Skip to main content
ARS Home » Southeast Area » New Orleans, Louisiana » Southern Regional Research Center » Food and Feed Safety Research » Research » Research Project #439725

Research Project: Modeling to Forecast Mycotoxin Breakout in U.S.-Grown Maize

Location: Food and Feed Safety Research

Project Number: 6054-41420-009-006-S
Project Type: Non-Assistance Cooperative Agreement

Start Date: Jun 1, 2021
End Date: May 30, 2025

Objective:
The goal of this research collaboration is to develop and implement predictive computer models for mycotoxin (aflatoxin and fumonisin) contamination in U.S. corn. Such models provide usable and practical information directly to the stakeholder to aid in crop management. In future projects, these models can also be used to estimate effects of predicted alterations in climatic conditions, such as elevated CO2 levels and elevated temperatures, on mycotoxin occurrence, which helps developing long-term mitigation strategies for U.S. food crops.

Approach:
Predictive modeling utilizes various computational tools, including artificial intelligence, machine learning and big data analysis algorithms. Many of the barriers to the development of U.S. models are due to lack of mycotoxin modeling capabilities and expertise in ARS. However, these technologies for data integration and successful modeling for prediction of mycotoxin breakout have been developed by cooperator. In order to develop predictive models that aid U.S. stakeholders, it is necessary to have access to large data sets with multiple years of aflatoxin contamination data measured from diverse geographic regions, as well as the relevant historical climatic data related to the sites. Limited sets of mycotoxin data are currently available from selected states in the United States. Other meta data that would need to be incorporated include the dates of sample collection and date of harvest, plus geographical location of the sampled fields, and rainfall, humidity, temperature data of the specific fields/years. Other anonymized data indicating the cultivar of corn that was analyzed and possible corn borer infection, would be very helpful in the development of the predictive models, which will be very useful to stakeholders, farmers and grain industries with the ultimate objective of controlling unwanted contamination of food and feed crops with dangerous levels of mycotoxins.