Individual Quality Drives Life History Variation in a Long-Lived Marine Predator

AbstractIndividual quality and environmental conditions may mask or interact with energetic trade-offs in life history evolution. Deconstructing these sources of variation is especially difficult in long-lived species that are rarely observed on timescales long enough to disentangle these effects. Here, we investigated relative support for variation in female quality and costs of reproduction as factors shaping differences in life history trajectories using a 32-year dataset of repeated reproductive measurements from 273 marked, known-age female gray seals (Halichoerus grypus). We defined individual reproductive investment using two traits, reproductive frequency (a female's probability of breeding) and provisioning performance (offspring weaning mass). Fitted hierarchical Bayesian models identified individual investment relative to conspecifics (over a female's entire life and in three age classes) and subsequently estimated how these investment metrics and the Atlantic Multidecadal Oscillation are associated with longevity. Individual differences (i.e., quality) contributed a large portion of the variance in reproductive traits. Females that consistently invest well in their offspring relative to other females survive longer. The best-supported model estimated survival as a function of age class-specific provisioning performance, where late-life performance was particularly variable and had the greatest impact on survival, possibly indicating individual variation in senescence. There was no evidence to support a trade-off in reproductive performance and survival at the individual level. Overall, these results suggest that in gray seals, individual quality is a stronger driver in life history variation than individual strategies resulting from energetic trade-offs.

Large offspring have enhanced lifetime reproductive success: Long-term carry-over effects of weaning size in gray seals (Halichoerus grypus)

An individual's size in early stages of life may be an important source of individual variation in lifetime reproductive performance, as size effects on ontogenetic development can have cascading physiological and behavioral consequences throughout life. Here, we explored how size-at-young influences subsequent reproductive performance in gray seals (Halichoerus grypus) using repeated encounter and reproductive data on a marked sample of 363 females that were measured for length after weaning, at ~4 weeks of age, and eventually recruited to the Sable Island breeding colony. Two reproductive traits were considered: provisioning performance (mass of weaned offspring), modeled using linear mixed effects models; and reproductive frequency (rate at which a female returns to breed), modeled using mixed effects multistate mark-recapture models. Mothers with the longest weaning lengths produced pups 8 kg heavier and were 20% more likely to breed in a given year than mothers with the shortest lengths. Correlation in body lengths between weaning and adult life stages, however, is weak: Longer pups do not grow to be longer than average adults. Thus, covariation between weaning length and future reproductive performance appears to be a carry-over effect, where the size advantages afforded in early juvenile stages may allow enhanced long-term performance in adulthood.

Prior choice and data requirements of Bayesian multivariate hierarchical models fit to tag‐recovery data: The need for power analyses

Recent empirical studies have quantified correlation between survival and recovery by estimating these parameters as correlated random effects with hierarchical Bayesian multivariate models fit to tag‐recovery data. In these applications, increasingly negative correlation between survival and recovery has been interpreted as evidence for increasingly additive harvest mortality. The power of these hierarchal models to detect nonzero correlations has rarely been evaluated, and these few studies have not focused on tag‐recovery data, which is a common data type. We assessed the power of multivariate hierarchical models to detect negative correlation between annual survival and recovery. Using three priors for multivariate normal distributions, we fit hierarchical effects models to a mallard (Anas platyrhychos) tag‐recovery data set and to simulated data with sample sizes corresponding to different levels of monitoring intensity. We also demonstrate more robust summary statistics for tag‐recovery data sets than total individuals tagged. Different priors led to substantially different estimates of correlation from the mallard data. Our power analysis of simulated data indicated most prior distribution and sample size combinations could not estimate strongly negative correlation with useful precision or accuracy. Many correlation estimates spanned the available parameter space (−1,1) and underestimated the magnitude of negative correlation. Only one prior combined with our most intensive monitoring scenario provided reliable results. Underestimating the magnitude of correlation coincided with overestimating the variability of annual survival, but not annual recovery. The inadequacy of prior distributions and sample size combinations previously assumed adequate for obtaining robust inference from tag‐recovery data represents a concern in the application of Bayesian hierarchical models to tag‐recovery data. Our analysis approach provides a means for examining prior influence and sample size on hierarchical models fit to capture–recapture data while emphasizing transferability of results between empirical and simulation studies.

Physiological tipping points in the relationship between foraging success and lifetime fitness of a long-lived mammal

Although anthropogenic change is often gradual, the impacts on animal populations may be precipitous if physiological processes create tipping points between energy gain, reproduction or survival. We use 25 years of behavioural, diet and demographic data from elephant seals to characterise their relationships with lifetime fitness. Survival and reproduction increased with mass gain during long foraging trips preceding the pupping seasons, and there was a threshold where individuals that gained an additional 4.8% of their body mass (26 kg, from 206 to 232 kg) increased lifetime reproductive success three-fold (from 1.8 to 4.9 pups). This was due to a two-fold increase in pupping probability (30% to 76%) and a 7% increase in reproductive lifespan (6.0 to 6.4 years). The sharp threshold between mass gain and reproduction may explain reproductive failure observed in many species and demonstrates how small, gradual reductions in prey from anthropogenic disturbance could have profound implications for animal populations.

Identification of familial networks reveals sex-specific density dependence in the dispersal and reproductive success of an endangered ungulate

Density is an important demographic parameter that is commonly overlooked in studies of wild populations. Here, we examined the effects of variable spatially explicit density on a range of demographic parameters in a wild population of a cryptic ungulate, boreal woodland caribou (Rangifer tarandus caribou). Using non-invasive genetic sampling, we applied spatial capture–recapture methods with landscape covariates to estimate the density of boreal woodland caribou across a 108,806 km2 study area. We then created a familial network from the reconstructed parent–offspring relationships to determine whether spatial density influenced sex-specific individual reproductive success, female pregnancy status, and dispersal distance. We found that animal density varied greatly in response to land cover types and disturbance; animal density was most influenced by landscape composition and distance to roads varying from 0 in areas with >20% deciduous cover to 270 caribou per 1,000 km2 in areas presenting contiguous older coniferous cover. We found that both male and female reproductive success varied with density, with males showing a higher probability of having offspring in higher-density areas, and the opposite for females. No differences were found in female pregnancy rates occurring in high- and low-density areas. Dispersal distances varied with density, with offspring moving shorter distances when parents were found in higher-density areas. Familial networks showed lower-closeness centrality and lower-degree centrality for females in higher-density areas, indicating that females found in higher-density areas tend to be less broadly associated with animals across the range. Although high-density areas do reflect good-quality caribou habitat, the observed decreased closeness and degree centrality measures, dispersal rates, and lower female recruitment rates suggest that remnant habitat patches across the landscape may create population sinks.

Demystifying individual heterogeneity

Among-individual variation in vital rates, such as mortality and birth rates, exists in nearly all populations. Recent studies suggest that this individual heterogeneity produces substantial life-history and fitness differences among individuals, which in turn scale up to influence population dynamics. However, our ability to understand the consequences of individual heterogeneity is limited by inconsistencies across conceptual frameworks in the field. Studies of individual heterogeneity remain filled with contradicting and ambiguous terminology that introduces risks of misunderstandings, conflicting models and unreliable conclusions. Here, we synthesise the existing literature into a single and comparatively straightforward framework with explicit terminology and definitions. This work introduces a distinction between potential vital rates and realised vital rates to develop a coherent framework that maps directly onto mathematical models of individual heterogeneity. We suggest the terms "fixed condition" and "dynamic condition" be used to distinguish potential vital rates that are permanent from those that can change throughout an individual's life. To illustrate, we connect the framework to quantitative genetics models and to common classes of statistical models used to infer individual heterogeneity. We also develop a population projection matrix model that provides an example of how our definitions are translated into precise quantitative terms.