Mapping soil cadmium content using multi-spectral satellite images and multiple-residual-stacking model : Incorporating information from homologous pollution and spectrally active materials
Tan, Chao ; Luan, Haijun LU ; He, Qiuhua ; Zheng, Yaling ; Lin, Zhenhong and Wang, Lanhui LU
(2025)
In Journal of Hazardous Materials
485.
- Abstract
Soil cadmium (Cd) contamination significantly threatens ecosystems and human health. Traditional geochemical investigation, although accurate, is impractical for wide-area and frequent monitoring applications. Multi-spectral satellite images combined with the homologous pollution information (HPI) and the spectral and content information of soil organic matter (SOMSCI) is an unconventional and promising approach for large-scale, dynamic soil heavy metal (SHM) monitoring. Based on a novel Multiple-Residual-Stacked (MRS) machine-learning framework, the study estimated the soil Cd content in Yueyang City, China, during the past decade (2014–2023) using Landsat 8 images. Within it, three feature construction methods and four models were... (More)
Soil cadmium (Cd) contamination significantly threatens ecosystems and human health. Traditional geochemical investigation, although accurate, is impractical for wide-area and frequent monitoring applications. Multi-spectral satellite images combined with the homologous pollution information (HPI) and the spectral and content information of soil organic matter (SOMSCI) is an unconventional and promising approach for large-scale, dynamic soil heavy metal (SHM) monitoring. Based on a novel Multiple-Residual-Stacked (MRS) machine-learning framework, the study estimated the soil Cd content in Yueyang City, China, during the past decade (2014–2023) using Landsat 8 images. Within it, three feature construction methods and four models were employed. The experimental results indicate that the XGB-MRS model incorporating HPI and SOMSCI significantly improved the estimation performance (RPD exceeded 90 %, R2, RMSE, and MAE exceeded 40 %). Moreover, against 243 ground samples during 2016–2022, the average overall estimation accuracy exceeded 80 %, validating the model's robustness and practicality. Furthermore, the descending order of contribution in the modelling is environmental auxiliary variables (55 %), HPI and SOMSCI (26 %), and spectral information (19 %). The fertilizer usage has direct (up to 2 years) and delayed (3–5 years) effects on soil Cd accumulation. Overall, our study provides a scalable framework for monitoring global SHM pollution using open-source multi-spectral satellite data.
(Less)
- author
- Tan, Chao
; Luan, Haijun
LU
; He, Qiuhua
; Zheng, Yaling
; Lin, Zhenhong
and Wang, Lanhui
LU
- organization
- publishing date
- 2025
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Homologous pollution, Landsat 8, Multiple-Residual-Stacking, Soil cadmium, Spectrally active materials
- in
- Journal of Hazardous Materials
- volume
- 485
- article number
- 136755
- publisher
- Elsevier
- external identifiers
-
- scopus:85211372983
- pmid:39667148
- ISSN
- 0304-3894
- DOI
- 10.1016/j.jhazmat.2024.136755
- language
- English
- LU publication?
- yes
- id
- df430b32-14ba-4754-b8e7-46817222bb89
- date added to LUP
- 2025-02-28 11:44:48
- date last changed
- 2025-07-04 22:52:37
@article{df430b32-14ba-4754-b8e7-46817222bb89,
abstract = {{<p>Soil cadmium (Cd) contamination significantly threatens ecosystems and human health. Traditional geochemical investigation, although accurate, is impractical for wide-area and frequent monitoring applications. Multi-spectral satellite images combined with the homologous pollution information (HPI) and the spectral and content information of soil organic matter (SOMSCI) is an unconventional and promising approach for large-scale, dynamic soil heavy metal (SHM) monitoring. Based on a novel Multiple-Residual-Stacked (MRS) machine-learning framework, the study estimated the soil Cd content in Yueyang City, China, during the past decade (2014–2023) using Landsat 8 images. Within it, three feature construction methods and four models were employed. The experimental results indicate that the XGB-MRS model incorporating HPI and SOMSCI significantly improved the estimation performance (RPD exceeded 90 %, R<sup>2</sup>, RMSE, and MAE exceeded 40 %). Moreover, against 243 ground samples during 2016–2022, the average overall estimation accuracy exceeded 80 %, validating the model's robustness and practicality. Furthermore, the descending order of contribution in the modelling is environmental auxiliary variables (55 %), HPI and SOMSCI (26 %), and spectral information (19 %). The fertilizer usage has direct (up to 2 years) and delayed (3–5 years) effects on soil Cd accumulation. Overall, our study provides a scalable framework for monitoring global SHM pollution using open-source multi-spectral satellite data.</p>}},
author = {{Tan, Chao and Luan, Haijun and He, Qiuhua and Zheng, Yaling and Lin, Zhenhong and Wang, Lanhui}},
issn = {{0304-3894}},
keywords = {{Homologous pollution; Landsat 8; Multiple-Residual-Stacking; Soil cadmium; Spectrally active materials}},
language = {{eng}},
publisher = {{Elsevier}},
series = {{Journal of Hazardous Materials}},
title = {{Mapping soil cadmium content using multi-spectral satellite images and multiple-residual-stacking model : Incorporating information from homologous pollution and spectrally active materials}},
url = {{http://dx.doi.org/10.1016/j.jhazmat.2024.136755}},
doi = {{10.1016/j.jhazmat.2024.136755}},
volume = {{485}},
year = {{2025}},
}