Survival and habitat use of mallard broods in the San Joaquin Valley, California

Thesis (M.S.)--Humboldt State University, Natural Resources and Science: Wildlife Management, 2000California's Central Valley is one of the most important wintering areas for North American waterfowl, but its importance as breeding habitat has only recently been recognized. Previous studies have shown that California mallards (Anas platyrhynchos) have nest success estimates above reported averages necessary for healthy recruitment, but duckling survival estimates in 2 separate studies have been below average recruitment estimates. The primary objective of my study was to determine if duckling survival limited mallard recruitment in the Grassland Region of the San Joaquin Valley (SJV). My specific objectives were: 1) estimate duckling survival rates from hatch to day 30, 2) document specific causes of mortality, 3) examine brood habitat preferences, and 4) determine if duckling survival varied among wetland types, especially with regard to recently created reverse-cycle wetlands. My study area encompassed the Los Banos and Salt Slough State Wildlife Management Areas and several adjacent duck clubs, and included 1,388 ha of wetlands and another 500 ha of intermixed grasslands and riparian areas. I radiomarked the adult female and 2 ducklings in each of 56 mallard broods in 1996 and 1997 to document duckling survival rates and brood habitat use. Total brood loss occurred in 27 of 53 broods (50.9%; 3 broods had undetermined fates). There was no evidence that radios reduced duckling survival; 30-day survival estimates from resighting data averaged 0.287 for radioed ducklings and 0.229 for non-radioed ducklings (Z = 0.51, P = 0.61). Thirty-day survival rates were below reported estimates for healthy populations, averaging 33.1% in 1996 and 19.3% in 1997. Survival differed between the Los Banos and Salt Slough study areas, but this difference was attributable to differences in habitat types. Ducklings had higher 29-day survival rates in reverse-cycle wetlands\ud (0.755 ?? 0.095) than in semi-perm anent/permanent wetlands (0.189 ?? 0.056) or moist-soil units (0.173 ?? 0.067). Avian and mammalian predators were the primary causes of death, comprising 39% (n = 31) and 22% (n = 18) of known mortalities. Distance from nest to first wetland and home-range sizes averaged 663 m (SE = 106 m) and 14.6 ha (SE = 3.2, n = 28), respectively. For initial brood movements, broods selected, in significantly greater proportion than available, reverse-cycle wetlands (P = 0.006) and tended to avoid perm anent wetlands (P = 0.09), but no selection was observed for semi-permanent or moist-soil wetlands (P = 0.22). For the remainder of the brood-rearing period (days 2-30), broods avoided semi-permanent (P = 0.0001) and permanent (P = 0.0010) wetlands and preferred reverse-cycle (P = 0.0001) wetlands. Brood movements resulted in 28% of all duckling mortalities. Unstable water levels, unreliable maintenance flows, or drawdowns associated with moist-soil management were the primary causes of brood movements and associated mortalities. Although overall duckling survival was low in my study, survival was very high in reverse-cycle wetlands. Broods exhibited a strong preference for this habitat type. My data strongly suggest that duckling survival could be improved by increasing the quantity and quality of reverse-cycle wetlands