MarineAware

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
public sources and inspectable assumptions
Bounded
claims tied to the sources actually retrieved
Correctable
versioned methods, limitations, and challenges

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.

Build-specific sourcesMissing-data statesMetocean contextOpen imageryInspectable scoring
Open the demonstrator ↗

Research questions across the maritime value chain

Carriers & operators · Marine insurers & P&I clubs · Commodity traders · Ports & terminals · Government & defense · Technical reviewers

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 operators

Identity & 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 analysts

Emissions & 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 analysts

Trade-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 observers

Voyage & 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 owners

Maritime 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 reviewers

Evidence 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

All case studies

Questions

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.