LIFE HISTORY, DISTRIBUTION, RUN SIZE, AND HARVEST OF SPRING CHINOOK SALMON IN THE SOUTH FORK TRINITY RIVER BASIN (continued)
The following results are repeated from the 1990-1991 Annual Report to allow the reader to follow the 1991 spring chinook cohort through the summer and fall, and to more clearly understand our methodologies and results.
We operated the Gates Weir for 80 days, from 28 April through 18 July 1991. During this period, both immigrant and emigrant traps were maintained. We captured, marked, and released 34 adult and nine grilse spring chinook, eight unspawned adult fall- or winter-run and 18 adult spring-run steelhead from the immigrant trap. We captured, examined, and released 39 out-migrant (spawned) adult winter-run steelhead from the emigrant trap (Table 1). We effectively marked 39 chinook salmon at the Gates Weir, 19 with Floy anchor tags and 20 with a RV fin-clip.
In 1991, we began catching spring chinook at the Gates Weir during the second week of May. The run appeared to reach its peak during early to mid-June (Table 1). We continued to catch fish until mid-July when we were forced to remove the weir due to excessively warm water temperatures (>21 oC). Based on snorkel observations, we feel that some spring chinook continued to enter the SFTR until the end of July. Therefore, the run timing in 1991 for spring chinook in the SFTR was from early May through late July.
Spring chinook captured at the Gates Weir averaged 60.4 cm FL (" 9.6 cm S.D.) (Figure 3). TFIP has established 55 cm FL as the length separating adults and grilse in the mainstem Trinity River. We may revise this length cut-off for SFTR fish based on scale analyses. However, we have not yet read enough scales to make a final determination. Length data for steelhead captured at the Gates Weir are being reported in the Annual Report for Job III and will not be discussed here.
Of the chinook captured and sexed at the Gates Weir, 26 were females and 12 were males. Small grilse may have gotten through the Gates Weir at a higher rate than larger adults, accounting for the lower male count. Of the chinook recaptured and sexed at the Forest Glen recovery weir, seven were females and six were males.
Recovery Weirs. No spring chinook were captured at the Hayfork Creek Weir, while 14 were captured at the Forest Glen Weir. Of these 14 fish, two were RV fin-clipped, indicating that they were processed through the Gates Weir (Table 2). No tagged fish were recovered. This number of mark recoveries was inadequate for a valid Petersen estimate.
Snorkel Surveys. We observed 66 spring chinook and nine spring-run steelhead during snorkel surveys. Twelve of the spring chinook were seen below the Gates Weir. Four of these fish may have moved above the weir after its removal. However, most remained below the weir until at least mid-September. All twelve were included in the run-size estimate.
Throughout the surveys, only three anchor-tagged spring chinook were seen. One additional spring chinook was observed with a LV fin-clip and an obvious tag scar. No fish were seen with the right ventral fin-clip. Therefore, we observed four marked spring chinook among the 54 seen above the weir. Unfortunately, this number of mark recoveries is insufficient for a statistically valid Petersen estimate. However, utilizing these numbers in the Petersen formula, and based on field observations, we estimate the run-size to be 232 fish.
In some pools, we observed spring chinook from a nearby bluff before conducting the snorkel survey. In one case, we observed four fish in a pool near Hidden Valley prior to entering the water. However, while snorkeling, only one fish could be found. Therefore, we believe the snorkel survey may only provide a lower estimate of the number of fish.
One of five spring chinook was observed to have shed its tag. Based on this small sample size, it appears that a tag shedding rate of 20 to 25% is possible.
No spring chinook were seen during surveys of lower Hayfork Creek or Grouse Creek.
Follow-up Observations at Holding Pools. Near the end of August and through mid-September, spring chinook numbers increased in each pool. We assumed fish we had seen during the snorkel survey in poor holding areas, such as glides and step-runs, had moved into holding pools. However, since we could not identify individual fish, this could not be confirmed. We also noted that, as September progressed, fish exhibited more and more chasing behavior and some pairing was apparent. This may be an important clue in determining when fish are nearing spawning condition. In the last several days of September, spring chinook began leaving pools and moving into glides and riffle areas, indicating the onset of spawning.
Spawner and Redd Surveys. We performed 30 individual surveys between 18 September and 6 November, and located 25 spring chinook redds. Twenty-four redds were distributed almost equally above and below Forest Glen (Figure 4); one was found well below Hyampom at RKM 38.3, and is not shown in Figure 4. Three spring chinook were seen to over-summer in the pool immediately upstream of this redd. Spring chinook spawning was most concentrated in river section E, near Silver Creek Ranch. All redds were typical for spring chinook salmon with regard to size, location in the stream, gravel size, current velocity, and water depth (Chapman 1943; Mattson 1948; Cramer & Hammack 1952; Lindsay & Jonasson 1989; Groot and Margolis 1991). No spring chinook were seen spawning in Hayfork Creek. The weather and water clarity were excellent during these surveys.
Based on observations, we estimated that there were between two and three spring chinook per redd. If this estimate is accurate, then only about 65 fish survived to spawn. However, it is conceivable that we missed a few redds in the upper river (sections A, B, C, and D), and that there was considerably more spawning occurring than we accounted for.
SFTR spring chinook were observed to complete redd construction in about 24 hours, with evidence of false redd activity in almost every case. Females could be found in the area of the redd for only three-to-four days after redd completion. They were never seen to defend their redd. Although individual fish could not be identified, in two instances we discovered two redds in isolated areas where only one female was observed. This led us to believe that females might dig more than one redd. Spawning commenced in the upper river on 3 October and progressed downstream. Spawning was completed by the end of the third week of October (26 October). Although, in a few instances, redds were within a few meters of each other, we did not observe any overlapping of redds.
Carcass Recovery Surveys. We were able to recover only one spring chinook carcass during our redd and carcass recovery surveys. This carcass was discovered on 11 October just above the Butter Creek pool. It was a fresh, unspawned 60-cm FL female with a clear LV fin-clip, but no apparent tag scar. No cause of death was apparent. This may have been the tagged fish we knew to be holding in the Butter Creek pool. Lindsay and Jonasson (1989) reported average pre-spawning mortality rates in wild spring chinook of 44% for the Deschutes River (Oregon) from 1977-81, with some years as high as 75%. The Rogue River (Oregon) experienced average rates of 12% for the same period (Lindsay and Jonasson 1989). For comparison, pre-spawn mortality rates for spring-run chinook in the mainstem Trinity River averaged 62.8% in 1990 (Zuspan 1992). Groot and Margolis (1991) reported that much lower rates (less than 10%) are more typical. High pre-spawn mortality rates are often associated with stress factors such as high water temperature, microbial agents, or a combination of the two (Groot and Margolis 1991).
We found remnants of chinook carcasses (various fins and skull bones) in several areas, but these observations are of little value except to indicate a degree of predation and scavenging which we felt was occurring. River otter and mink were the primary predators and scavengers. As evidenced by tracks and scat traces, bear activity was light in spring chinook spawning areas.
Other Observations. On several occasions during snorkel surveys, we observed spring chinook moving upstream through high-gradient riffles and step-runs when water temperatures were greater than 22.5 oC. On one occasion, the water temperature was 24 oC. It is noteworthy that these fish can not only tolerate these temperatures, but appear to be able to migrate in them. It may be that such warm temperatures motivate fish to move farther upstream.
Our snorkel observations suggest there is no shortage of over-summer holding habitat for the current population of spring chinook in the SFTR. We found numerous pools which appeared to be of adequate depth, had good in-stream cover, and good thermal stratification (cool bottom water) which were not being utilized by spring chinook. We documented eight spring chinook summer-holding pools spread throughout the SFTR, primarily above Hyampom (Figures 5 & 6). Each of these pools was occupied by at least three spring chinook; most had five or more. We also documented one pool downstream of Hyampom where spring chinook not only over-summered, but survived to spawn (see Spawner and Redd Surveys, p. 209).
During the juvenile trapping effort and other field observations, we saw numerous green sunfish in the SFTR, especially lower Hayfork Creek. On two occasions, we electrofished a large backwater pool about 1 km above the Gates Weir. We estimated as many as 200 sunfish in this pool, representing at least three age-classes. Stomach content analysis showed that the larger of these fish were piscivorous. Sunfish were also seen in the main river-channel. This question is beyond the scope of this project, but we are concerned that large numbers of these fish could be responsible for significant predation on emigrant juvenile salmonids.
We operated the weir at Sandy Bar for 26 days from 4 September to 1 October, attempting to capture late-entering spring chinook. During this period, we captured one grilse spring chinook, one adult fall chinook and two adult fall- or winter-run steelhead. During the period of operation, river flows were low and efficiency of the weir was considered 100%.
Water temperatures in the lower river during this period ranged from 19 to 28 oC, averaging 21 oC. The mainstem Trinity at the SFTR confluence is consistently much cooler (by 4 to 6 oC). We feel that this temperature difference would inhibit most chinook from entering the SFTR during August and September. The low number of spring chinook captured supports this idea. Also, during this time of year, flows in the SFTR can be so low as to make significant upstream migration difficult. Consequently, any late-entering spring chinook would likely enter the river with early-entering fall-run chinook. As a result, differentiating between spawners of the two races would be very difficult.
Scale Analysis. We were able to interpret 42 of 43 scale sets recovered from immigrant chinook marked and released at the Gates Weir this season. The one unreadable set was composed entirely of regenerated scales which precluded analysis. Thirty-seven scale sets (88%) showed an ocean-type juvenile life history, while five (12%) showed a stream-type life history.
Of the 42 scale sets, seven fish (17%) were two-year-olds (grilse), 12 (29%) were three-year-olds, 19 (45%) were four-year-olds, and four (9%) were five-year-olds (Figure 7). Interestingly, of the five stream-type fish, two were three-year-olds, while three were five-year-olds. Lindsay and Jonasson (1985) reported age classes by scale analysis for spring chinook from the John Day River as 1-5% three-year-olds, 54-89% four-year-olds, 8-44% five-year-olds. Virtually all these fish were stream-type (Lindsay 1985).
The average FL for returning SFTR spring chinook in age groups two- through five-years was 46.7, 59.3, 64.5, and 66.8 cm, respectively (Figure 7).
Juvenile Emigrant Trapping. We trapped at the Forest Glen site for 41 nights between 16 January and 9 July 1992. Over this period, we captured and released 490 young-of-the-year (YOY) and four yearling spring chinook. The emigration of yearling spring chinook was scattered throughout the winter and early spring. We captured the first spring chinook YOY on 9 April along with the last yearling. The peak of spring chinook YOY emigration at Forest Glen was between late April and early June (Figure 8).
Natural Stocks Assessment Project (NSAP) personnel trapped emigrant juvenile salmonids for 37 nights in the SFTR near Hyampom between 9 November 1991 and 15 July 1992, as part of Job III. Unfortunately, fall-run chinook are known to spawn above this trapping site. Therefore, juvenile chinook salmon captured here cannot be positively separated by race. Also, fewer chinook salmon were trapped at the Hyampom site compared to Forest Glen (Figure 9). Due to higher flows near Hyampom, trapping in the main channel was seldom feasible, and when possible, a smaller percentage of the river was trapped, accounting for the difference in catch.
NSAP's results were almost identical to ours in the number of yearling chinook salmon captured (six compared to four) and in the timing of yearling and YOY emigration (Figure 9). Only four of six yearling chinook captured by NSAP staff are shown in Figure 9; two were captured in early November (these were omitted from the graph so that Figures 8 and 9 could be shown with similar axes). In early April the average FL for chinook salmon trapped by NSAP staff at Hyampom was 53.6 mm, compared to 53.8 mm at our Forest Glen site. By the end of May the average FLs had increased to 74 mm at Hyampom and 67 mm at Forest Glen. This difference may represent the growth that occurs between these two points. However, comparisons between our Forest Glen data and NSAP's Hyampom data must be made with caution. The NSAP juvenile trap location is such that juvenile fall-run chinook may also be captured.
Only one juvenile chinook salmon was captured by NSAP staff in the Hayfork Creek trapping operation. Since no spring chinook were seen to over-summer or spawn in Hayfork Creek, we assume that this fish was from fall-run stock.
The catch-per-unit-effort (CPUE) at Forest Glen varied from zero to a high of 84. CPUE is defined as the number of fish caught per trap, per day. The average FL for YOY increased over the period from 53.8 to 91.7 mm (Figure 10). It was apparent that spring chinook YOY first captured at Forest Glen had been out of the gravel for some time, as evidenced by their relatively large size.
Yearling spring chinook measured 87, 89, 100, and 116 mm; the smallest caught first and the largest caught last. From our trapping operation and trapping performed by NSAP, it appears that yearling spring chinook emigrate during winter and early spring and do not mix with emigrant YOY to any significant extent. This life history strategy is represented by a small percentage of the population. It is not yet known how significant this life history strategy may be to the survival to spring chinook in the SFTR.
Coincident with this effort we captured and released 1,369 juvenile steelhead, 826 speckled dace, four green sunfish, five golden shiners, and a few thousand ammocetes.
Snorkel Survey Observations. We did not see yearling spring chinook during winter and early spring snorkel surveys. We observed five spring chinook YOY on 9 April, the first date we captured spring chinook YOY in fyke net traps at our Forest Glen trap site. These fish averaged 45 mm in length and were found in edgewater, among fine plant roots. Much of the SFTR in this section provides good edgewater habitat, as well as good habitat among large cobbles. Due to very cold water temperatures and fast flows, we were unable to document the use of the cobble habitat. Other habitat types in the section were apparently unused. No schooling behavior was noted in any habitat type.
Based on information presented by Alderdice and Velson (1978) and Dill (1969), we expected to see emergent fry in mid-March. Our field observations indicate that this time frame is accurate. Therefore, it took about 150 days from the time eggs were deposited until fry emerged from gravel. We also discovered that some fry take refuge in edgewater vegetation after emerging from the gravel. They appear hold there until foraging behavior and swimming abilities develop, a period of about 30 days. Reiser (1981) reported that fry were first seen in pools just upstream of redds. We did not see this behavior.
Seven individual creel surveys were performed during June of 1991. We saw no chinook salmon creeled, and no tags have been returned. CDFG staff members who reside in the Hyampom area received no reports of chinook salmon being caught anywhere in the SFTR this season. Based on these data, we believe the angler harvest of spring chinook in the SFTR in 1991 was zero.
Thermographs worked very well and will continue to be used to monitor river temperatures. Our primary purpose for monitoring river temperatures at the weir sites is to detect unacceptably high minimums where handling fish could cause lethal stress. Daily average water temperatures at the Gates Weir ranged from 14 to 22 oC between 29 April and 29 June. The Gates Weir was removed from operation on 24 June because minimum daily water temperatures reached 21 oC. Daily diurnal temperature variations ranged between 3 and 5 oC (Figure 11).
FIGURE 11.Water temperature variation at the Gates Weir in the South Fork Trinity River from 29 April through 24 June 1992.
During the 1992 season we operated the Gates Weir for 64 days, from 27 April through 7 July 1992. Late in the trapping period we were forced to suspend trapping operations intermittently due to excessively warm minimum water temperatures. During this period both immigrant and emigrant traps were maintained. We captured, marked, and released 39 adult and nine grilse spring chinook, one unspawned adult winter-run and 15 adult spring-run steelhead from the immigrant trap. We captured, examined, and released 65 out-migrant (spawned) adult fall- and winter-run steelhead from the emigrant trap (Table 3). Thirty-nine chinook were tagged with anchor tags and marked with a one-half LV fin-clip, and nine were implanted with radio tags and given a one-half RV fin-clip. Further analysis and discussion of these data will be covered in the next annual report where these spring chinook will be followed through the end of their spawning season.
During the 1992 adult trapping season, we examined 49 spring chinook and 81 steelhead at the Gates Weir. Only 28% of spring chinook showed scars this year, compared to 67% last year (Table 4). Conversely, 41% of the steelhead had scars this year, compared to only 15% last year. These numbers are significantly different (X2 = 10.0, X2 = 12.0, respectively). The composition of scar types on steelhead was generally comparable to last year with the exception of predator scars, which were up by 24%. Gill-net scars on spring chinook were 18% lower than last season. It has been reported by fisheries staff of the Hoopa Valley Tribe (M. Orcutt, telephone conversation), that the gill-net fishing effort for spring chinook in the lower Klamath-Trinity system was less intense this season compared to other years. Interestingly, we saw no hook scars on spring chinook this year compared to 13.9% last year. Angling regulations and open seasons were much more restrictive this year, and may have contributed to these observations.
1. Snorkel surveys should be performed in late July in as short a time frame as possible for a better delineation of spring-run chinook salmon distribution and a more accurate count for the Petersen estimate. The snorkel surveys should also be repeated near the end of August to document any re-distribution of adult and grilse salmon.
2. Follow-up observations of summer holding pools should be continued to document immigration and emigration, pre-spawning behavior, and to count the numbers of tagged and untagged chinook salmon in each pool.
3. An attempt should be made to evaluate the efficiency of the snorkel survey technique.
4. Discontinue the trapping operation aimed at the late-entering portion of the spring chinook salmon run since only one spring-run fish was caught this season. Also, during drought years, excessive water temperatures and low flows appear to inhibit, and possibly prohibit, any significant late-entering segment of the run. River temperatures should be monitored in the lower river to document extremes.
5. Consider using several different color-coded tags which would allow for the identification of individual fish during snorkel surveys, and especially during follow-up observation at holding pools.
6. Consider moving the SFTR recovery weir nearer to Hyampom in an effort to recapture more marked fish which would allow for a more valid population estimate.
7. Poor spawning gravel permeability and bedload movement may be affecting spring chinook salmon egg and alevin survival. Additional studies are needed in this area.
Alderdice, D. F., and F. P. J. Velson. 1978. Relation between temperature and incubation time for eggs of chinook salmon. Journal Fisheries Research Board of Canada, 35: 69-75.
Chapman, W. M. 1943. The spawning of chinook salmon in the main Columbia River. Copeia 1943: 168-170.
Cramer, F. K. and D. F. Hammack. 1952. Salmon research at Deer Creek, California. United States Fish and Wildlife Service. Special Scientific Reports, Fish. No. 67. 16 p.
Dill, L. M. 1969. The Sub-gravel behavior of pacific salmon larvae. p. 89-100. In: T. G. Northcote (ed.), Symposium on Salmon & Trout in Streams. H. R. MacMillian Lectures in Fisheries, Univ. of British Columbia, Vancouver, B.C. Canada.
Groot, C., and L. Margolis. 1991. Pacific Salmon Life Histories. UBC Press. University of British Columbia. Vancouver, B.C. Canada. 564 p.
LaFaunce, D. A. 1964. A king salmon spawning survey of the South Fork Trinity River. Calif. Dept. of Fish and Game, Mar. Res. Admin. Rep. No. 67-10., 1964. 13 p.
Lindsay, R. B. 1985. Study of wild spring chinook salmon in the John Day River system. Oregon Dept. of Fish and Wildlife. Available from Bonneville Power Administration, Div. of Fish and Wildlife, P.O. Box 3621, Portland, Oregon 97208. 119 p.
Lindsay, R. B., and B. Jonasson. 1989. Spring chinook salmon in the Deschutes River, Oregon. Oregon Dept. of Fish and Wildlife. Fish. Div. Information Reports, 89-4.
92 pp.
Mattson, C. 1948. Spawning ground studies of Willamette River spring chinook salmon. Oregon Fish Commission Resource Briefs, (12): 21-32.
Mills, T. 1986. Juvenile life history patterns of South Fork Trinity River chinook salmon as determined by adult scale analysis. Calif. Dept. of Fish and Game, Anad. Fish. Br. Admin. Rep. (manuscript). 20 p.
Reiser, D. W. 1981. Influence of streamflow reductions on salmonid embryo development and fry quality. Idaho Water & Energy Resources Research Institute, University of Idaho. 154 p.
Ricker, W. E. 1975. Computation and interpretation of biological statistics of fish populations. Bull. Fish. Res. Bd. Can. No. 191. 382 p.
Snyder, J. O. 1931. Salmon of the Klamath River, California. Calif. Div. of Fish and Game, Vol. 10, No. 4, p. 163-172.
Sullivan, C. 1989. Juvenile life history and age composition of mature fall chinook salmon to the Klamath River, 1984-1986. Masters Thesis. Humboldt State University, Arcata, California 95521. 69 p.
Zuspan, M. 1992. 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, 1989-90 season. June 1992. 140 p. Available from Calif. Dept. of Fish and Game, Inland Fish. Div., 1416 9th St., Sacramento, CA. 95814.
APPENDIX 1. Other sources of data.
