Testing methods to estimate abundance of Magellanic Penguins Spheniscus magellanicus

Abstract

Capsule: Using simulations in a geographical information system our results showed that systematic sampling with quadrats was the most accurate, precise and cost-effective method to survey Magellanic Penguin colonies. Aims: To determine which sampling method gives better estimations of penguin abundance. Methods: A virtual colony was generated deriving spatial parameters from a real survey and applying Kriging interpolation. Three sampling methods were then applied on this virtual colony: random sampling with quadrats; systematic sampling with quadrats; systematic sampling with fixed-width transects. The estimated abundance for each trial was compared to the abundance of the virtual colony to have a measure of accuracy and precision. Results: Systematic sampling with quadrats estimated penguin abundance better than random or systematic sampling with transects since it achieved 100% accuracy and great precision after sampling only 2.1% of the virtual colony. Conclusion: The use of a simulated colony allowed the comparison of several sampling methods traditionally used in Magellanic Penguin surveys. The results of this study are important in order to standardize sampling protocols for Magellanic Penguins and to have more comparable estimations to detect trends over time. Also, the methodological approach used here could be used to assess sampling methods for other colonial bird species.