Quantitative fish survey of the submarine canyons of the iSimangaliso Wetland Park
Master Thesis
2015
Permanent link to this Item
Authors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher
University of Cape Town
Department
Faculty
License
Series
Abstract
There have been no quantitative studies of fish species associated with the submarine canyons of the north east coast of South Africa. These canyons offer protection to coelacanths Latimeria chalumnae, fish of commercial importance, and a number of protected and endemic species. A fish survey was conducted by remotely operated vehicle (ROV) that captured video footage at depths between 60 m and 160 m. Seven canyons (Island Rock, South Island Rock, Wright, Jesser, Diepgat, Leadsman and Chaka) and one deep reef (Diepgat Deep Reef) spread along 78 km of shelf break were surveyed. Fish were identified and counted in 1143 30-second video segments. Patterns in diversity and abundance were investigated with respect to canyon, depth, and habitat type. The survey found 52 fish species from 23 families. The families Serranidae and Sparidae were well represented. The first submarine images of the critically endangered Seventy†Four seabream are particularly noteworthy. Only three Myliobatidae species represented the Chondrichthyans. Variation in Shannon-Wiener species diversity (H') was influenced by the diversity of habitat (Kruskall†Wallis p<0,0001) and depths (Kruskall†Wallis p<0,0001). Mann†Whitney post hoc tests showed cave to have a higher H' than sandy plain (p<0,001) and wall (p<0,001). H' for margin was greater than sandy plain (p=0,001) and wall (p<0,001). H' for rock outcrop was greater than sandy plain (p=0,006). Fish diversity increased with increasing depth until 90m, thereafter diversity decreased with subsequent depths. Habitat (Permanova p=0,0031) had the strongest influence on fish community composition. No north†south separation in terms of fish diversity (H') among canyons was detected. The results were consistent with similar studies. All canyons were adequately sampled as the rate of discovery of additional species per sample was ≤1%. A minimum of 80 30-second samples per canyon is recommended to survey fish. The survey methods employed during this study are recommended for surveying deep reef fish to allow for meaningful comparative studies.
Description
Keywords
Reference:
Geldenhuys, D. 2015. Quantitative fish survey of the submarine canyons of the iSimangaliso Wetland Park. University of Cape Town.