Building a Verified Aging Profile Database for Authentication

Data Is Cumulative Power

The more authenticated textiles you have measured and documented, the better your ability to evaluate the next unknown textile. Each data point — whether from a confirmed genuine textile, a confirmed forgery, or an inconclusive case — contributes to your understanding of how aging actually manifests in real objects.

What to Record for Each Textile

Object data:

• Type, fiber, dye identification, mordant identification
• Date of manufacture (confirmed or estimated)
• Geographic origin (confirmed or estimated)
• Provenance history (as complete as available)
• Authentication conclusion and confidence level

Measurement data:

• Spectral reflectance curves at multiple points
• Lab* values at measured points
• Photographs under controlled conditions (normal, UV, raking light)
• Any analytical results (FORS spectra, XRF data, HPLC results)

Degradation data:

• Model parameters used for prediction
• Predicted vs. actual comparison (ΔE, spectral comparison)
• Degradation factors identified (UV, humidity, oxidation, pollutant, biological)
• Notes on any unusual features

Organizing the Database

By dye type: All indigo textiles together, all madder textiles together, etc. This enables comparison of the same dye at different ages and under different conditions.

By era: All 18th-century textiles, all 19th-century, etc. This reveals era-specific patterns.

By outcome: Genuine, forgery, inconclusive. This enables comparison of genuine vs. fake characteristics.

Cross-referenced: Every entry searchable by dye, era, fiber, origin, condition, and authentication outcome.

Using the Database

For a new authentication case:

1. Identify the textile's dye, era, and claimed origin
2. Search the database for comparable genuine examples
3. Compare the suspect textile's degradation profile to the genuine examples
4. Search for comparable forgeries to check whether the suspect matches known fake patterns
5. Use the statistical range of genuine examples to assess whether the suspect falls within normal variation or outside it

For model calibration:

• Compare model predictions to actual measurements on dated, genuine textiles
• Identify systematic biases (does the model consistently over- or under-predict certain degradation types?)
• Adjust model parameters based on the comparison

Growing the Database

Active collection: Measure every textile you examine, regardless of whether it is a formal authentication case. Museum visits, auction previews, conservation lab access — every measurement adds data.

Collaborative sharing: Partner with other authentication specialists, conservation labs, and museum collections to share data. Larger databases are more powerful.

Published data: Incorporate published spectral data and degradation studies from conservation science literature.

The Competitive Advantage

An authentication specialist with a large, well-documented database has a significant competitive advantage:

• More confident opinions (backed by more comparison data)
• Faster assessments (the relevant comparison is already in the database)
• More defensible reports (statistical ranges from the database support the conclusion)
• Better pattern recognition (trends visible across a large dataset are invisible in small ones)

Ready to build your database with degradation model-calibrated measurements? Join the PigmentBoard waitlist.