Dall Sheep Demographics: A Peek at Intrinsic and Extrinsic Effects
AuthorBoan, Brianne V.
AdvisorStewart, Kelley M
Natural Resources and Environmental Science
StatisticsView Usage Statistics
Understanding population demographics requires an understanding of a multitude of factors, both extrinsic and intrinsic, that are important for reproduction, recruitment, and survival Nonetheless, physiological mechanisms underpinning reproduction, recruitment, and survival are often overlooked and only recently are being addressed to understand the root cause of demographic trends. In wildlife populations, fitness is measured in terms of reproductive success which is often manifested in population growth. Concerns arise when population counts reveal a declining trend without apparent cause. In south central Alaska, anecdotal evidence based on flight surveys of Dall sheep suggested a population decline since the mid-1990s. This population decline is what prompted our study. Dall sheep are an important source of revenue in the state of Alaska for both consumptive and non-consumptive use, however little research has been done on Dall sheep populations in south central Alaska over the last 25 years. Reviewing flight survey data between 1976 and 2013 in game management unit 14C (GMU 14C) suggested an increase in Dall sheep populations, peaking in 1996, and then a gradual decline. My objective was to determine potential extrinsic causalities for the population trend and link physiological health with expressed trends in pregnancy rates. In chapter 1, I focused on one game management unit (GMU 14C) in south central Alaska and in chapter 2, I studied 2 game management units (GMU 14C and GMU 13D) in south central Alaska. Both chapters involved capture work in 2012 and 2013. Pregnancy, parturition, and survival rates were evaluated through capture work done in GMU 14C and GMU 13D. Adult female Dall sheep were captured between mid-March and mid-April and fitted with VHF radio collars. We collected blood samples, hair samples (2013 only), assessed body condition, and determined age. Pregnancy was determined from blood collected from captured individuals. During parturition from early May to mid-June, collared females were monitored by air to observe neonates. Neonates were captured on foot, fitted with VHF radio collars, weighed, and sex and age were determined. Both collared adults and juveniles were monitored throughout the duration of the study to determine survival and cause of mortality. In chapter 1, I investigated specific demographic trends on pregnancy, parturition, and adult and juvenile survival rates and the relationship between extrinsic variables (weather and plant productivity) and reproductive output in GMU 14C. Pregnancy, parturition, and survival rates were determined from capture work in 2012 and 2013. Cause of mortality for both adults and juveniles was determined from field work. I used juvenile to adult female (age) ratios from flight survey data as an indicator of reproductive output. I used growing season length, onset, precipitation, length of winter, cumulated snowfall, and winter temperature variance to predict reproductive output. I used normalized difference vegetation index (NDVI) data, an indicator of plant productivity, to detect relationships between plant productivity variables and reproductive output. Pregnancy rates were 0.45 and 0.93 in 2012 and 2013, respectively. Parturition rates were 0.73 and 0.64 in 2012 and 2013, respectively and survival rates were 0.91 and 0.73 in 2012 and 2013, respectively. Leading causes of adult mortality were 40% avalanche and 40% predation. Juvenile survival rates were 0.59 and 0.44 in 2012 and 2013 respectively. The leading cause of mortality (79%) was predation. Late onset of the growing season and winter length had negative effects on reproductive output (i.e. fewer young relative to females). There was marginal support for growing season length (positive) and cumulative snowfall (negative). There was also marginal support for the effects of NDVI variables growth onset (negative), peak NDVI (positive), and peak date (positive) on reproductive output. In chapter 2, I investigated relationships between physiological health and reproduction. I was specifically interested in how reproductive efforts affect components of immune function and how cortisol levels, and indicator of chronic stress, affect reproduction. I focused on the constitutive branch of the immune system - specifically bacteria killing ability and the inflammatory response. Bacteria killing ability was evaluated using the bacteria killing assay. I used haptoglobin levels to evaluate the inflammatory response using a haptoglobin assay. Cortisol was quantified from the hair samples collected in 2013. I did not detect an effect of pregnancy on bacteria killing ability to inflammatory response in the 2012 data. Nonetheless, when pregnancy status was paired with study site, there was a negative effect on bacteria killing ability in 2013; however study site was the main driver of that equation. Study site was also the predictor variable for inflammatory response in 2013. These results suggest extrinsic factors are playing a greater role in constitutive immunocompetence. I observed no relationship between cortisol and pregnancy; however I observed marginal support for a negative relationship between maternal cortisol and birth weights of neonates. Understanding demographic processes is critical to set appropriate harvest regulations or to pinpoint where restoration efforts would be most effective. I demonstrated the ability to detect relationships between weather and reproductive output using survey data. Study site was the best predictor variable for immunocompetence, indicating external pressures, such as weather, forage availability, or pathogen presence, likely influenced physiological functions. Physiological health is traditionally viewed in terms of pregnancy and survival rates but those metrics do not describe mechanisms underpinning changes in population demographics. Understanding the role of cortisol and immune function can help answer questions regarding population trends. Applying modern analytical methods to historic and current data can inform comparisons with current and future datasets, and improve understanding of contemporary demographic trends.