Test of assumption

A comparison of overall substratum composition between the two sites can be used to test the assumption that control and impact sites are similar. In figure 5 the percent composition of both living coral (green) and non-living bottom types (blue) are shown for comparison at both sites. Both areas were characterized by a predominance of lobe coral (Porites lobata), and a mixture of coral rubble, coral boulders, and patches of coralline algae. At the impact site finger coral (Porites compressa), false brain coral (Pavona varians), and ridge coral (Pavona duerdeni) was more common than at the control site.

Overall the percent similarity between sites is 89%, indicating high similarity in the substratum composition at both sites. These results verify that sites are very similar and thus the major differences between them can likely be attributed to the human use differences (more divers at the impact site).

Changes in coral abundance

Both sites had similar mean percent live coral cover with the impact site at 54% and the control at 52% (figure 6). Between the one year survey period, the percent cover declined at both sites. Although the decline was greater at the  impact site (5.3%) relative to the control site (1.9%), neither of these declines were statistical significant (P< 0.05).

Bleached coral

The percent cover of bleached coral tips at sites in 1997 was higher at the impact (3.1%) site relative to the control (1.8%). However, these differences were not statistically significant (two-sample t-test, P< 0.05).

Broken coral

The percent cover of freshly broken coral at sites in 1997 was higher at the impact (7%) site relative to the control (3%). However, these differences were not statistically significant (two-sample t-test, P< 0.05).