Indexed in

License and use

Citations

Altmetrics

Analysis of institutional authors

Share

Factor analysis in residual soils of the Iberian Pyrite Belt (Spain): comparison between raw data, log-transformation data and compositional data

Publicated to:Geochemistry-Exploration Environment Analysis. 24 (2): geochem2024005- - 2024-05-27 24(2), DOI: https://doi.org/10.1144/geochem2024-005

Authors: Martín-Méndez, I; Llamas-Borrajo, J; Lois, AL; Locutura, J

Affiliations

Abstract

The Iberian Pyrite Belt (IPB) is a metallogenic province in SW Spain and Portugal hosting the largest concentration of massive sulfide deposits worldwide. Exploration campaigns in both the Spanish and Portuguese sectors of the IPB have increased recently due to the rise in metal prices. Within this framework, distinguishing geochemical features associated with natural phenomena and isolating geogenic anomalies from anthropogenic ones can pose a challenge. This contribution uses the residual soil geochemical database of the IPB (Spain) to examine numerous variables, encompassing major, minor and trace elements. Some of these variables commonly exhibit high correlations owing to consistent geochemical behaviour. However, the influence of anthropogenic factors tends to elevate data variability, occasionally masking the natural relationships that govern their distributions. We apply different treatments of data to develop factor analysis using log-transformed data, and centred log-ratio (clr) transformed data to compare and improve the geochemical interpretation of this important zone. Factor analysis has been developed with these results to compare with previously published research on factor analysis in raw data. Factor score interpolated maps were also generated using both lognormal and clr-transformed data to visualize better the distribution and the different geochemical associations in the IPB. This study shows the importance of the different data treatments and the improvement of the clr-transformed multivariate analysis to reduce the dilution or overestimation of the results of some elements that cause erroneous interpretations of the data.

Keywords