187 
Aerial Surveys 
Aerial breeding-pair and brood counts were obtained by flying along each 
irrigation ditch or stream, and around each lake. Altitudes and speeds varied, but 
averaged around 100 feet and 60 miles per hour indicated air speed, respectively. 
During breeding-pair counts, ducks were recorded by habitat type, species, and, 
where possible, sex, thus permitting a closer evaluation of air and ground results. 
It was found impossible to obtain accurate species identification of broods from the 
air; therefore, only the number of young were recorded. 
Two aircraft were used during the course of this problem. The first was 
a four-place Cessna 170, and the second a two-place Cessna 140. The alternate use 
of either craft did not seem to influence the number of birds counted, but it was noted 
that it was much easier to observe from the smaller plane because of structural 
differences. 
Statistical Analysis 
Covering the study area both on the ground and in the air permits compar- 
ison of the two counts. By dividing the ground count by the comparable air count, 
the air-to-ground ratio of the coverage can be established. The average ratio of 
several coverages may then be statistically tested by means of a contingency table 
and the chi-square test to determine whether the air-to-ground ratio was reasonably 
consistent between coverages. With a calculated chi-square value less than the 
tabular value of chi-square .5, it may be concluded that the average ratio is a good 
fit, and could be applied to future aerial counts to obtain an accurate estimate of the 
numbers actually on the ground. 
Results 
A total of five flights were made in 1951 and six in 1952 on the study area. 
The data thus derived emphasize the efficiency of aerial as compared with ground 
work. During spring counts it required roughly 35 man-hours to cover the study 
area adequately on the ground. By comparison, only one hour and 15 to 30 minutes 
flying time was needed for comparable aerial coverage. During the brood-counting 
season, after all ditches became filled with water, ground coverage time jumped to 
an average of about 86 man hours for the two years, and air time to slightly less 
than two hours. 
. Breeding-Pair Survey. 
Aerial breeding-pair counts in the Fort Collins area are best taken during 
late April, May and early June for it is during this time that most territories are 
still active. The total counts of territories for both 1951 and 1952 are tabulated in 
Table II. This table also lists the air-to-ground ratio for each coverage and gives the 
results of the chi-square test. Thus it was determined that average air-to-ground 
ratio found each year was consistent, being 2.23 for 1951, and 1.52 for 1952. This 
implies that in working with a large study area similar to the township it would be 
possible to adjust aerial counts by means of peneral correction factors such as those 
given above, The differences between the 1951 and 1952 ratios were caused by using 
different air observers each year. This will be discussed later, 
