YIELD O F POPULUS HYBRIDS IN " WOOD GRASS" AND O THER SHORT-ROTATION DENSI TY REGIMES Dean S. DeBell, William R. Harms, and John C. Zasada Principal Silviculturists USDA Forest Service, Forestry Sciences Laboratories Olympia, Washington 98502 Charleston, South Carolina 29407 Corvallis, Oregon 97331 ABS TRAC T A comprehensive test of effects of spacing on growth
and yield of Populus was established in western Washington
in spring 1986.
Two hybrid clones (D-Ol and H-11) were
planted at two woodgrass spacings (0.18 and 0. 30 m) and
three wider spacings (0. 50, 1. 00, and 2. 00 m).
All
treatments were replicated three times in a randomized
block design; all plots were fertilized, irrigated, and
weeded uniformly.
Heights and diameters of 100 trees in
each plot were measured after the first and second year.
Stems in the woodgrass spacings were harvested and weighed
after the 1986 growing season; stumps resprouted and the
Yields of
first coppice crop was harvested in fall 1987.
wider spaced plots were estimated from biomass equations
applied to height and diameter measurements of trees.
Growth of individual trees increased with increased
spacing in both years.
First-year yield was inversely
related to spacing, and therefore was highest in the
woodgrass spacings.
During the second year, growth per
tree and per hectare accelerated, especially in the wider
spacings.
Cumulative 2-year yield of woodgrass was
equalled or exceeded by standing woody biomass in the
0.5-m spacings of both clones and in the 1. 0-m spacing of
H-1 1.
These initial results coupled with other
information (costs and biomass characteristics) indicate
that woodgrass is less promising than wider spacings for
growing Populus biomass for energy.
DeBell, D. S., W. R. Harms and J. C. Zasada.
1988.
Yield of Populus hybrids
in "wood-grass" and other short-rotation density regimes.
In Proc., Institute
of Gas Technology Conference on Energy from Biomass and Wastes, New Orleans,
LA.
February 1988.
( In Press )
YIELD O F POPULUS h'lBRIDS IN "WOODGRASS" AND OTHER SnORT-ROTATION DENSITY REGIMES IKTRODUC TION
.current ideas about woody biomass farms have their
origin in the "silage sycamore" concept proposed more than
two decades ago in the southeastern United States
(1 1, 10). The concept included:
establishing
rapid-growing trees at dense spacings, applying intensive
cultural practices, harvesting in cycles of 10 years or
less, regenerating subsequent crops via sprouts or coppice
arising from stumps, and using a high degree of
mechanization. Trials were soon initiated elsewhere in
the country (4, 6, 8) with emphasis on Populus species and
hybrids.' During the first energy crisis, this
short-rotation approach \vas suggested as a means to
produce wood for energy (14)
Subsequently, -considerable
research has been done on biomass plantations for fiber
and energy, most of it funded by the Short Rotation Woody
Crops Program of the U.S. Department of Energy. In
general, such work has pointed toward improved
productivity and cost effectiveness if spacings are wider
and harvest cycles are longer than those evaluated in many
of the early trials. Stand densities of 2500 to 4000
trees per hectare (i.e., square spacings of 1.6 to 2.0 m)
a:1d rotations of 5 to 8 years are now perceived as optimum
=or bioenergy crops (12). Other on-going research
sugges ts that even wider spacing (up to 3.0 m) and longer rotations (up to 12 years) may be equally productive and perhaps preferable for some objectives. In the early 1980's, however, a nurseryman in Oregon
proposed a radical depart re from the above trends in
woody biomass farming (5). His system entailed
establishing a Populus hybrid at densities of 100 thousand
to
00 thousand rootstocks per hectare with annual
!J.a:::-vests of coppice.
Biomass yields were purported to
exceed 100 tonnes per hectare annually. Economic analyses
performed by coupling such yield data with estimated costs
sugges ted -::hat the system--dubbed "woodgrass"--compared
favorably with other short-rotation density regimes (15).
Considerable inte:::-est developed in the energy conversion
communi ty, but many forest biologists remained skeptical.
Was this skepticism well-founded or was it associated with
reluctance to accept innovation arising ou tside the
scientific es tablishmen ?
2
Managers of the Short Rotation Woody Crops Program
decided that a scientific evaluation of the woodgrass
concept was needed, and the Olympia Forestry Sciences
Laboratory of the U. S. Fo=est Service was selected to
conduct the investigation.
Our study compares two Populus
hybrids at five spacings--ranging from two woodgrass
spacings (0.18 and 0.30 m) to one approaching a
conventional pulpwood spacing (2.0 m). This paper
describes growth and development of the plantings during
the first 2 years. METHODS
The experimental site is 12 km east of Olympia,
Climate is mild with an average growing
Washington.
season of 190 frost-free days and a mean July temperature
°
of 16 C.
Precipitation averages more than 1000 mm per
year, falling mostly as rain from October through May;
summers are periodically dry. The land was previously
farmed for strawberry and hay crops, but it is now managed
by the Washington State Department of Natural Resources as
a Douglas-fir seed orchard. Topography is relatively
level, and the soil is Nisqually sandy loam.
The land was
prepared for planting by plowing and disking in winter
1985-86. The study was established as a factorial design with
two Populus clones and five spacing treatments, replicated
in three blocks.
One clone, D-0 1, is a Populus hybrid
(taxonomic identity unknown) developed originally at
University of Idaho and subsequently selected from a
Canadian planting by Dula's Nursery of Canby, Oregon (5).
· deltoides
· trichocarpa x
The other clone, H- 1 1, is a
hybrid developed and tested by University of Washington
and Washington State University (9).
Square spacings (m
by m) are 0.18, 0.30, 0. 50, 1.00, and 2. 00 m. Equivalent
trees per hectare are about 310,000, 110, 000, 40, 000,
The first two spacings (0.18 and 0.30
10, 000, and 2, 500.
m) are woodgrass treatments suggested to us by Dula (i. e.,
3 and 1 plants per square foot).
Size of treatment plots
varies with spacing, but all plots are sufficiently large
to provide at least 100 trees in the interior measurement
plot (400 trees for woodgrass harvests), and a border
around each measurement plot at least one-half as wide as
the projected height of trees at harvest.
Both clones were planted by hand as unrooted,
hardwood cuttings in late April 1986.
All cuttings were
30 em long and had a minimum upper diameter of 1 em; they
were planted 20 em deep with at least two healthy axillary
buds remaining above ground.
Supplemental nutrien s and water were provided
A pre-planting
uniformly in plots of all treatments.
application of fertilizer (16-16-16) provided the
equivalent of 100 kg per hectare each of nitrogen,
phosphorus, and potassium.
Additional fertilizer will be
applied as needed to main ain adequate nutrient status.
Plots are irrigated throughout the summer by a drip
system.
All plots were kept free of weeds by tilling and
hoeing the first year and by herbicides and hoeing the
second year. At the end of the first year, all positions
occupied by dead trees were replanted with unrooted
cuttings.
Survival, height, and basal diameter were recorded on
all plots.
Number of living and dead sprouts per
rootstock were also tallied after the second growing
season in woodgrass plots.
Yield data for the woodgrass
treatments were based on harvests after leaffall, at the
end of the first and second growing seasons, of 400 trees
Moisture contents were
in the center of each plot.
determined on subsamples to convert fresh weight to
°
ovendry (105 C) weight.
Yields for the wider spaced
plots were estimated from dry-weight biomass equations
applied to diameter and height measurements of the trees.
The equations were of the form:
ln W
=
where w
D
H
=
=
=
b
o
+
b
l
ln D2H ,
dry weight (g)
basal diameter (em)
height (m).
The equations were based on 72 trees for each clone and accounted for 95 to 97 percent of the variation in tree dry weight.
Estimated weights of all trees on each plot were summed, and the resul ting plot dry weights expanded by appropriate multipliers to provide yield per hectare. Plot means were calc lated for each variable and
displayed in tables or =igures to illustrate trends in
All data have been analyzed
development of the plantings.
hy standard AN OVA techniques, and treatment means were
compared by Tukey's test sing P <
0.05 as
he level of
significance.
=
RESULTS AND DISCUSSION
Survival at the end of the first growing season
averaged 96 percent for D-01 and 98 percent for H-11.
Average heights for D-01 and H-11 were 1.44 and 1.80 m,
respectively.
In the two woodgrass spacings, mean heights
of the two clones were very similar (Figure la). As
4
spacing widened from 0.18 to 1.00 m, mean height of both
clones increased, but the gain was greater for H-11 than
for D-01.
Mean heights for the clones at spacings of
0.50, 1.00, and 2.00 m differed by 50 em or more. Effects
of spacing on basal diameter (Figure 1b) were similar to
those for height. Both clones had similar diameters at
the 0.18- and 0.30-m spacings; mean diameter of both
clones was greater at wider spacings and gains were
greater for H-11.
First-year development of two Po:eulus clones as
Figure 1.
spacing
related to
a.
Height (m)
2.6
2.4
2.2
2.0
1.8
1.6
.....
..c
[J)
'Qi
I
1.4
1.2
1.0
0.8
0.6
0.4
o H-11
0
0.2
0.0
I
0.18
0.30
0.50
1.00
D-01
2.00
Spacing
b.
Basal diameter (mm)
:1
18
'-
O.J
O.J
;5
E
0
s
0
Ul
0
m
9
6
'I
(1 '
o H-11
0 D-01
C.18 0.30
0.50
Spacing
1.00
:?..00
Patterns of leaf, bud, and branch production also
differed markedly between clones and among spacings during
the first growing season. Average distances between
leaves (hence, axillary b ds) were greater in H- 1 1 (about
Moreover, H- 1 1 exhibited
4 em) than in D-0 1 (about 3 em).
sylleptic growth; that is, branches developed from
axillary buds during the same growing season in which the
buds formed.
The proportion of buds producing sylleptic
branches ranged from none in the densest woodgrass spacing
Growth in
to 3 1 percent in the widest (2.0 m) spacing.
D-Ol was predominantly proleptic with axillary buds
remaining dormant until the next growing season; no buds
produced sylleptic branches in the two wood-grass spacings
and only 3. 3 and 2.0 percent produced sylleptic branches
in the 1.0 m and 2.0 m spacings.
The substantially reduced first-year growth in the
woodgrass spacings as compared with growth at wider
spacings indicated that competition among plants was
sufficient to stress growth processes.
Contrasted with
trees in the 1. 0 m spacing, trees in the densest woodgrass
spacing averaged 4 1 percent shorter in height and 7 1
percent smaller in basal diameter.
Moreover, leaf area
per tree was already reduced by an average of 82 percent.
Because of the intense competition in the woodgrass plots,
and, in accord with Dula' s (5) procedures, we cut these
dense plots at the end of the growing season to establish
coppice.
Yields from the first (non-coppice) harvest of the
woodgrass spacings are shown with those of the second
(true coppice) harvest in Table 1.
First-year yields of
the 0. 18-m spacing were significantly higher than those of
the 0.30-m spacing but did not vary by clone.
Dry-matter
production averaged 4. 1 tonnes per hectare in the 0. 18-m
spacing, and 3.0 tonnes per hectare in the 0.30-m
spacing.
Although yields were about one-third higher in
the denser spacing, three times as many cuttings (200
percent more) had been planted.
This decreased growth
efficiency per tree is associated with increased
competitive stress in the denser spacing; it is also
reflected in differences between the two spacings in
height and diameter growth ( Figure 1).
Vigorous sprouts began to develop on the stumps in
early April, and growth was excellent throughout the
Yields from the second cutting were more
second year.
than double those of the first cutting, and ranged from
7.7 to 9.7 tonnes per hec tare (Table 1).
Production was
significantly greater in the denser spacing (9.0 vs. 8.2
tonnes per hectare). Clone D-0 1 tended to produce higher
yield than H- 1 1, but considerable variation existed within
the clonal treatments and differences were not significant
0.05.
Increased production of the woodgrass
at P <
=
6
spacings in the second ha est is associated with
increased growth of the dominant stern on each root stock
(cf. Figure 1 and Table 2) and a greater number of living
stems per plant for the D-01 clone.
Averaged over both
spacings, mean heights of D-01 and H-11 were 2.0 and 2.2 rn
in the second year as compared to 1.3 and 1.4 rn in the
first year; diameters were also greater in the second year
for both clones. Although dominant sprouts of H-11 were
larger than those of D-01, the tendency of D-01 to produce
higher yield per hectare resulted from dramatic clonal
differences in the initiation and survival of coppice
sprouts ( Table 2). Averaged across spacings, D-01
produced 7. 4 sprouts per rootstock but H-11 had only 4.6.
Differences in patterns of bud and branch production
during the previous season may be related to these clonal
differences in total sprout production.
Numbers of
axillary buds per centimeter were about one-third greater
in D-01 than in H-11; also, a larger percentage of the
H-11 buds sprouted the previous year.
The combined effect
of these two growth characteristics resulted in greater
numbers of vigorous axillary buds on the rootstocks of
D-01 than on H-11; such buds no doubt play a significant
role in sprout development, especially in the first
coppice cycle.
Table 1.
Dry rield of woodgrass in first and second years
after planting
Clone
Year 2
Second harvest
spacing
Year 1
First harvest
spacing
0.18 rn
0.30
rn
0.18 rn
- - - - -tonnes per hectare- -
0.30 rn
-
- - - -
D-01
4.0
3.1
9.7
8.7
E-ll
4.2
3.0
8.4
7.7
Averaae
4.1
3.0
9.0
8.2
1
Above-around, leafless biomass dried to constant
°
C.
weight a f 105
7
Table 2. Characteristics of woodgrass in second year
after planting--first coppice harvest
Clone
Rootstock
Spacing survival
-%-
D-0 1
H- 1 1
Tallest sprout
per rootstock
Diameter Height
-mrn-
Sprouts
per rootstock
Total Living
-m-
- - -no.-
-
-
0.18 m
96
8 c
1.8 b
5. 13 b
2.66 b
0.30
98
10 b
2. 1 b
9.7 1 a
7.06 a
Mean
97
9
2.0
7. 42
4.86
0. 18 m
90
8 c
1.9 b
3.89 b
1.29 b
0.30
97
12 a
2.6 a
5.22 b
1.36 b
Mean
94
10
2.2
4.56
1.32
Within a column, means followed by the same letter do not
differ significantly at P < 0.05.
In number of sprouts surviving, even greater
Such differences
differences existed between the clones.
were especially evident in the 0.30-m spacing where D-0 1
averaged 7. 1 living sprouts per rootstock and H- 1 1
averaged only 1.4. Apparently, H- 1 1 is less able than
D-0 1 to cope with the intense competition occurring in the
woodgrass spacings; the lower survival of H- 1 1 rootstocks
(90 percent) in the 0.18-m spacing is probably a similar reflection of differences in tolerance to competition. Growth in the other spacings also accelerated during
the second year ( Table 3). Mean height of D-0 1 increased
from 1.5 m to 3.5 m and mean diameters similarly grew from
Second-year growth of H-1 1 was even
12 mm to 31 mm.
greater; mean height increased from 2.1 to 5.5 m and mean
Although growth of
diameter grew from 20 mm to 42 mm .
both clones increased with spacing (to 2.0 m for diamete ,
and to 1.0 m for height), the improvement in growth was
substantially greater for H-11, and thus differences
between the clones became progressively larger with wider
spacing. Mean dry weight per plant, which reflects tren ds
in both height and diameter, well illustrates the
differential effect of spacing on the two clones; at 0.5-m
spacing, H-11 plants are 1.4 times heavier than D-01; at
1.0-m spacing, they are twice as heavy; and at 2.0-m
spacing, H-11 plants weigh about 2.6 times more than D-Ol
plants.
Apparently, clone H-11 can grow faster than D-01,
B
provided growth factors such as space are adequate.
At
very dense Epacings, dry-matter production o f the two
clones is similar; at wide spacing, t he superior capacity
o f H-11 for rapid growth is expressed, and its growth is
Moreover, this superior
dramatically greater than D-01.
capacity is probably associated to some degree with H-ll's
sylleptic growth patterns; that is, it was thereby able to
increase leaf area with increased spacing far more rapidly
than could D-01.
Table 3.
Characteristics o f other (non-woodgrass) short
rotation spacings a fter second year
Mean tree size
Clone
D-01
H-11
Spacing
Rootstock
survival
Diameter
-%-
-mm-
Height
-m-
Dry weight
-kg-
0.50
100
22 a
3.8 b
0. 36 a
1.00
99
32 b
3. 6 ab
0. 76 a
2.00
100
38 be
3.0 a
0. 96 ab
Mean
100
31
3. 5
0. 69
0. 50
99
26 ab
4. 8 c
0. 51 a
1. 00
100
43 c
6. 0 d
1. 52 b
2. 00
100
57 d
5.8 d
2. 46 c
Mean
100
42
5. 5
1. 50
Within a column, means followed by the same letter do not
di ffer significantly at P < 0.05.
Such differences in growth among spacings and clones
were evaluated in terms of cumulative (2-year) yield of
all treatments ( Figure 2).
The yields for woodgrass
spacings consist of the first and second harvests; yields
for the other spacings represent total woody biomass
standing after the second growing season.
Despite a
doubling of production in the woodgrass spacings with the
second (coppice) harvest, dry-matter production in some o f
the wider, non-coppiced treatments is even greater.
Thus,
production by H-11 at 0.5-m spacing was more than 21
tonnes per hectare -- at least 18 o f which were produced
during the second growing season.
This yield was
0
significantly higher than any woodgrass treatment; it
averaged 66 percent higher than the average woodgrass
yield. Moreover, cumulative yields of D-01 at 0.5-m
spacing and H-11 at 1. 0-m spacing were equal to or greater
than yield of any woodgrass treatment. Second-year
increment in these clone-spacing combinations was 13 to 15
tonnes per hectare -- nearly double the production
obtained in coppiced woodgrass plots for the year.
Cumulative two-year yield of woodgrass and Figure 2.
other short rotation spacings 24
...-0
.I:
L.. Q) a.
Ul Q) c
c
0
"'"' .._,
-c
Q)
>=
c
0
Q) >
:;:::; 0
:i E
::J u
D
lid]
22
20
D-01
H-11
Ff
18
16 14 .--.----
12
10
8
6
f='F
4
2
0
0.1 8
0.30
0.50
1 .00
I
2.00
Spacing (m)
I MPLICATIONS AND CONCLUSIONS
Growth and survival of trees during the 2-year period
have been excellent. Height and diameter were equal to or
greater than growth for comparable spacings at other
locations in the Pacific Northwest.
This successful
performance presumably resulted from a favorable
irrigation and fertilizer regime as well as excellent weed
control imposed uniformly on all treatments.
How did woodgrass yields in our experiment compare
with purported yields?
To make such a comparison, one
must first place the yields suggested by Dula (5) or Vyas
and Shen (15) on a common basis with those determined in
our study.
The report by Dula (5) suggests that at least
112 wet tonnes per hectare (50 wet tons per acre)-per year
10
of total above-ground yield (including leaves) can be His sample contained 37 percent leaves; thus, expected.
stems and branches weighed 71 tonnes.
Moisture content of t he stems and branches was 71 percent; thus dry woody Dula's yield was biomass weighed about 20 tonnes.
measured in a nursery environment in which long, narrow beds occupied only two-thirds of the total land dedicated to woodgrass production.
To be comparable to "solid" plantings which occupy the total land area, the woodgrass yield s hould t herefore be reduced by one-third.
Thus, the woodgrass yield indicated by Dula is equivalent to about
14 tonnes of dry woody biomass per hectare per year.
Our
coppice yields (second harvest) were about 9 to 10 tonnes
for D-01 and about 8 tonnes for H-11.
Previous work in
dense spacings (6, 7) has indicated that yield from the
second coppice harvest will be somewhat higher than that
obtained in the first coppice harvest.
We can probably
expect 10 to 12 tonnes in the next woodgrass harvest.
Our
experiment therefore indicated that the minimal woodgrass
yield suggested by Dula is not unreasonable, provided that
it is expressed on bases comparable to that conventionally
used to report short rotation yields--above-ground,
leafless, dry matter. Neither is it particularly high.
W hen compared on common bases, the woodgrass yield of 112
wet tonnes of total above-ground biomass per hectare
annually is similar to rates of production measured in
many short-rotation intensive culture trials.
Second-year
production in our wider-spaced treatments (0. 5- and 1. 0-m for H-11; 0.5-m for D-01) is equal to or greater than Dula's estimated annual woodgrass yield. How does productivity of woodgrass compare with that
of wider, short-rotation spacings?
Per-hectare production
during the first season was closely related to spacing,
with the dense woodgrass treatments greatly outproducing
t he wider spacings.
Leaf canopy closure occurred in all
spacings during the second year, and growth per tree and
per hectare accelerated--especially in the wider
spacings. We have already discussed the fact that 2-year
cumulative production of both clones in the 0.5-m spacing
and of H-11 in the 1. 0-m spacing equalled or exceeded that
of woodgrass. What can we expect in terms of growth and
yield in future years? Yield of the third woodgrass
harvest may be somewhat greater than that of the second
harvest, but subsequent yields are lik ly to decline
progressively (cf. 6, 7, 13).
In the wider spacings,
however, tree growth and per-hectare production are
expected to remain at least constant (and most probably
accelerate) for several years.
Thus, yields of the wider
spacings on rotations of three or more years will most
likely surpass cumulative yields from annual harvests of
woodgrass.
11
I.
-
The marked differences between clones in patterns of
bud and branch production and the interaction of these
traits with spacing may have contributed substantially to
yield differences in various clone-spacing treatments.
These findings suggest that patterns of bud and branch
production may be useful criteria for selecting superior
clones, and clonal performance should be screened at
spacings comparable to those at which they will be planted
in operational practice.
What is the potential role of woodgrass in the
production of biomass for conversion to energy? If yield
and cost of production are the primary criteria for
selection of a short-rotation density regime, spacings
other than woodgrass appear superior.
Yields in the wider
spacings are likely to be equal or greater than
woodgrass.
Moreover, establishment costs are
substantially higher for woodgrass.
Differences in
cutting costs alone are tremendous; at 10¢ per cutting,
such costs would be $31, 000 and $11, 000 per hectare for
the two woodgrass spacings as compared to $1000 per
hectare for the 1.0-m spacing.
Even if cuttings were only
1¢ each, total cutting costs per hectare for the
woodgrass spacings would be $3100 and $1100 versus only
$100 per hectare for the 1.0-m spacing--differences still
amounting to $1000 to $3000 per hectare.
Considerable
savings would therefore be needed in other management,
maintenance, harvest, or interest costs to overcome such
differences in establishment costs.
Despite the apparent
disadvantages of woodgrass in terms of yi eld and
production costs, the system could be desirable if
characteristics of the produced biomass were superior in
value to those of biomass grown by other short-rotation
Because of its younger age and smaller size,
systems.
woodgrass will have higher contents of bark, extractives,
nutrients, and moisture and a lower content of cellulose
than an equal biomass produced in a wider spacing on a
somewhat longer rotation (1, 2). Many of these differences
are considered negative traits in various systems of
conversion (3), but they might be beneficial for some
uses.
The initial results from our experiment, and other
current knowledge, therefore, indicate that woodgrass is
less promising than other short-rotation density regimes
for growing Populus biomass for energy.
REFERENCES CI TED
1.
Blankenhorn, P. R. , Bowersox, T. W., Kuklewski, K. M.,
and Stimely, G. Y.
"Effects of rotation, site, and clone
on the chemical composition of Populus hybrids. "
Wood and
Fiber Science 17 (3), 35 1-360.
(1985a).
12
2. Blankenhorn, P. R., Bowersox, T. W., Kuklewski, K. M., Stimely, G. Y., and Murphy, W. K.
com arison of selected fuel and chemical content valves for seven Populus hybrid clones."
vJood and Fiber Science 17(2), 148-158.
(1985b). 3. Butner, R. S., Elliott, D. C., Sealock Jr., L. J., and "Effect of biomass feedstock chemical and Pyne, J. W.
physical properties on energy conversion processes." Draft report submitted to U.S. Department of Energy, Short Rotation 'VJoody Crops Program. 99 p. (November 1987) . 4. Dawson, D. H., "History and organization of the maximum wood yield program, " In: Intensive plantation
culture:
Five years research. USDA For. Serv. Gen. Tech. Rep. NC-21. p. 104 (1976). 5. Dula, J. c., "Woodgrass production for energy application, " In: Argonne National Laboratory, Energy
from Biomass:
Building on a Generic Technology Base, Proceedings of the 2nd Technical Review Meeting.
Wilsonville, OR. April 23-25, 1984. 185 p. (1984). 6.
Geyer, W. A., Melichar, M. W., and Argent, R. M., A centuries old practice crucial to SRIC "Coppicing:
Proceedings, 22nd Annual Meeting of the forestry, " In:
Poplar Council of the United States.
Lawrence, KA.
June 25-27, 1985. p. 1-7. (1985). 7. Harrington, C. A. and DeBell, D. S., "Effects of irrigation, pulp mill sludge, and repeated coppicing on growth and yield of black cottonwood and red alder, " Can. J. For. Res. 14, 844-849 (1984).
8. Heilman, P. E., Peabody, D. V., Jr., DeBell, D. S.,
and Strand, R. F. , "A test of close-spacedr short-rotation
culture of black cottonwood, " Can. J. For. Res. 2,
- 456-459
(1972).
9. Heilman, P. E., and Stettler, R. F., "Genetic
variation and productivity of Populus trichocarpa and its
hybrids. II.
Biomass production in a 4-year plantation,
' 855-859 (1985).
Can. J. For. Res.
10. Herrick, A. M., and Brown, C. L., "A new concept in
cellulose production:
Silage sycamore, " Agric. Sci. Rev.
(4), 8-13 (1967).
11. McAlpine, R. G., Brown, C. L., Herrick, A. M., and
Ruark, H. E., "'Silage' sycamore", Forest Farmer 26(1),
6-7, 16 (1966).
13
12. Ranney, J. W , Wright, L. L., and Layton, P. A.1
"Hardwood energy crops:
The technology of intensive
culture ", J. For. 85(9), 17-28 (1987).
.
13.
Strong, T. F.
"Rotc.-:::ion length and repeated
harvesting influences Po:mlus coppice production. "
?or. Serv. Res. Note NC-XX.
(1988
in press).
USDA
-
Szego, G. C. and Kemp1 C. C.1 "Energy forests and
14.
fuel plantations ", Chemtech (5), 275-284 ( 1973).
15.
Vyas, A. D. and Shen, S. Y. , "Analysis of
short-rotation forests using the Argonne model for
selecting economic strategy (M OSES).
Argonne National
Laboratory Report ANL/CNSV-36, 50 p.
(1982).
14