INTERPRETING MEAN CHEMICAL-RATIOS FROM SIMPLE RANDOM COLLECTIONS OF SNOW-URINE SAMPLES

Urea nitrogen:creatinine ratios (UN:C) in urine-soaked snow samples we re lower for adult female elk (Cervus elaphus) in Yellowstone National Park, Wyoming, 1992-1994, than for adult males or calves, both within and between winters. These inter-class differences indicate that the traditional use of mean snow-urine ratios from simple random collectio ns to assess or compare the nutritional status of gregarious ungulates is unwise because mean ratios are influenced by the unknown proportio n of samples collected from each sex-age class. Using simulations base d on field data, we illustrate how apparent spatial and temporal diffe rences in mean ratios among sequential collections of snow-urine sampl es from various populations or the same population over time are confo unded with differences in sex-age composition among collections. Spati al and temporal comparisons of mean UN:C ratios are only interpretable when samples of known origin are used and comparisons are made within sex-age classes. The interpretation of mean UN:C ratios is further ha mpered because it is unclear exactly what UN:C ratios are indexing. Th ere is high unexplained within-animal variability in UN:C ratios over short periods, and ratios are an inconsistent index of nutritional sta tus until animals have lost >20% of their autumn body mass. Further re search on factors influencing UN:C ratios are needed as well as invest igations of alternative urinary metabolites that are more sensitive to moderate changes in nutritional status and are less affected by sex-a ge differences among sampled animals.