Supporting information for: Clade-specific elemental signatures across an Early Triassic marine fauna pave the way for deciphering the affinities of unidentifiable fossils
Data files
Aug 27, 2025 version files 147.11 MB
-
README.md
1.94 KB
-
S1_File_raw_data_Smith_et_al.zip
145.75 MB
-
S2_File_Supplementary_Materials_Smith_et_al.pdf
1.32 MB
-
S3_File_R_script_Smith_et_al.R
44.11 KB
Abstract
In palaeontology, the observation of morphological characters is at the heart of species determination. Nonetheless, since most fossils have undergone considerable morphological loss, distortion, and/or flattening throughout their taphonomic history, the use of visual techniques often remains limited. Complementary approaches such as geochemical analyses or molecular palaeontology are increasingly developed. However, them as well remain limited by the preservation state and diagenetic overprinting of the vast majority of fossils. Based on data obtained by state-of-the-art non-destructive synchrotron micro-X-ray fluorescence (µXRF) major-to-trace elemental mapping of Early Triassic Paris Biota fossils, we show here, at least within a single fossil fauna, the existence of a clade-specific elemental signature. Using complete multi-elemental µXRF spectra instead of elemental quantifications/concentrations, we set a data-formatting protocol that allows us to compare the morphology of the spectra. We then statistically demonstrate the existence of a geochemical discrimination between specimens of different clade despite intra-clade mineralogical variability, and build a “elemental-comparative taxonomic identification” model accordingly. The latter, that goes beyond the simple distinction of tissue nature or type of preservation, is all the more important as it appears to hold the potential to identify some hitherto unrecognizable specimens of the fossil record.
The supporting information consists of:
S1 - Raw data ("S1_File_raw_data_Smith_et_al.zip") including:
- Data S1 - Pictures of the analysed samples exported from PyMCA, with the studied zones highlighted by shaded areas. Files are sorted per experimental session (DiffAbs-2018, PUMA-2021, and PUMA-2024).
- Image zones_PUMA_2021: 96 documents (pictures in PDF format)
- Image zones_diffabs_2018: 84 documents (pictures in PDF format)
- Image zones_PUMA_2024: 84 documents (pictures in PDF format)
For each picture, the axes correspond to the pixel rows and columns, as set by default in PyMCA. File names follow the conventions in Table S1 and are completed with “sed” or “foss” to indicate whether the studied area corresponds to sediment or fossil material, respectively.
- Data S2 - Mean µ-XRF spectrum of each studied zone in respect to Data S1. The data presents the mean value of photons measured per pixel ("counts" column) per energy level ("energy" column) over the sampled area. The documents are in .CSV format.
- Data S3 - Mean µ-XRF spectrum of each studied zone regrouped per- specimens (following the Table S1 naming convention) such as to be statistically analysed using R software. The documents are in .CSV format.
S2 - PDF file ("S2_File_Supplementary_Materials_Smith_et_al.pdf") including:
- Figure S1
- Figure S2
- Supplementary text
- The µXRF spectra morphological descriptors approach
- Data acquisition setups
- Figure S3
- Table S1
- References
S3 - R data analysis script ("S3_File_R_script_Smith_et_al.R")