An evaluation of the suitability of non-specialist volunteer researchers for coral reef fish surveys. Mafia Island, Tanzania — A case study

Trials are described for testing and validating the use of non-specialist volunteer divers for visual censusing of reef fish populations. Volunteers showed significant improvement between two censuses in identifying 56 species in 30 genera. Thus, after 11 additional dives, the loss of precision compared with an experienced ‘control’ diver was reduced from 13% to 0·6%. In the second census there was no loss in consistency caused by the use of groups of divers rather than single divers. Even in the early stages of training the level of precision was sufficient to detect changes in composition at the trophic level. Coloured canvas strips were used as model ‘fish’ for estimating lengths underwater. After three trials divers were able to estimate lengths with an 80% accuracy in relation to the true distribution.