Sex- and age-specific resource selection and harvest mortality of elk: balancing disease risks with conservation benefits in a fragmented agricultural landscape

Abstract

Integrating the characteristics of landscape structure with species’ attributes that determine animal movement and consequently disease risk is a complex, yet critical step for effective conservation and disease management. I examined movement behavior of elk (Cervus canadensis manitobensis) in response to factors such as habitat fragmentation and predation risk (i.e., human hunting) to better understand the landscape-level risk of disease spread in Manitoba. Objectives of this thesis were to: (i) identify sex- and age-specific habitat corridors for elk movement to assess the potential risk of disease spread between elk sub-populations; and (ii) evaluate sex- and age-specific elk distribution and hunter-kill sites during the hunting season to understand the impact of hunting on elk resource selection and to improve the effectiveness of current disease control programs.

Elk in the Riding Mountain region of southwestern Manitoba are endemic with bovine tuberculosis (Mycobacterium bovis; TB) and are threatened by the imminent emergence of chronic wasting disease (CWD). I used collared elk locations from a combined dataset of 413 non-migratory female and male elk that were captured in and around Riding Mountain National Park and Duck Mountain Provincial Park and Forest between 2002 and 2011. Elk were fitted with either a GPS satellite collar (24 F; 12 M) or very high frequency (VHF) radio-transmitter (191 F; 186 M). In addition, I used 796 locations of hunter-killed female and male elk collected between 2003-2012 from the agricultural-dominated lands that surround the two protected areas. These data were used to develop resource selection function (RSF) models, which are powerful analytical tools that characterize and predict the selection of resources by animals. The RSF models integrated with graph theory revealed important sex-specific differences in resource selection during spring and summer (Mar-Aug) and identified potential habitat corridors between the two parks. The extent of connectivity across the fragmented agriculture-dominated landscape remained relatively constant across several spatial scales for both sexes and connectivity was greater for adult females compared with juvenile males. During the hunting season (Sept-Feb), sex- and age-specific differences in resource selection and hunter-kill sites were also evident. Adult males were rarely located outside of the parks, and were killed considerably less often than adult females and juvenile males, in close proximity to park boundaries. Adult females and juvenile males made some use of the agriculture-dominated landscape outside of the parks and were killed in a range of habitats located farther from the park boundaries. My research demonstrates that habitat connectivity and hunting are more critical than previously appreciated in limiting the potential spread of TB and CWD infected elk across this highly fragmented landscape. However, the implications for long-term conservation of elk remain an important concern. The risk of TB and CWD spread among elk sub-populations is low; however, the socioeconomic and ecological implications of these diseases remain significant. My thesis results include detailed maps of functionally connected areas that facilitate elk movement, and thus pinpoint local areas of disease management concern. I also provide an example of a small-scale manipulation of hunter effort linked to disease prevalence, with mapped areas where hunting can be targeted at the highest risk individuals for disease transmission. Resource managers can use these maps as a complementary tool to evaluate both the short- and long-term implications of habitat fragmentation and hunting efforts to effectively balance elk conservation and mitigate disease risks.