Population Dynamics of Newfoundland Moose Using Cohort Analysis

Abstract

Changes in abundance, recruitment rates, and age-specific harvest rates of 3 populations of moose (Alces alces) in Newfoundland were estimated with cohort analysis. Populations declined during 1966-73 due to overharvesting, but recovered in all areas following reduction of license quotas in the mid-1970's. Age-specific vulnerability to hunting was highest for young adults and lowest for calves and mid-aged adults. Differences in vulnerability were possibly related to hunter preferences, previous experience of moose with hunters, and frequency of movements by moose. Data on kills/day and moose seen/day obtained from hunter questionaires were positively correlated with moose abundance but negatively correlated with the number of hunters. The latter suggested that interference had a strong influence on hunter success. J. WILDL. MANAGE. 52(1):14-21 A challenging problem facing wildlife managers is how to measure changes in population abundance with sufficient accuracy to make reasonable management decisions. In general, 2 approaches have been employed. Where sufficient funding and manpower are available, many agencies attempt direct estimation of population abundance using aerial surveys, drive counts, or mark-recapture techniques. These methods are costly, time consuming, and often imprecise. Where resources are more limited, wildlife managers are often forced to rely on indirect indices of abundance such as pellet counts, track counts, or hunter success statistics. Such indices can show conflicting trends, and often no other information is available to evaluate which indices are meaningful. Fisheries scientists are faced with similar problems. Their predicament is perhaps even more acute because direct enumeration is impossible for most fish populations. In response, fisheries scientists have developed a wide variety of methods for assessing stock abundance based largely on catch returns from survey vessels or commercial fleets. A powerful technique to reconstruct historical changes in stock abundance from catch-at-age data is the cohort or virtual population analysis (Ricker 1948, Fry 1949, Pope 1972). We used cohort analysis to reconstruct historical changes in moose abundance in 3 areas of Newfoundland from kill-at-age data gathered during 1966-85. Population estimates for 1966-73 (corresponding to cohorts that have completely passed through the population) were used to estimate rates of recruitment and agespec fic vulnerability, and to calibrate indirect indices that have been used in the past to manage moose populations in Newfoundland. Both cohort analysis and indirect indices were then used to evaluate more recent trends in abundance. Funding for this work was provided by the Wildlife Division, Department of Culture, Recreation and Youth, Newfoundland. The Biological Data Centre, University of British Columbia, kindly provided computer facilities. Data were gathered with the assistance of W. A. Collins, S. M. Oosenbrug, and M. J. Strapp. We thank J. S. Collie, P. R. Krausman, M. F. Lapointe, C. J. Walters, G. H. Winters, and an anonymous reviewer for helpful discussions and comments on the manuscript.