have very small populations. Because the maxi- 

 mum number of spawners estimated on beaches 

 is less than the maximum estimated in streams 

 (table 14), the spawning ground potential of 

 the beaches is probably not in excess of that of 

 the streams. 



The largest escapement of sockeye salmon to 

 the Wood system since counting from towers be- 

 gan in 1953 was 2,209,000 in 1959 (table 8). 

 Although some spawning grounds were over- 

 crowded in 1959, as evidenced by a large mortal- 

 ity of eggs, most were greatly underutilized. 

 Escapements in 1955-62 varied between 289,000 

 and 2,209,000 sockeye salmon. Thus, the poten- 

 tial capacity of the spawning grounds in the 

 Wood system greatly exceeds the normal escape- 

 ments. 



Kvichak system. — The area of potential 

 spawning grounds in rivers, streams, and ponds 

 in the Kvichak system was estimated to be 839.7 

 hectares (table 12), and the space required per 

 spawning female was estimated to be 2 m.-' A 

 capacity of 4,199,500 spawning females is cal- 

 culated from these data ; on the basis of 1 :1 sex 

 ratio, the potential capacity of the nonlake 

 spawning grounds in the Kvichak system is 

 about 8,399,000 spawners. 



We do not have enough data to make a direct 

 estimate of the area of the potential beach 

 spawning grounds in the Kvichak system, but we 

 can estimate it indirectly. The area used by 

 3,706,000 spawners in 1961 on 16 of 21 recog- 

 nized spawning beaches was 11 1.5 hectares (table 

 12). The area used by the 14,630,000 spawners 

 of 1960 would have been 435 hectares if we 

 assume a constant relation between numbers of 

 spawners and area used (the ratio of spawners 

 in 1961 and 1960 was 1:3.9). Because only 75 

 percent of the beaches known to be used have 

 been surveyed, the estimate of 435 hectares is 

 further expanded by 4 3, to give 579.9 hectares 

 of beach actually used in 1960. Because only 

 beach areas used for spawning have been con- 

 sidered and the average use of potential spawn- 

 ing grounds in the streams has been 48 percent, 

 the amount of potential beach spawning grounds 

 was calculated by assuming the same rate. This 

 computation gives an estimate of 1,208 hectares 

 of potential beach spawning grounds with a ca- 

 pacity of 12,080,000 spawners. 



On the basis of the above information, we 

 postulate a potential spawning population of 



about 20,500,000 sockeye .salmon for the Kvichak 

 system. This figure is greater than the maximum 

 observed escapement through 1962. Escape- 

 ments since 1955 have varied between 251,000 

 and 14,630,000 (table 8).'" 



Naknek system. — The amount of potential 

 spawning ground in the Naknek .sy.stem, the 

 space required per female, and the capacity of 

 the spawning ground were calculated for each of 

 several types of streams. We made estimates 

 of the total potential of the beach spawning 

 grounds. 



The total number of potential redd sites — 

 calculated from estimates of the amount of 

 spawning ground available and the size of redd 

 sites — is 624,000 for the streams of the Naknek 

 .system (table 13). An estimate of 20,000 redd 

 sites for the beach spawning grounds of Gros- 

 venor Lake is based on past spawning ground 

 surveys. Other beach spawning grounds exist, 

 but their capacity is unknown. 



Since estimates of the size of redd sites were 

 based on observations of spawning density on 

 single days, the effect of spawning in waves or 

 surges must be considered. Three waves of fish 

 commonly enter the Brooks River (Hartman, 

 Merrell, and Painter, 1964), which is the major 

 interconnecting stream that produces .sockeye 

 salmon. The number of redd sites in the inter- 

 connecting streams has been expanded, there- 

 fore, by a factor of 3 to produce a more realistic 

 estimate of the capacity of these areas to receive 

 spawning salmon. 



If all of these factors are considered and a 

 1 :1 sex ratio is assumed, the estimated spawning 

 capacity of the Naknek system is about 670,000 

 females, or 1,340,000 fi.sh of both sexes. Escape- 

 ments in recent years varied between 273,000 

 and 2,232,000 salmon (table 8). Escapements 

 exceeded 1,260,000 twice in the 14 years from 

 1949 to 1962. 



Karluk system. — Estimates of the size of a 

 redd site for several tributaries in the Karluk 

 sy.stem are about 2 m.-, and because estimates 

 are not available for other areas, 2 m.- is used 

 for all. The estimated number of redd sites for 

 streams and lake beaches is 174,000 (table 13), 

 and (assuming a 1:1 sex ratio) the capacity is 



'" The escapement in 1965 was 24. :I million. Some of the spawning 

 grounds had few adult fish; others were so densely occupied that super- 

 imposition of ledds and retention of eggs caused a great loss of eggs. 

 On most spawning grounds, however, density appeared to be optima] 

 and confirmed the figure given above. 



SOCKEYE SALMON IN MAJOR RIVER SYSTEMS IN SOUTHWESTERN ALASKA 



425 



