Current Research

Partners in Success

The intersection between farmer experimentation and university research

Sun rising over a corn field with very young corn plants, and white gentlemen who is mostly unseen bending to touch an individual corn plant with affection

There is a critical need to co-create mechanisms that promote producer-led experimentation, while also creating a data pipeline for university research. Co-creation is significant because it improves the relevance of the science, while also improving the autonomy of producers for independent experimentation. Our rationale is that a co-creation process will enhance producer efforts to obtain higher quality data, improve their capacity to translate research findings, and facilitate gathering of usable on-farm data. We are partnering with other stakeholders to target a producer-initiated research question to evaluate our co-creation approach, innovating in participatory research, and simultaneously offering Extension opportunities in related topics to broader audiences. This work is supported by USDA NIFA CARE award #2022-68008-36356.

Climate Smart Partnership for Mississippi

Increasing climate-smart market opportunities for mid-south corn and soybean producers

A woman holding a giant clump of earth with grassy material on one side and exposed soil on the other

Achieving greenhouse gas (GHG) reductions in animal feed production systems will require the adoption of climate-smart practices, as well as dedicated research into quantifying GHG reduction benefits to verify a climate-smart commodity. However, adoption of conservation practices that can provide climate benefits, such as cover crops, remains low in comparison to other regions of the U.S., ranging from 1-10% in the mid-south. Therefore, the implementation and comprehensive assessment of a pilot climate-smart systems program in grain production operations is critical to document the environmental, agronomic, and economic benefits of adoption, and to demonstrate a viable and scalable opportunity to produce a climate-smart commodity in the mid-south. Climate-smart systems that 1) integrate with existing crop rotations, 2) support profitability and food security, and 3) minimize adoption risks for producers while demonstrating GHG emissions reductions are critical to overcoming producer barriers to conservation adoption. Failure to develop an approach that integrates with existing production systems, includes accurate GHG reduction estimates, and addresses immediate market demands, will result in decreased yield, increased direct costs of implementation, and limited ability to market and sell a climate-smart commodity for a value-added premium. Such impacts would ultimately increase undue burden and risk on our agricultural community and decrease conservation adoption. This project is funded by USDA NRCS award #NR233A750004G041.

Closing the Conservation Yield Gap

Collaboration with USDA ARS

sunrise over a cotton field in bloom with a barn shown in the distance

This work endeavors to develop a small uncrewed aerial system and related computational fluid dynamics model to monitor greenhouse gas flux at the farm-scale. The resulting system will be paired with an unmanned ground robot to form a collaborative robotics system for multi-level carbon sensing in crop fields. This work is supported by USDA ARS under agreement #58-6064-3-007.

Data Science for Decision Making

Applying deep learning models to automate identification of erroneous data points

AI-generated photorealistic image of small corn plants with a sunflare in the background. The image as a digital overlay meant to imply computer data associated with the corn plants.

Producers rely on automated yield monitoring devices to understand crop yield and yield variability in their farm fields. Yield data, is however, error prone and requires human experts to properly filter out invalid data points. This project investigates the role artificial intelligence, commonly referred to as "AI," can play in supporting agricultural decision making and farm management practices that arise from yield data. The hypothesis is that AI-based models can provide open, equitable, efficient, and objective yield data cleaning tools that automate and improve the quality of yield data processing. This work is supported by USDA NIFA DSFAS award #2024-67021-4252.