Tributaries to Hayfork Creek
Rusch Creek
This stream flows into Hayfork Creek from the west just below Hayfork Valley. A survey of 6.2 miles of Rusch Creek was conducted by the USFS from August 1-5, 1991 and subsequently reported by Mayo (1992). The stream flows through forested land. The lower reach surveyed has a moderately incised channel (Rosgen B-2) while the upper reach is deeply incised (A-3). During the survey, Rusch Creek stream temperatures ranged from 55 to 65°F, and averaged 60°F. Step runs were the dominant habitat type (39%) with lateral scour boulder pools the second most frequently occurring unit (12%). Pools totaled over 46% of all habitats by length but were restricted in depth. By volume, step runs made up 65% of all habitat and pools comprised only 29%.
Of the 705 juvenile steelhead counted during the habitat typing survey, 73% were young of the year (0+), 17% were yearlings (1+), and the remaining 10% were 2+. Highest densities of 0+ steelhead were found in glides, main channel pools and bedrock lateral scour pools. Surprisingly, young of the year showed little preference for low gradient riffles. Both older age classes of steelhead were found in greatest concentrations in dammed pools and step pools. Plunge pools were also favored by 2+ steelhead.
Substrate analysis showed that sand comprised a large percentage of the stream bottom (22%). Pools had a combined total of over 30% sand and fine sediment and spawning gravels had 29% of these components. Embeddedness of the stream averaged 32%. The high amount of sand in the channel is due to erodible, weathered diorite and associated sandy soils found in the watershed. Mayo (1992) suggests that large quantities are originating from roads that border the stream. Fine sediment and sand in the stream may be decreasing spawning success and invertebrate production. Lack of pool volume is certainly decreasing available rearing habitat. The full benefit of fish habitat improvement structures placed in Rusch Creek by the USFS and CCC may not be realized until erosion problems have been remedied (Gary Flosi, personal communication).Since sediment has only recently been contributed to the stream, a corresponding decrease in the fish population may not have occurred yet.
Salt Creek
Two habitat typing surveys have been conducted on Salt Creek. Wilcox et al. (1990) reported findings of a survey conducted by EBASCO Environmental from May 25-June 15, 1989 on the lower 11 miles of Salt Creek. The upper 3.8 miles of Salt Creek (on USFS lands) were habitat typed from June 26-August 9, 1989 by the Shasta Trinity National Forest staff (USFS, 1990e).
Lower Salt Creek: Just above its convergence with Hayfork Creek, Salt Creek has a mild stream gradient (0.8%). The channel is only slightly confined, dominated by cobble and gravel, and has areas with unstable banks (Rosgen C3). Above the Highway 3 bridge, Salt Creek becomes more confined, running through forest land, where conifers provide shade and their root masses assist in forming stable stream banks.
The most abundant habitat types (by length) were low gradient riffles (21%), main channel pools (16%), step runs (15%), glides (13%), and runs (11%). Shallow dammed pools and main channel pools made up 48% of the habitat by volume. Cobbles (38%) and gravel (23%) were the dominant particle sizes, but sand made up 20% of the substrate. Sand and fines comprised over 41% of the substrate in pools and embeddedness in spawning areas was 33%. Embeddedness over 25% decreases steelhead spawning success (Reiser and Bjornn, 1979). The high level of fine sediment was ascribed to bank erosion which was noted as a problem, particularly in areas downstream of Peanut, California.
Although the main form of cover in pools was over-hanging vegetation, pools had only 8% cover on average. Lack of riparian vegetation may be in part a result of over-grazing, as much of Salt Creek flows through farm and ranch land. Lack of stream shade also contributed to high stream temperatures, which exceeded 76°F. Stream temperatures later in summer, after the survey was conducted, were likely much higher. Low stream flows, due to agricultural and domestic diversions, compound problems of increasing water temperatures. All diversions noted in the survey did have fish screens.
Steelhead of all age classes, including adults, were seen during the survey. While the adult steelhead were described as summer steelhead, it is more likely that they were late winter steelhead that got trapped by low flows after spawning in April and May. Young of the year steelhead (0+) comprised 97% of all age classes present. Of the 4819 juvenile steelhead counted, only 83 were 1+ and 52 were 2+. Habitats most favored by 0+ steelhead were high gradient riffles and trench pools, although the latter habitat type was infrequent. Other run and riffle types also held high numbers of 0+, as did lateral scour bedrock pools. Older age steelhead juveniles were most abundant in trench pools. The second most favored habitat for 1+ and 2+ fish were high gradient riffles and cascades, respectively. Glides, main channel pools, and dammed pools were infrequently used by any age class.
The density of juvenile steelhead decreased progressively downstream, although a significant number of 0+ fish were found just above the convergence with Hayfork Creek. High stream temperatures in summer would probably preclude survival of these fish. Poor riparian conditions, low flows, and high water temperatures are the principal limiting factors in Salt Creek and all these problems increase progressively downstream. High levels of fine sediment also are restricting pool depth and possibly decreasing spawning success and invertebrate production. Aggradation associated with bank failure may be responsible for the high incidence of low gradient riffles and glides and low frequency of pools.
Upper Salt Creek: The headwaters of Salt Creek flow from USFS lands and the stream channel is confined in a narrow valley that is heavily forested (Rosgen B-2). The average stream gradient was calculated as 2.9%. Step runs (28%) were the most abundant habitat type (by length) followed by main channel pools (18%), low gradient riffles (10%), high gradient riffles (10%), and runs (8%). Substrate showed roughly equal components of bedrock, boulders, cobbles and gravel. Sand and fine sediment made up only 16% of the substrate in pools and 11% in flat water habitats.
Densities of older age juvenile steelhead (1+ and 2+) in this reach were significantly higher than those in lower Salt Creek. In addition, the USFS (1990e) survey found a much different proportion of year classes than in the lower reaches of Salt Creek. Young of the year comprised only 41% of steelhead present, yearlings made up 44%, and 2+ fish the remaining 15%. Older age steelhead may have moved upstream to this cold water refuge area to hold during summer while 0+ steelhead may have been forced downstream due to competition. Use of habitat by fish was also different in upper Salt Creek with dammed pools, lateral scour rootwad pools, and main channel pools most favored by all age classes of steelhead. Low densities of juvenile steelhead in riffle habitats may be due to insufficient depth and flow in this upstream reach or due to sampling error. The high standing crop of juvenile salmonids in upper Salt Creek is not surprising in light of cool water temperatures (51-62°F). The low levels of fine sediment in this reach also would permit high levels of invertebrate production.
Mills and Wilson (1991) found that downstream migration of juvenile steelhead peaks in lower Hayfork Creek between May 21 and June 1. The concentration of 0+ steelhead in lower Salt Creek found by Wilcox et al. (1990) were probably migrating downstream at the time of the survey in response to increasing water temperatures. Older steelhead juveniles migrate downstream earlier in the year. Those yearlings and 2+ steelhead remaining in Salt Creek may have moved upstream in response to decreasing flows and increasing water temperatures. The high proportion of 0+ steelhead and very low number of older age steelhead is very similar to that seen in Hayfork Creek below the East Fork (R3) by Frink et al. (1990).
Big Creek (Hayfork Valley)
Big Creek is a third order stream in which 12 miles of channel were habitat typed by the USFS in 1989 (USFS, 1990f). Big Creek flows from a forested watershed in a moderately confined channel with course textured alluvial terraces and relatively stable banks (Rosgen B-2). Water temperatures averaged 60°F with the maximum of 73°F measured in the lowest stream reach below agricultural diversions. Flow depletion for domestic and agricultural water supplies contributed to increased stream temperatures.
The most abundant habitat types (by length) were step runs (27%), glides (24%), low gradient riffles (19%), and main channel pools (9%). All pools together totaled 25% of units (by length) and 15% of all habitat by volume. Substrate composition was dominated by cobbles and gravel in most habitat types and sand and fine sediment covered only 16% of pool bottoms. Embeddedness in pool tail crests was 18%. Big Creek does not show symptoms of having excessive fine sediment although surface erosion from roads bordering the stream were noted as a problem in the report.
Juvenile steelhead were present throughout the Big Creek drainage, with the highest concentration of fish in the uppermost reach surveyed. Young of the year steelhead (0+) comprised 71% of the fish present while 1+ and 2+ made up 20% and 9%, respectively. Pools had higher concentrations of juvenile steelhead with rootwad lateral scour pools, dammed pools and main channel pools the most preferred habitat types. Glides and step runs were used by all age classes of steelhead but fish densities were much lower than in most pool types. No fish were found in high gradient riffles and few were found in low gradient riffles, cascades, and runs. The major limiting factor on Big Creek is apparently related to stream flow diversion. Downstream migration is hindered, stream habitat diminished and stream temperatures are elevated.
Potato Creek
Just under five miles of this second order tributary of the East Fork of Hayfork Creek was habitat typed in July of 1991 (Gilroy et al., 1992). No stream flows were measured but Potato Creek flowed sub-surface in just over 2% of habitat units. The stream gradient averaged 7% over the reach surveyed. Channel types varied between B-2, where the stream was moderately confined with a 4% gradient to A-2 and A-2A channels which were very confined with an 8% and 10% gradient, respectively. Water temperatures averaged 58°F with a range of 54-61°F.
Step runs were 47% of all habitat units by length and 85% by volume. Plunge pools, lateral scour bedrock pools, and main channel pools each constituted 9% of habitat types with pools making up 25% of all units, by length. By volume, however, pools only constituted 14% of the stream. While sand and fine sediment comprised only 20% of all substrate and constituted only 25% of the bottom of pools, embeddness in pools tails was very high (52%). Sediment problems were noted in some stream reaches due to stream side landslides.
Young of the year steelhead (0+) made up 23% of all juveniles present with 1+ and 2+ constituting 43% and 34%, respectively. Rootwad lateral scour pools, runs and glides showed highest concentrations of 0+ fish. Runs were most favored by 1+ and 2+ steelhead but a high number of 1+ also resided in main channel pools and lateral scour boulder formed pools. Preferred habitats for 2+ included almost all pool types with highest densities in step pools and main channel pools. Densities of 0+ steelhead in Potato Creek were low, and comparable to those found in Rattlesnake Creek and the East Fork of the South Fork. However, the number of 1+ and 2+ steelhead per unit of habitat area was in the upper range of all Trinity River tributaries. Low numbers of young of the year could be in response to due to low flows and few adult steelhead spawners in the previous year.
Gilroy et al. (1992) suggest that sediment sources in Potato Creek be studied further. Lack of pool depth, stream side landslides and the loss of surface flow in some reaches during low flows would all be consistent with aggradation.
Dubakella Creek
Approximately 2.6 miles of this creek was surveyed by Shasta Trinity National Forest staff from September 11-18, 1990 (Gilroy et al., 1993a). Dubakella Creek was divided into three reaches which were all very confined (Rosgen A-2) but varied in gradient: the lowest reach had a 4% gradient, the middle reach an 8% gradient, and the upper reach a 2% gradient. Stream temperatures ranged from 47-54°F during the survey with an average of 51°F.
In light of this relatively steep gradient, it is not surprising that step runs were the predominant habitat type, by length (46%). Lateral scour boulder pools (17%), high gradient riffles (6%), plunge pools (6%), and glides (6%) were the next most frequently occurring habitats. All pools combined comprised 36% of habitat types by length, but only approximately 12% by volume. Sand and fine sediment comprised only 8% of substrates in pools, one of the lowest levels of any tributary in the South Fork Trinity River basin. Spawning gravels made up 5% of the area of all habitat types surveyed, with fine sediment levels within them visually estimated at 21%. Embeddedness in pool tail crests was high at 48%.
Juvenile steelhead were counted in all but the upper-most stream reach. Young of the year (0+) comprised 49% of fish present while 1+ and 2+ constituted 12% and 39% respectively. Plunge pools had the highest densities of 0+ fish, with lateral scour boulder pools and high gradient riffles also being favored. Yearling steelhead (1+) were present in nearly equal densities in all pool types present. Main channel pools were most favored by 2+ fish with lateral scour rootwad formed pools and step pools also holding more of this age group.
Despite cool water temperatures and apparently low levels of fine sediment, 0+ and 1+ steelhead densities were quite low relative to other South Fork Trinity River sub-basins (Table 3-1). Conversely, densities of 2+ steelhead were on par with Butter Creek, upper Salt Creek, and Rattlesnake Creek. There are two plausible explanations for this apparent anomaly. Access to Dubakella Creek may have been limited for spawning steelhead in the two years prior to the survey due to low flows and difficult passage over cascades in Hayfork Creek just above the East Fork. The second possibility is that fish counted as 2+ might be resident rainbow trout, with little use of the stream by steelhead. Low fecundity of these fish, coupled with competition and predation, may be limiting new recruits. If the rainbow trout present are largely native, recreational fishing could also be a factor in keeping populations low.
Densities of various age classes of salmon and steelhead rearing in South Fork Trinity River basin streams vary widely. Habitat typing reports completed from Shasta-Trinity National Forest routinely compare fish densities from various streams within the Forest and from other Pacific Northwest streams (Everest et al., 1986; Hanson 1977). Relationships are explored between densities of juvenile salmonids in the South Fork Trinity River and its tributaries and key habitat variables, specifically temperature and levels of sand and fine sediment in pools (Table 3-2). Information from Wilcox and Johnson (in press) allows some comparison between habitat conditions in the South Fork Trinity River and other Klamath basin streams.
The South Fork Trinity River basin has few undisturbed control stream to assess habitat use by juvenile salmonids. Streams that are relatively undisturbed like Miner Creek and Bear Creek have not been habitat typed. Therefore, habitat typing data from Big French Creek (Cross 1989), a Trinity River tributary flowing from the Trinity Alps, is used for comparison to check "natural" habitat preference of salmonid juveniles. Questions are raised regarding whether relative use of various habitat types should be used to identify opportunities for development of fish habitat improvement projects.
Sedimentation of Streams as a Limiting Factor
Wilcox and Johnson (in press) found average sand and fine sediment concentrations totalling 22% in the South Fork Trinity River, which was significantly higher than in the main stem of the Trinity River and its tributaries (15%) or in Klamath tributaries (17%). This conclusion is to be expected given the watershed history, its inherent instability, and known problems from past flood events (Haskins and Irizarry, 1988). Fuller (1990) used information derived from habitat typing surveys on sand and fine sediment in pools (Figure 3-1) to characterize relative decreases in fish habitat and stream productivity. Comparison of sand and fine sediments in pools of the South Fork and its tributaries can be used in a similar fashion (Figure 3-4).
Many tributaries and reaches of the main South Fork Trinity basin seem to show an inverse relationship between density of juvenile salmonids and the amount of fine sediment and sand in pools (Table 3-2). Wilcox et al. (in press) found that young of the year steelhead (0+) showed a statistically significant inverse relationship to sand and fines in all Klamath basin tributaries.
Table 3-2. Combined juvenile steelhead densities with a comparison to stream temperatures
(in degrees F) and sand and fine sediment in pools.
% Sand &
Stream Name Fish1 Fines in Max. Ave. Date of Survey
Densities Pools Temp. Temp.
Eltapom Cr. .920 7% 65° 62° 8/1-8/5/91
Big French Cr. .420 12% 62° N/A 6/26-8/9/89
Plummer Cr. .161 12% 68° 62° 7/17-8/2/90
Butter Cr. .281 18% 65° 61° 8/12-29/90
Rusch Cr. .225 29% 65° 60° 8/1-8/5/91
Big Cr.2 N/A 16% 73° 60° 8/16-23/89
Rattlesnake Cr. .067 18% 68° N/A 8/10-9/9/89
Upper Salt Cr. .129 16% 62° N/A 6/14-8/9/89
Hayfork Cr R#13 .081 27% 76° 66° 6/12-29/89
Hayfork Cr R#24 N/A 30% 75° 67° 6/12-29/89
Hayfork Cr R#35 .061 40% 73° 68° 6/12-29/89
Hayfork Cr R#46 .055 23% 78° 64° 6/12-9/3/90
E.F. of S.F. .047 37% 64° 58° 6/28-8/9/89
Dubakella Cr. .022 8% 54° 51° 9/11-18/90
Upper S.F. .010 39% 76° 67° 6/26-8/9/89
Lower S.F. .029 32% 81° 69° 8/6-8/8/90
Lower Salt Cr. .182 41% 77° 64° 5/25-6/15/89
1 All age classes of steelhead combined (fish per sq. meter)
2 Big Creek in Hayfork warm only below diversions
3 Hayfork R#1 above Hayfork Valley to the East Fork
4 Hayfork R#2 is Hayfork Valley near town of Hayfork
5 Hayfork R#3 is reach below valley and above Nine Mile Bridge
6 Hayfork R#4 is above the East Fork of Hayfork Creek
The main stem of the South Fork Trinity River has some of the highest values of fines and sand in pools of all streams surveyed and the lowest fish densities of any available for reference in the Pacific Northwest. The South Fork, upstream of Slide Creek (39%), and the East Fork of the South Fork (37%) were even higher than Grouse Creek (34%) with respect to this parameter (Figure 3-4). In contrast, the steep gradient of the lower South Fork may have helped flush some excess fine sediment (32%), ranking it just under Grouse Creek. However, fish densities in the lower South Fork are extremely low, probably influenced by high water temperatures. The density of juvenile steelhead rearing in the East Fork of the South Fork are also very low, especially considering the year around cool water temperatures (Table 3-2).
Although the Hayfork Creek watershed has more resistent rock types than the remainder of the South Fork Trinity River watershed, the highest levels of fine sediment were found in this sub-basin (Figure 3-4). Fine sediment and sand in pools of lower Salt Creek and Hayfork Creek, below Hayfork Valley, constituted 41% and 40% of the substrate, respectively. Low stream gradients in these reaches, and/or several years of prolonged drought and low flows, could partially account for somewhat higher levels of fines and sand in pools.
The significant differences in fine sediment above and below agricultural lands in both Salt Creek and Hayfork Creek, however, suggest that levels are well above background. Frink et al. (1990) suggest that bank erosion related to grazing seem to be at the root of the sediment problem in the Hayfork area. Fish densities were found to be very low in these reaches, particularly those of 1+ and 2+ steelhead. The moderate number of 0+ steelhead in lower Salt Creek and in Hayfork Creek may have reflected the last pulse of out-migrating juveniles. This suggests fish densities in late summer may have been even lower
Figure 3-4. Fine sediment and sand in pool habitats in various tributaries to or reaches of the South Fork Trinity River. NOT AVAILABLE IN ELECTRONIC FORMAT
Rusch Creek fell within the higher range of values of fines and sand in pools (Figure 3-4). Mayo (1992) noted recent sediment problems stemming possibly linked to recently altered streamside roads. Fish densities remained relatively high in Rusch but it is possible that fish populations have not yet responded to changes in channel conditions. Therefore, continued monitoring may be necessary to determine if a future drop in productivity occurs.
Big Creek, upper Salt Creek, Potato Creek, Rattlesnake Creek, upper Hayfork Creek and Madden Creek all showed moderate levels of sand and fines in pools (15-25%). The former two streams have not had major problems with sediment while the latter four are showing recovery from past sediment incursions. Fish densities are intermediate for all these streams (Table 3-2).
Streams and their fish populations outside the South Fork Trinity River watershed, but within Six Rivers National Forest, such as Willow Creek, Camp Creek and Bluff Creek, all suffered major problems from sedimentation during past floods, but are now substantially recovered. These streams also fall within this moderate range of fines and sand in pools (Figure 3-1).
Big French Creek, which flows out of the Trinity Alp Wilderness, has experienced relatively low levels of erosion and has one of the lowest concentrations of fine sediment and sand in pools (12%) (Cross, 1989). It also has one of the highest fish densities (Tables 1 and 2). Cross (1989) noted that, due to low embeddedness in Big French Creek, interstitial spaces between cobbles and boulders provided the bulk of the cover for juvenile salmonids.
For comparison, Plummer Creek, Eltapom Creek and Dubakella Creek, in the South Fork Trinity River basin, also have levels of sand and fine sediment in this lower range (less than 15%), despite some past impacts from land management activities. The former two streams have very high fish densities. Low densities of juvenile steelhead in Dubakella Creek, despite low fine sediment and cool water temperatures, do not fit the pattern of all other streams in the South Fork Trinity River basin.
Wilcox and Johnson (in press) found that C-1 and C-3 channel types in the South Fork basin had very few 0+ steelhead and almost no 1+ and 2+. These flat, unconfined channels in alluvial valleys are by far the most susceptible to long term problems with sedimentation (Frissell and Liss, 1986), although they were once the prime areas for salmonid spawning and rearing. Lack of recovery can be due in part either to continuing sedimentation or to the lingering effects of past sediment incursions (Nawa et al., 1990). Chapter 3 continued