Non-invasive pigment identification at the Yungang Grottoes (Caves 7–8): a chemically corroborated spectral library
This study establishes a chemically corroborated, three-tier reflectance spectral library for the Yungang Grottoes (Caves 7–8) that significantly improves non-invasive pigment identification accuracy through a weighted matching algorithm, enabling a comprehensive pigment inventory and scalable hyperspectral mapping for future conservation efforts.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
The colors on ancient stone carvings are more than just decoration; they are a fragile record of history, telling us how artisans thought, what materials they could reach, and how a site has changed over centuries. However, these pigments are often hidden beneath layers of dust, weathering, and later repairs, making them difficult to read without causing damage. For decades, experts have relied on taking tiny physical samples to identify these colors, a method that is precise but destructive and limited to a few tiny spots. A newer approach uses light to "see" the chemical makeup of a surface from a distance. By bouncing a broad spectrum of light off a wall and measuring how it bounces back, scientists can create a unique optical fingerprint for each material. The challenge has been that these fingerprints are not universal; a red pigment on a cave wall in one part of the world might look different than a red pigment in another due to age, dirt, or the specific minerals used. Without a reliable library of these local fingerprints, the light-based method often produces guesses rather than answers.
At the Yungang Grottoes in China, a massive complex of fifth-century stone sculptures, researchers faced exactly this problem. The caves are covered in polychrome paintings that have suffered from centuries of weathering and repeated repainting, blurring the original palette. To solve this, a team of scientists from the Yungang Academy and several universities constructed a specialized, three-part library of light signatures specifically for these caves. They combined standard commercial pigments, references from public scientific databases, and a new, custom archive built directly from the cave walls themselves. This custom archive was created by taking a few tiny, verified samples from the caves, analyzing them with powerful chemical tools, and then recording the exact light signature of those same spots. This process allowed them to build a reference system that accounts for the specific aging and conditions of the Yungang site, rather than relying on generic, laboratory-perfect samples that might not match the real world.
The researchers then applied this library to over a thousand specific points across Caves 7 and 8. Instead of forcing every measurement to match a single color name, they developed a system that assigns a level of confidence to each identification. If the light signature matched a reference perfectly and was clearly distinct from other options, the system gave a high-confidence answer. If the match was ambiguous or the material was a complex mixture, the system flagged it for further review rather than guessing. This careful approach revealed that while many areas were covered in expected ancient pigments like red ochre and lapis lazuli, the caves also held evidence of later interventions. Chemical analysis confirmed that some of the green paint contained emerald green, a synthetic pigment invented in the 19th century, proving that parts of the caves were repainted long after the original construction. Similarly, the presence of a blue pigment known as Prussian blue, which was not available until the 18th century, was flagged in several spots, indicating later touch-ups that would have been missed by a simple visual inspection.
The study successfully mapped the distribution of these pigments, creating a unified inventory that distinguishes between original materials and later additions. They found that the red areas were dominated by iron-based earths and cinnabar, while the white sections were a mix of mica and gypsum, a mineral that had previously been difficult to identify without physical sampling. The team also demonstrated that this point-by-point method could be expanded to cover entire walls using a camera that captures light signatures across a wide area. This allowed them to generate a color-coded map of a ceiling section, revealing patterns of original painting and later repair that were invisible to the naked eye. By combining a tailored library with a strict system of confidence levels and chemical verification, the researchers established a reliable method for documenting the history of these fragile artworks. This approach not only provides a clear picture of what the caves looked like in different eras but also offers a sustainable way to monitor their condition over time without ever touching the stone.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.