Heterogeneity in individual life-history trajectories over age: quantifying the relative importance of ontogeny, selection, and environment

The quantification of how life-history traits vary with age is fundamental because age-specific variations affect evolution of traits and their influence on population dynamics. Changes in life-history traits with age occur at the individual level, but most previous studies of age-specific changes at the population level did not account for individual heterogeneity, a fundamental element shaping population responses and evolutionary processes. Although some recent studies have demonstrated the importance of individual heterogeneity to understand the ontogenetic processes observed in populations, very few studies have tried to quantify the relative contributions of these evolutionary processes. A major goal of this project was to partition and quantify the roles of ontogenetic (experience and senescence), selection (viability and fertility), and environmental processes (weather, density) on changes in life-history traits with age. To understand these processes at a higher level of biological organization, our aim was to compare the relative contribution of these processes among ungulate species that show contrasted dynamics and vary in their position along the slow-fast continuum (with the slower species having a generation time twice that of the faster species), as well as among populations of the same species living under contrasting environmental conditions. This, however, required improved biostatistical methods to model heterogeneity among individuals in a population. Therefore, one novel and key aspect of this project was to perform a statistical evaluation of the methods commonly used in longitudinal studies, and develop more effective alternative methods. This project was separated in two main streams: the assessment/development of analytical methods for studying life histories and their implementation in a standardized way to contrast evolution of traits among species. We first developed the use of mixture models to quantify the structure in individual variability and present their advantages over commonly used mixed models in a review paper entitled ?Assessing variation in life-history tactics within a population using mixture regression models: a practical guide for evolutionary ecologists?. The second part resulted in the used of mixture models to better understand life-history traits and how they change with age, contrasting the results in large comparative analyses. This resulted in 5 other papers. In a first paper entitled ?Cohort variation in individual body mass dissipates with age in large herbivores?, we developed different standardization methods to contrast the change in individual variation with age among 11 populations of ungulates and between the two sexes. In a second paper entitled ?Compensatory growth reduces chances of surviving to old age in wild ungulates?, we quantified compensatory growth in early age and evaluated whether it affected the chances to survive to old age. We showed an early-late life trade-off between compensatory growth and survival, the first evidence of such a trade-off in wild animals. In a third paper entitled ?Body size variation in ungulates: A formal test of Lomnicki (1983) hypothesis?, we are formally testing Lomnicki?s hypothesis (1983) for the first time, and we showed strong support in ungulates and other species. The fourth and fifth papers are part of a special theme issue that we have submitted to the Philosophical Transactions of the Royal Society B, entitled ?Ontogeny, adaptation, and chance in life-history trajectories: individual differences matter?. As guest Editors of this special issue, we will have an introduction paper that will provide a summary of the topic, its importance to ecology and evolution, and provide a terminology that will help homogenise definitions in the literature. We will also have a specific contribution to this issue that will be entitled ?Quantifying individual variability and its influence on life-history traits: different methods for different questions and contexts?, where we will review the different methods of assessing the influence of individual variability in life history. There were also four other papers that were published with students I supervised in relation with this project. These papers assessed individual variability and environmental effects in a specific ungulate species. To sum up, the project was intended to lead to two major scientific contributions: 1) to provide robust analytical methods of life-history strategies, which are standardized and hence greatly valuable for comparative analyses, and 2) to allow a better understanding of evolution of life history strategies and their link with environmental processes. As described above, both scientific contributions have been fulfilled and have resulted in six papers well on their way to be published in highly prestigious journals, in addition to the 4 papers already published in very good journals.

The quantification of how life history traits vary with age is fundamental because age-specific variations affect evolution of traits and their influence on population dynamics. Changes in life history traits with age occur at the individual level, but mo st previous studies of age-specific changes at the population level did not account for individual heterogeneity, a fundamental element shaping population responses and evolutionary processes. Although some recent studies have demonstrated the importance of individual heterogeneity to understand the ontogenetic processes observed in populations, very few studies have tried to quantify the relative contributions of these evolutionary processes. Here, our goal is to partition and quantify the roles of ontog enetic, selection, and environmental processes on changes in life-history traits with age, and to compare the contributions of these processes among species and traits. We will use longitudinal data on individually marked animals monitored from birth to d eath in 11 populations (8 ungulate species) studied intensively for 10-40 years. We will assess the relative contribution of each process to the mean and the variance of reproductive life-history traits and body size. We will compare the relative contribu tion of these processes among species that show contrasted dynamics and vary in their position along the slow-fast continuum, as well as among populations of the same species living under contrasting environmental conditions. That comparison, however, req uires improved biostatistical methods to model heterogeneity among individuals in a population. Therefore, one novel and key aspect of this project is to perform a statistical evaluation of the methods commonly used in longitudinal studies, and develop mo re effective alternative methods. The project will lead to two major scientific contributions: robust analytical methods of life history strategies and a better understanding of evolution of life history strategies.