Research

The USDA conducts regular surveillance for avian influenza infection in wild waterfowl across most of the US, but doesn’t have the resources to sample every location, or to sample frequently. The resulting estimates of AI prevalence on the landscape don’t incorporate diagnostic test sensitivity or specificity, and importantly they do not tell us anything about what mechanisms drive high prevalence in one area vs. another. We are developing a set of models that can estimate more accurate prevalence values in both sampled and unsampled locations, while also telling us more about what drives patterns of prevalence at large spatial scales.

Pathogens that infect migratory animals such as waterfowl can travel hundreds of miles a day, facing new environmental conditions and encountering many species of hosts. Understanding the movement of pathogens at such broad spatial scales is challenging since transmission cannot be directly observed and must be inferred from data gathered at much smaller spatial scales. In order to assess the risk of novel strains of avian influenza entering, and subsequently spreading though North America via migratory birds we must characterize the biological and environmental conditions under which these introductions occur. We are developing a spatial transmission model to determine the impact host movement and environmental covariates on when and where new viral introductions occur.

Migratory bird hosts are one of the primary routes by which new strains of avian influenza enter North America from Asia. In order to determine the frequency of these introductions, and to predict when and where they will occur in the future, we need a clear picture of host migration patterns. We are using a network approach to identify migratory communities that move between continents and identify key monitoring locations to detect this movement. We are also developing a model to combining multiple sources of movement data to generate an intercontinental migration network for east Asia and North America.

Several populations of deer on the west coast of the United States have shown widespread hair loss thought to be related to infestation by non-native louse species. For my master's work I examined the influence of ectoparasite abundance, endoparasite abundance, and micronutrient deficiency on the health and survival of an overwintering population of California mule deer affected by hair loss syndrome. I used generalized linear models to determine the influence of covariates on body condition, survival to the age of recruitment, and the incidence of hair loss.