Biovigilance Toolbox for Nationwide Biosurveillance of Air-borne Plant Pathobiome: Environmental and Landscape Drivers of Airborne Cereal Pathogens

Abstract

Fungal pathogens pose a constant threat to Canadian cereal production by reducing yield and grain quality. Many of these pathogens disperse via air currents, enabling rapid, long-distance spread. Moving beyond reactive disease management, this study employs a biovigilance framework to characterize and predict pathogen dynamics. We tested three primary hypotheses: that airborne pathogen distribution is driven by host-crop availability (spatial hypothesis), regulated by specific meteorological conditions (temporal hypothesis), and that species-level resolution reveals ecological patterns obscured at the genus level (taxonomic resolution hypothesis). Seven air samplers were deployed across seven provinces from 2020 to 2023, collecting 437 samples. Using ITS2 metabarcoding and the AODP bioinformatic pipeline, we generated 14.4 million high-quality reads. We identified 43 ASVs representing 15 species of cereal-associated pathogens. Supporting the spatial hypothesis, the wheat specialist Parastagonospora nodorum and various Ustilago species exhibited distributions strictly coupled with the prevalence of their respective host crops. In support of the temporal hypothesis, Fusarium, Blumeria graminis, and Puccinia species displayed climate-responsive dynamics, with abundance modulated by humidity and wind speed. Furthermore, consistent with our taxonomic resolution hypothesis, resolving pathogens to the species level allowed us to distinguish host-specific dispersal patterns within complex genera like Ustilago. This study demonstrates that integrating high-resolution airborne metabarcoding with landscape and climatic data provides a scalable framework for predictive disease surveillance and targeted intervention in complex agroecosystems. The pronounced seasonal peaks of Blumeria graminis associated with relative humidity and host presence within a 5 km radius further validate the temporal hypothesis, linking environmental suitability with host availability. This study demonstrates that integrating high-resolution airborne metabarcoding with landscape and climatic data can pinpoint high-risk periods and locations, offering a scalable framework for predictive disease surveillance and targeted intervention in complex agroecosystems.