Predicting Heavy Metal Adsorption on Soil with Machine Learning and Mapping Global Distribution of Soil Adsorption Capacities

Abstract

Studying heavy metal adsorption on soil is important for understanding the fate of heavy metals and properly assessing the related environmental risks. Existing experimental methods and traditional models for quantifying adsorption, however, are time-consuming and ineffective. In this study, we developed machine learning models for the soil adsorption of six heavy metals (Cd(II), Cr(VI), Cu(II), Pb(II), Ni(II), and Zn(II)) using 4420 data points (1105 soils) extracted from 150 journal articles. After a comprehensive comparison, our results showed that the gradient boosting decision tree had the best performance for a combined model based on all the data. The Shapley additive explanation method was used to identify the feature importance and the effects of these features on the adsorption, based on which six independent models were developed for the six metals to achieve better model performance than the combined model. Using these independent models, the global distribution of heavy metal adsorption capacities on soils was predicted with known soil properties. Reversed models, including one combined model for all the six metals and six independent models, were also built using the same data sets to predict the heavy metal concentration in water when the adsorbed amount is known for a soil/sediment.

Classification and Regression Machine Learning Models for Predicting Aerobic Ready and Inherent Biodegradation of Organic Chemicals in Water

“…A total of 13 common ML algorithms (Table S4) were screened during the regression model development, including adaptive boosting (AdaB), bagging, DT, a deep neural network, extra trees, gradient boosting (GradientB), KNN, Lasso, linear regression (Lr), RF, Ridge, support vector regression, and XGBoost. As there are numerous research articles, official documentations, posts, forums, and tutorials about these algorithms, we decided not to give introductions to them here. Note that all of these algorithms (as well as the chemical representations) were selected based on the literature as they were commonly used by other studies for environmental engineering applications .…”

“…The workflow of this study can be found in Figure . For those who have little ML knowledge, we recommend these reference papers, , which have step-by-step guidance for building ML models for general environmental applications.…”

Efficient Removal of Pb2+ from Water by Bamboo-Derived Thin-Walled Hollow Ellipsoidal Carbon-Based Adsorbent

“…Lead ion (Pb 2+ ) is one of the most toxic and frequently exposed heavy metals in the environment. − It can disturb the distribution and contents of trace metal elements (e.g., Fe and Zn) in the human body and destroy the enzyme system of the human body, thus causing diseases like autonomic nerve dysfunction, anemia, nausea, and dizziness. Furthermore, Pb 2+ are easily bound by albumin protein, thus leading to detrimental effects in the body .…”

“…In addition to the direct harm to the human body, Pb 2+ also pollutes the water, destroys the growth of aquatic animals and plants, shows strong accumulation in the soil, penetrates into the root systems of plants, affects the growth and development of plants, and reduces plant chlorophyll, thus hindering plant photosynthesis which in turn affects the entire ecosystem . Therefore, efficiently removing Pb 2+ from water has become a hot topic in sustainable chemistry and engineering. − …”

A Review of the Distribution Coefficient (<i>K<sub>d</sub></i>) of Some Selected Heavy Metals over the Last Decade (2012-2021)

“…The presence of heavy metals in soil and groundwater has long been recognized as a contentious issue worldwide. Because of their inherent accumulative and nondegradable properties, heavy metals in soil and groundwater pose significant environmental risks (Guo et al, 2018;Vatandoost et al, 2018;Yang et al, 2021).…”