Factors influencing the recruitment, growth and reproduction of a temperate reef fish, Forsterygion lapillum

Abstract

In order to gain a comprehensive understanding of population dynamics, it is vital to identify the key factors that contribute to variation in both survival and reproductive success throughout the life history of an organism. The population dynamics of reef fishes may be influenced by events occurring across multiple life stages, throughout the entire life cycle. For instance, the input of new individuals into a population (recruitment) is heavily shaped by mortality in the larval and juvenile stages, which is influenced by conspecific interactions and habitat characteristics. For individuals that survive, variation in somatic growth histories during development may influence mating success among fish that successfully recruit into the adult population, and particular developmental histories may receive a disproportionate amount of a population’s reproductive output. However, female preferences for particular phenotypes may also be modified by events occurring in adult life, such as parasite infection. Finally, absolute reproductive success (i.e., the number of offspring that survive to reproductive age) may be dependent upon early larval mortality of offspring, and variation in larval mortality among spawning sites could have consequences for metapopulation dynamics. In this thesis, I investigated how recruitment, growth and reproductive success varied among individuals of a small temperate reef fish, Forsterygion lapillum, the common triplefin, based upon their developmental histories, morphological traits, and habitat characteristics (including conspecific densities, regional locations, etc.). Specifically, I examined:

- how the spatial distribution and survival of juveniles is influenced by age-class interactions (Chapter 2) - verified methods to measure somatic growth rates during development using scale structures (Chapter 3) - explored how previous growth rates influence reproductive success (Chapter 4) - evaluated how reproductive success is modified by the presence of ectoparasites (Chapter 5) - and finally, assessed how natal origin modifies larval survival probabilities among offspring (Chapter 6).

The larvae of many reef fishes settle into habitats that are already occupied by adults, and interactions between age classes (intercohort interactions) may affect spatial variation in recruitment strength across settlement sites. In Chapter 2, I evaluated spatial covariation in juvenile and adult densities of F. lapillum (within the preferred settlement habitats of juveniles) to investigate correlations between adult and juvenile densities potentially caused by age-class interactions. The relationship between juvenile and adult densities followed a dome-shaped curve, with a negative correlation between juveniles and adults at higher adult densities. The shape of this curve was temporally variable, but was otherwise unaffected by particular features of the site (algal species identity). Using a laboratory-based experiment that used a “multiple predator effects” (MPE) design, I tested the hypothesis that increased settler mortality, caused by either (i) intercohort competition leading to enhanced predation risk or (ii) cannibalism by adults on juveniles, contributed to the observed negative relationship between juvenile and adult densities. Results suggested overall mortality attributable to cannibalism was low; however, smaller settlers appeared to be more vulnerable to cannibalism. There was no evidence that combined or interacting effects between predators (F. lapillum adults and Forsterygion varium [the variable triplefin]) increased predation risk in settlers of F. lapillum. Overall, these results highlight the potentially complex effects adult residents may have on shaping patterns of recruitment and the distributions of new juveniles.

Somatic growth rates through ontogeny are one of the most important metrics for understanding fish populations and in Chapter 3, I evaluate the use of spacing between growth increments on fish scales (called circuli) as a measurement technique for assessing historical growth in F. lapillum. First, I established the relationship between scale growth and body size, and determined how variable this relationship was among populations. The body-scale size relationship was strongly positive and was unaffected by gender; however, there did appear to be significant differences between certain populations. Second, I monitored somatic and fish scale growth in the laboratory to measure the relationship between somatic growth and spacing between growth increments (intercirculus spacing). New scale growth and circuli deposition were both positively correlated with somatic growth. Average intercirculus spacing was also positively correlated with somatic growth rate, but this appeared to differ between age/size classes, with the older and larger individuals showing a weaker relationship. Results suggest that intercirculus spacing can be used to determine previous growth histories, but may be limited to particular size/age ranges (e.g., juveniles).

In Chapter 4, I employ the techniques developed in Chapter 3 to examine how early growth rates (derived from fish scales) and male morphological traits explain variation in reproductive output between individual males in F. lapillum. I measured the reproductive success of breeding males in relation to their size and growth rates over the breeding season at two different spawning locations. Clutch size (number of eggs per nest) was highly variable among individuals over the study period; however, I detected a significant, albeit subtle, negative correlation between clutch size and growth rates after settlement. Although growth explained relatively small amounts of total variation, it was the only male trait I measured that significantly correlated with clutch size. The negative effects of faster growth on clutch size were greatest during the period of growth after settlement suggesting that growth at this early stage may be important for later reproductive success (early post-settlement).

In Chapter 5, I examined how infection with an ectoparasite modified reproductive success among individual males using a field survey. Females often preferentially mate with unparasitised males, and therefore parasitised males experience lowered reproductive success. In this study, individuals of greater total length were more likely to be infected with an ectoparasite, but were also more likely to have an egg nest. Parasite infection had no effect on reproductive success (either the presence of a nest, or the average surface area of eggs if a nest was present). Positive covariation in total length, reproductive success, and parasite infection potentially suggest that the influence of parasitic infection on reproductive success may depend upon the strength of selection for larger male body size. In addition, this study provides the first quantitative measurement of ectoparasite infection for both the focal parasite species (Caligus buechlerae) and the host (F. lapillum).

Finally, in Chapter 6, I explore how larval survival is mediated by spawning location. In marine reef fish, spatially isolated adult populations may be connected (i.e., have gene flow) via larval dispersal; however, differential larval survival between source populations may mediate both the degree of population connectivity as well as the reproductive success of individuals within those source populations. To evaluate variation in larval quality among different spawning locations, I conducted a laboratory assay to measure the potential effects of source population on larval time to starvation, as starvation is often proposed as a major source of mortality for larval fish. Average survival time was 3.75 days, but survival analysis indicated that starvation resistance did not differ between the two natal sources. For individual nests, mean larval size was negatively correlated with their mean survival time, although this was only apparent in larvae collected from one population (the south coast). My findings indicate that variation in larval traits between source populations does exist, but that on average, source populations had equal resistance to starvation. Given the differences between source populations in (i) the relationship between larval mortality and larval size (i.e., the absence of size effects in one source population) and (ii) overall variation in larval size (larger larvae on the south coast), the relative contribution of larvae from each source population may vary under certain conditions (e.g., low levels of food availability).

In conclusion, the field surveys and laboratory experiments conducted in this thesis demonstrate the potential for a variety of factors across multiple life history stages to influence recruitment, growth and reproduction. These findings suggest that factors across multiple life stages (e.g., conspecific density, previous growth histories, or spawning site) have the ability to influence individual success, and in turn populations. By carefully considering and integrating these factors into our studies of population dynamics, we may be able to gain a more comprehensive understanding of the spatio-temporal fluctuations in populations for marine reef fish.