Discrimination of ivory from extant and extinct elephant species using Raman spectroscopy : A potential non-destructive technique for combating illegal wildlife trade

Shepherd, Rebecca F. and Lister, Adrian M. and Roberts, Alice M. and Taylor, Adam M. and Kerns, Jemma Gillian (2024) Discrimination of ivory from extant and extinct elephant species using Raman spectroscopy : A potential non-destructive technique for combating illegal wildlife trade. PLoS One, 19 (4): e0299689. ISSN 1932-6203

Text (IvoryID_PONE_AAM_2024)
IvoryID_PONE_AAM_2024.pdf - Accepted Version
Download (1MB)

Abstract

The use of elephant ivory as a commodity is a factor in declining elephant populations. Despite recent worldwide elephant ivory trade bans, mammoth ivory trade remains unregulated. This complicates law enforcement efforts, as distinguishing between ivory from extant and extinct species requires costly, destructive and time consuming methods. Elephant and mammoth ivory mainly consists of dentine, a mineralized connective tissue that contains an organic collagenous component and an inorganic component of calcium phosphate minerals, similar in structure to hydroxyapatite crystals. Raman spectroscopy is a non-invasive laser-based technique that has previously been used for the study of bone and mineral chemistry. Ivory and bone have similar biochemical properties, making Raman spectroscopy a promising method for species identification based on ivory. This study aimed to test the hypothesis that it is possible to identify differences in the chemistry of mammoth and elephant ivory using Raman spectroscopy. Mammoth and elephant tusks were obtained from the Natural History Museum in London, UK. Included in this study were eight samples of ivory from Mammuthus primigenius, two samples of carved ivory bangles from Africa (Loxodonta species), and one cross section of a tusk from Elephas maximus. The ivory was scanned using an inVia Raman micro spectrometer equipped with a x50 objective lens and a 785nm laser. Spectra were acquired using line maps and individual spectral points were acquired randomly or at points of interest on all samples. The data was then analysed using principal component analysis (PCA) with use of an in-house MATLAB script. Univariate analysis of peak intensity ratios of phosphate to amide I and III peaks, and carbonate to phosphate peaks showed statistical differences (p