Evolutionary Robustness of the Storage Effect Coexistence Mechanism: Theoretical Investigations of the Evolution of Seed Germination Patterns in Annual Plants

Abstract

A fundamental question in ecology is how species with similar ecologies coexist. Species coexistence mechanisms are a potential answer, as they describe the ways that species can interact that allow for the persistence of multiple, interacting species. Species coexistence mechanisms have thus been the subject of intense study in ecology, yet little is known about how these mechanisms change as species evolve and the interactions between species evolve. Underlying all species coexistence mechanisms is the common property that they strengthen within-species density feedbacks relative to betweenspecies density feedbacks. Evolutionary processes that change the relative strengths of these feedbacks will modify the strength of a coexistence mechanism. The storage effect coexistence mechanism, which describes temporal niche partitioning of the physical environment, formalizes and makes clear the components that directly lead to a strengthening of within-species feedbacks relative to between-species feedbacks. The storage effect has two components: buffered population growth and density-dependent covariance between responses to the environment and competition. Here I present a study of the evolution of the two components of the storage effect in annual plant populations living in arid environments. For interacting annual plant species in arid environments, species have buffered population growth when they have a persistent between-year seed bank. Between-year seed banks occur when seeds have a propensity to remain dormant. Density-dependent covariance between responses to the environment and competition occur when seed germination patterns are species-specific. Species-specific germination patterns result from differential responses of species to the same environmental factors affecting germination. This study consists of theoretical investigations of (1) the evolution of seed dormancy, (2) the evolution of species-specific germination patterns, and (3) the joint evolution of seed dormancy and species-specific germination patterns. In each case, the evolutionary force of interest is natural selection. First, I present results on the evolution of seed dormancy, focusing on the effects of variation in seed germination across years. Yearly variation in germination is commonly observed in annual systems in arid environments and is facilitates temporal niche partitioning. I find that variable germination strengthens selection for dormancy under reasonable ecological conditions. Hence, seed dormancy, and by extension buffered population growth, are robust to selection under common ecological conditions. Second, I present results on the evolution of species-specific germination patterns. Species-specific germination patterns can result from character displacement. I ask how character displacement is affected by similarity of two interacting species in their growth traits. Annual plants in arid environments also respond strongly to the physical environment during growth as seedlings, sometimes in a species-specific manner. I find that character displacement can be both strengthened and weakened by environmental conditions affecting growth, depending on the similarity between species in their growth responses. Character displacement is weaker for similar species and is stronger for species with different growth responses. The results show that species-specific responses in one part of the life cycle, representing density-dependent covariance between environmental and competitive responses, can facilitate further divergence of species. Third, I present results on the joint evolution of seed dormancy and species-specific germination patterns. Germination in the study is assumed to be dependent on temperature at the time of rainfall. I find that there are two routes to the evolution of the storage effect. In the first, direct selection for seed dormancy leaves open ecological opportunities for competing species, which allows for the evolution of species-specific germination. In the second, disruptive selection favors the emergence of multiple, coexisting phenotypes that differ in their germination patterns. As the coexisting phenotypes diverge in germination patterns, dormancy is indirectly favored. Similarly, character displacement between species in sympatry can cause species-specific differences. Seed dormancy and species-specific germination evolve jointly in both cases. These results are the first to investigate the consequences of natural selection for multiple components of a coexistence mechanism. In most cases, adaptation in a fluctuating environment for two species favors the evolution of both components of the storage effect, which suggests it is robust to natural selection.