Heterogeneity of reproductive age increases the viability of semelparous populations

Abstract

Summary One important feature of natural populations is that individuals of the same age or functional class at a given time may vary in their demographic rates, such as their survival probabilities or fecundities. This demographic heterogeneity is due to the plastic response of individuals to spatially variable environmental conditions, to genetic variation or to maternal or cohort effects. Although demographic heterogeneity may have serious consequences, it is generally not included in models of population dynamics. The interactions of demographic heterogeneity with environmental stochasticity and density dependence are poorly understood, as are its consequences for population viability. We explored the effect of heterogeneity in age at reproduction on the dynamics and viability of semelparous populations. We developed individual‐based models according to which individuals had to follow a particular life path in a multifurcated life cycle. We used data from a well‐studied monocarpic plant, Centaurea corymbosa, to calibrate the models with realistic distributions of demographic parameters. Disregarding the trivial effect of changing generation time, we found that heterogeneous populations had higher viability than homogeneous populations due to the desynchronization of demographic processes among individuals. Heterogeneity buffers the negative impact of environmental stochasticity by reducing the covariation of life histories among individuals (i.e. the bet‐hedging effect). In addition, when negative density dependence is implemented, desynchronization in heterogeneous populations reduces intraspecific competition. Indeed, crowded cohorts, occurring randomly due to demographic or environmental stochasticity, can spread their offspring over several years, which limits negative interactions among individuals due to resource limitations (in this study, a fixed number of microsites for seedlings). These effects are likely to occur in every semelparous population. Our results allow us to better understand the functional links between the heterogeneity, desynchronization, dynamics and viability of populations facing spatial and temporal variations in their environments. These links suggest that demographic heterogeneity, regardless of its causes (genetic variation or plasticity), is an important component of population dynamics. Neglecting demographic heterogeneity leads to overestimating the level of synchronization among individuals, which may in turn strongly bias viability assessments in realistic environments.