Content uploaded by Robert Ziemer
Author content
All content in this area was uploaded by Robert Ziemer
Content may be subject to copyright.
343
Erosion and Sediment
Transport in
Pacific Rim
Steeplands. I.A.H.S.
Publ.
No. 132 (Christchurch, 1981)
Roots and the stability of forested slopes
R.R. Ziemer,
Forest Service, U.S. Dept. of Agric.,
1700
Bayview
Drive,
Arcata,#
California,
95521,
U.S.A.
Abstract.
Root
decay after timber cutting can lead to slope failure.
In
situ measurements of soil with tree roots showed that soil strength
increased linearly as root biomass increased.
Forests clear-felled
3 years earlier contained about one-third of the root biomass of
old-
growth forests.
Nearly all of the roots
<
2 mm in diameter were gone
from 7-year-old logged areas while about 30 percent of the
<
17 mm
fraction was found. Extensive brushfields occupied areas logged 12 to
24 years earlier. The biomass of brushfield roots
<
2 mm in diameter
was 80 percent of that in the uncut forest,
and fewer large roots were
found there than in the forest.
Roots
<
17 mm in diameter in the
brushfield accounted for 30 percent of that found in the forest, and
for total root biomass, only 10 percent.
Individual,
live brush roots
were twice as strong as conifer roots of the same size.
This difference
may partially compensate for reduced root biomass in brushfields.
Net
strength of the soil-root matrix in brushfields was about 70 percent
of that in uncut forests.
If soils are barely stable with a forest
cover,
the loss of root strength following clear-felling can seriously
affect slope stability.
Racines
et la
stabilite
des pentes
bois6es
R6sum6.
La
Pourriture
des racines
apr6s
coupe peut aboutir
ZL
des
glisse-
ments de terrain.
La
mesure
faite in situ du sol contenant des racines
344
d'arbre
a
ri?&ii!
que la
rhistance
du sol a augment&
d'une
facon
lin&aire
'a
mesure
que la biomasse racinaire a augment&.
Les for&s
exploiti?es
trois ans avant contenaient environ
les
deux tiers de la biomasse
racinaire qui se trouvait dans
le
vieux bois. Presque toutes les racines
'a
<
2
mm
de
diarn'etre
avaient disparu dans les terrains exploites sept
am
avant
taudisque
2
peu
prk
30 pour-cent de la fraction de
<
17 mm
s'y
trouvaient.
De
vastes surfaces couvertes de buissons remplissaient
les zones exploit&es de 12
'a
24 ans avant.
La biomasse des racines de
buisson
h
<
2 mm de
diam&tre
6tait
80 pour-cent de
celle
de la
forh
non-exploit&e. Et moins de grandes racines s'y trouvaient que dans la
forst.
Les
racines
h
<
17 mm de
diamhre
dans les terrains buissonneux
comprenaient 30 pour-cent de la biomasse
trouvge
dans la
fo&t,
et de la
biomasse racinaire
totale,
seulement 10 pour-cent.
Les
racines de
buisson vivantes individuelles
gtaient
deux fois plus
r&istantes
que
celles
des
conifhes
de la
m&e
taille.
Cette
diffgrence
peut en
partie
compenser pour la
rgduction
de la biomasse racinaire dans les terrains
buissonneux.
La rhistance nette de la
matrice
sol-racine dans les
terrains buissonneux
&tait
environ 70 pour-cent de celle des
forgts
non-
exploit&es.
Si le sol est
'a
peine stable avant l'exploitation forestiere,
la perte de
rhistance
racinaire suivant l'exploitation peut produire
des effets graves sur la
stabilit6
d'une
pente.
INTRODUCTION
Pacific rim steeplands contain some of the world's most unstable landscape
together with much of its most productive forests.
Little is known
about the interaction of timber harvesting and mass erosion.
Slides
occur more often on slopes where the forest has been removed than where
it remains.
Root systems of plants can increase stability of forested
slopes by anchoring through the soil mass into fractures in bedrock, by
crossing zones of weakness to more stable soil, and by providing interlocking
long fiberous binders within the weak soil mass. In deep soil, the
vertical anchoring effect of roots becomes negligible, and the other two
conditions predominate. After a forest is removed by fire or harvest,
the root system decays and the soil progressively weakens.
If the
forest slope is marginally stable,landslide frequency often increases
after trees are removed.
As deforested areas revegetate, the soil
mantle is again progressively reinforced as new roots occupy the soil.
IN SITU SHEAR STRENGTH OF FOREST SOILS
The strength of forest soils is difficult to measure directly,
Standard
346
the shear plane.
Endo
and Tsuruta found that even roots 1 to 2 mm in
diameter pulled from the soil blocks. How much roots contribute to the
strength of soil can be underestimated if they are pulled from the soil
block rather than break. Consequently,
I have developed a shear box
that shears a soil block along two parallel vertical planes.
The shear
box encloses the soil block, but remains open on two sides. The inside
dimensions of the shear box are 60 cm wide, 30 cm long, and
30
cm high.
Each open side is 30 cm square.
The box was field tested in the relatively simple soil-root system
of a mature
Pinus
contorta
stand growing on coastal sands in northern
--
California.
At each test site,
the front of the soil block to be
sheared was carefully excavated. A sharpened steel plate was then
pressed into the soil under the block, and the back of the soil block
was excavated. Steel plates covering the top, bottom, front and back of
the soil block were then bolted together. Stress was applied to the box
by a mechanical jack extended at a rate of 1.27 cm/min for about 7
minutes.
Stress was measured by using a proving ring mounted between
the jack and the shear box. The proving ring reading was recorded every
few seconds,
and force over time was plotted. The maximum force was
considered to represent the force required to shear the soil block. At
three sites,
the roots and soil within the shear plane were cut with a
knife before the test,and the force required to move the box and soil
block was measured.
This
force was considered to be the sliding friction.
The shear strength of the soil block was calculated as the maximum force
less the mean sliding friction.
The soil block was then carefully dissected, and detailed information
on the position of stones and roots in and adjacent to the two side
shear zones was recorded. Roots which are cut by the shear box, but
which extend undisturbed in one direction,
would
provide
strength to the
soil block similar to that reported by
Endo
and Tsuruta (1969) and
O'Loughlin
(1972).
These roots may pull out of the soil or they may
break.
However,
roots lying horizontally and aligned about normal to the
shear planes can continue undisturbed through the soil block and into
the soil on either side of the shear box.
These roots could not pull
out of the soil block without breaking.
Soil samples were examined for particle size, bulk density, aggregate
stability,
and soil moisture.
After the soil block was sieved, all roots
were extracted, washed, separated into live and dead fractions, subdivided
into
six size
classes,
dried at
70°C,
and weighed. From each test
location, data were collected for 24 soil and root variables.
Several
classes were combined, adding an additional eight variables.
Endo
and Tsuruta (1969) measured in situ the shear strength of
--
cultivated nursery soil densely planted with Alnus glutinosa saplings
which had an average stem diameter of
16 mm.
They reported 53 per cent
of the variation in soil "cohesion" was explained by the fresh weight of
roots.
O'Loughlin
(1972) evaluated the influence
of six soil variables and
one root variable on the in situ shear strength of glacial till subsoils
--
in a mixed old-growth
Pseudotsuga
menziesii, Thuja
plicata,
and Tsuga
heterophylla forest.
The fresh weight of roots was the most significant
of the variables,
accounting for 56 per cent of the variation in
soil
strength,
The
fresh weightof roots in the soil samples averaged 2.51
kg/m3
--
about
one-third of
that reported by
Endo
and Tsuruta.
Most forest biomass studies report oven-dry weight rather than
fresh weight.
I found the dry weight of roots predicted soil. shear
strength about as
well
as the fresh weight of roots.
348
To understand which variables are most useful in predicting the
force required to shear the soil block, I screened the 32 variables by
using all possible subsets regression and partial F-tests. The "best"
regression equation
contained
only a single variable -- the dry weight
of live roots
<
17
mm
in diameter (Fig. 1):
Strength
= 3.13 + 3.31 Biomass,
in which strength is in
kPa
and biomass in
kg/m'.
The equation explains
79 per cent of the variation in strength
(r*
= 0.79, n = 18) and is
statistically significant at the 1 per cent level
(Fl,l6
= 56.8).
The
mean biomass of the
<
17 mm live roots was 1.77
kg/m3,
which represents
64 per cent of the total root biomass.
Adding more variables did not
significantly improve the regression equation.
The squared correlation coefficient between the force required to
shear the soil block and the dry weight of live roots in the soil block
helps identify the ability of different root size-class groupings to
predict soil strength (Table
1)
.
The best relationship was between soil
strength and the biomass of live roots
<
17 mm in diameter.
Since
Endo
and Tsuruta (1969) and
O'Loughlin
(1972) used the total fresh root
weight,
the explained variance of their equations might also have been
improved if other size classes of roots and a distinction between live
and dead roots had been used.
In old-growth forests,
I
found many more horizontal roots than
vertical roots,
My equation probably underestimates the true
soil-
strengthening effect by roots because roots oriented vertically and
parallel to the axis of movement of the shear box were cut by the front,
back, and bottom plates of the box. I did not measure the orientation
of the roots,
but
would estimate that the root effect is underestimated
by at least one-half.
349
ROOT BIOMASS
The studies that have directly measured the in situ strength of soil
--
which contains tree roots have all identified the weight of roots in the
sample as the principal variable related to measured strength.
O'Loughlin
(1972) calculated that the root network accounted for 71 per cent of
shear strength at saturation of the till soils he studied on slopes of
35 degrees.
Bjorkhem et al. (1975) observed that the imposed load may
be 70 per cent greater before soil-rupture in soils with a root network
than in soils without roots.
Slope stability problems could easily develop as the tree root
system decays after timber cutting on steep slopes.
Little is known,
however,
about changes in root biomass after timber harvest and subsequent
revegetation.
To study these changes in root biomass, an area was
selected at 1300 m elevation in the Klamath Mountains of northwestern
California.
The soil is a gravely fine sandy loam derived from deeply
weathered diorite. The vegetative community is the white fir phase of
the mixed conifer forest.
Abies
concolor
dominates the old-growth
--
forest, but variable combinations of Pseudotsuga menziesii,
Pinus
ponderosa,
Pinus
lambertiana,and Libocedrus decurrens are found in the
coniferous component and Arbutus menziesii and
Castanopsis
chrysophylla
in the hardwood component.
When the forest is opened by fire or logging,
extensive brushfields,
composed principally of Ceanothus velutinus,
commonly occupy the openings for several decades. Even with site
preparation and conifer planting,
the brush has not been controlled
successfully
in the study area.
Roots were extracted from 400 soil cores collected from areas that
had beenclear-felled up to 24 years earlier. Each core contained about
350
3200 cc of soil. The variances within and between areas were so large
that no trends in root biomass with time after logging could be detected.
The two most obvious ways to reduce the variance are to increase either
the volume of individual samples or the number of samples.
The number
of samples required to determine a trend in biomass related to time
after logging was calculated to be in the tens of thousands. And so the
size of each sample was increased about 400 times to a volume of 1
l/3
m3.
A block of soil 1
m
square was excavated to a depth of
1 1/3 m in
1/13
1/3-m increments.
The
soil was screened and roots were extracted,
separated into live and dead components, washed, sorted into six size
classes, dried at
7O*C,
and weighed.
Only a few roots
were
found below
a depth of 1 m. Roots were taken from 103 soil blocks representing six
different ages of cutblocks up to 24 years and an uncut old-growth
stand.
For convenience,
the following discussion will refer to changes
over time,
even though data came from different areas.
This is probably
permissible because care was taken to select areas physically close to
one
another which had the same soil type and depth, slope, aspect,
elevation,
and forest density.
Within each of the seven study areas,
about six soil blocks were clustered around each of about three randomly
selected l-m by 2-m access pits.
The mean root biomass for each of the
size classes was calculated by using cluster sampling formulae (Fig. 2).
The
3-year-old
cutblock
supported an extensive cover of bracken
fern (Pteridium aquilinum). Through decay, root biomass decreased to
about two-thirds of that found in the uncut forest. The proportion of
the roots which had decayed decreases as the size of the roots increases.
Live Pteridium roots (rhizomes) returned to the
cutblock
about 10 per
cent of the live root biomass found in the uncut forest for
<
17-mm
roots.
351
Within 7 years after logging,
Pteridium had been replaced by widely
scattered brush and herbaceous species.
The live root biomass dropped
to about 3 per cent of that in the uncut forest.
The dead root component
continued to decline as decay progressed.
Essentially all of the
<
2-mm
roots had decayed within 7 years;
about 30 per cent of the
<
17-mm
dead
roots remained.
Within 12 years after logging,the cutblocks had revegetated with
Ceanothus velutinus rather than conifer regeneration, The brushfields
occupied all cutblocks that had been logged
12
to 24 years earlier. The
biomass of small live roots increased dramatically in these brushfields.
Within
12
years after cutting,
live roots
<
2 mm in diameter had recovered
to 82 per cent of that in the uncut forest.
Live root biomass of larger
roots, however,
recovered more slowly.
The biomass of live roots
<
17
mm
in diameter was about 30 per cent of that of the uncut forest, while
total live root biomass in the brushfields was only about 10 per cent of
that in the forest.
Virtually all but the largest of the roots which
were alive before harvest had now decayed.
Generally,
all that remained
of the dead roots were shells of bark surrounding completely decayed
wood.
Occasional sound resinous dead roots were found in all logged
areas
-- similar to results reported by Ziemer and
Swanston
(1977) in
southeast Alaska.
Such decay-resistant resinous roots accounted for,
perhaps, less than 5 per cent of the total forest root biomass in the
northern California study.
As non-resinous roots decay, however, the
resinous root fraction becomes the dominant component of residual root
biomass.
352
STRENGTH OF INDIVIDUAL ROOTS
Although roots generally tend to break in tension rather than shear
during slope failure,root shear strength is much easier to measure and
allows the study of larger roots than is allowed by tensile strength
apparatus.
Measurements of root tensile strength reported in the literature
have been limited to root diameters less than 15 mm, and most studies
have been conducted on roots smaller than 4 mm in diameter.
I recently developed a direct shear apparatus which satisfactorily
measures the shear strength of individual roots up to 50 mm in diameter
(Ziemer, 1978). Shear strength measured by this apparatus was closely
related to tensile strength measured by an apparatus developed by Burroughs
and Thomas (1977):
Tensile strength =
75 + 2.2 Shear strength,
in which strength is expressed in newtons.
The explained variance
(r2)
was 0.97.
No difference by species in the relationship was detected.
These results suggest that direct shear measurements can be used as a
surrogate of root tensile strength.
The strength of roots varies between tree species (Turmanina, 1965;
O'Loughlin, 1972; Ziemer and Swanston, 1977).
Segments of roots were
collected from the northern California root biomass study area for each
of the nine principal tree and brush species (Table 2). For each species,
approximately 10 roots were collected for each of six size classes,
ranging from 2 to 50 mm in diameter. Each root was sheared about five
times (Ziemer and Swanston, 1977; Ziemer, 1978).
A least-squares regression of log shear strength over log root
diameter yielded excellent results for each species.
The explained
variances for the nine equations ranged from
0.93
to 0.99.
Shear strength
353
was then calculated for each of the six root diameters.
The species
were ranked by decreasing root shear strength for each root diameter,
then by decreasing rank averages over all diameter classes (Table 2).
On the basis of Kendall's coefficient of concordance (Kendall,
1975),
the ranks within each diameter class agree in the overall order of the
species at the 1 per cent level of statistical significance.
As diameter increased,the species generally maintained their
relative position in the ranking.
Two species shifted in rank, however,
as diameter increased.
Arbutus had the weakest
2-mm
roots, but its
relative rank markedly changed as size of roots increased, and its large
roots were among the strongest. Libocedrus,
in contrast, had the
strongest 2-mm roots,
but its large roots were among the weakest.
The shrub and hardwood species have stronger roots than conifers.
Ceanothus roots,
a major component of brushfields, are about twice as
strong,
on the average,
as Abies and
Pinus
roots.
Ceanothus roots 2 mm
in diameter are about 1.35 times as strong as the conifer roots.
Their
biomass was about 82 per cent of that in the forest.
The difference in
strength more than compensates for the lower biomass.
The contribution
of these small roots to the strength of the soil is estimated to be
about 10 per cent greater in the brushfield than in the forest (i.e.,
1.35 times 0.82).
Brush roots
17-mm
in diameter are about 2.13 times
stronger than fir and pine roots,
but their biomass is only 32 per cent
of that of the forest.
And their relative contribution to
soi1
strength
would be about 68 per cent of that of the forest. The
50-mm
diameter
brush roots are about 3.76 times stronger than fir and pine roots, but
these brush roots account for only 4 per cent of the root biomass of
this size class found in the forest.
354
ESTIMATED
TRENDS IN SLOPE STRENGTH
Roots
<
17 mm in diameter are most useful in calculating relative changes
in slope strength because the biomass of that size class was the best
predictor of in situ shear strength.
--
As roots decay they lose both
biomass and strength. Root strength loss can be conservatively assumed
to be proportional to root biomass loss.
For a given biomass, the
strengthening effect of dead roots on the soil-root matrix is assumed to
be equal to that of live roots. Relative reinforcement of the soil
by roots was computed by summing over size/species classes the products
formed by multiplying the relative root strength by the proportion of
root biomass for each class (Fig. 3).
The dead root component in all study areas was predominately conifer
roots.
The relative reinforcement by dead roots follows an exponential
decline with time after logging (Fig. 3).
About half of the original
reinforcement was gone within 2 to 3 years after logging. Three-fourths
of the strength was lost within 8 years.
Essentially all of the reinforcement
by dead roots was gone within 25 years after logging.
The pattern of the relative reinforcement by live roots follows a
more complicated pattern.
Conifer roots contributed to live root biomass
only in the uncut forest. After logging,
live conifer roots were not a
significant component of the live root biomass.
In
3-
and
5-year-old
cutblocks,
the live root component consisted of Pteridium roots (rhizomes),
but they are only about one-fourth as strong as conifer roots and so
contributed little strength to the soil. Ceanothus began to occupy the
areas 7 years after logging and had completely revegetated the cutblocks
12 years after logging. The relative strength of brush roots
<
17 mm
in diameter weighted by the biomass in each size class, was 1.60 times
355
the strength of the conifer roots.
Consequently,the relative reinforcement
by live roots in the brushfields was about 60 per cent of that in the
uncut conifer forest. The brushfields apparently began to senesce,
and the relative effect of live roots in the
20-
to 25-year-old cutblocks
was about 40 per cent of that in the uncut forest.
The reinforcement by live roots and that by dead roots is the total
reinforcement action which the roots provide to the soil.
Within the
cutblocks studied, this reinforcement action drops to its low point about 7
years after logging,
when it was about 35 per cent of that of the uncut
forest.
When brush invaded the cutblocks,
root reinforcement recovered
to about 70 per cent of that in the uncut forest, but again decreased as
the brushfield aged and decay of the residual conifer roots progressed.
Given sufficient time without disturbance, the brushfield will
eventually yield to a coniferous forest.
If this transition is gradual,
reinforcement by roots will slowly return to that of the uncut forest as
projected (Fig. 3).
If the brush is killed in an attempt to establish
conifers more quickly, however,
the strength of the soil-root matrix
would be expected to again drop rapidly as the dead brush roots decay
and before the small conifer roots replace the brush roots.
Factor of safety analysis has been applied by
O'Loughlin
(1972),
Wu
et.
al
(1979),
and others to slopes with and without roots.
The factor
of safety provides an index of the relative stability of slopes.
If, for
a particular storm return period or soil water condition, the factor of
safety is less than 1,the slope is considered unstable.
Using this concept,
if the factor of safety is
1
when relative reinforcement was 0.3 (Fig.
3),
slides caused by loss of root strength would not be expected in the
logged areas because the minimum total reinforcement always exceeded 0.3.
356
If, however,
the factor of safety was 1 when reinforcement was 0.4,
areas logged from 4 to 8 years earlier would be vulnerable to increased
sliding, but younger or older cutblocks would not. Similar analyses,
based on locally obtained information,
would give land managers a tool
to evaluate the effect of logging and subsequent vegetative recovery on
sensitive slopes.
REFERENCES
Bjorkhem, U., Lundeberg, G. and Scholander, J. (1975) Root distribution
and compression strength in forest soils. Research Notes no. 22.
Departments of Forest Ecology and Forest Soils, Royal College of
Forestry, Stockholm, Sweden.
Burroughs,
E.R. and Thomas,B.R. (1977) Declining root strength in
Douglas-fir after felling as a factor in slope stability. U.S.
Dep. Agric. For. Serv. Res. Paper INT-190. Ogden, UT, USA.
Endo,
T. and Tsuruta,T. (1969) The effect of the tree's roots upon
the shear strength of soil.
1968 Annual Report, Hokkaido Branch,
Forest Experiment Station, 167-182.
English translation by Arata,
J.M.
and Ziemer, R.R., U.S. Dep. Agric. For. Serv., Arcata, CA, USA.
Golob,
T.B. and Silversides,
C.R. (1978) A study of a root-cutting
shear.
Forest Management Institute Information Report FMR-X-109.
Canadian Forestry Service, Environment Canada, Ottawa, Ontario, Canada.
Kendall, M.G. (1975) Rank Correlation Methods:
Ch. Griffin, London, UK
O'Loughlin,
C.L. (1972) An investigation of the stability of the
steepland forest soils in the Coast Mountains, southwest British
Columbia,
PhD.
dissertation,University of British Columbia.
Vancouver, B.C., Canada.
357
Turmanina,
V.I. (1965) The strength of tree roots.
Bul. Moscow
Sot.
Naturalists, Biol. Sec., vol. 70, no. 5, 36-45.
Wu, T.H.,
McKinnell,
W.P.
III
and Swanston, D.N. (1979) Strength of
tree roots and landslides on Prince of Wales Island, Alaska.
Can. Geotech.
J.,
vol. 16, no. 1, 19-33.
Ziemer,
R.R. (1978) An apparatus to measure the crosscut shearing
strength of roots.
Can.
J.
For.
Res.,
vol. 8, no. 1, 142-144.
Ziemer, R.R. and Swanston,
D.N. (1977) Root strength changes after
logging in southeast Alaska. U.S. Dep. Agric. For. Serv. Res.
Note PNW-306, Portland, OR, USA.
358
TABLE
1.
Correlation between in situ shear strength and the dry
--
weight of live roots,
by various size classes, for a mature
Pinus
contorta
stand growing on coastal sands, northern California.
Diameter of
Squared
live roots
correlation
(mm)
coefficient
<2
0.68
2-5
0.67
5-10
0.51
10-17
0.56
17-25
0.18
>25
0.03
Diameter of
Squared
live roots correlation
(mm)
coefficient
<2
0.68
<5
0.72
<l0
0.76
<l7
0.79
<25
0.72
Total 0.55
TABLE
2.
Live root shear strength and ranking (superscripts) by root diameter for species from the mixed
conifer area,
a/
northern California
Species
Sdueus
caIZicuq3a
Ceanothw
vehtinus
cas
tanupsis
chrysophy
Z
La
Arbutus
menxiesii
Pseudotsuga
menziesii
Zbocedrus
decurrens
flbies
coneoZor
Pinus
Zadertiana
Pinw
ponderosa
2
5
Root Diameter (mm)
10
17 25
50
~--~-~-l--=-
-shear strength
(N)-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
77
2
61
4
76 3
39
g
57
5
95
I
49
6
43
8
44
7
449
1
336
4
393
2
280
5
266
6
349
3
241
8
253
7
200
g
1600
1
1399
3
1400
2
1239
4
946
'
995
5
849
8
879
7
9
689
4096
2
4485
'
3781
4
3863
3
2658
5
2113
8
2286
6
2167
7
1871
'
7 974
3
10 888
1
7 866
4
8 830
2
5 819
5
4 355
7
4 767
6
4 063
8
3 972
'
25
436
5
58386
=
30
070
3
39
019
2
25
538
4
14
218
8
18
455
'
11
844
'
16
281
7
-
Species listed by decreasing rank averages over all diameter classes.
1 1
I
1
1
00
1
2
3
4
5
Biomass of <17mm live roots
(kg/m”)
FIGURE
1.
I
n
I
situ shear strength of
situ shear strength of the root-soil matrix the root-soil matrix increasedincreased
as the dry weight of live roots as the dry weight of live roots
<<
17 mm in diameter 17 mm in diameter
increased, in a mature increased, in a mature
Pinus
contorta
stand growing on stand growing on
coastal sands.coastal sands.
A. Live
roots
2kg
Years after logging
,2kg
B. Dead roots
---<25mm
------47mm
-
-
-
cl0mm
I-I---
<5mm
Years after logging
FIGURE 2.
Live root biomass increased and dead root biomassLive root biomass increased and dead root biomass
decreased with increasing time after clear-felling indecreased with increasing time after clear-felling in
the mixed conifer forest study area.the mixed conifer forest study area.
361
Years after logging
FIGURE 3.FIGURE 3.
The relative reinforcement of soils by live rootsThe relative reinforcement of soils by live roots
generally increased while generally increased while that by dead roots rapidlythat by dead roots rapidly
decreased with increasing time after clear-felling.decreased with increasing time after clear-felling.
The total reinforcement The total reinforcement by live and dead by live and dead roots droppedroots dropped
to a low point about 7 years after logging.to a low point about 7 years after logging.
