SELECTION OF SPAWNING SITES BY SOCKEYE SALMON 



IN SMALL STREAMS 



David T. Hoopes^ 



ABSTRACT 



This study was undertaken in three small western Alaska streams to identify factors 

 that influence selection of spawning sites by sockeye salmon. Temporal and spatial 

 distribution patterns were relatively constant from year to year despite large fluctuations 

 in the number of spawners. The selection of a spawning site was more closely related 

 to composition of the stream bottom than to gradient, water depth and velocity, or cover. 



The selection of spawning sites by salmon and 

 the factors that influence selection bear directly 

 on the problem of estimating the carrying capac- 

 ities of spawning grounds. Such estimates can- 

 not be developed for returning adults without 

 knowledge of what constitutes an acceptable 

 spawning area. In this study I relate the distri- 

 bution of spawning sockeye salmon, Oncorhyn- 

 chus nerka, in three small streams to the physical 

 characteristics of the streams. The work report- 

 ed was done at Brooks Lake, Alaska, in 1959, 

 1960, and 1961 and is one phase of an overall 

 investigation by the Bureau of Commercial Fish- 

 eries (now the National Marine Fisheries Ser- 

 vice) on the freshwater ecology of sockeye 

 salmon. 



The three study streams. Hidden, Up-a-tree, 

 and One Shot Creeks, are lateral' tributaries of 

 Brooks Lake (Figure 1). Hidden Creek enters 

 Brooks Lake in the southeastern shore about 

 10 km from the outlet, Brooks River. The other 

 two streams lie between Brooks River and Hid- 

 den Creek. Up-a-tree Creek, the longest lateral 

 stream in the Brooks Lake system, enters the 

 lake at the southeastern corner; One Shot Creek, 

 the shortest of the major spawning tributaries, 

 flows into the lake on the south shore not far 

 from Hidden Creek. Some physical features of 



' National Marine Fisheries Service, Auke Bay Fish- 

 eries Laboratory, Auke Bay, AK 99821. 



' The term "lateral tributaries" refers to streams en- 

 tering a lake from shores that lie roughly parallel to 

 the lake's long axis. 



the portions of the three streams accessible to 

 salmon are presented in Table 1. Included in 

 the table is the pool-riffle ratio, an indicator of 

 the proportion of stream length composed of 

 pools. What constitutes a pool is somewhat sub- 

 jective, but I considered a pool to be a place 

 where the stream was deeper and wider than 

 average, the current was appreciably slower 

 than immediately upstream or downstream, and 

 hiding places for fish were more extensive than 

 in adjacent parts of the stream. 



The three streams were alike in two significant 

 aspects — volume of flow and presence of beaver 

 dams. Flows in each stream varied from 0.23 

 to 0.34 m^s (cubic meter per second) (8.2 to 

 12.1 cfs) throughout the spawning season except 

 for some short periods of higher flows during 

 freshets. Beaver dams blocked the upstream mi- 

 gration of salmon except during brief periods 

 of unusually high water. 



DISTRIBUTION OF SPAWNERS 



The distribution of adult sockeye salmon was 

 observed closely and recorded from the time fish 

 arrived oflf the stream mouths until they had all 

 spawned. I recorded the number of fish ascend- 

 ing the streams each day and the distribution of 

 spawners along each stream on an average of 

 every 3 days. 



Sockeye salmon school oflf the mouths of trib- 

 utaries to Brooks Lake 1 to 3 weeks before 

 ascending the streams to spawn (Hartman and 



Manuscript accepted January 1972. 



FISHERY BULLETIN: VOL. 70, NO. 2. 1972. 



447 



