Published online by Cambridge University Press: 16 October 2008
The baited remote underwater video technique (BRUVS) was used to compare fish assemblages at three sites with different habitat characteristics: sandy bank (SB), rock reef (RR) and degraded coral reef (DCR) in Port Essington, Garig Gunak Barlu National Park, Northern Territory, Australia. The Carangidae family dominated, representing 35% of all recorded fish. The highest species number was recorded at RR followed by DCR and SB. The highest total fish number was recorded at DCR followed by SB and RR. Fish assemblages from all three sites are clearly different at high confidence level, but still overlapping and higher overlapping was found between SB and RR. Fish assemblages at all three studied sites contained some coral-associated species, fish fauna was less rich compared to coral reefs in the mouth of the bay and fish assemblages were noticeably different from typical coral fish communities. None of the fish assemblages at studied sites were presented by the ‘pure’, single habitat-associated association of species—at all three sites fish communities were a mixture of fish with various habitat preferences. Reef-associated species have a larger proportion at DCR and at RR their habitat complexity was higher; still soft-bottom habitat fish make substantial fractions in fish assemblages at both DCR and RR. Similarly, fish usually identified as ‘reef associated’ were found in notable proportions on SB. Fish with wide distribution and low selectivity in habitat preferences comprised a significant part in fish assemblages at all studied sites. In this study the BRUVS technique worked well in the area where diving visual surveys were impossible to implement because of high water turbidity. Another advantage of this method is the non-impact nature of visual survey; they can be used in long-term monitoring of fish assemblages at reference sites.