obtained from the Canadian government by the U. S. 
Fish and Wildlife Service. 
U. S. Federal Power Commission. The report, Water 
Power of Southeastern Alaska, published with the 
cooperation of the U. S. Forest Service, provides 
discharge rates and stream drainage areas, and other 
information about a number of important salmon 
streams. 
U._S. Forest Service. Data on stream characteristics 
and salmon escapements are available on several 
streams in records of studies conducted by this agency 
on the effects of logging on the physical makeup of 
streams. 
U._S. Geological Survey. Aerial photographs from 
this agency are the primary source for measurements 
of stream distances and areas and for valley features 
not visible from the ground. These photographs, 
which are of most of the streams in Southeastern 
Alaska, were made by the U. S. Navy (Patrol Squad- 
ron Four) in 1948. 
Local residents. Another source is the descriptive in- 
formation on several major streams provided by local 
residents. 
LIMITATIONS OF DATA 
Escapement estimates obtained by visual means 
are often limited in accuracy because fish are not seen 
in turbid water, under overhanging stream banks, or 
in areas inaccessible to observers. Actual counts 
throughout the duration of the salmon run past a count- 
ing weir or tower are relatively accurate estimates of 
total escapement. However, it is not economically 
feasible to establish a weir on each stream and escape- 
ment surveys are the only source of information for a 
large part of the area which must be covered. The 
value of the catalog as a history of the salmon escape- 
ments can be realized only if its limitations are fully 
known. 
Escapement Estimates 
Escapement estimates do not indicate the actual 
total escapement. At no time are all the salmon in the 
stream simultaneously since the spawning run extends 
over a period of weeks. Therefore, each escapement 
estimate is an index of the relative abundance at the 
time of survey. 
The maximum estimate determined by survey 
methods at about the peak of the run is used as an esti- 
mate of the relative abundance of the total escape- 
ment. Reliable indices of relative abundance from 
year to year can be made only if the surveys are com- 
parable. Evaluation of the following factors is 
necessary to determine the accuracy of the escape-~ 
ment estimates. 
Observers. --The escapement records are 
from many different observers. Variability in 
estimating the number of salmon in a given area by 
different observers should be considered in judging 
the accuracy of the data. In general, with more 
observers variability increases. 
Survey systems. --Different survey systems 
have been used by the various agencies. Reliability 
of the escapement estimates varies with the systems 
used, 
Survey systems that employ standard count- 
ing techniques over standard distances are the most 
reliable method now available for comparison of 
abundance between years, particularly when streams 
can be only partially covered. Standard survey 
distances in comparatively long streams were not 
widely used prior to 1949. 
Survey systems requiring periodic visits 
to each stream during the spawning period are more 
reliable for estimates of peak abundance than systems 
requiring only one or two visits. The peak period of 
abundance in a salmon stream is usually relatively 
short and one or two visits may miss the peak. 
Type of suryey. --Two basic methods for cov- 
ering the streams during escapement surveys are being 
used. 
The oldest method is the ground survey 
where the observer follows the stream course on foot 
or in a skiff with an outboard motor, Most parts of 
the stream can be closely observed by this type of 
survey. 
The newer method is aerial survey. This is 
a fast, economical means of covering a large number 
of streams in a short period of time over stream dis- 
tances greater than is possible on the ground. This 
method requires experienced personnel familiar with 
ground surveys as well. 
Aerial surveys are best suited for large 
rivers and streams where ground coverage is limited 
usually to the lower portion of the stream near the 
banks. Ground surveys are more reliable than aerial 
surveys on small streams that offer poor visibility from 
the air. 
Observation conditions. --Weather is an 
extremely important factor in the reliability of 
escapement estimates. Most streams are very difficult 
to travel on the ground during flooding. Visibility 
is also greatly reduced because of turbid water. Any 
