INFORMATION TO USERS
This reproduction was made from a copy of a document sent to us for microfilming.
While the most advanced technology has been used to photograph and reproduce
this document, the quality of the reproduction is heavily dependent upon the
quality of the material submitted.
The following explanation o f techniques is provided to help clarify markings or
notations which may appear on this reproduction.
1. The sign or “target” for pages apparently lacking from the document
photographed is “Missing Page(s)”. If it was possible to obtain the missing
page(s) or section, they are spliced into the film along with adjacent pages. This
may have necessitated cutting through an image and duplicating adjacent pages
to assure complete continuity.
2. When an image on the film is obliterated with a round black mark, it is an
indication of either blurred copy because of movement during exposure,
duplicate copy, or copyrighted materials that should not have been filmed. For
blurred pages, a good image of the page can be found in the adjacent frame. If
copyrighted materials were deleted, a target note will appear listing the pages in
the adjacent frame.
3. When a map, drawing or chart, etc., is part of the material being photographed,
a definite method of “sectioning” the material has been followed. It is
customary to begin filming at the upper left hand comer of a large sheet and to
continue from left to right in equal sections with small overlaps. If necessary,
sectioning is continued again—beginning below the first row and continuing on
until complete.
4. For illustrations that cannot be satisfactorily reproduced by xerographic
means, photographic prints can be purchased at additional cost and inserted
into your xerographic copy. These prints are available upon request from the
Dissertations Customer Services Department.
5. Some pages in any document may have indistinct print. In all cases the best
available copy has been filmed.

University
Microfilms
International
300 N. Zeeb Road
Ann Arbor, Ml 48106

8424452

M iller, R a y m o n d O y e n

WOODY-BIOMASS PRODUCTION IN MICHIGAN:
CULTURAL INVESTIGATIONS

Michigan S tate University

University
Microfilms
International

300 N. Zeeb Road, Ann Arbor, Ml 48106

SPECIES, GENOTYPE, AND

PH.D.

1984

WOODY-BIOMASS PRODUCTION IN MICHIGAN:
SPECIES, GENOTYPE, AND CULTURAL INVESTIGATIONS
By
Raymond Oyen Miller

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Forestry
1984

ABSTRACT
WOODY-BIOMASS PRODUCTION IN MICHIGAN:
SPECIES , GENOTYPE, AND CULTURAL INVESTIGATIONS
By
Raymond Oyen Miller
A stepwise approach was adopted in 1978 for developing
a comprehensive woody-biomass production system for Michigan.
The program consisted of four phases:

1. Identification of

the most promising biomass species through trial plantings on
abandoned agricultural fields and cleared forest stands,

2.

Preliminary yield comparisons of several species growing in
existing experimental plantations, 3. Species improvement
using standard tree improvement techniques, and 4.
Development of cultural techniques designed to optimize
woody-biomass yield from energy plantations.

This

dissertation summarizes results of research in each of these
areas.
Species recommendations for each of three climatic zones
in Michigan are based on survival and growth of 23 species at
nine old-field sites after four growing seasons.

Pinus

sylvestris, P. banksiana, P. resinosa, P. nigra x P.
densiflora, Larix leptolepis, Alnus glutinosa, and Picea
abies are recommended for use in the Upper Peninsula; Pinus
sylvestris, P. resinosa, P. nigra x P. densiflora, Larix
leptolepis, Alnus glutinosa, Picea abies, and Quercus robur
are recommended for use in the northern Lower Peninsula; and
Pinus sylvestris, Alnus glutinosa, Larix leptolepis, Populus,

Raymond Oyen Miller
Quercus robur, Fraxinus pennsylvanica, and Salix are
recommended for use in the southern Lower Peninsula of
Michigan.
Yield predictor equations were developed and yields
analyzed for 13 species.
older plantations

The best yielding species in the

(14- to 16-years old) were Pinus nigra x P.

densiflora and Betula alleghaniensis.

The best species in

the group of younger plantations (five- to nine-years old)
were a Populus hybrid mixture and Ailanthus altissima.
Seven provenance tests and one clonal test were
evaluated to develop preliminary seed-source and clone
recommendations for Acer rubrum, Betula alleghaniensis, B.
papyriferay Fraxinus americanar F. pennsylvanica, Juglans
nigra, Larix laricina, and Populus hybrids.

Test plantation

ages corresponded to the rotation lengths of short-rotation,
biomass plantations.

Gains in single-tree, oven-dry biomass

yield of selected over "local" sources can be as high as
37 percent with proper provenance selection.

Yield losses

due to improper pro ve na nc e selection can be as much as 48
percent of "local" source yield.

the memory of my brother
Erling John Miller
(1957-1978)

ACKNOWLEDGMENTS
The author thanks the members of the Guidance
Committee— Drs. J. Hanover

(Chairman), D. Dickmann, C. Ramm,

and J. Flore— for their assistance during the course of this
study.
Thanks are also due to those who assisted in the
collection of data and in the editing of this work— G. Howe,
A. Anderson, and M. Brady.
Special thanks are due to my family, and especially my
wife Dorothy,

for their patience and support during the years

of this project.
Financial support for this study was provided by the Oak
Ridge National Laboratory and the United States Department of
Energy under subcontract number 9053.

iii

TABLE OF CONTENTS
Page
LIST OF T A B L E S .......................................... vi
LIST OF F I G U R E S .......................................... ix
Chapter

1.

Woody-Biomass Production for Energy in
M i c h i g a n .....................................

1

Introduction ................................
1
Wood use in M i c h i g a n .......................... 3
The advantages of "energy plantations" . . . 4
Environmental concerns with "energy
plantations" ................................ 5
Production systems for Michigan
........... 6
Literature cited ............................ 3
2.

Short-rotation, Biomass-species Screening on
Old-field Sites in Michigan; Four-year
Results
......................................

9

Abstract ..................................... 9
Introduction ................................
9
Materials and methods
...................... H
Results and discussion ...................... 13
Literature cited ............................ 26
3.

Biomass Predictor Equations for Eleven Shortrotation Biomass Species Growing in Southern
M ichigan P l a n t a t i o n s .........................27
Abstract ......... . . . . . . . . . . . . . 2 7
Introduction ................................ 27
Materials and m e t h o d s ...................... 28
Single-tree oven-dry weight
d e t e r m i n a t i o n s ............................. 30
Model d e v e l o p m e n t ........................ 33
Results and discussion ...................... 35
Literature cited ............................. 40

4.

Biomass Yield Comparisons of Eleven Promising
Short-rotation Species in Southern Michigan
.42
A b s t r a c t ..................................... 42
I n t r o d u c t i o n ............
42
iv

V.'

Chapter
4.

Page
(Continued)
.....................44
Materials and methods
Results and discussion ...................... 48
Literature cited ............................. 53

5.

Spacing Effects on Biomass Yield of Young
Plantations of Ailanthus altissima and
Platanus occidentalis in Southern
M i c h i g a n ...................................... 55
Abstract . . . . . .
........................ 55
I n t r o d u c t i o n .................
..55
57
Materials and m e t h o d s ..........
A i l a n t h u s ................................ 57
Platanus ........... . . . . .
58
Results
..................................... 60
A i la nt hu s
................................ 60
Platanus ................................... 63
D i s c u s s i o n .................
63
Literature cited ............................ 67

6.

Provenance and Clonal Performance of Eight
Short-rotation Biomass-species in Southern
Michigan ....................................... 69

7.

Abstract ..................................... 69
I n t r o d u c t i o n .................
69
Materials and methods
...................... 73
Results and discussion ...................... 75
Red m a p l e ................................. 77
Yellow birch .............................. 60
White b i r c h .............................. 60
White a s h ................................. 61
Green a s h ................................. 62
Black w a l n u t .............................. 82
. . . . . 8 2
T a m a r a c k ..........
Hybrid poplar
............................ 83
C o n c l u s i o n ................................... 85
Literature c i t e d ............................. 86
Summary and Recommendations for Future
Research , . ...................................88

v

LIST OF TABLES
Table

Page

2.1.

A summary of characteristics of the nine Old­
field sites chosen for species trial plantations
in Michigan........................................... 12

2.2.

Species included in old-fieldr biomass production
t r i a l s ................................................13

2.3.

Analysis of variance for height growth of biomass
species in three climatic zones in Michigan after
4 growing seasons..................................... 16

2.4.

Kruskal-Wallis one-way test for differences in
survival among species in each of three climatic
zones in M i c h i g a n .................

17

2.5.

Average height and survival for the best seven
species and the best single species, and the best
hardwood and conifer plot in each climatic zone
in Michigan after 4 growing seasons................. 18

2.6.

Best performing species in old-field screening
trials in Michigan, showing superior traits of
each s p e c i e s ......................................... 20

3.1.

Species for which biomass predictor equations
were developed . . . . . . . . . . . . . . • • • * 2 9

3.2.

Characteristics of plantations from which trees
were sampled to develop yield predictor
equations ........................................... 31

3.3.

Parameters used for comparing regression
performance of yield predictor e q u a t i o n s ......... 34

3.4.

Regression models chosen for predicting wholetree, oven-dry biomass of eleven short rotation
species growing in southern Michigan test
p l a n t a t i o n s .......................................... 36

3.5.

Relative distribution of biomass within sampled
trees of eleven promising species
................ 38

4.1.

Species for which yield projections were made .
vi

. .45

Table

Page

4.2.

Plantations used to develop yield projections for
eleven promising biomass-species in southern
M i c h i g a n ............................................ 47

4.3.

Whole-tree, oven-dry biomass yields and areal
projections for eleven biomass species growing in
southern Michigan plantations ...................... 49

4.4.

Summary of yields reported by various authors for
biomass-species grown in north temperate
latitudes under low-intensity plantation
c u l t u r e .............

50

5.1.

Experimental design of ailanthus spacing trial at
.................... 59
Russ Forest, Decatur, Michigan

5.2.

Analysis of variance of spacing effects on
single-tree and areal biomass yields of ailanthus
five years after establishment in southern
Michigan
.....................

61

5.3.

Mean annual increment yields of ailanthus planted
at three spacings on Russ Forest, Decatur,
Michigan, at age f i v e ............................... 62

5.4.

Mean annual increment yields of sycamore planted
at two spacings at Kellogg Forest, Battle Creek,
Michigan, at age f i v e ............................... 62

5.5.

Estimated yields of sycamore and ailanthus at
various spacings assuming; 1. constant single­
tree yields and 2. 80% survival at age
five.. . . 65

6.1.

Estimated gains which may be achieved through
provenance, family, and individual-tree selection
of white spruce and jack pine in the Lake
States
.............................................. 72

6.2.

Species for which provenance recommendations
were m a d e .....................

74

6.3.

Plantations used for genotype evaluation of
biomass production of eight promising biomassspecies in southern Michigan
...................... 76

6.4.

Biomass yields of selected provenances and clones
as compared to "local" sources in genetic test
plantations of eight species in Michigan
......... 78

vii

Table
6.5.

Page
A summary of seed source recommendations for
the better biomass-species, for use in southern
M i c h i g a n .............................................79

viii

LIST OF FIGURES
Figure
2.1.

Page
Biomass production strategies ......................24

ix

CHAPTER 1
Woody-Biomass Production for Energy in Michigan

INTRODUCTION
The most predictable aspect of the world energy
situation is that sometime early in the next century, the
finite reserves of fossil fuels will inevitably shrink to
insignificance (Pollard,

1976).

One strategy for

alleviating the repercussions of this profound development
is the replacement of fossil fuels with non-fossil,
non-nuclear energy sources.

Several such alternatives are

now available; solar, geothermal, hydroelectric, wind, and
biomass.

All these sources are not only renewable but are

also clean from the pollution standpoint and do not disrupt
the temperature-regulating carbon dioxide balance in the
earth's atmosphere,

as does the burning of fossil fuels.

All five of the alternate energy sources mentioned above
can be used to generate electricity and/or heat, but only
biomass offers the additional benefit of being an alternate
carbon compound source.

This fact makes biomass an extremely

attractive substitute for fossil fuels in a society such as
ours, which is heavily dependent on synthetic polymer
products, liquid petroleum derivatives, and gaseous fuels.
Biomass, as wood fiber, contributes more than any other
non-fossil fuel to the global energy budget (Deudney and
Flavin, 1983).

In the United States, wood energy, as

firewood and mill and logging residues, currently provides
two to three percent of the total energy consumption, or 1.5
to 2.4 quads!./ (Waldrop,

1981; Yarosh,

1983).

This is

roughly equivalent to the energy provided from nuclear fuels.
The production of woody-biomass in "energy plantations",
to supplement the more conventional production techniques now
being used, has been investigated across the country since
the early 1970s.

Research scale plantations have been

yielding an average of 5 to 9 OD Mg ,ha” ^,yr”^ !/.

It has

been estimated that the use of genetically improved materials
and improved cultural practices in these plantations will
ultimately result in average yields of 15 to 20 OD Mg*ha“^*yr
by the year 2030 (Cannell and Smith, 1980; Van Hook and
Ranney,

1983).

If "energy plantation" grown materials are

combined with current supplies of woody-biomass, it may be
possible for wood to contribute 8 to 15 quads, or 8 to 15
percent, of the United State's energy requirements, by the
year

2030.
The state of Michigan is richly endowed with biomass and

a potential for vastly increasing its production,
particularly in "energy plantations."

This paper will use

Michigan as an example of how woody-biomass production is
currently used for energy, and as a case study of the
contribution of "energy plantations" to future energy
production.
1/Quadrillion British Thermal Units.
^/Oven-dry Megagrams (Metric tonnes) per hectare-year.

3

WOOD USE IN MICHIGAN
Fifty-three percent of Michigan's land area, or about
7.8 million hectares,

is forested.

7.0 million hectares are

classified as commercial forest land.

Because of this large

forest area, there is c u r r e n t l y an excess supply of wood in
the state, and in fact Michigan is a net wood exporter (Blyth
and Smith,

1980;

Vodak,

et al. 1981).

The timber industry in Michigan is primarily composed of
seven major fiber and paper companies, and about 300 small
and medium-sized sawmills (Zollner, 1980).

The home firewood

market accounts for a major portion of the lower quality
hardwood sales, and in some portions of the state is in
direct competition with pulpwood buyers.

Finally, there are

several new wood-electrical generating plants completed or
under construction, which will also create a substantial
demand for wood in this state.
At present, the demand for wood in Michigan has not
exceeded the supply, except in isolated areas.

This

condition is not likely to persist as more woody-biomass
energy production systems are introduced into the state, and
as traditional

forest product industries expand production.

Some paper manufacturers guard against this type of supply
shortage by establishing fiber plantations near their mills.
From this standpoint alone,

it is only logical that biomass-

energy producers should do the same.

4

THE ADVANTAGES OF "ENERGY PLANTATIONS"
Woody-biomass presents many advantages over current
energy systems:
burns cleanly,

it is low in sulfur and h e av y m e t a l s so it
it has a much smaller impact on the carbon

dioxide balance in the atmosphere, it presents no hazardous
waste disposal problems,

it is compatible with current

furnace and handling system designs, and it is renewable.
Woody-biomass can be produced on sites which are unsuitable
for conventional agriculture, and can be grown to ameliorate
existing pollution problems.
"Energy plantations" offer the additional benefits of
providing constant, reliable sources of woody-biomass with
controlled physical and chemical characteristics.
Plantations can be located close to the energy producer, and
be managed intensively to improve the productivity of each
hectare.

Productivity can be optimized through the proper

selection of species and genotypes for each site, and by
management practices specifically designed for short, highyielding rotations.

This fact is clearly demonstrated by

comparing the average productivity of commercial forest land
in Michigan of 1.1 OD Mg*ha"^*yr"l at age 50 (Hahn,

1982) to

that obtained in research plantations of a hybrid pine at
Michigan State University of about 8 OD Mg*ha”1*yr”1 at age
16 (Chapter 4).

These high yields were obtained with

cultural techniques which provided for weed control but did
not use extraordinary measures such as fertilization or

5
irrigation.
"Energy plantations" offer the only reasonable way of
incorporating improved genetic materials into the available
growing stock.

The importance of tree improvement in the

production of woody-biomass has been demonstrated repeatedly.
Studies at Michigan State University have demonstrated gains
of 8 to 37 percent over local seedlot yields through
provenance selection (Chapter 6).

Additional gains can be

obtained from continued breeding programs of the more
promising

species.

ENVIRONMENTAL CONCERNS WITH "ENERGY PLANTATIONS"
In 1979, the United States Department of Energy assessed
the environmental

impacts of wood-energy.

The report lists

five areas of environmental concerns:
1.

air pollution from particulates, hydrocarbons,
and carbon monoxide emissions.

2.

residential fires caused by wood stoves.

3.

soil erosion and stream sedimentation from
increased harvesting.

4.

nutrient depletion due to increased organic
matter removal.

5.

ecosystem impacts associated with extensive
forest harvesting.

The list of possible problems has been expanded by
Pimentel, et al. in 1984 to include:
6.

Wildlife habitat conversion.

6
7.

Reduced diversity of the plant communities
through monoculture and the resultant increase in
pest epidemics.

8.

Various economic impacts which would result from
the removal of cropland from agricultural
production.

The importance of the problems described above cannot be
overemphasized.

Protection of the forest ecosystem is of

paramount importance if silviculture is to avoid the
problems with which American agriculture is now faced;
chronic resource depletion.

namely

The social and economic demands

of a pluralistic society such as ours must also be
considered.

Past experience has shown that these needs can

have a stronger impact on public p o l i c y than a body of
scientific evidence.

Foresters and researchers working with

woody-biomass production systems must consider the
implications of their actions in a broad context.

It must be

emphasized, however, that none of these problems are
insurmountable.

Prudent application of available

technologies, coupled with research and education programs,
will help reduce negative impacts.

When the positive

contributions of energy from woody-biomass are considered
fully,

they most certainly outweigh the negative.

PRODUCTION SYSTEMS FOR MICHIGAN
Research conducted at Michigan State University and
elsewhere in the Lake States has provided the groundwork for

short-rotation "energy plantation" establishment.

Species

have been identified for various regions of the state,
inventory techniques have been developed, yields have been
projected for some species at various spacings, and tree
improvement programs have begun for many species.

This

dissertation will present the results of a major portion of
the woody-biomass research which was conducted at Michigan
State University over the last six years.
The research program was structured to:

first,

identify

the better species for biomass production in Michigan;
second,

to screen large numbers of families of these superior

species in large provenance/progeny tests; third, to identify
the genotypes which respond well to short-rotation culture
from among the better families in the provenance/progeny
tests; and fourth, to provide demonstration plantings to
encourage the application of research findings.

This

approach is recommended in other states as a comprehensive
way of developing "energy plantation" systems.

The urge to

take short-cuts at various stages is strong, but incorrect
decisions, based on incomplete information, can postpone the
day when successful commercial "energy plantations" are the
norm rather than just a theory.

8
LITERATURE CITED
Blyth, J.E. and W.B. Smith. 1982.
Pulpwood production in the
North-Central region by county, 1980.
USDA For. Serv.
Resource Bull. NC-59.
Cannell, M.G.R. and R.I. Smith. 1980.
Yields of minirotation
closely spaced hardwoods in temperate regions:
review
and appraisal.
For. Sci. 26(3):415-428.
Deudney, D. and C. Flavin.
1983.
Renewable Energy:
power to choose. W.W. Norton.
NY.
431pp.
Hahn,

The

J.T. 1982.
Timber resource of Michigan's southern
Lower Peninsula, 1980.
USDA For. Serv. Resource Bui.
NC-66.

Pimentel, C., C. Fried, L. Olson, et al.
1984.
Environmental and social costs of biomass energy.
B ioScience 34(2):89-94.
Pollard, W. G. 1976.
The long range prospects for solarderived fuels.
Amer. Sci.
64:509-513.
US Department of Energy. 1979.
Environmental readiness
document:
Wood combustion.
National Technical Info.
Serv.
Springfield, VA.
Van Hook, R.I. and J.W. Ranney. 1983.
Short rotation
intensive culture of hardwoods for wood energy
feedstocks.
An unpublished summary presented to US
Dept, of Energy, 1983.
Vodak, M.C., V.J. Rudolph, and J.T. Olson.
1981. An
evaluation of Michigan's forest cultivation program.
Mich. State Univ. Ag. Expt. Station Research Report No.
417.
Waldrop, M. M.
211:914.

1981.

Wood:

Fuel

of the future.

Sci.

Yarosh, T. 1983.
National fuelwood outlook.
Unpublished
presentation at "Wood Energy and its Challenges", SAF
meeting, Midland, MI.
Sept., 1983.
Zollner, J.A. 1980.
Michigan directory of primary wood
using companies.
MI Dept, of Nat. Res., For. Mgnt. Div.
Lansing, MI.

CHAPTER 2
Short-rotation, Biomass-species Screening on Old-field
Sites in Michigan; Four-year Results

ABSTRACT
Twenty-three species were tested on nine, abandoned,
agricultural fields throughout the state of Michigan to
determine which were best suited to maximum woody-biomass
production.

Planting recommendations for each of three

climatic zones in Michigan were made based on survival and
growth after four growing seasons.

Pinus sylvestris, P.

banksiana, P. resinosa, P. nigra x P. densiflora, Larix
leptolepis, Alnus glutinosa, and Picea abies are recommended
for use in the Upper Peninsula; Pinus sylvestris, P.
resinosa, P. nigra x P. densiflora, Larix leptolepis, Alnus
glutinosa, Picea abies, and Quercus robur are recommended for
use in the northern Lower Peninsula;

and Pinus sylvestris,

Alnus glutinosa, Larix leptolepis, Populus, Quercus robur,
Fraxinus pennsyl vanica, and Sal ix are recommended for use in
the southern Lower Peninsula of Michigan.

INTRODUCTION
Rising costs of petroleum and increasing concern about
the environmental effects of nuclear and fossil fuels have
prompted a search for alternate sources of energy.

Woody-

biomass is one of several alternatives which has been under
development as a replacement for solid, liquid, and gaseous
fossil fuels (Kelsey and Shafizadeh,

1979;

Inman,

1977).

10
Biomass used in each of these processes must meet certain
chemical, structural, and economic criteria which are often
dependent on the species and genotypes used and the
conditions under which it is grown.
The characteristics of woody-biomass can be best
controlled when trees are grown in plantations.

The use of

plantation culture allows the forester to control species
composition and genetic make-up of the stand more closely
than other regeneration systems.

In addition, plantation

management can improve yields, thus reducing the overall land
base required to produce a fixed amount of biomass (Inman,
1977).

The concept of producing pulpwood on short rotations

using intensive management practices was first proposed
eighteen years ago (McAlpine, et al. 1966).

Since then,

this

type of management has been applied to the production of
biomass-energy

(Einspahr,

1972;

Inman,

1977;

Schreiner,1970).

This project was designed to provide a preliminary
assessment several promising species for short-rotation
"energy-plantations" on abandoned agricultural fields (Old­
fields).

Old-fields were chosen because they are well suited

to mechanized planting and maintenance practices, and because
this type of land is common throughout the state.

11

MATERIALS AMD METHODS
Nine, old-field planting sites were chosen throughout
Michigan to represent a range of climates and soil types.
The sites were grouped into zones of similar climatic
conditions.

The Upper Peninsula

(U.P.) was designated Zone

1, the northern half of the Lower Peninsula

(northern L.P.),

was designated zone 2, and the southern half of the Lower
Peninsula

(southern L.P.) was designated Zone 3.

A summary

of the characteristics of the sites within each zone appears
in Table 2.1.

In general,

the shortest growing seasons and

coldest average temperatures are found in the U.P.

Growing

seasons are longer and the temperatures are warmer at the
more southerly latitudes.

Soil texture varies within each

zone.
Species were selected for testing on the basis of their
growth habits and site requirements (Dickmann, 1975; Vail,
1979).

Containerized seedlings were grown in Michigan State

University greenhouses, following procedures developed by
Hanover, et al. (1976).

All species could not be tested at

each site, but each species was planted in at least one
plantation in each zone.

The 23 species tested are listed in

T ab le 2.2.
Sites were cleared of brush and debris, and strips were
sprayed with Amitrol-T herbicide.
in the spring of 1979

Mine species were planted

(Table 2.2), and the remaining fourteen

were planted in the spring of 1980.

A sub-soiling, tree

12

Table 2.1.

Nearest
Town

A summary <
of characteristics of the nine old-field sites
chosen for species trial plantations in Michigan.

North
latitude

West
longitude

Soil texture
class

Frost free
days

(degrees.minutes)
1/

1/

Mean
annual
temp.
3/

ZONE 1 (Upper Peninsula)
Matchwood

46.35

89.25

Clay

120 days

40°F

Chatham

46.20

86.55

Sandy loam

100

42

Rexton

46.10

85.15

Sandy loam

120

41

Oak Ridge

46.20

84.10

Clay

140

41

ZONE 2 (northern Lower Peninsula)
Wellston

44.10

86.00

Sand

140

45

Tawas City

44.15

83.35

Sandy clay
loam

120

45

ZONE 3 (southern Lower Peninsula)
Six Lakes

43-30

85-10

Loamy sand

140

47

East Lansing

42.30

84.30

Sandy clay
loam

150

48

Augusta

42.25

85.20

Sandy loam

160

49

J_/ - Texture analysis performed by Michigan State University Soil Testing
Service.
2/ - Hill and Mawby, 1954.
3/ - Prom: Mean Temperature Maps, 1975*

Table 2.2. Species included in old-field, biomass production
trials.
COMMON NAME
Norway maple*
Ailanthus
European alder
Green ash*
European white birch
Honeylocust
European larch*
Japanese larch
Siberian Larch
Norway spruce
White spruce*
Jack pine
Kellogg hybrid pine
Red pine
Eastern white pine
Scotch pine
Sycamore
Hybrid poplar
Bigtooth aspen*
Trembling aspen*
English oak
Black locust*
Willow

LATIN NAME
Acer platanoides L.
Ailanthus altissima (Mill.) Swingle
Alnus glutinosa (L.) Gaertn.
Fraxinus pennsylvanica Marsh.
Betula ipendula Roth.
Gleditsxa tna ca nt ho s L.
Larix decidua Mill.
Larix leptolepis (Sieb. & Zucc.) Gord.
Larix siberlca Lebed.
Picea abies (L.) Karst.
Picea glauca (Moench.) Voss.
Pinus banksiana Lamb.
Pinus nigra Arnold, x
Pinus densiflora Sieb. & Zucc.
Pinus resinosa Ait.
Pinus strobus L.
Pinus sylvestris L.
Platanus occidental is L.
Populus
Populus grandidentata Michx.
Populus tremuioides Michx.
Quercus robur L.
Robinia pseudoacacia L.
Salix

* - These species were planted in the spring of 1979.
(all others were planted in the spring of 1980)

14
planter was used to establish the trees except at site number
four where the trees were hand-planted.

Trees were planted

on two-meter square spacings in a randomized block design.
Each plantation comprised four blocks of 23, single-species,
20-tree, rectangular plots.

Post-planting weed control was

accomplished by mowing between plantation rows and spraying
within rows with simazine (Princep) and glyphosate

(Roundup).

Heights of all trees were measured (to the nearest 5 cm) and
survival was recorded in the fall of 1982.
Height means were computed for all plots at each site.
An analysis of variance in height growth was performed for
each zone to test for differences among species and
plantations.

Plot mean heights were expressed as a percent

of the block mean to eliminate block effects,

and these

adjusted heights were averaged for each plantation within a
zone, then ranked to identify the species which grew best.
Plot survival was also averaged for each species within
each zone.

The Kruskal-Wal 1 is non-parametric one-way

analysis of variance (Steel and Torrie, 1960) was used to
test for significant differences in survival among species
within each zone.
In most cases, the trees in this study were too small to
be considered mature for biomass production, so it was
necessary to derive an estimate of productivity.
Productivity is a function of survival and growth, so plot
survival

(in percent) times average plot height (expressed as

a percent of the block mean) was computed to approximate the

15
way each species might respond in a production plantation
situation.

The seven best species (top 30%)

in each zone,

were selected based on this "index of productivity."

Changes

in species ranking within each zone were noted and trends
identified.
RESULTS AND DISCUSSION
Significant differences in height were found among
species within each zone (Table 2.3).

Significant

differences in survival were also found among species within
each zone (Table 2.4).

A summary of average height and

survival for the best seven species and the best single
species in each zone is presented in Table 2.5.

The

performance of the single best hardwood and conifer plot is
also presented for each zone.

Average heights and survivals

in the U.P. and northern L.P. are low (58 - 73cm height and
58 - 88% survival) but there are examples of plots which
performed much better (133cm height and 100% survival).
Average height growth was generally better in the southern
L.P.

(144 - 194cm) with individual plots growing as much as

384cm.

Most species were more difficult to establish

successfully in the U.P. and northern L.P. than in the
southern L.P.
These investigations have identified species groups for
use in the various climatic zones of Michigan.
Recommendations within each zone can only be made in a
general way, due to limitations of the experimental design.
The seven best species in each zone, based on the "index of

16

Table 2.3.

Analysis of variance for height growth of biomass species in
three climatic zones in Michigan after 4 growing seasons.

Source of variation

Degrees of freedom Mean square (cm)

F ratio

Zone 1. (Uppe:r peninsula)
Plantation
Species
Block (plantation)
Species x plantation
Error
Total

3
16

4986
5416

4 .42 **

12
48
168
247

1832
1224
191

6.41**

Zone 2. (northern Lower Peninsula)
Plantation
Species
Block (plantation)
Species x plantation
Error
Total

1
18

6
18
83
126

1169
2874
2224
306
93

0
9.39**
3.29**

Zone 5* (southern Lower Peninsula)
Plantation
Species
Block (plantation)
Species x plantation
Error
Total

2
16
9
32
139

22780
26574
2642
2228
631

198

** - significance at the 0 .0 1

level of probability.

11 .93**
3-53**

17

Table 2.4.

ZONE

Kruskal-Wallis one-way test for differences in
survival among species in each of three climatic
zones in Michigan.
NO. of CASES

CHI SQUARE VALUE

1

352

190.246**

2

180

126.749**

3

227

152.656**

** - significant at the 0.01 level of probability.

18

Table 2.5.

Zonel/

Average height and survival for the best seven species and
the best single species, and the best hardwood and conifer
plot in each climatic zone in Michigan after 4 growing
seasons.

Average zone height and
survival for the best
seven species single species

Best single plot height
and survival for
conifers
hardwoods

cm

%

cm

%

cm

%

cm

%

1

61

59

71

76

92

100

133

100

2

58

58

73

88

105

100

131

45

3

144

81

194

87

213

95

384

90

\J Zone
Zone
Zone

1 * Upper Peninsula of Michigan
2 * northern Lower Peninsula of Michigan
5 “ southern Lower Peninsula of Michigan

19
productivity", are listed in Table 2.6.

In general, conifers

were the better performing species in both the U.P. and the
northern L.P. while hardwoods such as hybrid poplar, green
ash, and willow only ranked well

in the southern L.P.

plantations.
The analysis of variance showed significant species x
plantation interaction within each zone (Table 2.3).

This

implies a degree of site specificity within the larger
"zones" for some of the selected species.
Scotch pine and jack pine ranked highly throughout
the U.P.
U.P.

The other five species which performed well

in the

(Table 2.6) do show some site specificity, but there

seem to be no patterns which can be linked to the measured
site

characteristics

(Table 2.1).

Scotch pine also ranked h i g h l y on both sites in the
northern L.P.

Jack pine, Kellogg hybrid pine, and English

oak outperformed the other species on the sandy, droughty
site in this zone.

These species did not perform as well as

Japanese larch and European alder on the

other site in the

zone, which had heavier and more poorly drained soils.

These

results are in keeping with expectations

based on the

silvical characteristics of the species;

pines and oaks are

better able to compete on dry sites while alders and larches
can better compete on moist sites.
European alder, Japanese larch and English oak performed
well on all three sites in the southern L.P.

Scotch pine

only performed well in relation to the others on the sandier

20

Table 2.6.

Best performing species in old-field screening
trials in Michigan, showing superior traits of
each species.

Species

Zone — /

Scotch pine
Red pine

1,2,3

1,2

Kellogg hybrid pine

1,2

Norway spruce
European alder
English oak

growth

survival

X

X

1

Jack pine

Japanese larch

Species had Superior
Performance in:

1,2,3

X
X
X
X

1,2
1,2,3

X
X

2,3

X

Green ash

3

X

Hybrid poplar

3

X

Willow

3

X

1/

Zone 1 is the Upper Peninsula, Zone 2 is the northern
Lower Peninsula, and Zone 3 is the southern Lower
Peninsula of Michigan.

21
site in the zone.

It would seem that the pines cannot

outproduce the hardwoods and larches, in terms of survival
and e a rly height growth, on the more fertile areas of the
southern L.P.

Hybrid poplar, ailanthus, and green ash ranked

best in relation to other species on the most southerly and
more fertile sites, but do not perform as well on the coarse
textured soils in the northern part of the zone.

Honeylocust

only ranked highly on the most southerly site where
temperatures were warm and the growing season was longest.
This is probably because honeylocust continues to grow until
the terminal bud is killed by frost, and the longer growing
seasons in the south allow more growth.
The species x site interactions described above are the
only ones which can be clearly identified from this test.

It

will be necessary in future investigations to increase the
number of sites on which tests are located to obtain a better
cross section of the types of sites available for biomass
plantations in Michigan.

This test has provided a

preliminary screening of many species.

Subsequent testing

could be restricted to the better species identified here.
This investigation, and others like it, contain several
sources of experimental error.

Most often these errors arise

because of the large size of the test, but in some cases are
due to unforeseen or uncontrollable complications.

The

following problems were thought to have affected this
investigation:
1.

Seedlings of some species were smaller than others

22
as a result of an attempt to grow and plant all species
simultaneously.

This affects seedling survival and

early growth.
2.

Conditions of planting and weed control varied

among sites.

The tree planter which was used to

establish these plantations has since been deemed
unsuitable for most sites in Michigan.

Also the weed

control applied to these plantations may have been
insufficient for some of the more sensitive species.
3.

Animal damage to seedlings such as deer browsing

and mouse girdling varied among sites and was severe
in some cases.

This type of damage was generally

sustained on hardwood species.
4.

Planting times, and therefore weather conditions,

were different for each site.
5.

The seedlings which were planted were grown in

greenhouses under accelerated growth conditions and
may have responded differently than nursery-grown
stock.

In addition, seedling health varied among

species and sites, to a certain extent.
All these factors tend to increase experimental error which,
in turn, reduces the discriminatory ability of the tests
and may adversely affect the evaluation of species
performance.
High survival and fast growth are traits which all
biomass species should possess.

Only those species

exhibiting both of these traits should be selected.

The

23
experimental error mentioned above may have caused a
reduction in survival or growth for some species.
Consequently,

it is not possible to determine whether

m oderate s u r v i v a l or growth rate in this test is due to
experimental error or inherent species characteristics.
Rather than eliminate species which may have survived and
grown poorly because of poor test conditions (Type II error),
it was decided to include borderline species for subsequent
testing and risk Type I error. Figure 2.1 shows growth
strategies which were observed in this study; note that
average survival may be offset by faster growth and slower
growth may be offset by high survival, each resulting in
superior

production.

Species which exhibited average survival

(such as

Japanese larch) or slower growth (such as red pine) were
identified among those with superior "index of
productivities" and marked for further investigation.

Jack

pine, for example, grew well in the U.P. but had only
moderate survival.

This species usually survives well when

p la nted in these areas and one is led to suspect that the
depression in survival is a result of experimental error
rather than inherent species characteristics.

Therefore,

future investigations of jack pine should concentrate on
improving survival.
Future tests should concentrate on the species
identified by these preliminary studies

(Table 2.6).

Alternate cultural techniques (especially weed control).

MAXIMUM PRODUCTION
^

HIGHER
SURVIV

SUPERIOR —
PRODUCTION

HIGHER
SURVIVAL

N5

FASTER
GROWT

AVERAGE
SURVIVAL

AVERAGE
SURVIVAL

POOR
SURVIVAL
Figure 2.1.

INFERIOR
PRODUCTION

Biomass production strategies.

POOR
SURVIVAL

4>

LOWE
GROWTH

25
seedling production, handling, and planting methods should be
investigated.

Genotype evaluation should be carried out for

all species identified as superior producers in these tests
to identify the best varieties for biomass production.
The genetic base for the species in this test was
limited.

Most species were represented by a small number of

families; sometimes by only one family.

This occurred

because of time constraints imposed at the outset of the
study, brought on by the need to grow all 23 species
simultaneously.

The effect of this narrow genetic base can

be most clearly observed in black locust.

The survival of

this species was poor throughout the test, but in the
southern L.P. black locust was among the top three species
in height growth.

This indicates that the family of black

locust used in this test was difficult to establish but was
an exceptional grower.

Tree improvement efforts are

currently being directed to improve the survival, growth,
and cold hardiness of black locust in Michigan plantations.
Tree improvement programs can lead to faster growing and
better surviving varieties of all species in this study,

in

most cases, these efforts are already under way in the
Forestry Department of Michigan State University.

This area

of research promises to produce substantial gains for all
biomass

species.

26

LITERATURE CITED
Anonymous.
1975.
Mean temperature maps.
Michigan Dept, of
Ag riculture, Michigan Weather Service.
Dickmann, D.I. 1975.
Plant materials appropriate for
intensive culture of wood-fiber in the North Central
Region.
Iowa State Journal of Research
49(3) :281-286.
Einspahr, D.W. 1972.
Wood and fiber production from short
rotation stands.
IN: Aspen Symposium Proceedings,
U.S.F.S. General Technical Report NC-1.
p 45-51.
Hanover, J.W., E. Young, W.A. Lemmien, and M. Van Slooten.
1976. Accelerated-Optimal-Growth:
A new concept in
tree production.
M.S.U. Ag. Expt. Stat. Research Report
No. 317, 16pp.
Hill,

E.B. and R.G. Mawby. 1954.
Types of farming in
Michigan. Michigan State College Agricultural
Experiment Station Special Bui. No. 206.

Inman, R.E. 1977.
Silvicultural biomass farms. Vol. I:
Summary. The MITRE Corporation/ METREK Division, Report
NO. MITRE-TR-7347-V1/LL.
Kelsey, R.G. and F. Shafizadeh. 1979.
Chemical
characteristics of wood residues and implications for
utilization.
IN: Symposium on Harvesting and
utilization opportunities for forest residues in the
northern Rocky Mountains.
U.S.D.A. For. Serv.
McAlpine, R.G., C.L. Brown, A.M. Herrick, and H.E. Ruark.
1966.
"Silage" sycamore.
Forest Farmer 26(l):6-7, 16.
Schreiner, E.J.
1970.
Mini-rotation
For. Serv. Res. Paper NE-174.

forestry.

U.S.D.A.

Steel, R.G. and J.H. Torrie. 1960. Pri nc ip le s and
Procedures of Statistics.
McGraw-Hill Book Co.,
New York. 481pp.

Inc..

Vail, C.W.
1979.
A preliminary screening of woody plants as
biomass crops on energy farms.
Final Report to U.S.
Dept, of Energy for Grant No.
DOE/ET/23124-T1.

CHAPTER 3
Biomass Predictor Equations for Eleven Short-rotation
Biomass Species Growing in Southern Michigan Plantations.

ABSTRACT
Non-destructive, single-tree, above-ground, biomass
yield predictor equations were developed to facilitate yield
estimates and comparisons among, and within species growing
in short-rotation energy plantations in southern Michigan.
Yield models have been developed for eleven, promising
species growing in Michigan, based on measurement of seven
to eight trees within each of eleven plantations.

Trees

were destructively sampled to determine whole-tree, oven-dry
weight and to assess the distribution of biomass within the
tree.

Test plantations were located in southern Michigan

and ranged in age from five to sixteen years.

Field data

required to use these predictor equations are restricted to
tree height and stem diameter (DBH in most cases), for ease
of application.

INTRODUCTION
Wood-energy is becoming important, again,

in the energy

budget of the United States as the price of other fuels
continues to increase.

As wood use increases,

foresters are

being asked to estimate stand biomass in addition to
pulpwood and timber volume.

Traditional forest inventories

have not been sufficient for predicting whole-tree biomass
yields on a single-stem or stand basis, nor have they been

27

28
sufficient for estimating residue biomass (unmerchantable
tops, branches, etc.).

Some investigators have been

developing inventory methods specifically for biomass
determinations by modifying existing stand tables or
generating new stand-biomass predictor tables (Adams, 1982;
Tritton, 1982).

This has facilitated inventories of older,

existing stands for biomass availability reports.
Biomass production in short-rotation, energy plantations
is a concept which has yet to be put into wide-spread
practice, but is being investigated by several research
organizations (USDOE, Gas Research Institute, major
universities,

and others).

The Department of Forestry at

Michigan State University has been conducting short-rotation
species and genotype suitability studies since 1978 with the
U.S. Department of Energy.

During these investigations it

became necessary to obtain non-destructive yield estimations
for a number of species.

Tables or equations for small

plantation-grown trees were unavailable, so investigations
were designed to provide preliminary yield predictor
equations for use in subsequent species and genotype
screening.

MATERIALS AND METHODS
Eleven species were selected (Table 3.1), based either
on their superior performance in previous tests or on their
rapid juvenile growth potential, and on the availability of
young test plantations.

One plantation of each species was

29

Table 3.1.

Species £or which biomass predictor equations
were developed.

Red maple

Acer rubrum L. var. rubrum

Ailanthus

Ailanthus altissima (Mill.) Swingle

Yellow birch

Betula alleghaniensis Britton

White birch

Betula papyrifera Marsh.

White ash

Fraxinus americana L.

Green ash

Fraxinus pennsylvanica Marsh.

Black walnut

Juglans nigra L.

Tamarack

Larix laricina

Kellogg hybrid pine

Pinus nigra Arnold x
P. densiflora Sieb. & Zucc.

Sycamore

Platanus occidentalis L.

Hybrid poplar

Populus

(Du Roi)

30
selected £or measurement.

All plantations are located in

the southern Lower Peninsula of Michigan at one of two
experimental forests owned by Michigan State University.
The plantations had each been established on 2.4m square
spacings (except for black walnut which was planted at a
3.1m square spacing)

and maintained under similar management

regimes, but were of different ages.

The characteristics of

each plantation are given in Table 3.2.
Tree diameters were measured at 137 cm (4.5 feet) above
the ground if the smallest trees in the plantation were
larger than 13 mm (0.5 inches) at this point,

otherwise they

were measured at 61 cm (2 feet) from the ground.

The range

of diameters within each plantation was determined to
establish diameter classes;

2.5 cm (1 inch) classes were used

if the range of diameters was 12 cm (5 inches) or less, and 5
cm (2 inch) classes were used if the range was greater than
12 cm.

A seven- or eight-tree random sample, stratified by

diameter class, was harvested from each plantation for
destructive analysis before leaf fall in 1982.

Each tree was

divided into four components: twig, branch, bark, and bole.
Oven-dry, whole-tree weights for the sample trees were
determined by obtaining and summing the weight of each
component.

Regressions using total height and diameter were

then used to predict whole-tree weight for each species.
Single-tree oven-dry weight determinations
Sample trees were removed from the plantations after

31

Characteristics of plantations from which trees
were sampled to develop yield predictor
equations.

Table 3.2.

Species

No. Trees
Sampled
5

-

Plantation
Age

Type Of
Test 1/

9 Year Old

Red maple

7

9

G

Ailanthus

7

5

S

White birch

8

7

G

White ash

8

6

G

Green ash

8

5

G

Sycamore

8

5

S

Hybrid poplar

8

8

C

15 & 16 Year Old
Yellow birch

8

15

G

Black walnut

8

16

G

Tamarack

7

16

G

Kellogg hybrid pine

7

16

H

1/

G
S
C
H

=
=
=
=

genetic provenance and/or progeny test.
spacing study.
hybrid clonal trial.
bulk hybrid seedling trial.

32
height and diameter were recorded.

Branches and upper stem

portion of the tree (bole diameter less than 2.5 cm or 1
inch) were removed and set aside. The bole was then
sectioned into 61 cm (2 foot)

lengths.

The diameter at the

top and bottom of each section was determined both inside
and outside the bark.

Inside and outside bark bole volume

was determined using the Smalian Log Rule.

The Smalian rule

was chosen to eliminate the need for mid-section diameter
measurements.

Disks were removed from the bottom of each

section and used to make specific gravity determinations of
4

the wood in that section.

Average specific gravity for the

bark was determined by sampling bark from each disc.

All

specific gravity determinations were based on the oven-dry
weight of the wood or bark and its v o l u m e at fiber
saturation point.
The o v e n-dry mass of each section of b o l e within a tree
was determined as the product of the section volume based on
DIB

(diameter inside bark) and it's corresponding specific

gravity.

Total b o l e mass was computed as the sum of the

section masses.

Bark volume was determined as the difference

of bole volume based on DOB (diameter outside bark) and
volume based on DIB.

Bark oven-dry mass was then computed as

the product of bark volume and bark specific gravity.
Branches removed from the main stem of each tree were
further separated in two components;

twigs (current year's

growth including leaves ) , and branches previous years'
growth.

Tamarack and the Kellogg hybrid pine were too large

33
to be handled expediently using these procedures, so branches
and twigs were treated as a single component (branches).
The fresh-weight of each component was recorded in the
field.

Samples were oven-dried to a constant weight at

105°C, and the moisture content was determined.

The oven-dry

weight of each component was computed as the product of
fresh-weight and one minus the moisture content.
Whole-tree biomass was calculated as the sum of all
component biomass.

A summary of the distribution of biomass

within the tree was developed by averaging the component
masses of each sampled tree for each species.
Model development
Several regression models which had been used by other
investigators were chosen for testing (Baskervi1 le, 1972;
Lavigne and Van Nostrand, 1982; et al.).

These models were

linear, non-linear, natural log-transformed, constrained, and
unconstrained models using functions of diameter and/or
height as independent variables to predict whole-tree, above­
ground, oven-dry biomass.

The individual-tree data were used

to solve for the regression coefficients of each model using
a least squares regression approach.

Correction factors for

the log-transformed regressions were calculated as described
by Sprugel

(1982), and incorporated into the "Y intercept"

term of each model.

Models were evaluated based on the five

criteria listed in Table 3.3.

The model which had the best

overall goodness-of-fit using these test criteria was
selected for each species.

Pearson's product-moment

34

Table 3.3.

S

Parameters used for comparing regression
performance of yield predictor equations,

(generalized coefficient of determination) — /
i - <*(y-y)2/£(y-y)2 )

RMSE

2/
(root mean square error)±
£(y-y)2/n

MAE

(mean absolute error) — /
|y-y|/n

MPE

(mean percent error)
$( (y-y)/y)/n
(index of agreement) — /
!-*(y-y)2/*<|y-y|+|y-y|)

Where s

1/
2/

y
y
y
n

=
=
=
=

observed yield in kg.
predicted yield in kg.
average observed yield in kg.
number of observations.

Payandeh (1981).
Willmott (1982) .

35
correlation coefficient (r) and the coefficient of
determination

(r^) were not used in comparing models because

of deficiencies in their ability to adequately predict
model accuracy especially for log-transformed and constrained
models

(Payandeh, 1981; Willmott,

1982).

RESULTS AND DISCUSSION
The regression models which were selected for use are
presented in Table 3.4.

Single-tree biomass yield models

have been shown to be relatively independent of site, age,
and spacing

(Alban and Laidly, 1982; Crow, 1983; and Jacobs

and Monteith, 1981).

This suggests that the models developed

in this study may have utility elsewhere in the Lake States.
Eight of the sampled plantations were genetic test and,
therefore, contained a diversity of genetic material.

This

may act to reduce the applicability of these models in
plantations composed of different or less diverse genetic
material.
The sample size used to develop these yield predictor
equations was limited to seven or eight trees for each
species.

Although this sample size has been used by others

(Albrektson, 1980), there is general agreement that larger
samples should be used for better confidence in the
predictive accuracy of the models

(Grove,

et al. 1982).

Models presented here should, therefore, be considered as
preliminary and used with caution until further sampling can
refine or confirm these results.

Table 3.4*

Regression models chosen for predicting whole-tree, oven-dry biomass of eleven short rotation species
growing in southern Michigan test plantations.
Residual Analysis
n

Species

Regression Model

-1/

S

RMSE

MAE

MPE

D

5 to 9-Year-Old
Red maple

7

Y

=

-2.70445 + 0.485D2 + 0-56793H - 0.03604D2H

-985

•480

•324

.170

•996

Ailanthus

7

Y

-

0.25567D2*01803 - 0.17879H

•911

2.810

1.985

.260

•976

Paper birch

8

Y

*

-2.34234 - 0.17524D2 + 1.01637H + 0.0515D2H

•951

1.482

1.19

.185

•987

White ash

8

Y

-

-0.02285D2 - 0.00642H + 0.05130D2H

•991

.156

.117

.103

.998

Green ash

8

LnY - -0.83160 + 1.93642LnD - 0«30172LnH

•919

.453

.307

.119

•979

Sycamore

8

Y - 0.2777D2,05504 - 0.18999H

•973

.576

.467

.417

•993

Hybrid poplar

8

Y - 0.17783D2

.962

1.698

1.426

.183

•990

15 and 16-Year-Old
Yellow birch

8

LnY - 0.09617 + 2.83563LnD - 1.40931LnH

•972

5.337

3.332

.096

.993

Black walnut

8

LnY - -3.70277 + 0.78726LnD + 2.40604LnH

•944

5-340

3.759

.192

•985

Tamarack

7

Y - 0.57058D2 - 0.8363H - 0.02128D2H

.978

3.646

2.578

.080

.994

Kellogg hybrid pine

7

LnY - -2.23651 + 2.05503LnD + 0.39002LnH

•990

4.128

2.660

.036

•997

\J

Where:

Y - Yield in Kg (oven-dry).
H * Total tree height in meters.
D * Diameter in cm. measured at 1.4m (4*5') above ground - except for white ash which was measured
at 0.67m (2') above the ground.

37
Accurate prediction of tree components' biomass using
conventional mensurational data, such as DBH and height, has
been shown to be difficult (Alban and Laidly, 1982; Alemdag,
1981; Alemdag and Horton, 1981).

For this reason,

and due

to a limited sample size, which would further compromise the
estimates, no attempt was made to develop regressions to
predict bole, branch, bark, or twig and leaf biomass
individually.

A summary of the average proportion of these

components was developed, instead, to provide an
understanding of biomass distribution for the eleven species
in these plantations

(Table 3.5).

For most species, the biomass of the bole is
approximately half of the tree's total biomass.

This

suggests that efforts should be made to harvest more than
just the bole.

Nutrient-rich leaves and twigs make up a

small percent of the total biomass of older trees, but in
some of the younger plantations they represent as much as
33 percent.

Harvesting systems may need to be designed to

accommodate the amount of fine materials (twigs and leaves)
for each species.

Further, if nutrient removal from the

site by harvesting is to be minimized, species which have
large portions of their biomass in leaves and twigs may be
less desirable.
Further investigations will be required to develop
better component estimates for these and other species at
various ages.

Such estimates are needed to make management

decisions regarding the type of harvesting system necessary

38

Table 3*5*

Relative distribution of biomass within sampled trees of
eleven promising species.

Species

Relative proportion of total biomass
Twig

Branch

Bark

Bole

Wood
specific
gravity

(% of whole-tree mass, oven-dry)
Red maple

1%

29%

\0%

54*

0.33

Ailanthus

20

22

5

53

0.42

Paper birch

18

32

4

46

0.36

White ash

33

33

3

31

0.43

Green ash

9

40

4

47

0.47

20

25

2

53

0.42

7

29

10

54

0.33

Sycamore
Hybrid poplar

Year Old Age Group
Yellow birch

5

43

4

48

0.49

Black walnut

4

35

12

49

0.47

Tamarack

—

38

8

54

0.35

Kellogg hybrid pine

—

58

7

35

0.32

to maximize biomass harvested and minimize nutrient removal
from the site

(Kimmins,

1977).

40
LITERATURE CITED
Adams, P.W. 1982.
Estimating biomass in northern Lower
Michigan forest stands.
For. Ecol. and Mgnt. 4:275-286.
Alban, D.H. and P.R. Laidly. 1982.
Generalized biomass
equations for jack and red pine in the Lake States.
Can. J. For. Res. 12:913-921.
Albrektson, A. 1980.
Relations between tree biomass
fractions and conventional silvicultural measurements.
Ecol. Bull.
32:315-327.
Alemdag, I.S. 1981.
Above-ground mass equations for six
hardwood species from natural stands of the research
forest at Petawawa.
Canadian Forestry.Service Inf.
Rep. PI-X-6.
______________ and K.W. Horton. 1981.
Single tree equations
for estimating biomass of trembling aspen, largetooth
aspen, and white birch in Ontario For. Chron.
57:169173.
Baskerville, G.L. 1972.
Use of logarithmic regression in
the es ti mation of p l an t biomass. Can. J. For. Res.
2:49-53.
Crow,

T.R. 1983.
Comparing biomass regressions by site and
stand age for red maple. Can. J. For Res. 13:283-288.

Grove, J.H., J.P. Barrett, and T.G. Gregoire.
1982. When is
n sufficiently large for regression estimation? Jour.
Envir. Mgnt. 15:229-237.
Jacobs, M.W. and D.B. Monteith.
1981.
Feasibility of
developing regional weight tables.
Jour. For.
79 (10):666-667.
Kimmins, J.P. 1977.
Evaluation of the consequences for
future tree productivity of the loss of nutrients in
whole-tree harvesting.
For. Ecol. Mgnt. 1:169-183.
Lavigne, M.B. and R.S. van Nostrand. 1981.
Biomass equations
for six tree species in central Newfoundland.
Newfld.
For. Res. Cent. Inf. Rep. N-X-199.
Panshin, A.J. and C. de Zeeuw. 1970. Textbook of Wood
Technology.
Vol. 1.
McGraw-Hill.
NY. 705pp.
Payandeh, B. 1981. Choosing regression models for biomass
prediction equations.
For. Chron.
57 (5):229-232.
Sprugel, D.G. 1982.
Correcting for bias in log-transformed
allometric equations.
Ecology 63(6):000-000

41
Tritton, L.N. and J.W. Hornbeck. 1982.
Biomass equations for
major tree species of the northeast.
USFS Gen. Tech.
Rep. NE-69.
Willmott, C.J. 1982.
Some comments on the evaluation of
model performance.
Bull. Amer. Meteorol. Soc.
63(11):1309-1313.

CHAPTER 4
Biomass Yield Comparisons of Eleven Promising
Short-rotation Species in Southern Michigan

ABSTRACT
Plantation yield information has traditionally been
available for mature (30- to 80-year-old) coniferous trees
of pulp and timber species.

Increased interest in short-

rotation energy-plantations has created a need for yield
data on young trees (five to fifteen years old) of species
with rapid juvenile growth.
plantations,

Eleven, single-species

from among the genetic and cultural tests at

Michigan State University, were selected for measurement and
yield assessment.

Average, single-tree, above-ground, oven-

dry biomass yield, expressed as a mean annual

increment was

determined for each plantation and projections were made to
approximate the areal yield which could be expected from
large-scale, biomass plantations.
The best yield in the older plantations (fourteen to
sixteen years old) was from the hybrid,
densiflora

(4.64 OD Kg*tree” ^*year“l).

Pinus nigra x P.
The best species

in

the group of younger plantations (five to nine years old) was
a Populus hybrid mixture

(1.13 OD Kg*tree” ^'year“^).

Other

species which also yielded well included Betula
alleghaniensis and Ailanthus altissima.
INTRODUCTION
The ability of a site to produce biomass remains fairly
constant over time, assuming a lack of catastrophic
42

43

disturbance.

Forest management attempts to shift a site's

biomass production to organisms, such as trees, capable of
producing usable products, such as timber or wood fiber
(Smith, 1962).
Cannell and Smith (1980) conducted an extensive review
of "short-rotation" plantation yield investigations, and
concluded that northern temperate sites of good fertility
were capable of producing a maximum of 10-15 oven-dry, metric
tonnes per hectare-year

(OD Mg*ha"^,yr“^) of usable biomass

at age five years, without extraordinary cultural treatments.
These yields were obtained only if the sites were fully
stocked throughout the rotation.
In contrast to the maximum yield postulated by Cannell
and Smith, the average production of commercial forest land
in the United States has been estimated to be only
Mg*ha"1,yr“-*- (Howlett and Gamache,

1977).

1.3 OD

Michigan's forests

are yielding an average of 1.1 OD Mg*ha”^*yr"^ (Bradley,
al., 1980; Hahn,
1982).

1982; Jakes, 1982; Smith,

et.

1982; and Spencer,

In addition, plantations managed to maximize pulpwood

yield in Michigan only produce approximately 3.8 OD M g ’ha"

l.yr~l (Lundgren,

1982).

Research is needed in Michigan to determine ways to
better capture the potential of forest sites so that woodybiomass yields might begin to approach the maximum proposed
by Cannell and Smith.

Michigan State University's Department

44
of Forestry has been working with the United States
Department of Energy and the Michigan Cooperative Tree
Improvement Program to screen potential biomass species and
identify the more promising ones for further

investigation.

Part of the screening process was to examine and compare the
yields of several species to determine which had the highest
biomass productivity rate. The purpose of these
investigations was to make comparisons among species based
on single-tree, whole-tree biomass expressed as mean annual
increment (MAI).

These single-tree yields were also

expanded to an areal basis for comparison to yields reported
by other authors, and to serve as preliminary estimates of
yields from short-rotation energy-plantations in Michigan.

MATERIALS AND METHODS
Energy-plantations should produce high yields on
relatively short rotations.

It is therefore necessary that

the trees used in these plantations grow rapidly during their
early years (Dickmann, 1975).

Eleven, single-species

plantations were selected from among the cultural and genetic
test plantations established by Michigan State University for
measurement and analysis of juvenile growth.

All eleven

species were known to have rapid juvenile growth,
are prodigious sprouters.

and four

Use of species that coppice would

eliminate the need for re-planting after harvesting.
The plantations used in these studies were located on
one of two Michigan State University Experimental Forests,

45

Table 4.1.

Species for which yield projections were made.

Common name

Latin name

Red maple

Acer rubrum L.

Ailanthus

Ailanthus altissima

Yellow birch

Betula alleghaniensis Britton

White birch

Betula papyrifera Marsh.

White ash

Fraxinus americana L.

Green ash

Fraxinus pennsylvanica Marsh.

Black walnut

Juglans nigra L.

Tamarack

Larix laricina

Kellogg hybrid pine
Sycamore

Pinus nigra Arnold x
P. densiflora Sieb. & Zucc.
Platanus occidentalis L.

Hybrid poplar

Populus

(Mill.) Swing!

(Du Roi)

46
located in southern Michigan (Table 4.2).

All plantations

were established and maintained using similar procedures.
Trees were planted on 2.4 x 2.4m spacings (except for black
walnut, which was established on 3 x 3m centers).
Eight of the eleven plantations were genetic tests
which were planted in randomized-block designs with multipletree plots.

In genetic test plantations, only those plots

containing seedlots which could be considered "local" were
included in the analysis,

to minimize the effect of genetic

variability on yield comparisons.

The analysis of genetic

variability was performed separately and is reported in
Chapter 6.
Heights and diameters of all "local-source" trees were
measured in the fall of 1982.

Whole-tree, above-ground,

oven-dry biomass was computed using the yield predictor
equations (Table 3.4), developed for each plantation in
Chapter 3.

Single-tree yield for each species was computed

by averaging the yields of each "local-source" tree.
Species were ranked within each of two age groups according
to calculated single-tree yield.

Grouping by age was done

in an attempt to minimize growth phase and between-tree
competition differences among plantations.

It was not

possible to eliminate yield differences due to site.
Areal yields were derived from the average, single-tree
yields calculated above to provide preliminary comparisons of
these species with yields reported by other investigators.
Estimates were derived for two spacings (2.3 x 2.3m and 1.8 x

Table 4.2.

P la n ta tio n s used to develop y ie ld p ro je c tio n s fo r e le v en promising biom ass-species
in southern Michigan.
Age
(yrs)

Species

Location of planting

Red maple
Ailanthus
Yellow birch
White birch
White ash
Green ash
Black walnut
Tamarack
Kellogg hybrid
pine
Sycamore
Hybrid poplar

Kellogg Forest
Russ Forest
Kellogg Forest
Russ Forest
Kellogg Forest
Russ Forest
Kellogg Forest
Kellogg Forest

9
5
15
7
6
5
16
14

Kellogg Forest
Kellogg Forest
Russ Forest

16
5
8

Type of test

No. Seedlots

Spacing
(m)

Provenance test
Spacing trial
Provenance test
Progeny test
Provenance test
Provenance test
Provenance test
Provenance test

94
bulk
35
224
35
43
19
33

2.4
various
2.4
2.4
2.4
2.4
3-1
2.4

Hybrid trial
Spacing/ interplanting
Clonal trial

bulk
bulk
40

2.4
various
2.4

48
1.8m) at two survival rates (80 and 90 percent).

It was

assumed that all trees would grow at the same rate as the
average-tree of the test plantation.
RESULTS AND DISCUSSION
Single-tree and areal yields computed for each species
are presented in Table 4.3.

Kellogg hybrid pine and yellow

birch are the two better yielding species in the older
plantation age group.

It may be possible to obtain between

6.9-12.2 and 4.1-8.2 OD Mg ,ha“1*yr"1, respectively,

in

energy-plantations of these species on 16-year rotations.
That is equivalent to 65.5 to 195.2 dry Mg per hectare or
29.2 to 87.1 dry tons per acre, at a rotation age of 16
years,

under non-intensive management practices.

Ailanthus and hybrid poplar were the two species which
had the highest yields of the younger plantations.
possible

to obtain between

It may be

1.4-2.9 and 1.5-3.0 OD Mg*ha~1*yr“1 ,

respectively, in energy-plantations with these species on
six—year rotations.

These yields are equivalent to 8.4 to

18.0 dry Mg per hectare or 3.7 to 8.0 dry tons per acre, at
a rotation age of six years, under non-intensive management
practices.

All plantations in this younger age group had

not yet fully occupied the planting sites.

Planting

densities might be increased to improve yields on rotations
of five to nine years.
A summary of yields reported by other investigators is
compared to yields estimated here in Table 4.4. Higher
yields have been obtained by some, but often the conditions

Table 4.3*

W hole-tree, oven-dry biomass y ie ld s and a re a l p ro je c tio n s fo r eleven biomass sp ecies growing in
southern Michigan p la n ta tio n s .
Projections Of Areal Yields
1.8 x 1.8m
-Spacing2.4 x 2.4m

Species

Age
(yrs)

No. Trees
Sampled

Mean Tree
Yield

-Survival80*

90*

80*

90*

gruuy
(Oven-Dry Mg’ha”1*yr”1) -

(Kg*tree”1*yr"1)
Kellogg
hybrid pine
Yellow birch
Tamarack
Black walnut

16
15
14
16

280
44
92
32

4.64
3*05
2.31
1.62

11.1
7.3
5-5
3-9

12.5
8.2
6.2
4.4

6.2
4.1
3.1
2.2

7.0
4*6
3-5
2.5

3.0
2.9
1.9
1.8
1.8
0.9
0.4

1.5
1.4
0.9
0.9
0.9
0.5
0.2

1.7
1.6
1.1
1.0
1.0
0.5
0.2

5 to 9-year-old group
Hybrid poplar
Ailanthus
White birch
Sycamore
Red maple
Green ash
White ash

7
5
7
5
9
5
6

1065
309
649
142
80
40
100

1.13
1.06
0.70
0.68
0.67
0.34
0.15

2.7
2.5
1.7
1.6
1.6
0.8
0.4

Table 4.4. Summary of yields reported by various authors
for biomass-species grown in north temperate
latitudes under low-intensity plantation culture.

Species

(yr)

Kellogg hybrid pine
European black alder
Cottonwood
Black locust
Yellow birch
Silver maple
Aspen
European white birch
Red pine
Sycamore
Hybrid poplar
Ailanthus
Jack pine
1/

16
6
6
6
15
6
7
6
60
5
7
5
13

(Mg*ha“**yr“^)
12.5
10.3
8.8
8.7
8.2
7.5
4.3
3.9
3.8
3.5
3.0
2.9
2.6

Site
Michigan
Indiana
Kansas
Kansas
Michigan
Kansas
Maine
Norway
Michigan
Indiana
Michigan
Michigan
Michigan

1-■ Frivold and Borchgrevink , 1981.
2- Callahan and Toth, 1978.
3- Howlett and Gamache, 1977.
4- Geyer and Naughton, 1981.
5- Taken from this paper.
6- Howe, 1983.
7- Lundgren, 1982.

Citation— /
5
2
4
4
5
4
3
1
7
2
5
5
6

51
under which these yields were obtained are extraordinary.
For example, fertilizers and irrigation schedules are often
used to ensure non-limiting nutrient and water conditions on
test sites.

Plantations are commonly established on "prime"

agricultural land or are afforded the type of cultural
treatments one would expect of an agricultural crop.
Comparisons have been limited to trees grown under nonintensi ve management strategies similar to those applied to
the test plantations of this study.

The assumption is that

lower management costs of non-intensive systems will be more
attractive to managers than the higher yields which result
from the more expensive, intensive systems.
It is obvious from data in Table 4.4 that further
cultural and genetic work is needed to improve the yields of
species such as ailanthus, hybrid poplar, aspen and others,
so that they will approach the theoretical maximum of 10-15
OD Mg*ha_1*yr_1 proposed by Cannell

and Smith

(1980).

The plantations measured in this study ranged in age
from 5- to 16-years, so that each species was at a different
point along its "Grand Period of Growth" curve

(Baker, 1950).

Interspecies comparisons have been restricted to plantations
of similar ages (7 +/- 2 years and 15 +/- 1 year)
attempt to minimize these differences.

in an

This approach is

reasonable if one assumes that rotation length will be set by
economic or operational factors and selections must be based
on biological efficiency at a common age and spacing.
The next step, for the better species identified in

52
th^se investigations, will be to measure MAI on different
spacings for several years, to determine the yields of
different combinations of spacings and rotation lengths.
These data will provide the basis for management
prescriptions for each species.
Yield information reported here is highly specific to
the ages, stocking levels, and sites of the test plantations
and therefore should be used only as preliminary guidelines
for species selection.

Extreme caution should be used when

projecting the yields reported here to other sites and other
management

systems.

53

LITERATURE CITED
Baker, F.S.
Hill.

1950.
Principles of Silviculture.
NY.
414pp.

McGraw-

Bradley, D.P., et. al. 1980.
The supply and energy
potential of forest resources in northern Wisconsin and
Michigan's Upper Peninsula.
USDA Forest Service Res.
Paper NC-182
21pp.
Callahan, J.C. and J.M. Toth.
1978.
Short-rotation fiber
production in Indiana yields and potential uses. Purdue
U n i v e r s i t y Ag. Expt. Sta. Bui. No. 194 38pp.
Cannell, M.G.R. and R.I. Smith. 1980. Yi e l d s of
minirotation closely spaced hardwoods in temperate
regions:
Review and appraisal.
For Sci. 26(3) :415-428.
Dickmann, D.I.
1975.
Plant materils appropriate for
intensive culture of wood-fiber in the North Central
Region.
Iowa State Jour, of Res.
49(3):281-286
Frivold, L.H. and I. Borchgrevink.
1981.
Biomass yield of
silver birch (Betula verrucosa Ehrh.) in a 6 years old
trial plantation at As, Norway.
(in Norwegian, summary
in English). Scientific Reports of the Agricultural
University of Norway 60:12, 17pp.
Geyer, W.A. and G.G. Naughton. 1981. Short rotation
forestry biomass yields and cost analysis in eastern
Kansas.
IN:
Proceedings...1981 Southern Forest Biomass
Workshop. Ed: C.A. Gresham.
Hahn,

J.T. 1982.
Timber resource of Michigan's southern
Lower Peninsula, 1980.
USDAFS Resource Bui. NC-66.
118pp.

Howe,

G.T. 1983.
Genetic evaluation of jack pine half-sib
families collected from the Lower Peninsula of Michigan.
IN: 1983 Annual Technical Report to Oak Ridge Natnl.
Lab. for Short Rotation Woody Crops Program sub-contract
no. 9053.
Michigan State University.

Howlett, K. and A. Gamache.
1977.
Silvicultural Biomass
Farms.
Vol. II.
The potential of short-rotation farms.
MTR-7347 (Vol. 2) The Mitre Corp., McLean, VA.
Jakes, P.J. 1982.
Timber resource of Michigan's northern
Lower Peninsula, 1980.
USDAFS Resource Bui. NC-62.
120pp.

54

Lundgren, A.L. 1982.
Can red pine in the Lake States
outproduce loblolly and slash pine in the South?
IN:
Proc. Artificial Regeneration of Conifers in the Upper
Great Lakes Region.
Mich. Tech. Univ., Houghton, MI.
Ed: Mroz, G.D. and J.F. Berner. p337-344.
Smith, D.M. 1962.
The Practice of Silviculture.
W i l e y and Sons, Inc. NY. 578pp.

John

Smith, W.B. 1982.
Timber resource of Michigan's eastern
Upper Peninsula, 1980.
USDAFS Resource Bui. NC-64.
102pp.
Spencer, J.S.
1982.
Timber resource of Michigan's western
Upper Peninsula, 1980.
USDAFS Resource Bui.
NC-60.
102pp.

CHAPTER 5
Spacing Effects on Biomass Yield of Young Plantations of
Ailanthus altissima and Platanus occidentalis
in Southern Michigan

ABSTRACT
Ailanthus altissima was planted at three spacings (1.2 x
1.2m, 1.8 x 1.8m, and 2.4 x 2.4m) and Pla ta nu s occidental is
was planted at two spacings (2.4 x 2.4m and 2.4 x 4.9m) to
determine the effect of planting density on biomass yield.
Both species showed increased areal yields at higher planting
densities at age 5.

Ailanthus produced 6 Mg*ha“^*yr“^ at

the 1.2 x 1.2m spacing and Platanus yielded 0.8 Mg*ha“^*yr”^
at the 2.4 x 2.4m spacing at age five.

Neither species

showed significant reduction in single-stem biomass yields
at the higher densities,

indicating that upper density

thresholds had not yet been reached and that maximum yields
at age five may be obtained at higher densities than those
tested

in this study.

INTRODUCTION
Rising petroleum prices have been responsible for a
gradual but dramatic increase in the use of wood for fuel in
recent years.

As the demand on the wood resource grows,

forest managers will be required to modify existing
silvicultural practices to increase yields of biomass-fuel.
An important consideration in the production of biomass-fuel
will be to maximize areal yields and thus reduce land-base

55

56
requirements.

This will involve using selected species and

genotypes in short-rotation, high density plantations.
Each species has an optimum range of planting densities
for a particular site for which biomass production is
maximized.

The "law of constant final yield" (Hozumi, et

al. 1956) states that:

Tree biomass yields of fully-stocked

stands are independent of the numbers of stems per hectare
over a wide range of spacings.

Thus, the productive

potential of the site is distributed among the individuals
which are present (Cannell, 1980; Ek and Dawson, 1976; Lloyd
and Jones, 1982).

The stocking levels common to current

forestry practice are designed to maximize production of
traditional forest products such as pulpwood and sawlogs, and
not to maximize biomass productivity.

Spacings much closer

than those now used are necessary to maximize biomass-yield
on short rotations.
Mean annual increment of biomass (MAI) tends to increase
with increasing planting density in young stands until
optimum stocking is reached (Alemdag and Stiell, 1982; Geyer,
1978).

In general there is no appreciable reduction in the

heat content of the wood produced in high density plantations
(Holt and Murphey, 1978; Maeglin, 1967).
however,

for stand densities to become too high resulting in

stand stagnation and a reduction in MAI
1977).

It is possible,

(Howlett and Gamache,

High densities could also lead to a higher incidence

of disease in the plantation, which ultimately reduces MAI
(Schipper, 1976).

57
Economic factors also affect the density at which
plantations will be established.

Seedling and planting costs

rise dramatically with increasing density.

Increased yields

from high-density plantations may not offset the high
establishment costs

(Harms,

1982).

Density/yield relationships need to be determined for
each biomass species to identify the procedures and problems
that will be associated with large-scale plantations.

Forest

managers may then apply their own economic constraints to
define the optimum system for their needs.
Two plantations were established to determine the
effects of spacing on dry-matter yield of young Ailanthus
altissima (Mill.) Swingle (ailanthus) and Platanus
occidentalis L. (sycamore) at experimental forests of
Michigan State University.

Five years after the plantations

were successfully established, yields were measured and
summarized to provide guidelines for future stocking studies
of these species.
MATERIALS AND METHODS
Ailanthus
The ailanthus plantation was established in 1976 on the
Russ Experimental Forest of Michigan State University.

Trees

were planted in three blocks (or replicates) at densities of
6719,

2986 and 1680 trees per hectare.

The young stems were

killed by frost during the winter of 1977-1978 but the stand
re-established itself by sprouting.

This resulted in

increased densities of 8317, 4564 and 3237 trees per hectare

by 1982 (Table 5.1).
Heights and diameters at 1.37m above the ground of all
trees in the p l a nt at i on were measured in the fall of 1982.
Whole-tree, above-ground, oven-dry biomass was computed for
each tree using yield predictor equations specifically
developed for each spacing, using techniques described in
Chapter

3.

For 1.2 m spacing:
Mass

=

0.03944’DBH2*92637

+ 0.14737*Ht

For both wider spacings:
Mass = 0.25567‘D B H 2*01803 - 0.17379‘Ht
W here:
Mass = Oven-dry biomass yield in kg*tree-3-*yr“^
DBH = Diameter at 1.37m above ground
Ht = Total tree height in meters
Areal estimates of biomass yields were established by
summing the individual-tree yields for each plot and
expanding these totals.

Analysis of variance was used to

detect spacing effects on average individual-tree yields and
on areal yields.
Platanus
The sycamore p l a nt at io n was planted in 1978 as part of a
long-term cultural research program of Michigan State
University.

It is located on the Kellogg Experimental Forest

near Battle Creek, Michigan.

The trees were planted at two

different fiber-production densities (1667 and 840 trees
per hectare);

there was no replication of densities.

This

Table 5.1.

Experimental design of ailanthus spacing trial at Russ
Forest, Decatur, Michigan. The change in density
from 1976 to 1982 is due to sprouting after frost
damage in 1978.

Conditions at
planting (1976)
spacing
trees per
(m)
hectare

Plot size
(hectares)

Conditions at
measurement (1982)
spacing
trees per
(m)
hectare

Average number
of Sprouts for
each original
seedling

1.22 x

1.22

6719

0.0156

1.10 x 1.10

8317

1.2

1.83 x

1.83

2986

0.0234

1.48 x 1.48

4564

1.5

2.44 x

2.44

1680

0.0327

1.76 x 1.76

3237

1.9

60
plantation was selected for analysis even though spacings
were much wider than those common to biomass plantations.
The mean survival for the plantation was 84 percent.
Trees were measured in the fall of 1982 and these data
were summarized in the same manner as ailanthus, above.

The

yield predictor equation used for sycamore was developed in
Chapter 3:
Mass
Where:

=

0.05731'DBH2’77994

+ 0.18707’Ht

Mass = OD biomass in kg*tree-1,year_1
DBH = Diameter at 1.37m above the ground
Ht = Total tree height in meters

RESULTS
Ailanthus
The analysis of variance (Table 5.2) showed no
significant effect of spacing on individual stem biomass
production.

Variability within a plot (sampling)

indicating uniformity of yield among stems.

is low,

Yield on an

areal basis was significantly affected by spacing;

the

highest yield was realized at the closest spacing, but there
was no significant difference between the two wider
spacings.

Yields of ailanthus at each of the spacings

tested are summarized in Table 5.3.

The highest areal

yields found for ailanthus averaged 6

OD Mg*ha”^*yr“ ^ and

ranged from 4.1 to 8.1 OD Mg *ha"^*yr”^ over all blocks with
8317 stems per hectare.

61

Table 5.2.

Analysis of variance of spacing effects on
single-tree and areal biomass yields of
ailanthus five years after establishment
in southern Michigan.

Single-tree yields
Source of variance
Spacing
Blocks
Error
Sampling
Total

(yield in oven -dry Kg'tree” ^ •year”^-)
d.f.
2
2
4
1021
1029

Mean square
1.231
78.149
4.668
0.785

F ratio
0.26 NS

Areal yields (yield in oven-dry Mg •ha“^-*year“l)
Spacing
Block
Error
Total

2
2
4
8

10.814
6.483
0.796

13.58 *

* - significant at the 95 percent level of confidence

62

Table 5.3. Mean annual increment yields of ailanthus
planted at three spacings on Russ Forest,
Decatur, Michigan, at age five.

Spacing
(meters)
1.1 x 1.1
1.5 x 1.5
1.8 x 1.8

No. trees
sampled

390
321
309

Whole-tree, oven-dry weight-^/
Single-tree
(Kg‘tree- 1 •year”1)
0.72 a
0.63 a
0.78 a

Areal
(Mg •ha” 1 'year"1 )
5.97 b
2. 86 c
2.54 c

1/ Treatment means not followed by the same letter are
significantly different at the 0.05 level of probability
according to the LSD criterion.

Table 5.4.

Spacing
(meters)

2.4 x 2.4
2.4 x 4.9

Mean annual increment yields of sycamore
planted at two spacings at Kellogg Forest,
Battle Creek, Michigan, at age five.
plot size
(ha)

0.0857
0.0785

No. trees
sampled

101
66

Whole-tree, oven-dry weight
Single-tree
Areal
(Kg*tree"1 *yr“1 ) (Mg*ha“1 *yr“1 )
0.68
0.65

0.82
0.55

63

Platanus
The lack of replication of spacings makes it impossible
to separate environmental

from spacing effects.

Since

previous investigations have shown that environmental effects
can be as large, or larger than those of spacing, doubts
about the finality of the conclusions drawn from these data
are raised.

Nevertheless,

it is of interest to examine the

trends in these data to identify directions for future tests
(Table 5.4).
There appear to be no differences in single-tree yields
between the two spacings.
0.66 OD kg*tree“^,year"^.

Each tree yields, on average,
The areal yield of trees planted

at 1667 trees per hectare is 0.82 OD M g ’ha'^-’year-1.

Trees

planted at 840 trees per hectare yielded 0.55 OD Mg*ha“
I'year-*-, which is exactly two thirds of the yield found at
the higher density.
DISCUSSION
Neither ailanthus nor sycamore showed reduced single­
tree yields at the higher densities tested in these
plantations, which indicates that the optimum planting
densities for these species, at this age, on these sites has
not yet been reached.

Single-tree yield would be expected to

decrease with increasing density when maximum MAI is being
produced (Cannell,

1980).

Future tests should concentrate on

examining spacings closer than 2.44 x 2.44m (1667 trees per
hectare)

for sypsmore and 1.10 x 1.10m (8317 trees per

64
hectare)

for ailanthus.

Since there were equivalent single-tree yields over all
spacings for both species,

increased areal yields with

increasing densities was expected.

The yield of sycamore,

for

example, increased in direct proportion to the number of
stems.

Yields of the two species at various spacings are

compared in Table 5.5.

Projections have been made beyond

the limits of the test plantations by making several
assumptions:
each spacing.

1.

The single-tree yield remained constant at

This yield was equal to that of the average

tree in the test plantations.

2.

S u r v i v a l at age five in

all plantings would be eighty percent of the number of trees
planted.

3.

The yields of the test plantations are

representative of other similar plantings.

The purpose of

these projections was to demonstrate what the maximum
theoretical yield of these species might be in closelyspaced, short-rotation, biomass plantations.

They are only

projections and should be used with caution.
Sycamore and ailanthus planted at spacings tested by
these investigations do not produce biomass at acceptable
rates.

Individual-tree yields are encouraging though, and

further tests should be conducted at closer spacings and on
other sites to determine if the estimates presented in Table
5.5 are reasonable.

These yields seem to be in keeping with

a theoretical maximum rate of 10 to 15 OD Mg*ha” ^-yr” ^
reported by Cannel and Smith (1980).
It is interesting to note that the rate of resprouting

65

Table 5.5.

Estimated yields of sycamore and ailanthus at
various spacings assuming; 1. constant single
tree yields and 2. 80% survival at age five.
Estimated yield______

Species

Single-tree yield-1/ 1344
(Kg*tree"^*year“^)

Ailanthus
Sycamore
—/

0.72
0.68

stems per hectare ----5375
22222

---1.0
0.9

(Mg*ha“^*year"^) ----3.9
3.7

16.0
15.1

Single-tree yields were derived from field measurements.

66
of ailanthus following frost-killing, seems to be related to
planting density (Table 5.1).

The number of sprouts per

seedling decreases with increasing density.

The result is to

make the stocking more uniform among the treatments once
resprouting has occurred.

This is an example of the

mechanism by which natural stands attempt to fully occupy a
site; that is, a stand which is under-stocked sprouts back
more vigorously than one which is more nearly fully-stocked.

67
LITERATURE CITED
Alemdag, I.S. and W.M. Stiell.
1982.
Spacing and age effects
on biomass production in red pine plantations.
For.
Chron. 58 (5) :220-224.
Cannell, M.G.R. 1980.
Productivity of closely-spaced young
poplar on agricultural soils in Britain.
Forestry
53(1):1-21.
_______________ and R.I. Smith. 1980.
Yields of minirotation
closely spaced hardwoods in temperate regions:
review
and appraisal.
For Sci. 26(3) :415-428.
Ek, A.R. and D.H. Dawson. 1976.
Yields of intensively grown
Populus: Actual and projected.
IN:
Intensive
Plantation Culture:
five years research, p 5-9.
USDA
For. Serv. Gen. Tech. Rep. NC-21.
Geyer, W.A. 1978.
Spacing and cutting cycle influence on
short-rotation silver maple yield.
Tree Planter's Notes
29(1):5-7,26.
Harms, W.R. 1982.
An empirical function for predicting
survival over a wide range of densities.
IN:
Proc.
Second Biennial Southern Silvicultural Research Conf.
USDA For. Serv. Gen. Tech Rep. SE-24. p334-337.
Holt,

D.M. and W.K. Murphey. 1978.
Properties of hybrid
poplar juvenile wood affected by silvicultural
treatments.
Wood Sci.
10 (4):198-203.

Howlett, K. and A. Gamache.
1977.
Silvicultural Biomass
Farms.
Vol. II. The biomass potential of shortrotation farms.
MTR-7347(vol 2). The Mitre Corp.
McLean, VA.
Hozumi, K., T. Asahira, and T. Kira. 1956.
Interspecific
competition among higher plants.
VI.
Effects of some
growth factors on the process of competition.
J. Inst.
Polytech., Osaka City University, Ser. D7:15-34.
Kennedy, H.E. 1975.
Influence of cutting cycle and spacing
on coppice sycamore yield.
USDA For. Serv. Res. Note
SO-193, 3pp.
Lloyd, F.T. and E.P. Jones Jr.
1982. Density effects on
height growth and its implications for site index
prediction and growth projections.
IN: Second Biennial
Southern Silvicultural Research Conference. USDA For.
Serv. Gen. Tech. Rep. SE-24. p329-333.
Maeglin, R.R. 1967.
Effect of tree spacing on weight yields
for red pine and jack pine. Jour. For. 65 (9) :647-650.

68
Schipper, A.L. 1976.
Poplar planting density influences
foliage disease.
IN:
Intensive Plantation Culture.
USDA For. Serv. Gen. Tech. Rep. NC-21. p81-84.

CHAPTER 6
Provenance and Clonal Performance of Eight
Short-rotation Biomass species in Southern Michigan

ABSTRACT
Seven provenance test plantations and one clonal test
plantation in Michigan were evaluated to develop preliminary
seed source and clone recommendations for eight, promising
biomass species: Acer rubrum (red maple), Betula
alleghaniensis

(yellow birch), B. papyrifera

Fraxinus americana

(white ash),

(white birch),

F. pennsy1vanica

(green ash),

Jug Ians nigra (black walnut), Larix laricina (tamarack), and
Populus hybrids.

Test plantations ranged in age from five to

sixteen years, which correspond to the rotation lengths of
short-rotation, biomass plantations.

Provenance or clone

performance was analyzed based on single-tree, oven-dry,
above-ground biomass yield, expressed as a mean annual
increment.
Gains in biomass yield over "local" sources can be as
high as 37 percent with proper provenance selection.

Yield

losses due to improper provenance selection can be as much as
48 percent of "local" source yield.
INTRODUCTION
Efforts to reduce the cost of energy and to increase the
"energy-independence" of the United States began in earnest
in the early 1970*s.

Many alternatives to fossil fuels have

been, and are being investigated.
69

In 1978, the U.S.

70
Department of Energy and Michigan State University began a
long-term project to define an effective system for producing
substantial quantities of woody-biomass for fuel on forest
sites and abandoned agricultural fields in Michigan.
A stepwise approach similar to that proposed by Inman
(1977) was adopted for developing a comprehensive woodybiomass production system for Michigan.
of four phases:

The program consists

1. Identification of the most promising

biomass species through trial plantings on abandoned
agricultural fields and cleared forest stands, 2.

Preliminary

yield comparisons of several species growing in existing
experimental plantations, 3. Species improvement using
standard tree improvement techniques, and 4. Development of
cultural techniques designed to optimize woody-biomass yield
from energy plantations.

Investigations have been conducted

in each phase of the program since 1978.

This paper will

summarize the results of the preliminary genetic screening
v/hich has been conducted on eight of the biomass species
under

investigation.
Forest tree improvement and biomass-tree improvement

have three common goals:

1. To maximize yield by selecting

genotypes with superior growth rates and wood qualities,

2.

To increase the site-adaptability of a species by selecting
less site-specific genotypes, and

3. To reduce the adverse

effect of insect and disease pests on stands by selecting
resistant genotypes.

The gain which can be achieved in each

of these three areas depends on the species itself, and the

71
intensity and type of tree improvement technique being used
(e.g. provenance selection, family selection, or individualtree selection).

Examples of the gains in height which

might be achieved for white spruce and jack pine under
different selection strategies have been provided by
Nienstaedt and Jeffers

(1976)

(Table 6.1).

Variability

within white spruce provenances and families is greater than
for jack pine.

As a result, gains from family and within

family selection of white spruce are higher than for jack
pine. If the species being investigated in this study
respond as white spruce does, the gains from provenance
selection, which are reported here, represent the minimum
gains which may be achieved by breeding.
The selection of genotypes for biomass production can be
based on a range of characteristics which include:
individual-tree growth rate, tolerance of competition in
densely planted stands, wood quality (e.g. specific gravity,
extractive content, or heat content),

insect and disease

resistance, and cultural treatment or site adaptability.
Each of these characteristics is important in determining the
quantity and quality of the biomass produced, but the
consensus among biomass researchers is to concentrate on
maximizing growth rates.

Genotypes with other desirable

characteristics may be selected from among this faster
growing group in a later phase (Ezell, et al. 1983; Frampton
and Rockwood,

1983; Giordano, 1969).

With this in mind,

selections in this study were based on single-tree, total

72

Table 6.1.

Estimated gains which may be achieved through
provenance, family, and individual-tree
selection of white spruce and jack pine in the
Lake States.
(Nienstaedt and Jeffers; 1976)

Type of selection

Comparison of superior seedlots
to control seedlots for:
jack pine
white spruce

Provenance^./

20 - 30 %

10 - 20 %

Half-sib Progeny

14 - 25 %

2 - 10 %

Tested Clonal

10 - 100 %

10 - 50 %

These figures represent gains in height from collection
within the better provenances.

73
above-ground, oven-dry biomass yield.

Subsequent testing in

higher density plantations is planned to identify genotypes
within the better yielding provenances for inclusion in
biomass-species seed orchards.
The forest tree improvement program at Michigan State
University began in the mid-1950's.

The approach taken for

most species has involved provenance testing to identify
racial trends, followed by progeny testing within the better
provenances to identify the better families, and subsequent
sexual or asexual propagation of proven superior individuals
when practical.

More than 45 species are currently being

tested in over 400 plantations throughout Michigan.

Eight of

the younger provenance and clonal test plantations were
selected for analysis in this study, based on the
desirability of the represented species for the production of
biomass on a short rotation basis.

MATERIALS AND METHODS
Species represented in test plantations are listed in
Table 6.2.

The plantations were located on either one of two

Experimental Forests owned by Michigan State University in
southern Michigan.

Each test was established with bare-root

seedlings, planted by machine between 1967 and 1978, and
maintained under similar low-intensity management regimes.
Trees were arranged in one to five-tree row p lots in a
randomized block design.

A summary of the characteristics

and experimental design of each plantation is presented in

74

Table 6.2.

Species for which provenance recommendations
were made.

Common name
Red maple
Yellow birch
White birch
White ash
Green ash
Black walnut
Tamarack
Hybrid poplar

Latin name
Acer rubrum L.
Betula alleghaniensis Britton
Betula papyrifera Marsh.
Fraxinus americana L.
Fraxinus pennsylvanica Marsh.
Juglans nigra L.
Larix laricina (Du Roi)
Populus

75
T a b le 6 . 3 .
All trees were measured for total height and diameter in
the fall of 1982.

Total, above-ground, single-tree biomass

was computed using predictor equations developed for each
plantation in Chapter 3 (Table 3.4).

Plot means for biomass

were computed and an analysis of variance was performed to
detect differences among seedlots and to estimate the
variance components needed to compute heritability.

The

plot means were expressed as a percent of the block mean to
minimize block effects on performance.

Yield for each

seedlot within a provenance was then averaged over all
blocks to provide an overall estimate of performance at each
plantation.

The average of all Michigan seedlots was

computed and used as a baseline (the "local" source).

The

gain or loss in biomass yield from provenance selection was
computed by applying the heritability to the difference in
performance of the best and worst provenances to the "local"
source performance.
Heritabilility and gain were not computed for hybrid
poplar because an adequate base-line was lacking in this
plantation, i.e. there is no valid "local" source of hybrid
poplar with which to compare other clones.

Therefore, yields

reported in this study represent the mean performance of the
best and worst clones over all blocks in the plantation.
RESULTS AND DISCUSSION
Significant differences in yield were found among
provenances for all species.

High variability indicates that

Table 6.3*

P la n ta tio n s used fo r genotype e v a lu a tio n of biomass production of eig h t promising
biom ass-species in southern Michigan.

Species

Nearest
town

Age
(Yrs from
planting)

Red maple

Augusta, HI

9

Yellow birch

Augusta, MI

15

White birch

Decatur, MI

7

White ash

Augusta, MI

Green ash

Type of test

No.
seedlots
(provenances)

No. of trees
per plot
(No. blocks)

1/2-sib progenyprovenance test

94 (23)

2 (6)

Provenance test

35 (20)

4 (5)

1/2-sib progenyprovenance test

224 (26)

1 (4)

6

1/2-sib progenyprovenance test

35 (23)

5 (5)

Decatur, MI

5

1/2-sib progenyprovenance test

43 (20)

4 (5)

Black walnut

Augusta, MI

16

Provenance test

19 (13)

4 (6)

Tamarack

Augusta, MI

14

Provenance test

33 (12)

4 (3)

Hybrid poplar

Decatur, MI

8

Clonal trial

40

4 (8)

—

77
ample opportunities exist for biomass breeding in all species
considered.

The gains and yields which can be expected from

plantations established with sources from the recommended
provenances are presented in Table 6.4 and a summary of
recommended provenances appears in Table 6.5.

Losses in

yield which would be expected to result from improper
provenance selection are also shown.

In all cases, the

advantages of selecting the better provenances and the
disadvantages of selecting the worst provenances are
substantial when compared to the "local" source average.
It is important to note that, for each species,

these

data were collected in a single plantation at a single age.
Heritability and gain are overestimated because of unknown
genotype x environment interaction variance.

All conclusions

are p re li mi na ry in nature and further testing on a range of
sites will be necessary to provide confirmation.
Red maple
Red maple families collected from the south-central
portion of Michigan yielded 15 percent better than the
"local" source in the test plantation.
yield,

The reduction in

if the worst provenance was selected,

be 33 percent.

is estimated to

The differences among provenances is highly

significant and heritabi1 ities are correspondingly high
(0.69),

indicating substantial opportunities exist for

obtaining gains in yield through breeding.

Townsend and

Harvey (1983) examined height and diameter growth of this and

78

Table 6.4.

Biomass yields of selected provenances and clones as compared
to "local" sources in genetic test plantations of eight
species in Michigan.

Effect of selection of the best or poorest
provenances or clones
Species
1/

h2

Best

Poorest
Gain

------------

%

Best
Poorest
Yield
— Kg*tree -1-yr-1

0.66

+37

-26

2.22

1.20

Yellow birch

0.29

+17

-15

3-57

2.59

Tamarack

0.68

+30

-48

3.00

1.20

Red maple

0.69

+ 15

-53

0.77

0.45

Hybrid poplar 2/ 0.78

—

-----

1.80

0.40

White birch

0.54

+30

-40

0.91

0.42

White ash

0.25

+ 8

-14

0.16

0.13

Green ash

0.61

CM

-25

0.44

0.26

+

Black walnut

\J - Species are listed from oldest to youngest.
2/ - The "best" selection of hybrid poplar here is based on the clone
which produced the highest yield among the canker- resistant
clones. Higher yields were observed for this species, but these
clones were heavily cankered.

Table 6.5.

A summary of seed source recommendations for
the better biomass species, for use in southern
Michigan.

Species

Recommended collection area

Yellow birch

Mackinac Co., MI.

White birch

Manistee, Mason, and Lake Counties, MI.

Black walnut

West Virginia

Tamarack

Alger Co., Michigan

Hybrid poplar

Northeast Clone NE-207
(Populus deltoides x P. trichocarpa)

80
four other plantations of this same material throughout the
Lake States, and arrived at similar conclusions about the
opportunities for breeding red maple.

They also noted a

significant genotype x environment interaction among
plantations which may indicate that regional or site-specific
source recommendations will be needed for this species,
yields of improved red maple are not outstanding (0.8
kg’tree_ ^*year"^).

It is not,

therefore,

recommended for use

in biomass plantations under these conditions.
Yellow birch
Differences among yellow birch provenances were
moderately significant and heritability was low (0.29).
Sources collected from the southeastern portion of Michigan’s
Upper Peninsula were superior and represent a 17 percent gain
over the "local" source tested.

These gains from provenance

selection are substantial, but Clausen (1973) has stated that
the variation within provenances is much greater than among
provenances for yellow birch.

This suggests greater gains

can be achieved through family and individual-tree testing.
Although yellow birch produced high annual yields in this
study

(3.57 kg*tree“^*year~^), establishment problems such as

animal damage and intolerance to weed competition may limit
the utility of this species in managed plantations.
White birch
A 30 percent gain in yield can be expected from white
birch collected in the northwestern portion of the Lower
Peninsula of Michigan when compared to the state-wide

81
average.

Differences among sources were highly significant

in the test plantation and heritability was fairly high
(0.54).

Substantial

reductions in yield

from improper provenance selection.
production rate demonstrated here

(40%) would result

If the individual-tree
(0„91 kg*tree“ ^*year"^)

can be sustained at higher densities than those in the test
plantation, white birch could be considered as a good
biomass species.
White ash
Seed source differences in the test plantation were only
moderately significant and heritability was low (0.25).
Moderate gains (8%) could be demonstrated from selecting
provenances in the Central Plains states.
the better

Yields achieved by

sources were so low (0.16 kg*tree“^*year"^) that

white ash should be discounted as a biomass species in
Michigan.
Rung and Clausen (1983) reported on four plantations
which contained materials similar to those of the plantation
measured in this study.

Growth in Lake States plantations

was less than that at other test locations.

They also noted

a strong genotype x environment interaction among the
plantations.

The poor yields of white ash experienced in

Michigan may have occurred because no provenances suited to
the conditions of the test site were included.

It is more

likely that the juvenile growth rate of white ash is slow and
therefore not suited to short-rotation plantings.

82
Green ash
Green ashr as with white ash, performed poorly in terms
of dry-matter production in these tests; superior sources
from Illinois yielded only 0.44 kg*tree"^*year"^.

The

reasons for this poor performance may be similar to those
discussed above for white ash.

However, there is more

variability among families for green ash than for white ash,
so the opportunities for genetic improvement may be greater.
Black walnut
Black walnut sources from West Virginia performed best
in this test, yielding 2.22 kg*tree”1,year” 1.

The gain from

using these sources was estimated to be 37 percent over the
"local" source.

Although black walnut is most commonly

thought of as a high-value timber and veneer species, the
yields demonstrated here show that it can also produce
substantial quantities of biomass on short-rotations.
Clausen (1983) examined this and six other similar
plantations of black walnut located throughout the Lake
States to determine how source rankings change over time.

He

found that the sources which grew best at age ten also were
the best at age 15.

This means that provenance selections

based on 16-year-old trees will hold for younger trees,
allowing some flexibility of rotation length.
Tamarack
Differences among tamarack provenances were highly
significant and heritability was very high (0.68).

Gains of

30 percent over the "local" source were demonstrated for the

83
Alger County, Michigan provenance.
provenance was 3.0 kg*tree”^*year

Yield from this
, which makes tamarack

appear very desirable as a biomass species.

Riemenschneider

and Jeffers (1981) found that the correlation between
14- and 9-year-old tamarack heights and diameters were
strong, which indicates that provenances which did well in
this test would also be suitable on slightly shorter
rotations.
Very little genetic work has been done with tamarack,
but the few tests which do exist have shown strong genotype x
environment interaction among plantations (Jeffers, 1975;
Riemenschneider and Jeffers,

1981).

This means that source

recommendations may need to be specifically tailored to the
site or region of planting.

Future work should attempt to

find less site-specific genotypes, which would make tamarack
easier to establish successfully.
Hybrid poplar
Hybrid poplar has received a great deal of attention as
a short-rotation biomass species.

New clones have been

identified which are better adapted to the conditions of the
Lake States, and they have been used to augment or replace
the Northeast clones planted in this test.

These newer

clones tend to be more resistant to the insect and disease
pests which plague most clones of hybrid poplar in Michigan
(Woods and Hanover,

1982).

This plantation serves as an

example of the range of yields and pest resistance of hybrid

84
poplar clones, rather than a source of data for making final
clonal recommendations.
Growth rates varied significantly among clones in the
test plantation and ranged from 0.4 to 3.5 kg*tree"^,year"^
at age eight.

The best yielding clone was NE-17 (Populus

nigra x P. deltoides); this was also the best clone at age
four when Brissette, et al.
plantation.

(1979) analyzed this same

Some clones which had done well at age four

were no longer superior at age nine, which is consistent
with the experiences of other investigators (Wilkinson,
1973).

This indicates that clones which yield well on very

short rotations (1-4 years) may not do well on longer
rotations (10-15 years), and therefore selections must be
tailored to rotation length in some cases.

Hybrid poplar

clones have also been shown to be very site-specific, so
testing of clones must be conducted on many sites to
identify and characterize these interactions (Dickmann and
Stuart,

1983).

Many clones of hybrid poplar have been identified as
having rapid juvenile growth, but this criteria for selection
in not sufficient in Michigan.

Severe problems have been

encountered with cankering diseases such as Septoria musiva
and subsequent infection by secondary pathogens such as
Fusarium soloni.

The test plantation is severely cankered.

Some clones had been killed completely, while others show
resistance (although no clone had escaped cankering
completely).

85
Canker resistance must be the prime consideration for
selection of hybrid poplar clones in Michigan.

The fastest

growing clone from among the more resistant clones was NE207 (Populus deltoides x P. trichocarpa), which yielded 1.8
kg*tree"^*year"^.

The lower yield must be accepted if the

h ea lt h of the trees and subsequent sprouts is to be
preserved.
CONCLUSION
The results summarized here represent the first efforts
to improve eight species for biomass production in Michigan.
Provenance and clonal recommendations have been made with the
understanding that the heritabilities and gains are probably
overestimated due to a lack of testing on multiple sites.
Further research efforts must concentrate on improving
yields, site adaptability, wood quality, and

pest resistance

of the more promising species, through progeny and clonal
testing,

and systematic breeding efforts.

Genetic

plantations are already being established at Michigan state
University for promising biomass species which have not
previously been tested in breeding programs.

Data from

these and other tests, as well as information being
generated in cultural experiments, will help to refine the
optimum short-rotation, woody-biomass production system for
Michigan in the years to come.

86
LITERATURE CITED
Brissette, J.C., J.W. Hanover, T.J. Stadt, and J.W. Hart.
1979.
improved poplars for Michigan through the
Michigan State Cooperative Tree Improvement Program.
IN: Proc. North American Poplar Council.
Manistee, MI.
1979. ppll5-122.
Clausen, K.E. 1973.
Genetics of yellow birch.
Serv. Res. Pap. WO-18, 28p.

USDA For.

Clausen, K.E. 1983.
Performance of black walnut provenances
after 15 years in 7 midwestern plantations.
IN:
Proc.
Third North Central Tree Improvement Conf. Wooster, OH.
August, 1983.
Dickmann, D.l. and K.W. Stuart.
1983. The C u lt ur e of
Poplars in Eastern North America. Michigan State
University, East Lansing, MI.
168pp.
Ezell, A.W., W.J. Lowe, W.K. Murphy, and J.A. Wright.
1983.
Plantation culture for energy production— the importance
of within-species selection.
Wood and Fibre Sci.
15(1):69-73.
Frampton, L.J. and D.L. Rockwood. 1983.
Genetic variation
traits important for energy utilization of sand and
slash pines.
Silv. Gen. 32(1-2):18-23.

in

Giordano, E. 1969.
interaction between breeding and
intensive culture.
Second World Consultation on Forest
Tree Breeding.
August, 1969, Washington, D.C.
Food and
Agriculture Organization of the United Nations.
FO-FTB69-12/1.
17p.
Inman, R.E. 1977.
Silvicultural Biomass Farms:
Summary.
Mitre Tech.
Rep. No. 7347, Vol 1.
Corp.
McLean, VA.

Vol.
I.
The Mitre

Jeffers, R.M. 1975.
Survival and height growth of tamarack
planted in northern Wisconsin.
USDA For. Serv. Res.
Note NC-190.
3p.
Rung,

F.H. and K.E. Clausen. 1983.
Provenance x environment
interactions in white ash.
IN:
Proc. Third North
Central Tree Improvement Conf. Wooster, OH. August,
1983.

Nienstaedt, H. and R.M. Jeffers. 1976.
Increased yields of
intensively managed plantations of improved jack pine
and white spruce.
IN:
Intensive Plantation Culture:
five years research. USDA For. Serv. Gen. Tech. Rep.
NC-21.
pp51-59.

87
Riemenschneider, D.E. and R.M. Jeffers. 1981.
Height and
diameter of tamarack seed sources in northern Wisconsin.
USDA For. Serv. Res. Paper NC-190. 6p.
Townsend, A.M. and W.R. Harvey.
1983.
Genetic analysis for
height and diameter growth of 9-year-old red maple
progenies in five plantations.
IN: Proc. Third North
Central Tree Improvement Conf. Wooster, OH. August,
1983.
Wilkinson, R.C. 1973.
Realized and estimated efficiency of
early selection in hybrid poplar clonal tests.
IN:
Proc. 21st Northeastern Forest Tree Improvement
Conference.
Fredericton, New Brunswick.
Canada.
August, 1973.
lllp.
Woods, R.F. and J.W. Hanover. 1982.
Growth of Imperial
Carolina poplar over a range of soil types in lower
Michigan.
Tree Planter's Notes 33(2):8-13.
Wright, J.W. 1976.
Academic Press

Introduction to Forest Genetics
Inc.
NY.
463p.

CHAPTER 7
Summary and Recommendations for Future Research

Research conducted since 1978 has provided the basic
information needed to begin production of woody-biomass in
short-rotation energy plantations in the Michigan, and has
identified areas where future research should be
concentrated.

Species screening has been conducted on

abandoned agricultural fields, on clearcut forest sites, and
several older single-species test plantings.

The data from

these tests suggests that the following species are suitable
for use in short-rotation energy plantations in Michigan:
Scotch pine, red pine, jack pine, Kellogg hybrid pine,
larches, hybrid poplar, hybrid aspens, ailanthus, and black
locust.

The conifers are better suited to the colder and

drier sites in the region and the hardwoods to the warmer and
more fertile sites.

Species-site matching can only be made

in a general way at present, due to the limited number of
locations tested.
Tree improvement programs and improved cultural
techniques promise to increase the site adaptability and
growth rates of all biomass species.

Many of the species

listed above have been investigated, and a great deal is
known regarding their genetics and their response to cultural
treatments.

These species are:

pine, and hybrid poplar.

Scotch pine, red pine, jack

Much less information is available

for the remaining five species:

88

Kellogg hybrid pine,

89
larches, hybrid aspens, ailanthus, and black locust.

Future

research should concentrate on this second set of species to
identify the best genetic materials and cultural systems for
energy production.

Several research projects have begun or

are scheduled to begin in the next two years at Michigan
State University which are designed to provide this
information for these "primary species."
Current projections of plantation yields have been based
on small

(<0.2 hectare) plots, which lends uncertainty to the

figures.

Some investigators have applied cultural treatments

which resemble those used in modern American agriculture to
high density (about 200,000 trees per hectare) energy
plantations and have obtained yields as high as 15-20 Mg*ha"
^•yr-3- in one to two years.

It is questionable,

however,

that the fertilizer and irrigation inputs and the extremely
high costs of establishment required to obtain these yields
can be justified.
Research conducted at Michigan State University was
based on the assumption that more conventional plantation
management systems would be better suited to economical
biomass production than the intensive cultural systems
described above.

Certain species, such as Kellogg hybrid

pine, have produced as much as 12 Mg*ha“ ^,yr‘“^ by age 16
years in test plantations using less intensive cultural
systems.

It is therefore recommended that biomass

plantations in the Lake States be established on moderately
close spacings (3000 - 5000 trees per hectare) and provided

90
with good weed control for the first two to four years, but
with no fertilization or irrigation.
Comprehensive genetic testing and breeding programs have
begun for each of the five primary species.

This area of

research has been shown to have great potential for improving
biomass yields.

Yield improvements are also being sought

through improvements to the cultural systems outlined above.
Large-scale spacing studies are planned for each of the five
primary species.

These investigations are designed to build

on the data reported in Chapter 5, and define the optimum
planting densities for each species on a range of sites.
Built into these investigations is a program to further
refine the biomass predictor equations developed in Chapter
3.

This will improve the ability of plantation managers to

inventory standing biomass and also provide accurate yield
figures needed for meaningful economic analysis.
Past research has concentrated on producing large
quantities of woody-biomass to be used as fuel in direct
combustion or gasification processes.

Another area of

current research at Michigan State University is designed to
look for alternative uses for the biomass which is produced
in short-rotation plantations.

These include:

Fiber for

pulp, feedstocks for chemical production processes, animal
feed, and nitrogen fertilizer substitutes.

All or part of

the trees grown may be used to produce these products.

This

will help to improve the economics for short-rotation
plantation systems and to reduce the national dependence on

91
fossil fuels for some of these commodities.
Short-rotation biomass plantations are currently only a
theory in the Lake States; no commercial production systems
exist.

This can be attributed to two major factors.

First,

the technology for growing and harvesting this type of
material is largely experimental, especially for the species
which are being recommended.

Second, there is currently an

excess supply of biomass available in natural stands and
older plantations in the region.
not be true for much longer.

The second condition will

Pressure on the available

timber resource is already being felt in certain parts of
Michigan as paper, pulp, and composition-board manufacturers
compete with home and small

industrial fuelwood markets.

Wood utilization is increasing rapidly and new supplies will
be needed soon.

It is important that plans be made soon to

meet rising demands, due to the long interval between
plantation establishment and final harvest (8-15 years).
The first problem can be easily solved through close
cooperation between industry and university researchers.
Plans are currently being made to involve several industries
in Michigan in commercial size plantings of the five primary
biomass species listed above.

These plantings will serve as

proving grounds for the theories developed in the laboratory
and in experimental plots, and provide a demonstration of the
feasibility of short-rotation biomans production.
Progress toward defining optimal woody-biomass
production systems for the Lake States has been substantial.

92
Species recommendations have been made, cultural systems have
been defined, and genetic improvement of many species has
begun.

Future research should concentrate in the following

areas:
1.

Refining species-site adaptability.

2.

Intensifying genetic screening and breeding
programs in the primary species.

3.

Refining cultural systems to identify the best
planting densities, weed control methods, and
harvesting

4.

techniques.

Improving the ability to predict yields on a range
of sites.

5.

Searching for additional uses for woody-biomass
produced under short-rotation culture.

6.

Establishing short-rotation systems as a viable
commercial alternative to current production
techniques.