Spatial Ecology of Reef Fish in Backreef and Coral Reef Habitats.
No Thumbnail Available
Files
Date
2006-05-08
Authors
Journal Title
Series/Report No.
Journal ISSN
Volume Title
Publisher
Abstract
Understanding the spatial population dynamics of organisms is essential for effective ecological conservation and management. Landscape ecological theories consider the habitat composition and structure of landscapes at multiple spatial scales as drivers for population and community patterns. Yet, many of these theories have evolved through study of terrestrial systems, and a formal, predictive marine spatial ecology is needed to account for the unique characteristics of marine species and their environments. Marine systems present several conceptual challenges to established spatial ecological theories, because (1) pelagic larval dispersal strategies and life-histories incorporating ontogenetic habitat shifts make it difficult to identify population boundaries, (2) the decoupling of adult, larval, and juvenile populations means that a population's demographic rates are not associated with a single habitat patch, but rather a heterogeneous mosaic of habitats, and (3) hydrodynamic processes complicate predictions of landscape connectivity.
My research contributes to the development of marine spatial ecology by addressing the effects of local habitat patch characteristics, regional landscape structure, and hydrodynamics upon dispersal and recruitment of marine populations at multiple spatial scales. My focus was on the important shallow, coastal ecosystems that often serve as nursery habitat for many fish and crustacean species, and where habitat and hydrodynamics are likely to both play important roles in organism dispersal and survival. My research included three related studies: (1) a computer simulation model of passive and active dispersal strategies evaluated how organism dispersal behavior and landscape structure interacted to influence dispersal and recruitment success; (2) a regression analyses tested the efficacy of both traditional (e.g. patch area, habitat diversity) and marine-specific (e.g. proximity to hydrodynamic corridors, habitat volume) landscape characteristics at multiple scales as predictors of juvenile fish population and community patterns at local scales (1 km) in Florida Keys mangroves; and (3) a retrospective analyses of island-wide, Caribbean habitat and fish population databases tested the degree to which adult reef fish population abundance and community structure correlate with nursery habitat variables.
I found that: (1) an organism's behavioral response to currents (whether to walk, swim, or drift) significantly influences organism recruitment success in heterogeneous landscapes and how organisms respond to landscape change; (2) the inclusion of marine-specific landscape characteristics (e.g. distance to nearest channel) greatly improves the predictive power of statistical models describing some fish population and community trends at local scales; and (3) the influence of nursery habitat on reef fish populations appears to be a localized effect that is minimized in regional, island-wide fish abundance datasets, such that regional-scale fish surveys likely highlight island-wide differences in reef structural complexity, rather than dependence on putative nursery habitats.
My research indicates that the spatial ecology of marine populations varies across spatial scales and among species. The behavioral strategies of dispersing individuals, as well as the patch- (100s meters) and landscape-scale (1 to 10s kilometers) characteristics of the habitat encountered by dispersing individuals, influence individuals' recruitment success and assemblage composition over larger scales. Even reef fish species in the same family or behavioral guild (e.g. feeding guild) have been observed to respond to distinct landscape and patch scale habitat features. These observations support arguments in favor of multispecies, multi-scale, organism-based approaches for reef fish conservation and management, rather than simpler habitat patch based approaches. Contour maps of species probability functions, in particular, could make excellent use of multivariate regression models to map predicted species distribution to (1) aid in the identification of potential population density or diversity hot-spots for each species, (2) evaluate the relative benefits of alternative reserve designs to each species, and (3) identify gradients in habitat quality that would aid in the development of stratified sampling design.
Description
Keywords
reef structural complexity, seascape ecology, cellular automata dispersal model, juvenile fish density, spatial scaling, passive-active dispersal gradient
Citation
Degree
PhD
Discipline
Marine, Earth and Atmospheric Sciences
