Historic changes in tributary channel conditions

Tributary channels throughout the basin have experienced wide variations in the degree of channel bed aggradation following major floods. In general, the effects of the 1964 flood, and the associated erosion and sedimentation, on the west side basins far outweighed the effects generated on the eastern upland areas of the watershed (USFS, 1990i)(Plate 2).

The degree of erosion, landsliding and channel bed aggradation which has occurred is largely a function of bedrock geology and land use history. Some watersheds, such as Pelletreau Creek and others draining erodible South Fork mountain terrain, have yielded large volumes of sediment and still show high rates of sediment transport, mobile, aggraded beds, and sediment storage in their channel systems. Within the South Fork Trinity River, debris slides are most common along oversteepened slopes of the inner gorge of the Grouse, Madden and Pelletreau Creeks (CDWR, 1979). Extensive landuse in the 1960's and damage in the 1964 flood occurred on Pelletreau, Kerlin, Mill Creek and Big Creek (Haskins and Irizarry, 1988).

A number of other South Fork tributaries show few current impacts and little evidence of historic, heavy aggradation (Haskins and Irizarry, 1988). For example, in the Hayfork Creek watershed, water quality and fish habitat problems are more related to past placer mining, irrigation, [grazing] and limited riparian vegetation than to channel sedimentation. In some of these tributary channels, fine sediments derived from road surface erosion, streambank trampling, grazing and loss of riparian vegetation has resulted in local bed aggradation and filling of pools with sediment. This has been locally identified in habitat surveys as potentially limiting rearing habitat and, thus, the recovery of the anadromous fishery.

Increased channel width caused by aggradation in tributary watersheds has been less common than filling of pools and the local deposition of fill terraces in these channels. Most tributaries are steeper than the main stem South Fork Trinity River, and have narrow, or no, floodplains adjacent to the active channel. Where widening has occurred, it has been restricted to the lower, low gradient portions of larger tributaries where they join the South Fork Trinity River. Because of the isolated occurrence of tributaries displaying significant channel widening, this is not thought to represent a limiting factor in the recovery of anadromous fisheries.

Widespread loss of riparian vegetation, especially along steep lower basin tributaries flowing off South Fork mountain (caused by debris torrenting), and along the low gradient reaches of some Hayfork Creek tributaries (caused by drought, water extraction and over grazing), has opened and exposed these channels to both warming during low flow summer periods and to erosion in the winter. In addition, logging on private lands, earlier logging on USFS lands and wildfires in 1987-88, have removed mature coniferous and/or deciduous riparian vegetation along many tributary streams throughout the South Fork Trinity River watershed. Loss of riparian vegetation has resulted in a loss of shading and a diminished input of large organic debris to tributary stream channels. These effects are locally manifested in increased summer water temperatures, reduced channel structure and complexity, fewer and smaller pools for rearing and diminished cover. Loss of riparian vegetation, and its slow recovery in many areas, is thought to be one factor limiting fisheries recovery.

Fire effects

Future erosion

The Upper South Fork Trinity River watershed has been heavily impacted by a combination of logging during the 1950's and 1960's, by recent wild fires, and by the effects of the 1964 storm and flood (Haskins and Irizarry, 1988). Although thought to be gradually recovering, many sub-watersheds are approaching or have already exceeded their Threshold-of-Concern for cumulative effects. Because of these effects, most sub-watersheds in the upper South Fork Trinity River remain sensitive to potential cumulative effects, especially in areas burned within the last decade (Haskins and Irizarry, 1988; USFS, 1990i)(Plate 2).

As yet, the 1987-1988 fires have had little apparent effect on the main stem of the South Fork Trinity River (USFS, 1990i). Future accelerated erosion and sediment yield from areas burned in the 1987 and 1988 wildfires on the Hayfork and Yolla Bolla districts could still be severe (Haskins, 1992, personal communication). With a large storm, water quality and fish habitat degradation are likely to occur in the intensively burned watersheds, and these effects are expected to be transmitted to the South Fork Trinity River (USFS, 1990I).

The severity of future erosion and sedimentation, and potential effect on downstream fishery habitat, will be dependent upon the magnitude and timing of the next storm. Fortunately, there have been no significant ("10 year recurrence interval) storms or floods to have affected the basin since the fires occurred (USFS, 1990a). Substantial erosion control and revegetation efforts have been undertaken to limit the immediate effects of the fires. The potential for significant erosional impacts are expected to gradually lessen as the recovery and revegetation period lengthens.

Surface erosion from areas burned in 1987 and 1988 is no longer likely to represent a significant future source of sediment to the stream system, or to act as a limiting factor to fisheries restoration and recovery. Grass seeding and the invasion of native vegetation over the last 5 years has provided good ground cover over much of the burned areas. Surface erosion, if it is to be significant in the future, will be localized on developing landslide areas and harsh sites with poor growing conditions.

Fluvial erosion (erosion from swales, stream channels and gullies) could still be an important source of future sediment from the burned areas. When the areas burned, not only did most of the standing vegetation (trees and brush) die, much of the downed trees and logs on the hillslopes were also consumed. This has removed valuable, stabilizing structure from low order swales and stream channels. Gullying and channel enlargement has occurred in some of the larger tributary streams which were burned, and heavy rainfall and runoff would likely trigger widespread fluvial sediment production from other affected channels. Accelerated fluvial erosion and sediment yield from stream channels and gullies in the burned areas could provide relatively large quantities of sediment to the South Fork Trinity River and its major tributaries, thereby further delaying and limiting recovery of the anadromous fishery.

Perhaps the greatest threat to fisheries resources stemming from the 1987 and 1988 burns comes from future slope stability and mass erosion problems. During the burns, there was complete loss of riparian and inner gorge coniferous vegetation along many smaller tributaries (Haskins, 1992, personal communication). Over the 15 to 20 year period following the fires, stabilizing root systems of conifer forests on north facing slopes will decay and slopes that once displayed marginal stability have now become much more susceptible to failure. On drier south facing slopes, hardwoods that burned have since resprouted and failures due to root decay may be less common.

The loss of vegetation has altered the subsurface hydrology of the steep streamside slopes and increased the likelihood of slope failure. Slopes are now likely to be wetter for longer periods during the year than when a thick vegetative cover was intercepting rainfall and transpiring moisture from the soil. USFS geologists have identified several, large landslides (total volume of 300,000 yds3) developing in the lower Blossom Creek drainage, apparently as a result of the loss of root strength in an area of unstable geologic materials, following the 1988 Hermit Fire (Haskins, 1992, personal communication).

Widespread failure of steep, streamside slopes could occur within recently burned areas if a large magnitude storm and flood were to occur within the next two decades (USFS, 1990i; Haskins, personal communication). Such an event could cause serious downstream sedimentation in affected tributaries and in the main stem South Fork Trinity River, and delay or limit the recovery of the anadromous fishery. The longer the recovery and revegetation period before the next flood, the less likely such a rainfall event will trigger serious slope instability and sediment delivery to the South Fork Trinity River.

In watersheds where large areas were burned by fires, there is a significant potential for long term cumulative effects (USFS, 1990i). These include North Yolla Bolly (in wilderness), Middle Watershed, Blossom Cabin Creek, and Lower Watershed (USFS, 1990i). Six-of-nine (6 of 9) watersheds, comprising 60% of the Penny Ridge burn area, are thought to be highly subject to cumulative watershed effects during the next large storm (USFS, 1990i).

Fire rehabilitation efforts

Sedimentation, from any source, will continue to be detrimental to the protection and recovery of South Fork Trinity River fisheries. Harr (1992), in an analysis of possible erosion from fire salvage logging in the South Fork Trinity River, stated "a catastrophic influx of sediment might destroy the populations quickly, but a more gradual influx of sediment beyond what the river can transport might result in a steady decline of fish populations."

Past efforts to control and prevent post-fire erosion in both the Hayfork and Yolla Bolla districts have apparently met with some success (USFS, 1990a). The reported success of the program is a combined result of proactive erosion prevention actions, undertaken in the years following the fires, and the absence of significant, large magnitude storms and floods since the fires occurred.

In spite of the preventive efforts undertaken on the burn areas, it is expected that surface erosion and downstream sedimentation would have still been a significant and serious problem if a large storm had occurred within the first 2 to 3 years following the fires. Itis also expected that serious future effects, enough to damage fish habitat and limit the recovery of anadromous salmonid populations, can still occur as a result of large storms occurring in the next 10 to 15 year period. Most of these effects will come from channel enlargement, gullying and slope failures in intensely burned watersheds (Haskins, 1992, personal communication).

Following the fires, emergency aerial and broadcast seeding was performed to reduce surface erosion and disperse overland flow. Revegetation efforts were largely successful and first year cover ranged from 10 to 90 percent (USFS, 1990a). This work has successfully limited the future volume of sediment that is likely to be derived from surface erosion processes on most burn sites.

Channel stabilization was undertaken in literally hundreds of locations using a variety of straw bale dams, log check dams, single fence check dams and head cut controls to prevent channel erosion during the first several years following the burns (USFS, 1990a). Structures were installed to prevent downcutting, trap and store eroded sediment, to establish grade control, to slow stream velocities and, ultimately, to prevent sediment from entering and impacting spawning gravels of the South Fork and its tributaries.

Structures installed in the channels of most ephemeral streams were largely non-functional because they have not carried significant runoff in the last four years. In other channels, structures quickly filled and were overtopped with sediment. Further contributions from upstream slopes and channels were then transported through the channel system, over the structures and into downstream areas (Ranken and Haskins, 1991).

Many straw bale and log structures observed in the field are in a state of disrepair. Straw bale structures were installed as temporary measures, and most have partially failed or completely decomposed and disappeared. Most of the sediment trapped behind these structures remains today, partially stabilized by grasses and shrubs (Ranken and Haskins, 1991). This sediment is still highly subject to erosion by future storm runoff, when it occurs.

Most channel and gully control structures observed in the field were placed too far apart in the channel systems to provide adequate grade control. Once structures were filled with sediment, they no longer served to store and prevent sediment from being transported to the South Fork Trinity River and its tributaries.

Finally, structures which are in some state of disrepair can end up causing additional erosion by deflecting streamflow in channel banks. It is important to place structures close enough together, to clean out (or stabilize) sediment deposited in sediment storage structures, and to maintain the longer lasting log structures until they are no longer needed. Installation of check dams requires a long term commitment to structural maintenance to ensure proper functioning until the root systems of planted trees and woody vegetation can provide a sufficiently stabilizing effect against future storm flows.

Some burned riparian zones were planted with willow and riparian vegetation as an Emergency Burn Area Rehabilitation (EBAR) measure following the 1988 Hermit fire. It has been estimated that it can take as long as 3 years for natural riparian vegetation to return in hotly burned streamside zones (USFS, 1990a).

As a rule, conifer planting in burned areas is not an EBAR treatment that is performed the first year following a burn. Conifer and hardwood planting is typically part of salvage logging operations and follows removal of merchantable trees. This delay in planting has been as long as four years on some burn areas, depending on preparation, approval and litigation associated with the salvage EIS. In areas that are not salvage logged, conifers might not be replanted. Burned slopes within the Wild and Scenic River corridor are not replanted (USFS, 1990i). Chapter 5

Table of Contents