Modeling the abiotic and host factors that determine tick load of Dermacentor variabilis on Peromyscus leucopus
Betsey C. York* and Justin Baraboo
Dr. Stephanie Fore and Dr. Hyun-Joo Kim, Faculty Mentors

The number of Dermacentor variabilis (American dog tick) on Peromyscus leucopus (white-footed mouse) is varied but what affects the tick load is unknown.We have been monitoring the number of ticks on mice since 2006. The objective of this study was to build statistical models to describe the number of larval ticks on a mouse based on various environmental and host variables. Model selection criterion were used to identify an appropriate model and significant variables. Zero-inflated negative binomial regression was the best fit among four commonly used count regression models. With all combinations of seven variable zero-inflated negative binomial regression models, host mass, host gender, habitat, degree days averaged over the previous sixty days, total precipitation ten days prior to sampling and day length on day of sampling were found to be significantly related to tick load.