Beware of Spatial Autocorrelation When Applying Machine Learning Algorithms to Borehole Geophysical Logs

Neil Terry; Carole D. Johnson; Frederick D. Day-Lewis; Beth Parker; Lee Slater

doi:10.1111/gwat.13081

Beware of Spatial Autocorrelation When Applying Machine Learning Algorithms to Borehole Geophysical Logs

Neil Terry, Carole D. Johnson, Frederick D. Day-Lewis, Beth Parker, Lee Slater

Rutgers Newark, Earth & Environmental Science

Research output: Contribution to journal › Comment/debate › peer-review

3 Scopus citations

Original language	English (US)
Pages (from-to)	315-319
Number of pages	5
Journal	Groundwater
Volume	59
Issue number	3
DOIs	https://doi.org/10.1111/gwat.13081
State	Published - May 1 2021

All Science Journal Classification (ASJC) codes

Water Science and Technology
Computers in Earth Sciences

Access to Document

10.1111/gwat.13081

Cite this

@article{8a28311022a340c491615471b72325dc,

title = "Beware of Spatial Autocorrelation When Applying Machine Learning Algorithms to Borehole Geophysical Logs",

author = "Neil Terry and Johnson, {Carole D.} and Day-Lewis, {Frederick D.} and Beth Parker and Lee Slater",

note = "Funding Information: The data described in this paper are available from Johnson et al. (2021). Rock core VOC data is provided by Dr. B. Parker's research program funded by site owners and a Canadian NSERC Industrial Research Chair (NSERC IRCPJ 363783-18) based at the University of Guelph since 2007, with G360 Institute for Groundwater Research contributions from: Maria Gorecka, M.Sc., as Senior Analytical Chemist and Laboratory Manager, performing all VOC contaminant analyses on the numerous rock core samples for the NAWC (NJ) and SSFL (CA) sites and oversight of the analyses by Stone Environmental for the Hydrite (WI) site; Peeter Pehme, Ph.D., for his expertise with advanced borehole geophysical and hydrophysical logging and supporting the field logging efforts at all three sites (NAWC, SSFL, and Hydrite); and Steven Chapman, MS.c., P.Eng., Amanda Pierce, M.Sc., P.Geo. and Jessica Meyer, Ph.D. as Senior Hydrogeologists with lead project management roles for G360 tasks including support of field work and site data/knowledge transfer to project participants for the NAWC (NJ), SSFL (CA) and Hydrite (WI) sites, respectively. Financial support for this research was provided by the Department of Defense Strategic Environmental Research and Development Program under project ER-2421, the U.S. Geological Survey Water Availability and Use Science Program, and the U.S. Geological Survey Toxic Substances Hydrology Program. We also thank Daniel Beckman, Jeremy White, and an anonymous reviewer for their help in reviewing and improving this technical commentary. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. Funding Information: The data described in this paper are available from Johnson et al. ( 2021 ). Rock core VOC data is provided by Dr. B. Parker's research program funded by site owners and a Canadian NSERC Industrial Research Chair (NSERC IRCPJ 363783‐18) based at the University of Guelph since 2007, with G360 Institute for Groundwater Research contributions from: Maria Gorecka, M.Sc., as Senior Analytical Chemist and Laboratory Manager, performing all VOC contaminant analyses on the numerous rock core samples for the NAWC (NJ) and SSFL (CA) sites and oversight of the analyses by Stone Environmental for the Hydrite (WI) site; Peeter Pehme, Ph.D., for his expertise with advanced borehole geophysical and hydrophysical logging and supporting the field logging efforts at all three sites (NAWC, SSFL, and Hydrite); and Steven Chapman, MS.c., P.Eng., Amanda Pierce, M.Sc., P.Geo. and Jessica Meyer, Ph.D. as Senior Hydrogeologists with lead project management roles for G360 tasks including support of field work and site data/knowledge transfer to project participants for the NAWC (NJ), SSFL (CA) and Hydrite (WI) sites, respectively. Financial support for this research was provided by the Department of Defense Strategic Environmental Research and Development Program under project ER‐2421, the U.S. Geological Survey Water Availability and Use Science Program, and the U.S. Geological Survey Toxic Substances Hydrology Program. We also thank Daniel Beckman, Jeremy White, and an anonymous reviewer for their help in reviewing and improving this technical commentary. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. ",

year = "2021",

month = may,

day = "1",

doi = "10.1111/gwat.13081",

language = "English (US)",

volume = "59",

pages = "315--319",

journal = "Groundwater",

issn = "0017-467X",

publisher = "Wiley-Blackwell",

number = "3",

}

TY - JOUR

T1 - Beware of Spatial Autocorrelation When Applying Machine Learning Algorithms to Borehole Geophysical Logs

AU - Terry, Neil

AU - Johnson, Carole D.

AU - Day-Lewis, Frederick D.

AU - Parker, Beth

AU - Slater, Lee

N1 - Funding Information: The data described in this paper are available from Johnson et al. (2021). Rock core VOC data is provided by Dr. B. Parker's research program funded by site owners and a Canadian NSERC Industrial Research Chair (NSERC IRCPJ 363783-18) based at the University of Guelph since 2007, with G360 Institute for Groundwater Research contributions from: Maria Gorecka, M.Sc., as Senior Analytical Chemist and Laboratory Manager, performing all VOC contaminant analyses on the numerous rock core samples for the NAWC (NJ) and SSFL (CA) sites and oversight of the analyses by Stone Environmental for the Hydrite (WI) site; Peeter Pehme, Ph.D., for his expertise with advanced borehole geophysical and hydrophysical logging and supporting the field logging efforts at all three sites (NAWC, SSFL, and Hydrite); and Steven Chapman, MS.c., P.Eng., Amanda Pierce, M.Sc., P.Geo. and Jessica Meyer, Ph.D. as Senior Hydrogeologists with lead project management roles for G360 tasks including support of field work and site data/knowledge transfer to project participants for the NAWC (NJ), SSFL (CA) and Hydrite (WI) sites, respectively. Financial support for this research was provided by the Department of Defense Strategic Environmental Research and Development Program under project ER-2421, the U.S. Geological Survey Water Availability and Use Science Program, and the U.S. Geological Survey Toxic Substances Hydrology Program. We also thank Daniel Beckman, Jeremy White, and an anonymous reviewer for their help in reviewing and improving this technical commentary. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. Funding Information: The data described in this paper are available from Johnson et al. ( 2021 ). Rock core VOC data is provided by Dr. B. Parker's research program funded by site owners and a Canadian NSERC Industrial Research Chair (NSERC IRCPJ 363783‐18) based at the University of Guelph since 2007, with G360 Institute for Groundwater Research contributions from: Maria Gorecka, M.Sc., as Senior Analytical Chemist and Laboratory Manager, performing all VOC contaminant analyses on the numerous rock core samples for the NAWC (NJ) and SSFL (CA) sites and oversight of the analyses by Stone Environmental for the Hydrite (WI) site; Peeter Pehme, Ph.D., for his expertise with advanced borehole geophysical and hydrophysical logging and supporting the field logging efforts at all three sites (NAWC, SSFL, and Hydrite); and Steven Chapman, MS.c., P.Eng., Amanda Pierce, M.Sc., P.Geo. and Jessica Meyer, Ph.D. as Senior Hydrogeologists with lead project management roles for G360 tasks including support of field work and site data/knowledge transfer to project participants for the NAWC (NJ), SSFL (CA) and Hydrite (WI) sites, respectively. Financial support for this research was provided by the Department of Defense Strategic Environmental Research and Development Program under project ER‐2421, the U.S. Geological Survey Water Availability and Use Science Program, and the U.S. Geological Survey Toxic Substances Hydrology Program. We also thank Daniel Beckman, Jeremy White, and an anonymous reviewer for their help in reviewing and improving this technical commentary. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

PY - 2021/5/1

Y1 - 2021/5/1

UR - http://www.scopus.com/inward/record.url?scp=85101441465&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85101441465&partnerID=8YFLogxK

U2 - 10.1111/gwat.13081

DO - 10.1111/gwat.13081

M3 - Comment/debate

C2 - 33521933

AN - SCOPUS:85101441465

SN - 0017-467X

VL - 59

SP - 315

EP - 319

JO - Groundwater

JF - Groundwater

IS - 3

ER -

Beware of Spatial Autocorrelation When Applying Machine Learning Algorithms to Borehole Geophysical Logs

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Cite this