Evaluation of Population Trend Estimates Calculated Using Capture‐Recapture and Population Projection Methods

Abstract

Concern has arisen recently that Northern Spotted Owl (Strix occidentalis) populations may be declining faster than suspected by Spotted Owl biologists. This is based on recent estimates by the USDI of the annual rate of change in population size (i.e., population trend) calculated using capture–recapture and population projection methods. These original reports estimated change in the territorial population but used an inappropriate demographic model which did not incorporate immigration from the floater population. This approach, however, would be appropriate for estimating trend in the total population (floaters and territorial birds) if the model assumptions were met. This raises the possibility that Northern Spotted Owl populations are declining at approximately the rate reported in the USDI studies (6‐8% per year) rather than at the rate previously suspected (1‐3% per year). I investigated this issue using a stochastic computer simulation in which population trends were estimated with capture–recapture and population projection methods and compared to the true population trends in the simulation. Factors that might cause bias in the trend estimate were identified and set to simulate the smallest and largest likely bias. The most important of these factors were the floater population and permanent emigration from the study site. Results showed that population trends in the simulation were underestimated by 0.03 to 0.13. Thus if the true annual rate of change in population size were 0.99 (i.e., a decline of 1%/yr), the estimate of 0.99 provided by the USDI studies would be 0.86 to 0.96. The simulations also provided new evidence, based on the past and current size of the floater population, indicating that Northern Spotted Owl populations have probably been declining at an average rate of ≤3% per year during the past few decades. My analysis indicates that the USDI studies should not be interpreted as providing evidence that owl populations are declining faster than previously assumed. We need careful investigation of trend estimates based on capture–recapture and demographic analyses before they are used as the basis for management decisions.