Skip to main content

Research Repository

Advanced Search

The singularity index for soil geochemical variables, and a mixture model for its interpretation

Lark, R.M.; Patton, M.; Ander, E.L.; Reay, D.M.

The singularity index for soil geochemical variables, and a mixture model for its interpretation Thumbnail


Authors

Profile Image

MURRAY LARK MURRAY.LARK@NOTTINGHAM.AC.UK
Professor of Geoinformatics

M. Patton

E.L. Ander

D.M. Reay



Abstract

A geochemical anomaly is a concentration of an element or other constituent in a medium (soil, sediment or surface water) which is unusual in its local setting. Geochemical anomalies may be interesting as indicators of processes such as point contamination or mineralizations. They may therefore be practically useful, indicating sources of pollution or mineral deposits which may be of economic value. As defined, a geochemical anomaly is not merely a large (or small) concentration of a constituent as compared to the marginal distribution. To detect anomalies we must therefore do more than simply map the spatial distribution of the constituent. One proposed approach makes use of a singularity index based on fractal representation of spatial variation. The singularity index can be computed from local concentration measures in nested windows. In this paper we propose an approach to compute threshold values for the index to identify enrichment and depletion anomalies, separate from background information. The approach is based on a mixture model for the singularity index, and it can be supported by computing a distribution for background values of the index by parametric bootstrapping from a robustly-estimated variogram model for the target constituent. This approach is illustrated here using data on elements in the soil in four settings in Great Britain and Ireland.

Journal Article Type Article
Acceptance Date Feb 21, 2018
Online Publication Date Mar 20, 2018
Publication Date Aug 1, 2018
Deposit Date Apr 12, 2018
Publicly Available Date Mar 21, 2019
Journal Geoderma
Print ISSN 0016-7061
Electronic ISSN 0016-7061
Publisher Elsevier
Peer Reviewed Peer Reviewed
Volume 323
DOI https://doi.org/10.1016/j.geoderma.2018.02.032
Keywords Geochemistry; Anomalies; Singularity; Fractal; Mixture model
Public URL https://nottingham-repository.worktribe.com/output/948734
Publisher URL https://www.sciencedirect.com/science/article/pii/S0016706117308765?via%3Dihub