Impact of Foliage Loss 



Removal of slope vegetation results in a temporary but signifi- 

 cant loss of foliage available for interception and transpiration of 

 water. This in turn leads to wetter conditions and higher 

 piezometric levels in a slope. Results of the Pine Creek study 

 support this conclusion (figs. 22 and 23) as do results of other 

 hydrologic investigations on effects of timber removal (Gray and 

 Brenner 1970; Bethlahmy 1962; Brenner 1973). The impact of 

 vegetation removal on soil moisture changes in a slope appears 

 to be most critical the first year after cutting. Studies by Hallin 

 (1 967) showed that, after 3 years, low vegetation that invades a 

 cutover site is nearly as effective as old-growth timber in deplet- 

 ing moisture. 



The results of soil water piezometry studies in Watersheds 

 Nos. 1 and 4 indicate that removal of vegetation by clearcut 

 logging can increase piezometric levels as much as 100 per- 

 cent. Critical piezometric levels shown in table 6, that is, tPie 

 minimum head required to cause slope failure assuming no soil 

 cohesion, were frequently exceeded. The occurrence of slides 

 in the slope above the road cut in both watersheds (figs. 1 0-13) 

 and in other watersheds in the vicinity (table 2) reflects the low 

 margin of safety under high piezometric conditions. On the 

 other hand, the absence of massive and pervasive slope fail- 

 ures suggests that some residual cohesion must be present. 

 Required cohesions for local stability in Watershed No. 4 for 

 different piezometric heads are also shown in table 6. These 

 cohesions could be provided in whole or in part by root rein- 

 forcement. 



Impact of Root Decay 



The importance of cohesion on stability was clearly estab- 

 lished in the preceding sensitivity analyses. The root rein- 

 forcement model coupled with root tensile strength and root 

 distribution data show that live roots can provide a large fraction 

 of the total or apparent cohesion present in granitic soils in the 

 batholith. Conversely, studies of root strength loss with time 

 after cutting (Burroughs and Thomas 1977) and landslide fre- 

 quency with time after cutting (Megahan and others 1978) 

 indicate progressive loss of root cohesion following clear- 

 cutting. Megahan's data suggest that landslides are most fre- 

 quent 4 to 1 years after logging. These data are consistent with 

 findings of other investigators (for example, Bishop and 

 Stevens 1964; Swanston and Walkotten 1970). The time of 

 minimum stability represents a crossover point between the 

 growth and decay curves of root systems of slope vegetation. 

 Root strength decline after tree felling is undoubtedly both 

 species and site dependent (Burroughs and Thomas 1977). 

 The timing or occurrence of slope failures is thus dependent on 

 the amount of residual stand on the slope and the rate of 

 establishment of new vegetation relative to the root strength 

 decline of previously cut trees (Kitamura and Namba 1966). 



Loss of Buttressing and 

 Soil Arching Action 



Analysis of spacing relationships and rooting morphology of 

 trees in forested slopes of the Idaho batholith indicate the soil 

 arching between trees may play an important role in restraining 

 soil movement. Several examples of buttressing action by 

 embedded tr^e trunks and root systems were observed (figs. 4 



and 5). Gonsior and Gardner (1 971 ) reported similar examples 

 in their analyses of slope failures in the Idaho batholith. They 

 recommended, in fact, that barriers of live trees should remain 

 undisturbed immediately below the toe of fill slopes and above 

 the cut slope. 



Removal of all large diameter stems by clearcutting, of 

 course, gradually eliminates any soil arching restraint or soil 

 arching action. The stumps will temporarily provide restraint, 

 but when the roots rot and decay these anchor points or "arch 

 abutments" will actually become zones of weakness in the 

 slope. This will occur because as roots rot and disappear voids 

 with no shear strength will be left behind, or infilling with weak 

 colloids may occur in the old root channels. In addition, former 

 root channels may provide entry points for water and thus 

 facilitate rapid buildup of pore pressures. 



Loss of Surcharge 



The sensitivity analysis reported showed that decreasing the 

 vertical surcharge (qo) by removing slope vegetation has a 

 beneficial influence on stability, but only a slight one. Under 

 certain conditions, surcharge can actually enhance stability. 

 Ward (1976) showed that this occurs under the following 

 circumstances; 



(Cs + Cr)<7w Hw tan 4) cos^p (10) 



This relationship shows that surcharge is beneficial for low 

 cohesion values, high piezometric levels, and relatively gentle 

 slopes. Assuming the worst case of maximum rise in 

 piezometric surface (Hw = H) and substituting the median 

 values in table 6 for the variables on the right-hand side of 

 equation (10), a limiting total cohesion of 0.57 Ib/in^ (3.9 kPa) 

 results. This cohesion is quite possible as an upper, limiting 

 value in many granitic slopes. In such slopes, surcharge from 

 the weight of trees would have at best a beneficial influence and 

 at worst a negligible effect as critical, saturated conditions de- 

 velop in the soil. 



MANAGEMENT IMPLICATIONS 



Measures to Minimize 

 Mass Erosion Hazard 



LOCATION mo SIZE 

 OF CLEARCUT AREAS 



The preceding analyses and findings indicate that many 

 slopes in the Idaho batholith are in a state of marginal or 

 metastable equilibrium. Such slopes are vulnerable to both 

 surficial and mass erosion when vegetation is removed by 

 clearcutting or by wildfire. In many instances, road construction 

 associated with timber harvesting appears to have a greater 

 impact than vegetation removal alone (fig. 3). On the other 

 hand, both may have synergistic and cumulative impacts on 

 stability that are hard to distinguish and separate. The slope 

 failures observed in the slope above the road cut in Watershed 

 No. 1 (figs. 1 and 1 1 ) are a case in point. The failures appear to 

 be associated with the road cut, but may have been caused in 

 part by wetter conditions in the slope above and by loss of some 

 root cohesion as a result of vegetation removal. 



It is not possible at this point to formulate precise rules for 

 location and size of ciearcuts to minimize mass erosion 



20 



