LIFE HISTORY, DISTRIBUTION, RUN SIZE AND ANGLER HARVEST OF STEELHEAD IN THE SOUTH FORK TRINITY RIVER BASIN (continued)
Sampling data collected by Project personnel are presented in Julian week (JW) format. Each JW is defined as one of a consecutive set of 52 seven-day (weekly) periods, beginning 1 January, regardless of the day of the week on which 1 January falls. The extra day in leap years is included in the ninth week, and the last day of the year is included in the 52nd week (Appendix 1). This procedure allows inter-annual comparisons of identical weekly periods.
Run timing of adult fall-run steelhead into the SFTR basin was monitored with an immigrant weir located at Sandy Bar in the South Fork Trinity River at river kilometer (RKM) 2.4 (Figure 1). The weir consisted of a series of panels, measuring 1.2 m high and 1.5 m wide constructed of 1.9-cm EMT (electrical metallic tubing) conduit welded to angle-iron frames with 3.2-cm horizontal bar-spacing. The panels were wired end-to-end and supported with metal fence posts. A trap (2.4 m wide x 2.4 m long x 1.2 m high), with sides constructed from the same weir panels, and flooring and top from marine plywood, was placed in the river thalweg with its fyke entrance facing downstream. The weir panels were tied in with the trap and extended outward across the river guiding upstream migrating fish into the trap. Small mesh netting was strung above the weir to prevent fish from jumping over.
Each steelhead captured was examined for fin clips, tags, and scars. Scars were categorized as: gill-net scars (nicks in the leading edges of dorsal and pectoral fins, sometimes combined with vertical white scars on the head); hook scars (of ocean origin when healed, of freshwater origin when not healed); predator scars (inverted 'V'-shaped marks, usually on the underbody); and other scars of unknown origin. Steelhead were measured to the nearest cm fork length (FL), and their sex recorded. A scale sample was removed from the left side of each weir-caught fish, from an area slightly posterior to the anterior insertion of the dorsal fin, just above the lateral line. Each scale sample was placed between waterproof paper within a coin envelope and labeled with collection date, collection site, method of collection, sex, and FL (cm) of the fish.
All adult steelhead in good condition were marked with a one-half right ventral (2RV) fin-clip. In addition, every third fish was tagged with a gray, discretely numbered $10-reward anchor tag, while the other two fish were tagged with green, discretely numbered non-reward anchor tags. This was a change from our previous years' procedure when all adult steelhead received reward tags. This was done in an attempt to discourage anglers from fishing primarily for money. The tags and fin clips were applied with the intention of computing a Petersen population estimate (Ricker 1975) based on the ratio of tagged to untagged fish observed in later recovery efforts (creel census and emigrant weirs). Angler harvest was to be estimated from reward tag returns.
To avoid excessive tagging mortality, we did not tag fish which were severely stressed by the weir capture and handling process, or those which appeared in generally poor physical condition.
During the operation of the Sandy Bar Weir all adult salmon caught were processed similarly to steelhead, except they were not tagged. Fish judged to be in poor condition were just identified and counted, then released to continue their upstream migration. Chinook salmon (O. tshawytscha) and coho salmon (O. kisutch) were given a right opercular punch (OPR) prior to their release. This was done so that investigators from other projects surveying the SFTR basin could identify salmon which had been caught at the Sandy Bar Weir.
Downstream emigrant weirs were used to capture post-spawning steelhead emigrating from the basin in order to recover fish tagged at the Sandy Bar Weir. These tagged fish, combined with those observed during the creel surveys in the basin, served as the recapture sample for our population estimates using the Petersen method of mark and recapture (Ricker 1975, p. 78, formula 3.7).
This season only one emigrant weir was installed due to high spring flows. We assembled a weir in the SFTR near the town of Forest Glen at RKM 89.6 (approximately 150 m downstream from the Highway 36 bridge). In past years we had also operated a weir on lower Hayfork Creek near the town of Hyampom (8.0 RKM upstream from the SFTR confluence).
We constructed an Alaskan-style weir at the Forest Glen site using a series of panels 3.2 m high and 3.0 m long, supported by wooden tripods set 2.4 m apart, and joined together to block the entire river. Each panel consisted of 1.9-cm EMT conduits set 2.9 cm apart (46 per panel), and secured through three aluminum channel sections on the face of the weir. A trap constructed of welded conduit panels and containing a fyke entrance was placed in the river thalweg with its entrance facing upstream. All steelhead recovered were measured (cm FL), and checked for spawning condition, tags, fin clips, and marks. Each fish was also sampled for scales, and given a OPR before being released downstream of the weir.
In addition to the downstream (emigrant) trap, we also installed an upstream (immigrant) trap to capture spring-run steelhead (also known as summer steelhead) and spring-run chinook salmon. This work was done in cooperation with CDFG's Trinity Fisheries Investigations Project, which was studying spring-run chinook salmon stocks in the SFTR basin. The Trinity Fisheries Investigations Project took over the operation of the Forest Glen Weir on 29 June 1993. This report covers catches of steelhead at the Forest Glen Weir through 30 June 1993.
Angler effort and harvest information for fall-run steelhead within two areas of the SFTR basin was determined from a systematic creel survey stratified by JW, section (upper/lower), day (weekend/weekday), and time periods (AM/PM; dawn to noon/noon to dusk, respectively).
Two sections of the SFTR basin were surveyed (Figure 2). The lower survey area extended from the confluence of the SFTR with the mainstem Trinity River to 22.5 km upstream. The upper area extended through the Hyampom Valley from RKM 33.0 to RKM 50.7. These two areas covered the river reaches fished by the majority of anglers, as the lack of public roads limits access. Angler access sites had been identified from past surveys.
Survey clerks followed a set route based on a predetermined schedule, and monitored each access site for anglers. Anglers observed were interviewed for number of hours fished that day, targeted species, success, angling method, and county and state of residence. Sport-caught salmonids observed were measured (cm FL), sexed, examined for fin clips and external tags, inspected for general body condition, and scale sampled. The number of any tag observed was recorded. We classified steelhead <25 cm FL as juveniles, >25 cm FL and <41 cm FL as half-pounders, and >41 cm FL as adults.
Data were extrapolated under the assumption that angling effort, angler numbers, and steelhead harvest were constant for the duration of each stratum sampled. A ratio of the number of legal fishing hours possible during the AM or PM stratum to the hours sampled during that stratum yielded a weighting factor which was used to expand observed angler numbers, angler hours, and steelhead harvest. Expanded estimates for strata not surveyed were calculated by using average values for strata from equivalent sampling periods (i.e., for a missing weekday evening survey, the mean of all weekday PM survey samples for that section during that JW was used). Expanded estimates and actual data were combined to give an estimate of sport harvest for the season in the SFTR basin.
All reward tags from Sandy Bar Weir that we observed during the creel surveys were left with the angler to return to us by mail. This was done so we could calculate an overall SFTR basin sport harvest for fall- and winter-run steelhead. The percentage of reward tags caught by anglers which were not returned to us (i.e., non-response) was calculated from the number of reward tags we observed during our creel surveys, and the number of those tags which were subsequently returned to us by mail. The sport harvest estimate used the number of reward tags returned by anglers divided by the non-response, and the number of tags applied at the Sandy Bar Weir. The estimated sport harvest was based on the following assumptions: 1) that all tagged fish caught in the sport fishery were recognized as such by anglers; 2) no tags were shed; and 3) there was no differential mortality between tagged and untagged fish.
Project personnel conducted walking surveys of tributary streams to the SFTR and Hayfork Creek to document steelhead spawning distribution and timing. The areas surveyed included tributaries in the Hyampom Valley near the towns of Hayfork and Wildwood, and tributaries in the upper SFTR basin near the town of Forest Glen (Figure 1). Specific creeks surveyed were selected to include those which had historically attracted spawning steelhead, and to replicate areas examined in previous CDFG surveys (Miller 1975; Mills and Wilson 1991; Rogers 1972, 1973; Wilson and Collins 1992; Wilson and Mills 1992).
Most streams were surveyed twice. During the first survey, two people walked and habitat-typed designated stream reaches recording the length and type of each habitat unit, observed spawning behavior, and individual redd locations. Each habitat unit was classified as either a cascade, pool, riffle, run, or step-run. Redds were flagged with surveyor's tape with the survey date and field notebook description number recorded on the tape. The tape was then attached to nearby structures such as root-wads, shrubs, or bushes. During the second survey, redd characteristics (area and depth), site descriptions (substrate and cover composition), and stream conditions (water velocities) were compiled for individual redds. New redds established since the first survey were included.
We evaluated steelhead spawning habitat by measuring the physical and hydraulic parameters of observed redds, and recording the characteristics and quality of the substrate and associated cover.
Length and width measurements were taken of each redd using a meter stick or tape measure, from the upstream end of the redd to the highest point of the tailspill, and perpendicularly across the widest point of the redd. An index of the surface area occupied by the redd area was calculated as the product of the length and width. Water depths were taken using a graduated top-setting wading rod, and water velocities were measured with an electronic flow meter. Two separate water velocity measurements were taken, mean water column velocity (MWCV) and fish-nose water velocity (FNWV). MWCV measurements were taken at 60% of the depth below the water surface, and FNWV measurements were taken 0.12 m above the substrate. Redd substrate composition was determined by assessing the average size of the dominant and subdominant components, and the percent embeddedness of each (Hampton 1988) (Table 1). The water velocity measurements and the substrate analysis were all made approximately 0.15 m upstream of the redd in an attempt to simulate prespawning hydraulic and substrate conditions. Distance to the closest cover, escape or resting place was noted, as well as the dominant habitat type in which the redd was located.
TABLE 1.Criteria used to describe the size of dominant and subdominant spawning gravel substrate.
|
Data code |
Substrate type |
Size range (mm) |
|
0 |
Fines |
< 4 |
|
1 |
Small gravel |
4-25 |
|
2 |
Medium gravel |
25-50 |
|
3 |
Large gravel |
50-75 |
|
4 |
Small cobble |
75-150 |
|
5 |
Medium cobble |
150-225 |
|
6 |
Large cobble |
225-300 |
|
7 |
Small boulder |
300-600 |
|
8 |
Large boulder |
>600 |
|
9 |
Bedrock |
We monitored juvenile steelhead emigration patterns by systematically trapping at two sites. One site was located in lower Hayfork Creek, 305 m upstream of its confluence with the SFTR. The other site was located in the SFTR, upstream of its confluence with Hayfork Creek, within 400 m of the Hyampom Road bridge at RKM 49.1 (Figure 1). When flow conditions permitted, we trapped once a week most of the year. During the spring period of peak juvenile steelhead emergence, 23 April - 22 July (JW 17-29), we increased trapping frequency to twice weekly.
Juvenile salmonids were captured using fyke nets attached to trap boxes. The nets were constructed of 1.3-cm nylon mesh, had a 1.8-m x 2.4-m upstream opening and extended 10.1 m to a 0.33-m x 0.33-m terminal end. Trap boxes were constructed of marine plywood and hardware cloth, and measured 0.8 m x 1.2 m x 0.5 m. One or two fyke-net traps were fished overnight in the river or stream, for 16- to 24-hour periods, and examined the following morning.
Captured fish were identified to species and enumerated. The first 50 individuals of each species removed from the traps were measured for FL (mm). Scale samples were systematically taken from a maximum of 10 individuals of each species, at each trap site, each sampling day.
Water temperature and stream flow were measured at the net opening each time the traps were set. Total volume of stream flow through the net was measured to the nearest 0.031 m/sec using either a pygmy meter or a Marsh-McBirney flow meter. Water temperatures were monitored using hand-held thermometers or digital recording thermographs.
Habitat use by juvenile steelhead was studied in Eltapom Creek (Figure 1) in the fall of 1992 (1-3 September). Prior to sampling, the creek was surveyed and delineated into units of five basic habitat types: cascades, pools, riffles, runs, and step-runs. Sampling was conducted by electrofishing. Habitat units sampled were randomly selected in proportion to the numeric abundance of each of the five basic habitat types. Our goal was to sample one-third of the units in each habitat type.
Sample units were isolated using block nets to prevent any immigration or emigration of fish, and then electrofished. All steelhead captured were counted, measured (mm FL), sampled for scales (first five per habitat unit), and then released.
Age delineation of fish captured was based on lengths. Fish <85 mm were classified as Age 0+, fish 86-150 mm as Age 1+, and fish >150 mm as Age 2+. Numbers of fish caught were used to determine the relative densities for each age group in each habitat type based on total catch-per-area. The number of fish caught in the same type of habitat unit was expanded, based on the relative densities multiplied by the total area available in each habitat type. These figures were then totaled to give a standing crop estimate for the entire stream.
After sampling, physical parameters of each unit were taken. We recorded air and water temperatures with hand-held thermometers. Water velocities (to the nearest 0.031 m/sec) were measured at 60% of the total depth from the surface along a line transverse to the flow at points one-quarter, one-half, and three-quarters of the way across the stream. Stream length and width were measured to the nearest 0.03 m.
