Standard

Mapping of potentially toxic elements in the urban topsoil of St. Petersburg (Russia) using regression kriging and random forest algorithms. / Сулейманов, Азамат Русланович; Поляков, Вячеслав Игоревич; Козлов, Александр; Абакумов, Евгений Васильевич; Кузьменко, Петр; Телягиссов, Салават.

In: Environmental Earth Sciences, Vol. 82, No. 23, 561, 29.11.2023.

Research output: Contribution to journalArticlepeer-review

Harvard

APA

Vancouver

Author

BibTeX

@article{7eae757617ba4fa6ba2181087323062b,
title = "Mapping of potentially toxic elements in the urban topsoil of St. Petersburg (Russia) using regression kriging and random forest algorithms",
abstract = "The world's largest cities are characterized by environmental pollution, which endangers human health and biological species. Therefore, environmental monitoring and mapping of pollution levels are important tasks. To this end, we applied regression kriging (RK) and random forest (RF) for digital mapping of potentially toxic elements (PTEs) on the territory of Saint Petersburg (Russia). We predicted concentrations of As, Cd, Cu, Hg, Ni, Pb and Zn PTEs based on 87 sites. The covariates were represented by remote sensing, terrain and anthropogenic variables. The results showed that the elements were characterized by high variability and their content varied between 1 and 37 mg/kg for As, 0.1 and 2.7 for Cd, 4.7 and 674 for Cu, 0.005 and 3.6 for Hg, 2.8 and 66.8 for Ni, 3.4 and 858.8 for Pb, 15.4 and 1306.5 for Zn. We found that RK was the best performing model for Cu, Hg, Ni, Pb and Zn, with the lowest mean absolute error (MAE), root mean squared error (RMSE), and highest coefficient of determination (R2) and Nash–Sutcliffe model efficiency coefficient (NSE) values. RF method was better for predicting As content, while the performance of the models for Cd was the same. Nevertheless, R2 values remained below 0.23, signifying a difficulty in effectively modeling PTEs within urbanized regions. The study provides insight into the benefits of combining regression analysis and kriging, and highlights the importance of considering multiple modeling approaches in spatial prediction of PTEs.",
keywords = "Digital soil mapping, Heavy metals, Kriging, Machine learning, Potentially toxic elements",
author = "Сулейманов, {Азамат Русланович} and Поляков, {Вячеслав Игоревич} and Александр Козлов and Абакумов, {Евгений Васильевич} and Петр Кузьменко and Салават Телягиссов",
year = "2023",
month = nov,
day = "29",
doi = "10.1007/s12665-023-11272-9",
language = "English",
volume = "82",
journal = "Environmental Earth Sciences",
issn = "1866-6280",
publisher = "Springer Nature",
number = "23",

}

RIS

TY - JOUR

T1 - Mapping of potentially toxic elements in the urban topsoil of St. Petersburg (Russia) using regression kriging and random forest algorithms

AU - Сулейманов, Азамат Русланович

AU - Поляков, Вячеслав Игоревич

AU - Козлов, Александр

AU - Абакумов, Евгений Васильевич

AU - Кузьменко, Петр

AU - Телягиссов, Салават

PY - 2023/11/29

Y1 - 2023/11/29

N2 - The world's largest cities are characterized by environmental pollution, which endangers human health and biological species. Therefore, environmental monitoring and mapping of pollution levels are important tasks. To this end, we applied regression kriging (RK) and random forest (RF) for digital mapping of potentially toxic elements (PTEs) on the territory of Saint Petersburg (Russia). We predicted concentrations of As, Cd, Cu, Hg, Ni, Pb and Zn PTEs based on 87 sites. The covariates were represented by remote sensing, terrain and anthropogenic variables. The results showed that the elements were characterized by high variability and their content varied between 1 and 37 mg/kg for As, 0.1 and 2.7 for Cd, 4.7 and 674 for Cu, 0.005 and 3.6 for Hg, 2.8 and 66.8 for Ni, 3.4 and 858.8 for Pb, 15.4 and 1306.5 for Zn. We found that RK was the best performing model for Cu, Hg, Ni, Pb and Zn, with the lowest mean absolute error (MAE), root mean squared error (RMSE), and highest coefficient of determination (R2) and Nash–Sutcliffe model efficiency coefficient (NSE) values. RF method was better for predicting As content, while the performance of the models for Cd was the same. Nevertheless, R2 values remained below 0.23, signifying a difficulty in effectively modeling PTEs within urbanized regions. The study provides insight into the benefits of combining regression analysis and kriging, and highlights the importance of considering multiple modeling approaches in spatial prediction of PTEs.

AB - The world's largest cities are characterized by environmental pollution, which endangers human health and biological species. Therefore, environmental monitoring and mapping of pollution levels are important tasks. To this end, we applied regression kriging (RK) and random forest (RF) for digital mapping of potentially toxic elements (PTEs) on the territory of Saint Petersburg (Russia). We predicted concentrations of As, Cd, Cu, Hg, Ni, Pb and Zn PTEs based on 87 sites. The covariates were represented by remote sensing, terrain and anthropogenic variables. The results showed that the elements were characterized by high variability and their content varied between 1 and 37 mg/kg for As, 0.1 and 2.7 for Cd, 4.7 and 674 for Cu, 0.005 and 3.6 for Hg, 2.8 and 66.8 for Ni, 3.4 and 858.8 for Pb, 15.4 and 1306.5 for Zn. We found that RK was the best performing model for Cu, Hg, Ni, Pb and Zn, with the lowest mean absolute error (MAE), root mean squared error (RMSE), and highest coefficient of determination (R2) and Nash–Sutcliffe model efficiency coefficient (NSE) values. RF method was better for predicting As content, while the performance of the models for Cd was the same. Nevertheless, R2 values remained below 0.23, signifying a difficulty in effectively modeling PTEs within urbanized regions. The study provides insight into the benefits of combining regression analysis and kriging, and highlights the importance of considering multiple modeling approaches in spatial prediction of PTEs.

KW - Digital soil mapping

KW - Heavy metals

KW - Kriging

KW - Machine learning

KW - Potentially toxic elements

UR - https://www.mendeley.com/catalogue/4a642529-fbfc-3a08-9382-963b7925aa05/

U2 - 10.1007/s12665-023-11272-9

DO - 10.1007/s12665-023-11272-9

M3 - Article

VL - 82

JO - Environmental Earth Sciences

JF - Environmental Earth Sciences

SN - 1866-6280

IS - 23

M1 - 561

ER -

ID: 114633405