Fig 1.
A map of all platforms surveyed by ROV during this study.
Platform code and position shown. Point of first sighting (2007) shown as a triangle (GI-93C).
Table 1.
A list of all platforms surveyed by ROV during this study.
Fig 2.
Percent wins in competition for space between Tubastraea micranthus and all other sessile epifauna pooled, by platform.
Percent competitive success plus 95% confidence limits shown. The competition variable is defined as the number of competitive successes over the total number of interactions within a taxonomic group. An overgrowth frequency of significantly >50% was considered to be an indicator of competitive success. A total of 15 platforms were surveyed. Those platforms possessing a number of interactions sufficient for quantitative analysis were included in the study. Data tested for significant variation from 1:1 ratio of competitive successes to losses. Range of number of interactions per platform (n): 18–361. Overall competitive success was significantly higher than the expected 50% level over all platforms (p < 0.01, Fisher’s Exact Test). Highly variable win frequencies between platforms (p < 0.001, Goodness of Fit Test, G-statistic). No significant sub-sets of platforms.
Fig 3.
Percent success in competition for space between Tubastraea coccinea and all other sessile epifauna pooled, presented by platform.
Mean plus 95% confidence limits shown. Range of number of interactions per platform (n): 217–1,283. See Fig 2 legend for additional details. Overall competitive success significantly higher than 50% (p < 0.001, Fisher’s Exact Test). Also, almost all platforms exhibited a competitive win frequency significantly higher than 50% (p < 0.01–0.001, Goodness of Fit Test, G-statistic), except ST-81A and ST-185B, which exhibited no significant difference from the 50% competitive success level (p > 0.05).
Fig 4.
Percent success in competition for space between Tubastraea micranthus vs. T. coccinea.
Here, the former species is the target organism. Presented by platform. Mean plus 95% confidence limits shown. See Fig 2 legend for additional details. No significant difference in overall competitive success (p > 0.05, Fisher’s Exact Test).
Fig 5.
Percent of successes in competition for space between Tubastraea coccinea vs. T. micranthus.
Here, T. coccinea was the target organism. Data presented by platform. Mean plus 95% confidence limits shown. Range of number of interactions per platform (n): 6–155. See Fig 2 legend for additional details. Overall competitive success significantly greater than 50% (p < 0.001, Fisher’s Exact Test). Significantly higher frequency of competitive successes on GI-93C than on other platforms (p < 0.001, Goodness of Fit Test, G-statistic) where competitive success frequency did not differ significantly from a 50% competitive success level (p>0.05
Fig 6.
Photographs of the six most abundant competitors of space against Tubastraea coccinea on oil/gas platforms in the study region, northern Gulf of Mexico.
Photos are still-captures from an ROV video. a = Xestospongia sp. with commensal zoanthids Parazoanthus catenularis; b = Dictyonella funicularis; c = Haliclona vansoesti; d = Xestospongia carbonaria; e = Mycale carmigropila; f = Phorbas amaranthus.
Fig 7.
Percent of success in competition for space between Tubastraea coccinea vs. Xestospongia sp. (with commensal zoanthids Parazoanthus catenularis), by platform.
Mean plus 95% confidence limits shown. Range of number of interactions per platform (n): 19–74. See Fig 2 legend for additional details. Overall competitive success significantly higher than the expected 50% (p < 0.001, Fisher’s Exact Test). Also, almost all platforms exhibited significantly high frequencies of competitive success (p < 0.05–0.001, Goodness of Fit Test, G-statistic). Only coral populations on MC-311A and GI-115A exhibited non-significant levels of competitive success (< 50% frequency of success, p > 0.05).
Fig 8.
Percent of competitive success in competition for space between Tubastraea coccinea and Dictyonella funicularis, by platform.
Mean plus 95% confidence limits shown. Range of number of interactions per platform (n): 13–83. See Fig 2 legend for additional details. Overall competitive successes significantly higher than 50% (p < 0.001, Fisher’s Exact Test). Highly variable win frequencies between platforms. Platforms exhibiting significant competitive success were GI-90A, MC-311A, GI-116A, MC-109A, SP-87D, and ST-185A (p < 0.01–0.001, Goodness of Fit Test, G-statistic). The remainder exhibited no significant difference with a 50% competitive success level (p > 0.05).
Fig 9.
Percent of successes in competition for space between Tubastraea coccinea and Haliclona vansoesti, by platform.
Mean plus 95% confidence limits shown. Range of number of interactions per platform (n): 16–60. See Fig 2 legend for additional details. No significant difference between overall observed frequency of competitive success, and an expected 1:1 ratio of the two competitors (p > 0.05, Fisher’s Exact Test). Highly variable success frequencies between platforms. T. coccinea populations on MC-311A, GI-115A, GI-116A, ST-185A, SP-89B, SP-87D, and ST-206A exhibited significant competitive success over this sponge (p < 0.05–0.01, Goodness of Fit Test, G-statistic). The GI-94B coral population exhibited significant competitive losses (p < 0.01). GI-93C, GI-90A, and MC-109A showed no significant variance from the expected 50% competitive success level (p >0.05).
Fig 10.
Percent of success in competition for space between Tubastraea coccinea and Xestospongia carbonaria, by platform.
Mean plus 95% confidence limits shown. Range of number of interactions per platform (n): 20–58. See Fig 2 legend for additional details. Overall competitive success significantly higher than 50% (p < 0.001, Fisher’s Exact Test). Highly variable success frequencies between platforms. T. coccinea on GI-93C, GI-90A, MC-311A, GI-115A, GI-116A, ST-206A, and ST-185B all exhibited significantly high competitive success (p < 0.01–001, Goodness of Fit Test, G-statistic). The remainder exhibited competitive success frequencies which did notvary significantly from 50% (p> 0.05).
Fig 11.
Percent success in competition for space between Tubastraea coccinea and Mycale carmigropila, by platform.
Mean plus 95% confidence limits shown. Range of number of interactions per platform (n): 13–150. See Fig 2 legend for additional details. Overall competitive success significantly higher than 50% (p < 0.001, Fisher’s Exact Test, Goodness of Fit Test, G-statistic). T. coccinea populations were highly variable with respect to competitive success for this interaction. T. coccinea on GI-93C, GI-90A, GI-115A, GI-116A, SP-89B, and ST-185A all exhibited significantly high competitive success (p < 0.01–001, Goodness of Fit Test, G-statistic). Populations on the remainder of the platforms exhibited frequency of success not significantly different from 50% (p> 0.05).
Fig 12.
Percent of wins in competition for space between Tubastraea coccinea and Phorbas amaranthus, by platform.
Mean plus 95% confidence limits shown. Range of number of interactions per platform (n): 17–150. See Fig 2 legend for additional details. Overall competitive success significantly higher than the 50% level (p < 0.01, Fisher’s Exact Test). Frequency of competitive success varied significantly between platforms. The following platforms exhibited significantly high competitive success frequencies: GI-93C, GI-90A, GI-116A, MC-109A, and SP89B (p < 0.05–0.01, Goodness of Fit Test, G-statistic). The remainder (most of the platforms) exhibited competitive success frequencies not significantly different from 50% (p > 0.05).