Research and methods
What MEPA is researching, and what it has not yet shown.
This page is the index of MEPA's research: the work, the methods, the publications and the limits. Each item says whether it is proposed, planned, in progress or published, so that a plan is never read as a result.
Where the research stands
Proposed
Levantine Lanes 90
A 90-day research pilot on the shipping lanes south of Cyprus, to test how public satellite observations and ship-movement data can support finding, assessing and documenting possible marine pollution. Methodology development is under way. No pilot results have been published.
Mixed
The monitoring system
Some work is done by hand today (analyst review of open satellite data and of reports from the public). Radar screening is a pilot scope. The look-alike filter, dark-vessel matching and the sealed evidence package are planned. Each part is marked on the Technology page.
Planned
MEPA Orbital
A concept for a satellite of MEPA's own. Planned, not funded, and not open for donations.
Published
Explainers, illustrative cases and one concept note
5 pieces. None reports measured results, and each says what it is on the page.
How to read what we publish
- Verified finding
- A result published with its data, method and review. MEPA has published none yet.
- Preliminary observation
- An early, unreviewed result. None is published.
- Hypothesis
- An idea to be tested. The Starlink V3 concept note is one.
- Plan or objective
- Work that has not been done. Every page marks it as planned.
- Illustration
- A written or AI-generated picture of how the system is meant to work. It is not a record of a real event.
Methods
The method is a chain: screen radar and optical imagery for possible slicks, filter out look-alikes, match a detection to the ships that could have been there, and package the evidence for the authority that can act. A person reviews every step that matters. The Technology page explains each link and marks it as done by hand, a pilot or planned.
- What the system looks at: six signals, and the status of each
- Signal from noise: separating oil from look-alikes
- Even with AIS off: matching a ship that is not transmitting
- From detection to authority: the reporting pipeline
- Open tools MEPA builds on
Publications
Concept note ·
Concept note: if we had the data from a network like Starlink V3
A theoretical look at what Starlink V3 is and is not, and what MEPA could do with data from such a network if it were ever shared.
Not: A finding, an agreement or a product. MEPA has no data from, and no partnership with, any satellite operator.
Illustrative case ·
Illustrative case: Response and Recovery
In this illustrative case, MEPA keeps watching after containment begins — tracking the response, shoreline and seagrass from orbit.
Not: A real incident or a real recovery. No outcome described here has been measured.
Explainer ·
Ghost Discharges
When a ship switches off its transponder, the slick it leaves could still be traced — to the vessel, and to the companies behind it.
Not: A finding about any ship, company or authority. An AIS gap is a lead, not proof of a discharge, and MEPA names no one on a single signal.
Illustrative case ·
Illustrative case: The Mediterranean Spill
From the first satellite flag to a referral authorities could act on: how an imagined Eastern Mediterranean spill would be detected, traced and documented.
Not: A real incident. No real vessel, company, authority or place is meant, and none of it is evidence of anything.
Explainer ·
Silent Spillways
Why small, repeated discharges go unseen — and how radar satellites, machine learning and analyst-reviewed alerts are designed to bring them to light.
Not: A report of detections, cases or measurements. No pilot results have been published, and none are claimed here.
To cite a piece: MEPA, title, date, and its address on www.mepa.ngo. Pieces carry a revision date when they change. Subscribe by feed.
Primary sources
MEPA works from open data and standard references rather than its own measurements. These are the primary sources behind the explainers. Listing a source does not mean its publisher endorses MEPA.
Limitations
- No pilot results are published. Nothing on this site reports how many slicks MEPA has found or how accurate its screening is.
- A radar detection is a lead, not proof. MEPA does not name a person or company on a single detection or a tip.
- Satellites revisit an area every few days, so some events will not be seen, and radar can mistake natural films and calm water for oil.
- Several images on the site are AI-generated illustrations of the system MEPA is building. They are not records of real events and are not evidence.
- MEPA is at an early stage, with a small team. It has not been independently audited.
Who is responsible
The research is directed by Santiago Puig Asensio, Director and Principal Investigator. How MEPA is governed, who controls the money and its disclosure about PolarVector, its investigations partner, are on For funders. MEPA's legal details are in the terms of use.
Found an error, or know better? Tell us. We would rather be corrected than be wrong.