An inventory of biodiversity data sources for conservation monitoring

even the organisations managing the data or the websites where data could be located were dif- ficult to identify easily. Categorising (or clustering) data sources was challenging. Clumping data under broad terms such as “state” was too general, and further subdivisions might be useful (perhaps based on taxa or biomes or geography). On the other hand, the CMP classifications for threats and actions were too detailed to make clustering under each individual action or threat a meaning- ful (or sometimes feasible) exercise. Once the new post-2020 biodiversity targets have been finalized to replace the Aichi Targets, we will try to cluster some of the data sources under each new target and its associated indicators (sensu [28]) Categorising sources was further compli- cated by the fact several databases contain multiple data sets (the FAO Fisheries and Aquacul- ture database being a good example) so can also be listed under several data categories (state, pressures/threats or responses/actions). Key actors involved in managing data sources of most use to biodiversity monitoring include BirdLife International, the Secretariat of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES), the UN Food and Agriculture Organi- sation (FAO), the International Union for Conservation of Nature (IUCN), the US National Oceanic and Atmospheric Administration (NOAA), UN Environment World Conservation Monitoring Centre (UNEP-WCMC), University of British Colombia (the Sea Around Us),

Wetlands International, the World Resources Institute (WRI), and the Zoological Society of London (ZSL). This reflects earlier analyses of data sets which noted that a small number of agencies curate most of the data. For example, Joppa et al [24] found six data providers account for more than a fifth of biodiversity threat data sets. Global data sources exist that are useful in providing time-series data of biodiversity mea- sures such as vertebrate population trends, habitat cover trends, species extinction risk, pro- tected area coverage and management effectiveness, fisheries and climate, though the frequency the data sets are updated varies and there are gaps in taxonomic and geographic cov- erage. Overall, across data sources, there was better coverage of some biomes, taxa, threats and responses than others. For example, vertebrates (especially large mammals and birds), forests, Fig 1. Data sources identified. A summary of the 145 data sources identified by data type, and the proportion known to have at least some data easily (i.e. instantly, openly and freely) accessible. https://doi.org/10.1371/journal.pone.0242923.g001 PLOS ONE Biodiversity data sources PLOS ONE | https://doi.org/10.1371/journal.pone.0242923 December 2, 2020 4 / 14