Detecting Resource Limitation in a Large Herbivore Population Is Enhanced With Measures of Nutritional Condition

Abstract

Resource limitation at the population level is a function of forage quality and its abundance relative to its per capita availability, which in turn, determines nutritional condition of individuals. Effects of resource limitation on population dynamics in ungulates often occur through predictable and sequential changes in vital rates, which can enable assessments of how resource limitation influences population growth. We tested theoretical predictions of bottom-up (i.e., resource limitation) forcing on moose (Alces alces) through the lens of vital rates by quantifying the relative influence of intrinsic measures of nutritional condition and extrinsic measures of remotely sensed environmental data on demographic rates. We measured rates of pregnancy, parturition, juvenile, and adult survival for 82 adult females in a population where predators largely were absent. Life stage simulation analyses (LSAs) indicated that interannual fluctuations in adult survival contributed to most of the variability in λ. We then extended the LSA to estimate vital rates as a function of bottom-up covariates to evaluate their influence on λ. We detected weak signatures of effects from environmental covariates that were remotely sensed and spatially explicit to each seasonal range. Instead, nutritional condition strongly influenced rates of pregnancy, parturition, and overwinter survival of adults, clearly implicating resource limitation on λ. Our findings depart from the classic life-history paradigm of population dynamics in ungulates in that adult survival was highly variable and generated most of the variability in population growth rates. At the surface, lack of variation explained by environmental covariates may suggest weak evidence of resource limitation in the population, when nutritional condition actually underpinned most demographics. We suggest that variability in vital rates and effects of resource limitation may depend on context more than previously appreciated, and density dependence can obfuscate the relationships between remotely sensed data and demographic rates.