Publication Date



Open access

Degree Type


Degree Name

Doctor of Philosophy (PHD)


Marine Biology and Fisheries (Marine)

Date of Defense


First Committee Member

John W. McManus

Second Committee Member

Diego Lirman

Third Committee Member

Larry E. Brand

Fourth Committee Member

Donald L. DeAngelis

Fifth Committee Member

Peter W. Glynn

Sixth Committee Member

Ligia Collado-Vides


Macroalgae are an important part of the coral reef ecosystem that has largely been overlooked. However, in the past few decades their abundances have increased and this has been attributed to combinations of coral mortality opening up space in the reef, decreased grazing and increased nutrient load in reefs. This dissertation illustrates a novel means of investigating the effect of various growth and disturbance factors on the dynamics of macroalgae at three different levels (individual, population and 3-species community). Macroalgae are modular and clonal organisms that have differing morphologies depending on the environment to which they are exposed. These traits were exploited in order to understand the factors that were acting on the dominant and common macroalgae in the Florida Reef Tract: Halimeda tuna, Halimeda opuntia and Dictyota sp. The agent-based model SPREAD (SPatially-explicit REef Algae Dynamics) was developed to incorporate the key morphogenetic characteristics of clonality and morphological plasticity. It revolves around the iteration of macroalgal module production in response to light, temperature, nutrients, and space availability, while fragmentation is the source for mortality or new individuals. These processes build the individual algae then the population. The model was parameterized through laboratory experiments, existing literature and databases and results were compared to independently collected field data from four study sites in the Florida Keys. SPREAD was run using a large range of light, temperature, nutrient and disturbance (fragmentation without survival) levels and yielded six morphological types for Halimeda tuna, and two each for Halimeda opuntia and Dictyota sp. The model morphological types that matched those measured in two inshore patch reefs (Cheeca Patch and Coral Gardens) and two offshore spur and groove reefs (Little Grecian and French Reef), were formed in conditions that were similar to the environmental (light, nutrient and disturbance) conditions in the field sites. There were also differences between species in the important factors that influenced their morphologies, wherein H. opuntia and Dictyota were more affected by disturbance than growth factors, while H. tuna morphology was affected by both. Allowing for fragmentation with survival in the model resulted in significantly higher population abundances (percent cover and density). The highest abundances were achieved under high fragment survival probabilities and a high disturbance level (but not large fragment sizes). Incorporating fragmentation with survival and simulating the variations in light, nutrients and disturbance between the inshore patch reefs and offshore spur and groove reefs in SPREAD led to comparable abundances of Halimeda in the virtual reef sites. Adding competition for space and light and epiphytism by Dictyota on the two Halimeda species suggests that it can regulate the populations of the three macroalgae. However, comparing model abundances to the field, competition may not be a strong regulating force for H. tuna in all the sites and H. opuntia in the patch reefs. H. opuntia in the offshore reefs is possibly competitively regulated. Although SPREAD was not able to capture the patterns in the population abundance of Dictyota, this points to the potential importance of other morphometrics not captured by the model, a variation in growth curves between reef habitats, or the differential contribution of sexual reproduction.


Benthic Composition; Phase Shift; Coral Reefs; Macroalgae; Agent-based Modeling