Sanctions-evading ship-to-ship oil transfer detection
Ship-to-ship oil transfers in international waters are a documented mechanism for laundering sanctioned crude from Iran, Venezuela and North Korea. Fusing wide-area SAR search with tasked high-resolution SAR confirmation exposes the transfers that AIS silence is designed to hide.
Sensors
- Sentinel-1 (ESA): C-band SAR, IW mode at 10 m ground resolution, 250 km swath. Revisit of 6 days per satellite over most ocean areas, reducing to 3 days with both satellites active. Used for wide-area search: detects vessel presence and proximity pairing but cannot reliably resolve mooring lines or fenders at this resolution.
- ICEYE SAR constellation: X-band SAR with Spot mode delivering approximately 0.5 m resolution and Strip mode at 3 m. Tasked on demand; typical tasking-to-delivery latency under 12 hours for priority orders. At 0.5 m, fender clusters and mooring geometry between hulls become visible, supporting confirmation imagery.
- Capella Space SAR: X-band SAR with Spotlight mode at approximately 0.5 m resolution. Similar on-demand tasking model to ICEYE. Independent constellation provides redundancy when cloud or orbital geometry prevents a single provider from collecting on a given pass.
- Airbus Pléiades Neo: Optical, 30 cm panchromatic resolution. Useful for confirmation in clear-sky conditions: hull markings, flag state, and hose-handling deck equipment are readable at this resolution. Useless through cloud or at night, which is precisely when many transfers are scheduled.
Why oil moves between hulls in the middle of the ocean
Sanctioned crude does not vanish at the point of embargo. It moves. UN Panel of Experts reports on Iran, Venezuela and North Korea have repeatedly documented ship-to-ship transfers as the primary mechanism for obscuring cargo origin. A sanctioned tanker offloads to a second vessel in international waters, often in known anchorage zones off the coasts of Malaysia, the UAE, West Africa or the eastern Mediterranean. The receiving vessel then proceeds to a buyer with a clean manifest and an AIS history that begins conveniently after the transfer.
The scale is not trivial. UN reports have described Iranian crude exports persisting at hundreds of thousands of barrels per day despite sanctions, with STS transfers in the Strait of Malacca approaches and off Fujairah cited explicitly. The transfers take hours, sometimes a full day for a VLCC cargo. That duration is the detection opportunity.
What a floating roof gives away
Radar does not care about darkness or cloud cover, which is why SAR is the primary sensor here. A vessel's superstructure, cranes, and deck equipment produce strong, characteristic radar returns in X-band and C-band imagery. Two VLCCs lying parallel at a separation of 5 to 15 metres produce a distinctive double-bright-line signature in SAR, quite different from vessels in normal transit or anchored independently. At Sentinel-1's 10 m IW resolution, the pair reads as an anomalously wide or elongated bright return. At ICEYE or Capella's 0.5 m Spotlight resolution, the individual structural elements of each hull are separable.
The physics matters for honest interpretation. C-band (5.4 GHz, roughly 5.5 cm wavelength) penetrates light rain and thin cloud. X-band (9.6 GHz, roughly 3.1 cm) gives finer resolution but is more attenuated by heavy precipitation. Neither wavelength reveals what is inside the cargo tanks. SAR confirms proximity and duration of contact; it does not confirm that hydrocarbons changed hands. That inferential step requires corroborating signals.
AIS silence as a corroborating signal, not a primary one
Vessels conducting illicit STS transfers typically disable or spoof their AIS transponders before and during the event. An AIS gap at a known anchorage zone, coinciding with a SAR detection of two vessels in close proximity, is strong corroborating evidence. It is not, by itself, proof: AIS outages occur for legitimate technical reasons, and SAR detects all vessels regardless of their legal status.
Spire Global and Aireon operate space-based AIS (S-AIS) receivers that provide near-global coverage with latency measured in minutes to low tens of minutes, far better than coastal VTS coverage. Comparing S-AIS vessel tracks against SAR detections allows analysts to flag vessels whose transponder history shows a gap precisely at the location and time of a SAR-detected proximity event. The combination of spatial, temporal and behavioural signals is what moves an observation from 'anomaly' to 'probable STS transfer'. Honest caveat: vessels can also loiter legitimately, anchor for weather, or conduct authorised bunkering. Context and pattern-of-life history are essential.
The revisit problem and how to work around it
Sentinel-1's 6-day revisit per satellite is the central operational constraint. A transfer lasting 12 hours in a 6-day window has a roughly 8% chance of being captured by a single Sentinel-1 pass. That probability is not good enough for systematic enforcement. The practical response is to concentrate search effort on known anchorage zones where STS activity is historically concentrated, and to use AIS anomaly alerts to cue tasked SAR collection.
When an S-AIS feed flags a vessel going dark in a known zone, a tasked ICEYE or Capella collection can be ordered within hours. The combination of a free, wide-area Sentinel-1 archive for pattern-of-life analysis and paid, on-demand X-band tasking for event confirmation is the operationally sensible architecture. Tasking costs money; Sentinel-1 does not. The workflow is: screen with open data, task on suspicion, confirm with high resolution.
Archive depth is also useful. Sentinel-1 data is available from 2014 onwards through the Copernicus Data Space Ecosystem. Retrospective analysis of a vessel's historical presence in known STS zones can establish a pattern that supports legal or diplomatic action, even when real-time collection was not in place.
What the imagery can and cannot prove
A 0.5 m SAR image showing two tankers in contact with visible fender geometry, combined with an AIS gap for both vessels at that location and time, is compelling evidence of an STS event. It does not, by itself, identify the cargo as sanctioned crude, confirm the flag state of either vessel with certainty, or establish the identity of the beneficial owner. Those questions require vessel identity cross-referencing against sanctions lists, IMO number databases, and ownership registries, work that sits in the intelligence-fusion layer above the imagery.
Optical confirmation from Pléiades Neo at 30 cm can read hull markings and sometimes flag state in clear conditions, which significantly tightens vessel identification. But operators conducting illicit transfers choose overcast nights for a reason. Cloud cover over the Gulf of Thailand or the Strait of Malacca can persist for days during the monsoon. Expecting optical confirmation on demand is unrealistic; SAR must carry the primary detection burden.
Satellize's analytics pipeline fuses open Sentinel-1 wide-area search with commercial tasking on client licence, applying the same general detection-and-fusion approach used in its Tonga crop-estimation work to maritime enforcement problems. The output is a structured alert package, not raw imagery, designed to support a compliance or legal team rather than a remote-sensing specialist.
Putting the evidence package together
An actionable STS detection report combines several layers: the SAR detection image with vessel outlines annotated, the S-AIS gap record with timestamps and coordinates, retrospective archive imagery showing the vessels' prior presence in the zone, and an identity cross-reference against published sanctions lists and IMO records. Each element addresses a different counter-argument that a sanctioned operator's legal team might raise.
Enforcement agencies and compliance teams at financial institutions financing tanker fleets are the primary consumers of this analysis. The UN Panel of Experts process has demonstrated that satellite evidence, properly documented, is admissible in UN reporting. The bar for internal compliance decisions is lower still. The question is not whether SAR can detect STS transfers; the published literature and operational programmes confirm it can. The question is whether the detection workflow is systematic and fast enough to matter.
Typical figures
| Wide-area search resolution (Sentinel-1 IW) | 10 m ground range, 250 km swath |
| Confirmation resolution (ICEYE / Capella Spotlight) | Approximately 0.5 m, swath 5–10 km |
| Optical confirmation resolution (Pléiades Neo) | 30 cm panchromatic; cloud-dependent |
| Sentinel-1 revisit (open ocean) | 6 days per satellite; 3 days with two satellites combined |
| Tasked SAR latency (ICEYE / Capella priority) | Typically under 12 hours from order to delivery |
| S-AIS coverage latency (Spire / Aireon) | Minutes to low tens of minutes for global ocean coverage |
| Minimum detectable vessel length (Sentinel-1 IW) | Approximately 30–50 m under good sea-state conditions; smaller vessels may be missed in high sea clutter |
| Sentinel-1 archive depth | From 2014 (Sentinel-1A launch); accessible via Copernicus Data Space Ecosystem |
| SAR frequency / wavelength | C-band: 5.4 GHz / ~5.6 cm (Sentinel-1); X-band: ~9.6 GHz / ~3.1 cm (ICEYE, Capella) |
| Deliverable formats | GeoTIFF annotated imagery, GeoJSON vessel detections, PDF alert report, structured CSV event log |
Analytics Satellize can run
| Wide-area vessel proximity detection | Constant false alarm rate (CFAR) ship detection on Sentinel-1 IW GRD scenes, followed by geometric clustering to flag vessel pairs within 50 m of each other | GeoJSON layer of candidate STS proximity events, updated on each Sentinel-1 overpass |
| AIS gap correlation alert | Temporal and spatial join of S-AIS track data against SAR detection positions; flags vessels whose transponder record shows a gap of more than 2 hours coinciding with a SAR proximity detection | Structured alert feed with vessel MMSI, gap duration, coordinates and SAR image thumbnail |
| High-resolution confirmation package | On-demand ICEYE or Capella Spotlight tasking triggered by AIS gap alert; analyst annotation of fender geometry, mooring configuration and hull separation distance | Annotated GeoTIFF with measurement overlays and written analyst assessment, delivered as PDF report |
| Vessel identity cross-reference | IMO number extraction from optical imagery where legible; cross-reference against OFAC, EU and UN consolidated sanctions lists and open ownership registries | Sanctions-list match table appended to confirmation report, with confidence rating and source citations |
| Pattern-of-life zone analysis | Retrospective CFAR detection across Sentinel-1 archive for nominated anchorage zones; heat-map of historical STS activity frequency by location and season | GIS layer (GeoTIFF / GeoPackage) showing detection density, with accompanying time-series chart |
| Vessel track reconstruction | Interpolation of S-AIS track segments before and after AIS gap, combined with SAR-derived position fix during gap, to reconstruct probable vessel route | KML / GeoJSON track file suitable for inclusion in compliance or legal documentation |
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.