This is a working manuscript currently accepted to be published in a peer review journal, Ecological Informatics.
Pizarro, V, A. G. Castillo, A. Piñones, and H. Samaniego. 2024. Spatial and Temporal Representation of Marine Fish Occurrences Available Online. Ecological Informatics 79:102403. doi: 10.1016/j.ecoinf.2023.102403.
Despite the 243,000 marine species described by 2022, our knowledge about the oceanic biodiversity is still incomplete. This knowledge gap carries potentially adverse and far-reaching consequences for the preservation of marine ecosystems, particularly in the context of the ongoing human-induced alterations to our biosphere and the rapid progression of climate change and global environmental shifts.
Recently, however, a large number of online repositories have emerged, which catalogue, store and distribute biodiversity information, including taxonomic and species occurrence data. FishBase, the Global Biodiversity Information Facility (GBIF) and the Ocean Biodiversity Information System (OBIS) are part of these publicly available repositories representing a variety of sources that have exploded in number. However, despite the incredible accumulation of biodiversity records, not all the information is actually useful, nor does it represent any new knowledge regarding global species richness patterns.
In this study, we assessed the spatial and temporal representativeness of marine fish records (order Actinopterygii) found in the GBIF and OBIS global repositories. The methodological framework that we developed relies on a series of non-parametric estimators for computing species richness from incidence data. This methodology employs hexagonal grids as sampling units that overlay marine bioregions across the globe.
Using standard ecological and spatial analysis tools, we identify regions that are adequately represented in terms of available records and therefore have more reliable data, as well as regions with few records that do not represent current species richness. We overlap these results with the location of marine protected areas and fishing exploitation zones to understand the anthropogenic effect on marine ichthyofauna. We additionally evaluate hypotheses regarding the taxonomic, geographic, and temporal distribution of information biases to deepen our current understanding of public records of species occurrences worldwide.
Considering that more than 40 years of information was analyzed, the results showed that, on a global scale, the primary data on marine fish available on GBIF and OBIS platforms are still far from being representative and complete. Only 1.14% of the records were useful for our analyses. In addition, we found that the information seems to be biased towards coastal areas, regions close to developed countries, and areas where there is a large fishing activity. Finally, the best represented species and families are those with a small body size, which use shallow habitats and are usually recognized as having commercial or cultural value.