Uncommon or cryptic? Challenges in estimating leopard seal abundance by conventional but state-of-the-art methods

• Published in: Deep Sea Research Part I: Oceanographic Research Papers
• Main Authors: Southwell, Colin, Paxton, Charles G. M., Borchers, David, Boveng, Peter, Rogers, Tracey, de la Mare, William K.
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2008
• Subjects: abundance; distribution; East Antarctica; haul-out; model-based inference; leopard seal; pack-ice; HYDRURGA-LEPTONYX; LOBODON-CARCINOPHAGUS; WINTER DISPERSAL; PRYDZ BAY; BEHAVIOR; Hydrurga; Prydz Bay; Antarc*; Antarctica; Hydrurga leptonyx; Leopard Seal; Leopard Seals; Lobodon carcinophagus
• Online Access: https://research-portal.st-andrews.ac.uk/en/publications/ab69c352-5c41-4fe4-8eb6-9a4aa0d91f06; https://doi.org/10.1016/j.dsr.2008.01.005; http://www.scopus.com/inward/record.url?scp=41549144279&partnerID=8YFLogxK

The method traditionally used to estimate pack-ice seal abundance employs sighting surveys from ships or ail-craft to estimate the number of seals hauled out oil the ice, combined with studies of haul-out behaviour to estimate the proportion of time spent on the ice. Application of this approach has been improved in recent times by developments in survey methodology and satellite technology that theoretically allow biases in the estimation of hauled-out abundance and haul-out behaviour to be accounted for that previously could not be addressed. A survey using these conventional but state-of-the-art methods was undertaken in the summer of 1999/2000 off east Antarctica between longitudes 64 degrees E and 150 degrees E to estimate the abundance of leopard (Hydrurga leptonyx) and other pack-ice seal species. Because they are either uncommon or very cryptic, very few leopard seals were encountered despite a large survey effort. This presented challenges in both application of the methods and analysis of the data. Abundance estimates were derived using a number of plausible predictive models. The model considered as the most reliable returned best estimates of 7300 and 12,100 for definite and definite plus probable leopard seal sightings, respectively, with 95%, confidence intervals of 3700-14,500 and 7100-23,400. These estimates are likely to be negatively biased and should be treated as minimum estimates only. (c) 2008 Elsevier Ltd. All rights reserved.