Building: Computer Science
Room: Computer Science 3
Date: 2016-12-06 03:15 PM – 03:30 PM
Last modified: 2016-10-16
Abstract
Citizen science monitoring programs have collected data on presence, abundance and distribution of butterfly species (Papilionoidea and Hesperoidea) across North America on a regular basis; some for up to forty years. These observations are fundamental for analyzing trends in butterfly populations over time and space in response to environmental changes, and are all the more valuable because butterflies, with differing requirements for larval and adult stages, are particularly good bioindicators of ecosystem disturbance. In the last 15 years, citizen scientist enthusiasm to participate in butterfly monitoring has sparked a tremendous expansion of new survey projects that would otherwise not be feasible. The North American Butterfly Monitoring Network (http://www.thebutterflynetwork.org/) tracks 30+ independent programs ranging in scope from local to continental, all which recruit local citizen scientists to collect field observations. No global naming authority yet exists for butterflies, so the programs operate using a taxon list of their own choosing, which may or may not derive from a published, authoritative list. To look at broad-scale patterns, data must be integrated across independent monitoring projects, but species and subspecies-level name conflicts significantly complicate integration of data among projects, and in some cases integration is impossible. To resolve as many issues as possible, we developed a data structure to act as a bridge in interpreting nomenclatural discrepancies. We aligned a cumulative total of 3201 species names and 3282 subspecies names from the three most recently published North American butterfly taxonomic checklists: NABA 2nd edition (2001), Opler and Warren (2003), and Pelham (2014) and North American species from the global Integrated Taxonomic Information Service (ITIS) to characterize and resolve all discrepancies. Pair-wise agreement between these base lists ranged from 78% to 95%. We worked with 10 programs to identify which of these base taxonomies their list most closely resembled and to record any name deviations from that base. The project taxon lists ranged in size from 80 to 244 taxa. None of the lists matched a base list exactly; the highest number of deviations between a project list and its base list was 22. Most deviations were due to generic-level disagreements. Our alignment unambiguously relates any name from any participating survey group to the equivalently defined taxonomic entity on any other participating group’s list. Defining comparable relationships between authorities allows essential cross-talk among the multitude of established and newly developing research agendas monitoring the dynamic occurrences of North American butterflies.