Document Type

Thesis - Open Access

Award Date


Degree Name

Master of Science (MS)


Wildlife Management


An evaluation of the roadside technique for censusing breeding waterfowl was conducted in east-central South Dakota. On the 54-mile transect used in 1968, number of coutns necessary to be within 20 percent of population mean with specified confidence of 0.90 for blue-winged teal (Anas discors) and gadwall (Anas strepera) was five and three using miles or wetlands. Within same limits counts needed for mallard (Anas platyrhynchos) were at least four for miles and three for wetlands. Shorter transects, used in 1967 required more counts. Transects 21 miles long required up to 10 counts based on miles and 11 based on wetlands. The seven-mile transect required 25 counts based on miles and 10 based on wetlands. When wetland numbers exceeded mile numbers more counts were necessary for miles and the reverse was true when mile numbers exceeded wetland numbers. Optimum period for censusing blue-winged teal was May 20 to June 6 in 1967, while optimum periods for mallard and gadwall were May 5 to 25 and May 25 to June 12. Optimum periods in 1968 were May 5 through 30 for mallard and May 23 through June 14 for blue-winged teal and gadwall. New vegetation growth did not generally affect observability of ducks until the third week in May during both years. Vegetation on each wetland was placed into of three categories: (1) 0 to 30 percent covered, (2) 35 to 60 percent covered, and (3) 65 to 100 percent covered. In 1967 fewer breeding pairs of blue-winged teal were observed on wetlands as vegetation increased. However, in 1968 no pattern of breeding pair means in vegetative classes within Type 3 wetlands was found and not all differences were significant. Significantly more pairs of blue-winged teal, gadwall and mallard were attracted to type 4 wetlands containing 35 to 60 percent cover than other cover classes. Significantly more ducks were observed on type 4 wetlands than Type 3 since Type 4 wetlands were larger and more permanent. However, more breeding pairs per acre were counted on smaller wetlands. Wind and light conditions exerted no significant influence on counts of breeding pairs of blue-winged teal, gadwall, and mallard on the Sinai study area. Flushing counts among vegetative classes for blue-winged teal and gadwall were significantly different for vegetative Class 3 but not Classes 1 or 2. Roadside and flushing counts were significantly different only in Class 2 for mallards. Observer bias should be recognized and when possible, compensated for in conducting waterfowl counts.

Library of Congress Subject Headings

Waterfowl -- South Dakota


Includes bibliographical references (pages 59-60)



Number of Pages



South Dakota State University


No Copyright - Non-Commercial Use Only