Maps, models, and marine vulnerability: Assessing the community distribution of seabirds at-sea

摘要: Abstract Whether considering the cumulative impact of chronic, small-size oil discharges or accidents associated with marine traffic and offshore gas development, seabirds face a variety threats in environment. Assessing vulnerabilities to maritime hazards requires an understanding their species distribution, means for combining this information across groups. Using at-sea survey data gathered as part regional monitoring program, efficient framework integrating multi-species was developed. Survey incorporated within distance sampling generate bias-corrected seabird densities area over 730,000 km2 size, which were used construct multiple distribution models (SDMs). The structural difficulties sparsely distributed individuals that also occur large, localized concentrations led use three modelling techniques potentially well suited type data: negative binomial, “hurdle”, random forest methods. Predicted abundances combined produce ensemble forecast, met exceeded accuracy predictions from individual models. Multi-species potential sensitivity maps developed identify core areas, confirming general importance physiographic features such shelf break bathymetry. Distribution seasonally influenced, spring winter standing out periods peak importance. When pollution layer derived aerial surveillance, vulnerability highest vicinity major ports (e.g., Halifax Sydney, Nova Scotia). However, map indicated lower but widespread levels oiling risk throughout shelf, presumably persistently high shipping ongoing petroleum extraction exploration. Outside region, migratory connectivity is expected expose wider network further underscores need coordinated routine collection hazard alongside distributional data.