An appraisal of two coral reef rapid monitoring manuals for gathering baseline data

There is concern worldwide that many coral reefs are suffering degradation and loss of biodiversity. In response, some rapid coral reef health monitoring manuals have been developed that purport to have wide biogeographical applicability and user friendliness. These manuals however, have been little tested in areas beyond their centers of origin. This study uses two of the manuals (involving line-intercept transects; English et al., 1997) and corallivorous chaetodont behavior (Crosby and Reese, 1996) to do a first, expeditionary assessment on a relatively inaccessible coral reef in the Seychelles, Western Indian Ocean. Future revisions of the manuals should perhaps consider the following points as extra guidance for newcomers to the methodologies. At any one time and site, at least ten, rather than five, transects should be done. Sensitivity of many fishes to line-transect diver activities suggests that the Stationary Visual Census Method is much preferred to line-intercept counts. Line-intercept coral data are well-illustrated by simple rank-abundance curves, flagged with species identifications. Such curves also apply well to specific fish taxa, but not to the general fish assemblage. Statistical correlations between fishes and benthos cannot be done from line-intercept data within transects as they are not independent variables, and this should be made explicit in the manuals. Stationary visual censuses would overcome this problem. Data gathered according to the practical manuals are amenable to multivariate analysis. Clarke and Warwick's (1994) manual and associated software is strongly recommended for follow-up analysis. Obligate corallivorous butterflyfishes clearly have potential for monitoring coral reef health but that the metric requires calibration and cross-referencing with other metrics. Further recommendations on coral reef health monitoring relative to the manuals are given, with emphasis on the importance of clearly defining the biodiversity and conservation questions before monitoring is started.