Multivariate analysis applied to X-ray fluorescence to assess soil contamination pathways: case studies of mass magnetic susceptibility in soils near abandoned coal and W/Sn mines

Milinovic, Jelena; Santos, Patrícia; Sant’Ovaia, Helena; Futuro, Aurora; Pereira, Carlos M.; Murton, Bramley J. ORCID: https://orcid.org/0000-0003-1522-1191; Flores, Deolinda; Azenha, Manuel. 2024 Multivariate analysis applied to X-ray fluorescence to assess soil contamination pathways: case studies of mass magnetic susceptibility in soils near abandoned coal and W/Sn mines. Environmental Geochemistry and Health, 46 (6). 10.1007/s10653-024-01988-3

Before downloading, please read NORA policies.

Preview

Text
s10653-024-01988-3.pdf - Published Version
Available under License Creative Commons Attribution 4.0.
Download (2MB) | Preview

Official URL: http://dx.doi.org/10.1007/s10653-024-01988-3

Abstract/Summary

Determining the origin and pathways of contaminants in the natural environment is key to informing any mitigation process. The mass magnetic susceptibility of soils allows a rapid method to measure the concentration of magnetic minerals, derived from anthropogenic activities such as mining or industrial processes, i.e., smelting metals (technogenic origin), or from the local bedrock (of geogenic origin). This is especially effective when combined with rapid geochemical analyses of soils. The use of multivariate analysis (MVA) elucidates complex multiple-component relationships between soil geochemistry and magnetic susceptibility. In the case of soil mining sites, X-ray fluorescence (XRF) spectroscopic data of soils contaminated by mine waste shows statistically significant relationships between magnetic susceptibility and some base metal species (e.g., Fe, Pb, Zn, etc.). Here, we show how qualitative and quantitative MVA methodologies can be used to assess soil contamination pathways using mass magnetic susceptibility and XRF spectra of soils near abandoned coal and W/Sn mines (NW Portugal). Principal component analysis (PCA) showed how the first two primary components (PC-1 + PC-2) explained 94% of the sample variability, grouped them according to their geochemistry and magnetic susceptibility in to geogenic and technogenic groups. Regression analyses showed a strong positive correlation (R2 > 0.95) between soil geochemistry and magnetic properties at the local scale. These parameters provided an insight into the multi-element variables that control magnetic susceptibility and indicated the possibility of efficient assessment of potentially contaminated sites through mass-specific soil magnetism.

Item Type:	Publication - Article
Digital Object Identifier (DOI):	10.1007/s10653-024-01988-3
ISSN:	0269-4042
Additional Keywords:	Metal contamination, Magnetic minerals, Geochemistry, XRF spectra, PCA and cluster analysis, Regression analyses
Date made live:	12 Jun 2024 12:06 +0 (UTC)
URI:	https://nora.nerc.ac.uk/id/eprint/537567

Item Type:

Publication - Article

Digital Object Identifier (DOI):

10.1007/s10653-024-01988-3

ISSN:

0269-4042

Additional Keywords:

Metal contamination, Magnetic minerals, Geochemistry, XRF spectra, PCA and cluster analysis, Regression analyses

Date made live:

12 Jun 2024 12:06 +0 (UTC)

URI:

https://nora.nerc.ac.uk/id/eprint/537567