Soil Analysis

Real-time soil analysis supports fertilizer decisions, field variability mapping, and soil carbon quantification in agricultural operations.

Business impact

• Nitrogen fertiliser savings — Stenon reports 20-40% nitrogen fertiliser savings in multi-year, multi-country customer use cases.
• Yield increases — Stenon reports 2-8% yield increases across several crops.
• Degraded grasslands covered by regeneration project — rTek uses VT0014 for a regeneration project involving 500,000 hectares of degraded grasslands in Kazakhstan.

Data requirements

• Soil sensor readings (sensor) — Stenon uses proprietary optical and electrical sensors to collect soil data in the field.
• Soil samples (tabular) — Perennial’s ATLAS-SOC model draws from a database of more than 350,000 soil samples.
• Local calibration data (tabular) — Perennial uses local calibration to adapt soil carbon estimates to the project area.

AI methods and techniques

• Machine learning — Perennial uses it in ATLAS-SOC to generate soil carbon estimates from soil samples and local calibration.
• Agronomic models — Stenon combines them with sensor data to translate soil measurements into fertilizer guidance.
• Digital soil mapping — Perennial uses it to estimate soil organic carbon across large areas with more data points than traditional sampling alone.

AI models and model families

ATLAS-SOC

Industries

Real-world evidence

2 documented case studies on record.

Companies using this: Stenon, rTek.