Every decision about the Posidonia meadow off Pollica rests on a survey taken once, years ago, by somebody who has not been back. This is the method that lets the people who are already on the water check it, every season, and keep the record.
Island Lab · community technology partner · v0.2
Arrow keys to move. Press C for contents.
The known Posidonia extent for this coast is the EMODnet layer sourced from ISPRA, surveyed 29 November 2017. It is the base for anchoring decisions, for protection lines, for anything a comune has to defend. Nine years on, it is still the only picture.
Professional re-survey is expensive, which is why the map is old rather than an accident of it. Nobody is neglecting the meadow. The cost of looking simply exceeds what a small municipality can spend on looking.
Meanwhile the boats go out every day, over the same water, and see it.
One polygon, one date: present or absent. It cannot say whether this year is worse than last, because there is no last.
Every hectare figure, every protection argument and every revenue ceiling multiplies through that one survey.
A temperature, light and position package small enough to mount on a working fishing boat costs less than a single professional survey day. The hardware stopped being the obstacle some years ago.
Exists, and is affordable. Sensors on a small vessel, a tablet, a route.
Has to be sound enough for science and simple enough for a working morning. That is a design problem, not a hardware one.
Who holds the raw data, who signs it, who archives it, who can produce it in five years. Usually unanswered.
Everything after this slide is evidence for that sentence: the protocol that makes it repeatable, the surface that makes it checkable, and the split that makes the independence real rather than stated.
The measurement is produced by the people whose coast it is, signed by a named reviewer, and archived by the municipality. Island Lab is not in the custody chain, on purpose.
One survey produces a number. A protocol produces a series, and a series is the only thing that can answer the question a coast actually has: is it getting better or worse, and did what we did about it work.
Four things have to be present at once, and the fourth is the one most monitoring programmes skip.
A survey is bought. A protocol is learned, and then it belongs to the people who learned it. That is the difference between a coast that has been measured and a coast that can measure itself.
The coast is drawn as a grid of cells. Every cell carries three separate facts, and keeping them separate is the entire point of the screen.
A boat ran this cell this season, with a timestamp, a vessel, a route and an operator behind the reading.
What the 2017 extent says is here. The prior picture, drawn as a prior rather than as the truth.
Where the two do not match. This is the output the whole product exists to produce.
The obvious way to build this screen is to test each reading against the 2017 polygon and drop whatever falls outside it. That treats the old map as ground truth, and it deletes exactly the evidence the survey went out to find.
So a reading beyond the mapped extent is rendered as disagreement between two sources, dated and attributed, and it stays. The meadow is allowed to be somewhere the map does not have it.
Absence of evidence and agreement are different states, and if the map ever draws them the same way the product is lying to the person reading it. Blank means nobody went there. It never means fine.
| Who | Holds | What that means in practice |
|---|---|---|
| Community and fleet | runs the survey | Boats run the lines and hold the raw data. The measurement is produced by the people whose coast it is, not delivered to them. |
| The comune | archives and decides | Takes custody of the published record as the official one, and routes what the comparison finds into a decision it can defend. |
| Island Lab | method and quality control | The method, the tooling and the quality control. Not the data, not the archive, not the decision. Nothing else. |
| Sign-off | open question | Nobody formally holds it yet. The director role carries authority in the schema and does not exist in the data model. |
The premise is that an untrained operator produces defensible data because a named reviewer checks and signs the run. Whether that reviewer is a municipal employee, an Island Lab role or a university secondment changes who is liable when a run is wrong, which makes it a governance decision before it is a technical one. It is shown open because it is open.
The South Cilento Masterplan spans thirteen coastal municipalities along the same stretch of water, with the same habitat, the same fleets and the same nine-year-old basemap problem.
The region configuration already travels. A second coast is a config swap: its own geometry, its own fleet, its own record, running the same seven steps. That is engineering that exists rather than a promise about a second phase.
Replicating a model means exporting a programme that belongs somewhere else. Carrying a method means each coast ends up with a capability of its own.
One coastline, thirteen administrations, one shared method and thirteen separate records.
No Posidonia carbon credit has issued anywhere, ever. Not a slow pipeline, an empty one.
Every blue-carbon price series in circulation is mangrove data wearing a blue-carbon label. Applying it to a Mediterranean seagrass meadow is an assumption, not a comparable.
Published restoration cost estimates span two orders of magnitude. We have not reconciled them, so we print the disagreement instead of averaging it away. unreconciled
Protection first, with restoration costed separately. Protecting a meadow that is already storing carbon is cheap, defensible and available now. Restoration is a real option with real costs, and it is priced on its own line rather than folded into a credit story that has not yet happened to anyone.
Where a number is not sourced yet, this work marks it pending rather than filling it with an estimate. A placeholder that reaches a careful reader costs more than a gap does.
Designed, written down, and running end to end. The seven steps are implemented, not illustrated.
What is on screen today is synthetic data driving a real pipeline. Every surface says so at the point of display rather than in a footnote.
The instruments have not reported since 8 July. The first real outing is the next stage of the work, and until it happens no reading here came off the water.
Designed and costed, and not built. It is the difference between confirming the meadow's condition and confirming its extent, and there is no sonar ingest anywhere in the system.
Island Lab is a Washington State LLC, founded by Sean Kolk and Kate Schnippering. We supply the method, the tooling and the quality control.
The community owns the data. We own the method. That split is deliberate, it is written into the accountability matrix, and it is what makes the independence checkable instead of asserted.
If Island Lab disappeared, the comune would still hold the record, still be able to produce it, and still have people trained to run the season. That is the test the arrangement is designed to pass.
Community: the instrument, the raw data, the archive, the decision.
Island Lab: the method, the tooling, the quality control.
Island Lab · sean@islandlab.dev · v0.2