Oral Technical Session: Ecology and Epidemiology

173-O

Networks of stored wheat: towards improving sampling and management strategies in the United States and Australia.
J. F. HERNANDEZ NOPSA (1), G. Daglish (2), D. Hagstrum (3), J. Leslie (4), T. Phillips (3), C. Scoglio (5), S. Thomas-Sharma (4), G. Walter (6), K. Garrett (7)
(1) Kansas State University and Plant Biosecurity Cooperative Research Centre (CRC), Canberra, Australia, Manhattan, KS, U.S.A.; (2) Department of Agriculture, Fisheries and Forestry, Agri-Science Queensland, Australia, and Plant Biosecurity Cooperative Research Centre (CRC), Canberra, Australia, Brisbane, Australia; (3) Department of Entomology, Kansas State University, Manhattan, KS, U.S.A.; (4) Department of Plant Pathology, Kansas State University, Manhattan, KS, U.S.A.; (5) Department of Electrical and Computer Engineering, Kansas State University, Manhattan, KS, U.S.A.; (6) School of Biological Sciences at the University of Queensland, Australia, Brisbane, Australia; (7) Department of Plant Pathology, Kansas State University, and Plant Biosecurity Cooperative Research Centre (CRC), Canberra, Australia, Manhattan, KS, U.S.A.

Wheat is the second most important staple worldwide. Insect, fungal, and mycotoxin contamination affect wheat grain quality during storage, and farmers and storage companies suffer economic losses due to mycotoxin contamination, trade limitation, and pesticide resistant insect populations. Stored grain moves from fields to storage structures (silos, elevators, depots), among storage structures, and to a final destination in the milling industry by train, truck, and barges. The network of stored grain movement may facilitate the dispersal of fungi, insect, and other contaminants among the nodes. We developed network models with 37 nodes (states) for the United States and 41 nodes (sites) for Australia (AU). Metrics such as average shortest path (2.1 and 6.7) and transitivity (0.38 and 0.35) were obtained for the US and AU, respectively. Analysis of shortest paths, betweenness centrality, and node degree highlighted the importance of KS, IL, and ID in the US and the Toowoomba, Natcha, and Malu in AU. A striking difference between the countries was that highly connected nodes were in the central US but in coastal AU, suggesting different optimal sampling and mitigation strategies for the two systems. Developing multilayer and interconnected network models of a) grain transportation, b) movement of fungi and insects, and c) management communication may be an important next step for understanding the risk of subpopulations of pesticide-resistant insects and mycotoxins.