A novel approach to documenting water diffusion in ancient obsidian artifacts via the complexity analysis of microscope images

Abstract

The diffusion of water through the surface of archaeological obsidian is mainly a temperature-dependent and concentration driven phenomenon that forms the basis of the hydration dating (OHD) method. For the first time, we apply a novel fractal approach, which explores the various attributes of images linked to the variation in water concentration and structural deficiencies within the hydrated layer. The microscopic images of hydrated layers on two obsidian specimens from Africa spanning thousands of years have been investigated. Seven complexity measures of fractal dimension (succolarity, lacunarity, correlation dimension, local connected fractal dimension (LCFD), fractal fragmented index (FFI), entropy and normalized Kolmogorov complexity (TIFF-LZW)), were applied to the hydration layer images. It was found that the effective determination of diffusion depth mainly comes from the application of succolarity, FFI and LCFD fractal indices, plus the structural deficiencies at the surface and the misconnected lacunas at the interface of the water diffusion layer and unaltered obsidian. The difference between fractal dimensions of the hydrated area and the non-hydrated images turned out to be a crucial quantitative parameter. Accordingly, we found that especially the succolarity, which is based on the capacity of a water fluid to percolate through the obsidian area in a given direction, was critical to determine the characteristics of the obsidian hydration layer. The FFI and LCFD estimate the compaction/fragmentation and connectivity, respectively. Therefore, the fractals play an important role in understanding the general diffusion in solids theory and the study aims at providing a new direction in the OHD and diffusion fields concerning the complex dynamic systems and structures which display a fractal pattern.