Satellite-data product validation
Make the quality and limitations of a satellite-derived product visible to its consumers.

The situation.
A processing pipeline produces a useful-looking map, but consumers cannot tell where it is reliable. Validation connects the intended use to reference evidence, quality flags and a repeatable assessment of errors.
A practical approach.
- 01
Set expectations
Define product meaning, permitted uses, required coverage and acceptable uncertainty.
- 02
Design validation
Choose reference data and sampling that reflect the conditions and locations of intended use.
- 03
Publish limitations
Attach quality evidence, provenance and known failure conditions to each product version.
What to validate
- Avoid validating against information already used to tune the method.
- Document reference-data limitations and differences in observation timing.
Typical outputs
- Product quality contract
- Validation dataset and assessment plan
- Versioned quality report
Example measures
- Quality by condition or region
- Coverage with validated reference evidence
- Products with complete provenance
Agree a baseline and acceptance criteria for the specific project.
Illustrative scenario and conceptual imagery. Scope, deliverables and acceptance evidence are agreed for each engagement; example measures are not achieved results.
