Related Experiment Video
Updated: May 1, 2026
Physical, Chemical and Biological Characterization of Six Biochars Produced for the Remediation of Contaminated Sites
Published on: November 28, 2014
Tree-structured parzen estimator optimized-automated machine learning assisted by meta-analysis for predicting
Bi-Ni Jiang1, Ying-Ying Zhang2, Zhi-Yong Zhang2
1School of Environment, Nanjing Normal University, Jiangsu Province Engineering Research Center of Environmental Risk Prevention and Emergency Response Technology, Jiangsu Engineering Lab of Water and Soil Eco-remediation, Wenyuan Road 1, Nanjing 210023, China; Institute of Agricultural Resources and Environment, Jiangsu Academy of Agricultural Sciences, Ministry of Agriculture and Rural Affairs, Liuhe Observation and Experimental Station of National Agro-Environment, Nanjing, 210014, China.
Biochar application in constructed wetlands (CWs) can mitigate greenhouse gas (GHG) emissions, but its effect on nitrous oxide (N2O) is uncertain. This study introduces a new metric and uses machine learning to accurately predict N2O mitigation, identifying key factors like COD/TN ratio for optimizing biochar
Area of Science:
- Environmental Science
- Climate Change Mitigation
- Wastewater Treatment
Background:
- Biochar is explored for climate change mitigation, but its impact on nitrous oxide (N2O) emissions in constructed wetlands (CWs) is unclear.
- Existing models struggle to accurately project biochar's N2O mitigation effect due to a lack of a standardized metric.
- Uncertainty in N2O emissions hinders accurate assessment of biochar's climate benefits.
Approach:
- A meta-analysis synthesized data from 80 global studies on biochar-amended CWs.
- A novel 'mitigation effect size' metric was developed for biochar's impact on N2O.
- Machine learning models, specifically extreme gradient boosting (XGBoost) optimized with Tree-structured Parzen Estimator (TPE), were employed for prediction.
Key Points:
- The TPE-XGBoost model achieved high accuracy in predicting N2O flux (R²=91.90%) and effect size (R²=92.61%).
- A high influent chemical oxygen demand/total nitrogen (COD/TN) ratio and COD removal efficiency significantly boosted N2O mitigation.
- The COD/TN ratio was a major contributor to both N2O flux and mitigation effect size.
Conclusions:
- This study provides a robust method for quantifying biochar's N2O mitigation potential in CWs.
- Optimizing CWs with high COD/TN ratios and utilizing granulated biochar from C-rich feedstocks can maximize climate benefits.
- Future biochar applications in CWs should consider these factors to enhance N2O mitigation and overall climate impact.
