Estimating sample size for landscape-scale mark-recapture studies of North American migratory tree bats

To assess changes in the condition of bat populations, biologists can monitor various population parameters, including abundance, population change, and survival (Kunz et al., 2009; O’Donnell, 2009). A common sampling and analysis framework used to estimate such population parameters is markrecapture methodology (Schwarz and Seber, 1999). However, a challenge to conducting such studies is determining the level of sampling required to produce reliable estimates, especially when sample size depends on many factors, including the detectability of the species, desired precision, and scope of study (Williams et al., 2001). Although population size (abundance) is a popular parameter of interest, satisfying the assumptions for models of abundance is more challenging than for models of other parameters, such as survival or population change (λ) (White et al., 1982). Alternately, survival is a parameter that can be estimated directly from mark-recapture studies, and model extensions provide estimates of λ (Pradel, 1996). Recently, the status of bat populations has been of particular interest because landscape-level impacts, such as wind energy development and whitenose syndrome, are reducing or threatening bat populations in North America (Kunz et al., 2007b; Frick et al., 2010b). Accurately assessing bat population status is challenging, primarily because of the difficulty in observing and detecting bats (Kunz et al., 2009). Compared to other wildlife species, bats are problematic because they are small, cryptic, nocturnal, inconspicuous when active, and live in sometimes inaccessible habitats. These challenges have not deterred bat marking studies dating back to 1916 in which unique bands were attached to the arm or leg of a bat (Allen, 1921). Acta Chiropterologica, 16(1): 231–239, 2014 PL ISSN 1508-1109 © Museum and Institute of Zoology PAS doi: 10.3161/150811014X683426