Ultrahigh Dimensional Variable Selection for Interpolation of Point Referenced Spatial Data: A Digital Soil Mapping Case Study

Benjamin R Fitzpatrick; David W Lamb; Kerrie Mengersen

doi:10.1371/journal.pone.0162489

Ultrahigh Dimensional Variable Selection for Interpolation of Point Referenced Spatial Data: A Digital Soil Mapping Case Study

PLoS One. 2016 Sep 7;11(9):e0162489. doi: 10.1371/journal.pone.0162489. eCollection 2016.

Authors

Benjamin R Fitzpatrick^{1

2

3}, David W Lamb^{2

4}, Kerrie Mengersen^{1

2

3

5}

Affiliations

¹ Mathematical Sciences School, Queensland University of Technology (QUT), Brisbane, QLD 4001, Australia.
² Cooperative Research Centre for Spatial Information (CRCSI), Carlton, VIC 3053, Australia.
³ Institute for Future Environments, Queensland University of Technology (QUT), Brisbane, QLD 4001, Australia.
⁴ Precision Agriculture Research Group, University of New England, Armidale, NSW 2351, Australia.
⁵ ARC Centre of Excellence for Mathematical and Statistical Frontiers, Queensland University of Technology (QUT), Brisbane, QLD 4001, Australia.

Abstract

Modern soil mapping is characterised by the need to interpolate point referenced (geostatistical) observations and the availability of large numbers of environmental characteristics for consideration as covariates to aid this interpolation. Modelling tasks of this nature also occur in other fields such as biogeography and environmental science. This analysis employs the Least Angle Regression (LAR) algorithm for fitting Least Absolute Shrinkage and Selection Operator (LASSO) penalized Multiple Linear Regressions models. This analysis demonstrates the efficiency of the LAR algorithm at selecting covariates to aid the interpolation of geostatistical soil carbon observations. Where an exhaustive search of the models that could be constructed from 800 potential covariate terms and 60 observations would be prohibitively demanding, LASSO variable selection is accomplished with trivial computational investment.

MeSH terms

Algorithms*
Carbon / analysis
Organic Chemicals / analysis
Reproducibility of Results
Soil*
Statistics as Topic*

Substances

Organic Chemicals
Soil
Carbon

Grants and funding

This work was funded by the CRC for Spatial Information (CRCSI), established and supported under the Australian Government Cooperative Research Centres Programme. One of the authors (BRF) wishes to acknowledge the receipt of a Postgraduate Scholarship from the CRCSI (http://www.crcsi.com.au/).