132 
AEC Reservation and should remain permanently as it now is. Unit II is that 
portion of the Columbia and the lower few miles of the Snake River which will be 
completely inundated in the impoundment behind McNary Dam, probably in late 
1952. Unit III embraces the future impoundment area of the proposed Celilo Falls 
Dam and upstream to McNary Dam. Unit IV includes the impoundment area of the 
proposed Ice Harbor Dam on the Lower Snake River upstream as far as Matthews. 
Within this river system, a distance of 218 miles, lie 86 islands used as goose 
nesting habitat. Upon completion of McNary Dam, 24 of these islands in Unit II 
will be lost for future use. There are 27 islands in Unit III below McNary Dam, 
14 in Unit IV, and the remaining 21 in Unit I. Each year, as time and manpower 
permitted, the amount of river surveyed has been increased, The breeding 
population figures are then reduced to birds per river mile so that an index can 
be applied to the entire habitat area. In 1952, 76 islands over a distance of 182 
miles were surveyed in April. Most of the islands in Unit I were re-checked in 
May by General Electric technical personnel from the Hanford AEC for late | 
nesting information. 
Nesting and hatching in 1952 was not only earlier than in 1951, but it was 
completed in a shorter span of time. The first known hatching date in 1951 was 
April 11 with the peak of the hatch occurring about May 10 to 15. The latest known 
hatching date was between June 10 and 12. On April 9 in 1952, 2.8 percent of 470 
nests observed had already hatched. Over 33 percent had hatched before April 24 
and less than 4 percent of the nests were still being incubated on May 13. Prior 
to banding, late in June, 97 complete and separate broods had been counted with a 
total of 445 goslings for an average brood size of 4.58, all age classes combined. 
The average size of 38 broods counted in 1951 was only 3.5 goslings per brood. 
The average size clutch of 509 nests in 1952 was 5.1 eggs, including incomplete 
clutches. The nesting loss was 3.7 percent of 509 nests observed. The causes 
ranged from human theft on one island through several types of predation elsewhere. 
The lowest nesting loss was 2.9 percent in Unit III up to a-high of 9.6 percent on 
the smaller Snake River in Unit IV. 
A 3-year summary of the resident goose population is included in Table V. 
With the exception of Units II and [II, pairs of geese have increased each year 
throughout the area, although in 1952, the total population dropped. This may have 
been due to an earlier season in which more of the yearlings and other non-breeders 
already had departed. Early, inconclusive banding returns indicate that sub-adult 
geese may spend their yearling summer wandering northward from the breeding 
grounds in the Pacific northwest. Then, after reaching sexual maturity, they 
remain south to nest. The greater number of nests observed each successive year | 
does not necessarily indicate a larger nesting population, but rather a more efficient 
method of locating nests in a shorter period of time on each island as experience is 
gained. Probably the most accurate indicator of the resident population fluctuation 
is the pairs of geese counted. The greatest effort is toward making this count as 
complete and accurate as possible each year. With an average of 4.63 pair of geese 
for each of the 218 miles, the 1952 breeding population index was 1,010 pair of 
geese compared with 970 pair in 1951. With nesting losses and average brood size 
accounted for, the 1952 production index will be about 4,450 young compared to 
3,300 in 1951, or an increased production of 25 percent. The goose production 
figures derived from this river inventory are separate and in addition to those 
contained in the general waterfowl inventory for the remainder of eastern and 
central Washington. 
