Biochar is widely promoted as a climate friendly soil amendment that can store carbon and improve crop growth. Yet scientists have long debated whether it always benefits soil ecosystems. A new study suggests the answer is more complex. By combining meta analysis with machine learning, researchers show that biochar can either help or harm soil organisms depending on how it is produced and applied.
The research, published in the journal Biochar, analyzed results from dozens of previous experiments to better understand how biochar affects plants, microbes, and soil animals. The findings provide one of the most comprehensive assessments to date of biochar's ecological impacts and offer new tools for predicting its effects before it is applied in the field.
Biochar is a carbon rich material created by heating biomass such as crop residues or wood under low oxygen conditions. Because it can improve soil structure and lock carbon in the ground for long periods, it is increasingly used in agriculture and climate mitigation strategies. However, different studies have reported conflicting results regarding its ecological impacts.
"To move toward sustainable biochar use, we need to understand not only its benefits but also the conditions under which it might cause harm," said the study's corresponding author. "Our work brings together a large body of data to clarify these patterns."
The research team compiled experimental results from 61 scientific studies, creating a dataset of 1,329 observations describing how biochar influences soil organisms. These data included responses of plants, microorganisms, and soil animals such as earthworms. The scientists first conducted a meta analysis to evaluate overall trends across the studies.
The results revealed that biochar's overall effect on soil organisms was close to neutral when all data were averaged together. However, important differences emerged when specific groups were examined. Plants generally benefited from biochar application, showing an average positive growth response. In contrast, animals and some microbes often experienced negative effects, especially in survival related measurements.
To better understand these patterns, the researchers then applied machine learning techniques. Using properties of the biochar and the soil, along with information about the tested organisms, they trained computer models to predict whether biochar would produce beneficial or hazardous outcomes.
The best performing model, based on a random forest algorithm, correctly classified biochar effects with an accuracy of about 79 percent. The model also identified several key factors that strongly influence biochar's impact on soil ecosystems.
Among the most important predictors were biochar pH, the amount applied to soil, the temperature used during biochar production, and the pH of the soil itself. These factors can determine whether biochar improves soil conditions or introduces stresses for organisms living in the soil.
For example, higher application rates were associated with more negative effects, likely because excessive biochar can bind nutrients and reduce their availability to organisms. Biochar produced at very high temperatures or with extremely high pH values also tended to increase ecological risks. Conversely, moderate application rates and lower production temperatures were generally linked with more beneficial outcomes.
The researchers say these insights could help guide safer and more effective biochar use in agriculture. Instead of assuming that biochar is always beneficial, farmers and land managers may need to consider both soil conditions and biochar properties before applying it.
The study also highlights gaps in current scientific knowledge. Many existing experiments focus mainly on plant growth, while relatively few examine impacts on soil animals or long term ecosystem changes. Expanding research in these areas could further improve predictive models and environmental guidelines.
"Our results show that biochar is neither simply a fertilizer nor a pollutant," the researchers noted. "Its effects depend on the interaction between the material and the soil ecosystem."
By combining meta analysis with artificial intelligence, the study offers a new approach for evaluating complex environmental technologies. The authors hope such data driven tools will support the responsible use of biochar as interest in carbon sequestration and sustainable agriculture continues to grow.
===
Journal Reference: Dong, Y., Tunali, M. & Nowack, B. Fertilizer or pollutant: analyzing the effects of biochar on soil organisms using machine learning. Biochar 8, 28 (2026).
https://doi.org/10.1007/s42773-025-00528-1
===
About Biochar
Biochar (e-ISSN: 2524-7867) is the first journal dedicated exclusively to biochar research, spanning agronomy, environmental science, and materials science. It publishes original studies on biochar production, processing, and applications—such as bioenergy, environmental remediation, soil enhancement, climate mitigation, water treatment, and sustainability analysis. The journal serves as an innovative and professional platform for global researchers to share advances in this rapidly expanding field.
