Search for: All records

Abstract Biodiversity collections in the United States hold over a billion specimens and are essential to understanding the history of life on Earth, as well as patterns of biodiversity in response to environmental change. Each specimen is linked by metadata to an organism's name and the place and time of its collection. Extensive data have been collected on Earth's geology, hydrology, climate, and organisms—past and present—but the data remain largely fragmented. We report in the present article on community discussions to develop a roadmap and identify action items for the Building an Integrated, Open, Findable, Accessible, Interoperable, and Reusable (BIOFAIR) Data Network, directly linking the various types of biological and environmental data. The roadmap is organized into five themes: stocktaking and gap analysis, technological capacity building, best practices, education and training, and community building. Together, these themes chart a path from initial resource inventories and skill building to infrastructure development, cross‑disciplinary collaboration, and the establishment of FAIR‑compliant workflows and governance. 

Abstract 1. The Poweshiek skipperling [Oarisma poweshiek(Parker, 1870; Lepidoptera: Hesperiidae)] is a federally endangered butterfly that was historically common in prairies of the upper Midwestern United States and Southern Manitoba, Canada. Rapid declines over the last 20 years have reduced the population numbers to four verified extant sites. The causes of Poweshiek skipperling decline are unknown. 2. We aggregated all known Poweshiek skipperling occurrence records to examine the spatiotemporal patterns of Poweshiek skipperling decline. Ecological niche models were developed for five time frames (1985, 1990, 1995, 2000 and 2005) and three spatial extents (eastern occupied range, western occupied range and total occupied range). We used a backward elimination method to investigate the effects of climate and land use on the ecological niche of Poweshiek skipperling. 3. Predictors of occurrence changed over time and across the geographical extent of Poweshiek skipperling. Land use covariates were retained in east models. In the west and total extent, climate variables contributed the most to model predictive power for the 1985, 1990 and 1995 models; land use variables contributed the most to model predictive power in the 2000 and 2005 models. 4. During the rapid decline in Poweshiek skipperling population numbers occurring at the turn of the century, probability of Poweshiek skipperling presence was being driven by proportion of natural land cover and distance to nearest grassland/wetland. Our results suggest that these land use variables are important landscape‐level variables to consider when developing risk assessments of extant populations and potential reintroduction sites.