You are here

Gyrfalcon nest distribution in Alaska based on a predictive GIS model

TitleGyrfalcon nest distribution in Alaska based on a predictive GIS model
Publication TypeJournal Article
Year of Publication2009
AuthorsBooms, TL, Huettmann, F, Schempf, PF
JournalPolar Biology

The gyrfalcon (Falco rusticolus) is an uncommon,little studied circumpolar Arctic bird that faces conservationconcerns. We used 455 historical nest locations,12 environmental abiotic predictor layers, GeographicInformation System (ArcGIS), and TreeNet modeling softwareto create a spatially explicit model predicting gyrfalconbreeding distribution and population size acrossAlaska. The model predicted that 75% of the state had arelative gyrfalcon nest occurrence index value of <20%(where essentially no nests are expected to occur) and 7%of the state had a value of >60%. Areas of high predictedoccurrence were in northern and western Alaska. The mostimportant predictor variable was soil type, followed by subsurfacegeology and vegetation type. Nine environmentalfactors were useful in predicting nest occurrence, indicatingcomplex multivariate habitat relationships exist. We estimatedthe breeding gyrfalcon population in Alaska is546 ยง 180 pairs. The model was 67% accurate at predictingnest occurrence with an area under the curve score of0.76 when assessed with independent data; this is a goodresult when considering its application to the entire state ofAlaska. Prediction accuracy estimates were as high as 97%using 10-fold cross validation of the training data. Themodel helps guide science-based management eVorts intimes of increasing and global pressures for this species andArctic landscapes.

Username Tag: