AN EVALUATION OF WATERFOWL SURVEYS ON THE

Fifteen simultaneous air and ground censuses of eight species of wintering sea ducks were con- ducted on a 22-mile section of the New Hampshire and Massachusetts coastline dunng 196S71. In com- parison wii surface counts, there was extreme varia'don in the aerial survey results even with identical aircraft, weather conditions, and observer. These discrepancies cast serious doubt on the usefulness of the aerial survey technique as an indicator of population trends of sea ducks. Size of aircraft, amount of cloud cover, and surface of ocean influenced the accuracy of aerial surveys, but these factors accounted for only a small part of the variability. A system of ecologically stratified ground censuses is recom- mended as an alternative to or a modification of the present aerial survey technique. Abstract: Fifteen simultaneous air and ground censuses of eight species of wintering sea ducks were con- ducted on a 22-mile section of the New Hampshire and Massachusetts coastline dunng 196S71. In com- parison wii surface counts, there was extreme varia'don in the aerial survey results even with identical aircraft, weather conditions, and observer. These discrepancies cast serious doubt on the usefulness of the aerial survey technique as an indicator of population trends of sea ducks. Size of aircraft, amount of cloud cover, and surface of ocean influenced the accuracy of aerial surveys, but these factors accounted for only a small part of the variability. A system of ecologically stratified ground censuses is recom- mended as an alternative to or a modification of the present aerial survey technique.