Direct contact is necessary for transmission of many diseases, and direct contact is much more frequent within than between social groups.  The social structure of ungulates can be greatly influenced by landscape structure, with larger but less cohesive groups and higher dispersal rates in more open landscapes.  Furthermore, crop harvest in agricultural landscapes causes dramatic seasonal shift in habitat characteristics that may reduce familiarity with neighboring groups or increase the use of shared space.  Our objective was to test whether pairwise direct contact rates between white-tailed deer differed between a landscape of contiguous forest cover (southern Illinois) and an agriculture-dominated landscape (east-central Illinois).  Further, we tested whether any difference between landscapes differs depending on group membership.  Contacts (simultaneous locations <10 to <100-m apart) and group membership were determined from GPS collar data from 30 deer monitored between 2002 and 2005 in southern Illinois and 21 deer monitored between 2006 and 2009 in east-central Illinois.  We used mixed-model logistic regression was used to test for a difference in contact rates between landscapes after accounting for joint space use (intersection or product of fixed-kernel utilization distributions), season, and group membership.

Results/Conclusions

The product of utilization distributions was a stronger predictor of a deer pair’s contact rate than was volume of utilization.  For all proximity criteria, contact rates were higher for within-group than between-group pairs except during summer.  All else being equal, baseline contact rates tended to be higher in southern Illinois, although the evidence was inconsistent across seasons and proximity criteria (0.0012 < P < 0.42).  Group x landscape interaction was consistently nonsignificant (P > 0.1).  Odds ratios comparing contact rates within vs. between groups were greatest in winter for all proximity criteria (15-30), whereas they were smallest (ca. 5) in summer at a 10-m criterion and in autumn at 50- and 100-m criteria.  This interaction between season and distance criterion highlights two mechanisms reducing the grouping effect:  avoidance of close contact by does with young fawns in summer vs. “stirring” due to breeding activity during fall.  Our findings indicate that the effect of social structure on direct contact rates between deer is robust to differences in landscape structure and point to the breeding season as a potential “hot moment” for between-group disease transmission among female deer, not just between males and females.