Dispersal of coral larvae: a modelling perspective on its determinants and implications
Blanco-Martin, Bernardo (2006) Dispersal of coral larvae: a modelling perspective on its determinants and implications. PhD thesis, James Cook University.
|PDF (Thesis front) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader|
|PDF (Thesis whole) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader|
In this thesis, I investigate the mechanisms and implications of coral larvae dispersal using spatially explicit and spatially realistic models. Chapter 1 presents an introduction to the relevant theoretical background and a thesis outline.
Chapter 2 investigates the effects of different combinations of larval type (brooders vs. spawners) and current pattern (non-directional flow vs. strongly directional flow) on settlement, connectivity among reefs by larvae, and evenness and diversity of sources of the larval input to reefs. Despite their simplicity, the models revealed complex dynamics, and significant differences in results were found for different larval types, hydrodynamics conditions, and individual reefs. Settlement was higher for brooders than spawners, because their larvae suffered lower mortality during shorter precompetent and competent periods. Increasing advection in the currents increased nonlocal settlement, however connectivity (the number of reefs connected) was not necessarily increased. The traditional view of connectivity being higher in spawners than brooders was often not supported when the currents were strong. Reefs or reefal systems with high evolutionary-scale connectivity did not necessarily have high ecological-scale connectivity. Moreover, different current patterns produced different effects on each larval type. These differences were quantitative for settlement and connectivity, but qualitative for the diversity of the settling larvae (an increase in advection increased diversity for brooders and decreased it for spawners).
Chapter 3 examines the effect of water retention at reefs on the settlement, connectivity, and evenness and diversity of sources of the larval input of different larval types under varying current patterns. Of the common predictions about the effects of an increase in the level of water retention at reefs only the increases in total and local settlement were fully supported. The other predictions (reduction in non-local settlement, connectivity, evenness, and diversity) were supported, but only at medium and high levels of water retention. At low retention levels and when the currents had a steady and strongly directional flow (the most common current conditions) the predictions were not met in many cases. In the absence of retention, settlement, connectivity, and diversity could be very low, especially for larvae with long precompetent and competent periods (i.e. spawners). It was concluded that some level of larval retention at the reefs must exist, at least for spawners, in order to obtain settlement and connectivity values comparable to those found by ecological and genetic studies. Moreover, only very small levels of retention were required to produce dramatic increases in settlement rate, and these levels did not substantially reduce connectivity or diversity. In fact, these levels of retention could increase the connectivity of spawners when the currents had a strongly directional flow, as their larvae could become trapped in the circulation of reefs other than their natal. Notably, the estimated level of retention for a typical reef in the central Great Barrier Reef (GBR) provided the most favourable conditions for spawning corals in the models (which included reef densities and current speeds similar to those found in this region); this retention level maximised evolutionary-scale connectivity, yielded a high ecological-scale connectivity, and vastly increased settlement in spawning corals.
Chapter 4 investigates settlement and the mechanisms determining it on the GBR. The models reproduced many patterns of settlement and connectivity previously described by empirical studies, including the latitudinal settlement patterns of spawners and their proportion in the total settlement. It was concluded that a significant part of the large spatial and temporal variation in recruitment observed in the field might be related to variation in the factors included in the models, rather than to purely stochastic variation as often assumed. Therefore, settlement may be predictable to a certain extent if enough physical and biological information is available at the relevant spatial and temporal scales. Discrepancies between the models and field observations most likely reflect the importance of factors not included in the models, such as the decrease in temperature with latitude. Although all the models for brooders predicted a decline in settlement from the central to the southernmost GBR, none of them accurately reproduced the recruitment patterns of the northern GBR (all underestimating recruitment in this area). The models that best reproduced the latitudinal patterns of recruitment of spawners included fecundity, indicating the importance of this process in determining regional-scale recruitment. The proportion of spawners in the total settlement was best predicted when the models included retention and adult abundance (fecundity data was not available for brooders). Large differences in settlement and connectivity patterns occurred in space (even among nearby reefs) and among taxa in the models, suggesting the danger of generalising results obtained by empirical studies sampling a small number of reefs and/or species.
Chapter 5 explores the putative relationship between the potential for dispersal provided by the larvae of scleractinian corals and the extent of their geographic ranges the Indo-Pacific (IP). Both the extent of the geographic range and the dispersal variables computed by the models varied substantially among the species in the study. Nevertheless, the potential for dispersal provided by the larvae of coral species was a poor predictor of the size of their geographic ranges. Notably, there was a tentative relationship between estimates of gene flow and some of the variables calculated by the dispersal models (particularly evolutionary-scale connectivity). The dispersal potential of the larvae is an important factor determining the geographic ranges of coral species, but the dispersal hypothesis per se cannot explain the geographic distributions of corals in the IP. The distribution of corals is also influenced by many other factors that mask the effect of the dispersal potential of the larvae in determining the geographic ranges of coral species in the IP.
Chapter 6 investigates the mechanisms by which the patterns of species richness in the IP may have arisen, using a ‘Topological Model’ of coral reef biogeography of this region that included speciation, dispersal, and local extinction processes. In the model dispersal occurred randomly (i.e. exclusively by diffusion) over a spatially realistic representation of the IP. This model produced a good approximation (better than previous models) to the present patterns of coral species richness in the IP. It was concluded that currents might not be the dominant factor shaping the geographic distribution of tropical inshore species as previously suggested, the topology of the habitats of the species being at least as important. In the ‘Topological Model’ the spatial attributes and relationships among habitats were sufficient for the central IP to act simultaneously as a centre of ‘origin’, ‘accumulation’, and ‘refuge’ (without the need of assuming particular properties for this area and/or the currents it receives), consequently generating a centre of species richness. The ‘Topological Model’, alone or in combination with a hydrodynamic model, can be used as a null-model against which the patterns of species richness in the IP can be contrasted.
The spatially explicit and spatially realistic models used in this thesis provided significant insights into the factors determining the dispersal and settlement of larvae and their consequences. These models can be used for guiding the conservation and management of marine populations; they should, however, be adapted to each specific problem, and contain adequate information about the physical conditions and organisms in question.
|Item Type:||Thesis (PhD)|
|Keywords:||coral, larval dispersal, Great Barrier Reef, GBR, reefs, spawning corals, brooders, spawners, currents, dispersal, settlement, connectivity, evenness, diversity, interaction, larval input, water retention, fecundity, scleractinian corals, Indo-Pacific ocean, modeling, modelling, competency, retention, spatial patterns|
|FoR Codes:||01 MATHEMATICAL SCIENCES > 0102 Applied Mathematics > 010202 Biological Mathematics @ 0%|
06 BIOLOGICAL SCIENCES > 0602 Ecology > 060205 Marine and Estuarine Ecology (incl Marine Ichthyology) @ 0%
06 BIOLOGICAL SCIENCES > 0603 Evolutionary Biology > 060302 Biogeography and Phylogeography @ 0%
|Deposited On:||13 Jan 2009 12:25|
|Last Modified:||02 Nov 2012 08:57|
Last 12 Months: 114
Repository Staff Only: item control page