A workflow to gather, clean, and standardize species occurrence records at scale for conservation applications

Biodiversity declines and pressing threats to wildlife necessitate data-driven approaches for conservation and management planning. Natural resource agencies, conservation practitioners, and researchers increasingly leverage digital species occurrence records. However, we still lack user-friendly and flexible tools to aggregate, clean, and standardize such data. These tools are particularly needed for speciose groups like invertebrates and plants whose natural histories are poorly or unevenly known. To provide such a tool and overcome limitations of existing options, we developed a workflow in R that encompasses the entire process of aggregating, cleaning, and standardizing occurrence records within one free software platform. It additionally can return information useful for assessing species' conservation status. Key features include user friendliness, flexibility in data sources and filtering, handling large volumes of data or multiple species, and high transparency. We demonstrate workflow functionality focusing on Texas native bees. We combined bee data from the Global Biodiversity Information Facility (GBIF) and several smaller datasets, then cleaned and standardized the combined dataset including removing duplicates. Processing 145,870 bee records took about an hour on a standard-configuration government computer. We used the resulting records in three case studies that demonstrated various uses conservation practitioners commonly implement. Case study 1 compiled a species list for Denton County, the second described phenology of bumble bees ( Bombus genus) in the Dallas-Fort Worth Metroplex, and the third assessed conservation status of Bombus pensylvanicus . Our workflow greatly facilitates addressing a range of questions at various spatial and temporal scales from one vetted dataset that are tailored to the needs of a specific research objective, management question, municipality, or audience.