Principal Component Analysis in monument conservation: Three application examples

Abstract

Abstract Multivariate statistics is a well-known and invaluable tool in archaeological science but its use is limited in monument restoration. The aim of this work is to demonstrate the effectiveness of Principal Component Analysis (PCA) on the characterization, technology and weathering condition investigation of building materials from historical monuments. Towards this aim, three examples are given: The first one is a provenance and technology investigation of the Aghia Sophia (Istanbul, Turkey) bricks, some of which had to be replaced due to weathering, during recent restoration works. It was proved by PCA that the original clay, used for the construction of the bricks, is not similar to the clay of other contemporary constructions in Istanbul but presents high similarity to the raw material of the bricks from a contemporary church in the island of Rhodes (Dodecanese, Greece). Additionally, the technology of the bricks was studied by mercury intrusion porosimetry, strength tests and Scanning Electron Microscopy. The use of PCA gives a very comprehensive way to present the difference in the technology of the dome bricks. The second presents a classification of mortars from medieval (Byzantine) monasteries, based on their microstructural characteristics (porosity, reverse hydraulicity ratio) and strength measurements. The PCA grouping gives an illustrative diagram depicting the correlation between mortar syntheses and resulting characteristics. The third case shows an example of the correlation between environmental pollution data and data from the weathering layers of marble surfaces (patina composition, orientation of the monument surface, etc.).