SALMON SPAWNER SURVEYS IN THE UPPER TRINITY RIVER BASIN (continued)
Sex was determined only for fish recovered from the mainstem Trinity River that were either condition-two unmarked fish, Program-marked fish, or flagged fish recaptured in the carcass
We determined the sex of 340 adult chinook salmon during the survey (49 spring-run and 291 fall-run). Of the adult spring-run chinook salmon observed, 53.1% were females, while adult fall-run fish were 54.1% females. The percentage of females in the survey was generally highest during the early and late weeks of the survey and lowest during the middle weeks (Figure 6). This seasonal trend in sex ratio was also noted in the previous two year's surveys (Zuspan 1992a, 1992b). However, the trend was not as pronounced this year as in the past. The preponderance of adult females in the chinook salmon run has been noted in all but two of the previous surveys and has ranged from 73.6% to 25.8% (Appendix 6). The preponderance of females among adult fish results when males return as grilse, thereby decreasing the number of males left to return as adults.
FIGURE 6. Percent females in the adult chinook salmon population observed in the mainstem Trinity River during the 1991-92 spawner survey.
We determined the sex of 109 coho, 60% (65) of which were females. For comparison, 42%, 57%, and 80% of the coho we examined in 1988, 1989, and 1990, respectively, were females (Zuspan 1991, 1992a, 1992b). Like female chinook salmon, coho salmon females were more prevalent early and late in survey (Figure 7).
FIGURE 7. Percent females in the adult coho salmon population observed in the mainstem Trinity River during the 1991-92 spawner survey.
Prespawning mortality was determined only for fish recovered in the mainstem Trinity River that were either condition-two unmarked fish, Program-marked fish, or flagged fish recaptured in the carcass survey.
We determined the spawning condition of 186 adult female chinook salmon, including 22 spring-run and 164 fall-run fish. Prespawning mortality was 0% (0/22) and 1.2% (2/164) for spring- and fall-run female chinook salmon, respectively.
The overall prespawning mortality rate of both races of female chinook salmon was the lowest on record, at 1%. For comparison, overall (spring- and fall-run) prespawning mortality of female chinook salmon has ranged from 1.5% to 44.9%, averaging 12.8% during previous surveys (Appendix 7).
Prespawning mortality of chinook salmon in the Trinity River basin appears to be related to spawner escapement and, therefore spawner density. Specifically, as spawner escapement increases so does prespawning mortality. The CDFG's Trinity River Project has developed chinook salmon escapement estimates in the Trinity River basin since 1978. Prespawning mortality rates are available for the periods of 1978 through 1982 and for 1987 to the present. During the periods where both escapement estimates and prespawning mortality rates are available, escapement has ranged from 6,135 to 100,913 while prespawning mortality rates have ranged from 1.1% to 44.9% (Table 6). With the exception of 1980, prespawning mortality generally increases with increasing escapement (Figure 8). The high prespawning mortality rate noted in 1980 may be a sampling error. During that year, only 63 female chinook were checked for spawning condition. A regression analysis of escapement and prespawning mortality indicates a significant correlation (R2=.406, P=0.048) even with the 1980 data included. Without the 1980 data, the significance is much greater (R2=.709. P=0.004).
TABLE 6. Adult chinook salmon natural spawner escapement estimates and adult female chinook salmon prespawning mortality rates for the Trinity River, 1978-1982 and 1987-1991.
FIGURE 8. Adult chinook salmon natural spawner escapement and adult female chinook salmon prespawning mortality rates for the Trinity River, 1978-1982 and 1987-1991.
Sixty-four adult female coho salmon were examined for spawning condition during the survey. None (0/64) of the fish examined died prior to spawning. For comparison, in 1988, 1989, and 1990, the prespawning mortality rate of adult female coho salmon was 25.6%, 6.2%, and 13%, respectively (Zuspan 1991, 1992a, 1992b). Coho prespawning mortality rates were not reported in surveys prior to 1988.
Salmon spawner distribution in the mainstem Trinity River is presented based on the seven-zone system first used in 1987 (Stempel [Appendix 1]). The results of Zones 5, 6 and 7 were combined this year because too few flagged chinook were recovered in these individual zones to make reliable estimates. Distribution estimates are for adult fish only. This is because grilse and adult salmon are recovered in the survey at different rates; a fact that would force us to stratify the distribution estimate. Also grilse are relatively unimportant to the spawner escapement as they are predominantly males and frequently do not spawn because of competition from larger, older males.
Mainstem Trinity River. We examined 678 adult chinook salmon this season, excluding flag recoveries. The numbers of chinook salmon spawners were greatest in upstream zones, decreasing from a high of 195 fish in Zone 1 to 65 fish in Zone 3 (Table 7). We recognize that carcass counts alone cannot be used to accurately describe distribution, because recovery efficiency can vary from zone to zone, due to differences in stream morphology. Therefore, the percentage of flags recovered for each zone was used to determine the recovery efficiency of that zone (Table 7). Even based on the total number of chinook salmon recovered divided by the different recovery efficiency rates for each zone, the percent of chinook salmon spawners decreased downstream in successive zones below Zone 1 (Table 7). Spawner density, in terms of spawners per river km, was highest in the uppermost section (98 spawners/km), and decreased steadily in a downstream direction (Table 7, Figure 9).
This pattern of higher chinook salmon spawning concentrations in the upstream sections has been noted in each of the five previous years (Zuspan 1991, 1992a, 1992), but was much less pronounced this year (Figure 9). Chinook salmon spawners were much more evenly distributed throughout the survey area this year.
It is possible that an increase in river flow during the late summer and fall was responsible for the more even distribution of spawners seen this year. In an attempt to keep river temperatures within specified criteria, the flow during the late summer and fall averaged about 150 cubic-feet-per-second (CFS) higher this year than in previous years (450 versus 300 CFS). While the higher flows probably lowered temperatures this year, they were not significantly lower than in previous years. It may be that the higher flows increased the holding and spawning habitat to a point that allowed chinook salmon to spawn in the lower reaches of river. It should also be noted that the decreases in spawner escapements over the last few years may have somehow caused spawners to distribute themselves more evenly. However, while there has been a steady decrease in annual spawner escapements in the last few years (Table 6), the proportion of spawners in each zone remained approximately the same until this year (Figure 9).
TABLE 7. Adult chinook salmon spawner distribution and estimated density, by river zone, in the 1991-92 Trinity River spawner survey.
FIGURE 9. Estimated adult chinook salmon spawning density, in spawners per river km, measured during the 1987 through 1991 mainstem Trinity River spawner surveys.
As noted in previous years (Zuspan 1991, 1992a, 1992b), a potential source of error in this estimate is the assumption that flagged chinook salmon carcasses are recovered only in the zone that they were originally flagged. If flagged fish are recovered in downstream zones, it would tend to increase the efficiency estimate in the recovery zone while decreasing the estimate in the flagging zone.
To determine the extent that carcasses drifted from one zone to another, fish flagged this year in each zone were again given a distinct hog ring color. Recoveries that were originally flagged in another zone were recorded as such. This season, all 87 of the flags were recovered in the same zone in which they were originally flagged. This indicates that carcass drifting had no effect on chinook distribution estimates, as was the case in the 1990-91 season (Zuspan 1992b). In the 1989-90 season the proportion of flags that drifted into other zones was less than 1% (Zuspan 1992a).
Tributaries. Spawning adult chinook salmon made very limited use of tributaries this year. Too few chinook salmon were observed to make a mark-and-recovery spawner estimate, so we used redd counts to describe spawner distribution, as was the case last year (Zuspan 1992b).
We located 51 salmon redds in seven of the nine tributaries surveyed this season. Since we could not differentiate a chinook from a coho salmon redd during the survey, we used the relative proportion of chinook and coho salmon observed in the individual tributaries to apportion the redds by species. Based on this apportioning, there were 35 chinook salmon redds observed this season with counts in the individual tributaries ranging from nine to zero (Table 8, Appendix 5).
TABLE 8. Salmon redd numbers and distribution observed in the 1991-92 Trinity River tributary spawner survey.
Mainstem Trinity River. We observed 127 adult coho salmon in the mainstem spawner survey this year, most of which were seen in Zones 1 and 2 (Table 9). We estimated the total number of coho salmon which spawned in each zone by dividing the actual number of carcasses observed by the recovery efficiencies for that zone that were developed from chinook salmon flag recoveries. Coho salmon spawning density was highest in Zone 4 (19 spawners/km) and ranged from 18 to 5 spawners per km in the other zones(Table 9).TABLE 9. Adult coho salmon spawner distribution and estimated density, by river zone, in the 1991-92 Trinity River spawner survey.
Tributaries. We recovered 12 coho salmon during the tributary surveys. They were recovered in Weaver Creek, North Fork Trinity River, and East Fork of the North Fork Trinity River (Appendix 5). When the observed redds were apportioned by species (see page 20), there were an estimated 16 coho redds observed in the tributary survey (Table 8). Estimated redd counts ranged from seven to zero in the individual tributaries.
We observed Program marks (spaghetti tags or operculum punches) on 14 spring-run and 59 fall-run chinook salmon in the mainstem Trinity River spawner survey. Program-marked spring- and fall-run chinook salmon were recovered from both Junction City and Willow Creek weirs (Table 10). Of the 73 Program-marked chinook we observed, 27 were condition-one fish and 46 were condition-two fish. Seventeen Program-marked coho, seven from Willow Creek Weir and ten from Junction City Weir, were also recovered in the mainstem Trinity River.
TABLE 10. Program-marked salmon recovered during the 1991-92 mainstem Trinity River spawner survey.
We used only condition-one chinook salmon to determine the actual percentage of Program-marked chinook salmon in the spawner survey. This is because we were more likely to correctly identify a Program mark on a fresh (i.e. condition-one) fish than one in an advanced state of decay. The percentage of condition-one salmon recovered in the survey which had been marked at the two tagging sites ranged from 3.3% to 7.7% for chinook salmon (Table 10).
We did not record the condition of coho salmon during the survey so we can not analyze the Program marks of condition-one fish. However, for fish of all conditions, 13.4% (17/127) of the coho salmon recovered were Program-marked (Table 10).
We recovered 30 chinook salmon and one coho salmon in the spawner survey which appeared to be Ad-clipped. Based on their CWTs, one was a spring-run chinook salmon, 26 were fall-run chinook salmon, and four fish did not have CWT's (Appendix 8). All of the CWT recoveries were of Trinity River Hatchery origin.
To minimize the number of Ad-clipped fish missed during the spawner survey, all fish recovered were passed through a coded-wire tag detector. Fish which produced a positive reading with the detector, regardless of the condition of their adipose fin, were considered Ad-clipped.
The percentage of Ad-clipped fish in the spawner survey is best estimated by considering only those Ad-clipped fish that had CWTs (Ad+CWT) and were condition-one fish, as fish in advanced decay (i.e. condition-two fish) were more likely to have shed their CWT. For example, the Ad+CWT rate of fall-run chinook salmon condition-two fish was only 3.5% (12/342) while for condition-one fish it was 7.5% (15/201). However, this method does not produce an estimate of Ad-clipped fish that can be directly compared with the estimate of Ad-clipped fish returning to the weirs or TRH. This is because we consider Ad-clipped fish in the spawner survey to be only those fish that have CWTs, while at the other sites they count fish with Ad-clips regardless of their having a CWT. To make the two estimates comparable, we expanded the number of Ad+CWT fish observed in the spawner survey by the CWT shedding rate for chinook salmon observed at TRH(%CWTs observed/ (1-(%at TRH with shed tags/100).. For example, of the 60 Ad-clipped spring-run chinook salmon observed at TRH, only 45 (75.0%) had CWTs indicating a 25.0% natural CWT shedding rate for these fish. For fall-run chinook the CWT shedding rate at TRH was 5.1% (16/317). Expanding our counts of Ad-clip+CWT fish in the spawner survey by the aforementioned CWT shedding rates, 0% and 7.9% of the spring- and fall-run chinook salmon observed in the spawner survey were Ad-clipped.
The percentage of Ad-clipped spring- and fall-run chinook salmon varied at the different recovery sites, probably as the result of hatchery-produced fish homing to the hatchery. Since naturally produced chinook salmon also spawn in the lower mainstem or its tributaries, we would expect the percentage of hatchery-produced, Ad-clipped chinook salmon in the population to increase at each sampling site proceeding upstream, and to be highest at the hatchery. The Ad-clipped chinook salmon rate was highest at the hatchery, intermediate at the weirs, and lowest in the mainstem Trinity River spawner survey (Table 11). Ad-clip rates in the spawner survey may have been less than at weirs downstream, as the weirs captured a fraction of all upstream migrants, both hatchery and natural fish, while the spawner survey emphasized in-river spawners which would be more likely to be naturally produced fish. The reason that chinook salmon trapped at Willow Creek had a slightly higher Ad-clip rate that those trapped at Junction City weir is unknown.
We cannot analyze the Ad-clip rates of coho salmon this year. This is because the returning adults are from a brood year that was not marked (Ad+CWT) at TRH.
TABLE 11. Numbers and percentages of adipose fin-clipped chinook salmon observed in the mainstem spawner survey and at three fixed locations in the Trinity River basin during the 1991-92 season.
We determined the incidence of hatchery-produced chinook salmon among the carcasses seen in the spawner survey by comparing the rate of Ad-clipped (hatchery-marked) chinook salmon at various locations within the river.
The percentage of Ad-clipped spring-run chinook salmon observed at the three locations in the Trinity River basin below Lewiston Dam ranged from 0% to 8.8% (Table 11), and were significantly different (X2=9.98, df=2, P=0.007) from each other.
During the previous three years, most (approximately 97%) of the spring-run chinook salmon recovered at TRH were of hatchery origin (Bill Heubach, Calif. Dept. of Fish and Game, pers. comm., based on expansions of CWT recoveries). This year, using the same methodology, only an estimated 65.4% of the spring-run chinook at TRH salmon were of TRH origin (Bill Heubach, Calif. Dept. of Fish and Game, pers. comm.). This apparent low rate is an artifact of the high CWT shedding rate for spring-run chinook this year. Since only Ad-clipped fish with CWTs can be used for the expansion, the 25% (only 45 of the 60 Ad-clipped spring-run chiook salmon entering TRH had coded-wire tags., which indicates a 25% shed rate for these fish) shedding rate for spring-run chinook salmon had the effect of decreasing the estimate of TRH-produced fish returning to the hatchery. We believe the actual percentage of TRH-produced spring-run chinook salmon returning to TRH this year is similar to previous years. Therefore, we assume that the 8.8% Ad-clip rate for spring-run fish observed at TRH represents a population of 100% TRH-origin chinook salmon. Since no condition-one Ad+CWT spring-run chinook salmon were recovered in our survey, we feel that essentially all of the spring-run chinook spawning in our survey zones were naturally produced (non-hatchery).
The Ad-clip percentage of fall-run chinook salmon ranged from 8.0% to 11.8% at the four sampling sites this season (Table 11). The differences in chinook salmon Ad-clip rates among the four sites is statically significant (X2=13.78, df=3, P=0.003).
Since most (93.3%) of the fall-run chinook recovered at TRH are estimated to be of hatchery origin (Bill Heubach, Calif. Dept. of Fish and Game, pers. comm., based on expansions of CWT recoveries), we assumed that the 11.8% Ad-clip rate for fall-run fish observed at TRH represents a population of 100% hatchery-produced chinook salmon. Since only 7.9% of the fall-run chinook salmon in the spawner survey were Ad-clipped, we estimated that 66.9% (7.9/11.8) were of hatchery origin, while the remaining 33.1% were naturally produced.
There are several assumptions which could be potential sources of error in using the aforementioned method to determine the incidence of hatchery fish spawning in the river. We assume that field personnel actually observed all possible Ad-clips in the survey. Using the strict protocol developed this year (i.e. using a tag detector on all fish and considering only condition-one fish) we feel we were successful at identifying essentially all Ad+CWT fish in the survey. We are also assuming that the probability of observing and recovering an Ad-clipped fish is the same in the survey as at the hatchery, and, most importantly, that ratios of Ad-clipped to unmarked hatchery fish are the same in the spawner survey as at TRH. Since different chinook salmon release groups are Ad-clipped at different rates, this last assumption is only valid if the various CWT groups occur in the spawner survey in the same proportions as among the fish recovered at TRH.
This is the fourth year of a multi-year effort of spawner surveys in the Trinity River basin. The following recommendations should be considered for inclusion in next year's spawner survey.
1. Spawner survey activities should be continued, with current objectives, in FY 1992-93 and beyond.
2. To increase the number and accuracy of our Ad-clip salmon recoveries, we should continue to pass all salmon through a tag detector. This should allow us to more reliably estimate the proportion of hatchery and naturally produced fish spawning in the wild.
3. Flows from Lewiston Dam should be increased during the late summer to mid-fall period from the base 300 CFS to approximately 450 CFS. The purpose of the higher flows would be to distribute chinook salmon spawners more evenly in the mainstem Trinity River. A more even distribution of spawners should also lead to a decrease in prespawning mortality. The increased flows could be especially important during years of high escapement when chinook salmon in the Trinity River have suffered unusually high prespawning mortality.
Gibbs, E. D. 1956. A report on king salmon, Oncorhynchus tshawytscha, in the upper Trinity River, 1955. Calif. Dept. Fish and Game, Inland Fish. Admin. Rep. No. 56-10. 14 p.
La Faunce, D. A. 1965. King (chinook) salmon spawning escapement in the upper Trinity River, 1963. Calif. Dept. Fish and Game, Mar. Res. Admin. Rep. No. 65-3. 10 p.
Moffett, J. W., and S. H. Smith. 1950. Biological investigations of the fishery resources of the Trinity River, Calif. USFWS Spec. Sci. Rep., Fish. Bull. No. 12. 71 p.
Rogers, D. W. 1970. A king salmon spawning escapement and spawning habitat survey in the upper Trinity River and its tributaries, 1968. Calif. Dept. Fish and Game, Anad. Fish. Admin. Rep. No. 70-16. 13 p.
Rogers, D. W. 1973. King salmon, Oncorhynchus tshawytscha, and silver salmon, Oncorhynchus kisutch, spawning escapement and spawning habitat in the upper Trinity River, 1970. Calif. Dept. Fish and Game, Anad. Fish. Admin. Rep. No. 73-10. 14 p.
Rogers, D. W. 1982. A spawning escapement survey of anadromous salmonids in the upper Trinity River, 1971. Calif. Dept. Fish and Game, Anad. Fish. Admin. Rep. No. 82-2. 11 p.
Smith, G. E. 1975. Anadromous salmonid spawning escapements in the upper Trinity River, 1969. Calif. Dept. Fish and Game, Anad. Fish. Admin. Rep. No. 75-7. 17 p.
Weber, G. 1965. North coast king salmon spawning stock survey 1956-57 season. Calif. Dept. Fish and Game, Mar. Res. Admin. Rep. No. 65-1. 34 p.
Zuspan, M. 1991. Salmon spawner surveys in the upper Trinity River Basin. Chapter I. Job I. p. 1-23. In: Carpenter, R. and K. Urquhart (eds.), Annual Report of the Trinity River Basin Salmon and Steelhead Monitoring Project, 1988-1989 Season. August 1991. 51 p. Available from Calif. Dept. Fish and Game, Inland Fish. Div., 1416 9th St., Sacramento, Ca. 95814.
Zuspan, M. 1992a. Salmon spawner surveys in the Upper Trinity River Basin. Chapter I. Job I. p. 1-29. In: Urquhart, K. (ed.), Annual Report of the Trinity River Basin Salmon and Steelhead Monitoring Project, 1989 -1990 Season. June 1992. 140 p. Available from Calif. Dept. Fish and Game, Inland Fish. Div., 1416 9th St., Sacramento, Ca. 95814.
Zuspan, M. 1992b. Salmon spawner surveys in the Upper Trinity River Basin. Chapter I. Job I. p. 1-31. In: Urquhart, K. and R. Carpenter (eds.), Annual Report of the Trinity River Basin Salmon and Steelhead Monitoring Project, 1990-1991 Season. December 1992. 186 p. Available from Calif. Dept. Fish and Game, Inland Fish. Div., 1416 9th St., Sacramento, Ca. 95814.
APPENDIX 1. Other sources of data.
APPENDIX 2. List of maps used to identify the river km of locations used during the 1991-92 Trinity River spawner survey.
APPENDIX 3. Summary of chinook salmon carcasses recovered during the 1991-92 mainstem Trinity River spawner survey.
APPENDIX 4. Summary of coho salmon carcasses recoverd during the 1991-92 mainstem Trinity River spawner survey.
APPENDIX 5. Summary of salmon carcasses and redds observed during the 1991-92 spawner surveys in tributaries to the Trinity River between Lewiston Dam and the North Fork Trinity River.
APPENDIX 6. Sex composition of adult chinook salmon observed during mainstem Trinity River spawner surveys from 1942-1991.
APPENDIX 7. Female chinook salmon prespawning mortality rates observed during mainstem Trinity River spawner surveys from 1942-1991.
APPENDIX 8. Release and recovery data for coded-wire-tagged salmon recovered in the 1991-92 mainstem Trinity River spawner survey.