CITRUS NURSERY SOURCE: How Artificial Intelligence Applications Could Enhance Nurseries

By Peter Chairs

The University of Florida Institute of Food and Agricultural Sciences’ (UF/IFAS) commitment to the use and development of artificial intelligence (AI) in production agriculture is fueling conversations throughout the citrus industry. For most of us, AI was not even on our radar a few years ago. Today, commercial and nursery growers are excited about the development of the UF/IFAS Center for Applied Artificial Intelligence in Agriculture and are dreaming about its possible near-term contributions.

Citrus Nursery Source reached out to Chris Gunter, chair of the UF/IFAS Horticultural Sciences Department, to casually brainstorm how citrus nurseries might benefit from advances in agricultural AI. Though some of the tools mentioned below may have been developed for other horticultural purposes, it is not unreasonable to assume that they may ultimately be adapted for use in nurseries. Economies of scale will likely be a determinant in the economic viability of these systems. Over time, AI tools will likely be scaled down, fine-tuned and become economically feasible for smaller operations. Here are some possible AI applications for citrus nurseries:

• Disease and pest detection. Vision and data interpretation technologies can search for and analyze important physical plant characteristics, such as injury, coloration and plant stresses. Early detection could make it possible to segregate affected plants and limit spread. Such systems can also be tuned to detect specific insect pests before plant damage occurs, allowing for earlier management response.
• Resource management. Fertilization and irrigation systems can be maximized using hyperspectral technologies to detect wilt and other issues. This can help minimize water waste and improve plant performance. Once abnormalities are detected, AI could conceivably enable automated decision-making and response.
• More efficient scouting. Labor will continue to be an issue for nurseries, and minimizing the need for manual scouting could prove highly beneficial. There would still be a need to ground-truth significant findings, but verification would be cheaper than 100% human scouting. Nursery management is perpetually on call. AI-driven scouting and monitoring will be no different. Human knowledge will be incorporated into algorithms driving the systems, and the systems will add knowledge as they go. The importance of around-the-clock monitoring will dictate that these systems cannot be shared among smaller companies. Smaller operations must wait until tools become affordable for their volume of output.
• Inventory management. Automated scouting can track maturity of plants throughout their growth cycle and fine-tune delivery schedules. It is conceivable that sorting and segregation may be done through robotic applications.
• Possible identification of zygotic liners. AI could detect small variations in liners prior to budding that might be missed by the human eye.
• Better management of physical structures. Water leaks, poorly performing exchange fans, etc. can be detected before more significant damage occurs.

Technology is changing and adapting at lightning speed. Some of the early AI applications will be focused on areas other than citrus nurseries, but nurseries will certainly stand to benefit from UF/IFAS taking the lead position in this arena.