IMIS | Flanders Marine Institute

Flanders Marine Institute

Platform for marine research

In:

IMIS

Publications | Institutes | Persons | Datasets | Projects | Maps
[ report an error in this record ]basket (0): add | show Printer-friendly version

Predictions of 27 Arctic pelagic seabird distributions using public environmental variables, assessed with colony data: a first digital IPY and GBIF open access synthesis platform
Huettmann, F.; Artukhin, Y.; Gilg, O.; Humphries, G. (2011). Predictions of 27 Arctic pelagic seabird distributions using public environmental variables, assessed with colony data: a first digital IPY and GBIF open access synthesis platform. Mar. Biodiv. Spec. Issue 41(1): 141-179
In: Marine Biodiversity. Springer: Berlin. ISSN 1867-1616, more
Peer reviewed article

Available in Authors 

Keywords
    Biodiversity; Colonies; Databases; Distribution; GIS; Marine birds; Marine

Authors  Top 
  • Huettmann, F.
  • Artukhin, Y.
  • Gilg, O.
  • Humphries, G.

Abstract
    We present a first compilation, quantification and summary of 27 seabird species presence data for north of the Arctic circle (>66 degrees latitude North) and the ice-free period (summer). For species names, we use several taxonomically valid online databases [Integrated Taxonomic Information System (ITIS), AviBase, 4 letter species codes of the American Ornithological Union (AOU), The British List 2000, taxonomic serial numbers TSNs, World Register of Marine Species (WORMS) and APHIA ID] allowing for a compatible taxonomic species cross-walk, and subsequent applications, e.g., phylogenies. Based on the data mining and machine learning RandomForest algorithm, and 26 environmental publicly available Geographic Information Systems (GIS) layers, we built 27 predictive seabird models based on public open access data archives such as the Global Biodiversity Information Facility (GBIF), North Pacific Pelagic Seabird Database (NPPSD) and PIROP database (in OBIS-Seamap). Model-prediction scenarios using pseudo-absence and expert-derived absence were run; aspatial and spatial model assessment metrics were applied. Further, we used an additional species model performance metric based on the best publicly available Arctic seabird colony location datasets compiled by the authors using digital and literature sources. The obtained models perform reasonably: from poor (only a few coastal species with low samples) to very high (many pelagic species). In compliance with data policies of the International Polar Year (IPY) and similar initiatives, data and models are documented with FGDC NBII metadata and publicly available online for further improvement, sustainability applications, synergy, and intellectual explorations in times of a global biodiversity, ocean and Arctic crisis.

 Top | Authors