Inventorying life: New standards and approaches for re-using and integrating taxonomic inventory data

Background/Question/Methods: Taxonomic inventories are routinely performed and serve as a critical means to assess local to regional biodiversity. However, reporting standards that would allow these inventories to be re-used, compared to one another, and further integrated with other sources of biodiversity data are lacking, impeding their broad utility. Here we provide a draft standard for taxonomic inventories, dubbed Humboldt Core, based on the outputs of a community input process and significant effort to refine initial outputs based on extensive testing of the standard against published inventory descriptions. We first describe different types of inventories and inventory processes, distinguishing between regional-scale species area checklists, and typically finer spatial and temporal grain surveys. We then provide an initial list of terms usable to describe all inventory types comprehensively, along with application profiles that utilize a smaller subset of terms and apply to either regional checklists or local-scale surveys

Results/Conclusions: We provide a very quick overview of terms from Humboldt Core needed to capture geospatial, temporal, taxonomic and habitat scope along with methodological descriptors related to assessment of sampling effort and inventory completeness. For the area species checklists application profile, we describe as well novel aspects such as effort in compiling previous data resources such as expert knowledge, museum specimen data, and prior literature, along with a method for assigning initial completeness metrics. We next describe how Humboldt Core has been successfully applied to a major effort to ingest and provision inventory data and metadata into the Map of Life project.  A key output of the metadata ingest is an initial assessment of reporting quality and completeness drawn from 129 published inventories of terrestrial vertebrates, which is only possible given the standardization of outputs afforded by Humboldt Core.  Our preliminary results show that only a minority of pubished area species checklists have sampled focal taxonomic groups completely, but many do utilize multiple inventory processes, including compiled (as opposed to primary-collected) data. We close by discussing next steps in further development, and integration into existing standards and ontologies, of this critical new standard for biodiversity science.