Global yield-suitability modeling for regenerative practices
Global yield-suitability modeling for regenerative practices is a research-based spatial analysis framework that identifies where regenerative farming methods—such as no-tillage, cover crops, agroforestry, and organic farming—can effectively increase crop yields globally. By using machine learning to integrate environmental data with field-scale experimental results, the model provides insights to guide sustainable agricultural policies and investments, helping to balance food security with soil health.
Overview
Global yield-suitability modeling for regenerative practices is a research-based spatial analysis tool designed to identify geographical locations where regenerative farming practices (RFP) can enhance crop yields. By integrating field-scale experimental data with global datasets on climate, soil, topography, and vegetation, this modeling framework provides a data-driven foundation for policymakers and agricultural stakeholders to target sustainable farming interventions. The research evaluates four key practices: no-tillage, cover crops, agroforestry, and organic farming. It helps address the challenge of increasing global food production while simultaneously improving soil health and mitigating environmental degradation.
Technical specifications
The modeling framework utilizes a Random Forest machine-learning algorithm to perform regression analysis, linking global spatial predictor variables against observed relative yield changes from meta-analyses. Key features include:
- Spatial resolution: Analysis conducted at a 5-arc-minute resolution.
- Predictive modeling: Uses Leave-Location-Out cross-validation (LLOCV) to account for spatial autocorrelation.
- Uncertainty quantification: Employs Quantile Regression Forests (QRF) to produce 90% uncertainty intervals for yield predictions.
- Data integration: Incorporates global datasets for bioclimatic variables, soil properties, topography, and land cover.
- Crop-specific insights: Provides suitability maps for various crop groups, including maize, wheat, cereals, and vegetables/fruits.
Technology readiness level
The technology is at TRL 3 (Analytical and experimental critical function and/or characteristic proof of concept). The methodology has been developed and published as a peer-reviewed global assessment, serving as a foundational research tool to illustrate potential outcomes. It is not currently a commercial product; the researchers emphasize that the maps are starting points for...
About University of Bristol
The University of Bristol is a comprehensive Russell Group research university with a strong record of collaborating with industry on translational R&D. Its new Temple Quarter Enterprise Campus provides shared labs, demonstrator environments, and a single front door for companies through Bristol Innovations. SETsquared Bristol—based at the Engine Shed—accelerates venture creation and connects corporates to a pipeline of startups and spin‑outs. Healthcare collaboration is embedded through strategic partnerships with regional NHS trusts, including the NIHR Bristol Biomedical Research Centre, enabling clinical translation at scale. Research is supported by competitive funding from UKRI councils and the National Institute for Health and Care Research, alongside industry partnerships.
delivered to your inbox.