Wood Duck Brood Movements, Habitat Use, and Survival Rates on Two Prairie Rivers in South Dakota

Wood ducks, have recently expanded their breeding range throughout South Dakota. Brood ecology may differ from traditional breeding areas, and between areas within South Dakota, due to difference in landscape. Wood duck brood movements, habitat use, and survival were studied by radio-marking hens nesting in boxes on the Big Sioux River, 1992-1994, and Maple River, 1993-1994, in South Dakota. The Big Sioux River is a lower perennial stream bordered by a narrow wooded corridor and a fairly wide floodplain. The surrounding area has a well integrated drainage system. The Maple River Supports cattail and bulrush along its margins due to the semipermanent water regime. The landscape has few trees and a non-integrated drainage system. Wood duck nest box use was near capacity with a large portion of dump nests on both study areas. Broods were highly mobile, in terms of travel distances (medians – 3.45 -6.65 km for each year and study area) and frequency of movements, in comparison with most wood duck brood studies. Movement was associated with changes in water conditions rather than movement at a particular age. The probability of moving to a new brood rearing areas was positively associated with weekly changes in river water height (P \u3c0.001) and negatively associated with the CV of daily changes in water height (P \u3c0.001). Fewer broods areas were used on the Big Sioux River in 1993 (P=0.05) when river discharge was highest. Total travel distance was longer (P=0.02) on the Maple River versus the Big Sioux River and was longer (P=0.04) on the Big Sioux River in 1992, when water levels were lowest of the 3 years studied. Broods on each study area used the river as a principle travel route. Hens with broods on the Big Sioux River selected forested wetlands most often and semipermanent wetlands least (P\u3c0.05). However, broods leaving the nest before 15 June in 1992, when the floodplain was dry, used more semipermanent wetlands and tributaries than broods later that year or in subsequent years. There was no trend in habitat use of successive brood rearing areas which would explain movements to new areas. Broods on the Maple River selected the river rather than surrounding temporary, seasonal, or semipermanent wetlands. Brood survival (≥1 duckling reared to 30 days) was higher (P=0.04) on the Big Sioux River than on the Maple River. Duckling survival did not differ among areas or years. Initial brood size (P=0.0002) and the date of nest exodus (P=0.003) were inversely related to duckling survival. In addition, duckling survival on the Maple River was positively correlated with the number of brood rearing areas (P=0.03) and negatively correlated with the age at the last brood count (P=0.05); there was a weak (P=0.11) negative association with overall travel distance. Overall duckling survival rates were comparable to previous studies (range-26-44% for each year and study area) Protection of forested wetlands and emergent and overhanging vegetation along streams would help to ensure brood rearing habitat in wood duck breeding areas. Nest boxes should be placed near water routes to potential brood rearing areas or near brood rearing habitat if travel is overland. The negative relationship between initial brood size and duckling survival indicates an additional negative effect of dump nesting which has not been detected in previous wood duck brood survival studies. Efforts should be made to reduce dumb nesting in breeding areas through nest box placement. The effect of brood size on duckling survival warrants further investigation, with efforts to account for correlated factors such as time of season