Chemical Fractions and Magnetic Simulation Based on Machine Learning for Trace Metals in a Sedimentary Column of Lake Taihu DOI Open Access
Hui Xiao, Tong Ke, Chen Li-ming

et al.

Water, Journal Year: 2024, Volume and Issue: 16(18), P. 2604 - 2604

Published: Sept. 14, 2024

In this study, the chemical fractions (CFs) of trace metal (TMs) and multiple magnetic parameters were analysed in sedimentary column from centre Lake Taihu. The column, measuring 53 cm length, was dated using 210Pb 137Cs to be 124 years old. Surface layers found contain significantly higher concentrations Cd, Co, Cu, Pb, Sb, Ti, Zn than middle bottom layers. core contained a substantial amount ferrimagnetic minerals. Most TMs present residual state, except for Mn Pb. Cd exhibited most significant variation with depth. pollution load index (PLI) indicated moderate levels region, whereas risk assessment code (RAC) classified as being heavily polluted. Multiple linear regression (MLR) random forest (RF), support vector machine (SVM), XGBoost (1.7.7.1) learning models used simulate RAC total concentration TMs, physical indicators sediments input variables. MLR model outperformed RF, SVM, simulating CFs R2 up 0.668 0.87. SHapley Additive exPlanations (SHAP) method reveals that χarm/χ is dominant factor influencing As models. For Co Cu RF models, C% N% exhibit greater contributions.

Language: Английский

Worldwide Examination of Magnetic Responses to Heavy Metal Pollution in Agricultural Soils DOI Creative Commons
Xuan-Xuan Zhao, Jiaxing Zhang, Ruijun Ma

et al.

Agriculture, Journal Year: 2024, Volume and Issue: 14(5), P. 702 - 702

Published: April 29, 2024

Over the last decade, a large number of studies have been conducted on heavy metals and magnetic susceptibility (χlf) measurement in soils. Yet, global understanding soil contamination responses remains elusive due to limited scope or sampling sites these studies. Hence, we attempted explore pollution proxy scale. Through meta-analysis data from 102 published studies, our research aimed provide worldwide overview metal agriculture We mapped geographic distribution nine (Cr, Cu, Zn, Pb, Ni, As, Cd, Mn, Fe) agricultural soils explored their sources contributions. Since 2011, The accumulation has escalated, with industrial activities (31.5%) being largest contributor, followed by inputs (27.1%), atmospheric deposition (22.66%), natural (18.74%). study reports χlf ranging 6.45 × 10−8 m3/kg 319.23 χfd 0.59% 12.85%, majority samples below 6%, indicating influence mainly human activities. Pearson’s correlation redundancy analysis show significant positive correlations Cu (r = 0.51–0.53) Mn Fe 0.50–0.53), while As were shown be key factors variation response. average load index 2.03 suggests moderate pollution, higher areas high χlf. Regression confirms is considered non-polluted 26×10−8 polluted above this threshold, all showing linear (R 0.72), that relationship between geochemical properties continues exist This provides new insights for large-scale quality assessment

Language: Английский

Citations

6

Evaluation of machine learning models for accurate prediction of heavy metals in coal mining region soils in Bangladesh DOI
Ram Proshad,

Krishno Chandra,

Maksudul Islam

et al.

Environmental Geochemistry and Health, Journal Year: 2025, Volume and Issue: 47(5)

Published: April 23, 2025

Language: Английский

Citations

0

Interpretable machine learning models reveal the partnership of microplastics and perfluoroalkyl substances in sediments at a century scale DOI
Ligang Deng, Kai Liu, Yifan Fan

et al.

Journal of Hazardous Materials, Journal Year: 2024, Volume and Issue: 486, P. 137018 - 137018

Published: Dec. 26, 2024

Language: Английский

Citations

1

Chemical Fractions and Magnetic Simulation Based on Machine Learning for Trace Metals in a Sedimentary Column of Lake Taihu DOI Open Access
Hui Xiao, Tong Ke, Chen Li-ming

et al.

Water, Journal Year: 2024, Volume and Issue: 16(18), P. 2604 - 2604

Published: Sept. 14, 2024

In this study, the chemical fractions (CFs) of trace metal (TMs) and multiple magnetic parameters were analysed in sedimentary column from centre Lake Taihu. The column, measuring 53 cm length, was dated using 210Pb 137Cs to be 124 years old. Surface layers found contain significantly higher concentrations Cd, Co, Cu, Pb, Sb, Ti, Zn than middle bottom layers. core contained a substantial amount ferrimagnetic minerals. Most TMs present residual state, except for Mn Pb. Cd exhibited most significant variation with depth. pollution load index (PLI) indicated moderate levels region, whereas risk assessment code (RAC) classified as being heavily polluted. Multiple linear regression (MLR) random forest (RF), support vector machine (SVM), XGBoost (1.7.7.1) learning models used simulate RAC total concentration TMs, physical indicators sediments input variables. MLR model outperformed RF, SVM, simulating CFs R2 up 0.668 0.87. SHapley Additive exPlanations (SHAP) method reveals that χarm/χ is dominant factor influencing As models. For Co Cu RF models, C% N% exhibit greater contributions.

Language: Английский

Citations

0