Watershed improvements and erosion control activities on the Yolla Bolla and Hayfork Districts began in earnest in 1988. Virtually all the earlier activities were in response to major wildland forest fires which occurred in 1987 and 1988. During 1988 and 1989, watershed improvements consisted primarily of labor intensive treatments aimed at controlling stream channel bed and bank erosion, reducing the risk of surface erosion on denuded hillslopes, and reestablishing older serial stage vegetation on the hillslope. Major activities included grass seeding, tree planting, construction of log and straw bale check dams, falling dead trees into stream channels and removing some woody debris.
Beginning in 1990, the thrust of watershed improvements shifted to the hillslope, and have largely focused either on fluvial and surface erosion problems at culverts, road cutbanks and on the road prism itself, or on stream channel restoration activities including stabilization of small streamside landslides and channel stored sediments, armoring channel knick-points and removing debris and stored sediments. Most of the work activities are very labor intensive. Erosion control measures intended to address or prevent landslide processes from occurring along roads and landings are noticeably absent from lists of annual accomplishments.
Treatments outlined in watershed assessment reports have been performed on only portions of a few areas which have been inventoried. Implementation funding for watershed work is requested through internal USFS sources such as programmed (appropriated) dollars, KV and WIN budget sources, and through grants received from the Trinity River Fish and Wildlife Restoration Program. KV monies are presently available at reduced levels as a result of recent reductions in timber harvesting on all the Districts.
The rate of implementation of identified watershed restoration and erosion prevention work has been slow, largely due to delays in finalizing inventory reports, hesitancy to request appropriate levels of funding to implement projects, a low rate of implementation proposal approval and funding from the TRFWRP and the USFS, and to a lesser degree, to very limited watershed staffing on the districts. As a result, sites which have been identified as potential sources of erosion which could impact spawning and rearing habitat are not being treated. This, in concert with the potential end for grant funding from the Trinity River Restoration Program, which is due to expire in 1995, may limit or delay salmonid recovery in tributaries that have not been treated.
1. On the Yolla Bolla District, work has been conducted on roads and selected stream channels in Gemmill Gulch, Hall City Creek, Goods Creek, Headwaters of Hayfork Creek and the South Fork, Prospect Creek, Dark Canyon Creek and Blossom Creek. Excluding fire rehabilitation measures, approximately $175,000 were expended in 1991 and 1992 on the Yolla Bolla District implementing identified erosion control work. The use of heavy equipment has been mainly limited to dump trucks, backhoes and graders to rock and grade roads, transport and supply rip-rap sized rock and materials for culvert inlet and outlet repair, and clean and armor inboard ditches. To date, only 0.75 mile of road has been obliterated (4 culverts removed and the road ripped), and a few landings have been ripped.
The work conducted at stream crossings and along roads is beneficial and important to protect fisheries resources, since erosion at these locations result in high delivery ratios of particle sizes (less than 2mm) which are particularly detrimental to fish. However, we believe the actual cumulative volumes of sediment prevented from entering stream channels is much less than projected by USFS personnel. Also, the lack of effort to identify and correct potentially large volumes of sediment expected to originate from mass movement processes in the South Fork Trinity River (excluding Hayfork Creek) may be a major shortcoming of the watershed rehabilitation program.
Watershed inventory reports for the STNF indicate a large amount of proposed stream channel rehabilitation projects are planned throughout basins in the South Fork. These generally fall into two broad categories: 1) structures to protect streambanks, and 2) structure to increase the quantity of large woody debris (LWD) in stream channels. The USFS feels justified in adding LWD structures to tributary stream channels in the South Fork in order to create better rearing conditions for steelhead and to improve fish habitat (Dick Irizarry, personal communication). Chapter III discusses pros and cons of habitat manipulation to improve fisheries.
In terms of structures to protect streambanks, it has been our experience that bare streambanks in forested and non-alluvial settings (ie. bedrock reaches of stream) are often perceived as active, on-going sediment sources, when in reality they may not be contributing sizeable quantities of additional sediment (which can be stabilized) to streams. In these situations, streamside landsliding and active bank erosion caused by concentrated stream flow frequently contributes much larger quantities of sediment to stream channels than does the subsequent surface erosion on the bare banks. The exception is in alluvial valleys and agricultural settings where bare streambanks are composed of poorly consolidated materials which can continue to erode at high rates once the vegetation is removed.
Revegetation of eroding streambanks and streamside landslides are very appropriate treatments and should be a major emphasis of riparian management practices in both forested and agricultural settings (Platts, 1990; Beschta, et. al., 1991; Mike Furniss, personal communication). However, the need for bank protection structures as a prerequisite to revegetation efforts should be carefully evaluated by professionals prior to implementation. These types of structures are typically expensive to install, and the overall effectiveness at controlling or arresting erosion is generally low, particularly in forested settings. In fact, structures which harden streambanks frequently cause erosion at some other downstream location. In alluvial valley settings, rather than installing a rock or log bank protection structure,it may be a sounder approach to physically lessen the streambank angle (ie. slope steepness) so that revegetation efforts will be successful.
Beschta, et. al. (1991), state that because of their frequent negative effects, structural alterations to stream channels (particularly hard structures which prevent channel adjustments) should be eliminated or seldom considered as a fish improvement strategy. The authors go on to state that the use of hard structures (e.g. low rock check dams, boulder clusters, rock riprap, rock berms and log weirs) has no scientific validation, nor is their any consensus amongst biologists regarding when structures should be added to a degraded riverine or riparian system, whether that be a forested or agricultural setting. The authors conclude that, in most instances, the recovery of streamside vegetation was the most efficient mechanism for improving fish habitat; and that the need for adding human-made structures was seldom justified.
We believe that most efforts to treat individual sites of eroding or unstable streambanks, or where the stream bed is very mobile, represent attempts to treat symptoms of much larger basin-scale problems. Prior to implementing any in-stream stabilization measures, the basin must be 1) evaluated for all sources of on-going and potential sediment production, 2) needed corrective and preventative erosion control activities on the roads and hillslopes must be underway, and 3) it can be demonstrated that in-stream conditions are in some state of recovery. In basins where channel conditions are degraded or where there is a high risk of additional sediment production from upland areas, in-stream structures intended to protect eroding banks and stabilize stored sediments should not be undertaken.
2. The Hayfork District has demonstrated a higher level of concern to mass movement by proposing a more extensive road obliteration program, and by proposing to treat several potential landslide sites. Based on a review of five watershed assessment reports (Rattlesnake, Tule, Lower Hayfork, Middle South Fork and Upper South Fork area), the Hayfork District has proposed obliterating 30 miles of road, closing 79 miles of road and performing erosion control work at 118 individual sites along roads to be retained. To date, only limited road obliteration and road closure projects in the Rattlesnake sub-watershed have been undertaken.
The Hayfork District has also proposed over 220 individual, labor intensive in-stream projects along selected stream channels in three of the basins inventoried. A majority of the projects are directed toward reestablishing vegetation, However, the remaining projects include a variety of in-stream structures intended to stabilize channel stored sediments and prevent additional bank erosion. As discussed earlier, these latter treatments will generally have a low likelihood of either being cost-effective or effective if upland "erosion proofing" efforts are not underway. To date, only in the Rattlesnake Creek basin has any in-stream work been performed as a result of USFS watershed assessments. It should be pointed out that the Rattlesnake Creek in-stream work is in addition to the extensive amount of in-stream habitat improvement projects constructed by the CCC's during the period 1986 to 1992 under the Model Steelhead Stream Demonstration Project Management Plan (MSSDP, see Chapter X).
3. Six Rivers National Forest has performed limited road removal activities in Grouse Creek, and none in other lowest South Fork tributaries. On both Six Rivers and Shasta-Trinity, there is a tendency among forest managers, engineering and forestry staff to oppose the formal decommission or obliteration of roads. This bias is still widespread even in light of the realization that forest road maintenance budgets are shrinking annually, and many tens to hundreds of miles of roads receive minimal to no preventative maintenance. These factors make it very difficult for watershed staff to integrate road removal as a treatment option for watershed restoration and fisheries improvement projects.
Six Rivers has worked with at least one private timberland owner to upgrade and reconstruct several logging roads in portions of the Grouse Creek basin. Work was funded by the landowner, even though public funding would have been available. There is an obvious reluctance of private land owners to allow watershed inventory work to be completed on their holdings, or to accept public funding for completion of recommended work if and where inventories have been completed. This general lack of interest and cooperation will have the effect of delaying or limiting the recovery of salmonid fisheries in erodible or unstable basins containing substantial private ownership.
As a rule, private timberland owners will not consider road obliteration or proactive road abandonment as viable rehabilitation treatments in impacted watersheds, nor will they voluntarily consider implementing watershed protection measures which limit their ability to harvest timber. These positions may, in some cases, be incompatible with the recovery of anadromous fishery resources in erodible and unstable watersheds.
Road upgrading is one treatment strategy that is receiving attention in discussions of new riparian management strategies for the four northern California forests (USFS, 1992a). Road upgrading would consist of the replacement of undersized culverts with structures that are designed to pass large storm flows (100 yr Tr, or greater) without failing. USFS roads typically have drainage structures with higher capacity than comparable crossings on private forest lands.
Lack of road upgrading, and the ubiquitous presence of undersized drainage structures on maintained and abandoned roads, are serious problems on private lands regulated by the California Forest Practices Act and Rules. Undersized culverts, deteriorating "humboldt" log crossings and potential fill failures on abandoned logging roads are a potentially serious factor that is likely to limit or delay the recovery of anadromous salmonid populations in some watersheds, especially those containing large holdings of private land which was roaded in the 1950's, 1960's or 1970's.
The recently released Forest Ecosystem Management Assessment Team (FEMAT) Report not only defined "key watersheds", but also directed federal agencies to perform Watershed Analysis in all key watersheds and selected adaptive management areas. The FEMAT report recommended that watershed restoration strategies be comprehensive, addressing both 1) watershed protection in the best habitats that remain (refugia) and 2) restoration of degraded habitats in an integrated program that moves ecosystems toward recovery and resilience. Watershed protection in the refugia sub-basins should involve: 1) reduction in management levels, particularly timber harvesting, 2) reduction in road densities through the "decommissioning or obliteration" of high sediment production risk roads and non-essential roads, and 3) upgrading roads which are needed for long term resource management needs.
Upland Watershed Assessment activities conducted in the South Fork Trinity River since 1988 give Shasta-Trinity and Six Rivers National Forests a leg up on many other Forests throughout the Pacific Northwest in implementing the intent of the FEMAT report. Forest personnel should be able to rapidly determine refugia sub-basins, fill in needed mass movement assessments, determine which roads are part of the "core road network" necessary for long-term management, and begin implementing watershed restoration activities for the purposes of fisheries recovery and Trinity County economic development.