Leaf magnetic properties as a method for predicting heavy metal concentrations in PM2.5 using support vector machine: A case study in Nanjing, China

Evaluation of atmospheric particulate matter pollution characteristics in Shanghai based on biomagnetic monitoring technology

Atmospheric particulate matter (PM) pollution is one of the world's most serious environmental challenges, and it poses a significant threat to environmental quality and human health. Biomagnetic monitoring of PM has great potential to improve spatial resolution and provide alternative indicators for large area measurements, with respect and complementary to standard air quality monitoring stations. In this study, 160 samples of evergreen plant leaves were collected from park green spaces within five different functional areas of Shanghai. Magnetic properties were investigated to understand the extent and nature of particulate pollution and the possible sources, and to assess the suitability of various plant leaves for urban particulate pollution monitoring. The results showed that magnetic particles of the plant leaf-adherent PM were predominantly composed of pseudo-single domain (PSD) and multi-domain (MD) ferrimagnetic particles. Magnolia grandiflora, as a large evergreen arbor with robust PM retention capabilities, proved to be a more suitable candidate for monitoring urban particulate pollution compared to Osmanthus fragrans, a small evergreen arbor, and Aucuba japonica Thunb. var. variegata and Photinia serratifolia, evergreen shrubs. Meanwhile, there were significant differences in the spatial distribution of the magnetic particle content and heavy metal enrichment of the samples, mainly showing regional variations of industrial area > traffic area > commercial area > residential area > clean area. Additionally, the combination with the results of scanning electron microscopy, shows that industrial production (metal smelting, coal burning), transport and other activities are the main sources of particulate pollution. Plant leaves can be used as an effective tool for urban particulate pollution monitoring and assessment of atmospheric particulate pollution characteristics, and the technique provided useful information on particle size, mineralogy and possible sources.

Reconstruction of a century of air pollution history in Nanjing, China, using trace elements in situ leaf specimens of Platanus × hispanica and Pittosporum tobira

Leaves can specifically uptake trace elements from the surrounding environment. And tree leaves are a good biological indicator for air pollution. Therefore, chemical analysis of leaf specifications can be used to reproduce a historical record of air pollution. To better understand the history of urban air pollution from the 1920s to the 2020s in Nanjing, China, leaf samples of two woody plants, Platanus × hispanica and Pittosporum tobira, were collected in this study as environmental indicators from different historical periods. These included historical herbarium specimens and current leaves from live trees. The concentrations of 10 trace elements were determined in the samples using ICP‒MS. Pollution indices were calculated, yielding the key findings. The historical leaf samples showed continuously increasing mean concentrations of the 10 trace elements over time, which significantly correlating with automobile quantities and the number of large-scale industrial enterprises (p < 0.05). Moreover, modern leaf trace element concentrations were significantly correlated with PM10, PM2.5, automobiles, large-scale industrial enterprises, and atmospheric factors, confirming these as sources. In addition to the historical growth trend, spatial heterogeneity was revealed in historical Platanus × hispanica leaf samples from the 14 sites in Nanjing. Changes in heavy metal trace element pollution distributions were consistent with transportation and industrial expansion, with homologous patterns across elements. Specifically, post 1980s increases were observed in the representative NJ2 (Zhongshan Botanical Garden) and the NJ5(Nanjing University) sites, with higher concentrations occurring at in the NJ5 contaminated site than at the NJ2 uncontaminated site. After 2009, the 10 element (except Cd) pollution indices in Platanus × hispanica leaves fluctuated but declined overall. This reconstruction of Nanjing's air pollution history demonstrates that ample environmental information can be extracted from plant leaf markers over time and space.

Pine needles as bioindicator and biomagnetic indicator of selected metals in the street dust, a case study from southeastern Iran

Among the different approaches currently being used to evaluate the contamination level of street dust, the magnetic susceptibility of dust and urban tree leaves has received little attention. The key objectives of this study were: (i) to investigate the feasibility of using pine needles as a bioindicator and biomagnetic indicator for estimating the concentration of selected metals in street dust, and (ii) to predict the contamination level of street dust by selected metals using magnetic susceptibility. Street dust and pine tree needle samples were taken from 60 locations in three adjacent cities in Kerman province (Kerman, Rafsanjan, and Sirjan), southeastern Iran. The total concentrations of selected metals, including Cu, Zn, Fe, Mn, Ni, and Pb, and the magnetic susceptibility (χlf and χhf) values of both pine tree needles and street dust samples were determined. Among the three cities studied, samples from Kerman showed the highest magnetic susceptibility and metal concentration values. This could be attributed to the larger size and much higher population density of this city, with more industrial activities and urban traffic than the other two cities investigated. The results also showed that the concentrations of metals in pine needles were strongly correlated (p < 0.01) with those in street dust. The magnetic susceptibility of pine needles and the concentrations of Fe, Pb, Zn, Cu, Ni, and Mn in street dust showed a statistically significant correlation (p < 0.01). A strong and statistically significant correlation (p < 0.01) was also found between magnetic susceptibility and the concentration of metals in pine needles. In conclusion, strong relationships between magnetic properties and metal concentrations of pine needles with those of street dust samples seem to make pine needles a good bioindicator and biomagnetic estimator of the contamination level of metals in street dust.

Assessing heavy metal pollution load index (PLI) in biomonitors and road dust from vehicular emission by magnetic properties modeling

Vehicular traffic occupies a significant place among the sources of air pollution, due to population and urban growth that has led to an excessive increase in the vehicle fleet worldwide, and in Costa Rica as well. Vehicle emissions generate greenhouse gases (GHGs), particulate matter (PM), and heavy metals (HMs), due to combustion products from fossil-fuel engines, tire wear, and brake linings. HMs are important because they cannot be degraded or destroyed naturally; however, they can be diluted by physicochemical agents and be incorporated into trophic chains where they can be bioaccumulated causing significant negative effects on human well-being and ecological quality. This study aimed to assess the HM pollution load in biomonitors and road dust from vehicular emissions by chemical analyses and magnetic properties modeling. For this purpose, chemical and magnetic property analyses were carried out on samples of road dust and leaves of Cupressus lusitanica Mill. and Casuarina equisetifolia L., which were sampled during 2 different years in the Greater Metropolitan Area of Costa Rica known as GAM. Contamination factor (CF) and pollution load index (PLI) results showed significant metal pollution in some of the study sites. Contamination by the metals V, Cr, and Zn was most commonly present in the biomonitors, and for road dust, they were Cr, Zn, and Pb. The PLI estimates obtained with the validated support vector machine (SVM) magnetic properties models were consistent (sensitivity, specificity, and precision) with those obtained by chemical analysis, demonstrating the feasibility of this method for the identification of this index of contamination.

Estimation of the soil arsenic concentration using a geographically weighted XGBoost model based on hyperspectral data

Considering the high toxicity of arsenic (As), its contamination of soil represents an alarming environmental and public health issue. Existing soil heavy metal concentration estimation models based on hyperspectral data ignore the spatial nonstationarity of the relationship between the soil spectrum and heavy metal concentration. A novel model (geographically weighted eXtreme gradient boosting or GW-XGBoost model) combining geographically weighted regression (GWR) method with XGBoost algorithm was proposed. The northeast district of Beijing, China, was chosen as a case study area to assess the effectiveness of the proposed model. The GW-XGBoost model was established to estimate the As concentration based on the typical spectrum of As and the spatial correlation between the spectrum and As concentration obtained using the GWR method, and the result was compared to that obtained with the XGBoost and GWR models. The accuracy of the GW-XGBoost model was obviously better than that of the other models (R²_GW-XGBoost = 0.90, R²_XGBoost = 0.48, and R²_GWR = 0.74). Therefore, the proposed model is reliable, as it considers the spatial correlation between the spectrum and As concentration.

Assessing magnetic properties of biomonitors and road dust as a screening method for air pollution monitoring

Particulate matter (PM) pollution is one of the world's most serious environmental challenges. Among PM components, atmospheric heavy metals (HMs) are considered one of the main pollutants responsible for causing significant negative impacts on human health, and ecological quality. This study aimed to assess environmental magnetism as a simple and rapid method that can be used to evaluate heavy metal contamination in urban areas from the relationships between magnetic properties and heavy metal concentrations. For this purpose, road dust and leaf samples of two common evergreen species (Cupressus lusitanica/Casuarina equisetifolia) were sampled simultaneously for 2 years at sites with different levels of traffic pollution. The results found significant statistical correlations between the magnetic properties and the chemical substances of the plants studied, as Fe, Cr and V showed an r ≥ 0.9 and Cr and Zn r ≥ 0.7 with χlf in C. equisetifolia. The frequency-dependent magnetic susceptibility was found to be between 0% and 14% for plants, and 0% and 2% for road dust, suggesting a rather dissimilar particle size distribution for plants, and a less important contribution from the more hazardous ultrafine superparamagnetic magnetite for both. Confirming that magnetic analyses can be used to distinguish different degrees of urban air pollution.

Comparison between machine linear regression (MLR) and support vector machine (SVM) as model generators for heavy metal assessment captured in biomonitors and road dust

Exposure to suspended particulate matter (PM), found in the air, is one of the most acute environmental problems that affect the health of modern society. Among the different airborne pollutants, heavy metals (HMs) are particularly relevant because they are bioaccumulated, impairing the functions of living beings. This study aimed to establish a method to predict heavy metal concentrations in leaves and road dust, through their magnetic properties measurements. For this purpose, machine learning, automatic linear regression (MLR), and support vector machine (SVM) were used to establish models for the prediction of airborne heavy metals based on leaves and road dust magnetic properties. Road dust samples and leaves of two common evergreen species (Cupressus lusitanica/Casuarina equisetifolia) were sampled simultaneously during two different years in the Great Metropolitan Area (GMA) of Costa Rica. MLR and SVM algorithms were used to establish the relationship between airborne heavy metal concentrations based on single (χlf) and multiple (χlf y χdf) leaf magnetic properties and road dust. Results showed that Fe, Cu, Cr, V, and Zn concentrations were well-simulated by SVM prediction models, with adjusted R² values ≥ 0.7 in both training and test stages. By contrast, the concentrations of Pb and Ni were not well-simulated, with adjusted R² values < 0.7 in both training and test stages. Heavy metal predicción models using magnetic properties of leaves from Casuarina equisetifolia, as collectors, yielded better prediction results than those based on the leaves of Cupressus lusitanica and road dust, showing relatively higher adjusted R² values and lower errors (MAE and RMSE) in both training and test stages. SVM proved to be the best prediction model with variations between single (χlf) and multiple (χlf y χdf) magnetic properties depending on the element studied.

A review on magnetic sensors for monitoring of hazardous pollutants in water resources

Water resources have long been of interest to humans and have become a serious issue in all aspects of human life. The disposal of hazardous pollutants in water resources is one of the biggest global concerns and poses many risks to human health and aquatic life. Therefore, the control of hazardous pollutants in water resources plays an important role, when it comes to evaluating water quality. Due to low toxicity, good electrical conductivity, facile functionalization, and easy preparation, magnetic materials have become a good alternative in recent years to control hazardous pollutants in water resources. In the present study, the idea of using magnetic sensors in controlling and monitoring of pharmaceuticals, pesticides, heavy metals, and organic pollutants have been reviewed. The water pollutants in drinking water, groundwater, surface water, and seawater have been discussed. The toxicology of water hazardous pollutants has also been reviewed. Then, the magnetic materials were discussed as sensors for controlling and monitoring pollutants. Finally, future remarks and perspectives on magnetic nanosensors for controlling hazardous pollutants in water resources and environmental applications were explained.

The state of art on the prediction of efficiency and modeling of the processes of pollutants removal based on machine learning

During the last few years, important advances have been made in big data exploration, complex pattern recognition and prediction of complex variables. Machine learning (ML) algorithms can efficiently analyze voluminous data, identify complex patterns and extract conclusions. In chemical engineering, the application of machine learning approaches has become highly attractive due to the growing complexity of this field. Machine learning allows computers to solve problems by learning from large data sets and provides researchers with an excellent opportunity to enhance the quality of predictions for the output variables of a chemical process. Its performance has been increasingly exploited to overcome a wide range of challenges in chemistry and chemical engineering, including improving computational chemistry, planning materials synthesis and modeling pollutant removal processes. In this review, we introduce this discipline in terms of its accessible to chemistry and highlight studies that illustrate in-depth the exploitation of machine learning. The main aim of the review paper is to answer these questions by analyzing physicochemical processes that exploit machine learning in organic and inorganic pollutants removal. In general, the purpose of this review is both to provide a summary of research related to the removal of various contaminants performed by ML models and to present future research needs in ML for contaminant removal.

Estimating PM2.5 concentration using the machine learning GA-SVM method to improve the land use regression model in Shaanxi, China

With rapid economic growth, urbanization and industrialization, fine particulate matter with aerodynamic diameters ≤ 2.5 µm (PM_2.5) has become a major pollutant and shows adverse effects on both human health and the atmospheric environment. Many studies on estimating PM_2.5 concentrations have been performed using statistical regression models and satellite remote sensing. However, the accuracy of PM_2.5 concentration estimates is limited by traditional regression models; machine learning methods have high predictive power, but fewer studies have been performed on the complementary advantages of different approaches. This study estimates PM_2.5 concentrations from satellite remote sensing-derived aerosol optical depth (AOD) products, meteorological data, terrain data and other predictors in 2015 in Shaanxi, China, using a combined genetic algorithm-support vector machine (GA-SVM) method, after which the spatial clustering pattern was explored at the season and year levels. The results indicated that temperature (r = -0.684), precipitation (r = -0.602) and normalized difference vegetation index (NDVI) (r = -0.523) were significantly negatively correlated with the PM_2.5 concentration, while AOD (r = 0.337) was significantly positively correlated with the PM_2.5 concentration. Compared to conventional land use regression (LUR) and SVM models and previous related studies, the GA-SVM method demonstrated a significantly better prediction accuracy of PM_2.5 concentration, with a higher 10-fold cross-validation coefficient of determination (R²) of 0.84 and lower root mean square error (RMSE) and mean absolute error (MAE) of 12.1 μg/m³ and 10.07 μg/m³, respectively. Y-scrambling test shows that the models have no chance correlation. The central and southern parts of Shaanxi have high PM_2.5 concentrations, which are mainly due to the pollutant emissions and meteorological and topographical conditions in those areas. There was a positive spatial agglomeration characteristic of regional PM_2.5 pollution, and the spatial spillover effect of PM_2.5 pollution for seasonal and annual variations does exist. In general, the GA-SVM method is robust and accurately estimates PM_2.5 concentrations via a novel modeling framework application and high-quality spatiotemporal information. It also has great significance for the exploration of PM_2.5 pollution estimation and high-precision mapping methods, especially early warning in high-risk areas. Finally, the prevention and control of atmospheric pollution should take pollution control measures from major cities and surrounding cities, and focus on the joint pollution control measures for plain cities.

Rapid monitoring of heavy metal pollution in lake water using nitrogen and phosphorus nutrients and physicochemical indicators by support vector machine

A novel method of predicting heavy metal concentration in lake water by support vector machine (SVM) model was developed, combined with low-cost, easy to obtain nutrients and physicochemical indicators as input variables. 115 surface water samples were collected from 23 sites in Chaohu Lake, China, during different hydrological periods. The particulate concentrations of heavy metals in water were much higher than the dissolved concentrations. According to Nemerow pollution index (Pi), pollution degrees by Fe, V, Mn and As ranged from heavy (2 ≤ Pi < 4) to serious (Pi ≥ 4). The concentrations of most heavy metals were the highest during the medium-water period and the lowest during the dry season. Non-metric Multidimensional Scaling Analysis confirmed heavy metal concentrations had slight spatial difference but relatively large seasonal variation. Redundancy Analysis indicated the close associations of heavy metals with nutrient and physicochemical indicators. When both nutrient and physicochemical indicators were used as input variables, the simulation effects for most elements in total and particulate were relatively better than those obtained using only nutrient or only physicochemical indicators. The simulation effects for As, Ba, Fe, Ti, V and Zn were generally good, based on their training R values of 0.847, 0.828, 0.856, 0.867, 0.817 and 0.893, respectively, as well as their test R values of 0.811, 0.836, 0.843, 0.873, 0.829 and 0.826, respectively; and meanwhile, in both the training and test stages, these metals also had relatively lower errors. The spatial distribution of heavy metals in Chaohu Lake was then predicted using the fully trained SVM models.

Tracking Airborne Pollution with Environmental Magnetism in A Medium-Sized African City

As in other parts of the world, air pollution over West and Central Africa has major health and meteorological impacts. Air quality assessment and its possible sanitary impact have become essential even in medium-sized towns, therefore amplifying the need for easy-to-implement monitoring methods with low environmental impact. We present here the potential of magnetic methods to monitor air quality at street level in the medium-sized city of Maroua (northern Cameroon) affected by dust-laden desert winds. More than five hundred (544) samples of bark and leaves taken from Neem trees in Maroua were analyzed. Magnetic susceptibility, saturation remanence, and S-ratio were found to determine the concentration and nature of magnetic particles. They are dominated by magnetite-like particle signals as a part of particulate emissions due to urban activities, including both traffic, composed of a substantial proportion of motorcycles, and wood burning for food preparation. We show that both bark and leaves from Neem trees are adequate passive bio-recorders. The use of both enables different times and heights to be sampled, allowing for the high-resolution monitoring, in terms of spatialization, of various urban environments. Particle emissions require assessment and screening that could be carried out rapidly and efficiently by magnetic methods on bio-recorders, even in cities impacted by dust-laden wind.

Magnetic properties and its application in the prediction of potentially toxic elements in aquatic products by machine learning

Magnetic measurement was provided to substitute for time-consuming conventional methods for determination of potentially toxic elements. Both the concentrations of 12 elements and 9 magnetic parameters were determined in 700 muscle tissue samples from the snail Bellamya aeruginosa, shrimp species Exopalaemon modestus and Macrobrachium nipponense, and fish species Hemisalanx prognathous Regan, Coilia ectenes taihuensis, and Culer alburnus Basilewsky collected from Chaohu Lake during different hydrological periods. Spherical and irregular iron oxide particles were observed in the muscle tissues of the studied aquatic products. A field survey of the exposure parameters in humans, such as per capita intake dose of local aquatic products, found no evidence that consumption of the tested species poses a potential health risk. Redundancy analysis revealed different degrees of correlation between the magnetic parameters and concentrations of elements in aquatic products. Back-propagation artificial neural network (BP-ANN) and support vector machine (SVM) models were applied to predict elemental concentrations in aquatic products, using magnetic parameters as input. SVM models performed well in predicting the presence of Cr and Ni, with R and index of agreement values of >0.8 in both training and validation stages as well as relatively low errors. The BP-ANN and SVM models both performed relatively poorly in predicting the presence of Cd and Zn in aquatic products, with R values between 0.333 and 0.718 for Cd and between 0.454 and 0.664 for Zn in training and validation stages. For most of the elements, a better R value was obtained with the SVM than with BP-ANN model. The R of Co, Cr, Cu, Ni, and Ti in the training and validation stages of snail in the SVM model were >0.8. This study is a first step in developing a novel approach allowing the rapid monitoring of potentially toxic elements concentrations in aquatic products.

Rapid diagnosis of heavy metal pollution in lake sediments based on environmental magnetism and machine learning

Environmental magnetism in combination with machine learning can be used to monitor heavy metal pollution in sediments. Magnetic parameters and heavy metal concentrations of sediments from Chaohu Lake (China) were analyzed. The magnetic measurements, high- and low-temperature curves, and hysteresis loops showed the primary magnetic minerals were ferrimagnetic minerals in sediments. For most metals, their concentrations were highest during the wet season and lowest during the medium-water period. Cd, Hg, and Zn were moderately enriched and Cd and Hg posed a considerable ecological risk. A redundancy analysis indicated a relationship between physicochemical indexes and magnetic parameters and heavy metal concentrations. An artificial neural network (ANN) and support vector machine (SVM) were used to construct six models to predict the heavy metal concentrations and ecological risk index. The inclusion of both the physicochemical indexes and magnetic parameters as input factors in the models were significantly ameliorated the simulation accuracy for the majority of heavy metals. The training and test R, for Be, Fe, Pb, Zn, As, Cu, and Cr were > 0.8. The SVM showed better performance and hence it has potential for the efficient and economical long-term tracking and monitoring of heavy metal pollution in lake sediments.

An insight into machine learning models era in simulating soil, water bodies and adsorption heavy metals: Review, challenges and solutions

The development of computer aid models for heavy metals (HMs) simulation has been remarkably advanced over the past two decades. Several machine learning (ML) models have been developed for modeling HMs over the past two decades with outstanding progress. Although there have been a noticeable number of diverse ML models investigations, it is essential to have an informative vision on the progression of those computer aid models. In the current short review covering the simulation of heavy metals in contaminated soil, water bodies and removal from aqueous solution, numerous aspects on the methodological and conceptual HMs modeling are reviewed and discussed in detail. For instance, the limitation of the classical analytical methods, types of heavy metal dataset, necessity for new versions of ML models exploration, HM input parameters selection, ML models internal parameters tuning, performance metrics selection and the types of the modelled HM. The current review provides few outlooks in understanding the underlying od the ML models application for HM simulation. Tackling these modeling aspects is significantly essential for ML developers and environmental scientists to obtain creditability and scientific consistency in the domain of environmental science. Based on the discussed modeling aspects, it was concluded several future research directions, which will promote environmental scientists for better understanding of the underlying HMs simulation.

Statistical Approaches for Forecasting Primary Air Pollutants: A Review

Air pollutant forecasting can be used to quantitatively estimate pollutant reduction trends. Combining bibliometrics with the evolutionary tree and Markov chain methods can achieve a superior quantitative analysis of research hotspots and trends. In this work, we adopted a bibliometric method to review the research status of statistical prediction methods for air pollution, used evolutionary trees to analyze the development trend of such research, and applied the Markov chain to predict future research trends for major air pollutants. The results indicate that papers mainly focused on the effects of air pollution on human diseases, urban pollution exposure models, and land use regression (LUR) methods. Particulate matter (PM), nitrogen oxides (NOx), and ozone (O3) were the most investigated pollutants. Artificial neural network (ANN) methods were preferred in studies of PM and O3, while LUR were more widely used in studies of NOx. Additionally, multi-method hybrid techniques gradually became the most widely used approach between 2010 and 2018. In the future, the statistical prediction of air pollution is expected to be based on a mixed method to simultaneously predict multiple pollutants, and the interaction between pollutants will be the most challenging aspect of research on air pollution prediction. The research results summarized in this paper provide technical support for the accurate prediction of atmospheric pollution and the emergency management of regional air quality.

Prediction of sediment heavy metal at the Australian Bays using newly developed hybrid artificial intelligence models

Hybrid artificial intelligence (AI) models are developed for sediment lead (Pb) prediction in two Bays (i.e., Bramble (BB) and Deception (DB)) stations, Australia. A feature selection (FS) algorithm called extreme gradient boosting (XGBoost) is proposed to abstract the correlated input parameters for the Pb prediction and validated against principal component of analysis (PCA), recursive feature elimination (RFE), and the genetic algorithm (GA). XGBoost model is applied using a grid search strategy (Grid-XGBoost) for predicting Pb and validated against the commonly used AI models, artificial neural network (ANN) and support vector machine (SVM). The input parameter selection approaches redimensioned the 21 parameters into 9-5 parameters without losing their learned information over the models' training phase. At the BB station, the mean absolute percentage error (MAPE) values (0.06, 0.32, 0.34, and 0.33) were achieved for the XGBoost-SVM, XGBoost-ANN, XGBoost-Grid-XGBoost, and Grid-XGBoost models, respectively. At the DB station, the lowest MAPE values, 0.25 and 0.24, were attained for the XGBoost-Grid-XGBoost and Grid-XGBoost models, respectively. Overall, the proposed hybrid AI models provided a reliable and robust computer aid technology for sediment Pb prediction that contribute to the best knowledge of environmental pollution monitoring and assessment.

Manganese (Mn) removal prediction using extreme gradient model

Exploring the Manganese (Mn) removal prediction with several independent variables is tremendously critical and indispensable to understand the pattern of removal process. Mn is one of the key heavy metals (HMs) stipulated by the WHO for the development of many attributes of the ecosystem in controlled quantity. In the present paper, an extreme gradient model (XGBoost) is proposed for Mn prediction. A compressive statistical analysis reveals the stochastics behaviour of the data prior to the prediction investigation. The main goal is to determine the Mn predictability of XGBoost algorithm with influencing factors such as D₂EHPA (M), Time (min), H₂SO₄ (M), NaCl (g/L), and EDTA (mM). The PCA biplot signifies the importance of the predictors. The XGBoost model validated against a diversity of data-driven models such as multilinear regression (MLR), support vector machine (SVM), and random forest (RF). The order of the applied models' performance are XGBoost > RF > SVM > MLR as per their R² and RMSE metrics over testing phase i.e. 20.88, 0.75, 0.61, 0.40, and 2.23, 3.01, 3.51, 6.38, respectively. Moreover, the Taylor diagram and Radar chart have drown to emphasize the XGBoost model efficiency, stability, and reliability. In respect of XGBoost model prediction, 'Time' predictor outperforms D₂EHPA, EDTA, H₂SO₄, and NaCl predictors in order.

Biomagnetic monitoring combined with support vector machine: a new opportunity for predicting particle-bound-heavy metals

Biomagnetic monitoring includes fast and simple methods to estimate airborne heavy metals. Leaves of Osmanthus fragrans Lour and Ligustrum lucidum Ait were collected simultaneously with PM₁₀ from a mega-city of China during one year. Magnetic properties of leaves and metal concentrations in PM₁₀ were analyzed. Metal concentrations were estimated using leaf magnetic properties and meteorological factors as input variables in support vector machine (SVM) models. The mean concentrations of many metals were highest in winter and lowest in summer. Hazard index for potentially toxic metals was 5.77, a level considered unsafe. The combined carcinogenic risk was higher than precautionary value (10^-4). Ferrimagnetic minerals were dominant magnetic minerals in leaves. Principal component analysis indicated iron & steel industry and soil dust were the common sources for many metals and magnetic minerals on leaves. However, the poor simulation results obtained with multiple linear regression confirmed strong nonlinear relationships between metal concentrations and leaf magnetic properties. SVM models including leaf magnetic variables as inputs yielded better simulation results for all elements. Simulations were promising for Ti, Cd and Zn, whereas relatively poor for Ni. Our study demonstrates the feasibility of prediction of airborne heavy metals based on biomagnetic monitoring of tree leaves.

A Comparative Study of Different Machine Learning Algorithms in Predicting the Content of Ilmenite in Titanium Placer

In this study, the ilmenite content in beach placer sand was estimated using seven soft computing techniques, namely random forest (RF), artificial neural network (ANN), k-nearest neighbors (kNN), cubist, support vector machine (SVM), stochastic gradient boosting (SGB), and classification and regression tree (CART). The 405 beach placer borehole samples were collected from Southern Suoi Nhum deposit, Binh Thuan province, Vietnam, to test the feasibility of these soft computing techniques in estimating ilmenite content. Heavy mineral analysis indicated that valuable minerals in the placer sand are zircon, ilmenite, leucoxene, rutile, anatase, and monazite. In this study, five materials, namely rutile, anatase, leucoxene, zircon, and monazite, were used as the input variables to estimate ilmenite content based on the above mentioned soft computing models. Of the whole dataset, 325 samples were used to build the regarded soft computing models; 80 remaining samples were used for the models’ verification. Root-mean-squared error (RMSE), determination coefficient (R2), a simple ranking method, and residuals analysis technique were used as the statistical criteria for assessing the model performances. The numerical experiments revealed that soft computing techniques are capable of estimating the content of ilmenite with high accuracy. The residuals analysis also indicated that the SGB model was the most suitable for determining the ilmenite content in the context of this research.

Tracking the magnetic carriers of heavy metals in contaminated soils based on X-ray microprobe techniques and wavelet transformation

Technogenic magnetic particles (TMPs) from industrial activities are major contamination sources of soils and dusts because they usually carry large amounts of heavy metals. The understanding of the association between TMPs and heavy metals in contaminated soils helps to trace the polluting sources and probing into the mechanism of magnetic phases enriched with heavy metals. In this study, we tracked the magnetic carries of heavy metals from different emission sources in steel industrial regions by using the synchrotron-based probe techniques and multiscale analytical methods. The μ-XRF mapping showed that TMPs contained various heavy metals, depending on their sources. The Fe K-edge μ-XANES revealed that the ferroalloy, pyrrhotite and TMPs in steel slag and coal ash were major magnetic phases in contaminated soils. Their relative content varied differently at the microscale. The multiscale analysis revealed that the heavy metals associated with magnetic phases exhibited pronounced scale dependence, depending on the size, type, and assemblage of different magnetic phases. Multiscale source apportionment revealed that the contamination sources varied differently at multiple scales. Heatmap analysis revealed that at 8-μm scale, Co, Cr, Cu and Mn were mainly derived from ferroalloy, while Ti, Zn and As from both ferroalloy and TMPs from coal ash.