Ecological Archives E096-267-A1
Ewan D. Wakefield, Ian R. Cleasby, Stuart Bearhop, Thomas W. Bodey, Rachel D. Davies, Peter I. Miller, Jason Newton, Stephen C. Votier, and Keith C. Hamer. 2015. Long-term individual foraging site fidelity–why some gannets don’t change their spots. Ecology 96:3058–3074. http://dx.doi.org/10.1890/14-1300.1
Appendix A. Sources and pre-processing of environmental data used in the analysis.
Gannets avoid crossing extensive land masses during foraging trips so we calculated colony distance using the R package gdistance (van Etten 2012) as that following the shortest path avoiding land (Wakefield et al. 2013). To describe bathymetry, we used the ETOPO2 Global Relief 2v2 data set (U.S. Department of Commerce, National Oceanic and Atmospheric Administration, National Geophysical Data Center, 2006) and to calculate seabed gradient we used the Slope tool in ARCGIS 10. The UK Marine Management Organisation supplied summary gridded Vessel Monitoring System data for 2007 – 2010. These comprised total vessel hours fishing by boats >15 m long using mobile gear in each 0.05 ° cell, which we converted to densities (h/km2). We were supplied with weekly and monthly mean NPP and SST by the NERC Earth Observation Data Acquisition and Analysis Service, Plymouth. NPP was estimated using MODIS chlorophyll, photosynthetically available radiation and SST (Smyth et al. 2005). SST was measured using the Advanced Very High Resolution Radiometer (AVHRR). Following (Miller 2009) we estimated the daily intensity and persistence of thermal fronts using AVHRR-SST data, then averaged these estimates by week and month giving a quantity we term front density (FD). The units of front density are degrees Kelvin/pixel distance (~ K km-1) but a number of weighting schemes are applied during its calculation (see (Miller 2009) for details) so we treat FD as a dimensionless relative index of frontal activity. All environmental indices were either supplied on an 8.998 × 10-3 degree (i.e., ~ 1 × 1 km resolution) grid or re-sampled, using bilinear interpolation, to this resolution. In order to improve data spread, prior to analysis we square-root transformed depth; double-square root transformed sea-floor slope, fishing effort and front density; and loge-transformed NPP.
Literature cited
Miller, P. 2009. Composite front maps for improved visibility of dynamic sea-surface features on cloudy SeaWiFS and AVHRR data. Journal of Marine Systems 78:327–336.
Smyth, T. J., G. H. Tilstone, and S. B. Groom. 2005. Integration of radiative transfer into satellite models of ocean primary production. Journal of Geophysical Research-Oceans 110.
van Etten, J. 2012. gdistance: distances and routes on geographical grids. R package version 1.1-2.
Wakefield, E. D., T. W. Bodey, S. Bearhop, J. Blackburn, K. Colhoun, R. Davies, R. G. Dwyer, J. A. Green, D. Grémillet, A. L. Jackson, M. J. Jessopp, A. Kane, R. H. W. Langston, A. Lescroël, S. Murray, M. Le Nuz, S. C. Patrick, C. Péron, L. M. Soanes, S. Wanless, S. C. Votier, and K. C. Hamer. 2013. Space partitioning without territoriality in gannets. Science 341:68–70.