Hypothesis-based Restoration Study For Mitigation of a Damaged SE Florida Coral Reef: A Work in Progress

T.P. Quinn1, E.A. Glynn1, R.E. Dodge1, K. Banks2, L. Fisher2, R.E. Spieler1

1National Coral Reef Institute, Nova Southeastern University Oceanographic Center (NSUOC),
8000 North Ocean Drive, Dania Beach, FL 33004
2Broward County Department of Planning and Environmental Protection, 218 S.W. 1st Avenue, Ft. Lauderdale, FL 33301


The United States Submarine Memphis (Figure 1) ran aground in approximately 10 meter depth on a coral reef off southeast Florida (Figure 2) February 25, 1993. Extensive physical damage to the reef substrate and injury to the coral community were attributed to the initial grounding and subsequent attempts to free the submarine from the impacted reef (Figures 3 and 4). The impact of the grounding was assessed, and the area of damage was determined through field and photographic studies.

An impacted area of 2,310 m2 was assessed with 1,205 m2 having been totally destroyed (Figures 3 and 4). In 1997, the State of Florida was awarded a settlement of $750,000 by the Federal government for environmental damages caused by the submarine grounding. A plan to perform hypothesis testing of restoration techniques was developed and initiated.

Using artificial reefs as experimental platforms, we are examining three restoration strategies: 1) the potential of enhancing coral recruitment through the use of coral larval attractants, 2) the effect of reef structure on the associated fish assemblages, and 3) the interaction between fish assemblages and coral recruitment and survival.


One hundred and sixty small artificial reef modules (Reef Balls™) were deployed in 11 m of water on a sand flat between reef tracks adjacent to the U.S.S. Memphis grounding site (Figure 5). The Reef Balls were organized into 40, 4-module reef units (quad) in a square configuration having approximately 4 m sides (Figure 6). The separation of individual Reef Balls (2 m) was judged sufficient to avoid interaction effects between Reef Balls in terms of coral settlement, but close enough for the 4 balls to function as a single reef unit in terms of fish recruitment. Each quad was located a minimum of 30 m from any hardbottom.

Coral Recruitment:

Settlement plates on each Reef Ball (Figure 7) are being used to test hypotheses on enhancing coral recruitment through the use of larval attractants. The settlement plates attached to each Reef Ball are treated with a potential attractant (iron, CaCO3, coral transplants) and compared with control plates (no attractant). Coral transplants are 4” cores drilled from large donor colonies (Figures 8 and 9). Eighty coral cores have been transplanted onto the Reef Ball modules (forty cores of each of two different species). Control corals occurring on the natural reef, and of comparable size to the donor corals, are being monitored for comparison of growth and mortality.

Larval Attractants:

Each individual Reef Ball in a quad will incorporate one of four different attractants on the settlement plates:

Coral Transplantation and Monitoring:

At quarterly intervals the donor corals, coral transplants, and control corals are being visually assessed to provide information on individual colony health, growth, and mortality.

Fish Recruitment:

The 40 quads are divided into 4 different levels of structural complexity (refuge) to test the hypothesis that multiple refuge size and the resultant diverse fish assemblages may affect coral recruitment, survival, and growth. One set of 10 quads has the void space of all the Reef Balls filled with large refuge structure (Figure 10). One set has the void spaces of all filled with small refuge structure (Figure 11). Another set is mixed having one Reef Ball empty, one with large refuge, and the last two with small refuge. The final set has the void space of all the Reef Balls empty. The assemblage of fishes (species, number, and size) associated with each quad is being recorded every three months by visual census.

Structural Complexity:

Each type of refuge is used for 10 quads.


Artificial reefs are commonly used to provide structure to damaged reef areas. This project has been designed to use artificial reefs to not only mitigate for lost reef structure but to provide experimental platforms to examine several restoration strategies. The examination of these strategies will aid in making reef restoration decisions that involve:

Back to NCRI Projects Page