Predictive Niche Modeling for the Identification of Maize Pathogens of Greatest concern in the United States

Maize is one of the world’s most valuable food crops with 717 million metric tons produced annually. Its economic significance worldwide is second only to rice. Given the importance of maize, it is crucial to understand the potential range of pests and pathogens that pose a significant risk to the crop. Ecological niche modeling is used to identify the environmental requirements of these pests and pathogens. Models can be built using existing occurrence data and records of environmental conditions such as vegetative coverage, isothermality, altitude, temperature, and precipitation. In this study, I use pest occurrence location data from the Global Biodiversity Information Facility and bioclimatic variables from WorldClim to create generalized additive models (GAM), maximum entropy (MaxENT) models, boosted regression trees (BRT), ensemble, and CLIMEX models to predict suitable habitat for maize pests and pathogens in the US. Distribution models were made of insect pests of highest concern, including Lepidopterans Autographa gamma, Chilo partellus, Helicoverpa armigera, Spodoptera litura, and Thaumatotibia leucotreta, Coleopterans Diabrotica speciosa and Heteronychus arator, and the Hemipteran Laodelphax striatellus. Each of the forty models were then used to make maps of the potential geographical range that highlights areas that would be most suitable to the greatest number of pests. Coastal areas are susceptible to most maize pests and these maps convey the levels of risk associated with land near an ocean. These maps can be used to efficiently direct preventative action to high-risk areas.