RED SALMON COUNTS FROM OBSERVATION TOWERS 



357 



liave also been used as means of determining the 

 escapements. However, detailed surveys of the 

 numerous and widely distributed spawnin<r areas 

 of Iliamna Lake in the past 5 years have failed 

 to reveal more than 20 percent of the estimated 

 total Kvichak migration as determined from 

 observation-tower counts. This pei'centage has 

 varied from year to year; also the relative number 

 of total spawners in each major spawning area 

 has varied from year to year, irrespective of the 

 size of the escapement. Hence, indices based on 

 counts of spawning fish in the Kvichak River 

 system appear to be highly unreliable. 



METHOD OF OBSERVATION 



The sample-count method of estimating escape- 

 ments is based on the migi'atory habits of red 

 salmon as they move up the river toward the area 

 in which the}' originated. Salmon fii"st appear in 

 the Kvichak River the latter part of June and 

 continue to pass upriver throughout July, with a 

 migration peak occurring near the middle of the 

 month. Once in the river the fish seek places where 

 currents are reduced. Since areas of low currents 

 are usually found near the bottom and next to the 

 banks, the migrants follow restricted paths close 

 to the shore where they can be readily observed. 

 This habit is quite consistent in sections above 

 tidewater where the river is confined to a single 

 channel and where there are swift midstream 

 currents. 



It has been found that current velocities 6 inches 

 from the bottom at the sites where the counts are 

 taken range from 1.55 to 2.13 ft. a second, while 

 those 6 inches from the surface range from 2.34 

 to 3.34 ft. a second. In 1959, the migrants passed 

 upriver at an average relative speed of 1.52 ft. a 

 second against the bottom currents. 



Varying water levels influence the paths taken 

 by the fish. Red salmon generally follow the banks 

 more closely when the water level is high because 

 of the relatively greater area of deep and quiet 

 water close to shore. Kvichak water levels at the 

 fir.st of July have varied more than 2 feet from 

 year to year, and usualh' rise from 8 to 14 inches 

 during that month. 



Fish apjiear in small se]>arate schools when the 

 size of the run is low. As the magnit nde increases, 

 the schools become larger and extend until the fisli 

 are ])assiiig in continuous bands. The separate 



schools behave somewhat erratically, but when the 

 migration intensity is high the fish are less wary 

 and follow the banks more closely. The direction 

 of migration is usually continuously upi-iver. 

 Only a few fish return downstream at the Kvichak 

 counting site, and these occur invariably near tlie 

 end of the migration. Slack water areas and 

 sloughs are utilized by some individuals to rest. 

 These fish conunouly bear injuries from gill nets 

 and predators in the ocean. 



From extensive beach seining the Kvichak River 

 escajDements have been found to consist almost 

 entirely of red salmon. Other species of salmon 

 occur in the seine samples largely near the end of 

 the migration, and make up less than 0.5 percent 

 of the total escapements. The other species are 

 omitted from the counts when identified from the 

 towers, and therefore the possible error caused by 

 including species other than red salmon is very 

 small. 



The visual counts are taken from towers erected 

 at suitable locations, aided by use of background 

 panels to silhouette the fish, turbidence reducers to 

 smooth the surface, and other special counting 

 aids. Once all migration paths can be observed 

 clearly, it is relatively simple to count during sam- 

 pling periods of predetermined length and to 

 estimate the total escapement with a high degree 

 of accuracy. 



Tower Locations 



The migratory habits of the fish and the physi- 

 cal characteristics of the Kvichak River limit the 

 number of suitable sites for observation towers. 

 The river below Kaskanak Flats is turbid and sub- 

 ject to tides that periodically revei'se current flow. 

 Migrants filter through the channels of the fiats 

 in numerous locations. Consequently, the most 

 practical counting sites are limited to the area 

 above the flats. The counting sites selected are 

 near the outlet of Iliamna Lake close to the village 

 of Igiugig and more than 50 miles ujjriver (fig. 1). 



For peak efficiency, towers should be located 

 where the moving bands of fisli are constricted in 

 width and jiass without deviating from near the 

 base of the structures. Since the river near Igiugig 

 is split by an island, three towei-s are required 

 (fig. 2). Towers Xo. 1 and No. 2 provide for 

 sampling of the two primary paths in tlie main 

 channel, which has a width of 380 feet and a 

 maximum dcptli of about U; feet. Tower No. 3 



