The development and trial of a method for surveying deep rock wall communities in Doubtful Sound using a Remote Operated Vehicle (ROV).
Major, Robert Neil
Rock wall communities and structuring processes in Doubtful Sound have been widely studied. However, due to the limitations of SCUBA diving this research has been limited to 30 m. This study attempts to further the understanding of the rock wall communities through the quantitative use of a remote operated vehicle (ROV) to observe these communities below the depths of SCUBA divers. The study developed a ROV methodology to quantitatively sample the rock wall communities of fiords. The developed ROV methodology was compared to the conventional methodology of digital photoquadrats taken by SCUBA divers at Tricky Cove in Doubtful Sound. The ROV methodology was observed to successfully sample the patterns of species richness and community composition across depth, however; the ROV underestimated the alpha species richness of the site by 45.5%, the point S and H’ of the samples by 33.9% and 22.3% respectively. The variations between the ROV and photoquadrat survey were concluded to be caused by the lower resolution of the ROV camera which missed small cryptic species (< 5cm in size). The ROV was deployed to a maximum water depth of 100 m at four sites, Deep Cove, Mid C4, Tricky Cove and Thompson Sound, across an inner to outer fiord gradient in Doubtful Sound. Variations in taxonomic abundance and community composition occurred across both vertical and horizontal spatial scales. Across sites, the surveys separated the communities into shallow and deep assemblages at ≈ 40 m. Shallow samples had a higher level of similarity among sites to samples from similar depth brackets, while deep samples had a higher level of similarity to other deep samples from the same site. Shallow samples followed an inner to outer fiord gradient of increasing taxonomic diversity and abundance, while deep samples did not have any along fiord gradient. These results were used to create a basic zonation profile of Doubtful Sound consisting of three zones; (1) a shallow low diversity zone from 0 – 8 m, characterised by Mytilus edulis galloprovincialis and Ulva spp.; (2) a shallow high diversity zone from 8 – 14 m, characterised by large populations of erect and encrusting sponges, encrusting ascidians, brachiopods and Antipathes fiordensis; (3) a deep barren zone from 40 – 100 m, characterised by Caryophyllia profunda and small populations of encrusting sponges and ascidians which were present under small overhangs and within crevices. The shallow communities were proposed to be structured by the LSL and biotic interactions, while the deep communities were proposed to be structured by sedimentation and basin processes. This research demonstrates the ability of an ROV to survey the patterns of diversity and abundance for rock wall communities in fiords, separating the communities into shallow and deep communities at 40 m. However, due to the low camera resolution of current ROVs, using an appropriate scale for surveys has been shown to be important.
Advisor: Lamare, Miles
Degree Name: Master of Science
Degree Discipline: Marine Science
Publisher: University of Otago
Keywords: Remote Operated Vehicle; Doubtful Sound; Fiordland; Marine Ecology
Research Type: Thesis