ML-based Biochar Optimization for Heavy Metal Remediation

In a recent study published in the journal Toxics, researchers explored the potential of machine learning (ML) models to predict biochar's effectiveness in remediating heavy metal (HM) contamination in soil-plant systems. They focused on understanding the complex interactions between biochar, soil, and plants, and how these affect HM immobilization and crop uptake.

Pearson’s correlation matrix of influencing factors. (X1: experiment scale, X2: type of crops, X3: application rate (%), X4: feedstock, X5: pyrolysis temperature, X6: clay content, X7: silt content, X8: sand content, X9: soil pH, X10: soil organic carbon, X11: crop duration, X12: total heavy metals in soil, X13: available heavy metals in soil, X14: type of heavy metals, and Y: immobilization efficiency). Image Credit: https://www.mdpi.com/2305-6304/12/8/575
Pearson’s correlation matrix of influencing factors. (X1: experiment scale, X2: type of crops, X3: application rate (%), X4: feedstock, X5: pyrolysis temperature, X6: clay content, X7: silt content, X8: sand content, X9: soil pH, X10: soil organic carbon, X11: crop duration, X12: total heavy metals in soil, X13: available heavy metals in soil, X14: type of heavy metals, and Y: immobilization efficiency). Image Credit: https://www.mdpi.com/2305-6304/12/8/575

Background

HM soil contamination is a global environmental concern, posing health risks throughout the food chain. Excessive HM levels in soils can lead to health problems, such as lung cancer, bone fractures, and kidney dysfunction. Reducing the bioavailability of HMs in soil-plant environments is the most effective way to mitigate these risks. In situ remediation techniques aim to immobilize HMs and prevent their absorption by crops.

Biochar has gained attention in recent years as a low-cost material for HM remediation. It is a carbon-rich product produced by heating biomass under limited oxygen. Biochar not only enhances soil nutrients and crop production but also reduces HM bioavailability by adsorbing them onto its surface.

About the Research

In this paper, the authors aimed to develop and evaluate ML models for predicting the immobilization efficiency of biochar-HM amendments in soil-plant systems. The analysis included 211 experimental datasets collected from pot and field studies, considering 14 influencing factors. These factors were categorized into three groups: soil characteristics (clay content, potential of hydrogen ion (pH), organic carbon, HM concentration, silt content, sand content, and soil organic carbon), biochar characteristics (application rate, feedstock, pyrolysis temperature), and crop characteristics (crop type, duration, and experiment scale).

The study employed four different ML models: partial least squares (PLS), linear regression (LR), random forest (RF), and support vector regression (SVR). These models were trained and tested using the collected dataset to predict the immobilization capacity of biochar-HM amendments. Furthermore, the researchers evaluated the relative importance of each influencing factor in determining the immobilization efficiency.

Research Findings

The outcomes showed that the RF model outperformed the other models in predicting the immobilization efficiency of HMs by biochar, with a coefficient of determination (R²) value of 0.5924. The feature importance analysis revealed that soil characteristics accounted for 79.7% of the influence on HM absorption by crops, followed by crop properties at 3.2% and biochar properties at 17.1%. The key factors that influenced the immobilization efficiency were the availability of HMs in the soil, the type of HMs, the soil clay content, and the biochar application rate and pyrolysis temperature.

The results also indicated that the application rate of biochar had a positive correlation with immobilization efficiency and that the pyrolysis temperature of biochar contributed significantly to its immobilization efficiency. Additionally, the study found that the species of HMs in the soil, such as lead (Pb), zinc (Zn), copper (Cu), and cadmium (Cd), affected the immobilization efficiency of biochar.

Furthermore, the authors developed the RF model to predict the bioaccumulation factors (BAFs) and changes in crop uptake (CCU) of HMs after biochar amendment. The model achieved R² values of 0.7338 and 0.6997 for BAFs and CCU, respectively, indicating its effectiveness in predicting HM bioavailability in soil-crop systems.

Applications

Predicting biochar's immobilization efficiency using ML has practical implications. By identifying key factors that affect biochar's performance, researchers and practitioners can optimize biochar production and application to improve HM remediation. The study highlights the importance of considering soil properties when designing remediation strategies, and facilitating site-specific approaches tailored to individual soils. Additionally, by predicting bioaccumulation and crop uptake of HMs, contamination risks can be assessed, helping farmers minimize HM absorption by crops.

The study also suggests that biochar can be optimized by adjusting the application rate and pyrolysis temperature to achieve maximum immobilization efficiency. Furthermore, the results indicate that the type of crop and its duration do not significantly affect HM immobilization efficiency, suggesting that biochar can be used as a general remediation strategy for various crops.

Conclusion

In summary, ML models proved effective for predicting the immobilization efficiency and bioavailability of HMs in biochar-amended soil-plant environments. The superior performance of the RF model demonstrated its ability to capture the complex non-linear relationships between various influencing factors and the HM immobilization process.

The researchers identified soil properties, such as HM availability and clay content, and biochar characteristics, including application rate and pyrolysis temperature, as key determinants of biochar's effectiveness in remediating HM-contaminated soils. Moving forward, they recommended optimizing biochar application factors and enhancing the models' predictive accuracy. Overall, their study marked an important advancement in using data-driven methods to understand and manage HM contamination in agricultural systems.

Journal reference:
  • Li, X.; & et, al. Predictive Machine Learning Model to Assess the Adsorption Efficiency of Biochar-Heavy Metals for Effective Remediation of Soil-Plant Environment. Toxics 2024, 12, 575. DOI: 10.3390/toxics12080575, https://www.mdpi.com/2305-6304/12/8/575
Muhammad Osama

Written by

Muhammad Osama

Muhammad Osama is a full-time data analytics consultant and freelance technical writer based in Delhi, India. He specializes in transforming complex technical concepts into accessible content. He has a Bachelor of Technology in Mechanical Engineering with specialization in AI & Robotics from Galgotias University, India, and he has extensive experience in technical content writing, data science and analytics, and artificial intelligence.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Osama, Muhammad. (2024, August 18). ML-based Biochar Optimization for Heavy Metal Remediation. AZoAi. Retrieved on October 08, 2024 from https://www.azoai.com/news/20240818/ML-based-Biochar-Optimization-for-Heavy-Metal-Remediation.aspx.

  • MLA

    Osama, Muhammad. "ML-based Biochar Optimization for Heavy Metal Remediation". AZoAi. 08 October 2024. <https://www.azoai.com/news/20240818/ML-based-Biochar-Optimization-for-Heavy-Metal-Remediation.aspx>.

  • Chicago

    Osama, Muhammad. "ML-based Biochar Optimization for Heavy Metal Remediation". AZoAi. https://www.azoai.com/news/20240818/ML-based-Biochar-Optimization-for-Heavy-Metal-Remediation.aspx. (accessed October 08, 2024).

  • Harvard

    Osama, Muhammad. 2024. ML-based Biochar Optimization for Heavy Metal Remediation. AZoAi, viewed 08 October 2024, https://www.azoai.com/news/20240818/ML-based-Biochar-Optimization-for-Heavy-Metal-Remediation.aspx.

Comments

The opinions expressed here are the views of the writer and do not necessarily reflect the views and opinions of AZoAi.
Post a new comment
Post

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.

You might also like...
Machine Learning Enhances Water Quality Monitoring