Risk-Based Viable Population MonitoringDAVID F. STAPLES,∗‡ MARK L. TAPER,∗AND BRADLEY B. SHEPARD∗†∗Montana State University, Department of Ecology, 310 Lewis Hall, Bozeman, MT 59717-3460, U.S.A.†Montana Fish, Wildlife, and Parks, Region 3, 1400 South 19th Avenue, Bozeman, MT 59718-5417, U.S.A.Abstract:We describe risk-based viable population monitoring, in which the monitoring indicator is a yearlyprediction of the probability that, within a given timeframe, the population abundance will decline below aprespecified level. Common abundance-based monitoring strategies usually have low power to detect declinesin threatened and endangered species and are largely reactive to declines. Comparisons of the population’sestimated risk of decline over time will help determine status in a more defensible manner than current mon-itoring methods. Monitoring risk is a more proactive approach; critical changes in the population’s status aremore likely to be demonstrated before a devastating decline than with abundance-based monitoring methods.In this framework, recovery is defined not as a single evaluation of long-term viability but as maintaininglow risk of decline for the next several generations. Effects of errors in risk prediction techniques are mitigatedthrough shorter prediction intervals, setting threshold abundances near current abundance, and explicitly in-corporating uncertainty in risk estimates. Viable population monitoring also intrinsically adjusts monitoringeffort relative to the population’s true status and exhibits considerable robustness to model misspecification.We present simulations showing that risk predictions made with a simple exponential growth model can beeffective monitoring indicators for population dynamics ranging from random walk to density dependencewith stable, decreasing, or increasing equilibrium. In analyses of time-series data for five species, risk-basedmonitoring warned of future declines and demonstrated secure status more effectively than statistical tests fortrend.Key Words:population modeling, population viability analysis, PVA, species recovery, threatened and endan-gered speciesMonitoreo de Poblaciones Viables con Base en RiesgosResumen:Describimos el monitoreo de poblaciones viables basado en riesgos, en el que el indicador delmonitoreo es la probabilidad de que la abundancia de la poblaci´on decline, en un per´ıodo de tiempo determi-nado, por debajo de un nivel predefinido. Las estrategias comunes de monitoreo basadas en la abundanciageneralmente tienen poco poder para detectar declinaciones de especies amenazadas y en peligro y son am-pliamente reactivas a las declinaciones. Las comparaciones del riesgo de declinaci´on estimado a lo largo deltiempo ayudar´an a determinar el estatus de una manera m´as defendible que con los m´etodos de monitoreoactuales. El monitoreo de riesgo es un m´etodo m´as preventivo; es m´as probable que los cambios cr´ıticos en elestatus de una poblaci´on sean evidentes antes de una declinaci´on devastadora que con m´etodos de monitoreobasados en la abundancia. En este marco, la recuperaci´on est´a definida como el mantenimiento de un bajoriesgo de declinaci´on para varias generaciones futuras y no solo como una evaluaci´on de la viabilidad a largoplazo. Los efectos de los errores de las t´ecnicas de predicci´on de riesgos son mitigados mediante intervalos depredicci´on m´as cortos, el ajuste de umbrales de abundancia cerca de la abundancia actual y la incorporaci´onexpl´ıcita de la incertidumbre en las estimaciones de riesgo. Intr´ınsecamente, el monitoreo de poblaciones vi-ables tambi´en ajusta el esfuerzo de monitoreo en relaci´on con el verdadero estatus de la poblaci´on y muestraconsiderable robustez ante errores de descripci´on del modelo. Presentamos simulaciones que muestran quelas predicciones de riesgo derivadas de un modelo simple de crecimiento exponencial pueden ser indicadoresefectivos del monitoreo de la din´amica poblacional que var´ıan de caminatas al azar hasta denso dependencia‡email [email protected] submitted November 8, 2004; revised manuscript accepted January 11, 2005.1908Conservation Biology 1908–1916C2005 Society for Conservation BiologyDOI: 10.1111/j.1523-1739.2005.00283.xStaples et al. Risk-Based Viable Population Monitoring 1909con equilibrio estable, decreciente o creciente. En el an´alisis de datos de series de tiempo para cinco especies, elmonitoreo basado en riesgo alert´o sobre futuras declinaciones y demostr´oelestatus seguro m´as efectivamenteque las pruebas estad´ısticas de la tendencia.Palabras Clave:an´alisis de viabilidad poblacional, AVP, especies amenazadas y en peligro, modelado de pobla-ciones, recuperaci´on de especiesIntroductionCorrectly evaluating the status of threatened and endan-gered populations is critical to detecting population de-clines that will further endanger the population and toverifying effectiveness of conservation actions. Monitor-ing to detect changes in status is frequently specified inrecovery plans (Morris et al. 2002). Unfortunately, de-tecting changes in status with commonly used monitor-ing strategies such as monitoring species abundance andstatistical tests (Thompson et al. 1998) for populationtrend can be difficult (Holmes & York 2003). Because ofhigh variability in ecological data arising both from nat-ural variation in population growth and from measure-ment errors in population estimates, a biologically dev-astating decline could occur before there is a reasonablechance to detect the decline. For example, at observedvariation levels common in bull trout (Salvelinus con-fluentus) data, a population could decline by more than60% before the power to significantly detect such a de-cline reached 0.80 (Maxell 1999). For species that havepresumably already declined to low levels, an inability toquickly detect further declines in population abundanceis a serious problem. Ideally, monitoring would reliablydetect a small (though biologically relevant) decline in ashort amount of time; this goal, however, is not realisticwith current methods for most populations. Monitoringspecies with abundance-based methods also leaves man-agers in a reactive posture because a decline might bedemonstrated only after it has seriously diminished thepopulation, even though the demographic problems thatcaused the decline occurred many years before the de-cline was “significant.”Natural