222 
WATERFOWL BREEDING GROUND SURVEY IN NEW YORK, 1953 
Dirck Benson, Donald D. Foley and Donald Schierbaum | 
Introduction 
The third consecutive aerial waterfowl production survey based upon a 
sampling plan devised by the Fish and Wildlife Service was completed during 
July 6 to 10, 1953. As in 1951 and 1952, when the work was more a pilot study, 
the same system and methods were used, although the coverages of certain strata 
this year were modified at the suggestion of the Waterfowl Biometrician, The 
flying has been done consistently by John A. Schempp and Ed Maunton. Figure I 
Shows the location of transects and cluster problem areas covered by air, and also 
the 4 rivers sampled by canoe by Benson and Foley for comparative purposes. 
As additional measurements of.the waterfowl populations, 1,471 birds were 
banded during the period June 15 to September 15, 1953 with Schierbaum correlating 
this phase of the work, Both drive-trapping and bait trapping were done, the former 
on an experimental scale in the western part of the State with the help of Oliver H. 
Hewitt and students of Cornell University. Bait trapping locations included Lake 
Alice, Kings Bay, Perch Lake, Oak Orchard, Montezuma Refuge (operated by the 
Fish and Wildlife Service), Tomhannock Reservoir and Basic Creek Reservoir. 
The breeding ground survey of sample areas by ground observers, begun in 
1949, was coordinated by the waterfowl project this year and considerably broadened 
and standardized. It is covered in another section of this report. 
Methods of Sampling and Area Covered 
The aerial survey methods consisted of two types. One was the straight-line 
transects over the largest or most important phytogeographic stratum of the State. 
The boundaries of these strata in 1951 and 1952 were taken from Braun's classifica- 
tion, but for 1953 were modified toward better agreement with forest and soil types 
and with waterfowl productivity. Figure 2 indicates the stratum as used for 1953. 
Transects were laid out as parallel strips 1/8-mile wide approximately at 
right angles to the drainage, of a sufficient sample mileage in each Stratum to give 
an error as a waterfowl index of not greater than plus or minus 30 percent. Due to 
the extremely sparse populations found on the Adirondack transects the first 2 years, 
transects were omitted from this stratum in 1953. Likewise, the 2 smallest strata 
(Champlain and New England) were disregarded. 
The cluster problem areas of each stratum included large rivers, lakes, 
swamps, and marshes which differed markedly from their surrounding habitat, and 
which presented special problems. Sufficient sampling of areas over 500 acres in 
size in each stratum was accomplished so that the resultant error should not have 
exceeded plus or minus 30 percent for marshes and swamps and 20 percent for 
rivers, while for lakes the minimum shoreline mileage recommended by the 
Waterfowl Biometrician was used. A strip 1/4-mile wide was covered over the 
marshes and swamps and converted into square miles for the production index, 
while for rivers and lakes a linear mileage basis was used, 
