Field-Scale Testing of Weed Identification and Mapping Tools for Accelerating Integrated Weed Management Adoption

PI: Erin Haramoto, Plant and Soil Sciences Title: Field-Scale Testing of Weed Identification and Mapping Tools for Accelerating Integrated Weed Management Adoption Source: subaward from ARS (funder is United Soybean Board) Abstract: This proposal will address the USB priority theme of sustainability by regionalizing crop and weed monitoring tools and user interfaces to accelerate the adoption of precision and sustainable agricultural practices by US soybean producers. Automation in agriculture, especially for monitoring crops (biomass) and weeds (density and biomass), is lacking and has substantial implications for evaluating and tailoring inputs (herbicides, non-chemical management, and costs) and the resulting outcomes (yield, return-on- investment, and weed seed production), and assisting decision-making for future planning. This proposal will conduct field-scale testing of two low-cost, easy-to-use weed and crop performance monitoring systems. Regional testing will facilitate improvements in weed density and biomass estimations for additional weed species. This project will also include the construction and beta-testing of a user- friendly interface that automates data analysis and visualization. Our goals are to refine precision agriculture technology (computer vision and artificial intelligence tools) to estimate and map weed density and biomass in US soybean production regions, and build a web-based application that automates data analysis and visualization for both farmers and researchers at the end of one year. Since such technology has many uses (described above), we envision a multi-year project that links the spatial distribution of climate, soils, crop yields, and weeds, enabling site-specific management. Additionally, our work will be publicly available for other research and development groups to continue improving precision “see-and-spray” and other technologies. The outputs of this project are expected to tremendously enhance both growers’ weed management and profitability as well as researchers’ ability to efficiently conduct multi-location field trials and scale-up from small plot research to whole field trials.