Modeling and optimizing the removal of cadmium by Sinapis alba L. from contaminated soil via Response Surface Methodology and Artificial Neural Networks during assisted phytoremediation with sewage sludge

Phytoremediation as a viable ecological and socioeconomic management strategy

Phytoremediation is an environmentally friendly alternative to traditional remediation technologies, notably for soil restoration and agricultural sustainability. This strategy makes use of marginal areas, incorporates biofortification processes, and expands crop alternatives. The ecological and economic benefits of phytoremediation are highlighted in this review. Native plant species provide cost-effective advantages and lower risks, while using invasive species to purify pollutants might be a potential solution to the dilemma of not removing them from the new habitat. Thus, strict management measures should be used to prevent the overgrowth of invasive species. The superior advantages of phytoremediation, including psychological and social improvements, make it a powerful tool for both successful cleanup and community well-being. Its ability to generate renewable biomass and adapt to a variety of uses strengthens its position in developing the bio-based economy. However, phytoremediation faces severe difficulties such as complex site circumstances and stakeholder doubts. Overcoming these challenges necessitates a comprehensive approach that balances economic viability, environmental protection, and community welfare. Incorporating regulatory standards such as ASTM and ISO demonstrates a commitment to long-term environmental sustainability, while also providing advice for unique nation-specific requirements. Finally, phytoremediation may contribute to a pleasant coexistence of human activity and the environment by navigating hurdles and embracing innovation.

A unified framework of response surface methodology and coalescing of Firefly with random forest algorithm for enhancing nano-phytoremediation efficiency of chromium via in vitro regenerated aquatic macrophyte coontail (Ceratophyllum demersum L.)

Nano-phytoremediation is a novel green technique to remove toxic pollutants from the environment. In vitro regenerated Ceratophyllum demersum (L.) plants were exposed to different concentrations of chromium (Cr) and exposure times in the presence of titania nanoparticles (TiO2NPs). Response surface methodology was used for multiple statistical analyses like regression analysis and optimizing plots. The supplementation of NPs significantly impacted Cr in water and Cr removal (%), whereas NP × exposure time (T) statistically regulated all output parameters. The Firefly metaheuristic algorithm and the random forest (Firefly-RF) machine learning algorithms were coalesced to optimize hyperparameters, aiming to achieve the highest level of accuracy in predicted models. The R2 scores were recorded as 0.956 for Cr in water, 0.987 for Cr in the plant, 0.992 for bioconcentration factor (BCF), and 0.957 for Cr removal through the Firefly-RF model. The findings illustrated superior prediction performance from the random forest models when compared to the response surface methodology. The conclusion is drawn that metal-based nanoparticles (NPs) can effectively be utilized for nano-phytoremediation of heavy metals. This study has uncovered a promising outlook for the utilization of nanoparticles in nano-phytoremediation. This study is expected to pave the way for future research on the topic, facilitating further exploration of various nanoparticles and a thorough evaluation of their potential in aquatic ecosystems.

Artificial intelligence models for validating and predicting the impact of chemical priming of hydrogen peroxide (H2O2) and light emitting diodes on in vitro grown industrial hemp (Cannabis sativa L.)

Industrial hemp (Cannabis sativa L.) is a highly recalcitrant plant under in vitro conditions that can be overcome by employing external stimuli. Hemp seeds were primed with 2.0-3.0% hydrogen peroxide (H2O2) followed by culture under different Light Emitting Diodes (LEDs) sources. Priming seeds with 2.0% yielded relatively high germination rate, growth, and other biochemical and enzymatic activities. The LED lights exerted a variable impact on Cannabis germination and enzymatic activities. Similarly, variable responses were observed for H2O2 × Blue-LEDs combination. The results were also analyzed by multiple regression analysis, followed by an investigation of the impact of both factors by Pareto chart and normal plots. The results were optimized by contour and surface plots for all parameters. Response surface optimizer optimized 2.0% H2O2 × 918 LUX LEDs for maximum scores of all output parameters. The results were predicted by employing Multilayer Perceptron (MLP), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost) algorithms. Moreover, the validity of these models was assessed by using six different performance metrics. MLP performed better than RF and XGBoost models, considering all six-performance metrics. Despite the differences in scores, the performance indicators for all examined models were quite close to each other. It can easily be concluded that all three models are capable of predicting and validating data for cannabis seeds primed with H2O2 and grown under different LED lights.

Optimizing nutrients from fly ash-amended soil through microbial-assisted phytoremediation using response surface methodology

Nutrients are vital ingredients to boost plant health. The availability of nutrients is limited in fly ash (FA) waste to properly implement phytoremediation. The research explored the integration of microbes and treated wastewater irrigation in phytoremediation to provide the necessary nutrients for plant growth in fly ash-amended soils. The Box-Behnken method was used to design the experimental layout for the pot study. Response surface methodology (RSM) was applied as the optimization approach to model predictions for nutrient accumulation. The implemented pot study attained the highest morphological indicators with a plastochron index of 33.40, an absolute growth rate of 2.63 cm/day, and a leaf area of 2681.68 cm2 and attained maximum biomass of 24.91 g for the treatments that included a mid-range of the variables. The combination of FA 14.98%, microbial dose 4.07 mL, and treated wastewater as the irrigation source was found to be the optimized combination for nitrogen and phosphorus accumulation of 212.4 and 8.867 mg/L.

Integrating artificial intelligence, machine learning, and deep learning approaches into remediation of contaminated sites: A review

The growing number of contaminated sites across the world pose a considerable threat to the environment and human health. Remediating such sites is a cumbersome process with the complexity originating from the need for extensive sampling and testing during site characterization. Selection and design of remediation technology is further complicated by the uncertainties surrounding contaminant attributes, concentration, as well as soil and groundwater properties, which influence the remediation efficiency. Additionally, challenges emerge in identifying contamination sources and monitoring the affected area. Often, these problems are overly simplified, and the data gathered is underutilized rendering the remediation process inefficient. The potential of artificial intelligence (AI), machine-learning (ML), and deep-learning (DL) to address these issues is noteworthy, as their emergence revolutionized the process of data management/analysis. Researchers across the world are increasingly leveraging AI/ML/DL to address remediation challenges. Current study aims to perform a comprehensive literature review on the integration of AI/ML/DL tools into contaminated site remediation. A brief introduction to various emerging and existing AI/ML/DL technologies is presented, followed by a comprehensive literature review. In essence, ML/DL based predictive models can facilitate a thorough understanding of contamination patterns, reducing the need for extensive soil and groundwater sampling. Additionally, AI/ML/DL algorithms can play a pivotal role in identifying optimal remediation strategies by analyzing historical data, simulating scenarios through surrogate models, parameter-optimization using nature inspired algorithms, and enhancing decision-making with AI-based tools. Overall, with supportive measures like open-data policies and data integration, AI/ML/DL possess the potential to revolutionize the practice of contaminated site remediation.

Effect of salinity on the potential cadmium phytoremediation from the polluted soil by carpobrotus rossii

Nowadays, toxic metals accumulation in soil texture due to anthropogenic activities is a major form of pollution, which can lead to worldwide concerns; however, there are many treatment methods to remove them from soil such as phytoremediation. The carpobrotus rossii, has shown great potential to tolerate high salinity and accumulate Cd from contaminated soils. The experiments, in this study, are analyzed and optimized by Central Composite Design (CCD) as method and using Response Surface Methodology (RSM) package in R software. The Cd removal by root and the whole plant followed the quadratic model and the R² values were 94.95 and 94.81, respectively. The results showed that a decrease in NaCl concentration in Cd-containing solution can increase the phytoremediation process of Cd by carpobrotus rossii, significantly. The optimum conditions for 58% Cd removal by the whole plant, predicted through a CCD response surface methodology model were as follows: initial Cd concentration of 49 mgKg^-1,NaCl concentration of 16 dSm^-1, time of 17 days, and pH of 6.5. C. rossii's potential in removing 58% of Cd under the obtained optimum condition from the modelling was evaluated in real condition in the laboratory. The results revealed that around 56% of the initial added Cd concentration was removed by carpobrotus rossii. As a take home message, carpobrotus rossii can be recommended as an efficient plant to remove heavy metals especially cadmium from soil and sediments in arid area which have a salty soil.

Phytoremediation of nitrogen and phosphorus pollutants from glass industry effluent by using water hyacinth (Eichhornia crassipes (Mart.) Solms): Application of RSM and ANN techniques for experimental optimization

The present study aimed to assess the efficiency of the water hyacinth (Eichhornia crassipes (Mart.) Solms) plant for the reduction of nitrogen and phosphorus pollutants from glass industry effluent (GIE) as batch mode phytoremediation experiments. For this, response surface methodology (RSM) and artificial neural networks (ANN) methods were adopted to evidence the optimization and prediction performances of E. crassipes for total Kjeldahl's nitrogen (TKN) and total phosphorus (TP) removal. The control parameters, i.e., GIE concentration (0, 50, and 100%) and plant density (1, 3, and 5 numbers) were used to optimize the best reduction conditions of TKN and TP. A quadratic model of RSM and feed-forward backpropagation algorithm-based logistic model (input layer: 2 neurons, hidden layer: 10 neurons, and output layer: 1 neuron) of ANN showed good fitness results for experimental optimization. Optimization results showed that maximum reduction of TKN (93.86%) and TP (87.43%) was achieved by using 60% of GIE concentration and nearly five plants. However, coefficient of determination (R²) values showed that ANN models (TKN: 0.9980; TP: 0.9899) were superior in terms of prediction performance as compared to RSM (TKN: 0.9888; TP: 0.9868). Therefore, the findings of this study concluded that E. crassipes can be effectively used to remediate nitrogen and phosphorus loads of GIE and minimize environmental hazards caused by its unsafe disposal.

Transcriptome sequencing of Brassica napus highlights the complex issues with soil supplementation with sewage sludge

The soil supplementation with sewage sludge (SS) has become a widespread method to improve soil quality, but its long-term possible consequences are still relatively unknown. SS may contain several groups of contaminants to which the biological responses of the organisms are still poorly understood mainly due to the mixture toxicity. In this context, RNA-seq has been used to assess the impact of the exposure to sewage sludge supplemented soil at the whole-transcriptome level in the Brassica napus (B. napus). Although the municipal sewage sludge passed all safety regulations set by the EU commission (86/278/EEC), soil supplementation with SS caused a significant (p < 0.05) increase in the content of lead (by 68.8%, 71.4% in plant shoots and roots, respectively), zinc (by 22.4% and 31.2%), nickel (by 67.0% and 30.2%), and copper (by 33.1% and 39.2%). The de-novo assembled transcriptome of B. napus identified 555 differently expressed genes (DEGs) in a response to sewage sludge supplementation at the false detection rate below 0.001 (FDR <0.001). Among them, 313 genes were up-regulated and 242 genes were down-regulated. The gene ontology analysis (GO) had shown, that significantly enriched GO groups included genes involved in photosynthesis, carbohydrate metabolism and photosystems repair (41 genes), response to oxidative stress (50 genes), response to pathogens (36 genes), response to xenobiotics (15 genes), and heavy metals (41 genes), cell death (8 genes), cell wall structure (15 genes). These results suggest a significant impact of contaminants in sewage sludge on plants transcriptome. The transcriptomic approach facilitated a better understanding of the molecular level of the potential toxicity of sewage sludge in B. napus. RNA-seq allowed for the identification of potential novel early-warning molecular markers of environmental contamination. This work highlights the crucial necessity for rapid legislation change concerning the allowable levels of contaminants in sewage sludge applied on land, to mitigate the possible adverse outcomes in the ecosystem after its use as a fertilizer.

The Effects of Wastewater Treatment Plant Failure on the Gulf of Gdansk (Southern Baltic Sea)

In August 2019 and during August/September 2020, the main collection system of the Wastewater Treatment Plant (WWTP) in Warsaw, Poland, malfunctioned. During that system failure, over 4.8 million m³ of untreated wastewater was dropped directly into the Vistula River in just a few days. It is currently considered as one of the largest known failures of WWTP worldwide. In order to assess the environmental impact, water samples were collected from 2 spots at the Vistula river estuary (406 and 415 km from the discharge location, respectively), and 4 spots at the Gulf of Gdansk, situated on the southern shore of the Baltic Sea. The sampling was conducted before the wastewater wave reached the Vistula river’s mouth, followed by daily sampling during 21 days after the malfunction occurred. The study showed the decline in water quality at the Vistula river estuary and the Baltic shore waters as the wave of wastewater reached those points, despite being situated over 400 km downstream from the place of the accident. Those changes included the reduction in the dissolved oxygen content (by 0.69-fold at its peak), the increase in Total Organic Carbon (TOC) (by 1.28-fold at its peak), nitrate-nitrogen (N-NO₃) (by 1.68-fold at its peak), phosphorous (P) (by 2.41-fold at its peak), conductivity (by 16.8-fold at its peak), and Chemical Oxygen Demand (COD) (by 1.84-fold). In the samples from the Vistula river, the decline in water quality was seen as incidental and lasted 2–3 days. Subsequently, the levels of physical and chemical parameters returned to the levels from before the accident. However, the changes in the Gulf of Gdańsk lasted significantly longer, especially on the West side of the Vistula river, where, even after 21 days from the initial accident, some parameters remained altered.

Effect of alkaline treatment on the removal of contaminants of emerging concern from municipal biosolids: Modelling and optimization of process parameters using RSM and ANN coupled GA

The current study aimed in enhancing the efficiency of alkaline treatment for CECs remediation in biosolids through the application of RSM and ANN. Due to the seasonal variation of CECs in biosolids, a complete CECs profile over a period of three years were performed. Out of 64 targeted CECs, 13 PhACs (70.1 μg/kg) and 10 pesticides (57.2 μg/kg) were detected in biosolids. In order to enhance the remediation efficiency of CECs by alkaline treatment, process parameters - pH (9.0-13.0), time (3.0-12.0 h) and biosolids age (1-28 days) were optimized by statistical modelling. Using Box-Behnken design, experiments were designed and the resultant data was employed as input for model building using RSM and ANN. The developed mathematical model for alkaline treatment of biosolids using ANN predicted CECs removal with 3.2-fold lower MSE and exhibited high regression coefficient (R² > 0.99) than the conventional RSM model. Further, the multiparameter model was optimized for achieving maximum of 95.7 % CECs removal using ANN-GA. On analyzing the acute toxicity of alkaline treated residual biosolids under the optimized conditions, a reduction in LC50 by an average of 2.1-fold than initial biosolids was observed. This study not only established the application of statistical modelling in the development of an efficient remediation strategy for biosolids, which can be further applied for large-scale remediation process, but also proved the reliability and efficiency of ANN-GA.

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.

Prediction of Plant Uptake and Translocation of Engineered Metallic Nanoparticles by Machine Learning

Machine learning was applied to predict the plant uptake and transport of engineered nanoparticles (ENPs). A back propagation neural network (BPNN) was used to predict the root concentration factor (RCF) and translocation factor (TF) of ENPs from their essential physicochemical properties (e.g., composition and size) and key external factors (e.g., exposure time and plant species). The relative importance of input variables was determined by sensitivity analysis, and gene-expression programming (GEP) was used to generate predictive equations. The BPNN model satisfactorily predicted the RCF and TF in both hydroponic and soil systems, with an R² higher than 0.8 for all simulations. Inclusion of the initial ENP concentration as an input variable further improved the accuracy of the BPNN for soil systems. Sensitivity analysis indicated that the composition of ENPs (e.g., metals vs metal oxides) is a major factor affecting RCF and TF values in a hydroponic system. However, the soil organic matter and clay contents are more dominant in a soil system. The GEP model (R² = 0.8088 and 0.8959 for RCF and TF values) generated more accurate predictive equations than the conventional regression model (R² = 0.5549 and 0.6664 for RCF and TF values) in a hydroponic system, which could guide the sustainable design of ENPs for agricultural applications.