New research outlines how AI tools can accelerate soil science by speeding up early-stage work. Soil systems are difficult to predict and are influenced by climate, weather patterns, and agricultural practices. But the authors stress that it will impact how we respond to “the world’s most pressing challenges”: food security and climate change.
Using these tools, AI has the potential to accelerate soil research and understanding of how it benefits food and climate systems, addressing global changes in soil security, climate resilience, and sustainable land use.
“A better understanding of soil could help land managers detect nutrient loss, water stress, compaction and erosion earlier, supporting more sustainable agriculture, better soil management and stronger climate adaptation,” said Budiman Minazny, lead author of the study and professor of soil and landscape modeling at the University of Sydney in Australia.
“We assessed the system’s ability to perform perceptual processing, strategic planning, and scientific reasoning. Our findings highlight the promise that multi-agent AI systems hold, with globally important impacts on soil, a valuable but perhaps undervalued resource.”
AI powers soil science
This study frontiers of sciencepoint out that current AI systems generally lack the integration, adaptability, and reasoning capabilities necessary to address the multifaceted and interconnected nature of soil systems.
“By working with experts, AI can help us better adapt to the complexity and ever-changing nature of soil ecosystems,” said lead author Alex McBratney, professor of digital agriculture and soil science at the University of Sydney in Australia.
“Unlike current machine learning tools that focus on isolated tasks, these systems can mimic scientific collaboration to a very sophisticated degree, combining reasoning, planning, and interdisciplinary insights to support researchers and drive major advances.”
The authors argue that AI has the potential to enhance areas beyond mapping.He said there is growing awareness of the vital importance of soils in the functioning of the planet, and soil science will continue to grow and thrive under scientist-driven AI.
Machine learning is currently being used in soil science approaches such as digital soil mapping and spectroscopy. However, the authors argue that AI has the potential to enhance these areas beyond mapping.
“Unlike traditional machine learning, these intelligent systems mimic scientific collaboration, combining reasoning, planning, and interdisciplinary insights to support human researchers.”
The study explains that AI applications range from digital soil twinning and microbiome monitoring to climate adaptation modeling, and hold promise for major advances in sustainable land use and soil carbon management.
imitate the scientific process
AI-designed soil strategies have the potential to transform nutrient utilization by optimizing fertilization rates. The researchers provided scientific literature for a review of multi-agent AI systems to generate ideas about how soils store carbon and what controls how much carbon soils can store.
The AI agent then proposed five hypotheses, including climate influences, biological and chemical controls, saturation thresholds, interdisciplinary feedbacks, and management strategies.
The study notes that the AI system successfully mimicked human scientific processes and provided output that was “beyond” what is currently in use, while being consistent with expert research.
Notably, AI also faces challenges around data quality, creativity, interpretability, and risk of error in the absence of human oversight.
The study argues that rather than replacing human expertise, AI has the potential to enhance it by allowing scientists to spend more time focusing on problems that require expert judgment.
“To fully realize this promising future, we must foster interdisciplinary collaboration, ensure equitable access to AI tools, and carefully address ethical challenges. Bridging digital innovation and real-world applications can unlock new levels of understanding, stewardship, and safety for one of our planet’s most important and enduring resources: our soils,” the study concludes.
