Winter feeding of elk in the Greater Yellowstone Ecosystem and its effects on disease dynamics

Abstract

Providing food to wildlife during periods when natural food is limited results in aggregations that may facilitate disease transmission. This is exemplified in western Wyoming where institutional feeding over the past century has aimed to mitigate wildlife-livestock conflict and minimize winter mortality of elk (Cervus canadensis). Here we review research across 23 winter feedgrounds where the most studied disease is brucellosis, caused by the bacterium Brucella abortus Traditional veterinary practices (vaccination, test-and-slaughter) have thus far been unable to control this disease in elk, which can spill over to cattle. Current disease-reduction efforts are being guided by ecological research on elk movement and density, reproduction, stress, co-infections and scavengers. Given the right tools, feedgrounds could provide opportunities for adaptive management of brucellosis through regular animal testing and population-level manipulations. Our analyses of several such manipulations highlight the value of a research-management partnership guided by hypothesis testing, despite the constraints of the sociopolitical environment. However, brucellosis is now spreading in unfed elk herds, while other diseases (e.g. chronic wasting disease) are of increasing concern at feedgrounds. Therefore experimental closures of feedgrounds, reduced feeding and lower elk populations merit consideration.This article is part of the theme issue 'Anthropogenic resource subsidies and host-parasite dynamics in wildlife'.

Keywords: Brucella abortus; brucellosis; chronic wasting disease; disease ecology; feedgrounds; wildlife–livestock conflict.

Figure 1.

There are 23 supplemental feedgrounds for elk in Wyoming. The National Elk Refuge, north of Jackson, is operated by the US Fish and Wildlife Service, while the remainder are operated by the Wyoming Game and Fish Department. (Online version in colour.)

Figure 2.

Hypothesized maintenance and reservoir hosts for B. abortus in the GYE during the three stages of the disease to date. Initially (a), cattle (bottom left) were a source population that infected bison (top left), fed elk (top right) and unfed elk (bottom right). After effective control measures were implemented in cattle (b) they were no longer a maintenance host but could be reinfected from fed elk. After 2000 (c), unfed elk became part of the reservoir community, able to maintain the infection in the absence of other host populations and are now a source of infection to cattle. Arrows depict established transmission paths. Arrow thickness denotes relative importance. (Online version in colour.)

Figure 3.

Seroprevalence trends from three unfed herd units in western Wyoming. Historically, unfed herds rarely exceeded 5% prevalence; however, seroprevalence of some unfed herds has crept upwards, now reaching the 10–40% seroprevalence observed on feedgrounds. Circle radius represents sample size; smoother lines were fit using generalized additive models with the mgcv package [25] in R [26]. (Online version in colour.)

Figure 4.

Test-and-slaughter of seropositive female elk between 2006 and 2010 reduced seroprevalence for brucellosis at the Muddy Creek feedground from 37% to 5%. In comparison, Scab Creek and Fall Creek received 2 years of treatment, the impacts of which are less clear. Points represent the proportion of seropositive animals tested in a given year, with 95% confidence intervals (bars). Vertical dotted lines represent years in which test-and-slaughter occurred at each site. Smoother lines were fit using generalized additive models with the mgcv package [25] in R [26]. (Online version in colour.)

Figure 5.

Seroprevalence estimates (empty circles, scaled to sample size) and rolling average feedground end date in the previous 8 years (solid circles) at Dell Creek and Greys River feedgrounds. Smoothed seroprevalence estimates (lines) were fit using generalized additive models in the mgcv package [25] in R [26]. (Online version in colour.)