INFERRING PROCESS FROM PATTERN: CAN TERRITORY OCCUPANCY PROVIDE INFORMATION ABOUT LIFE HISTORY PARAMETERS?

摘要: A significant problem in wildlife management is identifying "good" habitat for species within the short time frames demanded by policy makers. Statistical models of response presence/absence to predictor variables are one solution, widely known as modeling. We use a "virtual ecologist" test logistic regression means developing spatially explicit, individual-based simulation that allows quality influence either fecundity or survival with continuous scale. The basic question how good at where birth rates high and death low (i.e., "source" habitat)? find that, even when all important perfectly measured, there no error surveying interest, demographic stochasticity limiting effect localized dispersal generally prevent an explanation much more than half variation territory occupancy function quality. This true regardless whether influenced In addition, only detect on autocorrelated. based really measure ability reach colonize areas, not rates.