RED SALMON COUNTS FROM OBSERVATION TOWERS 



363 



Table 3. — Comparison of counts during the same period by different observers 



peared when the migration passed 450 fish per 

 5-minute interval, but even these errors followed 

 a random pattern and tended to average out. 

 However, because of the greater magnitude and 

 importance of these errors, it is imperative to in- 

 crease the number of samples taken with an in- 

 crease in intensity of the run. This greatly in- 

 creases the reliability of the calculated estimates. 



^25 



^20- 



<.* 

 <s 



a 



5 15- 



.8 

 I 10 H 



5 



I 





• • 





200 400 600 800 1000 

 Number of Fish Counted /5 Minute Period 



Figure. 8. — Relation between migration intensity and 

 variations between counters. 



When extremely heavy runs occur and samples 

 must be obtained by estimating groups of 10 fish, 

 counting errors are undoubtedly greater than 

 when individual fish can be tallied. Heavy runs 

 requiring such counting procedures are infrequent 

 and usually occur past one tower at a time. On 

 the Kvichak River, such runs appeared only dur- 

 ing the intense 1956 escapement (July 11-21) and 

 briefly in 1957 (July 12-14). 



Effect of Weather Conditions 



Sun glare, overcast skies, wind, and rain lower 

 visibility and increase the difficulties of obtaining 

 accurate sample counts. Glare may be trouble- 

 some during a 4-hour period daily when the sun 

 is bright and low over the water. Overcast skies 

 impart a dull appearance to the surface of the 

 water, while upriver winds and rain disturb the 

 surface and thereby distort the outlines of objects 

 under the surface. The effects of these adverse 

 conditions, combined with the normally turbulent 

 water surface, make turbulence reducers essential. 

 Then, with the aid of Polaroid glasses, the ob- 

 servers can count the passing fish accurately at 

 almost any part of a given day. 



Moderately turbid water, wliich occurs irregu- 

 larly throughout the season, is the most important 

 factor affecting the accuracy of the counts. This 

 turbidity is the result of heavy breakers against 

 the beach at the outlet of Iliamna Lake, which 

 occur with strong east winds. The pounding 

 causes excessive amounts of silt to be suspended 



