Agriculture needs better crops and better use of those crops under tightening land, water, nutrient and climate constraints. Genomics can generate and identify variation at scale, and high-throughput sensing can measure more phenotype than ever before. The limiting step is increasingly understanding what those measurements mean biologically, early enough to make a breeding or crop-management decision.
We combine crop physiology, deep phenotyping and causal computation to break equifinality, the problem that similar yield, stress or sensor signals can arise through different physiological routes.
Our aim is to make physiological state-space a practical organising layer for crop improvement, supporting earlier selection, physiologically meaningful crop models, and autonomous systems that can diagnose why a crop is departing from a desirable trajectory and identify the intervention most likely to improve productivity, resilience and resource efficiency.
BBC News · 25 Mar 2026
BBC News reported on the Greenbox acoustic network after its AI-assisted monitoring highlighted conservation-interest and unusual bird detections, including firecrest and pink…
Read more
BBC Sounds · 25 Mar 2026
BBC radio explored the high-tech research behind Greenbox: a distributed network of autonomous acoustic sensors, wireless data transfer and AI-assisted analysis being used to …
Read more
BBC News · 01 Aug 2025
BBC News covered the expansion of the Greenbox network, describing how solar-powered field units and artificial intelligence support continuous biodiversity monitoring before …
Read more