Community-based citizen science projects can support the distributional monitoring of fishes

Effective conservation and fisheries management requires data to capture demographic processes and range limits for each species to maximize population health and productivity. This need is constrained by limited funding and resources, particularly for countries with large land areas and coastlines as well as expansive exclusive economic zones. This imbalance means that monitoring efforts are often focused on targets of commercial and recreational fishing, which results in incomplete distributional records for non-target, small-bodied, and/or cryptic species. Community-based citizen science projects offer one potential alternative for scientists and fisheries managers needing this type of information but lacking sufficient resources to gather it. This study investigated whether data sourced from an online citizen science project (iNaturalist: Australasian Fishes) can assist in the distributional monitoring of a subset of fish species. Given the regional focus of this citizen science project, distributional data in the form of occurrence records for abundant, protected, and threatened fish species as assessed by the International Union for Conservation of Nature in Australia and New Zealand were explored. Data for important commercial and recreational fishery targets in New South Wales were also explored, as a case study of a large jurisdiction with extensive monitoring requirements. The occurrence records for some of these categories of fishes were well represented in the quality-filtered citizen science data set, particularly endemic fishes whose threat status had not yet been assessed and species not currently under any form of management. Despite gaps in coverage between major urban centres, citizen science data for the best represented endemic fishes were qualitatively comparable to the available geographic distributions for these species. We suggest that quality-filtered citizen science data can in fact be used to improve taxonomic representation and the geogra