Influence of habitat conditions on nesting activity of wood ducks (Aix Sponsa) in tree cavities
Richard A. Geboy, MS
Advisor: Robert Gates
Wood duck breeding biology has largely been studied in artificial nest boxes, but environmental influences on nesting biology must be studied in the context within which wood ducks evolved (i.e., natural tree cavities). Previous work suggested that intra- and inter-seasonal changes in abundance of invertebrate foods and/or accessibility of tree cavities may affect nesting season length and the breeding chronology of wood duck hens. This study examined the relationships of intra-seasonal and annual variation in invertebrate food abundance and leaf emergence with nesting chronology, nesting effort, and nest parasitism rates in a southern Illinois wood duck population that nests in natural cavities. Nest chronology, nesting effort, and parasitism rates were studied during 1994 – 1998 and 2001 – 2002. I modeled wetland conditions with local precipitation and stage of the Mississippi River. Wetland invertebrate food abundance was measured by sweep-net sampling in seasonal and semi-permanent wetlands in 2001 and 2002. Invertebrate sampling was completed during four nesting periods (egg laying, late egg laying/early incubation, incubation/early hatch, hatch) during 2001 – 2002. Weekly leaf emergence in 2001 and 2002 was measured with a Model-A Spherical Densiometer, from the onset of leaf emergence through the period of maximum upper story leaf coverage. Annual nesting effort of radiomarked hens known to incubate clutches ranged from 42% - 70% throughout the study. Clutch sizes >14 (indicative of nest parasitism) ranged from 8% - 43% during the study and varied inversely with nesting effort. High nesting effort was observed in years with the most stable water levels during egg laying and incubation. Invertebrate biomass did not differ throughout the 2001 (P = 0.758) and 2002 (P = 0.167) nesting seasons. In 2002, mean invertebrate densities during hatch were 1.7 times higher than during egg laying and incubation/early hatch, but did not differ from densities observed during late egg laying/early incubation (P = 0.047). Mean invertebrate densities in 2002 were not significantly different (P = 0.869) than those of 2001. Two commonly consumed invertebrates known to occur during nest initiation and hatch, Isopoda and Diptera (primarily Chironomidae), were 10.2% and 18.4% more numerous during the egg laying and hatch (P < 0.01), respectively. Leaf emergence in floodplain and upland habitats commenced during the week of 13-April in 2001 and 2002. Rates of leaf emergence did not differ between forest types in 2001 (P = 0.998) and across-year upland habitat (P = 0.396) comparisons. Forest types in 2002 (P = 0.006) and across-year floodplain habitat (P = 0.007) comparisons were significantly different. Data suggests that hens must arrive early on the breeding grounds in order to build reserves for egg laying when the appropriate invertebrate communities persist at elevated levels and before cavities become concealed in a dense forest canopy.