Hierarchical Bayesian modelling of plant pest invasions with human-mediated dispersal

Flight plan for the future: floatplane pilots and researchers team up to predict invasive species dispersal in Alaska

Aircraft can transport aquatic invasive species (AIS) from urban sources to remote waterbodies, yet little is known about this long-distance pathway. In North America and especially Alaska, aircraft with landing gear for water called floatplanes are used for recreation access to remote, often road-less wilderness destinations. Human-mediated dispersal of AIS is particularly concerning for the conservation of pristine wildlands, yet resource managers are often challenged by limited monitoring and response capacity given the vast areas they manage. We collected pathway data through a survey with floatplane pilots and used a Bayesian hierarchical model to inform early detection in a data-limited situation. The study was motivated by Alaska’s first known AIS, Elodea spp. (Elodea) and its floatplane-related dispersal. For 682 identified floatplane destinations, a Bayesian hierarchical model predicts the chance of flights originating from AIS source locations in freshwater and estimates the expected number of flights from these sources. Model predictions show the potential for broad spread across remote regions currently not known to have Elodea and informed monitoring and early detection efforts. Our result underlines the small window of opportunity for Arctic conservation strategies targeting an AIS free Arctic. We recommend management that focuses on long-distance connectivity, keeping urban sources free of AIS. We discuss applicability of the approach for other data-limited situations supporting data-informed AIS management responses.