Estimates of Annual Survival Rates of Trumpeter Swans Banded 1949-82 Near Red Rock Lakes National Wildlife Refuge, Montana

Abstract

Annual survival rates (S) were estimated from band recovery, recapture, and neck-collar resighting data for adult trumpeter swans (Cygnus buccinator) near Red Rock Lakes National Wildlife Refuge (RRLNWR), Montana. Analysis of banding and recovery data collected between 1949 and 1982 yielded an estimated S of 88%. Analysis of resightings of neck-collared birds was accomplished using 3 methods, and the estimated S averaged about 80%, based on data collected since 1971. Reasons why these estimates probably are low are discussed. The data collected over the past 30-35 years on this population have been inadequate; however, a larger and better planned banding and resighting program would be an effective monitoring system. J. WILDL. MANAGE. 50(2):218-221 Prior to the 20th Century, trumpeter swans were thought to be numerous in North America. Trumpeter swan skins made up a large proportion of swan skins purchased by the Hudson Bay Company (Swainson and Richardson 1832). By the early 1900's, the number of trumpeter swans in North America had dropped considerably and some biologists warned of possible extinction (Bent 1925). By 1982, there were 31 swans in Yellowstone Park, Wyoming; 26 in Centennial Valley, Montana; and only 12 others in the contiguous United States (Banko 1960). RRLNWR was established in 1935, and the number of swans there increased from 26 to about 600 by the mid-1950's (Fjetland 1974). Since 1954, the swan population at RRLNWR has declined to about 250 birds (Hansen 1973, Page 1974). Elsewhere, >2,000 swans nest in Alaska and a few nest in Canada. Hansen (1973) reported between 3,000 and 5,000 birds in North America at the end of the breeding seasons in the late 1960's. Hansen (1973:88) reported that the breeding habitat at RRLNWR became saturated with nesting pairs by the mid-1950's and that the number of birds declined due to ... a general reduction in the availability of submerged aquatic plants used as food. He wrote that temporary increases in the population after the mid-1950's were due to years with favorable weather which resulted in large numbers of offspring. Archibald (1977) reported winter food supplies to be the initial controlling factor in preventing population growth of the swans at RRLNWR. He attributed their increase in number by the mid-1950's to a supplemental winter feeding program and speculated that esting habitat availability was a reason for their recent decline. Swans at RRLNWR had been marked for many years, and these data allowed changes in survival to be assessed. The objective of this study was to estimate S and related parameters from banding data. We thank J. M. Dickson for providing program ANALYSE and L. L. McDonald for advice on procedures for treating the analysis of the combined recovery and terminal resighting data. A. N. Arnason and L. Johnston provided computer subroutines necessary to run program ANALYSE.