Please use this identifier to cite or link to this item:
The biogeography of Central Pacific coral reef fishes
|Musburger_Craig_r.pdf||Version for non-UH users. Copying/Printing is not permitted||1.64 MB||Adobe PDF||View/Open|
|Musburger_Craig_uh.pdf||Version for UH users||1.64 MB||Adobe PDF||View/Open|
|Title:||The biogeography of Central Pacific coral reef fishes|
|Authors:||Musburger, Craig Andrew|
|Keywords:||coral reef fishes|
|Issue Date:||May 2012|
|Publisher:||[Honolulu] : [University of Hawaii at Manoa], [May 2012]|
|Abstract:||The distribution and abundance of coral reef fishes is undoubtedly affected by a variety of factors. While much previous research into dispersal success of reef fishes has focused on species' ability to survive for long periods in the pelagic larval stage, this alone cannot explain biogeographic patterns. Consequently, a novel approach was developed to examine another potentially important factor contributing to dispersal ability: the number of offspring propagules produced at a source location. A large scale data collection effort was undertaken around islands of the Central and Western Pacific to facilitate testing of this hypothesis. Since direct measurement of larval output on this scale is not possible, a proxy for larval output--adult population density--was used as the predictive variable in models used to identify factors underlying dispersal patterns. In situ visual sampling was conducted over seven years on 38 islands across the Pacific. Over 750 species and 300,000 individual fish were identified and sampled along 700 transects. Analysis was focused on the large biogeographic barrier separating Johnston Atoll and the Hawaiian Archipelago from the rest of the study islands. The regions most likely to be sources of fish larvae reaching the Johnston-Hawaii system were sampled using robust, standardized, quantitative fish census methods. Methodological precautions were taken to ensure that the types of population estimates used in the analyses were robust in space and over time. Novel statistical methods were developed to evaluate the strength of ecological data and two types were identified as suitably robust and reliable; numeric density from belt-transect surveys and presence-absence records from large search area roving diver surveys. Biomass estimates and presence-absence data from belt transects were deemed inadequate or inappropriate for these analyses. Highly significant results of regression models confirmed a relationship between adult population density and long distance dispersal in 3 of 5 fish families included in the analysis. This finding is novel and offers a significant new approach to the study of the biogeography of marine species.|
|Description:||Ph.D. University of Hawaii at Manoa 2012.|
Includes bibliographical references.
|Appears in Collections:||Ph.D. - Zoology|
Items in ScholarSpace are protected by copyright, with all rights reserved, unless otherwise indicated.