Forensic soil provenancing in an urban/suburban setting: A sequential multivariate approach

Area of Science:

• Forensic geology
• Environmental science
• Geochemistry

Background:

• Soil provenancing is crucial for forensic and intelligence applications.
• Accurate soil analysis requires robust methodologies for identifying origins.
• Previous methods often lack comprehensive uncertainty quantification.

Purpose of the Study:

• To develop and test an empirical soil provenancing method using diverse analytical techniques.
• To assess the precision and accuracy of different analytical methods for soil provenancing.
• To evaluate the role of principal components (PCs) in improving provenance predictions.

Main Methods:

• Compositional data analysis of 268 topsoil samples from North Canberra.
• Utilized Fourier transform infrared spectroscopy (FTIR), magnetic susceptibility (MS), X-ray fluorescence (XRF), and inductively coupled plasma-mass spectrometry (ICP-MS).
• Employed sequential provenancing with interpolated raster grids, uncertainty propagation, and cumulative provenance maps.

Main Results:

• Fourier transform infrared spectroscopy (FTIR) and magnetic susceptibility (MS) provided the most precise and accurate mineralogical provenance predictions.
• X-ray fluorescence (XRF) and total ICP-MS (geochemistry) also offered high precision and accuracy.
• Including principal components (PCs) marginally improved provenancing performance.

Conclusions:

• Empirical soil provenancing, particularly using FTIR, MS, XRF, and ICP-MS, is a valuable tool for forensic and intelligence applications.
• Maximizing the number of analytes and analytical techniques enhances soil provenancing capabilities.
• The developed method demonstrates potential despite acknowledged limitations, offering a robust approach to soil origin determination.