Complex patterns in fossilized stromatolites revealed by hyperspectral imaging (400–2496 nm). Issue 5 (5th May 2016)

• Record Type: Journal Article
• Authors: Murphy, R. J.; Van Kranendonk, M. J.; Kelloway, S. J.; Wainwright, I. E.
• Is Part Of: Geobiology. Volume 14:Issue 5(2016)
• Page Start: 419
• Page End: 439
• Publication Date: 2016-05-05
• Subjects: Geobiology -- Periodicals; Biogeochemistry -- Periodicals; Ecology -- Periodicals; 551
• DOI: 10.1111/gbi.12184
• Languages: English
• ISSNs: 1472-4677

Abstract

Abstract: Hyperspectral imaging (400–2496 nm) was used to quantitatively map surface textures and compositional variations in stromatolites to determine whether complexity of textures could be used as evidence to support biogenicity in the absence of preserved biomarkers. Four samples of 2.72–2.4 Ga stromatolites from a variety of settings, encompassing marine and lacustrine environments, were selected for hyperspectral imaging. Images of the sawn surfaces of samples were processed to identify reflectance and mineral absorption features and quantify their intensity (as an index of mineral abundance) using automated feature extraction. Amounts of ferrous iron were quantified using a ratio of reflectance at 1650 and 1299 nm. Visible near infrared imagery (400–970 nm) did not reveal additional textural patterns to those obtained from visual inspection. Shortwave infrared imagery (1000–2496 nm), however, revealed complex laminar and convoluted patterns, including a distinctive texture of sharp peaks and broad, low troughs in one sample, similar to living tufted microbial mats. Spectral analysis revealed another sample to be composed of dolomite. Two other samples were dominated by calcite or chlorite ± illite. Large variations in amounts of ferrous iron were found, but ferric iron was exclusively located in the oxidation crust. Hyperspectral imaging revealed large differences between parts of a sample of biogenic and non‐biogenic origin. The former was characterized by calcite with varying amounts of ferrous iron, distributed in lenticular, convoluted patterns; the latter by Mg‐Fe chlorite with large amounts of aluminium silicate, distributed as fine laminar layers. All minerals identified by hyperspectral imaging were confirmed by thin section petrography and XRD analyses. Spatial statistics generated from quantitative minerals maps showed different patterns between these different parts of the sample. Thus, hyperspectral imaging was shown to be a powerful tool for detecting structures in stromatolites that could be used, together with other lines of evidence, to support biogenicity.