In this study, A. Altikat and his team focused on a serious environmental problem. They studied how pesticides like metamitron move through soil and contaminate water systems. These chemicals are widely used in agriculture, but they easily reach groundwater. This makes them harmful to ecosystems.
The researchers wanted a better and smarter solution. They tested biochar as a way to reduce pesticide pollution. They also explored modern tools like hyperspectral sensing and artificial intelligence. These tools helped them predict how well biochar works.
2. Materials and Methods
The team used three types of biochar: hazelnut shells, apricot kernel shells, and waste car tires. They mixed each biochar with soil at 5%, 15%, and 25%. This helped them test how well biochar reduces pesticide movement.
They measured pesticide levels using GC–MS, a precise lab method. They also used hyperspectral sensors to track light changes linked to pesticides.
To improve prediction, they applied machine learning models. These included PLSR, SVM, Random Forest, and ANN. This combination made the study more advanced and reliable.
3. Results
The results showed clear differences between biochar types. Apricot shell biochar performed the best. It reduced pesticide levels more than the other types. The lowest concentration appeared at the 25% application rate.
Waste tire biochar showed weaker performance. Its dense structure limited adsorption.
The study also found that wavelengths between 600–690 nm helped detect pesticide levels. Among all models, ANN gave the most accurate predictions.
4. Conclusion
This study shows that biochar can reduce pesticide pollution effectively. However, the type of biochar matters. Plant-based biochar works better than industrial waste-based biochar.
The study also shows the value of combining biochar with AI tools. These tools improve monitoring and prediction. Overall, this approach supports sustainable agriculture and better pollution control.
Reference
Altikat, A., Alma, M. H., Altikat, S., & Gürbüz, R. (2026). Integrating biochar characterization, hyperspectral signatures, and artificial neural networks for predictive modeling of metamitron leachate attenuation. International Journal of Environmental Science and Technology. https://doi.org/10.1007/s13762-026-07122-3
