Identifying factors influencing trace metal concentrations in urban residential soil using an optimal parameter-based geographical detector model

Australia's national citizen science program VegeSafe has collected and analysed over 26,000 residential garden soil samples for their trace metal concentrations, enabling a more comprehensive understanding of the factors influencing contamination. Here we analysed spatial data from 8221 soil s...

Full description

Bibliographic Details
Main Authors: Liu, X., Taylor, M.P., Song, Yongze, Aelion, C.M.
Format: Journal Article
Language:English
Published: 2025
Subjects:
Online Access:http://hdl.handle.net/20.500.11937/97968
_version_ 1848766345878962176
author Liu, X.
Taylor, M.P.
Song, Yongze
Aelion, C.M.
author_facet Liu, X.
Taylor, M.P.
Song, Yongze
Aelion, C.M.
author_sort Liu, X.
building Curtin Institutional Repository
collection Online Access
description Australia's national citizen science program VegeSafe has collected and analysed over 26,000 residential garden soil samples for their trace metal concentrations, enabling a more comprehensive understanding of the factors influencing contamination. Here we analysed spatial data from 8221 soil samples collected from 1828 homes across Greater Sydney, Australia's largest city, using an optimal parameter-based geographical detector (OPGD) model to quantify anthropogenic and natural factors influencing urban residential soil trace metal concentrations. The OPGD model identifies optimal spatial scales and discretization parameters, enhancing spatial stratified heterogeneity analysis. Results demonstrate anthropogenic factors, such as aged/painted home density, road density, and industrial trace metal emissions, primarily contribute to soil concentrations of arsenic (As), cadmium (Cd), chromium (Cr), copper (Cu), lead (Pb), and zinc (Zn). By contrast, natural factors including soil pH, regolith stability, and soil type dominate soil manganese (Mn) and nickel (Ni) concentrations. Strongest interactive effects typically involve an anthropogenic and a natural factor. Notably, 42.7 % of homes within the study area had at least one soil sample with Pb concentrations exceeding the Australian residential guideline of 300 mg/kg. Locations with potential risk of harm are identified to inform targeted mitigation strategies. Compared to machine learning methods, the OPGD model offers a more reliable and comprehensive assessment of urban residential soil trace metal contamination.
first_indexed 2025-11-14T11:49:41Z
format Journal Article
id curtin-20.500.11937-97968
institution Curtin University Malaysia
institution_category Local University
language eng
last_indexed 2025-11-14T11:49:41Z
publishDate 2025
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-979682025-07-22T06:02:02Z Identifying factors influencing trace metal concentrations in urban residential soil using an optimal parameter-based geographical detector model Liu, X. Taylor, M.P. Song, Yongze Aelion, C.M. Anthropogenic factors GIS Geo-detector Lead (Pb) exposure Machine learning Natural factors Risk assessment Spatial heterogeneity analysis Australia's national citizen science program VegeSafe has collected and analysed over 26,000 residential garden soil samples for their trace metal concentrations, enabling a more comprehensive understanding of the factors influencing contamination. Here we analysed spatial data from 8221 soil samples collected from 1828 homes across Greater Sydney, Australia's largest city, using an optimal parameter-based geographical detector (OPGD) model to quantify anthropogenic and natural factors influencing urban residential soil trace metal concentrations. The OPGD model identifies optimal spatial scales and discretization parameters, enhancing spatial stratified heterogeneity analysis. Results demonstrate anthropogenic factors, such as aged/painted home density, road density, and industrial trace metal emissions, primarily contribute to soil concentrations of arsenic (As), cadmium (Cd), chromium (Cr), copper (Cu), lead (Pb), and zinc (Zn). By contrast, natural factors including soil pH, regolith stability, and soil type dominate soil manganese (Mn) and nickel (Ni) concentrations. Strongest interactive effects typically involve an anthropogenic and a natural factor. Notably, 42.7 % of homes within the study area had at least one soil sample with Pb concentrations exceeding the Australian residential guideline of 300 mg/kg. Locations with potential risk of harm are identified to inform targeted mitigation strategies. Compared to machine learning methods, the OPGD model offers a more reliable and comprehensive assessment of urban residential soil trace metal contamination. 2025 Journal Article http://hdl.handle.net/20.500.11937/97968 10.1016/j.envres.2025.122045 eng http://creativecommons.org/licenses/by/4.0/ fulltext
spellingShingle Anthropogenic factors
GIS
Geo-detector
Lead (Pb) exposure
Machine learning
Natural factors
Risk assessment
Spatial heterogeneity analysis
Liu, X.
Taylor, M.P.
Song, Yongze
Aelion, C.M.
Identifying factors influencing trace metal concentrations in urban residential soil using an optimal parameter-based geographical detector model
title Identifying factors influencing trace metal concentrations in urban residential soil using an optimal parameter-based geographical detector model
title_full Identifying factors influencing trace metal concentrations in urban residential soil using an optimal parameter-based geographical detector model
title_fullStr Identifying factors influencing trace metal concentrations in urban residential soil using an optimal parameter-based geographical detector model
title_full_unstemmed Identifying factors influencing trace metal concentrations in urban residential soil using an optimal parameter-based geographical detector model
title_short Identifying factors influencing trace metal concentrations in urban residential soil using an optimal parameter-based geographical detector model
title_sort identifying factors influencing trace metal concentrations in urban residential soil using an optimal parameter-based geographical detector model
topic Anthropogenic factors
GIS
Geo-detector
Lead (Pb) exposure
Machine learning
Natural factors
Risk assessment
Spatial heterogeneity analysis
url http://hdl.handle.net/20.500.11937/97968