<p>Spatial variation in microhabitats, predation pressure, and competitor assemblages may create a landscape of selection pressures that drives spatial variation in phenotypes. Coral reef ecosystems provide a wide range of environmental variability and therefore an excellent opportunity to quantify and explore the potential effects of fitness landscapes on phenotypes of reef fish that inhabit these ecosystems. I evaluate patterns of variation in phenotypic traits of a common coral reef fish (Thalassoma hardwicke) across a prominent environmental gradient (from offshore to inshore within a lagoon system). I quantify phenotype-environment gradients established for cohorts of fish soon after their settlement, and how these relationships change through the time to infer selection gradients (Chapter 2). Specifically, I estimate the strength of selection on a set of early life-history traits estimated from otoliths (i.e., larval growth rates and pelagic larval duration), and morphological features (i.e., body condition and fin size).
</p><p><br></p>
Building on the results of Chapter 2, I conduct an observational field study to estimate the behavioural consequences of spatial variation in early life history traits for young T. hardwicke (Chapter 3). I quantify feeding frequency and agonistic interactions between young T. hardwicke and intra- and interspecific competitors, and evaluate these as a function of growth history traits. Growth history traits correlate positively with the frequency and direction of agonistic interactions. Species identity (i.e., which species were interacting with young T. hardwicke) is also important for determining the frequency and direction of agonistic interactions. Additionally, the size difference between T. hardwicke and the competitor also influenced the frequency and direction of agonistic interactions.
I use laboratory experiments to better understand the role of conspecifics on settlement choice of young T. hardwicke (Chapter 4). I evaluate the influence of growth histories on settlement choice in a laboratory experiment. Growth history does not significantly influence habitat choice with regards to conspecific presence for newly settled T. hardwicke. Additionally, fish that avoided habitats with conspecifics took longer to make a settlement choice, which may suggest that neophobic fish may choose habitats without conspecifics possibly to avoid competition.<div><br>I then use field experiments to evaluate the role of conspecifics on post-settlement survival of young T. hardwicke (Chapter 4), focusing on the role of conspecific size-differences and priority effects. I pair newly settled fish with larger conspecifics to evaluate the role of size-differences and priority effects on 1) frequency of agonistic interactions, and 2) post-settlement survival of newly settled T. hardwicke. I find no significant differences in either frequency of agonistic interactions or post-settlement survival.</div><div><br></div><div>The presence of phenotype-environment gradients in this system provides an excellent opportunity to test for phenotype-environment mismatches in young T. hardwicke in different environments. I set up a reciprocal transplant experiment in the field (Chapter 5) by comparing growth and survival of ‘control’ fish (i.e., fish remaining in their original environments) to that of ‘transplant’ fish (i.e., fish transplanted to a new environment). Transplant fish experience a significant reduction in survival, which suggests that phenotype-environment mismatch may be present in this system. I also found spatial differences in growth rates for treatment fish, suggesting the cost of phenotype-environment mismatches are context-dependent.
Overall, the observational and experimental components of my thesis suggest that patterns of settlement and subsequent post-settlement fitness are influenced by the interface between phenotypes and environment. I find significant spatial variation in phenotypes of newly settled T. hardwicke, and post-settlement survival is also spatially variable. Additionally, disrupting the established phenotype-environment gradients alters growth patterns and increases mortality. These results highlight the importance of context-dependence in understanding patterns of settlement and survival for young reef fish and illustrate the various roles of ecological processes that shape phenotypic distributions within ecosystems.</div>
Social and life-history correlates of hormonal partner compatibility in greylag geese (Anser anser)