01 / DATA PROVENANCE
Trace every layer
Provider, dataset, licence, spatial resolution, time coverage and retrieval time are retained separately. Provider catalogue dates describe availability, not a guarantee of coverage at a point.
SCIENCE & METHODOLOGY / V1.0
Trust is built into the method.
A useful ocean platform must show what is known, how it is known and where the evidence ends.
Measure first.
Act second.
Verify everything.
Act second.
Verify everything.
02 / REPRESENTATION
Imagery is not a point measurement
NASA GIBS serves visualisations. Happy Orbis does not reverse-engineer colour into precise numerical values. Missing or transparent pixels must not be interpreted as zero.
03 / TIME & COMPARISON
Compare like with like
Compare the same product and season where possible. Date comparison shows two imagery snapshots, not a statistically tested trend. MUR25 anomalies use the provider’s 2003–2014 climatology.
04 / UNCERTAINTY
Keep the categories separate
Observed profiles, satellite retrievals, interpolated analyses and forecasts have different uncertainty. A visual layer is not a complete ocean-health assessment.
05 / INTELLIGENCE
Source-based guidance
The current evidence guide uses curated source explanations and map actions. Generative Astra inference and scientific computation are not connected. The guide does not claim to diagnose local causes.
06 / VERIFICATION
No result without evidence
An intervention must retain its baseline, location, time, methodology, supporting records and reviewer. Estimated, measured and independently verified outcomes remain separate.
No composite ocean-health score, carbon-removal claim or verified intervention metric is published in this version. Version 1.0 covers imagery exploration, educational depth modelling and the Blue Route sensitivity calculator.