A holistic approach towards understanding population dynamics of a coral reef sponge

Understanding processes that contribute to population maintenance is critical to the management and conservation of species. Sponges (Phylum Porifera) are an evolutionary and ecologically significant group. However, information on biological and ecological processes, such as reproduction, larval dispersal, settlement, recruitment, survival and growth that influence sponge population dynamics is surprisingly limited. This study aimed to quantify pre- and post-settlement processes that affect the demographics and distribution of Carteriospongia foliascens (Thorectidae, Phyllospongiinae), a common coral reef sponge widely distributed from intertidal to mesophotic zones of the Great Barrier Reef (GBR). Intertidal populations of C. foliascens from the inshore central GBR were selected for this population study. Extreme morphological plasticity induced by local environmental conditions is inherent in sponges, hence it is imperative to investigate hidden diversity and identify robust taxonomic units prior to ecological studies to avoid the confounding interpretation of data due to species misidentification. Taxonomic and phylogenetic assessments of the study species and closely related taxa were therefore conducted, forming an essential preamble to the study proper. Foliose keratose sponges of the sub-family Phyllospongiinae (Dictyoceratida, Thorectidae: Strepsichordaia, Phyllospongia and Carteriospongia) are commonly found in intertidal and subtidal habitats of the Indo-Pacific. Lacking spicules, these sponges can be difficult to differentiate due to insufficient reliable morphological characters for species delineation. Molecular phylogenies inferred from the nuclear Internal Transcribed Spacer 2 region (ITS2) and morphometrics (19 characters; 52 character states) were used to identify evolutionarily significant units (ESUs; sensu Moritz) within foliose Phyllosponginiids collected from seven geographic locations across tropical eastern and Western Australia (Chapter 2). The ITS2 topology was congruent with the tree derived from the Bayesian inference of discrete morphological characters thereby supporting expected taxonomic relationships at the genus level, and the identification of five ESUs. However, phylogenies inferred from the ITS2 marker revealed multiple sequence clusters, some of which were characterized by distinct morphological features and specific geographic ranges. Results are discussed in light of taxonomic incongruences within this study, hidden sponge diversity and the potential role of vicariant events on present day distribution patterns. The identity of intertidal C. foliascens on the inshore GBR was confirmed and facilitated further assessments of population ecology. Sexual reproduction is integral to our understanding of population dynamics. The mode of sexuality and development, seasonality, sex ratios, gametogenesis, reproductive output and size at sexual maturity were established for C. foliascens, in the central GBR, over two reproductive cycles (Chapter 3). A population sexual productivity index (PoSPi) integrating key reproductive parameters was formulated to compare population larval supply over time. C. foliascens is reproductive all year round, gonochoric and viviparous, with larvae developing asynchronously throughout the mesohyl. The influence of environmental parameters relevant to C. foliascens reproduction (i.e. sea surface temperature [SST], photoperiod and rainfall) was also examined, and SST found to have the most significant effect on phenology. The reproduction of C. foliascens exhibited annual mono-cyclic patterns closely resembling SST fluctuations. Reproductive output was depressed at SST less than 23°C, and increased at temperatures above 23°C. Peak sperm release occurred at temperatures above 25°C, while peak larval release occurred during the annual temperature maxima (> 28°C). A two-fold increase in the maximum production of larvae (PoSPi) in C. foliascens occurred in the second reproductive cycle, following a depressed PoSPi in the first cycle. This reduction in PoSPi in the first reproductive cycle was associated with increases of SST by 1.25°C and rainfall by 380%, coinciding with one of the strongest La Niña events on record. Although C. foliascens has been reported from the intertidal to the mesophotic, the distribution of C. foliascens at inshore reefs of the GBR is restricted to the intertidal with no individuals evident in adjacent subtidal habitats. The potential influence of substrate limitation, and larval pre-settlement and settlement behaviour on adult population distribution were investigated (Chapter 4). The abundance of C. foliascens and substrate availability was first quantified to investigate the influence of substrate limitation on adult distribution. Pre-settlement processes of larval spawning, swimming speeds, phototaxis, vertical migration, and settlement to intertidal and subtidal substrate cues were also quantified. Notably, suitable settlement substrate (coral rubble) was not limiting in subtidal habitats. C. foliascens released up to 765 brooded larvae sponge⁻¹ day⁻¹ during the day, with larvae (80 % ± 5.77) being negatively phototactic and migrating to the bottom within 40 minutes from release. Subsequently, larvae (up to 58.67 % ± 2.91) migrated to the surface after the loss of the daylight cue (nightfall), and after 34 h post-release > 98.67 % (± 0.67) of larvae had adopted a benthic habit regardless of light conditions. Intertidal and subtidal biofilms initiated similar settlement responses, inducing faster (as early 6h post-release) and more successful metamorphosis (> 60 %) than unconditioned surfaces. C. foliascens has a high supply of larvae and larval behaviours that support recruitment to the subtidal. The absence of C. foliascens in subtidal habitats at inshore reefs is therefore proposed to be a consequence of post-settlement mortalities. Lastly, the effect of temperature, photoperiod, rainfall and habitat on post-settlement mortality, growth, asexual reproduction (fission), larval production and recruitment were assessed over 24 months at two locations characterized by distinct hydrodynamics (wave height, Chapter 5). Location-specific differences in growth, body size and fecundity for C. foliascens occurred and are attributed to water movement, with a higher wave height range corresponding to higher abundances of larger, more reproductive individuals. The positive effects of hydrodynamics on growth and the production of larvae also translated to higher levels of recruitment highlighting a potential stock-recruitment relationship in this species. C. foliascens showed no evidence of fission, and exhibited fluctuating growth trajectories in all size classes. Decreasing variability in growth corresponded with increasing size, reflecting growth trajectories for species with indeterminate growth. These results highlight the important role of habitat for post-settlement processes, production of larvae and recruitment in sessile invertebrate species with limited larval dispersal, such as brooding sponges. In summary, C. foliascens at intertidal sites on inshore reefs of the central GBR is highly reproductive, producing larvae with behaviours suggestive of low dispersal potential and endogenous recruitment. Strict intertidal distributions are linked to post-settlement processes that limit the occurrence of sponges in adjacent subtidal habitats. These subtidal regions are less than a hundred metres away from areas of high C. foliascens abundance. The self-recruiting and self-regulating nature of intertidal C. foliascens combined with the stock-recruitment relationship supported by this study, indicate the critical impact of habitat characteristics and external environmental parameters in determining recruitment levels and population structure for C. foliascens. Taken as a whole, this thesis represents a collection of systematic investigations that are critical for a comprehensive understanding of population dynamics in sponges, and sessile benthic invertebrates in general. The integration of datasets from all life history stages from reproduction to larval recruitment, and growth and mortality in juvenile and adult life stages, allows for realistic assessments of the vulnerability of sponge populations to predicted environmental and climatic changes