In recent years it has become evident that biological processes must be analyzed at multiple spatial scales to investigate how results may change across scales, which is often a non-linear pattern. Phenotypic plasticity in scleractinian corals has been studied widely, however the degree to which it exhibits spatial scale dependence has not been explored. The primary objective was to determine if coral morphology varies among traits and across spatial scales. To test for spatial scales of morphological variation, skeletal traits of two corals Pocillopora verrucosa, in French Polynesia, and Porites porites in the Virgin Islands were quantified across multiple spatial scales. The experimental design was constructed to exploit a nested ANOVA, in which all the factors were random and nested by spatial scale. To estimate which skeletal traits of Pocillopora verrucosa contributed most to overall coral shape, principal components analysis (PCA) was used define the multivariate morphology and to collapse the data into fewer variables that explained the majority of the variance. Nine skeletal traits were quantified in 160 colonies of P. verrucosa across two shores. To test for spatial scale dependence, components (PCs) then were tested across spatial scales. My findings show that 2 PC’s, which were associated with 2 different spatial scales, can explain 45% of the variation in morphology. Firstly, corallum dimensions (highest correlation coefficients in PC 1), exhibited significant variation at a spatial scale of 100’s of meters. Second, verrucae and corallite morphology (highest correlation coefficients in PC 2), varied significantly at a smaller scale of 10’s of meters. The majority variance (76%-84%) of both PC’s was unexplained by the tested spatial scales. To test for spatial scale associations of morphological variation in Porites porites, colonies were sampled along ~10 km of the southern coast of St. John. Corallum traits were quantified in 140 colonies, and traits were found to differ between sites. In both species of coral, some skeletal traits were conserved and did not exhibit much variation, and other traits were highly variable. These studies provide evidence that morphological variation is not consistent among skeletal traits and spatial scales for Pocillopora verrucosa or for Porites porites. Following the quantification of morphological traits, a skeletal trait of Pocillopora verrucosa, small projections called verrucae, which occur on branches of P. verrucosa was subjected to controlled flow speeds in a flume to explore the functional significance of this trait. Shear velocities around and above verrucae were experimentally estimated at two flow speeds, to test if verrucae influence rates of mass transfer at the coral surface. My results suggest that verrucae do not influence shear velocities, and thus are unlikely to affect mass transfer on the scale of our experiment. These findings were consistent with previous studies that indicate that small skeletal features may be less important than larger skeletal attributes in determining shear velocities above corals.
Thesis or Dissertation
Department of Biology, CSU Northridge