Comparative Census Procedures Using Single Mark‐Recapture Methods

Proportional abundance is defined as the ratio of the animal abundance of one population (N 2 ) to that of another (N 1 ), say K = N 2 /N'1. Confidence interval estimators of K for paired populations are presented using single mark—recapture studies. Two alternative estimators of K based on assumptions about catchability are constructed, and power curves developed using computer simulations. Tests of homogeneity are presented which test the assumption of equal catchability among animals of the two populations. When the homogeneity assumption is valid, the proportional abundance estimator based on two indices of relative abundance is from 2 to 20 times more efficient than the comparable estimator based on absolute abundance. Analytic approximations for power calculations and sampling precision are presented which may be used in the design of comparative census studies.