Noninvasive Assessments of the Nutritional and Reproductive Status of Free-ranging Herbivores

Résumé du livre

I detected high within-animal variance in urea nitrogen:creatinine (UN:C) ratios from urine deposited in snow by elk (Cervur elaphus) in Yellowstone Park, Wyoming. This substantial variability was unexpected and led to experiments that examined whether it was due to collection and measurement techniques, or reflected actual biological variability. Dilution by snow, measurement variability, and diurnal variations in metabolite excretion did not contribute substantially to the total within-animal variance. Thus, snow-urine ratios accurately estimate the true ratios of metabolites in an elk's bladder urine. The interpretation of UN:C ratios is complex, however, since ratios reflect nitrogen balance and are often an inconsistent index of overall nutritional stress. Adults and calves have different UN:C profiles during winter, which poses problems for using simple random collections of samples to compare the nutritional status of populations. The mean ratio for a collection is influenced by the proportion of samples obtained from each age class, thereby increasing the variance of the index. Comparisons of mean ratios are confounded by differences in the age-class composition of populations between collections. Allantoin and cortisol are also purported to index nutritional status in ungulates. I evaluated the sensitivity and interpretability of these metabolites by collecting sequential snow-urine samples from radiocollared, adult female elk in Yellowstone Park during consecutive winters (1991-1994) that varied in severity and forage availability. UN:C ratios were insensitive to differences in nutritional stress among years, whereas trends in cortisol:creatinine ratios were uninterpretable. Allantoin:creatinine ratios were sensitive to differences in nutrition and provided an interpretable biological signal; suggesting that this metabolite warrants further study. I also validated fecal steroid analysis as a means of diagnosing pregnancy in elk by measuring the concentrations of estrone conjugates, pregnanediol-3-glucuronide, and free progesterone in samples from pregnant and nonpregnant animals. Pregnant females had higher concentrations of these steroids than nonpregnant animals, especially later in gestation. Pregnancy diagnoses should be based upon multiple samples due to high within-animal variability in steroid concentrations.