Farmers have been using organic and regenerative practices for decades. Some of these techniques go back for millenia. Farmers apply these methods using knowledge from a lifetime of in-the-field learning and from networks of farm advisors. Every farm has a unique combination of climate, soil, and crops, requiring different practices and knowledge fine-tuned to each site.
For years farmers and researchers have worked together on scientifically evaluating growing practices using controlled and replicated experiments. Much of this work is done in collaboration with universities, the USDA, and other organizations such as the Organic Farming Research Foundation. Funding opportunities such as Conservation Innovation Grants provide support for exploring better ways to support healthy soil, water quality, and biodiversity.
We will use our machinery to assist with this kind of research. Since our field machines are involved in every step of the crop lifecycle (including soil preparation, planting, pest control, and harvesting), we obtain detailed field data, while doing useful field work, at no additional cost. Our machines use precision GPS and onboard cameras, so we can obtain measurements of the life history of each individual plant.
We will work with farmers and academic researchers who want this level of data to carry out research on various kinds of new farming practices. Research areas we would like to explore include:
- reducing tilling on large-scale organic farms
- using subsurface drip irrigation (enabled by reducing tilling) to save water
- estimating soil carbon levels using low-cost in-field sensors
- mechanical, flame-based, and laser-based weeding
- monitoring pests using detailed camera data, including multi-spectral imagery
- targeted releasing of beneficial insects based on the above pest data
In addition to explicit experiments, we can combine data across many farms to better understand how these issues and ideas vary across diverse soil types, weather conditions, and crop types. We are not assuming that massive data sets and machine learning will solve agriculture’s problems; black-box predictions may be far removed from what can be practically implemented on a farm, taking into account both biology and economics. We hope to support farmers and researchers as they explore these complex issues.