PARENT SESSION
Oral Session 13: Rare, Threatened, and Endangered Species I: Management I.
Presiding: H Balbach
Monday, August 2, 8:00 AM to 11:30 AM, Meeting Room D 135.

Minimizing the cost of threatened species management: Can error rates be optimized?

Field, Scott¹, Tyre, Andrew^*,2, Jonzen, Niclas¹, Rhodes, Jonathan¹, McCarthy, Michael³, Wintle, Brendan³, Possingham, Hugh¹, ¹ University of Queensland, Brisbane, Queensland, Australia² University of Nebraska-Lincoln, Lincoln, Nebraska³ University of Melbourne, Melbourne, Victoria, Australia

ABSTRACT- Decisions based on monitoring data form the cornerstone of effective threatened species management. However, when such data are analysed with frequentist statistics, decisions are prone to potentially costly errors. The convention of fixing alpha at 5 % and accepting the resulting statistical power ignores the fact that the cost of Type II errors (failing to detect a decline) are often greater than that of Type I errors (unnecessarily initiating recovery). A more defensible approach is to set optimal levels of significance and power by formulating the problem in a decision theory framework that accounts for the relative costs of the two kinds of errors. We propose a new method for doing this, by minimizing a function specifying the expected overall cost of monitoring and management. Using a case-study of managing a threatened koala population in eastern Australia, we show that for a species of such high economic value, Type II errors should never be tolerated and therefore monitoring to trigger recovery action is redundant. We identify a narrow range of Type 2:Type 1 error cost ratios for which an optimal alpha-level exists. Our analysis suggests that a Bayesian approach might provide a more logically coherent framework for this problem.

Key words: Phascolarctos cinereus, power analysis, optimal monitoring