In affected basins, several decades after torrenting, both inner gorge and upslope areas are still actively landsliding (Haskins, 1981). Up to 25% of the inner gorge zones are still in a state of active or incipient landsliding, often times for distances up to 1000 feet on either side of the channel. The prognosis for these impacted watersheds is not bright in the near future, as a significant portion of the zone remains unstable. These tributary watersheds remain vulnerable to large scale storm events which could cause severe setbacks to channel recovery. In the best of conditions, it could take several decades for channel margins to stabilize.

Finally, excessive upslope harvest levels were the third factor identified as contributing to downstream impacts in unlogged reaches of stream channel (MacCleery, 1974; Haskins, 1983). "The study appears to indicate that even if sensitive lands are left unmanaged, if activities are not dispersed on nonsensitive ground in the upper portions of watersheds, cumulative impacts can occur on the sensitive lands adjacent to stream channels in the lower watershed" ("sensitive lands," for this study, are comprised of the inner gorge and the peripheral zones described by Haskins et al., 1980).

MacCleery (1974), found there to be a direct link between the percent of a watershed heavily tractor logged and the susceptibility to stream channel damage. Without exception, if more than 45 to 50% of a watershed greater than 300 acres in size had been heavily logged prior to 1964, devastating stream channel scouring and gutting resulted from the 1964 flood. Extremely large volumes of runoff, and high peak flows, rather than logging slash were most likely the primary cause of failure in these watersheds. He concluded that "vegetation manipulation may be more important than culvert design in causing or preventing damage" (MacCleery, 1974).

Cumulative Watershed Effects: Analysis and Implementation

Study findings and observations on private and public lands on South Fork Mountain spurred the development and adoption of the cumulative effects model now employed by Shasta-Trinity National Forests, and as modified for several other National Forests. The procedure is geared towards identifying conditions likely to lead to excessive erosion and sedimentation.

"Sensitive ground" in a watershed is first identified. Sensitive lands for sediment impacts are considered to be the inner gorge areas and adjacent peripheral zones described by Kojan (1974; 1976) and Haskins, et. al. (1980). Sensitivity is related to erodibility, mass wasting hazards, slope gradient and peak streamflow characteristics. The Forest Service views this first step as critical... "defining sensitive lands, their relative hazard and appropriate management level is of supreme importance in watershed management" (Haskins, 1983). Although sensitive lands make up only a minor portion of managed ground, they were found to contain over 95% of the mass wasting which occurred during the 1964 storm and flood (Haskins, 1983).

A Threshold of Concern (TOC) is then established for the basin. This TOC is the maximum area of the basin that can be covered by roads and landings, and other areas of high runoff, at any one time. The area of a basin in roads and landings and other compacted land can be measured from aerial photos and is called the ERA (Equivalent Roaded Acres). The TOC (Threshold of Concern) represents a point where, if exceeded by the measured ERA, Forest Service geologists and land managers believe there is a significant risk of initiating cumulative watershed effects.

When the TOC is exceeded, the land manager has the option to plan for less intensive land management activities, defer the proposed activities or undertake watershed rehabilitation work which would reduce the watershed's current ERA value. For example, Six Rivers National Forest deferred timber operations in the Grouse Creek basin due to excessive ERA values brought about mostly through adjacent private land logging. Likewise, Shasta-Trinity National Forest deferred green sale plans for the upper South Fork Trinity River watershed as a result of elevated ERA values stemming from the wildfires of 1987 and 1988.

Shasta-Trinity National Forest has assigned Thresholds of Concern (TOC) values, in ERA terms, to three landscape sensitivity classes (low, moderate, high) in all sub-basins of the South Fork Trinity River watershed, based primarily on studies from the Oregon Coast Ranges (Harr et al., 1975; Harr, et. al. 1979). These are listed in Tables 5-2, 5-3 and 5-4, together with the ERA values as of 1988.

"Our posture is to assume the worst case scenario when evaluating the effects of management activities on peak streamflows... Harr et al. (1979) noted both significant increases in peak flows and channel degradation when a watershed had more than 12% of its area in roads, skid trails and landings. Since the watersheds they studied seemed highly sensitive we assigned a 14% ERA value to our most sensitive watersheds then arbitrarily assigned 16% to moderate and 18% to low sensitive watersheds" (Haskins, 1983).

The cumulative effects procedure represents a significant step forward in dispersing watershed impacts in space and time, so as to minimize adverse impacts to the basin. However, it is still likely to need refinement. Thus, in describing several South Fork Mountain watersheds impacted by the 1964 flood, Haskins (1983) reflected that although he felt the values were reasonable, "it is difficult to [determine] the appropriateness of the 14% TOC for watersheds having this level of sensitivity." Observations of watershed and stream channel condition need to be considered in re-evaluating the TOC for individual watersheds (eg., see Plate 2).

"The USFS has developed a good methodology to deal with the difficult problem of estimating cumulative watershed effects in managed watersheds." However, the method may be flawed in some situations. For example, watersheds listed in "poor condition" may still have a ERA value less than the "threshold of concern." The setting of the "threshold of concern" appears somewhat "arbitrary." In these cases, the threshold of concern must be lowered to match watershed and channel conditions. The fact that few watersheds exceed the threshold of concern "is more a function of the generally large sizes of the watersheds [selected] and the value chosen as the threshold. There are unquestionably numerous watersheds in the Trinity Basin-portion of the Forest that have suffered cumulative effects and are in degraded condition" (USDI, 1990).

There is another reason many watersheds did not exceed their threshold. Upstream private lands in a watershed were not included in the Forest Service calculations and analysis for a watershed's ERA (Veevaert, personal communication). According the Veevaert, inclusion of private lands sent many basins over their Threshold of Concern.

The National Forest views their methodology as a good tool to improve watershed conditions and treat significant existing watershed problems. We concur, and believe that continued refinements in methodology and TOC values for individual sub-watersheds will improve the results and outcome of planned land uses. Properly used, cumulative effect analysis has been found to be a useful planning tool for watershed improvement. Through limited harvesting, stands have been silviculturally "improved" and resulting funds spent to remedy a range of watershed problems.

Cumulative effect analysis has been performed for all of the larger sub-watersheds in Shasta-Trinity National Forest. Fifteen (15) subwatersheds comprising all Shasta-Trinity National Forest lands were grouped into three distinct areas with different watershed and fisheries conditions (Tables 5-2, 5-3 and 5-4):

a. Upper South Fork Trinity River

b. Hayfork Creek

c. Lower South Fork Trinity River

The Upper South Fork Trinity River watershed area was heavily impacted by a combination of logging impacts during the 1950's and 1960's, by the effects of the 1964 storm and flood, and by recent fires. Although thought to be gradually recovering, many sub-watersheds are approaching or have already exceeded their Threshold-of-Concern for cumulative effects.

Six of nine (6 of 9) watersheds, comprising 60% of this upper basin area, are highly subject to cumulative watershed effects during the next large storm (Haskins and Irizarry, 1988; USFS, 1990i). In watersheds where large portions were burned by fires, there is a significant potential for cumulative effects. These include North Yolla Bolly (in the wilderness), Middle Watershed, Blossom Cabin Creek, and Lower Watershed. Watershed degradation is likely to occur if and when a large storm occurs during the next several decades (Plate 2).

For the Hayfork Creek watershed group of basins, water quality and fish habitat problems are more related to past placer mining, irrigation, grazing and limited riparian vegetation (Haskins and Irizarry, 1988). Drafting water and poor riparian conditions in Hayfork Valley reduces summer flows and contributes to high water temperatures. Surface, stream bank and road erosion in Salt Creek, lower Hayfork Creek, Rusch Creek and Corral Creek has lead to infilling of pools and channel impaction in the lower main stem. Past placer mining in upper Hayfork Creek and the East Fork of Hayfork Creek has contributed to channel instability, broad, shallow channels, a lack of pools and poorly established riparian vegetation (Haskins and Irizarry, 1988). Grazing has contributed to channel instability in Hayfork Valley and Salt Creek, and water drafting in Tule, Salt, Big, Carr, Barker, Gulch and Hayfork Creeks has contributed to seasonal low flows and thermal warming.

In the Hayfork watershed, there are not widespread areas of sensitive lands, but local areas are steep, prone to extreme surface erosion or contain slope instabilities. Mass wasting has locally caused sediment problems, but is not widespread. The watershed can be characterized as being moderately sensitive to cumulative erosion and sedimentation effects. It is perhaps more sensitive to grazing and mining practices which directly cause channel impacts, or to agricultural and domestic water drawing practices which reduce stream flows and elevate summer stream temperatures.

In the Lower South Fork Trinity River reach, the main stem is heavily aggraded. The bulk of the 1964 sediment pulse lies in this reach, resulting in poor habitat conditions in the main channel. Riffles are shallow and wide, pool habitat is poor, riparian vegetation is lacking, and solar warming occurs (Haskins and Irizarry, 1988). Many sub-watersheds, especially those originating on South Fork Mountain, have had significant channel destabilization and serious slope stability problems. These problems continue to plague many sub-basins today and the threat of additional cumulative watershed effects remains high.

Most commonly, watersheds that are over the TOC are characterized by large blocks of selectively logged, understocked lands (Haskins, 1983). It is Forest Service policy that for watersheds already over the TOC, every new ERA of harvest proposed should be matched by rehabilitation of an equivalent or greater area of unnecessary road (Cruz, personal communication). Thus, recent timber sales in Prospect Creek in the Yolla Bolla District have included prescriptions for obliteration of old, unneeded roads.

According to Haskins and Irizarry (1988), several individual watershed areas of the South Fork Trinity River are considered to be approaching their Threshold of Concern (TOC) for cumulative watershed effects. Those watershed areas managed by Shasta-Trinity National Forest include:

1. Butter Creek

3. Rattlesnake Creek

5. East Fork South Fork Trinity River

6. Upper South Fork Trinity River

8. Hidden Valley

13. Upper Hayfork Creek

14. Hyampom, and

15. Gulch

Individual sub-basins within these watershed areas are listed in Tables 5-2, 5-3 and 5-4, and shown on Figure 5-2. Several basins managed by Six Rivers National Forest, but not reviewed by Haskins and Irizarry, are also likely to be at or over their Threshold of Concern (Veevaert, personal communication). These include Madden Creek and Grouse Creek in the lower South Fork watershed.

Figure 5-2. South Fork Trinity River sub-basin map (Haskins and Irizarry, 1988). NOT AVAILABLE IN ELECTRONIC FORMAT

Haskins and Irizarry (1988) conclude that these watershed areas need to be managed with more emphasis on preventing cumulative effects. Special management practices and mitigations should be considered. Because of the high levels of risk, proposed levels of harvest should be modified (lowered) in some basins. Special management practices are also likely to be required in these areas. Chapter 5 continued

Table of Contents