Factors Affecting Age Ratios of Samples of Canada Geese Caught with Cannon-Nets

Abstract

Widely differing opinions are held concerning the reliability of age-ratio statistics derived from samples of Canada geese trapped with cannon-nets. The 95 percent confidence limits of an age-ratio estimate of 41.3 percent immature derived from 35 catches involving 2,451 geese were 35.3-50.5 percent immature. These data indicate the potential for error when such data are used to analyze population dynamics. Based on a study of behavior of marked Canada geese of all ages and sexes and family units, utilizing radio telemetry, reasons are suggested for age-ratio fluctuations of cannon-net trap samples. Important interrelated factors causing such cannon-net trap age-ratio differences are time of year, time of day, food availability, behavior changes, length of time on bait, and weather. When these factors cause Canada geese to crowd densely onto a baited-trap site, a higher percentage of adults would usually be obtained than when trap catches represent different levels of aggression, family defense, and less crowded conditions. Alternative explanations are offered of data that led Funk and Grieb (1965) to conclude that age ratios from baited-trap samples reliably indicate population structure-a conclusion different from that reached in this paper. Realistic estimation of sex and age composition of a population provides the axiomatic necessary foundation for evaluating and understanding past reproductive successes which in turn provide the basis for predicting future population fluctuations. Predicting the size of an ensuing year's population is an important aspect of the management of Canada geese (Branta canadensis interior) in the Mississippi Flyway, for the legal hunter-harvest is calculated as a percentage of geese expected to make up the coming year's population. This estimate is based upon many factors, one of the more important being the estimate of the age composition of the population that returned in the spring to the breeding grounds. Throughout the United States, cannonnet trapping (Dill and Thornsberry 1950, and many subsequent technique modifications) is one of the major methods, and often the only method, used to capture Canada geese during the migration and winter periods. Many persons who have done considerable goose trapping with cannon-nets have learned that age ratios of individual catches may fluctuate greatly and, as a result, place little confidence in the use of such age ratios in population dynamics analyses. Until recently, however (Nass 1964, Funk and Grieb 1965), little statistical or methodological examination of age-ratio data from trap samples has been reported, and only tentative explanations have been offered of possible factors influencing such age-ratio fluctuations or biases. The purpose of this paper is to present an evaluation of the variability of trap-sample age ratios obtained from 35 cannon-net c tches involving 2,451 Canada geese (nearly all B. c. interior). Based on a study of Canada goose behavior including observations of marked individuals of all ages and sexes and family groups at bait sites, explanations are offered as to why age ratios of cannon-net catches may vary rather widely. Results are part of a more comprehensive study of Canada goose behavior in winter, utilizing radio telemetry. The program was supported chiefly by the National Science Foundation (GB-623); additional support was provided by the Cooperative Wildlife Research Laboratory, Southern Illinois Univ rsity, for which I am indebted to Dr. W. D. Klimstra, Director. Findings are