Evaluation of the effectiveness of three underwater reef fish monitoring methods in Fiji

Locally based and volunteer-based monitoring approaches have recently received greater attention as a cost-effective way to collect data on the environment and/or involve stakeholders in management. The present study compared three monitoring methods based on an underwater visual census (UVC) of reef fish in a customary area in Navakavu (Viti Levu Island, Fiji). Two non-scientific approaches involved (1) snorkellers from the village using vernacular names for five local focal fish taxa and (2) non-resident volunteers using the Reef Check target list of fish for the Indo-Pacific (nine focal taxa). These approaches were cross-checked with a comprehensive scientific survey (all fish recorded at the species level across 12 families). The three methods were compared in terms of accuracy, precision, and ability to highlight changes in fish abundance between the community-managed no-take zone (NTZ) and the fished area. We found that scientific and volunteer-based data were consistent in terms of means and variance of abundance estimates but that Reef Check focal taxa would not be suitable for monitoring the Navakavu marine area. On the contrary, the locally selected target taxa were appropriate but overestimates and imprecision were noticed in the fish counts made by the community members. Differences in abundance of two major target taxa (Lethrinidae and Mullidae) between the NTZ and the harvest area were not validated by the scientific observations. Calibration of community-based UVC is thus needed especially if information from the local biological monitoring program is used to guide management actions.