managing the salmon resources in southeastern 

 Alaska should investigate the practical feasibility of 

 establishing and maintaining an open-ended com- 

 puter simulation model of this resource for use as a 

 basic planning tool. Alternative policies could be 

 evaluated on this type of simulation model to deter- 

 mine which ones are most promising. Such a simula- 

 tion model would be an invaluable aid in designing 

 future tagging experiments. This type of model could 

 be used to optimize the distribution of tagging and 

 recovery effort and would aid in development of 

 techniques for analyzing and interpreting tagging 

 data. A more detailed discussion of the uses of simula- 

 tion models for planning resource management 

 strategies is given by Paulik (1967). 



SUMMARY 



Records of pink salmon tagging experiments 

 conducted by the U.S. Fish and Wildlife Service in 

 the northern part of southeastern Alaska each year 

 from 1938 through 1942 and in 1945 were transferred 

 to punch cards and analyzed using electronic data 

 processing equipment. 



During these 6 yr 55 releases were made: 20,472 fish 

 were tagged and 7,027 of the tags were recovered. No 

 spawning ground surveys were made, and only 30 

 tagged fish were recovered in streams. Total 

 recoveries from all sources other than streams varied 

 from a low of 28.6% in 1940 to a high of 38.7% in 1941; 

 79.8% of the recoveries were from traps and 16.1% 

 from seines. The analysis of tag recoveries is 

 supplemented by an analysis of the recorded daily 

 catches of pink salmon from about 100 traps that 

 operated in the northern part of southeastern Alaska 

 during these years. 



In most years (with the exception of 1941 and to a 

 lesser extent 1942) releases of tagged fish were 

 restricted to a few localities and were not well dis- 

 tributed in time. No multiple recaptures were 

 reported. Little information could be found on the size 

 and distribution of the seine fleet, and the recorded 

 data accompanying tag recoveries by seines were in- 

 complete. The analysis is restricted for the most part 

 to trap catches and trap recoveries. 



Recoveries in the southern part of southeastern 

 Alaska, i.e., to the south of Kuiu, Kupreanof, and 

 Mitkof islands, were infrequent from these releases, 

 which were made in the northern part. The lack of 

 movement to the south confirms findings of earlier in- 

 vestigators that exchange of pink salmon between the 

 northern and southern parts of southeastern Alaska is 

 of no practical significance and that the pink salmon 

 runs in these areas should be regarded as distinct from 

 one another and managed as two independent units. 



None of the tagged fish released in middle Chatham 

 Strait were recovered in Salisbury Sound on the west 

 coast of Baranof Island, which indicates that there 

 was no westward movement through Peril Strait. 



Limited tagging in Salisbury Sound in 1945 suggests 

 that Peril Strait is not an important entryway into in- 

 ner waters, but the experiments were too few to es- 

 tablish the magnitude of the eastward movement 

 through Peril Strait. 



The geographical distribution of recoveries showed 

 that one main migration route is through Icy Strait 

 and upper Chatham Strait to middle Chatham Strait; 

 from there the pink salmon disperse eastward into 

 Frederick Sound and Stephens Passage. The other 

 major passageway from the ocean is through lower 

 Chatham Strait into Frederick Sound and Stephens 

 Passage. While some segment of each tagged group 

 moved rapidly along one of the main migratory paths, 

 others lagged behind, with the result that tagged fish 

 were distributed along the entire way traversed and 

 recoveries continued to be made from all points along 

 the route until the end of the season. Some fish re- 

 leased in Icy Strait moved through upper and middle 

 Chatham Strait in 1 to 2 days and were found in 

 Frederick Sound within 3 to 5 days. 



The data show there was not a major exchange of 

 fish between lower Chatham Strait and middle 

 Chatham Strait. The stocks found in Icy Strait, upper 

 Chatham Strait, and middle Chatham Strait are to a 

 large extent distinct from those found in lower 

 Chatham Strait. 



Fish tagged later in the season exhibited a more 

 restricted range of movements about the point of 

 release than those tagged earlier. Groups tagged 

 earlier in the season contained larger proportions of 

 fish destined for spawning grounds distant from the 

 release points. 



The movement pattern of a group of tagged pink 

 salmon appears to be made up of a random compo- 

 nent superimposed upon a directional component. 

 The relative strengths of these two components vary 

 considerably from year to year. Also the particular 

 mixture of racial stocks in a given sector of the fishing 

 ground at any time during the season varies greatly 

 from year to year. 



Analysis of data on catch per trap per week con- 

 firmed the major migratory routes as determined from 

 the analysis of the tagging data. Catches increased in 

 an orderly fashion for traps along the northern entry 

 route. The buildup of catches in Frederick Sound and 

 Stephens Passage was earlier than in middle 

 Chatham Strait but never preceded the increase in 

 lower Chatham Strait. The timing of catch increases 

 in lower Chatham Strait was similar to that in inner 

 Icy Strait. The populations in Frederick Sound and 

 Stephens Passage were composed of a mixture of fish 

 that had entered through both the northern and the 

 southern entry ways. 



The rate of movement of the center of density of a 

 single release of tagged fish was estimated from 

 locations of median tag recaptures on successive days; 

 the rate of movement of the leading edge was es- 

 timated from the second farthest recovery each day. 



38 



