Žydelis et al. Dynamic habitat models: Using telemetry data to understand fisheries bycatch
Electronic supplementary material ESM 1: Telemetry data of Laysan and black-footed albatrosses and static plots of area utilization
Input telemetry data
Laysan and black-footed albatrosses were equipped with Lotek LTD 2400 light level and temperature-based geolocator archival tags and tracked during the post-breeding seasons (March – November) of 2004-2006 by the Tagging of Pacific Predators (TOPP) research group (Block et al. 2003).
Tagging was conducted at their nesting colony on Tern Island in the atoll of French Frigate Shoals, Northwestern Hawaiian Islands (23°52’ N 166°17’ W) at the end of the breeding season in June-July. Also, a few unsuccessful breeders were tagged in March-May before they left the colony. Tags were recovered when birds returned to the colony for the subsequent breeding season in November. Tracking data from 28 Laysan and 24 black-footed albatrosses were used in this study (Fig. ESM1-1). On average, Laysan albatross tracks lasted for 172 days (range 58-268 days) and black-footed albatross tracks lasted for 165 days (range 139-259 days).
Figure ESM1-1. Location fixes of Laysan and black-footed albatrosses as determined from geolocator archival tags.
Static maps of albatross area utilization
Static representation of Laysan and black-footed albatross distributions obtained applying kernel analysis (following methods described in Hyrenbach et al. 2002, Kappes et al. 2010) indicated different area utilization between years and little overlap with observed long-line sets of the Hawaii-based fleet and recorded albatross bycatch (Fig. ESM1-1). Varying areas of high utilization suggest that identified ‘hot spots’ have been driven by individual bird positions. This is not surprising, considering the vast area of the entire North Pacific used by these albatross species and limited sample of tracked birds (which on the other hand represents one of the largest dataset of seabird telemetry). It is known that Laysan and black-footed albatross distributions change depending on season (Hyrenbach et al. 2002, Fischer et al. 2009; Kappes et al. 2010; Suryan & Fischer 2010). However further stratification of our tracking data into shorter time periods would lead to even greater patchiness of static distribution plots, and subsequently make spatial overlays with fisheries observations nearly impossible. These obvious limitations in using static distribution maps obtained from telemetry data for integrating with fisheries and bycatch datasets indicated a need for an alternative solution. We therefore aimed to generalize albatross habitat preferences and search for such conditions at a place or period of interest. This resulted in development of dynamic habitat models and subsequent calculation of albatross habitat probabilities at fisheries locations.
Figure ESM1-2. Distribution of non-breeding Laysan and black-footed albatrosses in 2005 and 2006 as estimated by kernel analysis. Maps also present locations of observed longline sets and recorded bycatch of each species during 1997-2000.
1