1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
October 21, 2003
Canadian Journal of Forest Research
Running Head: Managing for Old-Growth Forest Structure
Managing for Late-Successional/Old-Growth Characteristics
in Northern Hardwood-Conifer Forests
William S. Keeton
Rubenstein School of Environment and Natural Resources
University of Vermont
Burlington, VT 05405
Telephone: (802) 656-2518, Fax (802) 656-2623
Email: William.keeton@uvm.edu
Keeton
2
1
ABSTRACT
2
In the northern hardwood region of North America managing for late-successional forest habitats
3
and functions is an important element of ecosystem management. Modified uneven-aged
4
practices may provide an alternative for accelerating rates of late-successional forest
5
development. I am testing this hypothesis using a system that promotes old-growth structure,
6
termed “Structural Complexity Enhancement (SCE).” This approach is compared against
7
conventional uneven-aged systems modified to enhance post-harvest structural retention. The
8
study is replicated at two mature, northern hardwood forests in Vermont, U.S.A. Treatments
9
were applied to 2 ha units. Uneven-aged treatments were replicated twice; the SCE treatment and
10
controls were each replicated four times. Structural objectives include multi-layered canopies,
11
elevated large snag and downed log densities, variable horizontal density, and re-allocation of
12
basal area to larger diameter classes. The latter objective is achieved using crown release and an
13
unconventional marking guide based on a rotated sigmoid diameter distribution. The marking
14
guide is also derived from a target basal area and tree size indicative of old-growth structure.
15
Forest structure data have been collected over two years pretreatment and three years post-
16
treatment. Fifty-year simulations of stand development were run in NE-TWIGS and FVS
17
comparing treatment and no treatment scenarios. SCE retained higher levels of multiple
18
structural attributes, achieved rotated sigmoid diameter distributions, and sustained these 50
19
years into the future, resulting in reallocated basal area. Late-successional structure will develop
20
faster and to a greater degree under SCE. Large tree (>50 cm dbh) recruitment will be impaired
21
under the conventional treatments, whereas recruitment will be accelerated under SCE.
22
2
Keeton
1
3
INTRODUCTION
2
Sustainable forestry practices across “matrix” landscapes, such as forestlands providing
3
connectivity between protected areas, contribute to the maintenance of biological diversity and
4
ecosystem functioning at sub-regional and regional scales (Lindenmayer and Franklin 2002).
5
The challenge lies in determining the mix of management approaches – including type, timing,
6
intensity, and spatial configuration of silvicultural treatments – necessary to achieve
7
sustainability objectives. In the northern hardwood region of the eastern United States and
8
Canada, “structure-“ (Keeton 2005) or “disturbance-based” (Mitchell 2002, Seymour et al. 2002)
9
silvicultural approaches are of interest for this type of matrix management. Structure-based
10
forestry focuses on the architecture of individual forest stands and their spatial arrangement, with
11
consideration given to the aggregate representation of multiple structural (or habitat) conditions
12
at landscape scales. Disturbance-based silviculture uses the range of structural and
13
compositional conditions associated with natural disturbance regimes as a guide for designing
14
and scheduling silvicultural treatments. These approaches are complementary. They share the
15
operational objective of managing for currently under-represented forest structures and age
16
classes (Franklin et al. 2002). In matrix landscapes of the northern forest region this would
17
include managing for late-successional structure, which is vastly under-represented relative to
18
pre-European settlement conditions (Mladenoff and Pastor 1993, Cogbill 2000, Lorimer 2001,
19
Lorimer and White 2003). In this study variants of uneven-aged silviculture are explored as a
20
means for promoting late-successional/old-growth structural characteristics in northern
21
hardwood-conifer forests. Alternate approaches for accelerating stand development rates are
22
tested.
23
3
Keeton
1
2
4
Why manage for old-growth structure in northern hardwood forests?
In the northeastern United States there has been considerable debate regarding differing
3
proposals for adjusting age class distributions on forested landscapes. Forest structure and
4
composition in pre-European settlement landscapes were spatially and temporally variable due to
5
geophysical heterogeneity, climate variability, and disturbances, both natural and anthropogenic
6
(Foster and Aber 2004). However, the relative availability of stand structures has changed as a
7
result of 19th century forest clearing, agricultural land-use, land abandonment, subsequent
8
reforestation, and twentieth century forest management. In northern hardwood forests, age-class
9
distributions have shifted from a predominance, pre-settlement, of old-growth forests to a present
10
dominance of mature (e.g. 40 to 80 year old) forests (Lorimer and Frelich 1994, Lorimer 2001,
11
Lorimer and White 2003). Lorimer (2001) estimates that the majority of pre-settlement forests
12
were old-growth, whereas young forests (up to 15 years old) comprised <1 to 13% of northern
13
hardwood regions. Definitions of old-growth northern hardwood-conifer forests generally use a
14
combination of age (ca. >150 years), human disturbance history, and structure (Hunter 1989,
15
Dunwiddie et al. 1996, Hunter and White 1997). Today forests with these characteristics occupy
16
less than 0.5% of the region (Davis 1996). Riparian (Keeton et al. 2005), habitat (see reviews in
17
Tyrell and Crow 1994b, Keddy and Drummond 1996, and McGee et al. 1999), and other
18
functions associated with late-successional forests have declined as a result.
19
With reforestation and successional development on old-fields, the availability of
20
grassland and early-successional forested habitats declined significantly during the 20th century.
21
Successional habitats are now re-approaching (Lorimer 2001) or, in some locales, possibly even
22
below (DeGraaf and Yamasaki 2003) pre-settlement levels. Even-aged, patch-cut, or large-
23
group selection harvesting methods are sometimes advocated as a means for enhancing the
4
Keeton
5
1
relative abundance of habitats for early-successional species (Hunter et al. 2001, King et al.
2
2001, Litvaitis 2001, DeGraaf and Yamasaki 2003). Similarly, researchers have suggested that a
3
portion of the landscape could be managed for late-successional and old-growth forests (Keeton
4
et al. 2001, Lorimer and White 2003). These proposals need not be mutually exclusive, but do
5
require differing silvicultural approaches where active manipulations are desired.
6
7
8
9
Experiments in alternative silviculture
The field of alternative silviculture has expanded greatly in recent decades (Petersen and
Maguire 2005). In moist temperate and boreal regions of North America these systems are
10
generally designed: 1) to more closely approximate natural disturbance and stand development
11
processes through structural retention during regeneration cutting (Swanson and Franklin 1992;
12
Franklin et al. 1997, 1999, and 2002; Marshall and Curtis 2005; Beese et al. 2005); 2) create
13
spatial patterns and small gap dynamics similar to those associated with fine-scale natural
14
disturbances (Seymour 2005); or 3) to accelerate rates of late-successional forest development in
15
mature and riparian stands through under-planting, variable density thinning, crown release, and
16
other methods (FEMAT 1993, Berg 1995, Singer and Lorimer 1997, Harrington et al. 2005).
17
In the northern hardwood region particular interest in alternative, structure-based
18
approaches has evolved from studies of old-growth northern hardwood and mixed hardwood-
19
conifer forests. These have demonstrated the ecological significance of specific structural
20
elements, such as large trees (live and dead), downed logs, multi-layered canopies, and
21
horizontal variations in stand density and gap mosaics (Tyrell and Crow 1994b, Dahir and
22
Lorimer 1996, McGee et al. 1999). Availability of these structures can be highly limited in
23
forests managed under conventional even and uneven-aged systems (Gore and Patterson 1985,
5
Keeton
6
1
McGee et al. 1999). As a result, managing for old-growth structural characteristics, either in part
2
or in full, is a proposed alternative silvicultural approach (Keddy and Drummond 1996, Lorimer
3
and White 2003).
4
5
6
Structural Complexity Enhancement
Despite theoretical discussions (e.g. Lorimer and Frelich 1994, Trombulak 1996), there
7
have been few field trials or experimental tests of late-successional/old-growth forest restoration
8
in northern hardwood forests. An untested hypothesis is that silvicultural practices can
9
accelerate rates of late-successional forest stand development (Franklin et al. 2002), promote
10
desired structural characteristics, and enhance associated ecosystem functions more than
11
conventional systems. This hypothesis is tested here using an approach, termed “Structural
12
Complexity Enhancement (SCE), that promotes old-growth structural characteristics while also
13
providing opportunities for low-intensity timber harvest (Table 1). SCE is compared against two
14
conventional uneven-aged systems (single-tree selection and group-selection) that are advocated
15
regionally for sustainable forestry, particularly where maintenance of continuous forest cover is
16
desired (Mladenoff and Pastor 1993, Nyland 1998). Conventional uneven-aged prescriptions
17
employed in this study are modified to increase post-harvest structural retention and to represent
18
best available practices. In addition, group-selection treatments are modified to approximate the
19
average canopy opening size associated with fine-scale natural disturbance events in New
20
England, based on the findings of Seymour et al. (2002).
21
The objectives for SCE are based on previous research describing old-growth northern
22
hardwood and mixed northern hardwood-conifer forests (Gore and Patterson 1985, Hunter 1989,
23
Woods and Cogbill 1994, Tyrrell and Crow 1994a and 1994b, Hunter and White 1997,
6
Keeton
1
Goodburn and Lorimer 1998, McGee et al. 1999, Ziegler 2000). They include multi-layered
2
canopies, elevated large snag and downed log densities, variable horizontal density, and re-
3
allocation of basal area to larger diameter classes. The latter objective is achieved, in part, using
4
an unconventional marking guide based on a rotated sigmoid target diameter distribution.
5
Rotated sigmoid diameter distributions have been widely discussed in the theoretical literature
6
(e.g. O’Hara 1998, Leak 2002) but their silvicultural utility has not been field tested. Sigmoidal
7
form is one of several possible distributions in eastern old-growth forests (Goodburn and
8
Lorimer 1999). These vary with disturbance history, species composition, and competitive
9
dynamics (Leak 1996 and 2002). The distribution offers advantages for late-successional
7
10
structural management because it allocates more growing space and basal area (and thus biomass
11
and structures associated with larger trees) to larger size classes. I predict this distribution is
12
sustainable in terms of recruitment, growth, and yield. If so, it would support O’Hara’s (1998)
13
assertion that there are naturally occurring alternatives to the negative exponential or “reverse-J”
14
curve typically used in uneven-aged forestry. It would also suggest that silviculturalists have
15
greater flexibility in managing stand structure, biodiversity, and other ecosystem functions in the
16
northern forest region than previously recognized.
17
18
METHODS
19
Experimental Design
20
The study is conducted at the Mount Mansfield State Forest (MMFS) and at the
21
University of Vermont’s Jericho Research Forest (JRF). The former is located on the western
22
slopes and the latter resides in the foothills of the northern Green Mountain Range in Vermont,
23
U.S.A.. Elevations range from 470 to 660 m (MMFS) and from 200 to 250 m (JRF) above sea
7
Keeton
8
1
level. Soils are primarily stony loams (MMSF) and loamy sands or sandy loams (JRF). Study
2
areas are mature (ca. 70-100 year old), northern hardwood-conifer forests. Tree demography is
3
multi-aged due to previous management entries, based on extensive pre-treatment tree coring
4
across size classes. Dominant overstory species include Acer saccharum (sugar maple), Fagus
5
grandifolia (American beech), and Betula alleghanienis (yellow birch). Tsuga canadensis
6
(eastern hemlock) is also co-dominant at JRF. There are minor components of Picea rubens (red
7
spruce) at MMFS and Acer rubrum (red maple) and Quercus rubra (red oak) at JRF.
8
9
There were three experimental manipulations. The first two were conventional unevenaged systems (single-tree selection and group-selection) modified to enhance post-harvest
10
structural retention. The modifications were based on a target residual basal area of 18.4 m2/ha,
11
maximum diameter of 60 cm, and q-factor (the ratio of the number of trees in each successively
12
larger size class) of 1.3. Group-selection cutting patches were each approximately 0.05 ha in size
13
which results in 8 to 9 groups per treatment unit.
14
The third treatment was Structural Complexity Enhancement (SCE). The marking guide
15
(the number of trees that must be cut in each size class to achieve the desired post-harvest
16
structure) was based on a rotated sigmoid target diameter distribution (number of trees per size
17
class). This was applied as a non-constant q-factor: 2.0 in the smallest sizes classes, 1.1 for
18
medium-sized trees, and 1.3 in the largest size classes. The marking guide was also derived from
19
a target (desired future condition) basal area (34 m2/ha.) and maximum diameter at breast height
20
(90 cm) indicative of old-growth structure. Accelerated growth in larger trees was promoted
21
through full (4 or 3-sided) and partial (2-sided) crown release. Prescriptions for enhancing
22
coarse woody debris (CWD) volume and density were based on stand potential (e.g. pre-harvest
23
CWD volume) and literature-derived targets. Snags were created by girdling diseased, dying, or
8
Keeton
9
1
poorly formed trees. Pre-treatment densities of low vigor trees were sufficient such that girdling
2
of healthy trees was not necessary to achieve snag prescriptions. On one SCE unit at each of the
3
two study areas, downed logs were created by pulling trees over, rather than felling, to create pits
4
and exposed root wads.
5
Each of the first two treatments (uneven-aged) was replicated twice at Mount Mansfield;
6
the third (SCE) was replicated four times, twice at each of the two study areas. Two un-
7
manipulated control units were located at each study area. Manipulations were randomly
8
assigned to treatment units. Treatment units were 2 ha in size and separated by 50 meter (min.)
9
unlogged buffers to minimize cross contamination of treatment effects. Experimental
10
manipulations (i.e. logging) were conducted on frozen ground in winter (Jan.-February) of 2003.
11
All treatments included retention of American beech exhibiting resistance to beech bark disease
12
(Nectria coccinea). Scattered mature red spruce were retained as seed trees to encourage
13
recolonization where this species was historically more abundant.
14
15
Data Collection
16
There are five, randomly placed, 0.1 ha permanent sampling plots in each treatment unit.
17
Within each plot, all live and dead trees > 5 cm dbh and > 1.37 m tall were permanently tagged,
18
measured, and recorded by species, diameter, height, and vigor/decay stage (1-7). Tree heights,
19
height to crown base, and crown width (2 radii) on each tagged tree were measured using an
20
Impulse 200 laser range finder. Tagged trees were mapped using an integrated laser range finder
21
and Mapstar Digital Electronic Compass. Plot and tree positions were geo-referenced using a
22
Trimble Pro XRS Global Positioning System. Downed log (logs > 10 cm diameter) volume by
23
decay class (1-5) was estimated using a line-intercept method while log densities were measured
9
Keeton 10
1
across 0.1 ha plots. Leaf Area Index (LAI) was measured at five points in each plot using a Li-
2
Cor 2000 meter. LAI values were post-processed to calibrate “below canopy” measurements
3
against ambient “above canopy” light measurements taken by a remotely placed Li-Cor meter.
4
Two dominant canopy trees per plot were cored at breast height to allow subsequent laboratory
5
determination of tree age and site index50 (sugar maple). Two years of pre-treatment and three
6
years of post-treatment data collection have been completed. Two sample plots (one SCE unit
7
and one single-tree selection unit) escaped treatment due to inoperable or wet terrain. Data from
8
these units were not included in analyses of post-treatment data.
9
10
11
Data Analysis
Sample data, including individual tree heights, crown dimensions, and CWD attributes,
12
were used as inputs for 3-dimensional modeling in the Stand Visualization System (SVS)
13
(McGaughey 1997). The Northeast Decision Model (NED-2) (Twery et al. 2005) was used to
14
generate stand structure metrics based on pre and post-harvest sample data. These included
15
aboveground biomass estimates based on species-specific allometric equations developed by
16
Jenkins et al. (2003). A Before/After/Control/Impact (Krebs 1999) statistical approach was used
17
to compare structural metrics pre- to post-harvest and among treatments. Tukey-tests were used
18
for the former while Analysis of Variance and post-hoc Bonferroni (CWD data) or Least
19
Significant Difference (LSD, overstory data) multiple comparisons were used for the latter.
20
Homogeneity of variance was tested using F-tests for both pre and post-harvest data. Spatial
21
autocorrelation tests (Riply 1981) using the Moran coefficient were performed on key response
22
variables. Response data were sorted by treatment and made spatially explicit using geo-
23
referenced plot positions. Spatial autocorrelation results were cross-checked against empirical
10
Keeton 11
1
variograms produced using S-Plus (Kaluzny et al. 1998) software. Pre to post-harvest shifts in
2
downed woody debris decay class distributions were assessed using the Kolmogorov-Smirnov
3
two-sample goodness of fit test.
4
Diameter distributions (pre and post-harvest) were also used as an indicator that
5
integrates vertical and horizontal structural responses to treatment. To determine whether SCE
6
successfully shifted diameter distributions towards the target rotated sigmoid form, pre- and
7
post-harvest and target distributions were log transformed to enhance sigmoidal tendencies (Leak
8
2002). Residual distributions were smoothed using a Friedman smoothing run in S-Plus
9
software. Kolmogorov-Smirnov two-sample goodness of fit tests were used to test for
10
statistically significant differences between transformed residual and target cumulative frequency
11
distributions. Residual distributions were created using real (sample) data for smaller diameter
12
classes (<70 cm dbh) and hypothetical (e.g. future potential) values for larger diameter classes
13
(>70 cm dbh). The latter borrowed values from the target distribution. Statistical tests, therefore,
14
evaluated whether residual distributions achieved that portion of the target distribution possible
15
given the pre-harvest structure.
16
NED-2 output was used for simulation modeling of stand development using two models:
17
the northeastern U.S. variant (NE-FVS, Bush 1995) of the USDA Forest Service’s Forest
18
Vegetation Simulator (Dixon 2003) and NE-TWIGS (Hilt and Teck 1989). The NE-FVS
19
modeling structure is based on NE-TWIGS, which is an individual tree-based, distance-
20
independent stand growth simulator. However, mortality and large-tree growth functions operate
21
slightly differently in NE-FVS and calculations are made every ten years, rather on the annual
22
time step employed in NE-TWIGS. For both models, simulations were run two ways: 1) using
11
Keeton 12
1
unit-specific site index values; and 2) holding site index constant (set to 65) to determine
2
simulation sensitivity to this source of variability.
3
Fifty-year projections of stand development were run for each treatment unit, including
4
controls, using both pre and post-harvest scenarios. Cumulative basal area increment (CBAI)
5
was calculated for each simulation run at 5 year intervals. Projections were normalized on a unit
6
by unit basis by calculating the differences in CBAI between “no-harvest” and “harvest”
7
scenarios at each time step. The Kolmogorov-Smirnov two-sample goodness of fit test was used
8
to test for differences between treatment groups along mean CBAI time series. The log-
9
likelihood ratio goodness of fit test (G test) was used to examine total CBAI developed after 50
10
years; response ratios (treatment vs. no treatment) were compared against a null ratio (no
11
treatment effect). Simulation modeling was also used to predict the number of large trees (two
12
classes: > 50 cm and > 60 cm dbh) developed after 50 years. Mean projected diameter
13
distributions by treatment were evaluated to determine whether either rotated sigmoid (SCE) or
14
negative exponential (conventional uneven-aged) were sustained over time. Projected basal area
15
distributions were generated to determine the corresponding effect on basal area re-allocation.
16
Single-tree selection and group selection treatments were classified as one group
17
(“conventional uneven-aged aged”) for analyses of simulation output (e.g. calculations of mean
18
projections by treatment) and other statistical tests. This was appropriate because: 1) there were
19
no significant differences among uneven-aged units in residual structure when data were
20
aggregated to the unit scale; and 2) quantitative prescriptions were the same for both, the only
21
difference being dispersed vs. aggregated harvesting. This difference was not relevant since
22
neither NED-2 output nor the simulations were spatially explicit.
23
12
Keeton 13
1
RESULTS
2
Pre-harvest Structure
3
Pre-treatment structure was not significantly different (α = 0.05) between treatment units
4
based on ANOVA results. This held for basal area, aboveground biomass, stem density, relative
5
density, and all other variables tested. Variance in structure among units assigned to the same
6
treatment was not statistically significant (α = 0.05). These results suggest there is a statistically
7
valid basis for comparisons of both pre to post harvest changes and with respect to differences in
8
residual structure among treatment units. Site index50 varied among units, ranging from
9
approximately 57 to 70.
10
11
12
Residual Stand Structure
Visualizations generated in SVS illustrate the high degree of structural complexity
13
maintained by both SCE and single-tree selection (Figure 1). High levels of residual structure
14
were maintained by all of the experimental treatments, including group selection. However,
15
there were distinct differences. Canopy closure was highest for SCE (mean 77%) and lowest for
16
single-tree selection (mean 64%). These results were statistically significant (P < 0.01). Canopy
17
closure, as expected, was most variable across group-selection units. Aggregate canopy closure
18
remained high in group-selection units because 70- 80% of each group-selection unit was
19
unlogged and thus maintained full pre-harvest structure. F tests of variance showed no
20
significant differences (α = 0.05) in post-harvest structure between similarly treated units. There
21
was no significant (α = 0.05) spatial autocorrelation among plots sorted by treatment; a result
22
confirmed by empirical variograms. These results suggest that similar treatments produced
23
consistent effects with respect to horizontal structure and spatial heterogeneity.
13
Keeton 14
1
There were significant differences (P < 0.001) in LAI responses among treatments.
2
Single-tree and group selection cuts reduced LAI by 20 and 30% respectively. LAI reductions
3
were lowest in SCE units (9%), indicating high retention of foliage bearing tree crowns and thus,
4
by inference, vertical complexity. LAI was significantly more spatially variable for both SCE (P
5
= 0.031) and group-selection (P = 0.010) compared to single tree selection; within-treatment
6
variance was not significantly different between SCE and group-selection units (P = 0.296).
7
These results are indicative of the high degree of horizontal structural variability expected for
8
group-selection. In SCE units, increased horizontal complexity was achieved through variable
9
density marking and clustered harvesting around crown-release trees.
10
Both SCE and conventional uneven-aged treatments resulted in structural changes pre to
11
post-harvest but these were statistically significant only for the conventional treatments. For
12
example, SCE resulted in reductions of 24% for aboveground biomass, 25% for total basal area,
13
27% for relative density, and 28% for stem density, though none of these changes were
14
significant at the 95% confidence level. By contrast conventional treatments caused the same
15
attributes to decline by 43, 42, 41, and 41% respectively, and these changes were highly
16
significant (P < 0.001). Stand structure did not change significantly in control units during the
17
course of this study (+/- 0.1 to 2% depending on the parameter).
18
Post-harvest aboveground biomass (P = 0.041), total basal area (P = 0.010), relative
19
density (P = 0.008), and stem density (P = 0.025) were significantly different among treatments
20
based on ANOVA and post-hoc comparisons. Conventional treatments resulted in significantly
21
lower aboveground biomass (P = 0.014), total basal area (P = 0.003), relative density (P =
22
0.002), and stem density (P = 0.008) in comparison to control units. SCE did not result in
23
statistically significant contrasts with controls. Residual basal area (P = 0.037) and relative
14
Keeton 15
1
density (P = 0.048) were significantly higher in SCE units compared to conventional uneven-
2
aged units; post-harvest biomass (P = 0.104) and stem density (P = 0.124) were not significantly
3
different between these treatments.
4
SCE shifted residual diameter distributions to a form indistinguishable from the target
5
rotated sigmoid form. There were no statistically significant differences (α = 0.05) between
6
residual and target distributions for any of the SCE units based on the goodness of fit test using
7
log transformed diameter distributions (Figure 2).
8
9
Crown Release and Vertical Development
10
Variable density harvesting resulted in crown release for 45 dominant trees per ha on
11
average in SCE units. When combined with the average pre-treatment number (20 per ha) of
12
large trees (> 50 cm dbh), this exceeded the future target of 55 large trees per ha. The excess
13
provides a “margin of safety” to accommodate canopy mortality. Crown release is likely to
14
accelerate growth rates in the affected dominant trees by 50% or more based on previous
15
modeling (e.g. Singer and Lorimer 1997). Dominant canopy trees were released across a range
16
of diameter classes (>25 cm dbh); the majority were fully, rather than partially, crown released.
17
Crown release also resulted in spatial aggregations of harvested trees, creating canopy openings
18
and variable tree densities. Elevated light availability associated with clustered harvesting
19
around crown release trees is likely to promote vertical differentiation of the canopy through
20
release and regeneration effects. This may increase foliage density in the emergent and lower
21
canopy layers over time.
22
15
Keeton 16
1
Coarse Woody Debris Enhancement
2
SCE prescriptions resulted in substantially elevated densities of both downed logs and
3
standing snags. Pulling trees over was successful in most cases at creating large exposed root
4
wads and pits. The structural complexity enhancement treatments increased CWD densities by
5
10 boles (> 30 cm dbh) per ha on average for snags and 12 boles (> 30 cm dbh) per ha. on
6
average for downed logs. Post harvest dead tree basal area was 39 % higher in SCE compared to
7
conventional treatments, although this was not statistically significant (P = 0.092) based on
8
multiple comparison results. Post-harvest dead basal area was significantly higher when SCE
9
units were compared to controls (P = 0.036); conventional units showed no significant
10
11
differences with controls.
There were statistically significant differences (P = 0.002) among treatments with respect
12
to downed log recruitment effects (Figure 3). Post-harvest downed log volumes were 140%
13
higher on average than pre-harvest levels in SCE units. Mean downed log volume increased
14
30% in the uneven-aged units, although this effect was not statistically significant relative to the
15
controls. There were only slight (5% mean) increases in control units due to natural mortality;
16
volumes declined in some control units. Background recruitment rates were thus not sufficient to
17
explain SCE treatment effects. Analyses of downed log decay class distributions in SCE units
18
showed, as expected, significant (P < 0.05) shifts towards less decayed logs due large inputs of
19
felled trees (Figure 4).
20
21
Projected Stand Development
16
Keeton 17
1
2
Basal area and aboveground biomass development
Simulation projections showed little sensitivity to site index. For example, holding site
3
index constant changed CBAI after 50 years by only 1.2 m2/ha on average. Results reported here
4
control for site index. In comparison to NE-TWIGS, FVS (northeastern variant) tended to
5
produce more conservative estimates of growth increment and large tree recruitment due to the
6
differences in model operation previously mentioned.
7
Simulations in NE-TWIGS suggest that live basal area under SCE will, on average,
8
approach 34 m2/ha after 50 years of development (Figure 5, top). This is 24 % (or 8 m2/ha)
9
higher than the mean for the conventional uneven-aged units. Projected basal area for SCE also
10
exceeds the mean predicted for control units by 13% (or 4.5 m2/ha). Conventional units were
11
projected to have basal areas still 12 % (or 3.6 m2/ha) below the control units after 50 years of
12
development. Projected basal area was more variable (Figure 5, top) among SCE units (+/- 2.9
13
m2/ha) compared to conventional units (+/- 0.6 m2/ha). Stand development projections differed
14
slightly depending on choice of model. While the magnitude of difference among treatments
15
was consistent in FVS projections, CBAI and end-of-run basal area were 12-18% lower, except
16
for control units in which CBAI was flat and thus consistent under both models.
17
The projections showed no significant differences in absolute growth rates between
18
treatment scenarios, as measured by CBAI and evaluated using goodness of fit tests (Dmax =
19
0.007, critical value of D0.05 = 0.307). When projected development with treatment is
20
normalized, to reflect the amount of development (specific to each unit) that would have been
21
expected with no treatment, the simulations indicate that CBAI will be slightly faster under
22
conventional systems (Figure 5, bottom). However, this difference was not statistically
23
significant (Dmax = 0.005, critical value of D0.05 = 0.183). Both SCE (P < 0.05) and conventional
17
Keeton 18
1
treatments (P < 0.01) are projected to significantly accelerate tree growth rates above that
2
expected with no treatment based on both NE-TWIGS and FVS modeling. Projected CBAI after
3
50 years was 2.7 m2/ha (no treatment) compared to 6.4 m2/ha (with treatment), on average, in
4
SCE units. For conventional units projected CBAI increased from 0.34 m2/ha (no treatment) to
5
5.3 m2/ha (with treatment). Therefore, differences in basal area development are largely
6
attributable to treatment effects on residual basal. Both conventional and SCE treatments
7
accelerate basal area increment, but SCE leaves more residual basal area and thus ultimately
8
results in higher basal areas compared to conventional uneven-aged treatments.
9
Neither SCE nor conventional treatments resulted in projected basal areas that exceeded
10
those projected for the corresponding units under a “no treatment” scenario. However, basal area
11
in SCE units recovered to within 89% of the no-treatment scenario, whereas conventional units
12
recovered to within 77% on average. This difference was statistically significant (P = 0.025).
13
Aboveground biomass is predicted to increase over the next 50 years in all of the
14
experimental units, including controls, based on the FVS simulations and NED-2 biomass
15
calculations. Living biomass increases 70.5, 66.1, and 73.6% on average for SCE, uneven-aged,
16
and control units respectively under a “no treatment” scenario. It increases 90.7 and 105.2%
17
following SCE and uneven-aged treatment, but starts from a post-harvest level that is 18.3 and
18
35.9% lower than pre-treatment respectively. Consequently, neither treatments achieve the
19
biomass they would have without treatment. However, after 50 years SCE results in
20
aboveground biomass that is 91.4% of that projected under no treatment, while the conventional
21
treatments result in 79.1% of the no treatment potential. Biomass production annual increment is
22
accelerated 5.1% for SCE and 1.9% for conventional treatments based on normalizing treatment
23
against no treatment scenarios.
18
Keeton 19
1
2
3
Large tree recruitment
SCE is projected to enhance rates of large tree recruitment over no treatment scenarios.
4
Based FVS projections, in SCE units there will be an average of 5 more large trees (> 50 cm
5
dbh) per ha than there would have been without treatment after 50 years. There will be 10 fewer
6
large trees per ha on average in the conventional units than would have developed in the absence
7
of timber harvesting. SCE results in an increase of 4 very large trees (> 60 cm dbh), while
8
conventional treatments produce 3 fewer very large trees than would have been recruited without
9
treatment.
10
11
12
Projected diameter and basal area distributions
Projections suggest that a rotated sigmoid diameter distribution will be sustained over 50
13
years in SCE units (Figure 6). Projected and target (i.e. desired future) rotated sigmoid
14
distributions were not significantly different (Dmax = 0.058, critical value of D0.05 = 0.430). This
15
held as long as densities in smallest two diameter classes (stems < 15 cm dbh) were held
16
constant. This was appropriate because both NE-TWIGS and FVS lack a seedling recruitment
17
model (there is only limited regeneration through stump spouting). Pronounced regeneration
18
deficiencies thus develop rapidly in model projections. This explains the low densities evident in
19
the smallest size classes in Figure 6. Uneven-aged units appear to maintain the negative
20
exponential distributions initially prescribed. However, they have significantly less recruitment
21
into larger size classes compared to the projected distributions for SCE (Figure 6).
22
23
Differences in projected diameter distributions and large tree recruitment explain related
changes in basal area distributions. Compared to stand development under a “no treatment”
19
Keeton 20
1
scenario SCE results in significant reallocation of basal area into the largest size classes (e.g. >
2
50 cm dbh). This includes a shift of basal area into the very largest size classes (>85 cm dbh)
3
that experience no basal area recruitment under the “no treatment” scenario (Figure 7, top). The
4
uneven-aged treatments – modified to include a low q-factor and relatively high maximum
5
diameter – also resulted in some reallocation of basal to larger size classes (Figure 7, bottom).
6
However, this reallocation was substantially lower than for SCE and did not include recruitment
7
into size classes larger than the projected “no treatment” distribution. In fact, the “no treatment”
8
simulations showed basal area development into the 75-80 cm diameter class, whereas this was
9
lacking for the corresponding units subjected to uneven-aged treatments (Figure 7, bottom).
10
11
12
DISCUSSION
Silvicultural techniques can be used effectively to promote development of old-growth
13
structural characteristics in northern hardwood and mixed northern hardwood-conifer forests.
14
Both the uneven-aged and SCE approaches tested maintain high levels of post-harvest structure
15
and canopy cover. However, SCE has the potential to maintain, enhance, or accelerate develop
16
of CWD, vertical differentiation of the canopy, basal area, aboveground biomass, large tree
17
recruitment, and other structural attributes to a greater degree. In addition, SCE results in a
18
rotated sigmoid diameter distribution that appears sustainable at least over 50 years, and
19
consequently reallocates growing space and aboveground structure into larger size classes. This
20
contributes to enhanced large tree structure, foliage biomass, and associated canopy complexity.
21
The higher levels of structural retention associated with SCE are indicative of lower intensity,
22
minimal impact forestry practices (McEvoy 2004).
20
Keeton 21
1
2
3
Accelerating rates of stand development
Both SCE and conventional uneven-aged treatments will result in accelerated tree growth
4
rates according to simulation projections. Since the conventional treatments retained less basal
5
area, their comparatively higher projected basal area increment is consistent with previous
6
research on growth responses to stocking density (Leak et al. 1987). It is important to note,
7
however, that neither the NE-TWIGS nor the FVS model is spatially explicit. Inter-stem
8
competition and tree growth rates are simulated as a function only of tree size and total stand
9
stocking (Bush 1995), and not the spatial position of individual trees. The models do not,
10
therefore, capture the effects of crown release employed in SCE. Crown release is likely to
11
partially arrest or dampen declining growth rates in older, dominant trees, leading to rates of
12
large tree development that are 50 to 100 % faster compared to no release scenarios (Singer and
13
Lorimer 1997). Consequently spatially explicit simulation modeling, capturing the horizontally
14
complex residual pattern produced by SCE treatment, may result in different, and possibly
15
enhanced, developmental projections for SCE. Regardless, an important effect of SCE is the
16
promotion of large tree recruitment. This process is impaired under conventional treatments that
17
include maximum diameter limits and negative exponential diameter distributions that provide
18
lower large tree recruitment potential.
19
Projected basal area and aboveground biomass are higher after 50 years of development
20
under SCE (in comparison to conventional treatments) due, primarily, to elevated post-harvest
21
structural retention. Substantial increases in rates of production over that likely without
22
treatment also contribute to these changes. However, while structural development in SCE units
23
is projected to surpass controls units, none of treatments are likely to develop basal areas or
21
Keeton 22
1
aboveground biomass exceeding levels that would have accumulated without treatment. This
2
provides a strong argument for passive restoration, rather than silvicultural manipulation, as an
3
approach for ultimately developing higher levels of structure associated with these parameters.
4
However, that conclusion does not account for the accelerated rates of large tree recruitment,
5
reallocation of basal area, and associated structural complexity projected for SCE. In this
6
respect, active silvicultural manipulation may offer some advantages.
7
It is also clear that the conventional uneven-aged approaches were less effective at
8
retaining and promoting late-successional/old-growth structural development. The results do
9
show that conventional approaches can be modified to reallocate basal area to some degree and
10
provide a relatively high degree of biomass and vertical complexity. However, uneven-aged
11
practices that employ maximum diameter limits impair and, in fact, reduce large tree recruitment
12
potential based on the results. In comparison, SCE ultimately results in higher levels of
13
structural complexity across a range of parameters. Therefore, SCE offers a useful alternative
14
for landowners interested in both low-intensity harvesting and promotion of old-growth
15
characteristics.
16
Coarse woody debris, both standing and downed plays a critical role in northern-
17
hardwood systems as habitat and in other ecosystem processes (Tyrrell and Crow 1994a, Hunter
18
1999). SCE resulted in significantly elevated CWD densities and volumes. However, it remains
19
uncertain whether this effect will persist until natural recruitment rates increase, or, alternatively,
20
whether CWD enhancement in mature stands has only transient or short-term management
21
applications. While long-term CWD persistence and accumulation dynamics are uncertain, it is
22
likely than decay class distributions will again shift over time towards better-decayed material.
23
This would render silviculturally enhanced CWD more biologically available in habitat and
22
Keeton 23
1
nutrient processes in the future (Gore and Patterson 1985, Tyrrell and Crow 1994a, Goodburn
2
and Lorimer 1998).
3
4
5
Applications for Structural Complexity Enhancement
SCE may have a variety of useful applications, ranging from old-growth restoration, to
6
riparian management, to low-intensity timber management and late-successional wildlife habitat
7
enhancement (Table 2). These will depend greatly on economic feasibility under a variety of site
8
quality, product, and market conditions (Niese and Strong 1992), which is the subject of an on-
9
going investigation (Keeton and Troy 2005). Depending on the specific application, SCE could
10
be employed to varying degrees or to a more limited extent. For instance, where timber
11
production is emphasized, a subset of SCE elements might be used. Other elements might be
12
avoided or employed at a lesser intensity. In this scenario, multiple stand entries would be
13
expected, but late-successional structural development would be lower compared to full SCE
14
implementation. Such an approach, however, would allow forest managers to build some degree
15
of old-growth associated structure into actively managed stands, while maintaining greater
16
timber management flexibility.
17
Intermediate applications of SCE could be employed where low-intensity, ecologically
18
sensitive management is required within riparian areas. Managing for forest structural
19
complexity along freshwater streams would be useful where the associated influences on in-
20
stream aquatic habitat conditions are desired (Keeton et al. 2005). Managers of protected areas,
21
conversely, might choose to employ SCE more fully as a restorative approach. They might enter
22
a stand once or twice, thereafter allowing accelerated successional processes to take over. The
23
degree of implementation and the number of stand entries will thus vary by application (Table 2).
24
23
Keeton 24
1
2
Silvicultural flexibility in managing stand structure
Forest managers have the flexibility to manage for a wide range of structural
3
characteristics and associated ecosystem functions in northern-hardwood conifer systems. Late-
4
successional/old-growth characteristics can be promoted through a variety of silvicultural
5
approaches, including those investigated in this and other studies (e.g. Singer and Lorimer 1997).
6
Uneven-aged systems provide some, but not all, late-successional structural characteristics or
7
provide them to a more limited extent based on the results reported here. Residual basal area,
8
maximum diameter, and q-factor can be modified singly or collectively, resulting in greater
9
structural retention. However, maximum diameter limits significantly retard the potential for
10
large tree (live and dead) recruitment based on the results. Stand development is thus
11
continuously truncated by multiple uneven-aged cutting entries.
12
Cutting to alternative diameter distributions provides another way to manipulate stand
13
structure for a range of objectives, including old-growth structural development. The results
14
show that SCE’s variable q-factor marking guide can be used to successfully achieve a rotated
15
sigmoid diameter distribution. In addition, the results indicate that this distribution is sustainable
16
in terms of growth and recruitment across size classes. Density-dependent mortality in middle
17
size classes does not appear to self-thin a stand to a negative exponential distribution, at least
18
over the time frame investigated in this study. This may due to reduced (or non-constant)
19
mortality rates in these size classes (O’Hara 1998, Goodburn and Lorimer 1999). For
20
silviculturalists this offers the flexibility to manage for stand structures in which basal area
21
distributions are shifted towards medium and larger size classes. They would also maintain
22
sufficient densities of smaller stems to ensure recruitment across all size classes. This would
23
provide benefits associated with larger dimension timber, large tree habitat functions, and carbon
24
Keeton 25
1
storage associated with greater biomass. Unconventional prescriptive diameter distributions,
2
such as the rotated sigmoid, combined with higher levels of residual basal area, very large (or no)
3
maximum diameters, and crown release are an effective alternative for retaining high levels of
4
post-harvest structure and for promoting accelerated stand development.
5
6
7
CONCLUSION
Foresters can manage for late-successional/old-growth structure where desired in
8
northern hardwood-conifer ecosystems. Active silvicultural manipulations as well as passive
9
approaches provide alternatives for achieving this objective. Restorative, low-intensity
10
approaches will not necessarily replace more traditional approaches. Rather, they provide
11
another tool for our silvicultural toolbag. Managing for late-successional forest characteristics
12
can be actively employed as an element of structure-based forestry and comprehensive
13
ecosystem management.
14
15
ACKNOWLEDGEMENTS
16
This research was supported by grants from the USDA CSREES National Research Initiative,
17
the Vermont Monitoring Cooperative, the Northeastern States Research Cooperative, and the
18
USDA McIntire-Stennis Forest Research Program. Assistance with NED software was provided
19
by Dr. Mark Twery, USDA Forest Service, Northeastern Research Station. Computer
20
administration support was provided by Jonathan Trigaux, University of Vermont. The author is
21
grateful to Donald R. Tobi, University of Vermont Entomology Research Laboratory, for
22
assistance throughout this project.
23
25
Keeton 26
1
LITERATURE CITED
2
Berg, D.R. 1995. Riparian silvicultural system design and assessment in the Pacific Northwest Cascade
3
4
Mountains, USA. Ecological Applications 5:87-96.
Beese, W.J., B.G. Dunsworth, and N.J. Smith. 2005. Variable-retention adaptive management experiments:
5
testing new approaches for managing British Columbia’s coastal forests. Page 55-64 in: C.E. Peterson
6
and D.A. Maguire (eds.). Balancing ecosystem values: innovative experiments for sustainable forestry.
7
General Technical Report PNW-GTR-635. USDA Forest Service, Pacific Northwest Research Station,
8
Portland, OR. 389 pp.
9
10
11
12
13
14
15
16
17
18
19
Bush, R.R. 1995. Northeastern TWIGS variant of the Forest Vegetation Simulator. Forest Management
Service Center. Available online at http:/ftp.fs.fed.us/pub/fmsc/fvs/doc/overviews/nevar.txt
Cogbill, C.V. 2000. Vegetation of the presettlement forests of northern New England and New York. Rhodora
102:250-276.
Dahir, S.E. and C.G. Lorimer. 1996. Variation in canopy gap formation among developmental stages of
northern hardwood stands. Canadian Journal of Forest Research 26:1875-1892.
Davis, M.B. 1996. Extent and location. Pages 18-34 in: M.B.Davis. (ed.). Eastern Old-Growth Forests:
Prospects for Rediscovery and Recovery. Island Press, Washington, D.C. 383 pp.
DeGaaf, R.M. and M. Yamasaki. 2003. Options for managing early-successional forest and shrubland bird
habitats in the northeastern United States. Forest Ecology and Management 185:179-191.
Dixon, Gary E. (comp.). 2003. Essential FVS: A User's Guide to the Forest Vegetation Simulator. Internal
20
Rep. Fort Collins, CO: U. S. Department of Agriculture, Forest Service, Forest Management Service
21
Center. 193p.
22
Dunwiddie, P., D. Foster, D. Leopold, and R.T. Leverett. 1996. Old-growth forests of southern New
23
England, New York, and Pennsylvania. Pages 126-143 in: M.B.Davis. (ed.). Eastern Old-Growth
24
Forests: Prospects for Rediscovery and Recovery. Island Press, Washington, D.C. 383 pp.
26
Keeton 27
1
(FEMAT) Forest Ecosystem Management Assessment Team. 1993. Forest ecosystem management: an
2
ecological, economic, and social assessment. USDA Forest Service, USDI Bureau of Land Management,
3
Government Printing Office, Washington, DC.
4
5
6
Foster, D.R. and J.D. Aber (eds.). 2004. Forests in Time: the Environmental Consequences of 1,000 years of
change in New England. Yale University Press, New Haven, CT.
Franklin, J.F., D.R. Berg, D.A. Thornburgh, and J.C. Tappeiner. 1997. Alternative silvicultural approaches
7
to timber harvesting: variable retention harvest system. Pages 111-140 in: K.A. Kohm and J.F. Franklin
8
(eds.). Creating a Forestry for the 21st Century: The Science of Ecosystem Management. Island Press,
9
Washington, DC. 475 pp.
10
11
Franklin, J.F., L.A. Norris, D.R. Berg, and G.R. Smith. 1999. The history of DEMO: an experiment in
regeneration harvest of northwestern forest ecosystems. Northwest Science 73: 3-11.
12
Franklin, J.F., T.A. Spies, R. Van Pelt, A. Carey, D. Thornburgh, D.R. Berg, D. Lindenmayer, M. Harmon,
13
W.S. Keeton, D.C. Shaw, K. Bible, and J. Chen. 2002. Disturbances and the structural development of
14
natural forest ecosystems with some implications for silviculture. Forest Ecology and Management
15
155:399-423.
16
17
Goodburn, J.M. and C.G. Lorimer. 1998. Cavity trees and coarse woody debris in old-growth and managed
northern hardwood forests in Wisconsin and Michigan. Canadian Journal of Forest Research 28:427-438.
18
Goodburn, J.M. and C.G. Lorimer 1999. Population structure in old-growth and managed northern
19
hardwoods: an examination of the balanced diameter distribution concept. Forest Ecology and
20
Management 118: 11-29.
21
22
Gore, J.A. and W.A. Patterson. 1985. Mass of downed wood in northern hardwood forests in New
Hampshire: potential effects of forest management. Canadian Journal of Forest Research 16:335-339.
23
Harrington, C.A., S.D. Roberts, and L.C. Brodie. 2005. Tree and understory responses to variable-density
24
thinning in western Washington. Page 97 – 106 in: C.E. Peterson and D.A. Maguire (eds.). Balancing
25
ecosystem values: innovative experiments for sustainable forestry. General Technical Report PNW-GTR-
26
635. USDA Forest Service, Pacific Northwest Research Station, Portland, OR. 389 pp.
27
Keeton 28
1
2
Hilt, D.E. and R.M. Teck. 1989. NE-TWIGS: An Individual Tree Growth and Yield Projection System for the
Northeastern United States. The Compiler 17(2):10-16.
3
Hunter, M. L.. 1989. What constitutes an old-growth stand? Journal of Forestry: 33-35.
4
Hunter, M.L. Jr. (ed.). 1999. Maintaining Biodiversity in Forested Ecosystems. Cambridge University Press,
5
6
7
Cambridge, UK.
Hunter, M. L. and A.S. White. 1997. Ecological thresholds and the definition of old-growth forest stands.
Natural Areas Journal 17: 292-296.
8
Hunter, W.C., D.A. Buechler, R.A. Canterbury, J.L. Confer, and P.B. Hamel. 2001. Conservation of
9
disturbance dependent birds in eastern North America. Wildlife Society Bulletin 29: 425-439.
10
11
12
13
14
15
Jenkins, J. C., D.C. Chomnacky, L.S. Heath, R.A. Birdsey. 2003. National scale biomass estimators for
United States tree species. Forest Science 49: 12-35
Kaluzny, S.P., S.C. Vega, T.P. Cardoso, and A.A. Shelly. 1998. S+ Spatial Stats: User’s Manual for Windows
and Unix. Springer, New York, NY. 325 pp.
Keddy, P.A. and C.G. Drummond. 1996. Ecological properties for the evaluation, management, and
restoration of temperate deciduous forest ecosystems. Ecological Applications 6:748-762.
16
Keeton, W.S. 2005. Managing for old-growth structure in northern hardwood forests. Pages 107-117 in:
17
C.E. Peterson and D.A. Maguire (eds.). Balancing ecosystem values: innovative experiments for
18
sustainable forestry. General Technical Report PNW-GTR-635. USDA Forest Service, Pacific Northwest
19
Research Station, Portland, OR. 389 pp.
20
Keeton, W.S., C.E. Kraft, D.R. Warren, and A.A. Millward. 2005. Effects of old-growth riparian forests on
21
Adirondack stream systems. Pages 39-43 in: K.P. Bennet (tech. ed.). Moving toward sustainable
22
forestry: lessons from old-growth forests. Proceedings of the 6th eastern old-growth conference,
23
Moultonborough, NH. University of New Hampshire Cooperative Extension Natural Resource Network
24
Report.
25
26
Keeton, W.S., A.M. Strong, D.R. Tobi, and S. Wilmot. 2001. Experimental test of structural complexity
enhancement in northern hardwood-hemlock forests. Pages 33-40 in: J.M. Hagan (ed.). Forest Structure:
28
Keeton 29
1
A Multi-Layered Conversation. Proceedings of the Forest Information Exchange, October 25, 2001,
2
Orono, Maine. 78 pp.
3
Keeton, W.S. and A. R. Troy. 2005. Balancing ecological and economic objectives while managing for late-
4
successional forest structure (abstract). Page 72 in: Ecologisation of economy as a key prerequisite for
5
sustainable development. Proceedings of the international conference, Sept. 22 –23, 2005, National
6
Forestry Engineering University of Ukraine, L’viv, Ukraine.
7
King, D.I., R.M. DeGraaf, and C.R. Griffin. 2001. Productivity of early-successional shrubland birds in
8
clearcuts and groupcuts in an eastern deciduous forest. Journal of Wildlife Management 65: 345-350
9
10
11
12
Krebs, C.J. 1999. Ecological Methodology (2nd edition). Addison Wesley Longman, Inc. Menlo Park, CA.
620 pp.
Leak, W.B. 1996. Long-term structural change in uneven-aged northern hardwoods. Forest Science 42:160165.
13
Leak, W.B. 2002. Origin of sigmoid diameter distributions. USDA Forest Service Research Paper NE-718.
14
Leak, W.B., D.S. Solomon, and P.S. DeBald. 1987. Silvicultural guide for northern hardwood types in the
15
16
17
18
19
20
21
22
23
24
Northeast (revised). USDA Forest Service Research Paper NE-603. 36 pp.
Lindenmayer, D.B. and J.F. Franklin. 2002. Conserving Forest Biodiversity: A Comprehensive Multiscaled
Approach. Island Press, Washington, D.C. 351 pp.
Litvaitis, J.A. 2001. Importance of early-successional habitats to mammals in eastern forests. Wildlife
Society Bulletin 29: 466-473.
Lorimer, C.G. 2001. Historical and ecological roles of disturbance in eastern North American forests: 9,000
years of change. Wildlife Society Bulletin 29:425-439.
Lorimer, C.G. and L.E. Frelich. 1994. Natural disturbance regimes in old-growth northern hardwoods,
implications for restoration efforts. Journal of Forestry 92:33-38
Lorimer, C.G. and A.S. White. 2003. Scale and frequency of natural disturbances in the northeastern U.S.,
25
implications fore early-successional forest habitats and regional age distributions. Forest Ecology and
26
Management 185: 41-64.
29
Keeton 30
1
Marshall, D.D. and R.O. Curtis. 2005. Evaluations of silvicultural options for harvesting Douglas-fir young-
2
growth production forests. Pages 119-126 in: C.E. Peterson and D.A. Maguire (eds.). Balancing
3
ecosystem values: innovative experiments for sustainable forestry. General Technical Report PNW-GTR-
4
635. USDA Forest Service, Pacific Northwest Research Station, Portland, OR. 389 pp.
5
McEvoy, T.J. 2004. Positive Impact Forestry. Island Press, Washington, D.C. 268 pp.
6
McGee, G.G., D.J. Leopold, and R.D. Nyland. 1999. Structural characteristics of old-growth, maturing, and
7
8
9
10
11
12
partially cut northern hardwood forests. Ecological Applications 9:1316-1329.
McGaughey, R.J. 1997. Visualizing forest stand dynamics using the stand visualization system. American
Society of Photogrammetry and Remote Sensing 4:248-257.
Mitchell, R.J., B.J. Palik, and M.L. Hunter. 2002. Natural disturbance as a guide to silviculture. Forest
Ecology and Management 155:315-317.
Mladenoff, D.J. and J. Pastor. 1993. Sustainable forest ecosystems in the northern hardwood and conifer
13
forest region: concepts and management. Pages 145-180 in: G.H. Aplet, N. Johnson, J.T. Olson, and V.A.
14
Sample (eds.). Defining Sustainable Forestry. Island Press, Washington, DC 328 pp.
15
16
Niese, J.N. and T.F. Strong. 1992. Economic and tree diversity trade-offs in managed northern hardwoods.
Canadian Journal of Forest Research 22:1807-1813.
17
Nyland, R.D. 1998. Selection system in northern hardwoods. Journal of Forestry 96:18-21.
18
O’Hara, K.L. 1998. Silviculture for structural diversity: a new look at multi-aged systems. Journal of Forestry
19
20
96:4-10.
Peterson, C.E. and D.A. Maguire (eds.). 2005. Balancing ecosystem values: innovative experiments for
21
sustainable forestry. General Technical Report PNW-GTR-635. USDA Forest Service, Pacific Northwest
22
Research Station, Portland, OR. 389 pp.
23
Ripley, B.D. 1981. Spatial Statistics. John Wiley and Sons, Inc. New York, NY. 252 pp.
24
Seymour, R.S. 2005. Integrating natural disturbance parameters into conventional silvicultural systems:
25
experience from the Acadian forest of northeastern North America. Pages 41-49 in: C.E. Peterson and
26
D.A. Maguire (eds.). Balancing ecosystem values: innovative experiments for sustainable forestry.
30
Keeton 31
1
General Technical Report PNW-GTR-635. USDA Forest Service, Pacific Northwest Research Station,
2
Portland, OR. 389 pp.
3
Seymour, R.S., A.S. White, and P.H. deMaynadier. 2002. Natural disturbance regimes in northeastern North
4
America: evaluating silvicultural systems using natural scales and frequencies. Forest Ecology and
5
Management 155:357-367.
6
7
8
9
10
11
Singer, M.T. and C.G. Lorimer. 1997. Crown release as a potential old-growth restoration approach in
northern hardwoods. Canadian Journal of Forest Research 27:1222-1232.
Stehman, S.V. and W.S. Overton. 1996. Spatial sampling. Pages 17-30 in: S.L. Arlinghaus (ed.). Practical
Handbook of Spatial Statistics. CRC Press, Boca Raton, FL. 307 pp.
Swanson, F.J. and J.F. Franklin. 1992. New forestry principles from ecosystem analysis of Pacific Northwest
forests. Ecological Applications 2:262-274.
12
Trombulak, S.C. 1996. The restoration of old-growth: why and how. Pages 305-320 in: M.B.Davis. (ed.).
13
Eastern Old-Growth Forests: Prospects for Rediscovery and Recovery. Island Press, Washington, D.C.
14
383 pp.
15
Twery, M.J., P.D. Knopp, S.A. Thomasma, H. M. Rauscher, D.E. Nute, W.D. Potter, F. Maier, J. Wang, M.
16
Dass, H. Uchiyama, A. Glende, and R.E. Hoffman. 2005. NED-2: A decision support system for
17
integrated forest ecosystem management. Computers and Electronics in Agriculture 49: 24-43.
18
Tyrrell, L.E. and T.R. Crow. 1994a. Dynamics of dead wood in old-growth hemlock-hardwood forests of
19
northern Wisconsin and northern Michigan. Canadian Journal of Forest Research 24:1672-1683.
20
Tyrrell, L.E. and T.R. Crow. 1994b. Structural characteristics of old-growth hemlock-hardwood forests in
21
22
23
24
relation to stand age. Ecology 75:370-386.
Woods, K.D. and C.V. Cogbill. 1994. Upland old-growth forests of Adirondack Park, New York, USA.
Natural Areas Journal 14:241-257.
Ziegler, S.S. 2000. A comparison of structural characteristics between old-growth and post-fire second-
25
growth hemlock-hardwood in Adirondack Park, New York, U.S.A.. Global Ecology and
26
Biogeography 9:373-389.
31
Keeton 32
1
Table 1. Structural objectives and the corresponding silvicultural techniques used to promote those
2
attributes in Structural Complexity Enhancement.
3
Structural Objective
Multi-layered canopy
Elevated large snag densities
Elevated downed woody debris densities and
volume
Variable horizontal density
Re-allocation of basal area to larger diameter
classes
Accelerated growth in largest trees
Silvicultural Technique
 Single tree selection using a target
diameter distribution
 Release advanced regeneration
 Establish new cohort
 Girdling of selected medium to large sized,
low vigor trees
 Felling and leaving, or
 Pulling over and leaving
 Harvest trees clustered around “release
trees”
 Variable density marking
 Rotated sigmoid diameter distribution
 High target basal area
 Maximum target tree size set at 90 cm dbh
 Full and partial crown release of largest,
healthiest trees
4
5
6
Table 2. Potential applications of Structural Complexity Enhancement as an approach for
7
incorporating old-growth structure into managed forests.
8
9
Application
# Entries
Late-Successional
Structural Development
Old-growth
promotion
One or possibly
two entries
High
Riparian
management
Single or multiple
Moderate to high
Timber emphasis
Multiple
Low to moderate
10
11
32
Keeton 33
1
List of Figures
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Figure 1. Output of the Stand Visualization System (SVS) contrasting a single-tree selection unit (above)
and a Structural Complexity Enhancement (SCE) unit (below) at the Mount Mansfield study area. Shown
are images of pre- and post-harvest stand structure for 1 ha. blocks. Shaded circles represent tree crowns
(with species-specific coloration) seen from a simulated aerial view. Note the high degree of post-harvest
structure (e.g. basal area and stem density), canopy closure, vertical complexity, and downed log densities
in the SCE unit. Note the similar, though lower, degree of structural retention in the single-tree selection
unit.
Figure 2. Pre-harvest, post-harvest, and target diameter distributions for the four SCE units, two at Mt.
Mansfield (above) and two at the University of Vermont’s Jericho Research Forest (below). Log
transformed post-harvest and target distributions are compared (top portion of graphs) using the
Kolmogorov-Smirnov two-sample goodness of fit test. There were no statistically significant differences.
Thus, the post-harvest distributions achieved the target rotated sigmoid distribution.
Figure 3. Downed log response to treatments. Shown are percent change from pre-harvest levels and
absolute change in volume (m3/ha). Error bars are ±1 standard error of the mean.
Figure 4. Mean pre and post-treatment downed log (> 10 cm) decay class distributions for SCE units.
Cumulative frequency distributions were significantly different based on Kolmogorov-Smirnov twosample goodness of fit tests. Decay class (DC) 1 is least decomposed while DC 5 is most decomposed.
Note the large input of fresh logs (DC 1) following treatment.
Figure 5. Results of NE-TWIGS stand development modeling. Shown are 50 year projections of posttreatment cumulative basal area (live tree) production (top) and normalized scenarios (post-harvest minus
pre-harvest cumulative basal area increment, bottom). Error bars are ±1 standard error of the mean.
Figure 6. Projected diameter distributions for 50 years into the future based on FVS simulations. Mean
distributions developed after structural complexity enhancement (SCE) and conventional uneven-aged
treatments are compared against the target or desired future condition prescribed for SCE. Note the close
fit with the target shown by SCE, and the disparity evident for the conventional treatment.
Figure 7. Projected live tree basal area distributions based on FVS simulations of stand development.
Mean distributions are shown for SCE units (top) and conventional uneven-aged units (bottom) projected
both with and without treatment. Error bars are ±1 standard error of the mean.
33
Keeton 34
1
Figure 1
2
3
Single-Tree Selection Unit
4
5
Post-Harvest
Pre-Harvest
6
7
8
9
10
11
12
Structural Complexity Enhancement Unit
Pre-Harvest
Post-Harvest
13
14
15
16
17
18
19
20
21
22
23
34
Keeton 35
1
Figure 2
3
Pre Harvest
Post Harvest
Target
Log of Target
Log of Post-harvest (Smoothed)
Log of Post Harvest
3
400
5
Trees per Hectare
350
4
2.5
2
300
Dmax = 0.036
250
Dcrit, 0.05 = 0.427
200
No Sign. Diff.
1.5
1
150
6
600
3
Pre Harvest
Post Harvest
Target
Log of Target
Log of Post-harvest (Smoothed)
Log of Post Harvest
500
Trees per Hectare
450
2.5
400
2
Dmax = 0.039
300
1.5
Dcrit, 0.05 = 0.425
No Sign. Diff.
200
1
100
0.5
Log of Trees per Hectare
500
Log of Trees per Hectare
2
100
0.5
50
7
0
0
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
0
0
90
10
15
20
25
30
35
Diameter Class (cm)
40
45
50
55
60
65
70
75
80
85
90
Diameter Class (cm)
8
3
Pre Harvest
Post Harvest
Target
Log of Target
Log of Post-harvest (Smoothed)
Log of Post Harvest
11
12
300
2
D max = 0.022
250
D crit, 0.05 = 0.421
1.5
No Sign. Diff.
200
150
1
0.5
0
2
D max = 0.024
200
D crit, 0.05 = 0.420
1.5
No Sign. Diff.
150
1
100
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
0.5
50
0
0
10
14
250
2.5
100
50
13
3
Pre Harvest
Post Harvest
Target
Log of Target
Log of Post-harvest (Smoothed)
Log of Post Harvest
300
2.5
Trees per Hectare
10
Trees per Hectare
350
Log of Trees per Hectare
400
9
350
0
10
90
Diameter Class (cm)
15
20
25
30
35
40
45
50
55
60
Diameter Class (cm)
15
16
17
18
19
20
21
22
23
35
65
70
75
80
85
90
Log of Trees per Hectare
450
Keeton 36
1
2
Figure 3
3
4
5
6
7
180
8
160
9
11
12
13
140
3
Percent or Volume (m /ha)
10
120
100
ANOVA: P = 0.002
Bonferroni Multiple Comparisons:
SCE ¹Control (P < 0.05)
SCE ¹. Uneven-aged (P < 0.05)
Uneven-aged ¹Control (P > 0.05)
80
60
14
40
15
20
16
0
17
Percent Change
Volume Increase
Controls
Conventional
Uneven-Aged
18
36
Structural Complexity
Enhancement
Keeton 37
1
2
Figure 4
3
4
30
D max = 0.301
6
25
3
0.9
D crit, 0.05 = 0.255
0.8
P < 0.05
Downed Log Volume (m /ha)
7
1
0.7
20
0.6
15
0.5
0.4
SCE Pre-harvest
10
0.3
SCE Post-harvest
5
0.2
Pre-harvest Cumulative
0.1
Post-harvest Cumulative
0
0
1
2
3
Decay Class
37
4
5
Cumulative Frequency
5
Keeton 38
1
Figure 5
2
40
35
2
4
Cumulative Basal Area Increment (m /ha)
3
5
6
7
8
9
30
25
Structural Complexity Enhancement
20
10
Conventional Uneven-Aged
Control Units
11
15
0
5
10
15
12
20
25
30
Projected Years
35
40
45
50
35
40
45
50
6
Structural Complexity Enhancement
15
16
17
18
2
14
Cumulative Basal Area Increment (m /ha)
13
Conventional Uneven-aged
5
4
3
2
1
19
0
20
0
5
10
15
20
25
30
Projected Years
21
22
23
24
38
Keeton 39
1
Figure 6
2
70
3
4
60
Target Rotated Sigmoid
Uneven-aged Treatments (Projected)
SCE Treatments (Projected)
5
Density (stems/ha.)
50
40
30
20
10
0
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95
Diameter Class (cm)
39
Keeton 40
1
Figure 7
2
3
4
14
5
No Treatment (Year 50)
SCE Treatment (Year 50)
Percent Basal Area
12
10
8
6
4
2
0
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95
Diameter Class (cm)
14
No Treatment (Year 50)
Uneven-aged Treatment (Year 50)
Percent Basal Area
12
10
8
6
4
2
0
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95
Diameter Class (cm)
40