erics □ & < -J O u a _ - l4 ■ :S 4s FISH AND WILDLIFE CIRCULAR 59 ISSUED AT JUNEAU • SEPTEMBER 1959 The collection of papers included in this Circular are intended to acquaint the fishing industry of Alaska and neighboring regions with some aspects and preliminary results of investigations which are being undertaken by the Bureau of Commercial Fisheries and its contractors in Alaska. UNITED STATES DEPARTMENT OF THE INTERIOR, FRED A. SEATON, SECRETARY H AND WILDLIFE SERVICE. ARNIE I SUOMELA, COMMISSIONER BUREAU OF COMMERCIAL FISHERIES, DONALD L. McKERNAN, DIRECTOR ALASKA REC-ION, JOHN T. HHARRETT, REGIONAL DIRECTOR contents PINK SALMON MIGRATIONS ALONG 1 1 THE ALASKA PENINSULA Fredrik V. Thorsteinson Bureau of Commercial Fisheries SALMON MIGRATIONS IN SOUTHEASTERN ALASKA 8 1 Wallace H. Noerenberg Fisheries Research Institute University of Washington THE UNEXPECTED APPEARANCE OF PINK SALMON 13 1 IN THE KVICHAK RIVER Clarence D. Becker Fisheries Research Institute University of Washington GRAVEL REMOVAL AND THE FISHERIES 14 1 Robert McVey Branch of River Basin Studies Bureau of Commercial Fisheries ESTIMATING ABUNDANCE OF SALMON FINGERLING 19 1 Howard S. Sears Bureau of Commercial Fisheries ODD YEARS 1947-1955 EVEN YEARS 1946-1954 Figure 1. Seasonal occurrence of pink salmon as shown by average daily trap catches of difference are apparent. An alternation of cycles exists in which the early runs, those be- fore July 1, are more abundant in even years. From 1946 to 1955, the early trap catch in even years was about 33 percent of the season's total; in odd years about 4 percent. Cannerymen of the district and others have long observed and speculated on the small size of pink salmon taken during the early season. These fish vary in average size from 25 to as high as 35 per standard 48-pound case, as compared to the general seasonal average of 18 to 25 per case. If the pink salmon found in June were to remain in the fishery until late July, it is quite possible that they would reach the larger size. Whether the fish are small because of the time of the year in which they are captured or because of a distinct racial character is as yet unknown- -it is a matter for further study. Description of Tagging Program The 1958 tagging program was carried out during most of the commercial fishing season. Salmonfor tagging were captured between June 23 and August 3 in commercial traps and by purse seines at sites between False Pass and Ivanof Bay (figure 2). Serially numbered plastic disc tags one-half inch in diameter were attached to the fish with nickel pins at the base of the dorsal fin. Thesetags bore a number, notice of reward, and the address where they were to be returned. To stimulate returns from the commercial fish- ery, tag-return envelopes and posters that de- scribed the program and detailed the information desired were distributed. A reward of one dollar was paid for each tag returned. Canneries in the region and Bureau of Commercial Fisheries em- ployees in the field collected tags and paid the rewards. To gain information on home stream des- tinations and timing of the runs in relationship to the fishing seasons, a tag recovery program in streams was carried out after the commercial fishing season. The tagging experiments are summarized in table 1. The number and percent recovered from commercial fisheries and the personal-use fisheries in Kuskokwim Bay and northward are also shown. Separation of local and migratory runs Pink salmon runs vary in time of occur- rence and have characteristic peaks of abundance in different sections of the Alaska coast. The characteristics of the Peninsula runs (the long period pink salmon are available to the fishery; two peaks of abundance in even years, and size variations within the season) indicate that the fishery is operating on a complex of races. In particular, it may be argued that pink salmon taken in June are intercepted while en route to other areas. Three points in support of this hy- pothesis can be demonstrated from results of the 1958 tagging program. 1. Returns from individual experiments - For comparison, in figure 3 the percent recovery from each experiment is shown in chronological order of tagging. The percent of tags returned per experiment varies sharply with the time of season. The extremely low recovery of fish from June experiments, which coincide with the first peak of the even-year runs, can be explained only by assuming that these fish were traveling through the district. Had these fish remained in the area, recoveries should have been as great as those from experiments between July 10 and 20, be- cause fish would have been available to the fishery for a longer period. The increase in returns during early July and the high rate of recoveries during mid-July are explainedby the entrance of increasing num- bers of pink salmon of local origin into the fish- ery. These for the mostpart are available to the fishery for a longer time. The decrease in re- turns at the end of July and early in August is due to the fact that chances for recapture were limit- ed by the ending of the fishing season on August 5. Stormy weather at this time hampered fishing ac- tivity, and pink salmon began to enter streams of the district, further reducing recovery. 2. Destinations of tagged pink salmon - The low rate of recovery from the early tagging experiments needs additional explanation before it can be completely accepted as evidence of traveling fish. The explanation lies in the desti- nation of the tagged fish. Pink salmon that were tagged during June and early July were first re- captured near the tagging sites and later at the following widely divergent points along the Bering Sea Coast: Makushin Bay and Dutch Harbor on Unalaska Island, Kvichak and Nushagak Bays in Bristol Bay, the Kanektok River which enters Kuskokwim Bay, the Yukon River, an unnamed creek near Nome in Norton Sound, and the Una- lakleet, Shaktoolik, and Koyuk Rivers, which enter Norton Sound. Three pink salmon were also captured in the Kodiak Island area. The scat- tered recoveries from Nushagak Bay and north, which comprise a major portion of the low total returnfromthe early experiments, are extremely significant, because no commercial fishing for pink salmon is carried on in the area (except in Nushagak Bay). The chances of tagged fish being recaptured are relatively small. Further, no effort was made to secure returns from this sec- tion of the coast. The small number of local re- coveries from these tagging experiments may repiesentthe first of the local spawners to enter the fishery. But more likely the low recovery rate reflects the rapid movement of fish out of the area. The dispersion of tagged pink salmon to widely separated points is evidence of the broad Figure 2. Tagging sites — Alaska Peninsula pink salmon tagging experiments, 1958 mixture of races that make up the June run in the Peninsula area. 3. Isolation of migratory runs-Time of occurrence of migratory runs in the fishery is best shown by comparing local and nonlocal tag recov- eries in relation to time tagged (figure 4). Most pink salmon bound for distant places have passed through the area by July 10; few traveling fish are found after July 1. Additional analysis is needed to determine if all pink salmon in the area during June are migrants, or if the runs consist of a mixture of local and nonlocal stocks. Pres- ent evidence indicates that most of the early-run fish should be considered migrants. From tag returns, it is apparent that local fish far out- number migrants after July 1. For practical purposes this date should be used to separate and distinguish the two groups. Patterns of movement within the area After July 1, when the catch can be con- sideredto be composed mainly of Peninsula fish, the patterns of local movement are of primary interest. Results of tagging after July 1 indicate that these runs can be differentiated both in space and in time. For the purposes of discussion and com- parison, the results of tagging experiments per- formed over short intervals of time at the same Table 1. Summary of pink salmon tagging experiments on the Alaska Peninsula Date tagged Tagging site Number tagged Number recovered Percent recovered June 23 Kelly Rock Trap 26 Kelly Rock Trap 28---------------- East Anchor Cove (Seine) 29 East Anchor Cove Trap 29 ---------- ------ Swedania Point Trap 30 East Anchor Cove Trap July 1 - Red Bluff Trap 2- - Pop of Trap 2 Boiler Point Trap 7- --.. .. Boiler Point Trap 9 - Kelly Rock Trap 10 Boiler Point Trap 15 Boiler Point Trap 16- — --- — -------- Boiler Point Trap 17 Boiler Point Trap 18-- -- Boiler Point Trap 18---------------- Ivanof Bay (Seine) 20 Ivanof Bay (Seine) 20 Belkofski Point (Seine) 21 -. Bold Cape Trap 22-- Bold Cape Trap 23 Bold Cape Trap 24 Popof Trap 27 Bold Cape Trap 27 Kelly Rock Trap 28 -- Kelly Rock Trap 29---------------- Swedania Point Trap 29 Settlement Point Trap 30 Settlement Point Trap August 1 Orzenoi Bay (Seine) 3---. -_--.--_-__- Moss Cape Trap 3 Kelly Rock Trap 155 8 5.2 166 13 7.8 142 7 4.9 137 8 5.8 74 10 13.5 85 5 5.9 229 50 21.8 115 32 27.8 60 14 23.3 216 100 46.3 211 61 28.9 311 174 55.9 257 182 70.8 377 213 56.5 209 105 50.2 210 118 56.2 95 64 67.4 257 149 58.0 81 45 55.6 124 47 37.9 101 41 40.6 198 91 46.0 344 125 36.3 202 83 41. 1 499 132 26.4 498 119 23.9 199 69 34.7 200 103 51.5 213 77 36.2 399 8 2.0 204 37 18.1 49 6 68 13.7 Total 7,064 2,358 33.4 23 26 28 29 29 30 I 2 2 7 9 10 15 16 17 18 18 20 20 21 22 23 24 27 27 28 29 29 30 3 3 June July Aug. Figure 3. Recoveries of pink salmon tagged in the Alaska Peninsula area during 1958 site have been combined. (Experiments carried out at traps located in the Shumagin Islands and at Swedania Point have been considered as one site and are termed Shumagin experiments.) Segregation of local runs-Figures 5 and 6 illustrate the dispersion of fish tagged at sites in Morzhovoi Bay, Pavlof Bay, and the Shumagins . They show there is little intermingling between pink salmon of the eastern and western sections of the coast. Pink salmon tagged in Morzhovoi Bay re- mained almost exclusively in waters at the south - westernpart of the Peninsula. Eighty-six percent of the recaptures were from this section--Cold Bay, Morzhovoi Bay, and Ikatan Bay. Only 4 percent of the returns were from areas east of Cold Bay. No salmon tagged at Morzhovoi Bay were recaptured east of the Shumagin Islands (figure 5). On the other hand, pink salmon that were tagged in the Shumagins showed a wide range inmovement, both in easterly and westerly direc- tions. The movement to areas west of Pavlof Bay was minor and decreased as the season pro- gressed. Tag recoveries in this section of the coast from the Shumagin experiments were as follows: July 1 to 9, 13 percent; July 24 to 28, 5 percent, and July 23 to August 3, 0 percent. Even more significant is the fact that recoveries from areas on the mainland east of the Shumagin Islands were almost entirely from Shumagin ex- periments. This indicates that runs in the west- ern and eastern section of the coast are indepen- dent. The Pavlof and Shumagin experiments il- lustrate further segregation. The Shumagin traps are located in outside waters, at outer points, capes and headlands. The wide range of recov- eries from tagging at these sites indicates that the gear deployed in these localities intercepts runs composed of a mixture of races. A similar pattern of dispersion was observed for all ex- periments carried out at the other outer points- East Anchor Cove, Bold Cape, Belkofski Point, and Moss Cape. In contrast, the distribution of fish tagged in bays was limited, indicating sepa- ration into distinct populations. Tagging experiments performed in Ivanof and Orzenoi Bays also demonstrate this clearly. Two experiments were carried out in Ivanof Bay; 95 pink salmon were tagged on July 18 and 257 on July 20. Returns from both experiments were extremely high, 67 and 58 percent, respectively, but not one recapture was reported from outside Ivanof Bay. Later in the season, on August 1, 399 pink salmon were tagged in Orzenoi Bay. Only 8 of these fish were recaptured in the com- mercial fishery . They were in the bay or adjacent waters. However, 173 of the tagged fish were found in the stream that enters Orzenoi Bay. Segregation into distinct units in the bays is a natural function of time--as the season pro- gresses and the fish approach sexual maturity, they seek out their natal streams — so these re- sults are to be expected. The significant point demonstrated by these experiments is that there is apparently little tendency on the part of the fish to leave once they have entered a bay. This is in contradiction to the results of the Morzhovoi Bay 100 □ local RE( ■ NON-LOCAL tr - ERIES 1 1 1 ] > 80 o u UJ — 60 u. o 1 ■- i i I 1 1 1 z 40 o a. UJ [— \ f 1 o. 2 0 r 1 E JfrLocol recoveries 100 % from (7 July to 3 Aug. 11 2 3 6 28 29 2 9 1 0 i r 9 5 17- » — ■*■ 3 Aug Figure 4. Comparison of local and nonlocal recoveries by time of tagging Figure 5. Dispersion of pink salmon tagged in Morzhovoi Bay and the Shumagins experiments, which showed considerable move- ment out of the bay. The difference and expla- nation lie in the fact that the Morzhovoi Bay ex- periments were carried out relatively early in the season when the salmon were still en route to their parent streams. Conclusion This report is a partial analysis of the 1958 tag recovery data. Much detail has been omitted, and much remains to be done before the analysis is complete. The distribution of fishing effort and the relative abundance of pink salmon in the various sections of the Peninsula area have not been considered in relation to the distribution of tag recoveries. This must be done before the true significance of these results can be stated. Yet, incomplete as they are, the results of this study have immediate value in the management of the fishery. During June the fishery is supported by runs of pink salmon bound for Bristol Bay and areas along the northeast Bering Sea coast. There is little intermingling of fish that appear in the fishery west of Bold Cape with those that appear in the waters of the Shumagin Islands and the mainland coast north and east of the Shuma- gins. Pink salmon bound for mainland bays north and east of the Shumagin Islands, particularly that reach of coast from Balboa Bay through Stepo- vak Bay, pass through the Shumagins . The fish- ery west of the Shumagins does not appear to draw on these runs. For practical purposes, the western limit of migration of fish passing through the Shumagins after July 1 should be considered as Pavlof Bay. The data indicate that the area comprised of Ika- tan Bay, Morzhovoi Bay , and Cold Bay should be considered as a unit. Once pink salmon enter bays of the area they show little tendency to leave, especially after mid-July. Further, populations within a given bay after this time tend to be distinct; tagging within bays showed very few individuals traveling to either adjacent or distant waters. Under these circumstances it can readily be seen that heavy fishing could reduce the runs to dangerously low levels . — Mokusn.n Boy Kodiok- Dutrict tTA ~ A(S)SHUMAGIN EXPERIMENTS, July 28,29 ^jP) P4VL0F BAY EXPERIMENTS, July 29,30 Figure 6. Dispersion of pink salmon tagged in Pavlof Bay and the Shumagins SALMON MIGRATIONS IN SOUTHEASTERN ALASKA Wallace H. Noerenberg A salmon tagging program conducted off the west coast of Prince of Wales Island in southeast- ern Alaska in 1957 and 1958 was planned primarily to determine destinations and migration routes of the runs of pink (Oncorhynchua gorbuscha) and red salmon (O. nerka)that are intercepted by the fishery in the vicinity of Noyes Island. The tag- ging was done by the Fisheries Researchlnstitute of the University of Washington under contract to the Bureau of Commercial Fisheries. The area studied is the region referred to by the Bureau of Commercial Fisheries as the West Coast Manage- ment District (figure 1). This report comprises only preliminary general summaries of the release and return of the tagged fish and some interpretation of results . Detailed analysis of the data has just been started. The timing and abundance of pink salmon runs at Noyes Island have been variable and, at times, apparently unrelated to the appearance of runs in adjacent inshore areas of southeastern Alaska. Catches of pink salmon at Noyes Island nearly always remain low throughout the early seine season (July). The general open season for the use of seines and traps begins during the last week of July, but catches usually do not be- come large until mid-August. The peak of the catch generally occurs during the last two weeks of August, following which the fishing season is closed. In contrast to the usual situation, the 1957 run of pink salmon at Noyes Island occurred very early in the season and was unusually heavy. Peak catches were made during the last week of July and the first week of August. Owing to excep- tionally calm weather and poor runs elsewhere in southeastern Alaska, almost the whole southeast- ern Alaska seine fleet appeared at Noyes Island during this period. In 1958 pink salmon catches were unusually poor at Noyes Island throughout the season, although a small peak occurred in mid-August. Scarcity of fish and poor weather limited the activity of seine boats. The atypical character of the pink salmon runs of the two years should be considered in assessing results of the tagging. Location and extent of tagging The West Coast Management District has been divided into two sections for the purpose of summarizing the tagging results in this report. The "outer section" consists of the west coasts of Noyes, Baker, Suemez, and Dall Islands and includes tagging locations at Cape Ulitka, Roller Bay, Cape Addington, and Granite Point. The "inner section" consists of all waters of the dis- trict east of the outer section and includes the tagging locations at Point Desconocida, Tranquil Point, Ruth Bay, and McLeod Bay. These loca- tions are shown in figure 1. The principal sources of fish for tagging in both 1957 and 1958 were two salmon traps on the north shore of Noyes Island at Cape Addington and Cape Ulitka. Additional tagging was done during the early seine fishery in the vicinity of these same capes on Noyes Island and at Granite Point, Baker Island, in 1958, and from other salmon traps in both years. 8 Figure 1. Dixon Entrance and vicinity showing 1957 and 1958 tagging locations Table 1 shows the numbers of pink and red salmon tagged each year in the outer and inner sections of the West Coast Management District. The dates of tagging are also given. In addition to the pink and red salmon listed in the table, taggings in 1958 included 492 chum (O. keta), 574 coho (O. kisutch), 9 king salmon (O. tshawytscha) , and 4 steelhead trout (Salmo gairdneri gairdneri). Preliminary results from returns Table 1 shows gross returns from tagging of both years for each of the following four com- ponents: American fishery, American spawning grounds, Canadian fishery, and Canadian spawn- ing grounds. It should be noted that, in total, the proportion of both pink and red salmon tags re- turned was about equal for the two years. Some components of these returns, however, exhibit very great differences between the years. During the 1958 season, American fishery returns came principally from inside waters, while in 1957 a large proportion of the tagged salmon came from the immediate vicinity of the tagging locations at Noyes Island. Despite the dispersion of the A- merican fleet from the immediate tagging areain 1958, the percentage recovery of tagged fish in- creased over 1957. The percentage of recovery on American spawning grounds was almost the same for the two years for pink salmon, but for red salmon it was slightly greater in 1958. Of major interest was the large decrease in 1958 of the percentage recovery of both species from the Canadian fishery. Spawning ground recoveries of tagged pink salmon in Canada were almost nil in 1958 (3 compared to 88 in 1957) but substan- tially more tagged red salmon were observed at the Babine River fence weir than in 1957 (95 com- pared to 10 in 1957). The Babine Lake system is the major red salmon producer of the Skeena River in Canada. Factors to be considered in the analysis of the returns are the differences in the patterns of Table 1. Summary of results from pink and red salmon experiments on the west coast of Prince of Wales Island — 1957 and 1958 Tagging location Percentage returned from: American American Canadian Canadian Dates Num- Num- of Year ber ber fishery spawning fi8hery spawning tagging tagged returned grounds grounds Total PINK SALMON Outer section 7/16-8/25 1957 11,478 7/ 9-8/24 1958 4,925 Inner section 8/17-9/ 2 1957 5,057 7/22-8/25 1958 3,885 5,058 26.9 1.4 15.1 0.7 44.1 1,868 31.4 3.2 3.3 0.04 37.9 2,024 23.9 16.1 0.02 0.02 40.0 1,804 36.5 9.7 0.3 0.0 46.4 Total 1957 16,535 1958 8,810 7,082 3,672 26.0 33.7 5.9 6.0 10.5 2.0 0.5 0.02 42.8 41.7 RED SALMON Outer section 7/25-8/11 1957 499 7/ 9-8/24 1958 1,483 187 18.6 0.0 16.8 2.0 37.5 613 25.6 0.5 9.0 6.3 41.3 Inner section* 7/22-8/25 1958 715 225 29.5 1.7 0.0 0.3 31.5 Total 1957 499 1958 2,198 187 18.6 0.0 16.8 2.0 37.5 838 27.1 0.9 6.1 4.6 38.7 ♦No rod salmon were tagged in this section in 1957 10 CAPE ADDINGTON CAPE ULITKA West Coast of Noyes, Baker $ Dall Islands a LU X. UJ > o O 0 LU a: 0 West Coast of Prince of Wales Island 0 ii i ■ T ■* i i ■ ■■ .1 I , J Clarence Strait i — n f* Revillagigedo Channel -i p— "P 10 0 20 Nass River District Skeena River District Central British Columbia T r 15 20 25 30 I 5 10 15 20 25 25 301 5 10 15 July August July August Figure 2. Percentage recovery of pink salmon in principal recovery districts from tagging experiments at Noyes Island, 1957 11 the returns between tagging locations in the West Coast District. Tagging each year from the traps in the inner section of the district at Point Des- conocida and McLeod Bay and all experiments at Ruth Bay in 1957 and Tranquil Point in 1958 re- vealed insignificant Canadian returns of both pink and red salmon. Most tags were recovered in the waters surrounding Prince of Wales Island, and as the season progressed these tags tended to be more and more associated with spawning grounds in the vicinity of the tagging locations. Because of this factor, and because there was a difference between the two years in the propor- tion of the total tagging accomplished in the outer and inner sections, the results obtained in the two sections should be considered independently. This breakdown is given in table 1. Seasonal variations have been noted in the percentage of recovery of pink salmon in the sev- eral recovery districts from the tagging experi- ments at Noyes Island in 1957 (figure 2). Pink salmon destined for Canadian and mainland areas of southeastern Alaska were most abundant in the tagging areas during late July and early August. Salmon tagged in mid -August consisted primarily of Clarence Strait stock, while late August fish were recovered from the fishery and spawning grounds of the west coast of Prince of Wales Is- land. The routes taken by a considerable number of the pink salmon, tagged at Noyes and Baker Islands in July and early August of both years , appeared to be southward. These were outside the other sections of the west coast fishery and via Dixon Entrance to Clarence Strait and main- land areas north and south of the International Border. Evidence of northward movement into Sumner Strait was almost completely lacking in 1957 but was apparent for a small proportion of the salmon tagged in 1958. In late August most pinksalmon fromNoyes Island moved directly inshore to the fishery of the inner waters of the West Coast District and adjacent spawning grounds. All waters in south- eastern Alaska, east of Clarence Strait, were closed to fishing after mid-August, and this may have had some effect upon the pattern of returns at this time of the season. Red salmon were important in the catches at Noyes Island during July and early August and migration patterns in most respects resembled those of the early pink salmon mentioned previ- ously. Recoveries came principally from south- ern Clarence Strait and the Skeena area in both years. A small part (5 fish or 1 percent of the salmon tagged) of the red salmon returns in 1957 came from Johnstone Strait, but no returns were reported from there in 1958 although three times as many red salmon were tagged at Noyes Island as in 1957. During both years, the movement of pink salmon tagged at Ruth and McLeod Bays in the southern part of the West Coast District was gen- erally northward along the east and west coasts of Prince of Wales Island. Returns from the pink salmon tagged at Point Desconocida, Heceta Is- land, in both years and from Tranquil Point in 1958 came almost entirely from the fishery and streams along the northwest coast of Prince of Wales Island adjacent to Klawak Inlet. Summary This is the initial report on this project, and the conclusions drawn at this time are only preliminary. In summarizing the results in re- gard to the presence of salmon bound for Canadian fishing waters, the following points are evident: 1. The American fishery recovered about 2j times as many pink salmon tags as the Canadian fishery in 1957; 16 times as many in 1958. 2. The American fishery recovered a slightly larger number of red salmon tags than did the Canadian fishery in 1957; more than 4 times as many in 1958. 3. Returns of both pink and red salmon from Canadian fishery and spawning grounds were con- fined almost exclusively to fish tagged at Noyes and Baker Islands in both years. 4. More Canadian returns came from salmon tagged during late July and early August of 1957 than from those tagged during mid and late August of 1957. 5. It is necessary to make a thorough analysis of the data to obtain estimates of the number of fish bound for Canada from fishing waters along the west coast of Prince of Wales Island. 12 THE UNEXPECTED APPEARANCE OF PINK SALMON IN THE KVICHAK RIVER Clarence D. Becker In the fall of 1958 there was abundant spawn- ing by pink salmon (Oncorhynchus gorbuscha) in the Kaskanak Flats area of the Kvichak River in Bristol Bay, Alaska. The Kvichak River, which is approximately 55 miles long, drains from Lake Iliamna and flows into Kvichak-Naknek Bay. The unexpected appearance of these fish occurred si- multaneously with a tremendous, and also unex- pected, run of pink salmon in the neighboring Nushagak River. Spawning area Extensive spawning occurred in the sur- veyed section of the Kaskanak Flats, a series of shallow channels extending uprive r approximately eight miles from the head of tidewater. Above the flats, spawning took place largely in shallow side branches of the main channel. The head of the Kvichak River at Igiugig was the upper limit of the spawning. I Occasional pink salmon were observed in the upper section of the river by biologists of the Fisheries Research Institute, who were working near the native village of Igiugig at the outlet of Lake Iliamna. However, no pink salmon spawn- ing surveys were performed on the Kvichak River prior to 1958. Surveys of red salmon spawning grounds in streams tributary to Lake Iliamna re- vealed no spawning pink salmon populations. No observations were made in the tidal area below Kaskanak Flats. On the basis of catch data and timing of the run, it is apparent that the 1958 pink salmon run to the Kvichak River escaped the commercial fishery almost entirely. Only about 11,000 pink salmon were deliveredto the cannery at Koggiung after the close of the red salmon season in late July. Personnel of the Fisheries Research Insti- tute made the following observations on the pink salmon escapement from September 9 to 12. Number present An accurate estimate of the total number of pink salmon could not be made because only one channel of Kaskanak Flats was surveyed, covering the navigable section of the river from Igiugig to Ole Creek. In this area alone, 10, 000 to 15, 000 pink salmon were spawning- -one or two pairs of fish per ten square feet of gravel. If it can be assumed that pink salmon were present in simi- lar numbers throughout the flats, then the spawn- ing population in the Kvichak River in 1958 would probably have been between 150, 000 and 250, 000. Time of spawning On September 9, the time of the main sur- vey, most pink salmon were paired on the grounds and actively spawning. The survey disclosed only occasional small groups of schooled fish. White fungus patches were present on about 50 percent of the live fish, and there were only a few dead salmon. This evidence indicates that the survey was made just after the peak of spawning. 13 GRAVEL REMOVAL AND THE FISHERIES Robert McVey Removal of gravel from streams offers a potential source of damage to Alaska's fishery re - sources. Although it has not been a matter for great concern in past years, continuing develop- ment will so increase gravel requirements for construction purposes that the problem may be- come serious. It is vital that Alaska's valuable fishery resources receive proper consideration during this period of development and growth. Since most major construction to date has been federally financed, and appropriations for these activities have fluctuated little from year to year, Alaska's sand and gravel requirements did not show any distinct upward trend the past several years. From 1950 to 1958, over 6 1/2 million short tons of gravel, worth approximately 6 million dollars, were used each year. Danger of damage to the fisheries would be eliminated if gravels were never taken from the streams. However, streams are often the only source from which gravels can be obtained in- expensively, and the quality of this stream-worn material frequently makes it the more desirable. Removal from streams is relatively simple and not costly since trucks can be loaded by dragline right at the stream bank. Although the need for additional construc- tion and road building might seem so important that it should override consideration of the fishery problem, a brief look at the value of the Alaskan salmon resource will show that to do this would be an error. Throughout the years salmon have provided the backbone of Alaska's economy. Re- cent years have witnessed a decline inproduction but the value of the annual pack ha6 remained high (figure 1). The value of some salmon runs should not be measured only in terms of dollars and cents. Certain native villages are highly dependent upon local runs of salmon for food; for these people, preservation of the resource is essential. Many people derive enjoyment each year from observ- ing the salmon runs as they enter the streams. The annual enactment of this drama will become increasingly valuable as a source of recreation. The sport fisheries of Alaska are sustained almost entirely by salmon, trout, char, and gray- ling, most of which are stream spawners. While a reliable estimate of the value of these fisheries is not available, their importance should not be underestimated. ■ PRODUCTION ■ VALUE I ■ i ' ' I ' ' 1925 1930 1935 1940 1945 1950 1955 Figure 1. Production and value of Alaskan canned salmon 14 Although occasional gravel removals might cause only a slight decrease in a stream's poten- tial productivity, experience has shown that com- plete spawning grounds may be lost if removals are carried on indiscriminately and on too large a scale. Siltation normally accompanies any re- moval operation in a greater or lesser degree, and the fine silt particles that are washed down- stream impair percolation of water through the gravels and tend to smother both eggs and fry. Heavy equipment used on the stream bed during removal operations frequently crushes the eggs and fry. Merely the activity of the operation it- self and the resultant turbidity prevents adults from spawning. More permanent effects are pro- duced when large portions of the actual spawning beds are removed. In addition, fish might be blocked from reaching long stretches of spawning area as a result of dewateringor diversion asso- ciated with gravel excavation. Examples of the effects of indiscriminate removal of gravels are conspicuous in a group of streams on the northeast side of Kodiak Island (figure 2). These streams, Russian River , Kalsin and Sargent Creeks, an unnamed stream which flows into Anton Larsen Bay, and a tributary of BuskinLake, historically supported valuable runs of pink and chum salmon and Dolly Varden char. Some of them also contained runs of red and silver salmon and steelhead trout. During World War II, it became necessary to construct numerous build- ings, airport runways, and an extensive network of roads. Great amounts of gravel were required, and the most expeditious method of obtaining it was by removal from stream beds. Owing to the emergency situation, there was little considera- tion given to the fishery values. Entire stretches of stream beds were removed to depths of 20 feet, and at some locations the tailings and silt from the washing and screening were permitted to re- enter the stream channel. Large stretches of spawningbeds were re- moved in Russian River (figure 3) and Sargent Creek. The re-entry of tailings from washing and screening processes so reduced the average size of the stream gravels that the stream bed is now unstable and the channel pattern shifts re- peatedly. As a result, the fish-producing poten- tial has been reduced. The effects on the tributary of BuskinLake, the east fork of Kalsin Creek, and the unnamed stream entering Anton Larsen Bay were of a dif- ferent nature. In these, streambottom materials for distances up to 2 miles above the removal site broke up and were washed downstream. Natural layering of the silt, sand, and gravel was com- pletely disrupted. As a result, stream flow was limited to subterranean seepage in the "slipped" area during periods of low water. Without sur- face flow these sections had no value as spawning grounds and, in addition, fish were not able to reach the large areas of spawning gravels farther upstream. Several miles of excellent spawning grounds were rendered completely inaccessible. This is believed to be relatively permanent since the removals described took place largely before 1944, and the gravels of only one of the streams that which enters Anton Larsen Bay have shown signs of stabilization since that time. Re- deposition of the layer of silt that prevents sub- terranean flow is largely dependent on the silt load carried by the stream. Thus, complete re- covery in these clear waters will require many years. It is possible that the decline in the fish runs of these streams could be the result of fac- tors other than gravel removals; however, the streams where the decline was most severe were Russian River and Sargent Creek, both of which were subjected to extensive removal operations for several years. Salonie Creek flows into the same bay and is comparable in many respects with these streams, but only minor amounts of gravel were removed from here and the fish runs have remained at a high level. A similar com- parison between the two forks of Kalsin Creek reveals that fish runs of the east fork from which large amounts of gravel were removed have de- clined markedly, while those of the west fork, which was subjected only to very limited removal operations, have remained in good condition. Although the undesirable effects of a single gravel removal project may seem insignificant in relation to the total Alaskan fishery resource, the aggregate effect of several projects is extremely significant. For this reason, the Bureau of Com- mercial Fisheries reviews each proposal for re- moval of stream gravels that is brought to its attention. Responsibility for this work rests with the Branch of River Basin Studies, which operates under the authority of the Fish and Wildlife Co- ordination Act. In Alaska, this Branch presently maintains a permanent staff of four biologists in Anchorage and three in Juneau. For information concerning gravel removal projects, the Branch of River Basin Studies relies on the District Management offices of the Bureau of Commercial Fisheries, the Bureau of Sport Fisheries and Wildlife, and on the three Federal agencies that arrange and administer gravel sales I 15 Cape Chiniak 5 Miles Figure 2. Streams on northeastern Kodiak Island affected by gravel removal during World War II 16 Gravels formerly utilized by spawning salmon have been windrowed for removal FWS photo by Jack Lentfer Gravels have been removed from large section of the original stream channel FWS photo by Jack Lentfer Figure 3. Views of Russian River showing changes brought about by gravel removals FWS photo by M. A. Monson Excavation includes parts of the natural stream *'fcjR*3rt --the F o re 8 1 Service, the Bureau of Land Manage- ment, and the Army Corps of Engineers. The Forest Service has jurisdiction over the National Forests down to the mean high tide line. Below the mean high tide line and on state-owned lands, the State of Alaska has jurisdiction. The Army Corps of Engineers is responsible for all navi- gable waters and reviews any projects on these waters that involve removal of gravels by means of dredging. To assure that each proposed pro- ject is carefully evaluated for its effect on the fishery, close liaison is maintained between the Bureau of Commercial Fisheries and the three administering agencies. An effective working arrangement between the Bureau of Commercial Fisheries and the Bu- reau of Public Roads assures that all gravel re- movals associated with the latter 's construction projects are reviewed in advance of making an application. This permits adequate considera- tion of a project and allows time for working out detailed recommendations for protection of the fishery resources. Upon being notified of a proposal for gravel removal, the Branch of River Basin Studies con- ducts afield survey of the area involved, often in the company of personnel from the administering agency. The proposed excavation is located in relation to the stream, and possibilities concern- ing removal methods and effects are discussed. In this way, both the construction interests and the fishery values receive on-the-spot consider- ation. Persons familiar with the project area are interviewed. If the project involves anadro- mous fish, several surveys of the area may be made by River Basin personnel to determine the number of fish that will be affected. Flow data concerning the streams may be obtained from the U. S. Geological Survey. Pre- vious fishery surveys of the Bureau of Commer- cial Fisheries, the Alaska Department of Fish and Game, and the Fisheries Research Institute, University of Washington, are examined. Quanti- tative information on fish populations of former years is usually very limited or nonexistent, since removal projects are often situated on streams that have not been routinely surveyed in the past. When information concerning a stream and its fishery resources has been examined, recom- mendations are formulated on methods and time of gravel removals. The Bureau of Commercial Fisheries policy concerning the removal of gra- vel from streams is based on the premise that. by sound management methods, Alaska's devel- opment can proceed concurrently with perpetua- tion of the fishery values. If the fisheries will be adversely affected, and means cannot be found to prevent these effects, the Bureau will recom- mend that other sources of gravel be sought; how- ever, itis usually possible toprescribe practices and dates of operation that will minimize or elim- inate injury to the fisheries. Recommendations on times for gravel re- movals are based on periods when no eggs, fry, or adults are present in the stream. In streams in which the dominant species are chum and pink salmon, this condition usually exists during June and July. For streams that contain trout or the other species of salmon, prescribing dates of op- eration is a much more difficult task since fry and fingerling are normally present throughout the year. Generally it is recommended that gravels be removed only from areas that are located away from the existing stream channel. However, if it is unavoidable to excavate adjacent to a stream channel, it is recommended that a dike or levee be built that will stop the stream from flowing into the excavation. Heavy equipment is usually prohibited from the stream bed. If washing or screening of gravels is required, provision must be made to prevent entry of silt-laden water into the stream channel. In most instances, special recommendations are made that apply to partic- ular characteristics of the project involved. When recommendations have been formu- lated, the Regional Director, Bureau of Commer- cial Fisheries, submits them to the administering agency. Since close liaison has been maintained throughout the entire process, approval of these recommendations is usually a routine matter and they are incorporated into the gravel permit. Summary Removal of gravels from stream beds has resulted in losses to Alaskan fishery resources in past years. Removal operations will be inten- sified as development of the State and its natural resources progresses. The importance of the fisheries warrants careful consideration of pro- posed gravel removal projects. To insure the protection of these resources, it is essential that each proposal be reviewed individually. 18 ESTIMATING ABUNDANCE OF SALMON FINGERLING Howard S. Sears Estimating the abundance of pink salmon (Oncorhynchus gorbuscha) fingerlings by means of various types of sampling gear was attempted in 1957 and 1958 in southeastern Alaska to establish an index of abundance that could be used to predict the size of future pink salmon runs. Fishing gear tested during 1957 included beach seines, set gill nets of variable mesh, an Isaacs Kidd midwater trawl, and a Canadian meter hoop net. Insufficient numbers offish were taken by these types of gear , and in 1958 experiments were conducted with a lampara bait seine. Operations were carried on in Ernest Sound and Clarence Strait and in Revillagigedo Channel as far south as Dixon Entrance. Results obtained were more encouraging than with the other nets that had been tested. A conventional lampara seine consists of a large central bag or bunt section and two tapered wings. It differs radically from the purse seine in that its mesh size is not uniform, and there are no purse rings or purse line. The bag section is of smallmesh andis about one -fifth of the total net length. The wings are tapered, and the mesh varies from about 5 inches at the base to as much as 14 and 16 inches at the wing tips. The function of the large wing mesh is to guide fish into the bag where they are held until the lead line is closed by pulling both wings simultaneously. The foot or lead line is considerably shorter than the cork line. The net is hung so that it is in the shape of a large scoop. The modified seine used in the 1958 ex- periments was 120 fathoms long at the cork line, 96 fathoms long at the lead line, and 12 fathoms deep at the point where the bag and wing bases meet. Mesh sizes were 6-inch, 6-thread Marlon in the wings; 3-inch, 6-thread Marlon in the apron (floor of net), and 1/2-inch, 26-thread Marlon in the bag. Dry weight of the net was 285 pounds. The method of fishing differed from that generally used by commercial bait seiners using a conventional net. Their method is to make a circling set from a towed skiff by paying out one wing, which is anchored or buoyed, followed by the bag and second wing. Setting of the net is com- plete when the end of the first wing is retrieved. Both wings are then pulled simultaneously until the bag is brought close to the skiff. Instead of a single towed skiff , two 18-foot flat bottom boats (called Cordova skiffs in Alaska) powered by outboard motors were used in the 1958 experiments. The bag and one wing were placed in one skiff, and the other wing was placed in the second skiff. A large circling set was made by setting the bag from the stern of the first skiff. The wings were then payed out from the respec- tive skiffs. After the wings were set, the two skiffs towed the wing ends together. The wings were then hauled aboard the skiffs. A schematic viewof the modified lampara seine in fishing po- sition is shown in figure 1. In the summer of 1958, 260 sets were made with the lampara seine to evaluate its usefulness in the prediction program. A total of 2,509 fing- erling salmon of all 5 species were taken. The higher catches of this type net over those used heretofore suggest that it may be possible with random sets to relate the average number of fing- erling per seine haul to the number of returning adults . I 19 o e n re a. E -D O ■J. MBL WHOI librf 5 WHSE 00111 GPO 993004