Band-Recovery Analysis and Survival Estimates for South Dakota Giant Canada Goose Populations

Survival and recovery rate estimates were calculated for South Dakota giant Canada goose (Branta canadensis maxima) populations. I analyzed Canada goose bandrecovery data obtained from the USGS\u27s Bird Banding Laboratory consisting of 26, 141 bandings and 6,837 recoveries (1955 - 95). Only Canada geese banded preseason (June - September) and shot or found dead during the hunting season were included in the analysis. Programs DBASE IV, PARADOX, and Band Analysis System were used for sorting banding and recovery data. Survival and recovery rate estimates were derived using Program ESTIMATE and INTERVAL. To increase sample size and to meet assumptions, testing for pooling of sexes, ages, status\u27 (status 3; normal, wild vs. status 2,4,6,8; restored), year periods (1967-76, 1977-86, and 1987 95) and regions (Waubay, eastern and western, SD) was conducted. Calculation of bandings by IO-year periods (i.e., 1967-76, 1977-86, 1987-95) indicated regional differences (P≤ 0.05) in banding effort. Banding efforts in western South Dakota were relatively constant (P=0.937) over the 29-year period while efforts in eastern South Dakota increased significantly (P=0.02) during the same interval. Banding efforts for the Waubay region showed significant increases (P=0.0001) over the 17-year period studied. Results from recovery rates (unadjusted, direct and indirect recoveries pooled) (± SE) derived using Program ESTIMATE and INTERVAL indicated an increase in recovery rates over time [3 .7 ± 0.7% (1967-76), 5.6 ± 0.9% (1977-86)] for normal, wild (ages and sexes pooled) Canada geese banded in western South Dakota. For this same status group (ages and sexes pooled), eastern South Dakota recovery rates increased from 2.6 ± 0.2% (1967-78) to 5.8 ± 0.3% (1987-95). Similarly, normal, wild geese (ages and sexes pooled) banded at Waubay showed increasing recovery rates of 5.6 ± 1 .7% (1959-66) and 7.9 ± 0.5% (1967-76). Adult (1968-79) and subadult (1976-78) recovery rates for restored flocks banded in western South Dakota were 6.4 ± 0.4% and 8. 1 ± 1.1 %. Recovery rates for restored flocks (ages and sexes pooled) banded in eastern South Dakota ranged from 4.0 ± 0.4% (1987-95) to 7.4 ± 0.7% (1977-86). The highest and lowest recovery rates derived for this study were for normal, wild subadults (sexes pooled) banded in the Waubay region (1 2.0 ± 2.7%) (1959-66) and eastern South Dakota (0.2 ± 0.8%) (1977-95). Program CENTROID was used to test for differences in recovery distributions. A difference occurred for both direct (P=0.0001) and indirect (P=0.0001) recovery distributions for normal, wild birds banded in western (1967-76) versus eastern (1967-95) South Dakota. Similarly, a difference existed for both direct (P=0.0001) and indirect (P=0.0001) recovery distributions for restored flocks (subadults) of Canada geese banded in western (1976-78) versus eastern (1977-86) South Dakota. Survival estimates were compared using Program CONTRAST. Akaike\u27s Information Criterion was used to select the most appropriate model. Pooled estimates (i.e., ages and sexes combined) of survival (± SE) for normal, wild (status 3) Canada goose flocks were 91.6 ± 19.6% (1967-76) and 68.6 ± 4.9 % (1977-86) and 74.1±4.6% (1967-78) and 81.9 ± 16.6% (1987 - 95) for western and eastern South Dakota, respectively. Pooled estimates of survival for restored Canada goose flocks (status 2,4,6) in eastern South Dakota were 59.0 ± 3.0% (1977 - 86) and 89.7 ± 3.7% (1987-95). Due to significant (P≤0.05) age class differences in survival estimates [Adults = 58.3 ± 2.2%, Subadults = 21.4 ± 5.6% (1968 -79)], pooling was not possible for restored flocks (status 2,4,6,8) of Canada geese in western South Dakota. Similarly, pooling across status groups was not possible regardless of banding region. In addition, effects of harvest regulations on survival of giant Canada geese were analyzed. Data were separated and analyzed for three sample periods (1960-76 considered historical, 1977-86 considered restrictive, and 1987-95 considered liberal) and two regions (Waubay corresponding to the historical period and eastern South Dakota corresponding to the restrictive and liberal periods, respectively) based on available data and recommendations from South Dakota Department of Game, Fish and Parks personnel. Bag limits (BL), season length (SL), and decoy restrictions (DR) varied across the three periods with more liberal regulations (BL = 2, SL x = 80.2 days, DR = no restrictions) occurring during the latter period. Using a multiple regression model of survival, the number of active adult Central Flyway waterfowl hunters, season length, and decoys was significant (P≤0.05), however, the overall regression model failed to explain a statistically significant amount of variation in survival estimates (P = 0.074). During the same three periods, the percentage of large Canada geese harvested in South Dakota increased (1967-76, x = 50.0%; 1977-86, x = 55.4%; 1987-94, x = 68.3%). The estimated survival rates documented for South Dakota giant Canada geese compare favorably to similar studies conducted in Michigan (x = 76.5%) and New Jersey (x = 82.8%). The general increase in survival estimates (in eastern South Dakota) obtained during this study is consistent with an increasing Canada goose population