Survival of Adult Female Bighorn Sheep Following a Pneumonia Epizootic

Beginning in the early 1900s, poly-factorial, poly-microbial pneumonia was identified as a disease affecting bighorn sheep (Ovis canadensis) and it continues to threaten bighorn populations, posing an ongoing management challenge. In May and June 2013, a pneumonia outbreak linked to the pathogenMycoplasma ovipneumoniaeled to an all-age die-off of desert bighorn sheep (O. c. nelsoni) at Old Dad Peak in the Kelso Mountains of the Mojave Desert in California, USA. Subsequently, we observed clinical signs of respiratory disease among bighorn sheep in multiple neighboring ranges. Our objective was to investigate post-outbreak survival of adult female bighorn across 9 populations from 2014 to 2017 in the Mojave Desert and evaluate the relationship betweenM. ovipneumoniaeinfection and survival, while testing effects of range factors that could potentially influence differences in adult female survival (i.e., forage quality, winter precipitation, population abundance). We fitted adult females with radio-collars following the outbreak and collected serum and nasal swab samples for competitive enzyme-linked immunosorbent assay (cELISA) and polymerase chain reaction (PCR) testing to determine exposure and infection status at time of capture. We tracked survival of 115 adult females with radio-collars and used the known-fate model in Program MARK to evaluate effects and estimate survival from November 2013 to March 2017. Annual survival was negatively correlated with positive infection status at capture but varied across populations with respect to differences in range conditions. Summer and autumn forage quality, as represented by mean normalized difference vegetation index (NDVI) values for summer and autumn, was positively correlated with overwinter survival, whereas winter precipitation (a proxy for winter severity) was negatively correlated with overwinter survival. Population abundance was negatively correlated with annual survival, suggesting a potential density-dependent effect. Model-averaged annual survival estimates ranged from 0.700 +/- 0.07 (SE) to 0.945 +/- 0.026 for infected individuals and 0.896 +/- 0.03 to 0.983 +/- 0.011 for uninfected individuals. We conclude that summer and autumn forage quality, indexed by NDVI, may partially offset the negative effect associated withM. ovipneumoniaeinfection on host survival. Our survival modeling results suggest that chronic infection may have afflicted adult females that were PCR-positive (i.e., infected withM. ovipneumoniae) at time of capture. We propose programmatic re-testing of infected individuals to assess pathogen persistence at the individual level and evaluate whether selective culling might potentially help to reduce prevalence and transmission within populations. (c) 2020 The Wildlife Society.