Open maritime evidence · AI · state capacity
Evidence before
maritime risk.
MarineAware is an independent research demonstrator created by Dipankar Sarkar. It examines how AIS, satellite, weather, and open reporting can support transparent maritime analysis—and when incomplete data requires an explicit “insufficient evidence.”
Open research demonstrator · live.marineaware.com
Inspect what the public build can and cannot establish
The demonstrator assembles available open reporting, AIS samples, metocean, and broad imagery across maritime areas. Each output must be read with its source coverage, freshness, and missing-data limits. It is research, not an operational risk picture.
Research questions across the maritime value chain
Research areas
Six classes of maritime evidence problem
These pages describe potential methods and decision contexts. They are a research agenda, not a claim that every capability is implemented or validated in the public demonstrator.
Maritime Domain Evidence
What can open sources establish beyond cooperative vessel reporting?
Public AIS is partial and cooperative. Satellite imagery, registries, weather, and reporting can add context, but they have different coverage, timing…
Researchers, public institutions, analysts, and maritime operatorsIdentity & Sanctions Evidence
Which vessel-risk indicators remain defensible when identities and behaviour change?
A list match is only one form of evidence. AIS gaps, identity changes, ownership records, transfers, and imagery can be relevant, but none proves misc…
Compliance researchers, insurers, public institutions, and maritime analystsEmissions & Regulatory Evidence
How should activity estimates, reported emissions, and policy rules be reconciled?
AIS-derived emissions are estimates, verified reporting has its own scope, and regulatory regimes use different boundaries and definitions. Combining …
Researchers, ship operators, policy teams, and sustainability analystsTrade-Flow Inference
Which conclusions survive gaps, latency, cargo uncertainty, and aggregation?
Port calls and vessel movement can illuminate trade, but cargo, utilisation, destination, ownership, and timing are not always observed. A useful flow…
Economic researchers, policy analysts, and maritime market observersVoyage & Port Modelling
When can open observations improve an arrival estimate?
An ETA model can look accurate on an easy sample and fail under congestion, weather, route change, anchorage, or incomplete history. The relevant ques…
Researchers, ports, terminals, carriers, and cargo ownersMaritime AI Claims
How should a maritime-AI capability claim be tested?
A polished maritime-AI demonstration can hide purchased data, narrow test conditions, leakage, weak ground truth, or dependence on one provider. Techn…
Researchers, boards, investors, public institutions, and technical reviewersEvidence architecture
The difference between assembling signals
and supporting a conclusion
More feeds do not automatically create better evidence. A serious method records which source was present, what it observed, how fresh it was, and what alternative explanations remain. AI can help organise that work, but it cannot repair missing observation by prose.
Sources keep their roles
AIS, imagery, weather, reporting, and registries answer different questions. Combining them must not erase their distinct coverage and failure modes.
AI stays in its lane
Models may orchestrate tools or narrate already-computed evidence. They do not become sensors or independent corroboration.
Missing is visible
A source that failed or returned no item remains missing—not zero, calm, or evidence that nothing occurred.
Conclusions are bounded
A defensible output exposes provenance, time, assumptions, and uncertainty and can abstain when the question cannot be answered.
Research cases
Preserved evidence and bounded conclusions
Insights
Field notes on the rules moving the market
From open source to operational: scaling a live OSINT maritime picture with proprietary data and fine-tuned models
We built a live, open-data maritime domain awareness dashboard — fusing free AIS, dated OSINT reporting, metocean and satellite across 12 global chokepoints. Here is what open sources can and cannot show, and how the same fusion engine scales with licensed AIS and SAR, multispectral analysis, domain-fine-tuned models, and the wave of satellites launching through 2026–2027.
Jul 11, 2026 MethodsFine-tuning vision models for multispectral maritime analysis
Generalist vision models are weak sensors on satellite imagery. Detecting and classifying vessels across SAR, optical and multispectral bands needs models fine-tuned on maritime data. Here is why domain fine-tuning matters, what multispectral and hyperspectral sensing adds, and how it compounds with the satellites launching in 2026–2027.
Jul 10, 2026 Agentic AIHow AI agents are changing the way we analyze AIS data
AIS analysis is moving from hand-written geofence queries and static dashboards to agentic pipelines that plan a multi-step investigation, call the right tools, and explain the answer. Here is what changes and what stays hard.
Jun 30, 2026Questions
Straight answers
What is MarineAware? +
MarineAware is an independent public research demonstrator created by Dipankar Sarkar. It examines how AIS, imagery, weather, reporting, and AI can support transparent maritime analysis and how an accountable system should behave when evidence is incomplete.
Does no captured AIS position mean no vessel was present? +
No. An empty sample can reflect receiver coverage, a collection window, connection failure, rate limits, legitimate switch-off, deliberate non-broadcast, or true absence. Intent or absence requires appropriate corroboration.
Is the public demonstrator an operational surveillance service? +
No. It does not claim government deployment, classified access, persistent global coverage, validated client outcomes, or suitability for navigation, security, sanctions, legal, insurance, investment, or trading decisions.
What collaboration is appropriate? +
MarineAware welcomes methods critique, seminars, reproducible open-data cases, and bounded research collaboration on maritime evidence, accountable AI, and institutional decision-making.
Pressure-test a maritime evidence question.
MarineAware welcomes research critique, methods discussions, seminar invitations, and bounded collaboration proposals.