Model comparison table for Pradel’s temporal symmetry analysis of data collected from a population of northern raccoons (<i>Procyon lotor</i>) at the Ordway-Swisher Biological Station, Florida, from September 1992 to December 1998.

We ran model sets investigating the effects of sex, time, month, season, and additive and interactive effects of sex with the other variables on four parameters: recapture probability, p; apparent survival probability, φ; recruitment rate, f; and realized population growth rate, λ. In section (A), we used the recruitment parameterization; in section (B), we used the lambda parameterization. Only the top ten models are included for each set. K is the number of parameters, AICc is Akaike’s Information Criterion corrected for small sample size, ΔAICc is the difference between each model’s AICc and the AICc of the top-ranked model, weight is the Akaike weight or model probability, and deviance is the model deviance.我们构建多组模型集，用以探究性别、时间、月份、季节，以及性别与其余变量的加性效应和交互效应对4个种群参数的影响：重捕概率（recapture probability）p、表观存活概率（apparent survival probability）φ、补充率（recruitment rate）f，以及实际种群增长率（realized population growth rate）λ。在(A)部分，我们采用补充率参数化方案；在(B)部分，我们采用种群增长率参数化方案。每组模型集仅纳入排名前十的候选模型。其中K为参数个数，AICc即针对小样本量校正的赤池信息准则（Akaike’s Information Criterion corrected for small sample size），ΔAICc为单模型AICc值与最优模型AICc值的差值，weight为赤池权重，或称模型概率，deviance为模型偏差。