and the late run (in July and August) has pri- 

 marily fish bound for Chignik Lake. 



Fuirhik system. — The distribution of sockeye 

 salmon on the spawning grounds in the Karluk 

 system has been well documented from data col- 

 lected on weekly surveys carried out on foot 

 each summer in 1947-62. Salmon that enter the 

 system during May, June, and July tend to oc- 

 cupy all tributaries; those entering during 

 August and September tend to occupy only the 

 large tributaries at the south end of the lake, the 

 lake beaches, and the outlet river.' 



Sockeye salmon in Karluk Lake have a 

 marked temporal as well as spatial segregation 

 by age group (Rounsefell, 1958; also see foot- 

 note 9) . In general, salmon that spend the long- 

 est time in the ocean (3 years or more) return 

 early in the spawning season and those that 

 spend the least time at sea return late. Since late 

 spawners tend to occupy the large terminal 

 streams, lake beaches, and the upper end of Kar- 

 luk River, these areas usually receive a dispro- 

 portionately large number of smaller 2-ocean 

 fish. 



The Chignik and Karluk systems contrast 

 sharply with the systems tributary to Bristol 

 Bay in having temporal segregation of spawn- 

 ing units in the trunk river. The absence of 

 marked segregation in Bristol Bay systems may 

 be the result of the short duration of these runs. 

 In general, the Bristol Bay runs last about 4 

 weeks, and more than 50 percent of each run 

 enters the fishing areas and trunk streams dur- 

 ing a few days in early July. In the Chignik and 

 Karluk systems the escapements enter the trunk 

 stream over a 2- to 4-month period. 



POTENTIAL SPAWNING POPULATION 



The physical capacity of the rivers or beaches 

 to contain adult sockeye salmon imposes a limit 

 on the number of fish that could spawn in each 

 system. It was obvious, however, that a large 

 portion of most of the rivers and beaches was 

 in fact unsuitable for spawning and that the 

 amount of space required for a pair of salmon to 

 spawn varied among different types of spawn- 

 ing grounds. Therefore, to estimate the capacity 

 of each river or beach in terms of numbers of 

 females we determined the area of potential 



spawning ground and divided this figure by the 

 estimated area required per female. 



Estimates of Area of Potential Spawning Grounds 



The amount of potential spawning ground for 

 streams and for lake beaches was measured to 

 determine if the area of suitable spawning 

 grounds might limit salmon production in any 

 of the river systems. The data obtained are first 

 approximations because complete and" accurate 

 surveys of all spawning grounds could not be 

 completed in 2 years. Estimates of suitable po- 

 tential spawning area were made for the Wood, 

 Kvichak, Naknek, and Karluk systems (table 

 12). 



During the 2-year study period, catalogs of 

 spawning grounds, which contain descriptions 

 of the physical characteristics and estimates of 

 the area of the potential or utilized spawning 

 grounds and estimates of annual escapements 



T.\BLE \2.- Estimates of surface area of potential sockeye 

 salmon spawning grounds in streams and lake beaches of 

 Wood, Kvichak, Naknek, and Karluk systems 



9 Owen. John B.. Charles Y. Conkle, and Robert F. Raleigh. 1962. 

 Factors possibly affecting production of sockeye salmon in Karluk 

 River. Alaska. Bur. Connmer. Fish. Biol. Lab., Auke Bay. Alaska. 

 [Manuscript, 57 pp.] 



' Estimates based on grave! types only. 



- Minimum estimate based on incomplete information. 



^ No estimate made. 



' Excludes North Arm: North Arm is considered separately because 

 it constitutes a rearing area for juvenile salmon discrete from the rest 

 of Naknek Lake. 



^ Only the beach spawning areas of Thumb and O'Malley Lakes are 

 included here. 



422 



U.S. FISH AND WILDLIFE SERVICE 



