Including behavioral data in demographic models improves estimates of population viability

tion biology attracted a great deal of attention in past years (Curio 1996; Caro 1998; Sutherland 1998). While there has been wide recognition that behavior can profoundly influence the dynamics of populations (Sutherland 1996; Houston and McNamara 1997), empirical research linking behavior to demography is sparse (Anholt 1997; Levin et al. 2000). Furthermore, a framework for applying behavioral knowledge to real-world conservation problems has not yet been developed (Anthony and Blumstein 2000). Individual behavioral strategies influence how animals respond to perturbation, so understanding factors that influence behavior over short time scales may provide important information for developing accurate population models (Sutherland 1988; Durant 1998). Theory about small populations suggests a number of potential links between behavior and population viability, as mediated by a population’s sex ratio (Anthony and Blumstein 2000). Specifically, a species’ mating system dictates the similarity between the actual sex ratio (ASR) and the operational sex ratio (OSR) of the population. While the ASR reflects the true ratio of females to males, the OSR is the ratio of breeding females per breeding male. These sex-based differences are typically ignored in population models used for conservation, since only the female portion of the population is explicitly modeled. The assumption is that only females determine the population growth rate. In theory, the mating system links behaviors such as mate defense and territoriality indirectly to population growth via the difference between ASR and OSR. Here, I distinguish between genetic and demographic measures of population viability and argue that the link between behavior (mediated through the OSR) and demographic population viability cannot be ignored. Population geneticists typically quantify the genetic viability of a population using the mean and variance of successful gametes per individual as a proxy for genetic variability (Lande 1994). By this measure, small populations with biased sex ratios have lower genetic viability than populations with even sex ratios of the same population size. By contrast, population biologists typically assess the demographic viability of a population based exclusively on means and variances in female fecundity, survival, and abundance. This assumes either that these parameters are similar for males, or that in males they 419