Paper Number: 035014
An ASAE Meeting Presentation
Preliminary Soil Disturbance Assessment of a Skyline Logging Operation Performing Five Silvicultural
Prescriptions
John Klepac, General Engineer
USDA Forest Service, Southern Research Station, Auburn, Alabama, USA
Stephen E. Reutebuch,
USDA Forest Service, PNW Research Station, Seattle, Washington, USA
Written for presentation at the
2003 ASAE Annual International Meeting
Sponsored by ASAE Riviera Hotel and Convention Center Las Vegas, Nevada, USA
27- 30 July 2003 Abstract.
Surface soil disturbance was compared among five different skyline logging units (commercial
thinning, group selection, patch cut, 2-age, and clearcut) and a control unit. Undisturbed area was
inversely related to harvested volume per acre and 8 percent higher than previously observed with
ground-based harvesting methods. No clear trends were observed across treatments, likely
reflecting the over-riding effect of topography on equipment selection and operating methods and
their associated disturbance. In units where terrain slope allowed increased use of a feller-buncher,
soil disturbance likewise increased. Overall, area of exposed soil was 9 percent lower and area
covered with litter following harvest was 9 percent higher than previously observed with groundbased harvesting methods. Severe soil disturbance (soil exposed with ruts greater than 4 inches
deep) was limited and similar between treatment units (0.8-2.7 percent of area). Newly deposited
litter generally increased with increasing volume harvested per acre.
Keywords. Harvesting, soil disturbance, silvicultural prescription, skid trail, erosion
The authors are solely responsible for the content of this technical presentation. The technical presentation does not necessarily
reflect the official position of the American Society of Agricultural Engineers (ASAE), and its printing and distribution does not
constitute an endorsement of views which may be expressed. Technical presentations are not subject to the formal peer review
process by ASAE editorial committees; therefore, they are not to be presented as refereed publications. Citation of this work should
state that it is from an ASAE meeting paper EXAMPLE: Author's Last Name, Initials. 2003. Title of Presentation. ASAE Paper No.
03xxxx. St. Joseph, Mich.: ASAE. For information about securing permission to reprint or reproduce a technical presentation, please
contact ASAE at hq@asae.org or 269-429-0300 (2950 Niles Road, St. Joseph, MI 49085-9659 USA).
Introduction
In recent years public concern has increased in the Pacific Northwest about the visual and
ecological effects of clearcut harvesting. In response, natural resource managers are testing
alternative silvicultural approaches that include various levels of harvest intensity. A thorough
understanding of these alternative silvicultural regimes, in terms of stand productivity, public
acceptability, production costs, and site impacts, is needed.
Harvest activity causes soil to be displaced, compacted, scarred, or churned. Removing the
organic layer and exposing the mineral soil can result in erosion problems and increased
sedimentation in streams. Amounts of erosion and stream sedimentation following logging may
vary directly with the degree of disturbance caused by timber removal (Dyrness, 1965). The
severity and frequency of soil disturbance varies with factors such as type of equipment and skill
of operators, soil type, moisture content, slope, time of year, and skid trail orientation (Allen and
others, 1999). Megahan (1980) summarized results from 9 prior soil disturbance studies by
harvesting systems in the Pacific Northwest. He reported that tractor operations (excluding
roads) disturbed an average of 21 percent (range 7-36 percent) of the activity area compared
to13 percent (range 1-32 percent) for ground cable systems (jammer, grapple, and highlead), 8
percent (range 3-12 percent) for skyline, and 4 percent (range 1-5 percent) for aerial systems
(helicopter). Unfortunately, a standardized soil disturbance rating system was not used in these
9 studies, explaining the wide ranges in reported disturbance and making comparison of results
of little value. Klepac and others (1999) found that, for a mechanized ground-based harvest
system, as harvest volume per acre increased, the percentage of undisturbed area decreased.
In this paper, we examine the level of soil disturbance associated with alternative silvicultural
approaches on steep terrain, where aerial cable logging is required. Five units that were
harvested under different removal intensities were surveyed, as well as a control unit that
received no treatment. A comparison of percent soil surface disturbance for seven disturbance
classes is presented. Percent area in primary and secondary skid trails and processing sites for
each unit is also reported.
Study Area and Stand Conditions
The soil disturbance survey reported here is the second replication block of a much larger, longterm evaluation of alternative silvicultural options for managing second-growth forests in
western Washington (Curtis and others, in press). The study site is known as the Copper Ridge
site, a name used for the timber sale associated with the harvest treatments. The study area is
located about 10 miles southwest of Olympia, Washington, on land owned by the Department of
Natural Resources (DNR). The topography of the site is steep, with most of the area having a
slope from 40 to 70 percent, although some small ridge top areas are relatively flat with gentler
slopes (10-30 percent). Elevation varies within the units from about 600 to 1,600 feet.
Soils in the study area are silt loams and gravelly silt loams. They are deep (50—60 inches),
well-drained soils formed in residuum and colluvium derived from underlying basalt (Pringle,
1990). The main limitation of these soils with regards to harvesting operations is their
steepness. Annual precipitation in the area is 60—80 in. Runoff from these soils is medium,
and associated erosion is moderate.
The site is covered with 60- to 70-year-old second-growth forests that naturally regenerated
after the original forest was clearcut and burned in the early part of the twentieth century. The
1
overstory is primarily Douglas-fir (Pseudotsuga menziesii) and western hemlock (Tsuga
heterophylla) with small components of red alder (Alnus rubra), and western redcedar (Thuja
plicata). Table 1 provides preliminary estimates of pre- and post-treatment stand
characteristics. Table 2 gives preliminary estimates of volume removed in each treatment unit.
Silvicultural Prescriptions
The study evaluated six silvicultural prescriptions for regenerating Douglas-fir (Pseudotsuga
menziesii) and are described as follows (Curtis and others, in press):
• Control--no harvesting was prescribed. The area will be allowed to grow, unthinned, for 60­
70 years additional years.
• Clearcut--all of the merchantable and unmerchantable trees were cut. The area will be
replanted and managed as an even-aged stand on a 60-year rotation schedule.
• Two-age--basal area was reduced by approximately 80 percent, leaving approximately 17
large dominate and co-dominate over-story trees per acre. The area will be under-planted
with a mixture of Douglas-fir, western hemlock, and red-cedar. The residual over-story trees
will be allowed to grow for 65-70 years.
• Thinning--basal area was reduced by approximately 20-30 percent. The unit will be
repeatedly thinned by a similar amount every 15 years for 60-70 years.
• Group selection with thinning--basal area was reduced by approximately 40 percent. The
area was first designed as a regular thinning unit, and then small, scattered areas (up to 1.5
acres) were marked to create openings over approximately 20 percent of the unit. The
openings will be planted with a mixture of Douglas-fir, western hemlock, and red-cedar. A
similar treatment will be applied every 15 years.
• Patch cut with thinning--approximately 20 percent of the unit was clearcut in four large,
scattered patches (1.5-5 acres in size). The remaining area surrounding these patches was
thinned. For the areas that were only thinned, the basal area was reduced by approximately
20-30 percent. The patches will be planted with Douglas-fir. A similar treatment will be
applied every 15 years.
Incomplete Thinning Unit Treatment
Unfortunately, the thinning treatment was only completed on a portion of the planned treatment
unit (approximately 8 ac out of 18.7 ac). After harvesting on the thinning unit began it was
decided that it was not economically feasible to continue; therefore, harvesting was stopped. To
obtain an estimate of the disturbance caused by the thinning treatment, observations collected
from the harvested thinning unit were combined with observations from thinned areas of the
patch cut, which was adjacent to the thinning unit.
Harvest System
Harvesting began in May 2002 and ended the following September. The units were
predominately cable logged with a Washington Ironworks 78SL yarding crane built in 1974. The
yarder was rigged as a gravity-return (shotgun) skyline system throughout most of the harvest,
with 2-4 chokers.
The yarder was equipped with a ¾-inch 1600-ft skyline. In much of the area, only partial
suspension of logs was obtained during yarding. Merchantable limits included a 12-ft length
with a 6-inch small end diameter. Stems were limbed and topped at the landing with a
processor mounted on a tracked carrier.
2
Table 1. Preliminary pre- and post-harvest stand characteristics of the six units.
Treatment
Control2
Clearcut
2-Age
Thinning3
Group Selection
Patch Cut4
1
2
3
4
Pre-harvest (2000)
Basal Area Stocking Diameter1
(sq ft/ac)
(trees/ac)
(in)
---206
214
13.3
239
137
17.9
242
150
17.2
242
218
14.3
260
171
16.7
Post-harvest (2003)
Basal Area Stocking Diameter1
(sq ft/ac)
(trees/ac)
(in)
238
226
13.9
0
0
50
17
23.1
n/a
n/a
126
98
15.4
164
84
18.9
Post-harvest Change (2003)
Basal Area Stocking Diameter1
(%)
(%)
(%)
----100
-100
-100
-79
-88
+29
n/a
n/a
n/a
-48
-55
+8
-37
-51
+13
Quadratic mean diameter: The Control unit was only measured in 2003. Only a small portion of the thinning unit was actually thinned, therefore unit-wide estimates are not available. Includes the entire unit, both areas that were clearcut (patches) and areas that were only thinned surrounding the patches. Table 3. Summary of sample size for each unit. Spacing is
given between and along transects.
Table 2. Preliminary estimates of volume harvested from each
of the six units (Scribner board foot volume to a 6-in
top, diameter-inside-bark, in 32-ft logs (16-ft log
minimum).
Treatment
Control
Clearcut
2-Age
Thinning1
Group
Selection
Patch Cut
Pre-harvest
Volume
(MBF/ac)
38.5
27.7
53.0
51.5
39.9
Net Volume
Harvested
(MBF/ac)
0
27.7
40.3
n/a
18.0
50.1
15.7
Silvicultural
Area
Spacing
No. of
Prescription
(ac)
(ft)
Observations
50 x 20
8282
Thinning
~81
Group Selection
58.4
47 x 48
1059
Patch Cut
73.9
100 x 50
716
2-Age
49.2
66 x 35
711
Clearcut
25.5
66 x 21
637
Control
27.6
66 x 25
477
1
Only a portion of the planned 18.7 acres in the thinning unit
was actually thinned.
2
Total includes 248 points from the thinning unit and 580
points from the thinned portion of the patch cut unit.
1
Only a small portion of the thinning unit was actually thinned,
therefore unit-wide estimates are not available.
3
Terrain in some portions of each unit was gentle enough for ground equipment to operate,
allowing these areas to be mechanically felled by a Timbco 445 feller-buncher. Steeper areas
were hand felled. In small areas of the clearcut, patch cut, and group selection units, the terrain
was flat enough to allow ground skidding by a rubber-tire skidder. The steepest unit was the 2­
age unit, followed by the group selection, control, patch cut, thinning, and clearcut units.
All of the units were harvested during the same year by a single contract logging crew using the
same array of equipment. When possible, the area was skidded uphill or cross-slope.
Methods
An assessment of soil surface disturbance was performed using the Point Transect Method
(McMahon, 1995) for each of the six units. The survey was conducted during October 2003.
Parallel transect lines were established along unit boundaries at a pre-determined spacing and
along the contour such that transects were generally perpendicular to yarding corridors. Soil
disturbance sample points at a pre-determined spacing were located along each transect. At
each point a visual assessment of the soil surface disturbance was made. Spacing between
transect lines and between points along a transect line varied depending on the size of the unit
being surveyed, with a goal of obtaining between 600-800 sample points per unit. Since it was
assumed that soil surface disturbance within the Control unit would have very little variability, a
goal of obtaining around 400 observations was targeted. Compass and pacing were used to
determine direction of transect lines and distance between points. Soil surface disturbance was
classified using the following classes:
1. Undisturbed – no evidence of machine or log movement over the point.
2. Disturbed with litter in place – litter was still in place with evidence of machine or
log movement over the point. Typically this class had machine track marks.
3. Litter removed and soil exposed – bare soil visible. Disturbance could be from
machine or log movements.
4. Litter and soil mixed – litter layer and soil were mixed together by machine or log
movements.
5. Soil exposed with rutting over 4 inches deep – points where machine travel or log
skidding had created significant ruts.
6. Non-soil – standing trees, stumps, logs, rocks.
7. Slash pile – piles of debris left from the logging operation.
8. Newly deposited litter – points that had been completely covered during the logging
operation with fresh needles, leaves, and fine branches, totally obscuring underlying
litter/soil conditions.
In addition to classifying the level of soil disturbance at each point, sample points located at
landings and in yarding corridors were also identified. For the group selection and patch cut
units points that fell within a group or patch were identified so that disturbance assessments
could be made within thinned areas, within group and patch areas, and for each unit overall.
Some units contained areas that were logged using a rubber-tire skidder because terrain was
conducive to this type of machinery. Other areas consisted of riparian zones where no logging
was allowed. Points that fell in either of these areas were coded so they could be eliminated
from the analysis.
4
Results and Discussion
The unit area, sample point spacing, and sample size for each unit is shown in Table 3. Percent
area in each disturbance class was computed for each treatment unit.
Observed Disturbance by Silvicultural Treatment
As expected, the control unit contained the highest undisturbed area (fig. 1). Comparing the
harvest treatment units, there was a general trend of increasing amount of undisturbed area as
harvest intensity decreased. For all treatments, the observed undisturbed area was greater by
an average of 8 percent (range 6-10) when compared to undisturbed area associated with the
same treatments harvested with a ground-based system (Klepac and others, 1999).
100
P ercent
80
60
40
20
0
Control
Group
Patch
2-Age
Clearcut
Thin
Treatment
Figure 1. Percent of area undisturbed.
The amount of area disturbed with litter in place did not differ much among prescriptions and
ranged from 9 percent for the group selection to approximately 16 percent for both the patch cut
and clearcut prescriptions (fig. 2). This type of disturbance was mainly caused by machine tires
or tracks and rolling or sliding logs. In this study a tracked feller-buncher was generally used to
fell trees in areas with slopes less than 40 percent.
Disturbed with litter removed and soil exposed is normally a result of heavy machine or log
movement over the area and is an indication of the amount of area occupied by skid trails or
yarding corridors. In this survey this type of disturbance was mainly located in yarding corridors
where logs were transported from the stump to the landing by the cable yarder. Results indicate
that the patch cut had the highest amount of disturbed with litter removed and soil exposed
(fig. 3) at over 8 percent of the area, followed by the clearcut (6.8), thinning (6.4), 2-age (4.5),
and group selection (3.7).
The survey revealed that the patch cut also had the highest amount of litter and soil mixed
(fig. 4) with 12.4 percent, followed by the thinning (11.5), clearcut (11), 2-age (9.7), and group
selection (7.6).
5
20
Percent
15
10
5
0
Control
Group
Patch
2-Age
Clearcut
Thin
Treatment
Figure 2. Percent of area disturbed with litter in place.
10
Percent
8
6
4
2
0
Control
Group
Patch
2-Age
Clearcut
Thin
Treatment
Figure 3. Percent of area with litter removed and soil exposed.
14
12
Percent
10
8
6
4
2
0
Control
Group
Patch
2-Age
Treatment
Figure 4. Percent of area with litter and soil mixed.
6
Clearcut
Thin
The clearcut and patch cut both had the same level of disturbance with soil exposed with
rutting over 4 inches deep (fig. 5) at 2.7 percent, followed by the thinning (2.4), 2-age (1.3),
and group selection (0.8).
3.0
Percent
2.5
2.0
1.5
1.0
0.5
0.0
Control
Group
Patch
2-Age
Clearcut
Thin
Treatment
Figure 5. Percent of area soil exposed with 4-inch or deeper rutting.
Differences in the non-soil class are only partially the result of the prescriptions. This class
characterizes on a treatment unit basis the area covered by debris, rocks, stumps, and standing
trees. If a unit had a high concentration of rocks, stumps, or down logs that existed prior to
harvest, then it will be reflected in the percentage for this class. Figure 6 summarizes percent
area for the non-soil class.
14
12
Percent
10
8
6
4
2
0
Control
Group
Patch
2-Age
Clearcut
Thin
Treatment
Figure 6. Percent of area with non-soil categories.
The percentage of observations which landed on slash piles (piles of limbs and tops from
harvesting) was 2.5 percent for the patch cut and 1.9 percent for the thinning, followed by the 2­
age at 1.1 percent (fig. 7). The clearcut and group selection were nearly the same with 0.6 and
0.7 percent, respectively.
7
3.0
Percent
2.5
2.0
1.5
1.0
0.5
0.0
Control
Group
Patch
2-Age
Clearcut
Thin
Treatment
Figure 7. Percent of area with slash piles.
During the harvest operation fine branches and needles from harvested trees were deposited on
the forest floor. In the survey these fine branches and needles were classified as newly
deposited litter and results are summarized in Figure 8. As might be expected, the frequency
of newly deposited litter increased with volume harvested per acre. The 2-age prescription
had the highest amount of this class at nearly 40 percent, followed by the clearcut with 32
percent. This is typical since these two treatments are the most intense and generate the most
new ground debris. The group selection resulted in nearly 22 percent of the area having new
litter, followed by the patch cut (13) and the thinning (11).
50
P ercent
40
30
20
10
0
Control
Group
Patch
2-Age
Treatment
Figure 8. Percent of area with newly deposited litter.
8
Clearcut
Thin
Combined Disturbance Classes
Exposed soil is at risk of eroding more rapidly than areas covered with litter. The control unit
had 91 percent litter cover (from the three disturbance classes containing some form of litter
cover: undisturbed, disturbed with litter in place, and newly deposited litter), whereas the
harvest treatment units had combined litter cover ranging from 54 (clearcut) to 68 (thinning)
percent. The skyline logging treatments in this study had on average 9 percent higher litter
cover than previously observed with ground-based harvesting methods used to apply the same
treatments (Klepac and others, 1999).
Likewise, combining the three disturbance classes containing some form of exposed soil (soil
exposed and litter removed, litter and soil mixed, and soil exposed with rutting over 4
inches deep), it was found that between 12 (group selection) and 24 (patch cut) percent of the
area had exposed soil. This is on average 9 percent lower exposed soil than previously
observed with ground-based harvesting methods used to apply the same treatments (Klepac
and others, 1999).
Area in Landings, Corridors, and Trafficked by Feller-Buncher
As displayed in Figure 9 the percent area in landings was similar among prescriptions, ranging
from 0.2 percent for the clearcut to 1.6 percent for the 2-age. No observations were recorded
on landings for the thinning. Percent area in corridors was also similar among prescriptions and
ranged from nearly 3 percent for the clearcut to 6.6 percent for the 2-age.
The amount of area trafficked by the feller-buncher was highly variable among prescriptions,
ranging from near 0 (group selection) to 20 (clearcut) percent. However, this was more of a
slope effect rather than a prescription effect. The group selection and 2-age units had more
challenging terrain than the clearcut, thinning, and patch cut units, which is reflected in the low
percentage of feller-buncher traffic for these two units.
25
Percent
20
Landing
15
Corridor
10
Feller-Buncher
5
0
Thin
Group
Patch
2-Age
Clearcut
Treatment
Figure 9. Percent of area in landings, corridors and trafficked by feller-buncher.
This increased use of the feller-buncher in the clearcut, patch, and thinning units likely
contributed to increased percentages of the following disturbance classes: disturbed with
9
litter in place (fig 2), litter removed and soil exposed (fig. 3), litter and soil mixed (fig. 4),
and soil exposed with rutting over 4 inches deep (fig. 5).
Conclusions
The only trends found for the treatments were that newly deposited litter increased and
undisturbed area decreased with increasing volume harvested per acre. No other trends were
observed for the remaining disturbance classes. One might have expected soil disturbance to
increase with volume harvested per acre; however, differences in terrain conditions (slope
steepness and shape) caused differences in equipment and methods used for felling and
yarding trees, which in turn caused differences in soil disturbance unrelated to treatment.
Portions of the clearcut, patch cut, and thinning units had relatively gentle terrain that allowed
greater use of a mechanical feller-buncher. This resulted in increased levels of disturbance due
to movements of the machine unrelated to silvicultural treatment. This mixing of mechanical
and hand felling within skyline units has become a common practice as fell-bunchers have
become common and have improved their capacity to work on steeper slopes. Therefore, it
was felt that including areas that were mechanically felled in the analysis was representative of
cable operations in the Pacific Northwest.
When compared to using a ground-based system for applying the harvest treatments, the
skyline system had an average of 8 percent more undisturbed area, 9 percent more litter cover,
and 9 percent less exposed soil across all treatments. Although this reduced level of
disturbance may reduce erosion due to raindrop impact, overall erosion could still be greater
from the skyline treatment units due to their increased steepness. More research is needed to
evaluate erosion and sediment yields from the two harvesting systems.
Finally, the results of the survey showed relatively low levels of severe soil disturbance
associated with any treatment, illustrating that on stable soils all of the treatments can be
successfully applied with skyline logging if reasonable care is taken in planning and
implementing the treatments.
Acknowledgements
The authors would like to thank the dedicated Forest Service field personnel who assisted with
data collection. Likewise, we thank the Washington State Department of Natural Resources for
providing the study site and for implementing necessary silvicultural prescriptions.
References
Allen, M. M.; M. Taratoot; P. W. Adams. 1999. Soil compaction and disturbance from skyline
and mechanized partial cuttings for multiple resource objectives in western and
northeastern Oregon, U.S.A. In: Proceedings of the International Mountain Logging 10th
Pacific Northwest Skyline Symposium. Eds: J. Sessions and W. Chung. March 28-April 1,
1999. Corvallis, OR. Dept. of Forest Engineering, Oregon State University, Corvallis, OR:
107-117.
Curtis, R.O., D.M. Marshall, and D.S. DeBell, (eds.). In press. Silvicultural options for younggrowth Douglas-fir forests: The Capitol Forest Study—establishment and first results.
U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station,
Portland, Oregon, General Technical Report PNW-GTR-XXX.
10
Dyrness, C.T. 1965. Soil surface condition following tractor and high-lead logging in the
Oregon Cascades. Journal of Forestry 63:272-275.
Klepac, J.F., and S.E. Reutebuch, and R.B. Rummer. 1999. An assessment of soil disturbance
from five harvesting intensities. ASAE Paper No. 99-5052. St. Joseph, Mich.: ASAE.
Megahan, W.F. 1980. Nonpoint source forestry activities in the western United States: results
of recent research and research needs. In: Proceedings of forestry and water quality: what
course in the 80s? An analysis of environmental and economic issues. Washington, DC:
Water Pollution Control Federation: 92-151.
McMahon, S. 1995. Accuracy of two ground survey methods for assessing site disturbance.
Journal of Forest Engineering, 6(2):27-33.
Pringle, R.F. 1990. Soil survey of Thurston County, Washington. United States Department of
Agriculture, Soil Conservation Service ; in cooperation with Washington State Department
of Natural Resources and Washington State University, Agricultural Research Center.
11