UNDERSTANDING THE CAROLINGIAN ARTIST THROUGH A NOVEL WORKFLOW INVOLVING MULTIMODAL REMOTE AND IN-SITU SPECTRAL IMAGING AND SPECTROSCOPIC TECHNIQUES

UNDERSTANDING THE CAROLINGIAN ARTIST THROUGH A NOVEL WORKFLOW INVOLVING MULTIMODAL REMOTE AND IN-SITU SPECTRAL IMAGING AND SPECTROSCOPIC TECHNIQUES Liggins, Florence (Nottingham Trent University) - Detalle, Vincent (Centre National de la Recherche Scientifique (CNRS), Paris, France) - Romano, Francesco (CNR, Istituto di Scienze del Patrimonio Culturale (ISPC-CNR), Catania, Italy) - Cassetti, Patrick (Foundation pro Monastery of St. John, Müstair) - Liang, Haida (Nottingham Trent University) F. Liggins, S. Kogou, C.S. Cheung, A. Vichi, A. Suzuki, V. Detalle, X Bai, M. Botticelli, C. Caliri, E. Ravan, F.P. Romano, C. Tennenini, P. Cassitti, H. Liang The analysis of historical wall paintings can invite several obstacles, such as scale, complexity, diversity of materials, in substrate, pigment and binder. Locating restorative work can further add complication, especially where the work has been applied in a heterogenous manner. The Benedictine convent of St. John, a UNESCO world heritage site in Müstair, is host to both original Carolingian paintings (8th – 9th Century), and 20th Century restorative work and is therefore a prime example of where multiple methods are needed for large-scale analysis. We demonstrate a novel workflow for thorough analysis of large-scale wall paintings which starts with a survey using a ground-based high spatial resolution remote spectral imaging system (PRISMS), which can scan entire building interiors at distances of order ~10m, providing 10-band reflectance spectra in the visible to near infrared (VNIR) (450- 850nm)[1]. Hyperspectral imaging is also carried out with higher spectral resolution (~nm) covering the VNIR and shortwave infrared (SWIR) (1000-2500nm)[2]. These datasets are processed using in-house developed machine learning methods to provide material distribution maps based on reflectance spectra[2,3] such that further analysis only needs to be performed in targeted areas. Point-based analyses provide necessary complementary information in material identification, in this case ground-based remote Raman spectroscopy[4], close-range Laser-induced Breakdown Spectroscopy (LIBS), X-ray Diffraction (XRD), X-ray Fluorescence (XRF) and Fourier Transform Infrared (FTIR) spectroscopy. Whilst demonstrating this workflow on the murals in Müstair through an IPERION HS MOLAB campaign[5], we’re able to identify pigments from the original Carolingian artist’s palette and define to what extent these materials are still distributed across the wall paintings. [1] https://doi.org/10.1016/j.isprsjprs.2014.05.011 [2]https://doi.org/10.1038/s41598-020-76457-9 [3]http://dx.doi.org/10.1186/s40494-022-00765-8 [4]https://doi.org/10.1364/OE.27.031338 [5]Molab Iperion HS: https://www.iperionhs.eu/molab/.