TEMPORAL EFFECTS OF MECHANICAL TREATMENT ON WINTER
MOOSE BROWSE IN SOUTH-CENTRAL ALASKA

Sharon Smythe1, Dana Sanchez1, and Ricardo Mata-Gonzalez2

1Fisheries and Wildlife Department, Oregon State University, 104 Nash Hall, Corvallis, Oregon 97331,
USA; 2Animal and Rangeland Sciences, Oregon State University, 120 Withycombe, Corvallis, Oregon
97331, USA

ABSTRACT: Sites containing winter browse species utilized by moose on the Copper River Delta of
south-central Alaska were mechanically treated (hydraulic-axed) to counteract possible earthquake-
related increases in less-preferred forage species, and to measure treatment effects on biomass, height,
nutritional quality (crude protein, lignin, and tannin), utilization, and snow burial on preferred (willow
[Salix spp.]) and less-preferred forage species (sweetgale [Myrica gale], cottonwood [Populus tricho-
carpa], and alder [Alnus viridis sinuata]) within 3 winter scenarios (mild, moderate, and severe). Sites
were treated in 4 winters (1990–1992, 2008, 2010, and 2012) within 5 stand types in 20 sites varying
from 0.9–63.4 ha. We found few significant differences in biomass, height, nutritional quality, utiliza-
tion, and snow burial relative to controls. However, our ability to detect differences may have been
limited by sample size (n = 1–9), as visual comparison suggests hydraulic-axing may be an effective
method for increasing willow biomass while reducing alder biomass without influencing nutritional
quality. However, because treated willows were shorter than untreated willows, treatment may result
in less preferred forage for moose in severe winters with deep snow. Our results have implications
for habitat management of moose but further research is needed to determine incremental and long-
term effects of treatment on willow growth and productivity.

ALCES VOL. 51: 135–147 (2015)

Key words: Alaska, Alces alces gigas, Alnus viridis sinuata, Copper River Delta, forage biomass,
hydraulic axing, Myrica gale, nutrition, Populous trichocarpa, Salix spp.

Since many deer species in North Amer-
ica rely on early-successional forage, habitat
management efforts commonly delay forest
succession through mechanical treatment
via shearing, crushing, or axing of overstory
vegetation (Scotter 1980, Hundertmark
et al. 1990, Renecker and Schwartz 1997,
Thompson and Stewart 1997, Suring and
Sterne 1998). Mechanical treatment (hydraulic-
axing) was applied on a limited scale to
increase availability of preferred winter forage
for an Alaskan moose (Alces alces gigas)
population on the Copper River Delta (CRD,
Stephenson et al. 1998), the location of this
study (Fig. 1).

Moose were introduced to the CRD from
1949–1958 to establish a harvestable pop-
ulation, having likely been excluded by
topography (MacCracken et al. 1997). With
a potential range encompassing >54,000 ha,
the more managed and hunted western sub-
population has since grown to >600 animals
(C. Westing, Alaska Department of Fish and
Game, unpublished data). However, intense
winter winds through the Copper River
canyon, variable snow depths, and snow
drifting can restrict winter range access to
4,800–12,900 ha (MacCracken et al. 1997,
Stephenson et al. 2006). This seasonal effect
constrains accessible browse and has

Sharon Smythe, 104 Nash Hall. Oregon State University, Corvallis, Oregon 97331, USA,
sharonsmythe77@gmail.com

135

mailto:sharonsmythe77@gmail.com


historically been thought to limit adult
moose survival (Regelin et al. 1985,
Schwartz et al. 1988, MacCracken et al.
1997). Furthermore, a 9.2 magnitude earth-
quake in 1964 uplifted the area by 1.0–4.0 m
(Grantz et al. 1964, Ferrians 1966, Plafker
1969, Stover and Coffman 1993), initiating
changes in hydrology, soil salinity, and vege-
tation, including an acceleration of succession
in some stands to stages with increased pro-
duction of less-preferred browse (Thilenius
1990, 2008).

Managers are concerned that the com-
bined effects of winter range restrictions
and earthquake-initiated vegetation changes
might limit the performance or persis-
tence of this locally important population

(MacCracken et al. 1997, Stephenson et al.
2006). As a result, the USDA Forest Service
Cordova Ranger Station initiated exper-
imental treatments of moose habitat with
hydraulic-axing machines (hereafter hydro-
axing) which use rotary axes to cut down
and splinter trees or shrubs up to 15 cm in
diameter (Stephenson et al. 1998). Initial
treatment plots were cut in 1990–1992,
followed by additional plots in 2008, 2010,
and 2012 (M. Burcham, USDA Forest
Service Cordova Ranger District, personal
communication, Stephenson et al. 1998).
Because wintering CRD moose depend on
5 willow species (feltleaf willow, Barclays
willow, undergreen willow, Hookers willow,
and Sitka willow [S. alexensis, S. barclayi,

Fig. 1. Sites mechanically-treated (hydraulic-axed) in 1990–1992, 2008, 2010, and 2012 on the
west Copper River Delta of south-central Alaska to improve the availability of willow forage for
wintering moose.

136

MECHANICAL TREATMENT OF MOOSE BROWSE – SMYTHE ET AL. ALCES VOL. 51, 2015



S. commutata, S. hookeri, S. sitchensis,
respectively]), and only occasionally on
black cottonwood (Populus trichocarpa),
sweetgale (Myrica gale), and Sitka alder
(Alnus viridis sinuata) (MacCracken et al.
1997), treatments have focused on increas-
ing the willow component of stands. In the
Kenai National Forest willows re-sprouted
following mechanical treatment whereas
mature red alder (A. rubra) experienced high
mortality (Oldemeyer and Regelin 1980,
Harrington 1984). Thus, most treatments on
the CRD were sited on alder-dominated
stands with remnant willow components,
though spruce-cottonwood-, sweetgale-, and
willow-dominated stands have also been
treated (Table 1).

Stephenson et al. (1998) evaluated the
success of the initial (1990–1992) treatments
1–3 years post-treatment, and found that
alder mass generally declined and Sitka wil-
low mass increased in treated sites. How-
ever, responses in biomass and utilization
by other browse species varied by stand or
were statistically precluded by sample size
(Stephenson et al. 1998). In addition, mean

height of browse in treated stands was often
less than in controls, and snow-buried
browse varied by location, treatment, and
stand type. It was hypothesized that Sitka
willow at full height (5 m) in alder- and
willow-dominated stands would be especially
important in winters with deep snow and
heavy drifting. Therefore, it is possible that
extensive treatment might increase the prev-
alence of shorter willows, coincidentally
limiting browse available to moose in severe
winters. However, hydro-axing effects in this
system have not been studied beyond the
first 3 years post-treatment.

Our objectives were to 1) evaluate
species-specific and time-since-treatment re-
sponses of available biomass, height, nutri-
tional quality, and moose utilization of winter
browse species to hydro-axing 1, 3, 5, and
23 years post-treatment, and 2) estimate
how biomass availability within treated sites
varies with snow depth (winter severity). Our
results will assist managers in assessing the
relative benefits of hydro-axing to maintain
willow availability for moose in a dynamic
ecosystem.

Table 1. Characteristics of mechanically treated (hydraulic-axed) sites sampled (2012–2013) for moose
browse species on the western region of the Copper River Delta, Alaska, including site age (years since
treatment), control stand type, soil type, area (ha), and sampling replicates. Soil types include AST =
alluvium and stream terrace deposits, OPN = glacial outwash plains, nonforested, and GM =
undifferentiated glacial moraines (Davidson and Harnish 1978).

Age (yr) Winter Treated Control Stand Types Soil Type Replicates (n) Size (ha)

1 2012–2013 Spruce-cottonwood AST 1 57.9
Alder AST 2 23.9, 63.4

3 2010–2011 Alder OPN 1 3.4
Sweetgale AST 3 8.0, 3.4, 5.7

5 2008–2009 Spruce-cottonwood GM 2 10.7, 7.6
Willow AST 2 11.8, 10.5

22–23 1990–1991 & 1991–1992 Spruce-hemlock OPN 2 0.9, 1.5
Alder AST/OPN 2 3.0, 2.2

Alder-willow AST 2 0.9, 4.9

Willow AST 1 1.5

Sweetgale OPN 2 2.6, 0.8

ALCES VOL. 51, 2015 SMYTHE ET AL. – MECHANICAL TREATMENT OF MOOSE BROWSE

137



METHODS
Study Area

The CRD lies within the Chugach Na-
tional Forest and is bordered by 3 glaciers,
the Chugach Mountain Range, and the Gulf
of Alaska (Fig. 1). As the largest continuous
wetland in the Pacific Northwest, it extends
120 km along the coast and supports abun-
dant early-successional browse in a moist,
relatively mild climate, lengthy growing
season, and continuous channel changes
by glacial streams and the Copper River
(Christensen 2000, Kesti et al. 2007,
Thilenius 2008). Using a map derived from
Satellite Pour l’Observation de la Terre
(SPOT version 5 [SPOT5], 2011, Red Castle
Resources, Inc.), we identified 7 stand types
that produce winter moose forage: spruce-
hemlock, spruce-cottonwood, cottonwood,
alder, alder-willow, willow, and sweetgale
(Viereck 1992). Spruce-hemlock, spruce-
cottonwood, alder, and sweetgale can all
form late-successional stands depending on
hydrology, but alder-willow, willow, and
sweetgale stands are generally considered
early-successional (Boggs 2000).

Drainage and desalination resulting from
the 1964 earthquake increased the distribu-
tion of spruce-hemlock and alder stands,
while accelerating succession or increasing
the composition of willow, alder, Sitka
spruce (Picea sitchensis), and western hem-
lock (Tsuga heterophylla) within some
stands (Boggs 2000, Stephenson et al.
2006, Thilenius 2008). Total winter snow
depths range from 83.3–548.6 cm (1994–
2013; ACRC 2014), and the area also
receives substantial rainfall (annual mean of
236 cm), frequently interspersed within peri-
ods of snowfall (Kesti et al. 2007). This phe-
nomenon varies with winter severity, which
can significantly affect snow accumulation,
drifting, and compaction. Thus, efforts to
understand the complex interactions among
snow depth, moose behavior, and browse
availability are complicated and challenging.

Treatments and Data Collection
Prior to initial treatments, managers sub-

jectively rated the suitability of potential
treatment sites as high, medium, or low
using factors of willow composition, en-
croachment by other woody species, and
the level of understory organic matter
(M. Burcham, USDA Forest Service Cordova
Ranger District, unpublished data); only
highly suitable sites were treated. Due to the
logistical difficulty of moving heavy equip-
ment through wetlands, treatment occurred
during winters with sufficiently frozen
ground, and sites were partially determined
by road access. Managers refined their site
selection techniques after the 1990–1992
treatments, selecting stands with the greatest
potential for increased willow production. In
total, the Forest Service treated approximately
300 ha from 1990–2012. Treatments were
applied to 32 sites in 5 stand types varying
from 0.9–63.4 ha in the east-central, mid-
central, and north-central regions of the
west Delta (Table 1; Fig. 1). All sites were
unfenced and open (available) to moose.

We sampled sites in August–September
2012–2013 and April–May 2013 to capture
pre-winter available biomass and over-
winter utilization and nutrition, respectively.
Because of logistical difficulties and differ-
ences in moose browsing pressure among
sites, we selected 20 comparable sites treated
in the east-central and mid-central region of
the Delta (Table 1; Fig. 1). We randomized
sampling plots in treated sites and untreated
adjacent controls, categorizing each site by
the current control stand type. Our study
plots consisted of 3 random-start belt tran-
sects (1 × 10 m) separated by 5 m and
running north, north, and east, respectively.

We estimated the forage biomass avail-
able to moose (total biomass of twigs with
diameters ≤8.3 mm; g/stem) with basal
diameter-mass regression equations (Table 2;
MacCracken and Van Ballenberghe 1993,
Stephenson et al.1998). At every 0.5 m along

138

MECHANICAL TREATMENT OF MOOSE BROWSE – SMYTHE ET AL. ALCES VOL. 51, 2015



Table 2. Regression equations used to estimate species-specific available biomass (g/stem) and biomass consumed (g/twig) by moose wintering on the Copper
River Delta, Alaska, USA.

Time Since Treatment

Browse Species 1 Yeara 3 Yearsa 5 Yearsb 22–23 Yearsb Untreatedc Consumptionc

Cottonwood = exp(−4.22) (BD2.85) = 0.64 (BD) = 0.15 (BD1.97) g– = 2.37 (BD) = 0.04 (bD2.6)
Alder = exp(−3.89) (BD2.77) = exp(−2.45) (BD1.8) = 0.03 (BD2.58) = 4.12 (BD) = 2.33 (BD) d= 0.03 + 0.06 (bD2.5) or = 0.34 (bD4)
Sitka willow = exp(−3.16) (BD2.52) = exp(−0.93) (BD1.46) = 0.13 (BD2.02) = 0.21 (BD1.8) = 11.07 (lnBD) = 0.03 + 0.06 (bD2.5)
Barclay willow e,f= exp(−3.50) (BD2.72) f= 0.98 (BD) = 1.74 (BD) = 2.56 (BD) e= 0.14 (BD1.93) = 0.05 + 0.03 (bD2.7)
Hooker’s willow e,f= exp(−3.50) (BD2.72) f= 0.98 (BD) = 0.11 (BD2.09) = 1.43 (BD) e= 0.18 (BD1.80) = 0.05 + 0.03 (bD2.7)
Undergreen willow = exp(−3.12) (BD2.48) = 0.56 (BD) = 1.51 (BD) = 1.40 (BD) = 0.55 (BD) = 0.05 +0.03 (bD2.7)
Sweetgale = 0.12 (BD) = 0.22 (BD) = 1.26 (BD) = 1.70 (BD) = exp(−3.33) (BD2.15) d= 0.05 + 0.03 (bD2.7) or = 0.12 (bD2)

Available biomass and biomass consumed equations are derived from measurements of basal diameters (BD, mm) and bite diameters (bD, mm), respectively. Available
biomass equations were developed in both mechanically-treated (hydraulic-axed) and untreated control sites. Treated site equations are presented according to their site
age (time since treatment, as of sampling in 2012 & 2013).
aDeveloped by Stephenson et al. (1998).
bDeveloped by Smythe et al. (current).
cDeveloped by MacCracken and Van Ballenberghe (1993).
dRevised by Stephenson et al. (1998).
eRevised by Smythe et al. (current); negative added to coefficient.
fSeparate equations were not developed for Hooker’s and Barclay willows (Smythe unpublished).
gSample size was insufficient to develop a regression equation.

A
L
C
E
S
V
O
L
.
5
1
,
2
0
1
5

S
M
Y
T
H
E
E
T
A
L
.
–
M
E
C
H
A
N
IC
A
L
T
R
E
A
T
M
E
N
T
O
F
M
O
O
S
E
B
R
O
W
S
E

1
3
9



the belt transects, we measured basal di-
ameters (mm; above the moss layer) of the
3 stems closest to the transect line. Past
research indicated that very large stem basal
diameters (>60.0 mm) increased regression
equation heteroskedasticity (MacCracken
and Van Ballenberghe 1993). Thus, with
such stems we instead measured a branch
diameter and estimated how many equiva-
lent branches were on the stem. Within the
belt transects, we calculated stem density
(stems/belt; stems/ha), measured shrub
height (m) on 3 replicates of every species,
and estimated the available biomass (%)
on each stem in 1-m vertical increments
from 0–6 m to reflect the range of moose
winter browsing heights, depending on
CRD snow pack conditions (T. Joyce,
USDA Forest Service Cordova Ranger
District, personal communication). We cal-
culated the total available biomass (kg/ha;
stem biomass × stem density) of every spe-
cies in each plot.

To calculate moose utilization, we mea-
sured every instance of browsing (bite dia-
meters) on the closest 0.5 m stem. We
estimated biomass consumed (g/twig) with
bite diameter-mass regression equations
(MacCracken and Van Ballenberghe 1993)
and summed the biomass removed per stem
(g/stem). We collected nutritional samples
of every browse species found at each plot,
stored them fresh-frozen, removed all leaves,
and sent them to the Washington State Uni-
versity Wildlife Habitat and Nutrition Lab
(Pullman) for analysis.

We developed 3 winter scenarios (mild,
moderate, and severe) by summarizing data
on mean winter snow depth (cm) from
1917–2012 collected by the Alaska Climate
Research Center (ACRC 2014) at Cordova’s
“Mudhole Smith” Airport weather station.
We could not accurately model the inter-
action between snow depth, snow compac-
tion, and biomass available within the moose
browsing window (0.5–3.0 m without snow).

Instead, we estimated the overall change in
available biomass of browse in each plot
according to our estimates of mean snow
depth under each winter scenario, assuming
that moose browsing height increased equally
with snow depth.

To evaluate differences between treated
sites and their controls, we used t-tests to
compare individual browse species and total
plot available biomass, height, crude protein,
lignin, tannin, and utilization, as well as the
ratio of willow:alder biomass. Individual
willow species effects did not differ signifi-
cantly and willow counts were pooled; felt-
leaf willow was not observed in any plot
and was removed from analyses; the 1990–
1992 treatments were analyzed as a single
treatment because they were not documented
separately. Furthermore, because we found
few differences in time-since-treatment
effects across stand types, we pooled all
stand types for time-since-treatment analyses
and used analysis of variance (ANOVA) to
compare treatments across time and winter
scenarios.

RESULTS
Treated willow, sweetgale, and total

available biomass in 1990–1992 sites were
higher than at control sites (P = 0.05,
0.003, and 0.001, respectively; Table 3) No
other differences were found between treated
and control sites in available biomass of any
browse species or treatment year (Table 3).
When weighted according to their untreated
control (cut/control × 100), neither the rela-
tive total available biomass nor the relative
total willow biomass differed significantly
across times-since-treatment (Fig. 2). Treated
alders in 2012 plots were shorter than in
controls (P = 0.03). There was no significant
effect on average willow height for time-
since-treatment (Fig. 2), but the average
treated willow was shorter than the average
control willow (P = 0.003). There were no
significant differences in nutritional quality

140

MECHANICAL TREATMENT OF MOOSE BROWSE – SMYTHE ET AL. ALCES VOL. 51, 2015



Table 3. Species-specific and total mean (±SD) available biomass (kg/ha), height (m), crude protein (%), lignin (%), tannin (mg/g), and use (%) of winter browse
for moose in mechanically treated (cut, via hydraulic-ax) and untreated (control) sites on the Copper River Delta, Alaska, USA.

Browse Species Age (yr) Treatment Biomass (kg/ha) Height (m) Crude Protein (%) Lignin (%) Tannin (mg/g) Use (%)

Black cottonwood 1 Cut 10.89 (−) 1.0 (−) a– a– a– a–
Control 2343.00 (−) 6.0 (−) a– a– a– a–

3 Cut b– b– b– b– b– b–
Control b– b– b– b– b– b–

5 Cut 15.18 (11.05) 2.3 (1.2) 8.16 (2.56) 12.47 (0.65) 0.00 (0.00) 18.47 (0.32)

Control 573.53 (522.92) 4.0 (2.8) 5.45 (−) 13.28 (−) 0.00 (0.00) 0.00 (0.00)
23 Cut b– b– b– b– b– –

Control 21.49 (51.19) 4.5 (2.1) 4.74 (−) 18.7 (−) 0.00 (−) 18.47 (−)
Sitka alder 1 Cut 18.15 (13.43)* 1.0 (0.0)** a– a– a– a–

Control 605.42 (307.10)* 4.7 (1.2)** a– a– a– a–
3 Cut 3.78 (4.99) 1.5 (0.7) c7.64 (−) c14.7 (−) c31.6 (−) 57.05 (15.20)

Control 138.59 (240.04) 6.0 (−) c7.64 (−) c14.7 (−) c31.6 (−) 0.40 (−)
5 Cut b– b– b– b– b– b–

Control 125.48 (149.59) 4.0 (0.0) 7.64 (−) 14.7 (−) 31.6 (−) 0.00 (0.00)
23 Cut 143.42 (430.25) 5.0 (−) 7.64 (−) 14.7 (−) 31.6 (−) 0.00 (−)

Control 257.49 (429.99) 4.5 (1.29) 7.64 (−) 14.7 (−) 31.6 (−) 7.17 (12.41)
Willow spp. 1 Cut 78.13 (75.01) 1.3 (0.6) a– a– a– a–

Control 279.81 (253.04) 3.9 (1.9) a– a– a– a–
3 Cut 386.19 (416.60) 1.4 (0.5) 7.04 (0.71) 11.87 (0.48)* 49.07 (17.64) 14.50 (10.87)

Control 405.41 (244.35) 2.3 (0.7) 6.91 (0.89) 15.47 (2.08)* 44.51 (19.55) 12.67 (5.03)

5 Cut 550.79 (370.05) 1.6 (0.5) 7.91 (1.13) 15.53 (1.47) 32.28 (30.92) 3.25 (4.27)

Control 260.67 (112.35) 3.7 (1.5) 6.85 (1.18) 13.71 (1.29) 43.52 (3.03) 0.00 (0.00)

23 Cut 1225.01 (614.71)** 2.0 (0.5) 7.06 (0.54) 15.60 (1.67) 48.26 (16.92) 16.17 (15.45)

Control 522.89 (408.90)** 2.5 (0.7) 7.07 (0.64) 15.61 (0.54) 37.48 (30.56) 11.83 (8.35)

Table 3 continued . . . .

A
L
C
E
S
V
O
L
.
5
1
,
2
0
1
5

S
M
Y
T
H
E
E
T
A
L
.
–
M
E
C
H
A
N
IC
A
L
T
R
E
A
T
M
E
N
T
O
F
M
O
O
S
E
B
R
O
W
S
E

1
4
1



Table 3 continued

Browse Species Age (yr) Treatment Biomass (kg/ha) Height (m) Crude Protein (%) Lignin (%) Tannin (mg/g) Use (%)

Sweetgale 1 Cut 21.06 (36.47) 1.0 (−) a– a– a– a–
Control 0.04 (0.08) 1.0 (−) a– a– a– a–

3 Cut 76.63 (86.54) 1.0 (0.0) 8.50 (0.64) 22.42 (0.86) 44.53 (1.33) 53.00 (23.07)

Control 250.13 (221.49) 1.0 (0.0) 6.85 (−) 22.61 (−) 98.90 (−) 33.50 (21.92)
5 Cut 403.28 (547.33) 1.0 (0.0) 6.75 (−) 17.00 (−) 41.00 (−) 10.80 (6.22)

Control b– b– b– b– b– b–
23 Cut 503.02 (560.63)** 1.0 (0.0) 7.53 (0.46)* 21.73 (0.59)* 56.98 (27.95)* 7.75 (8.18)

Control 56.30 (103.77)** 1.0 (0.0) 6.91 (0.11)* 22.51 (0.17)* 95.37 (6.12)* 40.00 (40.15)

Total winter 1 Cut 120.96 (80.93) 1.17 (0.29)* a– a– a– a–
Control 1666.26 (1292.39) 4.13 (1.63)* a– a– a– a–

3 Cut 466.59 (476.44) 1.35 (0.47) 7.52 (0.27) 14.80 (1.27) 43.66 (6.70) 23.28 (12.00)

Control 794.12 (212.15) 2.38 (1.19) 6.94 (0.80) 16.40 (2.53) 51.06 (23.13) 16.00 (15.09)

5 Cut 969.24 (852.05) 1.76 (0.77) 8.13 (1.50) 14.47 (0.23) 23.79 (23.96) 5.00 (5.83)

Control 959.69 (663.60) 3.44 (1.50) 6.81 (0.55) 14.13 (0.83) 30.43 (8.34) 0.00 (0.00)

23 Cut 1871.44 (711.48)** 1.96 (0.56)* 7.19 (0.38) 16.77 (2.04) 46.76 (16.92) 10.67 (4.23)

Control 858.17 (454.79)** 2.86 (1.01)* 7.06 (0.60) 16.20 (1.30) 40.75 (29.50) 9.33 (7.58)

Treated sites were sampled 1, 3, 5, or 23 years post-treatment (age) in 2012 & 2013.
aRe-growth of sites had not occurred by the time of spring nutritional sampling in one-year-old sites, but had occurred by the time of fall biomass sampling.
bSpecies did not occur in site.
cAlder samples combined for nutritional analysis.
*t-test, P = 0.06–0.10 between cut and control.
**t-test, P ≤ 0.05 between cut and control.

1
4
2

M
E
C
H
A
N
IC
A
L
T
R
E
A
T
M
E
N
T
O
F
M
O
O
S
E
B
R
O
W
S
E
–
S
M
Y
T
H
E
E
T
A
L
.

A
L
C
E
S
V
O
L
.
5
1
,
2
0
1
5



Fig. 2. Total relative biomass (cut/uncut × 100, ±SD) of all winter browse species,
relative biomass of willows (Salix spp.), and mean heights of treated willows
available to wintering moose within mechanically treated (via hydraulic-ax
between 1990–2012) sites on the Copper River Delta, Alaska as of 2012–2013
sampling. The dashed line represents the point at which treated sites have
recovered pre-treatment biomass (100%) or the mean height of untreated willows
(2.85 m). Relative biomass across the 4 treatments was not significantly different
(P = 0.15 and 0.13, respectively, 3 df). The average treated willow is significantly
shorter than the average untreated willow (P = 0.003), but treated willow heights
across treatment years are not significantly different (P = 0.13, 3 df).

ALCES VOL. 51, 2015 SMYTHE ET AL. – MECHANICAL TREATMENT OF MOOSE BROWSE

143



or utilization across any comparison. The
ratio of willow:alder in treated sites was
higher than in control sites at 23 years
post-treatment (treated = 1163.37, control =
205.82, P = 0.004), though treated sites 1,
3, and 5 years post-treatment were not differ-
ent (treated = 11.26, 323.63, 550.79, respect-
ively; control = 0.77, 360.11, and 74.38,
respectively). All treatment years were differ-
ent (P = 0.02, 3 df).

The 3 winter scenarios (mild, moderate,
and severe) occurred 49, 29, and 11 times,
respectively, with 6 winters uncategorized
due to missing data. Mean snow depth dif-
fered by scenario; 11.4 cm (±9.9–12.9),
25.8 cm (±23.3–28.3), and 63.9 cm (±47.4–
80.4), respectively. Total available biomass
across times-since-treatment varied signifi-
cantly by scenario (P = 0.007–0.03, 4 df;
Fig. 3). Total available biomass in treated
1990–1992 plots also differed across scen-
arios (P = 0.04, 3 df), declining 61% from
mild to severe winters. Further, available
willow biomass across times-since-treatment
varied significantly by scenario (P = 0.01–
0.05, 4 df; Fig. 3). Treated willow biomass
in the 2008 plot differed across scenarios
(P = 0.05, 3 df), declining 95% from mild
to severe winters.

DISCUSSION
Our data indicate that hydro-axing pro-

duces more total and willow biomass, with
the effect increasing over time. Given the
observed variability, our a posteriori power
analyses suggested sample sizes of 9–17
would be necessary to detect significance in
comparisons of willow-only or all-species
browse; however, treatment caused signifi-
cant increase in the ratio of willow:alder
over time. Our results support those of
Harrington (1984), and further suggest that
hydro-axing can be an effective method
to increase willow biomass and counter
ecologically-initiated (including earthquake-
influenced hydrological or successional)

increases in alder. Hydro-axing did not influ-
ence the nutritional quality of the treated
browse, as suggested by the lack of differ-
ence in crude protein, lignin, tannins, and
utilization by moose. Bowyer et al. (2001)
reported similar findings for treated feltleaf
willow in interior Alaska, whereas Rea and
Gillingham (2001) measured nutritional dif-
ferences in Scouler’s willow (Salix scouleri-
ana); however, both studies were short-term
(≤3 years post-treatment).

The high variability in height (m) of
treated willows makes it difficult to deter-
mine if hydro-axing affects final regrowth
height and the biomass available to moose
across winter scenarios. Because the average
treated willow is shorter, yet more produc-
tive than the average untreated willow,
hydro-axing may be causing a bushier
growth form in treated willows, with more
biomass concentrated in many smaller
shoots on recovering stems. A changed
architecture may explain the larger decrease
in available biomass relative to controls in
1990–1992 treated sites as winter severity
and snow depth increased. However, after
23 years of regrowth, mean available bio-
mass in severe winters was similar to the
mean available biomass provided by con-
trols, suggesting that overall availability of
treated biomass may compensate for losses
due to snow burial. If so, hydro-axing would
be an effective tool for increasing biomass
available to moose in mild and moderate
winters, while maintaining “normal” avail-
ability in severe winters, given sufficient
time for regrowth. Given the large gap be-
tween the 2008 and 1990–1992 treatments,
we were unable to determine the regrowth
asymptote or the minimal time required for
winter browse species to recover sufficiently
from treatment to provide equivalent (or po-
tentially increased) biomass during severe
winters.

Overall, our results indicate that mechan-
ical treatment of moose winter browse

144

MECHANICAL TREATMENT OF MOOSE BROWSE – SMYTHE ET AL. ALCES VOL. 51, 2015



species via hydro-axing has potential to re-
duce alder and increase willow biomass for
wintering moose on the CRD. However, ex-
tensive treatment could limit browse avail-
ability during extreme winter scenarios
(deep snow) until regrowth occurs in a few
decades. Managers should be cautious in

applying this technique across large areas
concurrently. Furthermore, monitoring at
more frequent intervals should determine
the temporal development and long-term
effects of mechanical treatment on moose
forage in the CRD. This study provides a
substantial summary of the effects of

Fig. 3. Reductions in total and willow (Salix spp.) biomass (kg/ha, ±CI) available to moose due to
mean snow depths in 3 winter scenarios (mild, moderate, severe) in mechanically treated (via
hydraulic-ax) sites cut over 4 years (1990–1992, 2008, 2010, and 2012) on the Copper River
Delta, Alaska. Sites were sampled in 2012–2013. All biomass differences within winter scenarios
are significant (P = 0.007–0.03 and 0.01–0.05, respectively, 4 df), and the 1990–1992 across-
scenario differences are significant (P = 0.04, 3 df).

ALCES VOL. 51, 2015 SMYTHE ET AL. – MECHANICAL TREATMENT OF MOOSE BROWSE

145



mechanical treatment on winter browse spe-
cies, and should provide habitat managers
of the CRD and similar areas with a useful
structure for current management decisions
and further research.

ACKNOWLEDGEMENTS
Special thanks go to the managers at

the Cordova Ranger District of the US Forest
Service and the Alaska Department of Fish
and Game, who provided critical financial,
logistical, and personal support, including
T. Joyce, E. Cooper, M. Burcham, C. West-
ing, D. Crowley, and many seasonal employ-
ees. G. Reeves, USFS-PNW Research
Station, provided vital support in arranging
and implementing this project.

REFERENCES

ALASKA CLIMATE RESEARCH CENTER (ACRC).
2014. Applied Climate Information Sys-
tem Daily Data Browser of University
of Alaska Fairbanks. <http://climate.gi.
alaska.edu/acis_data> (accessed May
2014).

BOGGS, K. 2000. Classification of Commu-
nity Types, Successional Sequences, and
Landscapes of the Copper River Delta,
Alaska. U.S. Department of Agriculture,
Forest Service, Pacific Northwest
Research Station, Portland, Oregon, USA.

BOWYER, T. R., B. M. PIERCE, L. K. DUFFY,
and D. A. HAGGERSTROM. 2001. Sexual
segregation in moose: effects of habitat
manipulation. Alces 37: 109–122.

CHRISTENSEN, H. H. 2000. Alaska’s Copper
River: Humankind in a Changing World.
U.S. Department of Agriculture, Forest
Service, Pacific Northwest Research
Station, Portland, Oregon, USA.

DAVIDSON, D., and C. HARNISH. 1978. Soil and
Water Resource Inventory of the Copper
River Delta. U.S. Department of Agricul-
ture, Forest Service, Chugach National
Forest, Anchorage, Alaska, USA.

FERRIANS, O. J. Jr. 1966. Effects of the Earth-
quake of March 27, 1964 in the Copper

River Basin Area, Alaska. Geological
Survey Professional Paper 543 - E.
U.S. Government Printing Office,
Washington, D.C., USA.

GRANTZ,A.,G.PLAFKER,andR.KACHADOORIAN.
1964. Alaska’s Good Friday Earthquake,
March 27, 1964: A Preliminary Geologic
Evaluation. U.S. Department of the
Interior, Geological Survey, Washington,
D.C., USA.

HARRINGTON, C. A. 1984. Factors influencing
initial sprouting of red alder. Canadian
Journal of Forest Research 14: 357–361.

HUNDERTMARK, K. J., W. L. EBERHARDT, and
E. BAIL. 1990. Winter habitat use by
moose in southeastern Alaska: implica-
tions for forest management. Alces 26:
108–114.

KESTI, S., M. BURCHAM, B. CAMPBELL,
D. DAVIDSON, R. DEVELICE, C. HUBER,
T. JOYCE, D. LANG, B. MACFARLANE,
D. SHERMAN, and L. YARBOROUGH. 2007.
West Copper River Delta Landscape
Assessment. U.S. Department of Agricul-
ture, Forest Service, Chugach National
Forest, Anchorage, Alaska, USA.

MACCRACKEN,J.G., andV. VAN BALLENBERGHE.
1993. Mass-diameter regressions for
moose browse on the Copper River Delta,
Alaska. Journal of Range Management
46: 302–308.

———, ———, and J. M. PEEK. 1997. Habi-
tat relationships of moose on the Copper
River Delta in coastal south-central
Alaska. Wildlife Monographs 136: 3–52.

OLDEMEYER, J., and W. REGELIN. 1980.
Response of vegetation to tree crushing
in Alaska. Alces 16: 429–443.

PLAFKER, G. 1969. Tectonics of the March 27,
1964 Alaska Earthquake. U.S. Geological
Survey Professional Paper 543-I. U.S.
Government Printing Office, Washington,
D.C., USA.

REA, R. V., and M. P. GILLINGHAM. 2001. The
impact of the timing of brush manage-
ment on the nutritional value of woody
browse for moose Alces alces. Journal
of Applied Ecology 38: 710–719.

146

MECHANICAL TREATMENT OF MOOSE BROWSE – SMYTHE ET AL. ALCES VOL. 51, 2015

http://climate.gi.alaska.edu/acis_data
http://climate.gi.alaska.edu/acis_data


REGELIN, W. L., C. C. SCHWARTZ, and A. W.
FRANZMANN. 1985. Seasonal energy
metabolism of adult moose. The Journal
of Wildlife Management 49: 388–393.

RENECKER, L. A., and C. C. SCHWARTZ. 1997.
Food habits and feeding behavior. Pages
403–440 in A. W. FRANZMANN and C. C.
SCHWARTZ, editors. Ecology and Manage-
ment of the North American Moose.
Smithsonian Institution Press, Washington,
D.C., USA.

SCHWARTZ, C. C., M. E. HUBBERT, and A. W.
FRANZMANN. 1988. Energy requirements
of adult moose for winter maintenance.
The Journal of Wildlife Management
52: 26–33.

SCOTTER, G. W. 1980. Management of wild
ungulate habitat in the western United
States and Canada: a review. Journal of
Range Management 33: 16–27.

STEPHENSON, T. R., V. VAN BALLENBERGHE, and
J. M. Peek. 1998. Response of moose
forages to mechanical cutting on the Cop-
per River Delta, Alaska. Alces 34:
479–494.

———, ———, ———, and J. G.
MacCracken. 2006. Spatio-temporal
constraints on moose habitat and carrying
capacity in coastal Alaska: vegetation

succession and climate. Rangeland Ecol-
ogy & Management 59: 359–372.

STOVER, C. W., J. L. COFFMAN. 1993. Seismicity
of the United States, 1568-1989 (Revised).
U.S. Geological Survey Professional
Paper 1527.

SURING, L., and C. STERNE. 1998. Winter habi-
tat use by moose in south-central Alaska.
Alces 34: 139–147.

THILENIUS, J. F. 1990. Woody plant succession
on earthquake-uplifted coastal wetlands
of the Copper River Delta, Alaska. Forest
Ecology and Management 33: 439–462.

———. 2008. Phytosociology and succes-
sion on earthquake-uplifted coastal wet-
lands, Copper River Delta, Alaska. U.S.
Government Printing Office, Washington,
D.C., USA.

THOMPSON, I. D., and R. W. STEWART. 1997.
Management of moose habitat. Pages
377–402 in A. W. FRANZMANN and C. C.
SCHWARTZ, editors. Ecology and Manage-
ment of the North American Moose.
Smithsonian Intitution Press, Washington,
D.C., USA.

VIERECK, L. A. 1992. The Alaska vegetation
classification. U.S. Department of Agri-
culture, Forest Service, Pacific Northwest
Research Station, Portland, Oregon, USA.

ALCES VOL. 51, 2015 SMYTHE ET AL. – MECHANICAL TREATMENT OF MOOSE BROWSE

147


	TEMPORAL EFFECTS OF MECHANICAL TREATMENT ON WINTER MOOSE BROWSE IN SOUTH-ENTRAL ALASKA
	METHODS
	Study Area
	Treatments and Data Collection

	RESULTS
	DISCUSSION
	ACKNOWLEDGEMENTS
	REFERENCES