189 
Table III - Air to Ground Comparison of Counts by Habitat Types, 1951 and 1952 
ee ee ne ee ee SS Sen. By sabatet Fypes, Fol amie te 
Average Ratio Chi-square Calculations 
Habitat Type Ground /Air 1951 1952 
2 2 
1951 1952 x x 65 (2DF) X x°.5 (4DF) 
Large Lakes 2.36 1. 66 .712* 1, 386 8.872 3.357 
Small Lakes and 1. 82 . 89 ~ 922% 1.386 .201* 3.357 
Ponds’ 
Irrigation Canals 1, 36 1,44 . 838% 1.386 2.504% 3.357 
Irrigation Ditches 2.30 1.26 .197% 1.386 1.750% 2.366 (3DF) 
Drain Ditches 7.81 2.76 Insufficient Data 2.776 2.366 
River 5.27 1.60 Insufficient Data . 543% 3.357 
Creeks 3.56 1.81 Insufficient Data 601% 1. 386 (2DF) 
Slough 2.96 2.14 1,932 1.386 3.424 3.57 

* Good fit - average ratio can be applied. | 
Sources of Variation Between 1951 and 1952 
As mentioned previously, the air-to-ground ratio differed between the two 
years. This was apparently caused by using a different observer for each year. 
The first year all aerial observing was done by the writer, while the second year's 
observations were made by Ervin L. Boeker. The writer had previously done no 
aerial counting of waterfowl and had very little flying experience. Mr. Boeker, on 
the other hand, was a capable pilot with experience in aerial census. Obvisouly, 
experience is one of the prime factors in the differences between the ratios obtained 
for each year. 
It is possible to point out where experience helps in counting. By referring 
to Table III, it may be seen that without exception Mr. Boeker's air-to-ground ratio 
for each habitat type was much smaller than the writer's; however, the differences 
were larger on those types with heavy cover as river, creeks and drains. Therefore, 
it is possible to conclude that an experienced observer will see an even greater per- 
cent of birds on those types with dense cover than will an inexperienced observer. 
Experience also plays an important part in observing all species of ducks. 
This is emphasized by Table IV, which shows the percent of ground-counted species 
seen from the air. Aithough the difference between mallard-like ducks was not great’ 
for the two years, only 8 percent of the teals were observed in 1951 as compared to 
59 percent in 1952. 
