Paper 1 · March 2027
What the current ocean carbon observing system costs
The first network-by-network cost of GO-SHIP, the ship lines, moorings, floats and data synthesis, next to the uncertainty it buys.
Starts nowWith the programme partners
Paper 2
What makes the system MRV-grade, and the technology advances to prioritise
The accuracy and density verification requires, and the advances in sensors, platforms, autonomy and models to get there. Advances are scoped for the shared system; some will also serve individual projects, and we say so.
Delivers detection thresholdsDelivers a technology roadmap
Paper 3
What an upgraded shared MRV system costs and saves, versus scale
What it saves per verified tonne compared with projects building their own MRV, and the scale at which it pays for itself.
Delivers shared-system costingDelivers per-project MRV costs
The deliverables are results in their own right, with their own users: the thresholds and roadmap for whoever designs the next observing system; the two costings for policy analysts and the registries. They arrive before the papers. Each paper is written only if its numbers come back; without new numbers it would be one more position paper.
Under all three papers
Economics and policy
Economics
Verifying a tonne must cost less than the tonne is worth. That sets the bar for the technology in paper 2 (what we prioritise has to be affordable for the value and revenue a verified tonne brings) and the case for sharing in paper 3 (the savings curve). The carbon price enters only as the yardstick for value, never as the object of estimation.
Policy
Rules decide what every project must monitor and what counts as verified, and the system has to be designed to those rules. Two rule sets, kept distinct: the London Convention and Protocol assessment frameworks for marine geoengineering research, adopted 9 October 2026, say what an experiment must monitor and report; the registries decide what counts as a verified tonne. And a shared system exists only if nations, programmes and buyers agree on who pays for it, the Argo model.
How papers 2 and 3 fit together
Who should measure the background ocean?
Every carbon-removal project must detect its own signal against the background ocean. The project can measure its own signal with the sensors and autonomy paper 2 describes. The background is the same for every project in a region and no project can measure it alone. It is the public good, and it is what the shared system measures.
Paper 2 is about quality (accuracy and density), which has no scale attached. Paper 3 holds the technology fixed and varies who deploys it: every project measuring the background for itself, N times over, or one shared system measuring it once. The saving is the duplication avoided, and it grows with the number of projects.
Scale means two things, and paper 3 needs both. The scale of demand (number of projects, tonnes per year) drives the per-project total roughly linearly. The scale of supply (how dense the shared system must be) is not a free choice; it comes from paper 2's requirement and is costed with paper 1's unit costs. So paper 3's result is a curve, not a number.
A · Total MRV cost per year
B · Cost per verified tonne
Each project measures the background itself
One shared system measures it once
Illustrative shapes, not data. These curves are what paper 3 would estimate. Inputs: unit costs (paper 1); the accuracy and density required (paper 2); what projects pay for MRV today (published cost models and the registries).