INTEGRATING COUNT EFFORT BY SEASONALLY CORRECTING ANIMAL POPULATION ESTIMATES (ICESCAPE): A METHOD FOR ESTIMATING ABUNDANCE AND ITS UNCERTAINTY FROM COUNT DATA USING ADÉLIE PENGUINS AS A CASE STUDY

This work describes a parametric bootstrap model (Davison and Hinkley 1997) for standardising animal count data to a common reference point of breeding chronological for species showing temporal availability to sampling methodology. ICESCAPE (Integrating Count Effort by Seasonally Correcting Animal Population Estimates) is suite of routines that implements a general abundance estimator accounting for availability bias, detection bias, and sampling fractions less than unity (Pollock et al. 2004, Southwell 2004a). Within this resampling framework all measures of uncertainty associated with originally published counts are propagated through to final adjusted estimates. Adjustment for availability bias is achieved by standardising counts to a common reference point of breeding chronology by applying an adjustment factor based on independently measured time-series of availability throughout a breeding season. Such series are typically collected at only a limited number of sites, so a search algorithm is used to determine surrogate availability information for a site when none exists. Importantly, standardisation in this way allows site-specific estimates to be aggregated to achieve region-scale population estimates. By way of illustration, the method is applied to several examples of published studies of Adélie penguin abundance at breeding sites in Antarctica. These examples focus on adjusting counts of adults to an effective number of breeding pairs, although the software has been developed to accommodate adjustment and aggregation of other count objects typical for penguin species (e.g. occupied nest or chick counts). While tailored for Adélie penguins, the method and implementation is sufficiently general to be easily adapted for other colonial land-breeding species showing seasonal variation in availability to sampling methodology.