spire AIS message latency | Minutes to low tens of minutes |
| Sentinel-1 archive depth | From April 2014 (Sentinel-1A launch) |
| Planet SuperDove optical resolution | 3-5 m, 8 spectral bands |
| Planet SuperDove revisit | Daily over most offshore fields, cloud-permitting |
Analytics Satellize can run
| PSV visit-frequency time series per platform | SAR vessel detection (CFAR thresholding) combined with proximity-to-installation geometry; time-series aggregation | Monthly CSV or GeoJSON time series with visit count, vessel count per event and confidence flag |
| AIS-dark vessel presence alert | SAR detection cross-referenced against Spire AIS; gap detection where no MMSI matches the SAR return within 500 m of platform | Near-real-time alert (email or API push) with SAR scene timestamp, platform ID and estimated vessel dimensions |
| Platform operational-status classification | Visit-frequency deviation from historical baseline; thresholds set per platform from Sentinel-1 archive | Weekly status report: active, reduced-activity, maintenance-pattern or idle, with supporting imagery |
| Vessel identity confirmation report | AIS track matching plus optional Planet optical pass for hull-colour and marking corroboration | PDF or structured JSON report per vessel event, including MMSI (where available), flag state and visit duration estimate |
| Field-level supply-vessel traffic density map | Multi-epoch SAR detection aggregation across all installations in a defined field boundary | GIS layer (GeoTIFF or Shapefile) showing visit-count heat map by platform, updated monthly |
| Anomaly dossier for sanctions-screening workflow | Dark-vessel events combined with flag-state and ownership screening against public sanctions lists | Structured anomaly dossier per flagged event, suitable for compliance team review; explicitly labelled as indicative, not determinative |
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.