I:

I an
t. it.
.1.

 

 

. -
1. ,F. .wWqui
in. .. x n ”a

mum». am.

It:
e... .13..
A .| £51.:
r g c y.
;I

#NﬁJQ‘!

. . a

. u. am»:
.

a ..
9 .UN‘. J. a.-

....

 

THESIS

2000!

 

 

This is to certify that the
thesis entitled

BIOENERGY CROPPING SYSTEMS ON MARGINAL
LAND

presented by

STEPHANIE LYN SMITH

has been accepted towards fulﬁllment
of the requirements for the

MS.

 

 

P

.3 “\l N
<04 ’ ALA

 

Major Professor’s Signature
, /
r» /8/ a 7
I r

Date

MS U is an Afﬁrmative Action/Equal Opportunity Employer

l
PLACE IN RETURN BOX to remove this checkout from your record.

TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KrlProleccsPres/CIRC/DateDue indd

 

BIOENERGY CROPPING SYSTEMS ON MARGINAL LAND
By

Stephanie Lyn Smith

A THESIS

Submitted to
Michigan State University
in partial futﬁllment of the requirements
for the degree of

MASTERS OF SCIENCE

Crop & Soil Sciences

2009

ABSTRACT
BIOENERGY CROPPING SYSTEMS ON MARGINAL LAND
By
Stephanie Lyn Smith

Field trials were conducted in 2007 and 2008 to evaluate the efﬁcacy of
identiﬁed potential bioenergy crops production on marginal land sites.
Treatments of crop including com (Zea mays ssp. mays), soybean (Glycine max
(L.) Merr.), canola (Brassica napus L. var. napus), sunﬂower (Helianthus
annuus), and switchgrass (Panicum virgatum L.) were evaluated to determine the
highest yielding crop which would produce the most fuel speciﬁc to each selected
location. Sites selected for the study included a remediated brownﬁeld site
(ROSE), two northern marginal sites (LC and UP), and a traditionally cropped
site in Ingham County (MSU). Yield of soybean, sunﬂOwer, canola, and
switchgrass differed between locations; however the traditionally cropped site did
not consistently provide higher yields. Measured soil characteristics of fertility did
not predict yield returns. Ethanol produced from corn grain was signiﬁcant
(p<0.05) among locations, and as latitude increased, ethanol yield decreased.
Total oil of soybean responded negatively to increases in latitude however, total
oil in canola beneﬁted from increased latitudes. Fatty acid analysis provided
results indicating northern locations yielded higher levels of oleic acid in soybean
and canola. Switchgrass stands, equal in age, regardless of location, were not
signiﬁcantly different. Considering all factors, additional acreage including
brownﬁelds and marginal land sites should be explored to increase crop acreage

and biomass available for conversion to bio-based fuel sources.

ACKNOWLEDGMENTS

Many people have aided and helped me with this project. I would
like to sincerely thank the members of my graduate committee, Dr. Kurt
Thelen, Dr. Bruce Dale, Dr. Russell Freed, and Dr. Philip Robertson for
their guidance and support throughout this project. I would like to extend
gratuitous thanks to everyone at the Agronomy Farm and in particular Bill
Widdicombe and Keith Dysinger and volumes of thanks for my lab
members and in particular Sarah MacDonald and Juan Gao. I would also
like to thank Christian Kapp for all of his labor and data collection for
. chapter 3.

Very special thanks are in order for my parents and brother who
have instilled in me the love of agriculture and the determination to

achieve my goals.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................ vi
LIST OF FIGURES .......................................................................................... viii

CHAPTER 1: Review of Literature

Introduction ..................................................................................................... 1
Rationale .......................................................................................................... 2
Land Classiﬁcations ........................................................................... 5
Energy ........................................................................................... 12
Crop Conversion to BioEnergy ............................................................ 14
Potential BioEnergy Commodities ........................................................ 16
Corn .......................................................................................... 17
Soybean ..................................................................................... 21
Canola ....................................................................................... 25
Sunﬂower ................................................................................... 28
Switchgrass ................................................................................. 31
Concerns of BioEnergy Crops ............................................................. 36
LITERATURE CITED ........................................................................ 38

CHAPTER 2: Use of Regulatory Brownﬁeld Sites for BioEnergy Production

ABSTRACT .................................................................................... 52
INTRODUCTION .............................................................................. 53
MATERIALS AND METHODS ............................................................ 55
Site Description ............................................................................... 55
Contaminant Analysis ....................................................................... 56
Cultural and Management Practices ..................................................... 58
Harvest and Handling ........................................................................ 60
Corn Grain Ethanol ........................................................................... 61
Fatty Acid Methyl Esters ..................................................................... 62
Total Oil .......................................................................................... 64
Total Crystalline Sugar ...................................................................... 65
Statistical Analysis ............................................................................ 66
Economic Analysis ........................................................................... 66
RESULTS AND DISCUSSION ............................................................ 67
Contaminant Analysis ....................................................................... 67
Environmental Conditions .................................................................. 68
Agronomic Yield ............................................................................... 70
Corn Grain Ethanol Yield ................................................................... 73
Fatty Acid Methyl Esters Yield ............................................................ 75
Total Oil Yield .................................................................................. 77
Total Crystalline Sugar Yield ........................................................................... 78

TABLE OF CONTENTS CONTINUED:

Economic Analysis ........................................................................... 80
CONCLUSIONS .............................................................................. 82
APPENDICES ................................................................................. 84
LITERATURE CITED ........................................................................ 86

CHAPTER 3: Use of Agronomically Marginal Land for BioEnergy
Production

ABSTRACT .................................................................................... 91
INTRODUCTION .............................................................................. 92
MATERIALS AND METHODS ............................................................ 94
Site Description ................................................................................ 94
Cultural and Management Practices ..................................................... 95
Harvest and Handling ........................................................................ 98
Corn Grain Ethanol .......................................................................... 100
Fatty Acid Methyl Esters .................................................................... 101
Total Oil ......................................................................................... 102
Total Crystalline Sugar ..................................................................... 103
Statistical Analysis ........................................................................... 104
Economic Analysis ........................................................................... 105
RESULTS AND DISCUSSION ........................................................... 105
Environmental Conditions 105
Agronomic Yield .............................................................................. 108
Corn Grain Ethanol Yield ................................................................... 112
Fatty Acid Methyl Esters Yield ........................................................... 114
Total Oil Yield ................................................................................. 1 17
Total Crystalline Sugar Yield .......................................................................... 120
Total Fuel Yield ............................................................................... 122
Economic Analysis ........................................................................... 123
CONCLUSIONS .............................................................................. 124
APPENDICES ................................................................................. 126
LITERATURE CITED ........................................................................ 129

CHAPTER 4: Green Peach Aphid (Myzus persicae Sulzer) Response to
Increased BioEnergy Crop Production

ABSTRACT .................................................................................... 134
INTRODUCTION ............................................................................. 135
MATERIALS AND METHODS ........................................................... 142
RESULTS AND DISCUSSION ........................................................... 144
CONCLUSIONS .............................................................................. 144
LITERATURE CITED ........................................................................ 146

LIST OF TABLES

CHAPTER 2:

1.

Site characteristics of location, elevation, soil type, land

capability class, pH, P and K levels .......................................

Planting date, seeding rate, and variety selection for MSU

and ROSE .......................................................................
Postemergence herbicide applications and timings ...................

Harvest dates at MSU and ROSE

Total monthly precipitation (cm) at MSU (East Lansing, MI)
and ROSE (Rose Township, MI) in 2007-2008. Long-term

means (1970-2000) included for comparison ...........................

Average monthly air temperatures (°C) at MSU
(East Lansing, MI) and ROSE (Rose Township, MI) in

2007-2008. Long term means (1970-2000) included for

comparison“

. Monthly and yearly accumulation of Growing Degree Day

units at MSU (East Lansing, MI) and ROSE
(Rose Township, MI) during 2007 and 2008, calculated

using a base temperature of 10°C .........................................

Mean crop yields of bioenergy crops in 2007 and 2008 at
MSU (n=4) and ROSE (n=6). Means within a year and crop
followed by a common letter are not different at (p < 0.05)

using Fisher’s Protected Least Signiﬁcant Difference test ..........

Results of fatty acid methyl esters (FAME) analysis of
oilseed crops at MSU and ROSE, 2007 and 2008 dates
combined (mean percent of total oil content by weight); %
FAME followed by a different letter within a fatty acid are
signiﬁcant at p < 0.05 using Fisher’s Protected Least

Signiﬁcant Difference test ....................................................

10. Economic return utilizing marginal land (ROSE) compared

to traditional cropping system (MSU) acreage ........................

vi

...... 55

..... 58

...... 59

60

..... 68

...69

...... 70

..... 72

...... 75

...... 81

CHAPTER 3:

1.

Site characteristics of location, elevation, soil type, land

capability class, pH, P and levels .........................................

. Planting date, seeding rate, and seed selection for MSU,
LC, and UP ........................................................................

Postemergence herbicide applications and timings ....................

Harvest dates for given crops at MSU, LC, and UP ....................

Total monthly precipitation (cm) at MSU (East Lansing, MI),
LC (Lake City, MI), and UP (Chatham, MI) in 2007-2008.

Long-term means (1970-2000) included for comparison ..............

Average monthly air temperatures (°C) at MSU
(East Lansing, MI), LC (Lake City, MI), and UP

(Chatham, MI) 2007- 2008. ”Long term means ...(1 970-2000)

included for comparison”

Monthly and yearly accumulation of Growing Degree Day
units at MSU (East Lansing, MI), LC (Lake City, MI), and UP

(Chatham, Ml) during 2007 and 2008, calculated using a

base temperature of 10°C...

Mean crop yields of bioenergy crops in 2007 and 2008 at MSU
(n=4), LC (n=6), and UP (n=6). Means within a year and crop
followed by a common letter are not different at (p < 0.05)

using Fisher’s Protected Least Signiﬁcant Difference test. ..........

Results of fatty acid methyl ester (FAME) analysis of oilseed
crops at MSU, LC, and UP, 2007 and 2008 dates combined
(mean percent of total oil content by weight); % FAME
Followed by a differing letter within a fatty acid are statistically
signiﬁcant at p < 0.05 using Fisher’s Protected Least

Signiﬁcant Difference test .....................................................

10. Total fuel yielded from crops grown on marginal sites

(LC and UP) compared to a traditional cropping system

(MSU) ..... . .........................................................................

11. Economic return utilizing marginal land (LC and UP)

compared to traditional (MSU) cropping system acreage .............

vii

....94

....96
97

....98

....106

107

...108

..... 109

....115

....122

123

LIST OF FIGURES
CHAPTER 2:
1. PCB soil sampling results from ROSE during 2007 (n=90) ............... 67

2. Corn grain ethanol yields for 2007 (MSU and ROSE) and
2008 (MSU); * represents no statistical signiﬁcance at
p < 0.05 using Fisher’s Protected Least Signiﬁcant
Difference test ........................................................................ 74

3. Mean total oil content (% by weight) of soybean, canola,
and sunﬂower of oilseed crops at MSU and ROSE, 2007 and
2008 dates combined; * represents no signiﬁcance at p < 0.05
using Fisher’s Protected Least Signiﬁcant Difference test ................ 77

4. Average crystalline cellulose content of switchgrass at MSU
And ROSE, in pg cellulose/mg switchgrass. Location means
followed by a differing letters are signiﬁcant at p < 0.05 using
Fisher’s Protected Least Signiﬁcant Difference test ........................ 79

CHAPTER 3:

1. Crop yield response and standard errors (o=0.05) at
three latitudes 42°N (MSU), 44°N (LC), and 46°N (UP) ................... 111

2. Two-year average corn grain ethanol yields at MSU, LC,
and UP, 2007 and 2008 data combined; means followed
by a differing letter are statistically signiﬁcant at a p < 0.05
using Fisher’s Protected Least Signiﬁcant Difference test ............... 113

3. Mean total oil content (% by weight) of soybean, canola,
And sunﬂower of oilseed crops at MSU, LC, and UP, 2007
And 2008 dates combined; means followed by a differing
letter are statistically signiﬁcant while * represents no
statistical signiﬁcance at p < 0.05 using Fisher’s Protected
Least Signiﬁcant Difference test .......... . ...................................... 1 1 8

4. Average crystalline cellulose content of switchgrass at
MSU, LC, and UP in pg cellulose/mg switchgrass; means
followed by a differing letter are statistically signiﬁcant at
p < 0.05 using Fisher’s Protected Least Signiﬁcant -
Difference test ........................................................................ 120

viii

Chapter 1. Review of Literature
INTRODUCTION

In 2009, the operational capacity of Michigan ethanol plants reached 780
million liters requiring 25% of the state’s corn acreage (United States Department
of Agriculture, National Agriculture Statistics Service, 2008). Additional acreage
will require exploration into land not currently under production agriculture.
Therefore there is an interest in producing bioenergy crops on marginally
productive lands which can include USDA soil classiﬁcations of III, IV, as well as
VI and VII. Growing biofuel crops on marginal land can offset some of the strain
imposed on the food supply by biofuels. Petroleum based fuel supplies are not
going to be sufﬁcient to meet our energy needs because global fossil fuels are
ﬁnite in supply and are becoming more expensive to extract as supplies are
diminished. Hubbert (1956) projected the world’s oil reserves will be depleted in
2045. Meanwhile, the US. is the top ranked nation for petroleum consumption
using more than 20 million barrels of petroleum a day (Organization of Petroleum
Exporting Countries, 2007). In 2008, world population reached 6.7 billion and in
2050 it is projected 9.4 billion people will inhabit the earth (United States Census
Bureau, 2008). This will place unprecedented demands on food and energy
resources. Bioenergy crops production on marginal land is increasingly seen as
an essential component of future energy plans.

Studies indicated that switchgrass is competitive under marginal
conditions (Mooney et al., 2005) and that yields of corn grain and stover removal

can be similar to switchgrass yield under marginal environments (Varvel et al.,

2008), yet no multi-crop comparisons with starch, cellulose, and oilseeds have
been reported. Additionally, little information is available on the growth and
quality of bioenergy crop production on remediated brownﬁeld sites. It is
currently unknown if marginal land can produce adequate yield to offset capital
investment or if crops grown on various types of marginal land will yield fuel of
adequate quality.

RATIONALE

In 2008, world population reached 6.7 billion and in 2050, it is projected
9.4 billion people will inhabit the Earth (United States Census Bureau, 2008). Of
the total number of inhabitants, the life expectancy of these individuals is also
going to increase. In 1996, life expectancy on a wortdly basis was 63.1 years
while following 2007, expectancy was 65.8 years (United States Census Bureau
lntemational, 2008). An increase of population and projected lifespan will result
in an increased demand for resources essential for life. Clean water, feed, food,
ﬁber, and fuel will be demands of the growing population which will need to be
met. To meet these critical human needs, marginal land will need to be explored
for energy production to not disrupt the US. food supply chain.

According to several estimates, the world’s oil reserves will be depleted in
approximately 40 years leaving the world in a fuel shortage resulting in increased
oil prices (Dhugga, 2007; Hubbert, 1956). The US. does not have adequate
petroleum reserves to meet the needs of the growing population as many feel
as though US. oil had peak production in the 1970’s and now does not have a

means to meet the growing population’s fossil fuel needs of the country (Hirsch

et al., 2005). Political instability of oil producing regions coupled with the growing
population demand for fuel and increased prices of petroleum prices have
spurred interest in alternative fuel sources such as renewable fuels or biofuels,
which have gained market strength over the last few years (Dhugga, 2007).

Alternative energy has had a long proven history in the US. and the world,
one which has been forgotten due to relatively low petroleum prices. Biofuels
evolved with the internal compression engine and ﬁrst reached the public in 1898
when Rudolph Diesel demonstrated the compression ignition engine at the
World's Exhibition in Paris. The engine operated using peanut oil which is
considered the original biodiesel. Henry Ford’s ﬁrst vehicles, the 1896
Quadracycle and later in 1908 the Model T, were engineered to operate on pure
ethanol which was derived from com grain alcohol (Solomon et al., 2007).
During times of petroleum shortages such as World War .' and II and the energy
crisis of the 1970’s, as well as increased petroleum prices and environmental
degradation, the United States has reverted to the renewable liquid fuels ethanol
and biodiesel (Nass et al., 2007). Today with decreased oil reserves, increasing
population base, rising costs of petroleum, political fallout, and increased carbon
emissions in the atmosphere, biofuels and renewable energy research appears
quite attractive to consumers and is required to meet our societal needs.

Energy security critically impacts foreign policies. A domestic energy
source would create an economic stimulus in communities as well as ensure a
dependable fuel source (Sutterﬁeld, 2007). The US. is vulnerable to oil supply

disruptions because the global supply of oil is concentrated geographically as

66% of proven oil reserves are located in the Middle East and only 5 percent are
located in North America (NEPDG, 2001). This geographic concentration of oil
also funnels governmental power to control energy in the hands of a few
producers. Many of these leading countries in world oil production are either
ruled by unstable governments or maintain adversarial relationships with the
United States (Luft, 2005). US. dependency on these countries for oil
endangers our national security (Sutterﬁeld, 2007).

Climate change in addition to energy security requires large-scale
transition to biofuels (Farrell et al., 2006). Renewable fuels have potential to
reduce green house gas emissions (GHG) and global warming potential (GWP)
(Grifﬁn and Lave, 2006; Sheehan et al., 1998). Furthermore, Fronning et al.
(2008) found fertilization regimes incorporating compost and manure
amendments in bioenergy crops can further improve their carbon footprint. Hill et
al. (2006) found corn grain ethanol yielded 25% and biodiesel yielded 93% more
energy than what was invested in its production. Additionally they found GHG
are reduced 12% by the production and combustion of ethanol and 41% by
biodiesel. Schmer et al. (2008) showed a life-cycle analysis estimated that
ethanol from switchgrass averaged 94% lower GHG than from gasoline. For
reasons of efﬁciency of petroleum use and GHG emissions, Farrell et al. (2006)
suggest biofuels will require cellulosic technology.

Challenges of bioenergy crop production arise when determining crop
production location. As population continues to increase, more crop yield will

need to be achieved to meet dietary needs. Between the years of 1982 and

1997, 4.5 million hectares of cropland were converted to urban land uses
(Greene and Stager, 2001 ). In 2002, US. Department of Agriculture estimated
176,700,000 hectares in the US. were cropped, down from 180,200,000
hectares in 1997. In the ﬁve year period between 1997 and 2002, 4.5 million
hectares of agricultural land in the US. shifted from rural to urban uses (United
States Department of Agriculture, 2002). Additionally, the Energy Independence
and Security Act of 2007 now requires 36 billion gallons of biofuel to be produced
from renewable sources found within the United States by 2022.

Increased yield from production agriculture will need to be achieved on
less fertile land base areas as human population expands into area devoted to
agriculture and exploration into marginal lands for bioenergy production is
necessary. Currently, the allocation of crops for bio-based fuel sources has
. raised tensions. The food verse fuel debate challenges human ethics whether
food crops should be taken out of the food supply and converted to fuel. Third
world countries around the wortd are facing limited food supplies while developed
countries convert food into fuel. Crop production on marginal lands can ease
competition between acres devoted to bioenergy and food production.

LAND CLASSIFICATIONS

The United States Department of Agriculture has 8 land capability classes
for agricultural suitability. The Land Capability Classiﬁcation is a system of
grouping soils primarily on the basis of their ability to produce common cultivated
crops and pasture plants without deteriorating soil properties over a long period

of time. Capability classes are based primarily on soil type, slope, and drainage.

Classes I and II are classiﬁed as prime agricultural land with few limitations that
would restrict use. Classes Ill and IV are capable of producing crops yet there
are severe limitations that can include slope, erosion, ﬂooding, low-moisture
holding capacity and fertility, salinity or sodium, and adverse climatic conditions.
Classes I through IV are considered capable of producing cultivated crops with
good management practices including conservation. Classes III and IV are
comprised of various land features that make them environmentally sensitive
lands available to be enrolled in the Conservation Reserve Program, in which
landowners receive govemmental aid for preserving environmentally sensitive
acres. Class V has restricted use due to a high water table while classes VI and
VII are best suited for perennial vegetative species, but may be capable of
producing some specialized crops with highly intensive management. Classes
VI and _VII may need to implement a vegetative cover used to prevent severe
erosion because of severe erosion potential. Land capability class Vlll soils have
limitations which restrict use making soils not suitable for managed vegetative
production (Natural Resources Conservation Service, 2009). Primarily, marginal
land encompasses soils of III, IV, as well as VI and VII classiﬁcation.

In the contiguous 48 states in the US, 174.6 million hectares are
identiﬁed as cropland (USDA, 1999) however not all acres available for crop
production are cultivated as 7.7 million hectares are left idle. The Conservation
Reserve Program (CRP) accounts for 12.1 million hectares (USDA FSA, 1999).
Marginal cropping lands can be enrolled in the Conservation resen/e program

authorized by the Food Security Act of 1985, and administered by the Natural

Resources Conservation Service (NRCS) of the United States Department of
Agriculture (USDA). Landowners receive ﬁnancial assistance for implementing
environmental protection practices including planting grasses, trees, or other
conservation cover. The CRP started in part to the 1985 Farm Bill and sets
asides environmentally sensitive hectares under 10 to 15 year contracts.
Through 2009, 15.9 million hectares are enrolled in the CRP program with
acreage dropping to 12.9 million 2010 through 2012 (United States Department
of Agriculture, 2008).

All acres should not be converted out of CRP for bioenergy production on
marginal lands. Hectares enrolled in buffer strips to protect water quality,
classiﬁed as wetlands, essential to watershed management, and critical habitat
hectares in Wildlife Conservation Priority Areas should not be taken out of CRP.
According to Walsh et al. (2003) 6.8 million hectares of the 12.1 million hectares
of CRP are available for conversion to growing bioenergy crops for economic
reasons including crop prices, costs of inputs, and attainable yields under
marginal conditions. Under current legislation, harvest of CRP is prohibited
except under emergency conditions and does not make biomass available for
bioenergy conversion.

Marginal land is land which is commonly seen as unsuitable for cultivation
of arable crops. Maximum or optimum cropland production depends on the
growing season environment and the skill of the producer to identify and then

eliminate or minimize factors that can reduce yield potential. There are numerous

factors affecting crop yield potential. These can be categorized into three main
factors including climate, soil, and crop (Havlin et al., 2005).

Worldwide, the dominant stress reducing crop yield and directly correlating
to marginal land is plant available water. Water is the primary inﬂuencing
management factor when determining productivity of agricultural cropping
systems (Stone et al., 2002; Jiang and Thelen, 2004). Aridic soil moisture
regimes constitute dry conditions for half the growing season. During growing
seasons, rainfall is low in most months and levels of evapotranspiration are high,
limiting moisture recharge in arid regions (Foth, 1978). Continuous moisture
stress accounts for 36.5 million square kilometers in the US which occupies
27.9% of the total available landbase (USDA-ERS, 2003). In northern climates,
low temperatures limit yield and comprise 21.8 million square kilometers of the
earth’s potential landbase. These regions do not accumulate adequate growing
degree days (GDD) for most crops such that plant growth and consequently crop
yield is limited.

There are four main physical characteristics used to determine the
capability of land: soil texture, drainage, slope, and erosion. Soil factors primarily
affecting marginality can include low base fertility, low productivity, coarse soil
texture, high erodibility, soil and ground water contaminants, and highly disturbed
soils. Soil fertility is the status of a soil with respect to its ability to supply
essential nutrients for plant growth without having a toxic concentration of any
elements (Foth, 1978). Soil productivity is characterized as the ability of a soil to

produce an adequate amount of plant growth with optimum management, which

includes issues of fertility, drainage, and other environmental factors (Foth,
1978). All productive soils are fertile, however not all fertile soils are productive.

Depending on the range and characteristics of fertility, soils can support
plant growth even though land may be termed marginal. Typically, coarse soil
texture affects fertility of land bases however soils with high Cation Exchange
Capacities can be classiﬁed marginal. The Cation Exchange Capacity (CEC) of
soil is strongly affected by the nature and amount of clay minerals or organic
matter present within a soil. The CEC of soils dictate the nutrient availability of
cations soils are able to adsorb and make accessible for plants. High levels of
(50-100 meq/100 Q) can be associated with organic soils, and clay and clay loam
soils (20-50 meq/100 9). Coarse textured soils have substantially lower CEC
values (3-5 meq/100 g) and can be classiﬁed as marginal due to low nutrient
availability (Havlin et al., 2005). I

Soil erosion, which is dependent on soil type and slope, can determine
marginality. Soil erosion occurs when soil is removed through the action of wind
and water at a greater rate than it is formed. Erodibilty of soil is inﬂuenced
primarily by soil stmcture speciﬁcally by the proportion, size, and bulk density of
the erodible soil particles as well as climate and topography (Havlin et al., 2005).
Erodibility is prevalent on land capability classes of III and IV as well as Vl-Vlll.
Erosion is higher on land with greater than 6% slope. Erosion removes soil
horizons O and A consisting of organic soil material and mineral soil material,
which is typically nutrient laden. The O and A horizons are zones in which roots

permeate most frequently. Erosion is a natural process of weathering, yet

accelerated erosion occurring more rapidly than by normal, natural, geological
erosion is detrimental for nutrient availability and optimal plant growth. Tolerable
soil loss due to erosion for sustainability must be below 11.2 t ha‘1 as it takes 30
years to form one inch of topsoil in Michigan (Havlin et al., 2005).

Other examples of marginal soils can include climatic variables. Climatic
factors are dependent on geography and can include shortened growing seasons
and low precipitation. Growing Degree Days (GDD) track crop maturity and is
calculated using the average of daily high and low temperatures and subtracting
a base temperature of 10°C. Locations receiving more GDD have increased
potential to support crop growth and to achieve higher yield totals (Swanson and
Wilhelm, 1996). Locations with shortened growing seasons accumulate minimal
GDD. Precipitation is location dependent and vital for crop growth. A location
may accumulate high GDD however if rainfall is limited, the location will still be
considered marginal.

There are also additional acres of public and privately owned marginal
land which are idle and not enrolled in the CRP can support growth of bioenergy
crops. Brownﬁelds and Superfund sites are industrial areas that have
contaminated soils and groundwater from past chemical contamination and
abandoned hazardous waste, but hold value for economic development or
production agriculture if the pollutant is controlled and removed (Licht and
lsebrands, 2004). Total acreage designated with superfund or brownﬁeld status
is unknown as new locations are continuously discovered. In June 2008, 68

Superfund sites were identiﬁed in Michigan by the Environmental Protection

10

Agency comprising a land area of approximately 2,000 hectares. An additional
3,645 brownﬁelds were identiﬁed with acreage unknown (Environmental
Protection Agency, 2008). Site areas range from city blocks to hundreds of
acres. These lands are not enrolled in production agriculture systems and
depending on composition could be utilized for bioenergy crop production.
Superfund and brownﬁeld sites are regulated to preclude food crop production,
but conversion to energy sources is acceptable (Environmental Protection
Agency, 2004). Crops cultivated in areas of known pollution are acceptable
sources of biomass for bioenergy (Licht and lsebrands, 2004) and may prove to
phytoremediate an area as some plants are better for taking up heavy metals or
volatile organics (Westphal and lsebrands, 2001). A heterogeneous environment
suggested by Lafortezza et al. (2004) is an approach which optimizes biomass
and promotes ecological concepts. Plants take up essentially all nutrients as
water soluble compounds through the roots or foliage. Thus, for uptake to occur,
contaminants must reach the plant through water in the root zone (Licht et al.,
2001)

Fallow agricultural ﬁelds are also potential acreage that can be used to
cultivate bioenergy crops. In 2007, 2.7 million hectares in Michigan were
devoted to production agriculture. According to the Land Policy. Institute at
Michigan State University, 4.6 million additional hectares are potentially available
for production agriculture in Michigan. These lands are not currently cropped,
pastured, forested, or wetlands and not currently under development (Land

Policy Institute, MSU, 2007). While all available acres should not be used for

11

production agriculture for reasons of habitat preservation, naturalized acres, and
sensitive acreage, Michigan producers could use a portion of the available land
for bioenergy production. Ceotto (2007) warns of meat and milk markets
suffering if marginal grazing lands are transitioned to bioenergy production. Idle
land, whether it is a site with low fertility or contaminated soil to produce a
beneﬁcial feedstock is an efﬁcient land use strategy.

The utility of using marginal land to produce bioenergy crops on a
sustainable basis will depend upon economics and environmental practicality and
proﬁtability. Keeping the cost of biofuels low will stimulate consumers to seek '
alternative fuels and perhaps the environmental implications will keep consumers
supportive of a renewable approach. Additionally, incorporation of cap and
trading of carbon and carbon tax legislation may make biofuels economical for
consumers and producers alike. Information on production amendments, soil
structure and characteristics, site accessibility for ﬁeld scale farm equipment,
public perception and environmental ramiﬁcations is required to assess the long-
terrn sustainability of biofuels as well as production of biofuel crops on marginally
productive soils (Gleeson, 2007).

ENERGY

Daily, more than 20 million barrels of petroleum were consumed in the
United States (Organization of Petroleum Exporting Countries, 2007).
Approximately 182,000 Mg of carbon dioxide was emitted daily due to 2007
petroleum consumption. Petroleum has been utilized due to BTU efﬁciency and

ease of conversion to energy. Cost and environmental degradation has spurred

12

bioenergy interest as well as published net energy gain of corn ethanol and
biodiesel at 25 and 93% (Hill et al., 2006).

Solar radiation striking the earth on an annual basis is equivalent to
178,000 terawatts, which equates to 15,000 times that of current global energy
consumption (Food and Agriculture Organization of the United Nations, 1997).
Most plants and all ﬁeld crops are photoautotrophic meaning they are able to
synthesize food directly from inorganic compounds using light energy. Plants
convert light into chemical energy with a 6% maximum photosynthetic efﬁciency.
Photosynthetic efﬁciency can vary with the frequency of the light being
converted, light intensity, temperature and proportion of 002 in atmosphere (Zhu,
2008). Using crops for bioenergy production, sunlight is used to grow biomass.
Conversion of biomass into fuel and consumption of biofuel is, in theory, the
combustion of sunlight. Biofuel crops are renewable and protective of the
environment when sustainable methods are employed to manage, harvest and
process the crops.

In recent years, a new round of enthusiasm in biomass and bioenergy has
been initiated with the recognition that the global crude oil reserve is ﬁnite, and its
depletion is occurring much faster than previously predicted. In addition, the
environmental deterioration resultingfrom the over consumption of petroleum
derived products, especially transportation fuels, is threatening environmental
sustainability. Using bioenergy feedstocks as an energy source captured and

harvested from solar energy is environmentally sustainable.

13

CROP CONVERSION TO BIOENERGY

Plants have provided food, feed, and ﬁber for the world. Converting grain
and biomass efﬁciently to deliver a liquid fuel source has not been
commercialized. Many conversion systems have been investigated.
Fermentation and transesteriﬁcation have been used to convert starch crops,
biomass, and oilseeds into a fuel source.

Corn grain fermentation has occurred throughout world development.
Historically, corn was fermented to create an alcohol. The US. saw potential for
the use of ethanol as a renewable fuel source due to the fuel crisis of the 1970’s.
The widely used milling process to convert corn grain starch to ethanol requires
three days for the fermentation process (Solomon et al., 2007).

Lignocellulosic or cellulosic ethanol production converts plant materials
into sugars and then ferrnents the sugars into ethanol. Lignocellulosics are
comprised of three primary constituents: lignin, cellulose and hemi-cellulose.
Lignin chemistry is close to that of low rank coal. It has a high heating value and
usually is regarded as a fuel resource, however is not a source of ethanol. Hemi-
cellulose and cellulose are long chain polymers of ﬁve-carbon sugars (pentoses)
and six carbon sugars (hexoses). Hemi-cellulose is mostly a polymer of xylose,
while cellulose is almost entirely a polymer of glucose. The bonds linking the ‘
sugar monomers together in the hemi-cellulose and cellulose polymers can be
chemically broken by hydrolysis. Following hydrolysis, the sugars can be
processed to produce ethanol or other products (Kaylen et al., 2000). High

quality ethanol feedstock has zero nitrogen content and is high in cellulose.

14

Nitrogen reduces the conversion efﬁciency of fuel into energy and can become

an air pollutant after combustion. Therefore, zero or low nitrogen fertilizer
applications and a single yearly harvest after the plants have died back fully in
the winter produce the best feedstock as well as the highest amount of above
ground biomass for ethanol production (Westphal and lsebrands, 2001).
Problems arising with cellulosic ethanol stems from conversion economics (Lynd
et al., 1991 ), transportation and storage bottlenecks (Lofstedt, 1996), and
producer agronomic adoption. As of 2007 there is no market for switchgrass as
an energy crop (Epplin et al., 2007).

Ethanol has gained support as the US Environmental Protection Agency
(EPA) has shown interest in making gasoline blends more environmentally
friendly. In the 1980’s, the EPA started to phase out leaded gasoline. In
attempts to ﬁnd another octane booster and volume extender, ethanol was used
in the blend. The rapid growth of the ethanol industry can be attributed to
phasing out the additive methyl-tertiary—butyl ether (MTBE) (Solomon et al.,
2007) and as research indicates ethanol is relatively water soluble and MTBE is
potentially carcinogenic (Belpoggi et al., 1995).

Corn based ethanol has some important potentially negative
environmental impacts. Increased soil erosion, loss of biodiversity, and higher
volatile organic compound and NO)( pollution are all hazards of agronomic crop
production (Pimental and Palzek, 2005). By employing marginal lands and
conservation practices, these risks can be minimized. Corn based ethanol,

provides a positive energy return on investment and a 10-15 percent reduction in

15

CO; emissions (Farrell et al., 2006; Solomon, 1980; Shapouri et al., 2002;
Hammerschlag, 2006).

Biodiesel or fatty acid methyl esters (FAME) results from the process of
using transesteriﬁcation to turn vegetable oils into esters. Using an alkaline
catalyst, vegetable oil is separated into respective fatty acid chains and into alkyl
esters. The process yields methyl esters and glycerin (Ma and Hanna, 1999).
Five fatty acids are produced in biodiesel production including; palmitic acid
(C16:0), stearic acid (018:0), oleic acid (C18:1), linoleic acid (018:2) and
linolenic acid (C1823). Among these, palmitic and stearic have saturated
structures, which are stable during storage and have higher melting points (62 °C
and 69.3 °C). While saturated fatty acids are very useful in improving food taste,
saturated fatty acids in biodiesel contribute to cold ﬂow issues associated with
the fuel. The unsaturated fatty acids, such as oleic (13.4 °C), linoleic (-5 °C) and
linolenic (-11.3 °C), have lower melting points and have more unstable
properties. In the biodiesel as well as food industry, optimizing oleic fatty acid is
most ideal as the fatty acid has a relatively lower melting point than saturated
fatty acids and longer shelf life than polyunsaturated fatty acids (Dardanelli et al.,
2006; Piper and Boote, 1999; Primomo et al., 2002; Wilcox and Cavins, 1992).

POTENTIAL BIOENERGY COMMODITIES

National energy demands may possibly be met utilizing domestic
commodities. Biodiesel and ethanol can be produced from current food crops as
well as biomass. Three oil seed crops including soybean, canola, and sunﬂower

can be used to derive biodiesel while corn grain and switchgrass biomass are

16

candidates for ethanol production. Biodiesel, derived from triglycerides by
transesteriﬁcation with methanol, has attracted considerable attention as a
renewable, biodegradable, and nontoxic fuel. Ethanol or ethyl alcohol is made
from the fermentation of sugar which can be found in the starch of corn grain or
cellulose within cell walls of all living plants.

No ideal bioenergy crop has yet to be identiﬁed. Raghu et al. (2006)
outlined characteristics of ideal ecological traits of bio-energy crops.
Characteristics include C4 photosynthesis, long canopy duration, perennial
lifecycle, no known pests or diseases, rapid growth in the spring to outcompete
weeds, seed sterility, partitioning nutrients to below ground components in the
fall, high water use efﬁciency, and most importantly high fuel yields per area.
According to Raghu et al. (2006) outlined ecological traits, tended to favor a
cellulosic ethanol crop for use in bio-energy production. Marginal lands have the
ability to support growth of all potential bioenergy crops.

CORN

Corn grain (Zea mays ssp. mays) in 2008, is the current crop used to
produce ethanol in the US. Corn is a C4 warm weather annual grass which was
domesticated in Mesoamerica. Prior to domestication, the wild grass was native
to Mexico. The oldest carbon dated corn is believed to be cultivated around
5000 BC. The record originated in the Valley of Tehuacan in Mexico (Magness
et al., 1971). The crop has been manipulated by humans through hybridization

to achieve higher grain and biomass yields.

17

Corn is a high yielding C4 grass which conducts photosynthesis efﬁciently
due to a 002 concentrating mechanism that mitigates or eliminates
photorespiratory carbon loss (Specht, et al. 1999). This pathway allows C4
grasses to withstand periods of water stress and high temperatures (Kemp and
Williams, 1980). Stresses of moisture availability and temperature can be
intensiﬁed under marginal conditions.

Corn grain can be classiﬁed into ﬁve categories consisting of dent, ﬂint,
ﬂour, pop, and waxy corn (Smith, 1995). Dent corn is the corn that is primarily
cultivated for food, feed, ﬁber, and fuel use and encompasses the majority of
acres in the United States. The endosperrn of dent corn, which consists of
starches, is vital for seedling growth and is utilized by the ethanol production
industry. The endosperrn accounts for about 82% of kernel dry weight and is the
source of energy and protein for the germinating seed. A non-fermentable
portion, the germ, is the embryo from which a new corn plant develops. About 25
percent of the germ is corn oil which is high in polyunsaturated fats which can be
utilized for biodiesel production. The pericap or bran protects the seed before
germination and preserves nutrient value in the kernel. After fermentation,
distillers grain (wet or dry), used for animal rations, consists of the germ and bran
(National Corn Growers Association, 2007). Whole grain composition totals 72%
starch and 3.8% oil.

Corn yields and subsequently energy yields have benefited from
advancements and improvements in production agriculture. Corn yields in the

United States have increased over the last 100 years as 1908 yield totals

18

averaged 1.7 Mg ha'1 while in 2008 totals averaged 9.7 Mg ha'1 (United States
Department of Agriculture, National Agriculture Statistics Service, 2008).
Changes in national averages of corn yield involves many variables including the
use of hybrids, increased fertilizer use, longer seed ﬁlling periods, tolerance to
high plant populations, improved plant greening characteristics during seed
ﬁlling, more upright leaves, decreases in seed protein concentration (Duvick,
2005), and increased stress tolerance (Tollenaar and Wu, 1999). Management
practices also contribute to higher yield. Higher plant populations (Duvick, 2005),
rates of fertilizer, especially nitrogen, introduction of herbicides and improved
weed control, earlier planting date, and narrower rows (T royer, 2004;
Widdicombe and Thelen, 2002) have all increased corn yield. Additionally,
herbicide tolerant crops have increased corn yields by improved weed control
and soil water conversion due to reducing tillage requirements. Under marginal
conditions, crop stresses will be ampliﬁed and despite crop improvements, yield
totals may be negatively affected.

Corn requires a high investment of inputs to maximize grain yield and
economic return. Nitrogen signiﬁcantly affects yield of corn grain. In Michigan,
nitrogen amendments employing the Maximum Return To Nitrogen concept
should total approximately 112 kg N ha". With the development of new hybrids .
employing nitrogen efﬁciency, nitrogen recommendations have decreased 0.5-
0.7 kilograms of nitrogen per hectare, equating to each bushel requiring

approximately 0.9 kg N (Wamcke, 2008). These inputs have the potential to limit

19

economic return especially on marginal land where lower yields may make corn
yields cost prohibitive.

Planting population is vital when sowing a corn grain crop for optimum fuel
return. Corn yield and ethanol yield responds signiﬁcantly to total plant density.
Sarlangue et al. (2007) found optimum plant density to range between 10.3-13.7
plants rn'2 due to biomass plasticity and reproductive partitioning. Yield response
to plant density is greatest for shorter season hybrids, intermediate for medium
duration, and lowest for full season hybrids (Widdicombe and Thelen, 2002). At
low plant density, short-season hybrids intercept less radiation and produce less
biomass than full-season hybrids. Planting populations will reduce water
availability in soils and particularly in marginal soils in which water availability is a
concern. Higher populations will result in higher grain yield; however marginal
land may not be capable of providing adequate water to support high
populations.

Total energy yield of com grain ethanol has been widely debated. On
average, 25.4 kg of corn will yield 10.6 liters of ethanol, 8.2 kg dried distillers
grain and 8.2 kg 002 (Thelen, personal communication). Pimentel and Patzek
(2005) found 29% more fossil energy was being used to produce ethanol than it
produced. Since release, this ﬁnding has been disputed and discounted by many
(Kavanagh, 2006). Other studies found starch based ethanol produced from
corn grain yields 25% more energy than invested (Hill et al., 2006) and a total net

energy ratio of 1.38:1 (Lorenz and Morris, 1995), respectively.

20

Controversy surrounds the collection of crop residues particularly corn
stover for conversion into bio-based fuels. Sustainable energy production from
crop residues must take into consideration that residue removal could increase
erosion, reduce crop productivity, and deplete soil carbon and nutrients.
According to Graham et al. (2007) approximately 30% of stover could be
removed for energy production in given counties in the Midwest. Removal of
30% is based solely on erosion control and does not take into consideration the
necessity to maintain or enhance soil organic matter and tilth. Soil moisture and
nutrient replacement are two more major concerns of residue removal ampliﬁed
in marginal areas as all stover must be returned to the ﬁeld to maintain soil
moisture for the next crop (Wilhelm et al., 2004).

SOYBEAN

Soybean (Glycine Max (L.) Merr.) in 2008, is the current crop used to
produce biodiesel in the US. Soybean is native of East Asia. The soybean was
introduced as a forage crop in North America in 1765. In 1908, 20,000 hectares
were grown for forage production (Probst and Judd, 1973). Soybean was not a
major grain crop in the United States until the 1920’s (Hymowitz and Harlan,
1983). Production statistics were ﬁrst reported in 1924 when the average
national yield was 0.74 t ha". In 2008, 22.9 million hectares were planted and
yields averaged 2.69 t ha'1 (United States Department of Agriculture, National
Agriculture Statistics Service, 2008), however, in recent years national yield data
suggest soybean yields are showing plateau under known agronomic techniques

(Nafziger, 2004).

21

Soybean is a leguminous crop. Soybean root nodules containing

symbiotic Rhizobium bacteria reduce N2 ﬁom the atmosphere and convert to NH3

which can be utilized by the plant (Phillips, 1980). In the symbiotic relationship
Rhizobium bacteria supply the soybean with a nitrogen source while Rhizobium
gains carbohydrates (Hoeft et al., 2000). Soybeans provide 33.7 kg N ha‘1 for
the following crop (Wamcke et al., 2004) and in the US. supply 5.7 T9 N
annually (Herridge et al., 2008). Under marginal land conditions, the production
of N for subsequent crops is positive. A rotational crop, planted following a
soybean crop, will beneﬁt from additional N provided by soybean. Leguminous
crops such as soybeans will lower the inputs needed in marginal land rotations
which impacts the cropping system by increasing economic returns.

Soybean is a legume with a C3 carbon ﬁxation photorespiratory pathway.

Carbon losses due to C3 biology substantially reduce photosynthetic output

particularly under warmer temperatures and high water stress conditions (Specht

et al., 1999). The reduced photosynthetic output will negatively affect crop yield
under periods of crop stress. Marginal land crop stresses of water availability,
high temperatures, low fertility and erosion may increase negative responses
which would limit the utility of soybean as a bioenergy crop on marginal lands.

Soybean has been cultivated traditionally as a protein crop used for

livestock feed. Livestock production beneﬁts from soybean attributes as soybean

seed is comprised of approximately 40% protein, 25% carbohydrates, and 20%
oil. Poultry and swine consume a majority of the crop at 54 and 26% (American

Soybean Association, 2007). Domestic use of soybean, though low in

22

I -O_h.

percentage, is dominated by frying oils, as a component of margarine and salad
dressing, and in the production of inks, paints, and cosmetics (Sheaffer and
Moncada, 2008).

Changes and successes of the soybean used as a grain crop include
longer seed ﬁlling periods (Gay et al., 1980) and decreased lodging and often
shorter plants (Specht and Williams, 1984). In some cultivars, seed protein
decreases with increases in seed oil concentration (Morrison et al., 1999) have
negatively inﬂuenced the United States soybean export market. Soybean grain
yields have also beneﬁted from narrow rows (Bullock et al., 1998). Many credit
earlier planting, adoption of conservation tillage techniques, and reductions in
harvest losses for increasing yield (Heatherly and Elmore, 2004). Glyphosate [N-
(phosphonylmethyl)-glycine)] resistant soybean used with glyphosate in
agronomic systems enables a post herbicide application program which kills a
wide range of weed species without reducing yields (Nelson and Renner, 1999)
and reduces tillage requirements vital in conserving soil moisture. Marginal land
production will beneﬁt from recent advances in soybean agronomy.

Soybean oil and fatty acid methyl esters (FAME) are produced by
transesteriﬁcation of soybean. Total oil content in soybean averages
approximately 20 percent oil per volume dependent on maturity zone (Piper and
Boote, 1999). The quality of soybean oil depends upon the fatty acid methyl
ester (FAME) proﬁle of ﬁve major fatty acids (Primomo, 2002). The ﬁve major
FAME comprise over 98% of the fatty acids in soybean oil (Canakci and Gerpen,

2001; Dubois et al., 2007). On average FAME of soybean oil consists of (g kg '1)

23

linoleic 540, stearic 40, linolenic 70, palmitic 110, oleic 240 weight per volume of
oil respectively (Primomo, 2002). In the food and biodiesel industry, optimizing
oleic acid is ideal as it has a relatively lower melting point than the saturated fatty
acids (palmitic and stearic) and longer shelf life than polyunsaturated fatty acids
(linoleic and linolenic).

FAME quantities are dependent on environmental conditions (Primomo,
2002). FAME quantities decrease when subjected to environmental stresses
such as drought and temperature (Dombos and Mullen, 1992; Wolf et al.,1982).
Linolenic and linoleic acid decreased and oleic acid increased as the temperature
increased while palmitic and stearic acids remained unchanged. Additionally,
Karaaslan found sowing date affected fatty acid composition (2008). A study
performed by Howell and Collins (1957), found fatty acid levels were not affected
by environmental factors such as photoperiod, light intensity and quality, nitrogen,
phosphorus, potassium, and sulfur nutrition, or the addition of manure or plant
residues. Under marginal land conditions, FAME proﬁles are likely to show
variation. Marginal land conditions are not ideal for soybean production due to
environmental stresses and are likely to inﬂuence soy-diesel components.

The esters of soybean oil have similar physical properties, such as
. viscosity, cetane number, and energy content to No. 2 diesel fuel, while reducing
carbon, sulfur and nitrogen exhaust emissions (Knothe and Dunn, 2001). 27.2
kg of soybean grain yields 1.7 liters of soybean oil (Thelen, personal
communication). Co-products‘of soy-diesel per 27.2 kg of grain include 21.8 kg

of soybean meal which can be used as a protein source in livestock rations as

24

well as 0.5 kg of glycerin which can be used to produce soaps and cosmetics.
Soybean oil has potential for use as an oil alternative however it has some
limitations for use in direct-injection engines due to high viscosity, low volatility,
and the polyunsaturated character of the triglycerides (Schwab et al., 1988). In
2005, it was reported that 260 million liters of soybean oil was converted to
biodiesel (Urbanchuk, 2006) requiring (based on national yield data) 8.5% of the
total US yield (United States Department of Agriculture, National Agriculture
Statistics Service, 2008). Increasing soybean production on marginal lands
could increase national total yields as well as lower the percentageof commodity
allocated for fuel conversion.
CANOLA

Agronomic production of rapeseed can be traced back 3,000 years ago to
India. The crop was ﬁrst cultivated as an oil source for lamps and later was used
for cooking oil (Hougen and Stefansson, 1983). The name canola (Brassica
napus L. var. napus), adopted in 1979, was used to describe a rapeseed cultivar
with “double low” or less than 2% erucic acid. In 1985, canola and rapeseed
were separated into two distinct species (Shahidi, 1990). In 2007, the United
States harvested 2.9 million hectares and averaged 1,401 kg ha'1 totaling 4,063
million kilograms. In 2008, United States acreage decreased to 2.4 million
hectares while yield increased to 1,697 kg ha'1 totaling 4,073 million kilograms
(United States Department of Agriculture, National Agriculture Statistics Service,

2008)

25

Canola is a C3 oilseed crop that can be planted in the fall in the southern
US. and southern Europe, or spring in the northern US, Canada, and northern
Europe (Gunstone, 2004). Reduced photosynthetic output will negatively affect
crop yield under periods of crop stress. Soil conditions for optimum canola yield
include medium textured, well drained soils. Soil pH between 6.0 and 7.0 is
optimal, and yields may be reduced signiﬁcantly if pH is below 5.5 (Boyles et al.,
2008). Fall canola requires approximately 40.6 to 45.8 cm of precipitation
through the growing season while spring varieties require 20.3 to 21.1 cm of
precipitation (Oplinger, 1989).

Canola is a candidate for marginal land situations due to the deep
penetrating taproot which can breakup hardpans and improve soil structure
(Gunston, 2004). Increases in canola production can also be linked to growers
concem of Fusarium head blight (Fusarium gramenearum Schwabe) and other
disease infestations causing low yields in competing cereal crops to which canola
is not affected (Hanson et al., 2001) in addition to increased canola oil use as a
healthy cooking oil and biodiesel feedstock. Additionally, ﬁeld machinery used in
other cereal crop production is similar making transition to canola production
unproblematic.

Like other primary cash crops, canola has been subjected to genetic
modiﬁcation to increase agronomic productivity. The leading alteration has been
herbicide tolerance, mainly glyphosate resistance. Research conducted by
Brookes and Barfoot (2006) indicates use of glyphosate resistant canola

contributed to 6% higher canola yield in the US and 11% in Canada.

26

Additionally, the convenience of controlling a wide spectrum of weeds with a
single post emergence application and increase economic return explain the
rapid adoption of this technology (Harker et al., 2000). By the use of glyphosate
resistant canola, pest pressure is not ampliﬁed as ﬂeas beetles, cabbage
seedpod weevil, and lygus bugs do not exhibit a preference to invade and
damage conventional over herbicide tolerant canola (Caramo and Blackshaw,
2007). Herbicide resistance reduces tillage necessary for optimum grain yield
and conserves critical soil moisture essential for marginal land environments.

One potential problem of genetically modiﬁed canola hybrids was
addressed by Hansen et al. (2001). Research indicated canola can show
introgression under natural conditions from B. napus to B. rapa. Flax (Linum
usitatissimum L.) has the potential to hybridize transgenic traits with weedy and
wild relatives (Jhala, 2008) and pollen ﬂow from herbicide resistant canola
cultivars may occur to weedy pests and unaltered canola genetics.

Canola oil conventionally is used as cooking oil. Health beneﬁts of low
saturated fats (7%), high levels of monounsaturates (61%), no trans fat, and a
good source of polyunsaturated omega-3 fatty acid (Canola Council of Canada,
2008) which has been linked to lower the risk of coronary heart disease (Kris-
Etherton, 2002) has prompted agronomic production.

Canola seed content averages 40% oil and 23% protein (Oplinger, 1989).
Canola oil is highly suitable for production of quality biodiesel as 60% of the oil is
comprised of monounsaturated oleic acid'which demonstrates good stability and

winter operability. Canola is the biggest source of feedstock for biodiesel

27

production in the world and has become more attractive as rapeseed breeders
have improved yield levels of 2.9 ton oil ha”1 in Northern Germany (Puppan,
2002). In the US, canola can yield 11.0 liters per 22.7 kg and a hectare can yield
1122 liters (Thelen, personal communication). Rapeseed contributes 84% of the
raw material in the world used for production of biodiesel production while
soybean and sunﬂower accounted for 1 and 13%, respectively (Korbitz, 2000).
SUNFLOWER

Sunﬂower (Helianthus annuus) is native to North America. Archeological
evidence suggests Native American tribes cultivated sunﬂower for grain in
present day Arizona and New Mexico around 3000 BC The crop was cultivated
for commercialization in Russia and in 1880; a cultivar named ‘Mammoth
Russian’ was available. The crop originally was used to augment poultry dietary
rations and in 1926 sunﬂower seed was crushed for oil. In the 19703,
hybridization of sunﬂower began and traits providing additional yield and oil
enhancement as well as disease resistance were incorporated (Schneiter, 1997).

Acreage of sunﬂower in the United States in 2007 reached 970,000
hectares and averaged 1,610 kg ha'1 accumulating a total crop of 8,050 million
kilograms. In 2008, the number of acres increased to 5.9 million hectares as
average yield increased to 1,623 kg ha'1 totaling 9,576 million kilograms (United
States Department of Agriculture, National Agriculture Statistics Service, 2008).

Sunﬂower is a 03 oil seed which planted in the spring. Reduced
photosynthetic output due to C3 pathways can negatively affect crop yield under

periods of crop stress. The stem can grow as high as 3 meters and the ﬂower

28

head can reach 30 cm diameter. Sunﬂower has a well developed, deeply
penetrating root system, which following establishment can withstand droughty
conditions but depletes soil water more than sorghum, corn, and canola (Stone et
al., 2002; Merrill et al., 2007). The critical growth period for sunﬂower grain yield
is the R3 stage when sensitivity to environmental conditions is prevailing.
(Schneiter, 1997). In this stage, sunﬂower is in the reproductive stage before
ﬂowering in which the immature bud elongates more than 2.0 cm above the
nearest leaf (Schneiter and Miller, 1981 ).

Sunﬂower is typically cultivated for direct human consumption. Confection
sunﬂower seeds are normally black with white stripes and are approximately ﬁve
eighths of an inch long. The heavy hull accounts for approximately half the
weight of the seed and is loosely ﬁxed to the kernel inside to enhance
consumption. Oil sunﬂower seeds are black and much smaller than confection.
Oil seeds have been bred to enhance oil content (National Sunﬂower
Association, 2008).

As with oil seeds, fatty acid proﬁles determine oil conversion efﬁciency.
lzquierdo et al. (2006) determined the critical period for fatty acid composition in
sunﬂower at 100 to 300 degree days after ﬂowering. Sunﬂower oil contains
approximately 90% unsaturated fatty acids, combined oleic and linoleic, and 10%
saturated fatty acids, palmitic and stearic (Steer and Seiler, 1990; lzquierdo et
al., 2006). Sunﬂower oil does not contain linolenic fatty acids. Oleic acid content
in sunﬂower is cultivar dependent. Traditional sunﬂower has an oleic acid

content of 14 to 39%, mid-oleic acid sunﬂower of 42—72% oleic acid, marketed as

29

NuSun, Mycogen, Breckenridge, MN, and high-oleic acid sunﬂower containing
75—91% oleic acid (Hajimahmoodi et al., 2005).

Fatty acid levels are quite variable as levels are known to be affected by
night temperatures during the grain ﬁlling period, cultivar, environmental stress,
and fertility (Zheljazkov et al., 2008). Modeling performed by lzquierdo et al.
(2006) found the relationship between oleic acid and night time temperature was

linear up to the maximum level reached at 226° C. In growth chamber

experiments, oleic acid ranged from 39 to 52% among differing temperature

treatments. Increasing night temperature from 20° to 23° C resulted in a higher

oleic and lower linoleic acid concentration. Palmitic acid concentration was
highest at the 28° and 20° C temperature regime while stearic acid concentration
was not temperature dependent (lzquierdo et al., 2006). Research conducted by
Zheljazkov et al. (2008) found that increased N fertilization reduced oil
concentration 4% in sunﬂower. However, more seed was produced which
subsequently increased oil yield per acre. Oleic, the major fatty acid constituent
of sunﬂower oil, responded a positive 4.3% with additional N amendments. Oleic
acid response to N under marginal land conditions can be problematic as it is
closely tied to N fertility while temperatures do not appear to be detrimental.
Currently, sunﬂower oil is produced primarily for human consumption
purposes. Oleic acid and saturated fatty acids are desirable in frying, because
unlike polyunsaturated fatty acids, do not need to be hydrogenated before use.
Sunﬂower oils have high stability in frying applications and have a long shelf life,

which makes the oil valuable. Nearly 3 billion kg of sunﬂower per year is

30

marketed for use in frying oils (Zheljazkov et al., 2008). Additionally, sunﬂower
has protein levels of 26% and is the fourth highest source of protein supplement
behind soybean, cottonseed, and canola meals (Herridge, 2008).

Sunﬂower is able to produce 5.7 liters of biodiesel per 12.2 kg. Yield of
sunﬂower can produce 720 liters hectare (Thelen, personal communication). The
cake co-product can also be used as a component of cattle rations. Biodiesel
production from sunﬂower will have to compete with the export seed and crude
oil market which may place a higher value on the seed and oil versus its use as a
biodiesel feedstock. Additionally, costs associated with processing and reﬁning
signiﬁcantly increases the cost of the oil as it applies to the transesteriﬁcation
process as a biodiesel feedstock cost. Price increases may make biodiesel

derived from sunﬂower cost prohibitive (Nelson et al., 1994).

SWITCHGRASS

Switchgrass (Panicum virgatum L.) is a warm season perennial grass
native to the tallgrass prairies once found throughout the eastern two-thirds of
the United States before Europeans arrived (Hitchcock, 1935). It was widely
considered an obstacle to progress in agriculture and in the 19th century, the
famous “sodbusters” cleared the land of tall grass prairies. Switchgrass has a
relatively stiff stem, which the name suggests, and has been used by Native
Americans before European arrival. Switchgrass as a crop has primarily been
used as a forage and since 1980 has'it been considered an option for bioenergy

(Parrish and Fike, 2005).

31

As a C4 carbon ﬁxation plant, switchgrass can ﬁx carbon by multiple
metabolic pathways with high water use efﬁciency. In general C4 plants will
produce 30% more biomass per unit of water than other species. Grasses with
C4 metabolism are also better adapted to arid production areas where moisture is
limited (Moss et al., 1969).

Switchgrass is classiﬁed as a lignocellulosic crop. Its cell wall is high in
cellulose and lignin that can be used either by direct burning to produce
electricity or heat or digested to sugars, which can be fermented to produce
ethanol. As a perennial crop, switchgrass can survive less intensive
management and grow better on poorer quality, marginal lands than annual row
crops such as corn or soybean. The economic return of production to ethanol
from lignocellulosic sources surpasses that of annual crops in intensely managed
production systems (McLaughlin and Kszos, 2005).

The majority of land for perennial grass production is projected to come
from the reallocation of existing cropland and land currently enrolled in the
Conservation Reserve Program (CRP) (Schmer et al., 2007). Switchgrass has
been used extensively on acreages set aside in CRP to minimize erosion (Jewett
et al., 1996; Moser and Vogel, 1995). According to Walsh et al. (2003) nearly 17
million ha'1 of cropped, idle, pasture, and CRP lands could be converted to
switchgrass production for economic reasons including crop prices, costs of
inputs, and attainable yields resulting in an annual production level of 155 million
Mg. Switchgrass is most productive on moderately-well to well drained soils on

medium fertility and at a soil pH range of 5.5 to 7.0. However, seedlings are

32

quite tolerant of marginal and acid soils (Bona and Belesky, 1992) and
switchgrass roots have been reported to grow at pH 3.7 (Stucky et al., 1980).
Since switchgrass is a polymorphic species with two distinct ecotypes, lowland
and upland (anken and Estes, 1975) plants are able to withstand varying
conditions. Lowland types are found on ﬂoodplains and are able to survive short-
term ﬂooding while upland types occur in upland areas which may be droughty.

Switchgrass has a dense canopy and extensive root system which
reduces erosion from wind and water (Ichizen et al., 2005). The dense canopy
following establishment also suppresses weed populations (United States
Department of Agriculture, 2000). Variations in switchgrass growth and yield can
best be predicted by accumulated growing degree days (Mitchell et al., 2001).
As a result, switchgrass can produce a high yield of biomass across a wide range
of climatic scenarios and is suitable for growth on marginal, highly erodible
droughty soils (Casler et al., 2007).

Switchgrass has been shown to facilitate the breakdown or removal of soil
contaminants though bioremediation. Contaminants including herbicides
(Belden and Coats, 2004) and polychlorinated biphenyls (Dzantor et al., 2000)
are a few contaminants removed by the native grass. Other bioenergy
feedstocks such as poplar and willow have been used for bioremediation and are
environmentally acceptable sources of biomass for bioenergy (Vassilev et al.,
2004). Bioremediation properties may prove to be vital for switchgrass success

as a bioenergy crop On marginal lands.

33

Seeding rates of switchgrass are vital for stand success. Limitations to
switchgrass production include seed which has increased dormancy and is slow
to establish. Additionally, no herbicides or fungicides are currently labeled for
weed and disease control in switchgrass. In a study conducted by Mooney et al.
(2008), seeding rates varying from 1.13-5.67 kg did not make a signiﬁcant
difference on switchgrass yield on differing marginal soils in Tennessee. It has
been suggested, switchgrass cultivated on marginal land should be seeded at
rates of 4.5-6.7 kg ha'1 to account for seed dormancy.

Switchgrass grown as an energy crop has potential to reduce atmospheric
C02 accumulation by replacing fossil fuels and sequestering C (Bransby et al.,
1998). Studies performed by Gebhart et al. (1994) found land enrolled in CRP
and marginal land as part of a perennial grass system improves C sequestration
in comparison with annualrow crops and especially at deeper soil depths such
as 30-90 cm (Liebig et al, 2004). Ma et al. (2000) found over long term studies,
switchgrass stands can increase soil C levels 28-45%. Research conducted by
Lemus and Lal (2005) found C sequestration rates ranging from 0.6 to 3.0 Mg C
ha'1 yr'1 in degraded soils.

Switchgrass is a highly productive warm season perennial grass. Yield
potential is highly dependent upon latitude as well as the survivability of a
switchgrass stand (Casler et al., 2004). Yields Iof switchgrass are calculated on
dry matter content. Switchgrass moisture at harvest should be around 15% or
less to facilitate quick baling and transport. Yields lower than 4 Mg ha'1 per year

were common in the19703. As the crop has progressed into a bioenergy crop

as compared to a forage crop, yield has increased. Today, switchgrass yields
will frequently total 10 Mg ha'1 per year. Under intense management practices
and fertilizer amendments, yields have reached 20 Mg ha‘1 per year and higher
(Parrish and Fike 2005) in research plots however grower yields could be 35-
45% lower due to ﬁeld landscape inconsistencies, and windrowing and baling
tasks (Monti et al., 2009). Under management practices with no additional N or
manure amendments, yields have reached 9 Mg ha'1 (Lee et al., 2007). Varvel
et al. (2007) found under marginal conditions, switchgrass compared to corn
grain and stover yield, when converted to ethanol, showed no differences in fuel
yield using the same production practices and landbase.

Cutting system and timing also affects productivity of switchgrass as a
bioenergy crop. Multiple cutting systems have been suggested to increase yield
as is done in clover crops. Reynolds et al. (2000) found that a two cut system,
an early summer cutting and one following a killing frost, yielded more biomass
than a one cut post frost cutting. Conversely, Sanderson et al. (1999) found
reduced total seasonal yields as harvest frequency increased. Research
conducted by Lee et al. (2007) looked at alternating harvest years. Cutting
systems every year yielded more dry matter biomass than an alternate year
cutting system. Subsequently, due to seed rain and reduced stress on stand,
switchgrass stands were denser in the alternate year systems than compared to
the yearly cutting cycle. Casler and Boe (2003) found a mid-August harvest in
the north-central USA reduced stand density over time compared with a fall

harvest, yet in the Midwest maximum switchgrass yields occurred at this time

35

(Vogel et al., 2002). Delaying cutting from fall until spring to provide animal cover
during the winter decreased yield 40% (Adler et al., 2006). It is vital to leave six
inches of stubble following harvest to help trap snow and protect the root crowns
from winterkill (Westphal and lsebrands, 2001).

Switchgrass, a native grass covering many ecological niches, is not
susceptible to many diseases and pests. It has a wide adaptability to disease
and pest pressure which is likely due to high levels of genetic diversity
associated with open pollinated reproduction, a deep, strong root system, and
efﬁcient C4 physiological metabolism. Smut caused by a fungus Tilletia
maclaganii, has been identiﬁed as a threat to switchgrass yield. Smut is
prevalent in low yielding ﬁelds having characteristics of premature ﬂowering,
stunting, purple pigmentation of glumes, and replacement of seeds by fungal
spores (Gravert et al., 2000). Fields with smut infestations greater than 40%
yielded less than 30-40% of the expected biomass (Paige Marie Thomsen-
Johnson,2001)

In 2009, switchgrass is not converted commercially to ethanol. Many
reports have indicated fuel yield per acre, yet average yield results equate to 8
tons dry matter per acre yielding 273 liters per ton resulting in 5387 liters per
hectare (Thelen, personal communication). Cellulosic crops have the potential to
yield more fuel than grain based feedstocks, yet are more difﬁcult to process.

CONCERNS OF BIOENERGY CROPS
There are numerous potential bioenergy crops. A single, all purpose

bioenergy crop has not been identiﬁed. Determining selection for a given region

36

is vital for bioenergy production success. Of the previously identiﬁed crops,
some characteristics have the capability to be problematic. Canola has the
ability to either hybridize or introgress glyphosate resistant genes into unaltered
canola or weedy relatives through the same process as ﬂax (Linum usitatissimum
L.) (Jhala, 2008). Sorghum and sunﬂower also have the ability for some
transgenes to introgress into wild relatives shattercane (Sorghum bicolor),
Johnsongrass (Sorghum halepense) and common sunﬂower (Helianthus annuus)
(Stewart et al., 2003). Sorghum halepense at the time was deemed a beneﬁcial
introduced forage grass which inevitability has become an invasive weed in 16
US. states (McWhorter, 1993). Sterility has been bred into several species
however work with polyploidy species does not always guarantee sterility (Gray
et al.,1991). Additionally, proliﬁc seed production exhibited by species such as
switchgrass has the ability to create populations in unintended areas. Seed
inputs by Panicum species can range from 10,050 to 500,000 seeds (Vengris,
1975). To help plants conserve energy, most biomass is harvested after
senescence in the ﬁeld which can facilitate dispersal of seed to unintended
areas. Bioenergy feedstocks are being selected from taxa that produce highly
competitive stands that thrive with minimal human intervention (DiTomaso et al.,

2007) and research focusing on invasive characteristics is needed.

37

LITERATURE CITED

Adler, P., M. Sanderson, A. Boateng, P. Weimer, and H. Jung. 2006. Biomass
yield and biofuel quality of switchgrass harvested in fall or spring. Agron. J'.
98:1518-1525.

American Soybean Association Soy Stats. 2007. Soybeans...the miracle crop.
Available at http://www.soystats.corn/default.htm (veriﬁed 01 Dec. 2008).

Belden, J. and J. Coats. 2004. Effect of grasses on herbicide fate in the soil
column: Inﬁltration of runoff, movement, and degradation. Environ. Toxicol.
Chem. 23:2251—2258.

Belpoggi, F ., M. Soffritti, and C. Maltoni. 1995. Methyl-tertiary-butyl ether
(MTBE)-a gasoline additive-causes testicular and lymphohaematopoietic
cancers in rats. Toxicol Ind Health. 11(2):119-49.

Bona, L. and D. Belesky. 1992. Evaluation of switchgrass entries for acid
soil tolerance. Commun. Soil Sci. Plant Anal. 23:1827—1841.

Boyles, M., T., Peeper, C. Medlin. 2008. Winter canola planting guide for the
southern great plains. PSS-2131.

Bransby, D., S. McLaughlin, and D. Parrish. 1998. A review of carbon and
nitrogen balances in switchgrass grown for energy. Biomass and Bioenergy.
14:379-384.

Brookes, G. and P. Barfoot. 2006. GM crops: The ﬁrst ten years - global socio-
economic and environmental impacts. lntemational Service for the Acquisition of
Agri-Biotech Applications. No. 36. Available at
httpzllwww.isaaa.org/ResourceslPublicationslbriefs/36/download/isaaa-brief-36-
2006.pdf (veriﬁed 07 Dec. 2008).

Brunken, J. and J. Estes. 1975. Cytological and morphological variation in
Panicum virgatum L. Southwest Nat. 19:379-385.

Bullock, D., S. Khan, and A. Rayburn. 1998. Soybean yield response to narrow
rows is largely due to enhanced early growth. Crop Sci. 38:1011—1016.

Cairns, John. 2007. Corn for the starving or ethanol for fuel?: An ethical
dilemma for members of the automobile culture. Sci. & Soc. 5(2) 117-126.

California Foundation for Agriculture in the Classroom. 2008. Corn commodity

fact sheet. Available at http://www.cfaitc.org/Commodity/pdf/Com.pdf (veriﬁed 01
Dec. 2008).

38

Canakci, M., and JV. Gerpen. 2001. Biodiesel production from oils and fats with
high free fatty acids. Transactions. ASAE 44:1429-1436.

Canola Council of Canada. 2008. Canola facts: Trans fat and public health.
Available at http://www.canola-council.org/facts_trans.aspx (veriﬁed 07 Dec.
2008)

Carcarno and Blackshaw. 2007. Insect pest incidence and injury to herbicide-
tolerant canola in western Canada. Agronomy J.

Casler, M., and A. Boe. 2003. Cultivar 3 environment interactions in
switchgrass. Crop Sci. 43:2226—2233.

Casler, M., K. Vogel, C. Taliaferro, N. Ehlke, J. Berdahl, E. Brummer, R.
Kallenbach, C. West, and R. Mitchell. 2007. Latitudinal and longitudinal
adaptation of switchgrass populations. Crop Sci. 47:2249—2260.

Casler, M., K. Vogel, C. Taliaferro, and R. Wynia. 2004. Latitudinal
adaptation of switchgrass populations. Crop Sci. 44:293—303.

Ceotto, E. 2007. Grasslands for bioenergy production. A review. Agron.
Sustain. Dev. 28: 47—55.

Dardanelli, J., M. Balzarini, M. Martinez, M. Cuniberti, S. Resnik, S. Ramunda, R.
Herrero, and H. Baigorri. 2006. Soybean maturity groups, environments,

and their interaction deﬁne mega-environments for seed composition in
Argentina. Crop Sci 46:1939-1947.

Dhugga, K. S. 2007. Maize biomass yield and composition for biofuels. Crop
Sci. 47:2211-2227.

DiTomaso, J., J. Barney, and A. Fox. 2007. Biofuel feedstocks: The rise of
future invasions. CAST Commentary.

Dombos, D. and R. Mullen. 1992. Soybean seed protein and oil contents and
fatty acid composition adjustments by drought and temperature. Journal of the
American Oil Chemists' Society. 69: 3.

Dubois, V., S. Breton, M. Linder, J. Fanni, and M. Parmentier. 2007. Fatty acid
proﬁles of 80 vegetable oils with regard to their nutritional potential. European
Journal of Lipid Science and Technology 109:710—732.

Duvick, D. 2005. The contribution of breeding to yield advances in Zea maize
(L.). Advances in Agronomy. 86283-145.

39

Dzantor, E., T., Chekol, and L., Vough. 2000. Feasibility of using forage
grasses and legumes for phytoremediation of organic pollutants. J. Environ.
Sci. Health. Part A 35:1645—1661.

Edwards, J., L. Purcell, and E. Vories. 2005. Light interception and yield
potential of short-season maize (Zea mays L.) Hybrids in the Midsouth. Agron. J.
97:225-234.

Energy lnforrnation Administration. 2007. Petroleum basic statistics. Available
at http://tonto.eia.doe.gov/country/index.cfm (veriﬁed 07 Feb. 2009).

Environmental Protection Agency. 2008. Brownﬁelds and land revitalization.
Available at http:llwww.epa.gov/brownﬁelds/ (veriﬁed 10 June 2008).

Environmental Protection Agency, 2004. LEDC parcels of land at the south point
plant superfund site are ready for industrial reuse. Available at
http://www.epa.govlsuperfund/programs/recycle/pdf/southptrfr.pdf (veriﬁed 10
June 2008).

Epplin, F., C. Clark, R. Roberts , and S. Hwang. 2007. Challenges to the
development of a dedicated energy crop. American Journal of Agricultural
Economics. 89(5) 1296-1302.

Farrell, A., R. Plevin, B. Turner, A. Jones, M. O’Hare, and D. Kammen. 2006.
. Ethanol can contribute to energy and environmental goals. Science.
31 1(5760):506-508.
Food and Agriculture Organization of the United Nations. 1997. Photosynthesis
and biomass. Available at http://www.fao.org/docrep/w7241e/w7241e05.htm
#1 .2.1% 20photosynthetic%20efﬁciency (veriﬁed 01 Dec. 2008).
Foth, H.D. 1978. Fundamentals of soil science. Wiley, New York, NY.

Freedman, B., E. Pryde, and T. Mounts. 1984. Variables affecting the yields of
fatty esters from transesteriﬁed vegetable oils. JAOCS. 61 :1638-1643.

Gay, 8., D. Egli, D. Reicosky. 1980. Physiological aspects of yield improvement
in soybeans. Agron. J. 72:387-391.

Gebhart, D., H. Johnson, H. Mayeux, and H. Polley. 1994. The CRP increases
soil organic carbon. J. Soil Water Conserv. 49:488-492.

Gleeson, A. 2007. Phytoextraction of lead from contaminated soil by Panicum
virgatum L. (Switchgrass) and Associated Growth Responses.

40

Graham, R. R. Nelson, J. Sheehan, R. Perlack, and L. Wright. 2007. Current
and potential U.S. corn stover supplies. Agron. J. 99:1-11.

Gravert, C., L. Tiffany, and G. Munkvold. 2000. Disease outbreak of smut
caused by Tilletia maclaganii on cultivated switchgrass in Iowa. 84:5.

Gray A. J. et al. 1991. Ploidy number does not equal sterility. Adv. Ecol. Res.
21, 1.

Greene, R. and J. Stager. 2001. Rangeland to cropland conversions as
replacement land for prime farmland lost to urban development. The Social
Science Journal. 38:4 543-555.

Gunstone, F. 2004. Rapeseed and canola oil: Production, processing,
properties and uses. Published by CRC Press.

Hajimahmoodi, M., Y. Vander Heyden, N. Sadeghi, B. Jannat, M. Oveisi and S.
Shahbazian. 2005. Gas-chromatographic fatty-acid ﬁngerprints and partial least
squares modeling as a basis for the simultaneous determination of edible oil
mixtures. .Talanta. 66:5:1108-1 1 16.

Hammerschlag R. 2006. Ethanol’s return on investment: a survey of the
literature 1990-present. Environmental Science and Technology. 40(6):1744-
1750. ,

Hansen, L., H. Siegismund, and R. Jorgensen. 2001. lntrogression between
oilseed rape (Brassica napus L.) and its weedy relative B. rapa L. in a natural
population. Genetic Resources and Crop Evolution. 48: 621-627.

Harker, K., R. Blackshaw, K. Kirkland, D. Derksen, and D. Wall. 2000.
Herbicide-tolerant canola: Weed control and yield comparisons in western
Canada. Can. J. Plant Sci. 80:647-654.

Havlin, J. L., J.D. Beaton, S.L. Tisdale, and W.L. Nelson. 2005. Soil fertility and
fertilizers, an introduction to nutrient management. Pearson Education, Inc.

Heatherly, LG, and R.W. Elmore. 2004. Managing inputs for peak production.
p. 451-536. In HR. Boerma and J.E. Specht (ed.) Soybeans: Improvement,
production, and uses. 3rd ed. ASA, CSSA, SSSA, Madison, WI.

Herridge, D., M. Peoples and R. Boddey. 2008. Global inputs of biological
nitrogen ﬁxation in agricultural systems. Plant and Soil. 311(1-2).

Hesley, J. (Ed). 1994. Sunﬂower meal use in livestock rations. National
Sunﬂower Association, Bismarck, ND.

41

Hill, J., E. Nelson, D. Tilman, S. Polasky, and D. Tiffany. 2006. Environmental,
economic, and energetic costs and beneﬁts of biodiesel and ethanol biofuels.
Proceedings of the National Academy of Sciences, 103, 11206—11210.

Hitchcock, A. S. 1935. Manual of the grasses of the United States. United States
Department of Agriculture, Washington, DC.

Hoeft, R., E. Nafziger, R. Johnson, and S. Aldrich. 2000. Modern corn and
soybean production.

Hougen, F., and B. Stefansson. 1983. Rapeseed. Advances in cereal science
and technology. 52261-289.

Howell, R. and F. Collins. 1957. Factors affecting linolenic and linoleic acid
content of soybean oil. Agron J. 49:593-597.

Hirsch, R.L., R. Bezdek, R. Wendling. 2005. Peaking of world oil production:
impacts, mitigation, and risk management, report to US. Dept. of Energy - Natl.
Energy Technol. Lab. Available at http:llwww.netl.doe.govlpublications
/others/pdflOiI_Peaking_NETL.pdf (veriﬁed 17 Dec. 2008).

Hubbert, M. K. 1956. Nuclear energy and the fossil fuels: American petroleum
institute drilling and production practice. p. 7-25.

Hymowitz, T. and J. Harlan. 1983. Introduction of soybean to North America by
Samuel Bowen in 1765. Econ. Bot. 37:371-379.

lchizen, N., H Takahashi, T. Nishio, G. Liu, D. Li, and J, Huang. 2005. Impacts
of switchgrass (Panicum virgatum L.) planting on soil erosion in the hills of the
Loess Plateau in China. Weed Biology and Management. 5:1:31-34.

lzquierdo, N., L. Aguirreza' bal, F. Andrade, and M. Cantarero. 2006. Modeling
the response of fatty acid composition to temperature in a traditional sunﬂower
hybrid. Agron. J. 98:451—461.

Jewett, J., C. Sheaffer, R. Moon, N. Martin, D. Barnes, D. Breitbach, and N.
Jordan. 1996. A survey of CRP land in Minnesota: ll. Weeds on CRP land. J.
Prod. Agric. 9:535-542.

Jhala, A., M. Hall, and J. Hall. 2008. Potential hybridization of ﬂax with weedy
and wild relatives: An avenue for movement of engineered genes? Crop Science.
48:825-840.

Jiang, P., and K. Thelen. 2004. Effect of soil and topographic properties on crop
yield in a North-central com-soybean cropping system. Agron. J. 96:252-258.

42

Kaylen, M., D. Van Dyne, Y. Choi and M. Blasé. 2000. Economic feasibility of
producing ethanol from lignocellulosic feedstocks. Bioresource Technology.
72(1):19-32.

Karaaslan, D. 2008. Determination of total protein and fatty acid composition of
soybean seed as affected by sowing dates. Asian Journal of Chemistry. 20:
767-775.

Kavanagh, E. 2006. Letters. Science. 312.

Kemp, P., G. Williams. 1980. A physiological basis for niche separation
between Agropyron smithii (C3) and Bouteloua gracilis (C4). Ecology. 61 :846-
858.

Korbitz, W. 2000. World-wide trends in production and marketing of biodiesel.
Proc. Altener Seminar, University of Technology in Graz, Graz.

Kris-Etherton, P., W. Harris, and L. Appel. 2002. Fish consumption, ﬁsh
oil, omega-3 fatty acids, and cardiovascular disease. Circulation
1 06:2747—2757.

Lafortezza, R, G. Sanesi, B. Pace, R. C. Corry, and R. Brown. 2004. Planning
for the rehabilitation of brownﬁeld sites. a landscape ecological perspective.
Brownﬁeld Sites ll. WIT Press.

Lal, R. 2008. Soils and sustainable agriculture. A review. Agronomy for
sustainable development. 28: 57-64.

Land Policy Institute, Michigan State University. 2007. National land base pixel
count.

Lee, D., V. Owens, and J. Doolittle. 2007. Switchgrass and soil carbon
sequestration response to ammonium nitrate, manure, and harvest frequency on
conservation reserve program land. Agron. J. 99:462-468.

Lemus, R. and R. Lal. 2005. Bioenergy crops and carbon sequestration.
Critical Reviews in Plant Sciences. 2421-21.

Licht, L., E. Aitchison, W. Schnabel, M. English, and M. Kaemf. 2001. Landﬁll
capping with woodland ecosystems. Practice Periodical of Hazardous, Toxic and
Radioactive Waste Management. 5(4):175—184.

Licht, Louis A. and JG. lsebrands. 2004. Linking phytoremediated pollutant

removal to biomass economic opportunities. Biomass and Bioenergy 28: 203-
218.

43

Liebig, M., H. Johnson, J. Hanson, and A. Frank. 2005. Soil carbon under
switchgrass stands and cultivated cropland. Biomass and Bioenergy 28:347-
354.

Lofstedt, R. 1996. The use of biomass energy in a regional context: the case of
Vaxjo Energy, Sweden. Biomass and Bioenergy. 11:1 3342.

Lorenz, D. and D. Morris. 1995. How much energy does it take to make a gallon
of ethanol? Available at http://www.ilsr.org/carbo/ethanol/netethan.html (veriﬁed
12 Feb. 2009).

Luft, G. 2005. America’s oil dependence and its implications for US. Middle
East policy. Brieﬁng presentation before Senate Foreign Relations Subcommittee
on Near Eastern and South Asian Affairs, Washington, DC.

Lynd, L., J. Cushman, R. Nichols, and C. Wyman. 1991. Fuel ethanol from
cellulosic biomass. Science. 251 (4999):1318-1323.

Ma, F. and M. Hanna. 1999. Biodiesel production: a review. Bioresource
Technology 70:1-15.

Ma, 2., C. Wood, and D. Bransby. 2000. Soil management impacts on soil
carbon sequestration by switchgrass. Biomass and Bioenergy. 18:469-477.

Magness, J.R., G.M. Markle, and CC. Compton. 1971. Food and feed crops of
the United States. lnterregional Research Project IR-4, IR Bul. 1 (Bul. 828 New
Jersey Agr. Expt. Sta.).

McLaughlin, S., and L. Kszos. 2005. Plant Development of switchgrass
(Panicum virgatum) as a bioenergy feedstock in the United States. Biomass and
Bioenergy. 28:515-535.

McWhorter, C. G. 1993. Weed Sci. 41, 669.

Merrill, 8., D. Tanaka, J. Krupinsky, M. Liebig, and J. Hanson. 2007. Soil water
depletion and recharge under ten crop species and applications to the Principles
of dynamic cropping systems. Agron. J. 99:931—938.

Mitchell, M., J. Fritz, K. Moore, L. Moser, K. Vogel, D. Redfeam, and D. Wester.
2001. Predicting forage quality in switchgrass and big bluestem. Agron. Joumal.
93:1 18-124.

Monti, A., S. Fazio, G. Venturi. 2009. The discrepancy between 'plot and ﬁeld
yields: Harvest and storage losses of switchgrass. Biomass and Bioenergy 33:5
841-847.

Mooney, D., R. Roberts, B. English, D. Tyler, J. Larson. 2008. Switchgrass
production in marginal environments: A comparative economic analysis across
four west Tennessee landscapes. Agricultural and Applied Economics
Association 2008 Annual Meeting, July 27-29, 2008, Orlando, Florida, 6403.

Morrison, M., H. Voldeng, and E. Cober. 1999. Physiological changes from ﬁfty-
eight years of genetic improvement of short-season cultivars in Canada. Agron.
J. 91:685—689.

Moser, L. and K. Vogel. 1995. Switchgrass, big bluestem, and indiangrass. p.
409—420. In: R.F. Barnes, D.A. Miller, and C.J. Nelson (eds.), Forages. Vol 1, An
introduction to grassland agriculture. Iowa State Univ. Press, Ames.

Moss, D., E. Krenzer, and W. Brun. 1969. Carbon dioxide compensation points in
related plant species. Science 164:187—188.

Nafziger, E. 2004. Soybean production in the Midwestern USA: Technologies for
sustainable and stable yields. p. 523—530. In F. Moscardi et al. (ed.) Proc. Vll
World Soybean Conf. Embrapa, Londrina, Brazil.

Nass, L. L., P. A. A. Pereira, and D. Ellis. 2007. Biofuels in Brazil: An overview.
Crop Sci 47:2228-2237.

National Com Growers Association. 2007. Building markets for distillers grains.
Available at http: //ncga. com/building-markets-distillers-grains (veriﬁed 01 Dec.
2008).

Natural Resources Conservation Service. 2009. Land Capability Classes.
Available at http:/lwww.mn.nrcs.usda.gov/technical/nrilﬁndings/cropland_lcc.htm
(veriﬁed 07 Feb. 2009).

National Sunﬂower Association. 2008. Sunﬂower seed/kemel. Available at
http:/Iwww.sunﬂowemsa.comlseed/ (veriﬁed 07 Dec 2008).

National Energy Policy Development Group. 2001. National energy policy.
(ISBN 0-16-050814-2). Washington, DC: US. Government Printing Ofﬁce.

Nelson, R., S. HoWell, and J. Weber. 1994. Potential feedstock supply and
costs for biodiesel production. Presented at the Sixth National Bioenergy
Conference, Reno/Sparks, Nevada. October 2-6. Available at

http: l/www. biodiesel. org/resources/reportsdatabase/reports/gen/19941006_gen-
290. pdf (veriﬁed 13 Feb. 2008).

Nelson, K. and K. Renner. 2001. Soybean growth and development as affected

by glyphosate and postemergence herbicide tank mixtures. Agron. Joumal.
93:428-434.

45

Ohlsson, l. 1974. Row spacing in spring-sown oilseed crops. p. 212—215. in
Proc. Int. Rapskongress, 4th, Giessen, West Germany. 4—8 June 1974.

Oplinger, E., L. Hardman, E. Gritton, J. Doll, and K. Kelling. 1989. Alternative
ﬁeld crops manual, canola (rapeseed). Ext. Bull. Nov. 1989. Univ. of Wisconsin,
Madison, WI.

Organization of Petroleum Exporting Countries. 2007. Annual statistical bulletin.
Available at http:/lwww.opec.org/library/Annual%208tatistical%20Bulletin/
pdf/ASB2007.pdf (veriﬁed 23 April 2009).

Padgette, S., N. Taylor, D. Nida, M. Bailey, J. MacDonald, L. Holden, and R.
Fuchs. 1996. The composition of glyphosate-tolerant soybean seeds is
equivalent to that of conventional soybeans. Journal of Nutrition. 126:3 702-716.

Parrish, D. and J. Fike. 2005. The biology and agronomy of switchgrass for
biofuels , critical reviews. Plant Sciences. 24:5,423-459.

Phillips, D. 1980. Efﬁciency of symbiotic nitrogen ﬁxation in legumes. Annual
Review of Plant Physiology. 31: 29-49.

Pimental, D. and T. Patzek. 2005. Ethanol production using corn, switchgrass
and wood; Biodiesel Production Using Soybean and Sunﬂower. Natural
resources Research. 14(1):65-76.

Piper, E. and K. Boote. 1999. Temperature and cultivar effects on soybean
seed oil and protein concentrations. Journal of the American Oil Chemists'
Society. 76:10.

Primomo, V., D. Falk, G. Ablett, J. Tanner, and l. Rajcan. 2002. Genotype-
environment interactions, stability, and agronomic performance of soybean with
altered fatty acid proﬁles. Crop Sci. 42:37-44.

Probst, A. and R. Judd. 1973. Origin, US. history and development, and world
distribution. In Soybeans: improvement, production, and uses. Edited by B.E.
Caldwell. Amerimn Society of Agronomy Publ. No. 16. American Society of
Agronomy, Madison, Wis. pp. 1—15.

Puppan, D. 2002. Environmental evaluation of biofuels. Periodica Polytechnica
Ser. Soc. Man. Sci. 10:1 95—116.

Raghu, S., R. Anderson, C. Daehler, A. Davis, R. Wiedenmann, D. Simberloff,
and R. Mack. 2006. Adding biofuels to the invasive species ﬁre? Science
31321742.

Raven, P., R. Evert, and S. Eichhorn. 2005. Biology of Plants, 7th Edition. New
York: W.H. Freeman and Company Publishers, 124-127.

46

Reynolds, J., C. Walker and M. Kirchner. 2000. Nitrogen removal in switchgrass
biomass under two harvest systems. Biomass and Bioenergy. 19:281-286.
Rubel, A., R. Rinne and D. Canvin. 1972. Protein, oil, and fatty acid in
developing soybean seeds. Crop Sci. 12:739-741.

Sanderson, M., J. Read, and L. R. Reed. 1999. Harvest management of
switchgrass for biomass feedstock and forage production. Agron J 91 :5-10.

Sariangue, T., F. Andrade, P. Calviﬁo and L. Purcell. 2007. Why do hybrids
respond differently to variations in plant density? Agron. J. 99:984-991.

Schmer, M., K. Vogel, R. Mitchell, and R. Perrin. 2008. Net energy of cellulosic
ethanol from switchgrass. PNAS. 105(2)2464—469.

Schneiter, A. 1997. Sunﬂower technology and production. The American
Society of Agronomy No. 35 1-19.

Schneiter, A., and JP Miller. 1981. Description of sunﬂower growth stages. Crop
Sci. 21:901-903.

Schwab, A., G. Dykstra, E. Selke, S. Sorenson and E. Pryde. 1988. Diesel fuel
from thermal decomposition of soybean oil. Journal of the American Oil
Chemists' Society. 65:11.

Shahidi, F. 1990. Canola and rapeseed: production, chemistry, nutrition, and
processing technology. Springer.

Shapouri, H., J. Dufﬁeld, and M. Wang. 2002. The energy balance of corn
ethanol: an update: USDA, Ofﬁce of Energy Policy and New Uses, Agricultural
Economics. Rept. No. 813. 14 p.

Sheaffer, C. and K. Moncada. 2008. Introduction to agronomy, food, crops, and
environment CENGAGE Delmar Learning, Clifton Park, NY.

Smith, W. 1995. Crop production: evolution, history, and technology. Published
by John Wiley and Sons, Inc.

Solomon, B. 1980. Agricultural energy: debunking the net energy loss myth.
The Environmental Professional. 2(3/4):292-295.

Solomon B., J. Barnes, and K. Halvorsen. 2007. Grain and cellulosic ethanol:
History, economics, and energy policy. Biomass and Bioenergy. 31:416-425.

Specht, J., D. Hume, and S. Kumudini. 1999. Soybean yield potential—A
genetic and physiological perspective. Crop Sci. 39:1560—1570.

47

Specht J., and J. Williams. Chapter 3. Contribution of genetic technology to
soybean productivity — Retrospect and prospect. In: Fehr W.R., ed. Genetic
contribution to yield gains of ﬁve major crop plants. Madison, WI: CSSA and
ASA, 1984:49-74 CSSA Special Publ. 7.

Steer, B. and G. Seiler. 1990. Changes in fatty acid composition of sunﬂower
(Helianthus annuus L.) seeds in response to time of nitrogen application, supply
rates and defoliation. J. Sci. Food Agric. 51 :1 1—26.

Stewart, N., M. Halfhill and S. Warwick. 2003. Transgene introgression from
genetically modiﬁed crops to their wild relatives. Nature. 4:806-817.

Stewart, M. Halfhill, and S. Wanvick. 2003. Genetic modiﬁcation: Transgene
introgression from genetically modiﬁed crops to their wild relatives. Nature
Reviews Genetics. 42806-817.

Stone, L., D. Goodrum, A. Schlegel, M. Jaafar, and A. Khan. 2002. Water
depletion depth of grain sorghum and sunﬂower in the central high plains. Agron.
J. 94:936-943.

Stucky, D., J. Bauer, and T. Lindsey. 1980. Restoration of acidic mine
spoils with sewage sludge: l. Revegetation. Reclamation Review. 3:129—139.

Summers, C., D. Haviland, and L. Godfrey. 2008. UC IPM Pest Management
Guildlines: Sugarbeet. UC ANR Publication 3469.

Sutterﬁeld, L. 2007. United States oil dependency: An overview of the
desperate times that have imprisoned our foreign policy and the desperate
measures that may be required to liberate it. Available at
http:/lbakercenter.utk.edu/mainlﬁleslbakerJournalsNolume1/Sutterﬁeld.pdf
(veriﬁed 24 April 2009).

Thelen, K. 2007. Personal communication.
Thomsen-Johnson, P. 2001. The effects of head smut (Tilletia maclaganii) in
switchgrass ﬁelds used for biomass production in southern Iowa. Master’s

Thesis.

Tollenaar, M.and J. Wu. 1999. Yield Improvement in temperate maize is
attributable to greater stress tolerance. Crop Sci. 39:1597-1604.

Troyer, A. 2004. Popular and persistent corn gen'nplasm in 70 centuries of

evolution. In C.W. Smith (ed.) Corn: Origin, history, technology, and production.
John Wiley and Sons, New York (in press).

48

United States Census Bureau. 2008. Infant mortality rates and deaths, and life
expectancy at birth, by sex. lntemational Data Base. Available at
http:/Iwww.census.gov/cgi-bin/ipc/idbagg (veriﬁed 04 June 2008).

United States Census Bureau lntemational. 2008. Total midyear population for
the world21950-2050. lntemational Data Base. Available at
http:/lwww.census.gov/ipclwwwlidb/worldpop.html (veriﬁed 04 June 2008).

United States Department of Agriculture, Economic Research Service. 2003. Ag
Economic Report No. 823.

United States Department of Agriculture. 2008. Food, Conservation, and Energy
Act. Available at http:/lwww.fsa.usda.gov/lntemetlFSA_File/
2008focrpsummary.pdf (veriﬁed 10 June 2008).

United States Department of Agriculture, Farm Service Agency. 1999. The
Conservation Reserve Program, 18th Signup. Washington, DC.

United States Department of Agriculture, National Agriculture Statistical Service.
1999. Census of Agriculture — 1997. Washington, DC.

United States Department of Agriculture, National Agriculture Statistics Service.
2008. Corn yields 1908-2008. Available at http .l/.www nass. usda. gov/index. asp#
(veriﬁed 01 Nov. 2008).

United States Department of Agriculture, National Agriculture Statistics Service.
2002. United States census highlights. Available at http:/lwww.nass.usda.gov/
Census/Pull_Data_Census (veriﬁed 04 June 2008).

United States Department of Agriculture, Natural Resources Conservation
Service. 2000. Belleview-Perry sprayﬁeld plant materials adaption study.
Available at http:/lwww.plant-materials.nrcs.usda.gov/pubs/ﬂpmcpr7213.pdf
(veriﬁed 07 Dec. 2008).

United States Department of Agriculture, Natural Resources Conservation
Service. 2008. Conservation Reserve Program. Available at
http:/lwww.nrcs.usda.gov/programs/crp/ (veriﬁed 04 June 2008).

Varvel, G., K. Vogel, R. Miotchell, R. Follett, and J. Kimble. 2007. Comparison
of corn and switchgrass on marginal soils for bioenergy. Biomass and
Bioenergy.

Vassilev, A., J. Schwitzguebel, T. Thewys, D. van der Lelie, and J. Vangronsveld.

2004. The use of plants for remediation of metal-contaminated soils. The
Scientiﬁc World Journal. 429-34.

49

Vengris, J. 1975. Field growth habits of fall panicum. Proc Northeast Weed Sci.
Soc. 29:8.

Vogel, K., J. Brejda, D. Walters, and D. Buxton. 2002. Switchgrass biomass
production in the Midwest USA: Harvest and nitrogen management. Agron. J.
94:413-420.

Walsh, M., D. De La Torre Ugarte, H. Shapouri, and S. Slinsky. 2003.
Bioenergy crop production in the United States. Environmental and Resource
Economics 24: 313—333.

Wamcke, D. 2008. New MSU Nitrogen Recommendation for Field Corn.
Michigan State University. Available at httpzl/www.msue.msu.edu/objects
/content_revision/ download.cfm/item_id.4691 15/workspace_id.-30/New%20
MSU%20 reccomendations%20for%20Field%20Com%2008.pdf/ (veriﬁed 27
April 2009).

Wamcke, D., J. Dahl, L. Jacobs, and C. Laboski. 2004. Nutrient
Recommendation for ﬁeld crops in Michigan. Michigan State University
Extension Bulletin E-2904.

Westphal, L. M., and lsebrands, JG. 2001. Phytoremediation of Chicago’s
brownﬁelds: Consideration of ecological approaches and social issues. Paper
presented at the Brownﬁelds 2001 Conference, Chicago, IL.

Widdicombe, W. and K. Thelen. 2002. Row width and plant density effects on
corn grain production in the northern corn belt. Agronomy Journal. 9421020-
1023.

Wilcox, J.R., and J.F. Cavins. 1992. Normal and low linolenic acid soybean
strains - Response to planting date. Crop Science 32:1248-1251.

Wilhelm, W., J. Johnson, J. Hatﬁeld, W. Voorhees, and D. Linden. 2004. Crop
and soil productivity response to corn residue removal: A literature review. Agron.
J. 9621-17.

Witze, Alexandra. .2007. That’s oil folks... Nature Vol 445.

Wolf, R., J. Cavins, R. Kleiman, and L. Black. 1982. Effect of temperature on
soybean seed constituents: Oil, protein, moisture, fatty acids, amino acids and
sugars. Journal of the American Oil Chemists' Society. 59:5.

Zheljazkov, V., B. Vick, M. Ebelhar, N. Buehring, B. Baldwin, T. Astatkie, and J.

Miller. 2008. Yield, oil content, and composition of sunﬂower grown at
multiple locations In Mississippi. Agron. J. 100:635—642.

50

Zhu, X., 8. Long, D. Ort. 2008. "What is the maximum efﬁciency with which
photosynthesis can convert solar energy into biomass?” Curr. Opin. Biotechnol.
19:1 53.

51

Chapter 2. Use of Rggulatogy Brownﬁeld Sites for BioEnergy Production
ABSTRACT

Agriculture faces unique challenges as increasing world population places
unprecedented demands on food, feed, and energy resources. Agricultural
resources are becoming increasingly stretched to meet the demands of both food
and fuel production. Marginal land regulated for consumption purposes could be
utilized for bioenergy crop production. In 2006-2008, we evaluated corn (Zea
mays L. ssp. mays), soybean (Glycine Max (L.) Merr.), canola (Brassica napus L.
var. napus), sunﬂower (Helianthus annuus L.), and switchgrass (Panicum
virgatum L.) at two sites to determine the highest-yielding crop for bioenergy
production. Sites selected for the study included a remediated brownﬁeld site in
Rose Township (ROSE), Oakland County, 42° 43’N, 83° 37’W (ROSE) and a
historically cropped site at East Lansing (MSU), Ingham County, 42° 41 ’N, 89°
29’W, serving as the control. Comparisons of biofuel quality components were
also conducted. Yield of soybean, sunﬂower, canola, and switchgrass differed
between locations; however, the historically cropped site did not consistently
provide higher yields than ROSE. Ethanol yield from corn grain (leg) and the
total oil yield of oilseed crops was not signiﬁcantly different (p<0.05) between
locations. Concentrations of palmitic (C16:0), stearic (C1820), oleic (C18:1),
linoleic (C1822), and linolenic (C1823) fatty acids differed by location, yet oleic
acid concentrations, vital in biodiesel production, across all crops were not
signiﬁcantly different. Total crystalline cellulose levels from switchgrass were
higher at MSU compared to ROSE in 2007 yielding 16% more crystalline

cellulose; conversely sampling collection process were reﬁned in 2008 showing

52

no differences between locations. Pollutant scans results indicated detectable
PCB presence in soils from ROSE; however, transfer to grain or biomass was not
detected. Results suggest that brownﬁeld sites should be explored to increase

crop acreage and biomass availability for conversion to bio-based fuel sources.

INTRODUCTION

Energy production from agricultural crops in the United States has
become an ethical dilemma (Calms, 2007). Crops once devoted for food, feed,
and ﬁber production are now allotted to energy production, potentially reducing
the amount of commodities available for human consumption. If cultivation of
marginal acreage not currently under production was explored to increase
bioenergy crop production, the conﬂict of using consumption crops for energy
production would be alleviated (Walsh et al., 2003). One way to diminish the
competition is to produce bioenergy feedstock on land deemed unﬁt for the
production of food crops such as superfund and brownﬁeld sites (Environmental
Protection Agency, 2004). Brownﬁeld and superfund sites are industrial areas
where the soil and water have been contaminated from chemical or abandoned

hazardous waste.

Total brownﬁeld acreage in the United States has not been identiﬁed. In
2007'in Michigan, the Department of Environmental Quality identiﬁed 68
superfund sites with total acreage exceeding 2,025 hectares. An additional
3,645 sites have been identiﬁed as polluted with acreage unknown (Michigan
Department of Environmental Quality, 2007). In the entire US, there are over

500,000 brownﬁeld sites (Reddy et al., 1999). Soil and groundwater from

53

historical industrial dump sites can contain pollutants such as polychlorinated
biphenyls (PCBs), polyvinyl chloride (PVCs), heavy metals, petroleum products
and various solvents. These materials are persistent in soils and are not easily

removed.

Perennial grasses and native vegetation have been used to
phytoremediate brownﬁeld sites, however plant material generated from these
sites has not been sought for use (Belden and Coats, 2004; Dzantor et al., 2000;
Vassilev et al., 2004). In efforts to produce large quantities of grain and biomass
to supplement energy needs, agronomic principles will need to be reﬁned to
address sustainability of such cropping systems. Additionally, the quality of fuel
resulting from brownﬁeld acreage will also need to be evaluated to ensure

contaminants are not released during direct combustion.

1 Studies have been conducted to evaluate switchgrass cultural practices
under marginal conditions (Mooney et al., 2005) and to compare corn grain and
stover yield with switchgrass yield in marginal environments (Varvel et al., 2008).
However, there is little information available on multi-crop comparisons among
corn (Zea mays ssp. mays), soybean (Glycine Max (L.) Merr.), canola (Brassica
napus L. var. napus), sunﬂower (Helianthus annuus), and switchgrass (Panicum
virgatum L.) on marginal soils such as those on remediated brownﬁeld sites. It is
currently unknown whether crops grown on various types of brownﬁelds will yield

proﬁtable quantities of quality fuel.

In this study we seek to 1) evaluate bioenergy crop yield on a remediated
brownﬁeld site; 2) identify the optimum bioenergy crop for that speciﬁc
brownﬁeld; and 3) analyze fuel derived from brownﬁeld sites and make
comparisons to traditional cropping systems. A regulatory brownﬁeld located in
Rose Township, MI, that had been subjected to previous remediation under the

auspices of the USEPA superfund program was used as the study site.

MATERIALS AND METHODS

Site Description
In 2006, 2007, and 2008 ﬁeld experiments were conducted at a regulatory

brownﬁeld at Rose Township Dump Site (ROSE) in Oakland County, Michigan.
A second site on non-contaminated agricultural land was established at the
Michigan State University Agronomy Research Farm (MSU) in Ingham County to
serve as an experimental cheek for crop yield and quality analyses. Crops
evaluated include: com, soybean, canola, sunﬂower, and switchgrass. Site

characteristics can be found in Table 2.1.

Table 2.1. Site characteristics of location, elevation, soil type, land capability
class, pH, P and K levels.

 

 

Land
Site Location Elevation Soil Type Capability pH P K
m Class meq

MSU

42° 41'N 261 Capac Loam (fine-loamy, mixed, II 6.0 3.3 8.1

89' 29'W mesic, Aeric Ochraqualfs)
ROSE

42' 43'N 317 Marlette sandy loam (ﬁne- II 7.6 2.1 4.7

83' 37'W loamy, mixed, mesic Haplic

Glossudalfs)

 

55

Site history at MSU included a previous crop of soybean in 2005 while at
ROSE there was not a previous crop. ROSE was a fallow ﬁeld undergoing
succession and according to site history and soil characteristics, would be
deemed unﬁt for food crop production. The site consists of an upland area which
is almost completely surrounded by hydric soils. The pollutants found in soil and
groundwater at Rose township before remediation included solvents, paints,
polychlorinated biphenyls (PCBs), phthalates, lead battery sludges, chromium,
and volatile organic compounds (VOCs) such as toluene, benzene, and
trichloroethene, which breaks down into vinyl chloride (USEPA, 1987). Michigan
Department of Environmental Quality initiated removal and remediation action at
the site in 1980. During 1992, a mobile incinerator was constructed at the site
and was used to treat over 16,000 cubic meters (34,500 metric tons) of PCB-
bearing surface soils by incineration. In addition to incineration, an air stripping
groundwater pump and a soil vapor extraction (SVE) unit was also installed to
remove PCBs and VOCs, respectively. In 2006, an independent EPA capture
zone analysis, along with data from monitoring wells at the edge of the plume,
showed that pollutants were still persistent and complete capture was not

occurring (Fayoumi, 2008).

Contaminant Analysis

Soil and grain were extracted for contaminant analysis using accelerated
solvent extraction, ASE 200 (Dionex, Sunnyvale, CA). Ten grams of dry grain or
soil plus an equal amount of sodium sulfate (Sigma, St. Louis, M0) were placed

into an 11 ml extraction cell ﬁtted with a glass ﬁber ﬁlter, (PIN 047017, Dionex,

56

Sunnyvale, CA) and ﬁtted with a stainless steel frit at the outlet. Sodium sulfate
was not used in grain sample analysis. The dead volume of the cell was ﬁlled
with Ottawa sand ($23-3, Fisher Scientiﬁc, Pittsburg, PA). The extraction cell
was closed and automatically moved into a heatable extraction oven, ﬁlled with
solvent, and pressure was pressurized to 6900 kPa. Following the extraction
period, the extract was drained into a collection vial purged with compressed
nitrogen. The extraction temperature was 105°C, with a preheat period of 6 min,
and static extraction periods of 5 min with one cycle and a ﬂush volume of 60%
using 50% acetone and 50% hexane.

After extraction, the collection vials were weighed to determine extract
mass and recapped with undamaged septa. ASE samples were transferred to
gas chromatography (GC) vials and analyzed on an Agilent GC 6890 (Agilent,
Santa Clara, CA) equipped with a 30 m x 0.32 mm x 0.25 pm Supelco Equity 5
capillary column (Supelco, Bellefonte, PA) and HP 7673 autosampler (Agilent
Technologies, Santa Clara, CA). An electron capture detector (GC-ECD)
(Agilent Micro ECD, Santa Clara, CA) was used to detect electron-absorbing
compounds; parameters included: an increase in oven temperature from 120°C
to 300°C at 9°C per minute, injection volume of 1 pl with split ratio 4:1, inlet
temperature of 250°C, and a detector temperature of 310°C. Ultra high purity
(99.999%) helium was used as a carrier gas at ﬂow of 30 ml/min. Ten
representative peaks were chosen based on the fraction of total area and

uniqueness in the Aroclor 1248 standard as compared to the Aroclor 1254

57

standard (Supelco, Bellefonte, PA). The sum of these peaks was taken to

represent the total PCB content of the soil and grain samples.

Cultural and Management Practices

At the noncontaminated MSU site, plot preparation included spring chisel
plowing prior to planting. Management strategies for ROSE differed from MSU
as previous cropping systems required differing management. A glyphosate
application on 3 May 2006 followed by controlled burning and shrub removal on

24 May 2006.

Seeding dates, rates, and varieties are shown in Table 2.2. Varieties were

selected according to Michigan maturity zones.

Table 2.2. Planting date, seeding rate, and variety selection for MSU and ROSE.

 

 

Date Rate Variety
2007 2008 ha ‘1
Corn May 21 May 13 74,1311 Dekalb 38H66-RR2
Soybean May 21 May 13 444,785+ Diehl Fields DF8242RR
Canola May 21 April 30 6.721 Dekalb DKL 38-ZSRR
Sunflower May 21 May 13 74,131+ Seeds 2000 Charger
Switchgrass May 21" - 11.2" Forage First Forestburg_

 

*Reseeded switchgrass stand
1' Seeding rate in seeds ha'1
1 Seeding rate in kg ha'1

”Seeding rate in kg pure live seed ha’1

58

Planting of annual crops followed in 2007 and 2008 in a canola-com-
soybean-sunﬂower rotation with switchgrass remaining in dedicated established
plots. Switchgrass was originally planted in 2006 and no-till replanted in 2007

due to low populations.

In 2007, corn, soybean, and sunﬂower were planted with a custom built
planter utilizing John Deere row and vacuum units (Deere & Company, Moline,
IL) and canola and switchgrass were planted with an Almaco drill (Almaco,
Nevada, IA). In 2008, corn, soybean, and sunﬂower were planted with a custom
planter utilizing John Deere row units equipped with MaterMacc (MaterMacc, San
Vito al Tagliamento, Italy) metering units while canola was planted with an

Almaco drill.

Herbicide labels were followed for application and timing was crop

dependent as shown in table 2.3.

Table 2.3. Postemergence herbicide applications and timing.

 

 

Glyphosate Glufosinate Atrazine Quinclorac
Date (2007, 2008)
Corn May 23, May 15 - - -
Soybean May 23, May 15 - - -
Canola May 23, May 15 - - -
Sunflower - May 24, May 24 - -
Switﬂgrass - - May 12, May 10 *, May 10

 

* No application was made in 2007

Nitrogen soil amendments were side-dressed at rates of 112 kg ha'1 in

corn and sunﬂower. Other crops received no fertilization.

59

Harvest and Handling

.In 2006, crops were planted and harvested to establish the rotational
system, but yield totals were not recorded. Harvest dates for 2007 and 2008 are

presented in Table 2.4.

Table 2.4. Harvest dates at MSU and ROSE.

 

 

 

Crop Harvest Date
MSU ROSE
2007 2008 2007 2008
Corn Oct. 12 Oct. 29 Oct. 21 t
Soybean Oct. 11 Oct. 22 Oct. 11 Oct. 23
Sunflower Sept. 10 Sept. 17 Sept. 11 Sept. 20
Canola f July 23 Aug. 22 July 24

 

Switcﬂass Nov.1 Oct. 29 Oct. 30 Oct. 29

TCrops not harvested due to wildlife damage.

In 2007, corn and soybean grain yield was determined by whole-plot harvests
using a small-plot combine (Massey Ferguson 8XP, AGCO, Batavia, IL), and by
ﬂail harvester (Carter Manufacturing Co. Inc., Brookston, IN) for switchgrass.
Two sunﬂower and canola rows, 1 m in length with three replicates per plot, were
harvested by hand. In 2008, only subplot areas not affected by wildlife damage
were harvested. ln corn, 1.5 by 12.9 m was hand-harvested from two areas of
each plot. In soybean, 1.1 by 12.9 m was hand-harvested from three areas of
each plot. In canola, three areas 1.5 by 12.2 m were swathed (Swift Machine,
Swift Current, SK), then threshed (Almaco, Nevada, IA). In sunﬂower, three
areas 1.5 by 1 m were hand-harvested and sunﬂower heads were threshed

(Almaco, Nevada, IA). In switchgrass, three 50.8 by 50.8 cm quadrats were

60

hand-harvested at a cutting height of 7.6 cm from the soil surface. To eliminate
potential bias associated with wheel tracks and implement overlap, only the

center rows of each planter pass were harvested.

Grain moisture levels were determined using a DlCKEY-john GAC 2100
grain analysis computer (DlCKEY-john, Auburn, IL). Grain samples were oven-
dried at 60°C to 15.5, 13.0, 8.5, and 10.0% storage moisture for corn, soybean,
canola, and sunﬂower, respectively. Switchgrass samples were weighed in the
ﬁeld, oven dried at 100°C for 4 days, and weighed again to determine dry matter

content.

Corn Grain Ethanol

Triplicates of 25 9 ground corn grain samples were used for ethanol
analysis following a gravimetric, 72 h fermentation method described by Lemuz
I et al. (2004). Each 25 9 ground corn sample was mixed with 75 ml deionized
water and 27 pl alpha-amylase (9000-85-5, Sigma-Aldrich Inc., Allentown, PA) to
break the 01-6 linkage in the starch and form single glucose molecules in an
Erlenmeyer ﬂask ﬁtted with a rubber cap and needle vent. The slurry was
incubated in a 90°C oven for 90 min with occasional shaking. The slurry
temperature was then lowered to 60°C, and the pH was adjusted to 4.1-4.5 with
hydrochloric acid (EMD Chemicals Inc. Gibbstown, NJ). The liqueﬁed corn slurry
was sacchariﬁed and fermented simultaneously by adding 20 pl gluco-amylase
(9032-08-0, Sigma-Aldrich Inc.), 5 ml yeast inoculum, and 0.4 ml 1M NH4SO4
(EM Science, Gibbstown, NJ). The ﬂask was placed in a 60°C oven for 2 h

before it was placed in an incubator shaker at 32°C at 150 rpm for 72 h (Model

61

C25, New Brunswick Scientiﬁc, Edison, NJ). The yeast inoculum was prepared
according to Wang et al. (1999). Brieﬂy, active dry yeast (Fleischmann’s yeast,
Fenton, MO) was dispersed in 99 ml of 0.1% sterile peptone water (DOT
Scientiﬁc Inc. Burton, MI) and prewarrned at 38°C. The ﬂask with the slurry was
weighed together before and after the incubation, and the difference in weight
was considered as the mass of C02 released during the fermentation. Assuming
100% recovery, total ethanol produced was computed using a 1:1 molar ratio of

ethanol to C02 and the following conversion:

n = ((W1-W2)/44) x 46

where n is the total ethanol yield in grams, W1 in grams is the weight of the
solution before fermentation, W2 in grams is the weight of the solution following
fermentation, 44 is the molecular weight of C02, and 46 is the molecular weight

of ethanol.

Fatly Acid Methyl Esters

Fatty acid methyl ester (FAME) quantities in oil seed crops were
determined using a protocol adopted from Hammond (1991). Subsamples of
soybean, canola, and sunﬂower seed were collected at the time of harvest.
Whole grain samples of 1 g were placed in a 50-well, stainless steel plate and
crushed using a hydraulic press to break seed coats and physically extrude oil.
For oil extraction, 1 ml hexane was added to each sample of crushed seed. The
reaction was covered with a glass plate to ensure the hexane did not evaporate
and was left for 2 h to complete the reaction. A volume of 100 pl hexane and oil

solution was removed from the crushing tray, mixed with 500 pl of 1 N sodium

62

methoxide and transferred to a GC-FID autosampler vial. Vials were placed on a
Labnet Vortex Mixer (Labnet lntemational, lnc., Woodbridge, NJ) for 30-60 min
until oil droplets were no longer ﬂoating in the sodium methoxide solution. After
shaking, 1 ml of hexane and 150 pl of distilled water were added to each vial.
Following inversion of the vial, the solution was allowed to sit for 20 min to
complete the reaction. The addition of hexane and water results in the formation
of two distinct layers within the solution. The upper, lower density layer of the
solution containing FAME was removed and placed in another GC-F ID
autosampler vial while the lower layer was discarded. Samples were stored at
4°C until analysis.

Following the FAME reaction, samples were analyzed using GC-FID. A
6890 Agilent GC-F ID equipped with a 30 m x 0.53 mm x 0.5 pm Supelco wax 10
capillary column (SUPELCO lnc., Bellefonte, PA)and HP 7673 autosampler
(Agilent Technologies, Santa Clara, CA) were used to analyze FAME samples.
Sample vials were loaded into the autosampler tray and a sample sequence was
entered. GC-FID parameters included: an increase in oven temperature from
195°C to 240°C at 4°C/min and holding at 240°C for 2 min, injection volume of 1
pl with split ratio 20:1, inlet temperature of 220°C, and detector temperature of
290°C. Ultra high purity (99.999%) helium was used as carrier gas at a ﬂow rate
of 3.3 ml/min. Concentrations of the ﬁve major FAMEs were corrected with
external standard (Supelco, Bellefonte, PA) calibration curves (0.05, 0.1, 0.2, 0.5,

1, 2, 5 mg/ml). Quadruplicate analyses were conducted for all samples.

63

Total Oil

Total oil content was determined using a protocol developed by Matthaus
and Briihl (2001). An automated accelerated solvent extraction system ASE 200
(Dionex, Sunnyvale, CA) with 11 ml stainless steel sample cells was used for
total oil extraction. Moisture content of ground oilseed samples was measured
using a Moisture Analyzer A&D MF-50 (A&D Company Ltd, San Jose, CA).
Three 9 of sample was transferred to an extraction cell, ﬁtted with a cellulose
ﬁlter, and ﬁtted with a stainless steel frit at the outlet. The dead volume of the
cell was ﬁlled with diatomaceous earth (Dionex, Sunnyvale, CA). The extraction
cell was closed and automatically moved into the heatable extraction oven, ﬁlled
with solvent, and pressure was adjusted. Following the extraction period, the
extract was drained off into a collection vial which was purged with compressed
nitrogen to determine oil composition. The extraction temperature was 105°C,
with a preheat period of 6 minutes, and static extraction periods of 10 minutes
with 2 cycles and a ﬂush volume of 70% using hexane as a solvent were
selected. Pressure was held at 1000 psi throughout the extraction. After
processing, the extraction solvent hexane was evaporated by purging the
samples with oxygen free compressed nitrogen gas. The residual hexane was
evaporated in an oven at 105°C for approximately 60 min. The total oil content
was calculated using AOCS Ofﬁcial Method AM 2-93‘(AOCS, 2000):

%C = 100 x Qw/ (Wx (1- moisture%))

in which Qw in grams is the total oil extracted, Win grams is the weight of the

ground sample, and moisture% is the moisture percentage of the ground sample.

Total Cmstalline Sugar

Total crystalline cellulose content was calculated as pg/mg cell wall
material and analyses were performed in triplicate. Dried switchgrass samples
were ground to a ﬁne powder and pigments, proteins, lipids, and DNA were
removed by washing the samples with organic solvents (York at al., 1985). The
residue was de-starched with an amylase treatment resulting in only cell wall
material remaining. The composition of the wall matrix polysaccharides was
obtained by treating the samples with triﬂuoroacetic acid. Derivatization of the
solubilized monosaccharides into their corresponding alditol-acetates and
separation and quantiﬁcation was conducted using gas chromatography

employing a mass spectrometer (GC-MS) (Albersheim et al., 1967).

The remaining residue was washed with Updegraff reagent, an acid mix
that results in stripping of further hemicelluloses and amorphous glucan
(Updegraf, 1969). The remaining residue of crystalline cellulose was hydrolyzed
with sulfuric acid (Seaman hydrolysis) (Selvendran and O'Neill, 1987) and the
resulting monosaccharide quantiﬁed with a colorimetric assay (anthrone assay).
Sugars extracted include rhamnose, arabinose, mannose, galactose, glucose,
xylose, and arabinose. Levels of sugars are crop dependent and only whole
plant switchgrass samples were analyzed for cellulosic ethanol potential. The
matrix polysaccharide sugar composition gives indications about the presence
and abundance of wall polymers. In the grasses, arabinose and xylose represent
the hemicellulose arabinoxylan, mannose, the occurrence of glucomannan, and

glucose represents the presence of amorphous cellulose.

65

Statistical Anglvsis

The experiments at all locations were established in a randomized
complete block design (RCBD) with six replications at ROSE and four at MSU.
Crop yield data analysis was conducted separately for each crop by year due to
differing environmental conditions in 2007 and 2008. Gravimetric ethanol, total
oil, FAME, and cellulose component yields were determined using a two year
average to determine signiﬁcance. Analysis of variance (ANOVA) was
performed on crop yield and fuel quality and quantity using PROC MIXED in SAS
(SAS Institute, 2009). The location treatments are considered ﬁxed effects while
blocks are considered a random effect. Results are based on the statistical
model Yij = u + Location: + Block, + e. Normality of residuals was assessed using
PROC UNIVARIATE and Levene’s test was used to test homogeneity of
variances. When signiﬁcant effects of treatment occurred, means were
compared using Fisher’s Protected Least Signiﬁcant Difference (LSD). Unless

othenrvise stated, differences are considered signiﬁcant at an alpha level of 0.05.

Economic Analysis

Enterprise budgets and machine and work rates estimates were based on
Stein (2009). Seed prices were based on actual prices of purchased seed in
2007 and 2008. Fertilizer costs and herbicide prices were determined by
Jorgensen Farms Elevator (Williamston, MI). Commodity prices were
determined by National Agricultural Statistics Service for October 2008 (NAAS,

USDA, 2008).

66

RESULTS AND DISCUSSION

Contaminant Analysis
Analysis of soil samples taken from ROSE in 2007 indicated PCB

presence above a GC-ECD detection level of 0.28 ppm in each plot of block 4,
regardless of bioenergy crop (Figure 2.1). A detection limit level of 0.28 ppm is a
distinguishable PCB level discounting false-positives and discrepancies in GC

instrumentation.

 

9 13°19

 

 

 

 

 

 

 

 

 

 

SWIG

 

 

 

7W9

 

 

S ’POIO

 

 

INS

 

 

 

 

 

 

I ’Polﬂ

 

 

 

 

 

 

 

Undetectable PCB levels
Detectable PCB levels

I ’2. ”'0‘ .
l P '
.7 A ‘I

'mtlti‘Axx-J Jo.-

   

Figure 2.1. PCB soil sampling results from ROSE during 2007 (n=90).

67

PCB presence in block 4 may have been caused by improper capping
procedures.

Pollutant analyses were also conducted on crop grain samples (n=24),
with the exception of switchgrass, to determine potential plant uptake and
translocation. No substance peaks or detection limits were identiﬁable on GC-
ECD chromatograms including those for grain from block 4. Therefore,
combustion of bio-based fuels produced from agronomic feedstock at ROSE
would not adversely affect the environment.

Environmentjal Conditions

In 2007, precipitation at MSU was below average during the month of July
(Table 2.5), and rain received late in the season slowed crop drying. In 2008,
precipitation received during seedling establishment was below the 30—year

average and rain received late in the season again slowed crop drying.

Table 2.5. Total monthly precipitation (cm) at MSU (East Lansing, MI) and
ROSE (Rose Township, MI) in 2007-2008. Long term-means (1970-2000)
included for comparison.

 

Location Year Mar. Apr. May June July Aug. Sept. Oct.
MSU 2007 5.6 6.6 9.7 8.9 1.2 14.0 5.3 10.3
2008 4.0 4.4 3.0 11.3 7.6 1.7 18.6 4.1
30 yr. 5.7 7.5 7.5 . 9.3 7.9 7.8 6.3 5.5
ROSE 2007 8.0 5.7 6.3 5.3 3.5 16.6 3.5 6.5
2008 5.3 2.6 5.5 13.1 6.2 3.2 20.0 3.4
30 yr. 5.3 7.2 7.0 8.9 7.0 7.5 6.2 6.1

 

 

 

Precipitation at ROSE during the grain ﬁlling period of 2007 was below

normal and rainfall during cn'tical crop dry-down periods was above normal.

68

Droughty periods occurred during cn'tical grain ﬁlling periods at both locations in
2007, negatively affecting yield. In 2008, precipitation was below the 30-year
average during critical crop establishment periods, as well as in August,
adversely affecting yield. Growing season precipitation was consistent with the
30-year average; however, the timing of rainfall events was irregular. Water
availability, especially under marginal conditions, will be the primary
management factor to determine productivity of bioenergy cropping systems
(Stone et al., 2002; Jiang and Thelen, 2004). Crops with inherently high water

use efﬁciency will achieve higher crop yields.

Due to the relative proximity of the experimental plots, average monthly air
temperatures were similar. In 2007, average temperatures at MSU and ROSE

tended to be slightly above 30-year means (Table 2.6).

Table 2.6. Average monthly air temperatures (°C) at MSU (East Lansing, MI)
and ROSE (Rose Township, MI) in 2007-2008. Long-term means (1970-2000)
included for comparison.

 

Location Year Mar. Apr. May June July Aug. Sept. Oct.
MSU 2007 3.7 7.1 15.6 19.8 20.6 21.4 17.6 7.1
2008 -0.2 9.9 12.6 19.6 21.5 20.7 17.4 9.7

30 yr. 0.6 7.9 13.8 19.4 21.4 20.7 16.5 11.0

ROSE 2007 3.3 7.0 15.2 20.3 20.9 21.7 17.9 13.8
2008 -0.4 9.7 12.4 19.9 21.3 20.4 17.3 8.9

30 yr. 1.1 8.3 14.7 20.1 22.3 21.3 17.3 11.6

 

 

 

In 2008, temperatures at MSU were near the 30-year average while ROSE
experienced a colder than normal year. However, 2008 data were almost

identical at the two sites (Table 2.6).

69

Crop maturity can be predicted using daily average high temperatures and

low temperatures to calculate Growing Degree Day units (GDD). GDD are

calculated as the average of daily high and low temperatures less a base

temperature of 10°C. In 2007, temperatures were warmer (Table 2.6) and more

GDD were accumulated (Table 2.7) than in 2008, at both sites.

Table 2.7. Monthly and yearly accumulation of Growing Degree Day units at
MSU (East Lansing, MI) and ROSE (Rose Township, MI) during 2007 and 2008,
calculated using a base temperature of 10°C.

 

 

 

Mar. Apr. May June July Aug. Sept. Oct. Total GDD

MSU 2007 26.4 38.4 176.4 295.6 326.4 354.1 230.2 140.4 1587.9
2008 0.1 68.9 85.8 286.9 356.8 330.0 222.8 53.2 1404.6

ROSE 2007 21.3 37.7 163.2 309.9 337.1 362.9 238.9 142.3 1613.3
2008 0.0 64.4 78.4 296.4 352.2 321.3 217.9 38.0 1368.7

 

In 2007, ROSE accumulated 1613 GDD while MSU had 1588 GDD, a 2%
difference. Conversely, in 2008, MSU accumulated more GDD than ROSE, a
3% difference. Summed GDD over both years indicated similar conditions at
MSU and ROSE, with less than 1% difference between the two locations.
Locations which receive more GDD have the potential to achieve higher yield
totals (Swanson and Wilhelm, 1996). Given that GDD accumulations were
similar at the two locations, temperature had little effect on yield differences.
Agronomic Yield

Differences between sites included climate, soil type, shading, crop
demands, and logistics requiring differing harvest dates (T able 2.4). Canola, on

average, contains 40% oil, making the seed a very attractive food and energy

70

source for birds, in particular the American Goldﬁnch (Carduelis tn'stis L.). In
2008, swathing canola before it reached physiological maturity resulted in 5%
higher yields (8. Smith and K. Thelen, unpublished data). Canola was swathed
when 60-75% of the seeds were black and contained 30-40% moisture as
recommended by Oklahoma State University Extension (2007). To protect
sunﬂower crop yield in both years, plots were hand harvested and dried when
they had reached fully physiological maturity (R9) but had not yet dried to
acceptable mechanical harvest moisture percentages or storage moistures of

9.5%.

Statistical analysis determined that blocking within locations was not a
signiﬁcant factor (p > 0.05). Therefore, crop yield totals from all plots per crop
were averaged to determine a yield mean for each location. MSU and ROSE

yield means differed (p < 0.05) in both years (Table 2.8).

71

Table 2.8. Mean crop yields of bioenergy crops in 2007 and 2008 at MSU (n=4)
and ROSE (n=6). Means within a year and crop followed by a common letter are
not different at (p < 0.05) using Fisher’s Protected Least Signiﬁcant Difference
test.

 

 

 

 

Crop Yield (Mg ha'l)
2007 2008

MSU ROSE MSU ROSE
Corn 4.87 a 3.91 a 9.89 1
Soybean 1.83 a 2.40 a 2.10 b 2.56 a
Sunﬂower 2.54 a 1.16 b 2.70 a ‘4 2.66 a
Canola 1‘ ‘l 0.04 b 0.41 a
Switchgrass 0.61 b 4.14 a 7.39 a 7.46 a

 

TYield data not computed due to wildlife damage; samples retained for quality analysis.

tYield data not computed due to wildlife damage.

In 2007, grain yield of corn and soybean did not differ between locations.
Sunﬂower yields at ROSE were lower at 1.16 compared to 2.54 Mg ha’1 at MSU.
Conversely, biomass yield of switchgrass was higher at 4.14 Mg ha'1 at ROSE
compared to 0.61 Mg ha'1 at MSU. Yield comparisons of canola could not be
made due to wildlife damage. In 2008, canola and soybean yields at ROSE were
0.41 and 2.56, respectively, compared to 0.04 and 2.10 Mg ha" at MSU, while
sunﬂower and switchgrass yields were not statistically different. Comparisons of
2008 corn yield could not be made as plots were not harvested at ROSE due to

wildlife damage.

Crop establishment as well as favorable environmental conditions, are
vital for optimum crop yield. In 2008, crop yields at both locations were higher

than 2007 as crop management practices were reﬁned the second year of the

72

study. In particular, switchgrass biomass yields were low in 2007 partly due to
whole-plot harvests while quadrants harvests were used in 2008. Additionally,
switchgrass physiology dictates higher yield as a switchgrass stand matures as
2008 switchgrass yields were consistent with that of other marginal locations as
found by Schmer et al. (2008). Weed species including spotted knapweed
(Centaurea maculosa Lam.) and common lambsquarters (Chenopodr'um album
L.) slowed early crop growth and competed with crops as determined by low
sunﬂower yields. All together, these factors were likely more inﬂuential in
determining yields than the differences in soil fertility levels between the two

sites.

Corn Grain Ethanol Yield

Gravimetric fermentation of corn grain is standard industry practice for
ethanol production. In 2007, there was no difference in ethanol yield from MSU
and ROSE following fermentation (Figure 2.2). Ethanol yields at MSU and ROSE
of 0.4530 and 0.4532 L ethanol / kg corn were consistent with industry average

yields of 0.4173 ng (Bothast and Schlicher, 2004).

73

 

Corn Grain Ethanol Yield

 

0.6000

 

0.5000

 

0.4000 —_:——‘-:_' '1

 

. MSU
I ROSE

0.3000 —

L/ kg

 

0.2000 <————-'

 

0.1000 ~——-— 1 _

 

 

 

 

0.0000
2007 2008

 

 

 

Figure 2.2. Corn grain ethanol yields for 2007 (MSU and ROSE) and 2008
(MSU); ' represents no statistical signiﬁcance at p < 0.05 using Fisher’s
Protected Least Signiﬁcant Difference test.

Comparable ethanol yields were likely due to equal latitude and similar GDD of
the two sites. In 2008, corn grain from MSU yielded 0.4595 L/kg, which was only
a 1.4% ng increase from 2007.

Factors that inﬂuence the ethanol yield potential of corn grain are difﬁcult
to use as yield predictors because they are not well understood. Dein et al.
(2002) found that the starch content of corn did not correlate well with ethanol
yields. They suggested that ethanol yield may be affected by climatic conditions
and locations. The present research indicates that, similar to grain yield, ethanol
yield was not affected by the relatively low soil fertility at the ROSE brownﬁeld

site.

74

Faﬁy Acid Methvl Efsters Yielg

Biodiesel, which consists of fatty acid methyl esters (FAME), is derived
from triglycerides through a process termed transesteriﬁcation. Following
triglyceride extraction, ﬁve FAME compounds are of particular interest for
biodiesel production, including palmitic C16:0, stearic C18:0, oleic C18:1, linoleic
C182, and linolenic C18:3 fatty acids. FAME proﬁles will vary among crops and
growing conditions. FAME levels were determined using gas chromatography
with a ﬂame ionization detector (CG-F ID), with combined two-year results

indicating minimal signiﬁcant differences between MSU and ROSE (Table 2.9).

Table 2.9. Results of fatty acid methyl ester (FAME) analysis of oilseed crops at
MSU and ROSE, 2007 and 2008 dates combined (mean percent of total oil
content by weight); % FAME followed by a differing letter within a fatty acid are
signiﬁcant at p < 0.05 using Fisher’s Protected Least Signiﬁcant Difference test.

% fatty acid methyl esters/ml crop oil

 

 

 

n Palmitic Stearic Oleic Linoleic Linolenic

Soybean

MSU 32 11.68 a 5.33 a 24.63 50.92 b 7.44

ROSE 48 10.29 b 4.89 b 24.96 51.61 a 8.25
Canola

MSU 32 5.11 a 2.27 a 60.65 22.60 a 9.37

ROSE 48 4.78 a 1.96 b 60.21 23.31 a 9.74
Sunﬂower

MSU 32 5.18 a 4.74 a 77.36 12.63 a

ROSE 48 5.02 a 5.37 a 78.04 11.48 a

 

1' Sunﬂower oil does not contain linolenic FAME.

75

In soybean, the palmitic, stearic, linoleic, and linolenic FAME percentages
differed signiﬁcantly between MSU and ROSE and data were consistent with
those determined by Goering et al. (1982). Palmitic acid is the most common
saturated fatty acid found in plants (Goering et al.,1982; Kincs, 1985). Because
they are highly saturated, palmitic and stearic acids have high viscosity, causing
gelling and coagulation problems in diesel engines (Pryde, 1983; Ma and Hanna,
1999); therefore, low levels of palmitic and stearic acid are desirable. Linoleic
and linolenic FAME are unsaturated omega-6 fatty acids. Biodiesel with high
levels of linoleic and linolenic fatty acids, due to their unsaturation, has reduced
storage duration. Given these factors, optimum biodiesel is obtained by
optimizing oleic fatty acid levels. In soybean, canola, and sunﬂower, oleic acid
levels were not signiﬁcantly different between MSU and ROSE, suggesting
similar fuel quality. Stearic fatty acid levels in canola were lower at ROSE than at
MSU. In sunﬂower, no differences in levels of all FAME tested were found (Table

2.9).

lzquierdo et al. (2002) found that the oleic: linoleic ratio depends on the
temperature during the early stages of oil synthesis in oilseeds. Due to close
proximity and latitude of the two research locations, no temperature differences
(Table 2.6) were experienced, resulting in no differences in FAME levels (Table
2.9). Thus, brownﬁeld acreage can be expected to provide adequate oilseed
quality and can be used to increase the oilseed supply available for conversion to

biodiesel fuels.

76

Total Oil Yield

Two-year average total oil content for the oilseed crops did not differ

between MSU (n=8) and ROSE (n=12) (Figure 2.3). Total oil yield did not appear

to be affected by soil fertility levels.

 

Percent Oll Content

Soybean

 

Oil Quantities

 

Canola

Sunﬂower

 

 

Figure 2.3. Mean total oil content (% by weight) of soybean, canola, and
sunﬂower of oilseed crops at MSU and ROSE, 2007 and 2008 dates combined;
* represents no signiﬁcance at p < 0.05 using Fisher’s Protected Least Signiﬁcant

Difference test.

The range of soybean total oil content was 21.4 to 21.5% of seed dry mass,

which is consistent with data reported by Dardanelli et al. (2006).

Sunﬂower seed accumulates oil in the ﬁnal stage of reproductive growth,

. and oil content can differ with environmental conditions including soil type,

shading, agronomic practices, and precipitation (Zheljazkov et al. 2008). Seed

oil concentration is important for growers and processors. For example, sunﬂower

77

crushing plants in North Dakota and Kansas offer a 2% price premium for every
1% oil above 40% (National Sunﬂower Association 2007). No location
differences were observed with sunﬂower at MSU yielding 57.6 and ROSE
yielding 54.3% oil. However, these percentages may be slight overestimates
because a small fraction of seed coats could not be removed from the samples
and was included in the oil extraction process.

Jackson (2000) found that seed oil content of canola was reduced by
increasing applications of N fertilizer. At both MSU and ROSE, canola was
treated with the same, low-input fertilizer regime with no additional N
amendments. Canola total oil content can also be reduced by low levels of
precipitation during critical oilseed ﬁlling and oil accumulation periods (Chen et al.
2005). In the absence of N supplementation and water stress, canola oil content
at both MSU (36.7%) and ROSE (40.9%) fell within the range reported by
Sosulskl et al. (1988). Again, the brownﬁeld site produced oilseed crops
including canola, soybean, and sunﬂower of equal oil content as the conventional

agronomic site (Figure 2.3).

Total Crystalline Cellulose Yield

In 2007, average crystalline cellulose content of switchgrass was higher at
MSU than at ROSE (Figure 2.4). In 2008, no difference between locations of

switchgrass crystalline cellulose content was detected.

78

 

Crystalline Cellulose

400 , a

 

ab

 

MSU
IROSE

PSI/"l9

 

 

2007 2008

 

 

 

Figure 2.4. Average crystalline cellulose content of switchgrass at MSU and
ROSE, in pg cellulose/mg switchgrass. Location means followed by a differing
letters are signiﬁcant at p < 0.05 using Fisher’s Protected Least Signiﬁcant
Difference test.

Switchgrass grown at MSU yielded higher levels of crystalline cellulose as
compared to ROSE in 2007 but not 2008. Stand maturity may inﬂuence
increased cellulose content (Gomide et al., 1969). MSU switchgrass yielded
12% less cellulose in 2007 than in 2008. In 2008, stands yielded equal biomass
and crystalline cellulose levels at MSU and ROSE. In 2007, second-year stands
at MSU and ROSE produced signiﬁcantly lower crystalline cellulose content as
compared with three-year stands in 2008. ROSE produced signiﬁcantly less
cellulose than MSU in 2007. Levels of crystalline cellulose of weed species,
though not quantiﬁed, could have decreased levels at ROSE.

Switchgrass is a photoperiod-sensitive plant. Switchgrass cellulose
concentration and composition can be inﬂuenced by many environmental factors

(Moser and Vogel 1995). Temperature is the environmental factor that most

79

strongly affects plant cell walls, with high temperatures causing increased rate of
reproductive maturation. Temperature and water stress can affect morphological
development by resulting in higher cell wall concentrations, with increases
occurring in both polysaccharide and lignin fractions reducing total sugar (Buxton
and Fales, 1994). Temperature and GDD accumulation were similar at both sites
and was not likely a factor in total cellulose content.

Patton and Gieseker (1942) reported differences in cellulose and lignin
content depending upon maturity as well as grass species. Switchgrass grown in
2007 at MSU contained 29% more crystalline cellulose than that grown at ROSE,
while 2008 results were not signiﬁcant. Differences between locations can, in
part, be attributed to percent stand. Pure switchgrass samples could not be
collected from ROSE in 2007 due to inconsistent stands. Over two years and
two different sampling practices, similarities in total cellulose at ROSE were
present when pure switchgrass samples were collected due to similar fertility

base stand issues.

Economic Analysis

Producers planning to implement bioenergy cropping rotations in their
farming operations will make decisions based ultimately on optimizing economic
returns. Under marginal land conditions, positive economic outcomes will be

possible if the input costs are limited.

80

 

Table 2.10. Economic return utilizing marginal land (ROSE) compared to
traditional cropping system (MSU) acreage.

 

 

 

Crop Economic Outcomes
2007 2008
MSU ROSE MSU ROSE
Corn - + + X
Soybean - + + +
Sunﬂower + + + +
Canola X - X -
Switchgrass - - + +

 

 

- Represents an economic net loss
+ Represents an economic net gain
X Indicates the crop was not harvested

In 2007, estimated economic returns were positive only for sunﬂower at
MSU while corn, soybean, and sunﬂower returned economic gains at ROSE.
However at MSU, returns from corn, soybean, and sunﬂower in 2008 all provided
a ﬁnancial advantage. In both years and sites, canola yields were reduced by
wildlife damage. Although, switchgrass does not currently have market value as
a bioenergy crop, however $41.50 Mg‘1 (dry weight basis) price projections were
adapted from Sokhansanj et al. (2009). Production and economic analysis of a
perennial crop lends to different inputs depending on maturity of the stand.
Results of switchgrass economics indicate a proﬁt following the establishment
year. Agronomic and management practices need to be reﬁned for all bioenergy
crops to be proﬁtable under marginal conditions. Under ideal ﬁeld conditions, all
crops have the potential to be proﬁtable. More years of research are needed to

determine modeling for economic analysis.

81

Determining the optimum bioenergy crop to be grown for proﬁt on
remediated brownﬁeld sites cannot be easily determined. Currently, cellulosic
ethanol is not commercially produced on a large scale in the US. McLuughlin
and Kszos (2005) have reported that the economic return of ethanol production
from lignocellulosic sources surpasses that of annual crops. Under current
industry practices, and given environmental conditions at ROSE during 2007 and
2008, the greatest economic return was obtained from corn and sunﬂower. Once
cellulosic production technologies are reﬁned, switchgrass production on

marginal sites may prove to return the highest economic gains.
CONCLUSIONS

The ROSE brownﬁeld site provided bioenergy crop yield and quality
similar to that from noncontaminated agricultural land, suggesting the potential
for similar brownﬁeld sites across the Midwest to also be viable options for
bioenergy production. Soil analysis indicated PCB levels in block 4 at ROSE;
however, no traces of contaminant were found in the feedstock. Fuel quantity
and quality including ethanol, total oil, crystalline cellulose, and FAME produced
from marginal brownﬁeld sites showed minimal differences when compared to
areas with traditionally high fertility. Preliminary economic analysis supports
potential investment in economic endeavors on brownﬁeld acreage. Perennial
cellulosic crops such as switchgrass hold great promise for production on
brownﬁeld sites where water availability and soil fertility will limit yield of annual
crops. Increasing available feedstock for energy production on alternative

production sites eases the stress on market demand for food crops. However,

82

brownﬁeld sites should be approached on a case-by-case basis as some
pollutants may still prove to be a detriment to the environment following fuel
conversion and combustion. Sustainability of long-term cropping systems on
brownﬁeld sites is vital and must be further explored; for example, soil fertility
degradation of marginal lands under long-term crop production needs further

research in efforts to model fertility trends.

83

All

o—clml

[=1 [—0

APPENDICES

Appendix 2.1. Economic analysis of MSU in 2007 utilizing actual yields.

 

 

 

3 ha '1
MSU 2007 Canola Corn Soybean Sunﬂower Switchgrass
TOTAL REVENUE 169 968 581 4623 28
CASH EXPENSES
Seed 71 287 83 37 92
Fertilizer 19 70 19 70 19
Herbicides 43 30 30 56 6
Energy 316 386 319 386 782
Machinery and Repairs 217 219 178 213 536
Labor 25 25 25 25 25
TOTAL CASH EXPENSES 691 1016 654 787 1459
REVENUE ABOVE CASH EXPENSES -522 -48 -73 3836 -1431

 

Appendix 2.2. Economic analysis of ROSE in 2007 utilizing actual yields.

 

 

 

5 ha '1
ROSE 2007 Canola Corn Soybean Sunflower Switchgrass
TOTAL REVENUE 64 1179 866 2118 110
CASH EXPENSES '
Seed 71 287 83 37 92
Fertilizer 19 70 19 70 19
Herbicides 43 30 30 56 6
Energy 316 386 319 386 782
Machinery and Repairs 217 219 178 213 536
Labor 25 25 25 25 25
TOTAL CASH EXPENSES 691 1016 654 787 1459
REVENUE ABOVE CASH EXPENSES -627 163 212 1331 -1349

 

Appendix 2.3. Economic analysis of MSU in 2008 utilizing actual yields.

 

 

 

5 ha '1
MSU 2008 Canola Corn Soybean Sunﬂower Switchgrass
TOTAL REVENUE 23 1548 676 1380 338
CASH EXPENSES
Seed 71 287 83 37 0
Fertilizer 19 70 19 70 19
Herbicides 43 30 30 56 9
Energy 316 386 319 386 36
Machinery and Repairs 217 219 178 213 140
Labor 25 25 25 25 25
TOTAL CASH EXPENSES 691 1016 654 787 229
REVENUE ABOVE CASH EXPENSES -668 532 22 593 109

 

Appendix 2.4. Economic analysis of ROSE in 2008 utilizing actual yields.

 

 

 

ROSE 2008 Canola Soybean Sunﬂower Switchgﬁiss
TOTAL REVENUE 169 833 2403 341
CASH EXPENSES
Seed 71 83 37 0
Fertilizer 19 19 70 19
Herbicides 43 30 S6 9
Energy 316 319 386 36
Machinery and Repairs 217 178 213 140
Labor 25 25 25 25
TOTAL CASH EXPENSES 691 654 787 229
REVENUE ABOVE CASH EXPENSES -522 179 1616 112

 

85

ii

LITERATURE CITED

Albersheim, P., D. Nevins, P. English, and A. Karr. 1967. A method for the
analysis of sugars in plant cell wall polysaccharides by gas-liquid
chromatography. Carbohydr. Res. 5:340-345.

A008. 2000. AM 2-93: Determination of oil content in oilseeds. In Ofﬁcial
methods and recommended practices of the American Oil Chemists’ Society.
Champaign.

Belden, J. and J. Coats. 2004. Effect of grasses on herbicide fate in the soil
column: Inﬁltration of runoff, movement, and degradation. Environ. Toxicol.
Chem. 23:2251—2258.

Bothast, R. and M. Schlicher. 2004. Biotechnological process for conversion of
corn into ethanol. Appl. Microbiol. Biotechnol. 67: 19-25.

Burton, J. 1987. Quantitative genetics: results relevant to soybean breeding In:
J.R. Wilcox, Editors, Soybeans: Improvement Production and Uses (2nd edn.),
ASA, CSSA, and SSSA, Madison, WI (1987) Agron. Monogr. 16.

Buxton, D. and S. Fales. 1994. Plant environment and quality. p. 155-199. In G.
C. Fahey, Jr., M. Collins, D. R. Mertens, and L. E. Moser (eds). Forage Quality,
Evaluation, and Utilization. ASA-CSSA-SSSA, Madison, WI.

Cairns, John. 2007. Corn for the starving or ethanol for fuel?: An ethical
dilemma for members of the automobile culture. Sci. & Soc. 5(2) 117-126.

Chen, C., G. Jackson, K. Neill, D. Wichman, G. Johnson, and D. Johnson. 2005.
Determining the feasibility of early seeding canola in the northern Great Plains.
Agron. J. 97:1252—1262.

Dardanelli, J., M. Balzarini, M. Martinez, M. Cuniberti, S. Resnik, S. Ramunda, R.
Herrero, and H. Baigorri. 2006. Soybean maturity groups, environments, and
their interaction deﬁne mega-environments for seed composition in Argentina.
Crop Sci 46:1939-1947.

Dein, B., R. Bothast, L. lten, L. Barrios, and S. Eckhoff. 2002. Fate of Bt protein
and inﬂuence of corn hybrid on ethanol production. Cereal Chem. 79:582-585.

Dzantor, E., T., Chekol, and L., Vough. 2000. Feasibility of using forage

grasses and legumes for phytoremediation of organic pollutants. J. Environ.
Sci. Health. Part A 35:1645—1661.

86

Environmental Protection Agency, 2004. LEDC Parcels of Land at the South
Point Plant Superfund Site Are Ready for Industrial Reuse. Available at
http:/[www.epa.gov/superfund/programs/recyCle/pdf/southptrfr.pdf (veriﬁed 10
June 2008).

Environmental Protection Agency. 2008. Michigan Brownﬁeld Acreage.
Personal communication.

Fayoumi, N. 2008. Rose Township Dump. Environmental Protection Agency.
Available at http:/Iwww.epa.gov/R58uper/npl/michigan/MlD980499842.htm
(veriﬁed 05 Jan 2009).

Goering, C., A. Schwab, M. Daugherty, E. Pryde, and A. Heakin. 1982. Fuel
properties of eleven oils. Trans. ASAE 25: 1472-1483.

Gomide, J., C. Noller, G. Mott, J. Conrad and D. Hill. 1969. Effect of plant age
and nitrogen fertilization on the chemical composition and in vitro cellulose
digestibility of tropical grasses. Agron. J. 61:116-120.

Hammond, EC. 1991. Organization of rapid analysis of lipids in many individual
plants. Springer-Verlag, Berlin.

lzquierdo, N., L. Aguirrezabal, F. Andrade and V. Pereyra. 2002. Night
temperature affects fatty acid composition in sunﬂower oil depending on the
hybrid and the phenological stage. Field Crops Research 77:115—126.

Jackson, G. 2000. Effects of nitrogen and sulfur on canola yield and nutrient
uptake. Agron. J. 92:644—649.

Jiang, P. and K. Thelen. 2004. Effect of soil and topographic properties on crop
yield in a North-central com-soybean cropping system. Agron. J. 96:252—258.

Kincs, F. 1985. Meat fat formulation. JAOCS 62: 815-818.

Kimak, H., E. Dogan, M. Alpaslan, S. Celik, E. Boydak, and O. Copur. 2008.
Drought Stress Imposed at Different Reproductive Stages Inﬂuences Growth,
Yield and Seed Composition of Soybean. Philippine Agricultural Scientist 91:3
261—268.

Lemuz, C.R., B.S. Dien, M.E. Tumbleson, V. Singh and K.D. Rausch. 2004.
Ethanol yield determination for dry grind corn processing. Proceedings corn
utilization and technology conference. pp61.

Ma, F. and M. Hanna. 1999. Biodiesel production: a review. Bioresource
Technology 70: 1-15.

87

Matthaus, B. and L. Bmhl. 2001. Comparison of different methods for the
determination of the oil content in oilseeds. JAOCS 78:1 95-102.

McLaughlin, S. and L. Kszos. 2005. Development of switchgrass (Panicum
virgatum L.) as a bioenergy feedstock in the United States. Biomass and
Bioenergy. 28:515—535.

Michigan Department of Environmental Quality. 2007. Fiscal year 2007
Legislative Report. Available at http:/[www.michigan.gov/documents/deq/
FY2007LegislativeReport_229422_7.pdf (veriﬁed 12 March 2009).

Morrison, M., N. McLaughlin, E. Cober, and G. Butler. 2006. When is short-
season soybean most susceptible to water stress? Canadian Journal of Plant
Science. 86:5 Special Issue: Sp. Iss. SI 1327-1331.

Tﬂwu ...-

Moser, L. and K. Vogel. 1995. Switchgrass, big bluestem, and indiangrass.
Chapter 32. pp. 409—420. In R .F. Barnes, D. A. Miller, and C. J. Nelson (eds).
Forages, 5th ed. Vol.l: An introduction to grassland agriculture. Iowa State Univ.
Press, Ames, IA.

National Agricultural Statistics Service, United States Department of Agriculture.
Agricultural prices. October, 2008. Available at http:/lusda.mannlib.comell.edu
lusda/nass/AgriPric/l20003/2008/AgriPric-10-31-2008.pdf (veriﬁed 23 Feb. 2008).

National Sunﬂower Association. 2007. Available at http:/lwww.sunﬂowemsa.org
(veriﬁed 23 Jan. 2009).

Oklahoma State University Extension. 2007. Canola crop production. Available
at http:/lwww.canola.okstate.edu/cropproduction/harvesting/index.htm (veriﬁed
07 Jan. 2008).

Patton, A. and L. Gieseker. 1942. Seasonal changes in the lignin and cellulose
content of some Montana Grasses. J. Anim Sci. 1:22-26.

Pryde, E. 1983. Vegetable oil as diesel fuel: Overview. JAOCS 60: 1557-1558.

Reddy, K, J. Adams, and C. Richardson. 1999. Potential technologies for
remediation of brownﬁelds. Practice Periodical of Hazardous, Toxic, and
Radioactive Waste Management. 3(2): 61-68.

Rondanini, D., R. Savin and A.J. Hall. 2003. Dynamics of fruit growth and oil
quality of sunﬂower (Helianthus annuus L.) exposed to brief intervals of high
temperature during grain ﬁlling. Field Crops Res. 83:79-90.

[SAS] Statistical Analysis Systems. 2009. SAS/STAT User’s Guide. Release
9.1.3. Cary, NC: Statistical Analysis Systems Institute.

88

Schmer, M., K. Vogel, R. Mitchell, and R. Perrin. 2008. Net energy of cellulosic
ethanol from switchgrass. PNAS 105 (2): 464-469.

Selvendran, R. and O'Neill, M. 1987. Isolation and analysis of cell walls from
plant material. In Methods of Biochemical Analysis, G. David, ed. John Wiley 8
Sons, pp. 25—153.

Sokhansanj, S., M. Sudhagar, A. Turhollow, A. Kumar, D. Bransby, L. Lynd, and
M. Laser. 2009. Large-scale production, harvest and logistics of switchgrass
(Panicum virgatum L.) - current technology and envisioning a mature technology.
Biofuels, Bioprod. Bioref. 3:124—1 41 .

Sosulskl, K., F. Sosulskl, and E. Coxworth. 1988. Carbohydrase Hydrolysis of I!
Canola to Enhance Oil Extraction with Hexane. Journal of the American Oil .1
Chemists' Society 65:3. I3,

Stein, D. 2009. Crop enterprise budgets. Available at
http:/lwww.msu.edu/user/steind (veriﬁed 25 Feb 2009).

Stone, L., D. Goodrum, A. Schlegel, M. Jaafar, and A. Khan. 2002. Water
Depletion Depth of Grain Sorghum and Sunﬂower in the Central High Plains.
Agron. J. 94:936—943.

Summers, C., D. Haviland, and L. Godfrey. 2008. UC IPM Pest Management
Guildlines: Sugarbeet. UC ANR Publication 3469.

Swanson, S.P., Wilhelm, W.W., 1996. Planting date and residue rate effects on
growth, partitioning, and yield of com. Agron. J. 88.

United States Environmental Protection Agency. 1987. EPA Superfund
Record of Decision. EPA/ROD/R05-87/052.

Updegraf, OM. 1969. Semimicro determination of cellulose in biological
materials. Anal. Biochem. 322420 - 424.

Vassilev, A., J. Schwitzguebel, T. Thewys, D. van der Lelie, and J. Vangronsveld.
2004. The Use of 'Plants for Remediation of Metal-Contaminated Soils. The
Scientiﬁc World Journal. 419-34.

Walsh, M., D. De La Torre Ugarte, H. Shapouri, and S. Slinsky. 2003.
Bioenergy crop production in the United States. Environmental and Resource
Economics 24: 313—333.

89

Wang, S., W. lngledew, K. Thomas, K. Sosulski, and F. Sosulski. 1999.
Optimization of fermentation temperature and mash speciﬁc gravity for fuel
alcohol production. Cereal Chemistry 76:82-86.

York, W.S., Darvill, A.G., McNeil, T., Stevenson, T.T., and Albersheim, P. (1985)
Isolation and characterization of plant cell walls and cell wall components. Meth.
Enzymol. 118:3-40.

Zheljazkov, V., B. Vick, M. Ebelhar, N. Buehring, B. Baldwin, T. Astatkie, and J.
Miller. 2008. Yield, Oil Content, and Composition of Sunﬂower Grown at
Multiple Locations in Mississippi. Agron. J. 100:635—642.

90

Chapter 3. Use of Agronomically Marginal Land for BioEnergy Production
ABSTRACT

Agriculture faces unique challenges as increasing world population places
unprecedented demands on food, feed, and energy resources. Agricultural
resources are becoming increasingly stretched to meet the demands of both food
and fuel production. Marginal land could be utilized for bioenergy crop
production. In 2007 and 2008, we evaluated corn (Zea mays L. ssp. mays),
soybean (Glycine Max (L.) Merr.), canola (Brassica napus L. var. napus),
sunﬂower (Helianthus annuus L.), and switchgrass (Panicum virgatum L.) at
three sites to determine the highest-yielding crop for bioenergy production. Sites
selected for the study included two marginal sites, one at Lake City (LC),
Missaukee County, 44° 18’N, 85° 12’W (LC), the other at Chatham (UP), Alger
County, 46° 20’N, 86° 54’W, for comparison with a historically cropped site at
East Lansing (MSU), Ingham County, 42° 41’N, 89° 29’W. Comparisons of
biofuel crop quality components were also conducted. Yield of all crops differed
among locations; however, the historically cropped site did not consistently
provide higher yields than LC and UP. Ethanol produced from corn grain (L/kg)
differed signiﬁcantly (p<0.05) between locations: as latitude increased, ethanol
yield decreased. Total oil content of soybean was reduced with increases in
latitude, while the reverse was true of canola. Concentrations of palmitic (C16:0),
stearic (C1820), oleic (C18:1), linoleic (C182), and linolenic (C18z3) fatty acids
showed that northern locations yielded higher levels of oleic acid for soybean and
canola, while no differences were detected in sunﬂower. Total crystalline

cellulose levels from switchgrass were highest at MSU in 2008; however, the

91

MSU stand was one year older. Stands equal in age, regardless of location,

were not signiﬁcantly different in cellulose content.

INTRODUCTION

Energy production from agricultural crops in the United States has
become an ethical dilemma (Calms, 2007). Crops once reserved for food, feed,
and ﬁber are now allotted to energy production, potentially reducing the amount

of commodities available for human consumption. Expanding bioenergy crop

,.'I""-;'-P--='_.. _., _. a!

production acreage into areas not currently cultivated for agronomic yield could
offset the demand for consumption crops designated for energy production
(Walsh et al., 2003). One way to diminish this competition is to produce
bioenergy feedstock in areas not currently being used for production agriculture.
Marginal land is commonly. seen as unsuitable for cultivation of arable
crops and can be classiﬁed using various factors. The United States Department
of Agriculture has eight land capability classes based on level of agricultural
suitability. Classes Ill and IV are capable of producing crops yet have severe
use limitations, which include features such as slope, erosion, ﬂooding, low
moisture holding capacity and fertility, salinity, and adverse climatic conditions

(Natural Resources Conservation Service, 2009).

In the contiguous 48 states of the US, 174.6 million hectares are
identiﬁed as cropland (USDA, NAAS, 1999). The Conservation Reserve
. Program (CRP), which grants government payment for protection of sensitive
lands with capability classes of III and IV primarily, compensates owners for 12.1

million hectares nationwide (USDA FSA, 1999). According to Walsh et al.

92

(2003), 6.8 million hectares of the 12.1 million hectares of CRP could be used to
grow bioenergy crops in addition to 7.7 million hectares classiﬁed as idle.
Continuous moisture stress, the most limiting factor of crop yield, affects 3.7
billion hectares in the US. which contribute to 27.9% of the total available
landbase. Low temperatures and short growing seasons in northern climates
limit yield and occur on 2.2 billion hectares of the earth’s potential landbase

(USDA-ERS, 2003). Utilizing lands where marginal conditions limit production

a- '2‘]

could supplement the growing needs for renewable fuels.

(£15...-

Switchgrass has been studied to determine if this perennial grass is
competitive under marginal conditions (Mooney et al., 2005). Additionally,
comparisons between corn grain and stover yield and switchgrass biomass have
been explored in marginal environments (Varvel et al., 2008). However, little
information is available on multi-crop comparisons between corn (Zea mays ssp.
mays), soybean (Glycine Max (L.) Merr.), canola (Brassica napus L. var. napus),
sunﬂower (Helianthus annuus), and switchgrass (Panicum virgatum L.)
production on marginal soils. It is currently unknown whether marginal land can
produce adequate yield to offset capital investments, or whether crops grown on

various types of marginal land will yield biofuel of adequate quality.

US. food and energy independence could be attained by efﬁcient land
use strategies for bioenergy production, including cultivation of idle marginal
land, if fuel quality standards are met. In this study we sought to 1) evaluate

bioenergy crop yield on two marginal sites; 2) identify the optimal bioenergy crop

93

for each location; and 3) compare the quality of fuel derived from marginal sites

to that of fuel derived from traditional cropping systems.

MATERIALS AND METHODS

Site Description

Field experiments were conducted in 2006-2008 at marginal ﬁeld sites in
Chatham (UP), at the MSU Upper Peninsula Experiment Station, Alger County,
and in Lake City (LC), at the Lake City Experiment Station, Missaukee County in
2007. A third site of high fertility agricultural land was established in East
Lansing (MSU), at the Michigan State University Agronomy Research Farm,
Ingham County in 2006 to serve as an experimental control for crop yield and
quality analyses. Crops evaluated included com, soybean, canola, sunﬂower,
and switchgrass; sunﬂower was omitted at UP. Site characteristics are listed in

table 3.1.

Table 3.1. Site characteristics of location, elevation, soil type, land capability
class, pH, P and K levels.

 

 

Land
Site Location Elevation Soil Type Capability pH P K
m Class meq
MSU
42° 41'N 261 Capac Loam (ﬁne-loamy, mixed, II 6.0 3.3 8.1
89° 29W mesic, Aeric Ochraqualfs)
LC Montcalm-Graycalm complex
44° 18’N 377 (coarse-loamy, mixed, frigid Eutric III 5.6 6.0 1.1
85° 12'W Glossoboralfs and Alﬁc
Upidsamments
UP .
46° 20’N 368 Eben very cobbly sandy loam Vl 7.5 3.1 7.6
86° 54'W (sandy-skeletal, mixed frigid

Pachic Hapludolls)

Site history at LC included a previous crop of alfalfa in 2006 while UP was

planted to corn grown for silage. In 2005 at MSU, the previous crop was

soybean.

Cultural and Management Practices

Tillage methods took into account site-speciﬁc soil characteristics and
local production practices. At MSU and LC, plot preparation included spring
chisel plowing prior to planting. UP site preparation was minimal as crops were
no-tilled planted in 2007 and 2008. Soil features including location, elevation,

soil type, land capability class, pH, and soil fertility levels for all locations are

shown in Table 3.1.

Seeding dates, rates, and varieties are shown in Table 3.2. Varieties were

selected according to Michigan maturity zones.

95

 

Table 3.2

. Planting date, seeding rate, and seed selection for MSU, LC, and UP.

 

 

 

 

 

 

 

Crop Location Date Rate Variety
2007 2003 ha '1
Corn MSU May 21 May 13 74,131+ Dekalb 38H66-RR2
LC May 21 May 13 74,131+ Dekalb 38H66—RR2
UP May 20 May 14 74,131+ Dekalb DKC 33-11
Soybean MSU May 21 May 13 444,785+ Diehl Fields DF8242RR
LC May 21 May 13 444,785+ Diehl Fields DF8242RR
UP June 06 June 04 555,986+ Wolf River Valley WRV 2905
Canola MSU May 21 April 30 6.721: Dekalb DKL 38-25RR
LC May 21 April 30 6.721 Dekalb DKL 38-25RR
UP May 11 May 08 8.401 Dekalb DKL34-55
Sunflower MSU May 21 May 13 74,131+ Seeds 2000 Charger
LC May 21 May 13 74,131+ Seeds 2000 Charger
UP - - - -
Switchgrass MSU May 21* - 11.2# Forage First Forestburg
LC May 21 - 11.2# Forage First Forestburg
UP June 27 - 11.211 Ernst Conservation Seeds Cave in Rock
*Reseeded switchgrass stand

‘I' Seeding rate in seeds ha'1

:tSeeding rate In kg ha“1

# Seeding rate in kg pure live seed ha'1

Planting of annual crops at MSU and LC followed a canola-corn-soybean-

sunﬂower rotation with switchgrass remaining in dedicated plots. Switchgrass

was initially planted in 2006 and was no-till replanted at MSU in 2007 due to low

populations. At UP, planting of annual crops followed a canola-com-soybean

rotation with switchgrass remaining in dedicated plots.

At MSU and LC in 2007, corn, soybean, and sunﬂower were planted with

a custom built planter utilizing John Deere row and vacuum units (Deere 8

Company, Moline, IL), while canola and switchgrass were planted with an

96

 

Almaco drill (Almaco, Nevada, IA). In 2008, corn, soybean, and sunﬂower were

planted with a custom planter utilizing John Deere row units equipped with

MaterMacc (MaterMacc, San Vito al Tagliamento, Italy) metering units, while

canola was planted with an Almaco drill. At UP in 2007 and 2008, a John Deere

7000 no-till vacuum planter was used to plant corn and a Lilliston no-till seed drill

(Agco, Duluth, GA) was used to plant canola, soybean and switchgrass.

Herbicide labels were followed for application rates and timing was crop-

dependent as shown in Table 3.3.

Table 3.3. Postemergence herbicide applications and timing.

 

 

 

 

 

 

Crop Location Glyphosate Glufosinate Atrazine Acetochlor Quinclorac
Date (2007, 2008)
Corn MSU 5/23, 5/ 15
LC 5/23, 5/15
UP ‘ 5/31, 5/22 5/31, 5/22
Soybean MSU 5/23, 5/ 15
LC 5/23, 5/15
UP 7/9, 6/27
Canola MSU 5/23, 5/15
LC 5/23, 5/ 15
UP 6/9, 6/9
Sunﬂower MSU 5/24, 5/24
LC 5/24, 5/24
Switchgrass MSU 5/12, 5/ 10 *, 5/10
LC 5/12, 5/ 10
UP *, 6/4

 

* No application was made in 2007

Nitrogen soil amendments were side-dressed at rates of 112 kg ha'1 in

corn and sunﬂower. Other crops received no fertilization.

97

Harvest and Handling

In 2006, crops at MSU were planted and harvested to establish the
rotational system, but yield totals were not recorded. LC and UP crops were

established in 2007. Harvest dates for 2007 and 2008 are presented in Table

 

 

 

3.4.

Table 3.4. Harvest dates at MSU, LC, and UP. W
l I
|.

Crop Harvest Date '9"
MSU LC UP
2007 2008 2007 2008 2007 2008
Corn Oct. 12 Oct. 29 Oct. 21 Oct. 29 Oct. 25 Nov. 3

Soybean Oct. 11 Oct. 22 Oct. 11 Oct. 23 Oct. 30 Oct. 31
Sunﬂower Sept. 10 Sept. 17 Sept. 11 Sept. 20 - -

Canola + July 22 Aug. 22 July 24 Aug. 23 Aug. 11
Switchgﬁrass Nov. 1 Oct. 29 Oct. 30 Oct. 29 1 Oct. 31

1'Crop not harvested due to wildlife damage.

#Crop not harvested.

In 2007 at MSU and LC, corn and soybean grain yields were determined by
whole- plot harvests using a small plot combine (Massey Ferguson 8XP, AGCO,
Batavia, IL), and by ﬂail harvester (Carter Manufacturing Co. Inc., Brookston, IN)
for switchgrass. Two sunﬂower and canola rows, 1 m in length with three
replicates per plot, were harvested by hand. At UP, corn, soybean, and canola
were whole-plot harvested using a small-plot combine (Massey Ferguson 8XP, .
AGCO, Batavia, IL). Likewise, canola was whole-plot harvested using a small-

plot combine (Almaco, Nevada, IA). In 2008, only subplot areas not affected by

98

wildlife damage were harvested. In com, 1.5 by 12.9 m was hand-harvested
from two areas of each plot. In soybean 1.1 by 12.9 m was hand-harvested from
three areas of each plot. In canola, three areas 1.5 by 12.2 m were swathed
(Swift Machine, Swift Current, SK), then threshed (Almaco, Nevada, IA). In
sunﬂower, three areas 1.5 by 1 m were hand-harvested and sunﬂower heads
were threshed (Almaco, Nevada, IA). In switchgrass, three 50.8 by 50.8 cm
quadrats were hand-harvested at a cutting height of 7.6 cm from the soil surface.
At UP, whole-plot harvests were used for corn, soybean, and canola. In
switchgrass, three 50.8 by 50.8 cm quadrats were hand-harvested at a cutting
height of 7.6 cm from the soil surface. To eliminate potential bias associated with
wheel tracks and implement overiap, only the center rows of each planter pass

were harvested.

Grain moisture levels were determined using a DICKEY-john GAC 2100
grain analysis computer (DlCKEY-john, Auburn, IL). Grain samples were oven-
dried at 60°C to 15.5, 13.0, 8.5, and 10.0% storage moisture for corn, soybean,
canola, and sunﬂower, respectively. Switchgrass samples were weighed in the
ﬁeld, oven dried at 100°C for 4 days, and weighed again to determine dry matter

content.

At UP, switchgrass stands in 2007 did not have enough carbohydrate
reserves for cutting. Additionally, above-ground biomass was not substantial and
was left standing throughout the winter. Sunﬂower was not planted at UP as the

site’s climatic features and latitude are not conducive to sunﬂower production.

99

Corn Gra_in Ethalal

Triplicates of 25 9 ground corn grain samples were used for ethanol
analysis following a gravimetric, 72 h fermentation method described by Lemuz
et al. (2004). Each 25 9 ground corn sample was mixed with 75 ml deionized
water and 27 ul alpha-amylase (9000-85-5, Sigma-Aldrich Inc., Allentown, PA) to
break the 01-6 linkage in the starch and form single glucose molecules in an
Erlenmeyer ﬂask ﬁtted with a rubber cap and needle vent. The slurry was
incubated in a 90°C oven for 90 min with occasional shaking. The slurry
temperature was then lowered to 60°C, and the pH was adjusted to 4.1-4.5 with
hydrochloric acid (EMD Chemicals Inc. Gibbstown, NJ). The liqueﬁed com slurry
was sacchariﬁed and fermented simultaneously by adding 20 ul glucoamylase
(9032-08—0, Sigma-Aldrich Inc.), 5 ml yeast inoculum and 0.4 ml 1M NH4SO4 (EM
Science, Gibbstown, NJ). The ﬂask was placed in a 60°C oven for 2 h before it
was placed in an incubator shaker at 32°C at 150 rpm for 72 h (Model 025, New
answick Scientiﬁc, Edison, NJ). The yeast inoculum was prepared according to
Wang et al. (1999). Brieﬂy, active dry yeast (Fleischmann’s yeast, Fenton, MO)
was dispersed in 99 ml of 0.1% sterile peptone water (DOT Scientiﬁc Inc. Burton,
MI) and prewarmed at 38°C. The ﬂask with the slurry was weighed together
before and after the incubation, and the difference in weight was considered as
the mass of 002 released during the fermentation. Assuming 100% recovery,
total ethanol produced was computed using a 1:1 molar ratio of ethanol to C02

and the following conversion:

n = ((W1-W2)/44) x 46

100

I
'4 ".~
I

where n is the total ethanol yield in grams, W1 in grams is the weight of the
solution before fermentation, W2 in grams is the weight of the solution following
fermentation, 44 is the molecular weight of 002, and 46 is the molecular weight

of ethanol.

Fem Acid Methyl Esters

Fatty acid methyl ester (FAME) quantities in oil seed crops were using a
protocol adopted from Hammond (1991). Subsamples of soybean, canola, and
sunﬂower seed were collected at the time of harvest. Whole grain samples of 1 g
were placed in a 50-well, stainless steel plate and crushed using a hydraulic
press to break seed coats and physically extrude oil. For oil extraction, 1 ml
hexane was added to each sample of crushed seed. The reaction was covered
with a glass plate to ensure the hexane did not evaporate and was left for 2 h to
complete the reaction. A volume of 100 pl hexane and oil solution was removed
from the cmshing tray, mixed with 500 pl of 1 N sodium methoxide and
transferred to a GC-FID autosampler vial. Vials were placed on a Labnet Vortex
Mixer (Labnet lntemational, Inc., Woodbridge, NJ) for 30-60 min until oil droplets
were no longer ﬂoating in the sodium methoxide solution. After shaking, 1 ml of
hexane and 150 pl of distilled water were added to each vial. Following inversion
of the vial, the solution was allowed to sit for 20 min to complete the reaction.
The addition of hexane and water results in the formation of two distinct layers
within the solution. The upper, lower density layer of the solution containing
FAME was removed and placed in anOther GC-FID autosampler vial while the

lower layer was discarded. Samples were stored at 4°C until analysis.

101

Following the FAME reaction, samples were analyzed using GC-F ID. A
6890 Agilent GC-FID equipped with a 30 m x 0.53 mm x 0.5 pm Supelcowax 10
capillary column (SUPELCO lnc., Bellefonte, PA) and HP 7673 autosampler
(Agilent Technologies, Santa Clara, CA) were used to analyze FAME samples.
Sample vials were loaded into the autosampler tray and a sample sequence was
entered. GC-Fl D parameters included: an increase in oven temperature from
195°C to 240°C at 4°C/min and holding at 240°C for 2 min, injection volume of 1
pl with split ratio 20:1, inlet temperature of 220°C, and detector temperature of
290°C. Ultra high purity (99.999%) helium was used as carrier gas at a ﬂow rate
of 3.3 ml/min. Concentrations of the ﬁve major FAMEs were corrected with
external standard (Supelco, Bellefonte, PA) calibration curves (0.05, 0.1, 0.2, 0.5,
1, 2, 5 mg/ml ). Quadruplicate analyses were conducted for all samples.
Total il

Total oil content was determined using a protocol developed by Matthaus
and Brt’rhl (2001). An automated accelerated solvent extraction system ASE 200
(Dionex, Sunnyvale, CA) with 11 ml stainless steel sample cells was used for
total oil extraction. Moisture content of ground oilseed samples was measured
using a Moisture Analyzer A&D MF-50 (A&D Company Ltd, San Jose, CA).
Three 9 of sample was transferred to an extraction cell, ﬁtted with a cellulose
ﬁlter, and ﬁtted with a stainless steel frit at the outlet. The dead volume of the
cell was ﬁlled with diatomaceous earth (Dionex, Sunnyvale, CA). The extraction
cell was closed and automatically moved into the heatable extraction oven, ﬁlled

with solvent, and pressure was adjusted. Following the extraction period, the

102

 

 

extract was drained off into a collection vial which was purged with compressed
nitrogen to determine oil composition. The extraction temperature was 105°C,
with a preheat period of 6 minutes, and static extraction periods of 10 minutes
with 2 cycles and a ﬂush volume of 70% using hexane as a solvent were
selected. Pressure was held at 1000 psi throughout the extraction. After
processing, the extraction solvent hexane was evaporated by purging the
samples with oxygen free compressed nitrogen gas. The residual hexane was

evaporated in an oven at 105°C for approximately 60 min. The total oil content

 

was calculated using AOCS Ofﬁcial Method AM 2-93 (AOCS, 2000):

%C = 100 x Qw / (W x (1- moisture%))
in which Qw in grams is the total oil extracted, Win grams is the weight of the
ground sample, and moisture% is the moisture percentage of the ground sample.

Total Crystalline Sugar

Total crystalline cellulose content was calculated as pglmg cell wall
material and analyses were performed in triplicate. Dried switchgrass samples
were ground to a ﬁne powder and pigments, proteins, lipids, and DNA were
removed by washing the samples with organic solvents (York at al., 1985). The
residue was de-starched with an amylase treatment resulting in only cell wall
material remaining. The composition of the wall matrix polysaccharides was
obtained by treating the samples with triﬂuoroacetic acid. Derivatization of the
solubilized monosaccharides into their corresponding alditol-acetates and
separation and quantiﬁcation was conducted using gas chromatography

employing a mass spectrometer (GC-MS) (Albersheim et al., 1967).

103

The remaining residue was washed with Updegraff reagent, an acid mix
that results in stripping of further hemicelluloses and amorphous glucan
(Updegraf, 1969). The remaining residue of crystalline cellulose was hydrolyzed
with sulfuric acid (Seaman hydrolysis) (Selvendran and O'Neill, 1987) and the
resulting monosaccharide quantiﬁed with a colorimetric assay (anthrone assay).

Sugars extracted include rhamnose, arabinose, mannose, galactose, glucose,

xylose, and arabinose. Levels of sugars are crop dependent and only whole

’ uh; '_

plant switchgrass samples were analyzed for cellulosic ethanol potential. The

r-

matrix polysaccharide sugar composition gives indications about the presence
and abundance of wall polymers. In the grasses, arabinose and xylose represent
the hemicellulose arabinoxylan, mannose, the occurrence of gluco-mannan, and

glucose represents the presence of amorphous cellulose.

Statistical Analysis

The experiments at all locations were established in a randomized
complete block design (RCBD) with six replications at LC and UP while MSU had
four. Crop yield data analysis was conducted separately for each crop by year
due to differing environmental conditions in 2007 and 2008. Gravimetric ethanol,
total oil, FAME, and cellulose component yields were determined using a two
year average to determine signiﬁcance. Analysis of variance (ANOVA) was
performed on crop yield and fuel quality and quantity using PROC MIXED in SAS
(SAS Institute, 2009). The location treatments are considered ﬁxed effects while
blocks are considered a random effect. Results are based on the statistical

model Yij = p + Locationi + Block; + e. Normality of residuals was assessed using

104

PROC UNIVARIATE and Levene’s test was used to test homogeneity of
variances. When signiﬁcant effects of treatment occurred, means were
compared using Fisher’s Protected Least Signiﬁcant Difference (LSD). Unless

othenlvise stated, differences are considered signiﬁcant at an alpha level of 0.05.

Economic Analysis

Enterprise budgets and machine and work rates estimates were based on

 

Stein (2009). Seed prices were based on actual prices of purchased seed in
2007 and 2008. Fertilizer costs and herbicide prices were determined by 1.1..
Jorgensen Farms Elevator (Williamston, Ml). Commodity prices were

determined by National Agricultural Statistics Service for October 2008 (NAAS,

USDA, 2008).

RESULTS AND DISCUSSION
Environmental Conditions

In 2007, precipitation at MSU was below average during the month of July
(Table 3.5) and rain received late in the season slowed crop drying. In 2008,
precipitation received during seedling establishment was below the 30-year

average and rain received late in the season again slowed crop drying.

105

Table 3.5. Total monthly precipitation (cm) at MSU (East Lansing, MI), LC (Lake
City, MI), and UP (Chatham, MI) in 2007-2008. Long-term means (1970-2000)
included for comparison.

 

Location Year Mar. Apr. May June July Auﬁg. Sept. Oct.
MSU 2007 5.6 6.6 9.7 8.9 1.2 14.0 5.3 10.3
2008 4.0 4.3 3.0 11.3 7.6 1.7 18.6 4.1
30 yr. 5.7 7.5 7.5 9.3 7.9 7.8 6.3 5.5
LC 2007 8.1 5.8 3.5 6.0 4.3 5.8 4.1 5.9
2008 4.7 8.9 4.6 18.2 10.0 5.5 7.5 6.8
30 yr. 4.3 7.3 6.8 7.9 8.3 7.7 8.3 6.8

 

 

 

up 2007 10.3 16.3 5.5 5.2 4.4 2.0 12.5 13.3
2008 5.4 16.0 9.3 9.8 4.5 2.7 9.6 5.9 _.
30 yr. 6.3 8.0 6.5 7.4 8.8 8.4 10.2 9.5 ‘

 

In 2007, LC and UP experienced droughty conditions during critical crop
growth and grain ﬁlling periods. At UP, rainfall was unevenly distributed during
most of July, with very little rain received over a 44-day span from July 15 to
August 27, resulting in drought conditions. At LC, starting the same day,
droughty conditions spanned a 35-day period to August 18. Weather patterns
were slightly different in the southern part of the state at MSU as the site
experienced only a span of 25-days, lasting from June 28 to July 22, of less than
0.2 cm of rainfall. In 2008, precipitation at MSU and LC was closer to average
than 2007. MSU and LC experienced abundant rainfall in June. At UP in 2008,
rainfall was abundant in April during crop emergence and conversely was scarce

during crop reproductive and grain ﬁlling stages.

In 2007, monthly temperatures at MSU, LC, and UP were slightly above

average (Table 3.6).

106

Table 3.6. Average monthly air temperatures (°C) at MSU (East Lansing, MI), LC
(Lake City, MI), and UP (Chatham, Ml) 2007-2008. Long term means (1970-
2000) included for comparison.

 

Location Year Mar. Apr. May June July Pig. Sept. Oct.
MSU 2007 3.7 7.1 15.6 19.8 20.6 21.4 17.6 7.1
2008 -0.2 9.9 12.6 19.6 21.5 20.7 17.4 9.7

30 yr. 0.6 7.9 13.8 19.4 21.4 20.7 16.5 11.0

 

 

LC 2007 2.7 5.6 14.9 19.0 19.9 20.5 17.3 13.0
2008 -2.0 8.7 10.4 18.0 20.1 19.7 16.4 8.8
30 yr. -2.6 5.8 12.1 17.2 19.6 18.5 14.2 8.6 T

 

UP 2007 —0.6 3.0 12.3 17.2 19.1 18.3 14.9 10.6 ’0
2008 -5.3 3.9 8.2 15.4 18.4 17.9 14.1 7.3
30 yr. -3.1 3.9 11.1 16.1 18.6 18.1 13.9 7.8 L"-

 

Air temperatures recorded at MSU were below the 30-year average during the
months of April, July, and October. Temperature at LC were cooler in April, and
colder in both March and April at UP. Temperatures in 2008 were different
between sites since MSU and LC experienced higher temperatures than normal
while they were cooler at UP. The 30-year means were surpassed in all months
at MSU with exceptions of March, May, and October and at LC in all but March

and May. UP recorded cooler temperatures only in September.

Crop maturity can be predicted using daily average high temperatures and
low temperatures to calculate Growing Degree Days (GDD). GDD are calculated
as the average of daily high and low temperatures less a base temperature of
10°C. In 2007, temperatures were on average warmer (Table 3.6) and more

GDD were accumulated (Table 3.7) than in 2008 at all locations.

107

Table 3.7. Monthly and yearly accumulation of Growing Degree Day units at

MSU (East Lansing, MI), LC (Lake City, MI), and UP (Chatham, Ml) during 2007
and 2008, calculated using a base temperature of 10°C.

 

 

 

 

Location Year Mar. Apr. May June July Aug. Sept. Oct. Total GDD
MSU 2007 26.4 38.4 176.4 295.6 326.4 354.1 230.2 140.4 1587.9
2008 0.1 68.9 85.8 286.9 356.8 330.0 222.8 53.2 1404.6
LC 2007 14.7 22.4 155.5 269.6 312.4 324.6 218.6 128.7 1446.6
2008 0.0 62.2 40.4 240.6 310.9 302.0 192.8 37.1 1186.1
UP 2007 1.6 14.3 102.9 219.4 281.0 257.0 153.8 77.4 1107.4
2008 0.0 7.6 15.6 163.4 260.9 244.1 123.9 20.6 836.1

 

In 2007, MSU accumulated 1588 GDD while LC and UP accumulated
1447 and 1108 GDD, respectively. Results for 2007 indicate MSU accumulated
9% more GDD than LC and 30% more than UP. In 2008, MSU accumulated
1404 GDD while LC and UP accumulated 1186 and 836 GDD, respectively.
Results for 2008 indicate MSU accumulated 16% more GDD than LC and 40%
more than UP. Summed GDD over both years indicated dissimilar conditions at
MSU, LC, and UP. MSU accumulated 12% and 35% more GDD than LC and
UP, respectively. LC accumulated 26% more GDD than UP. Locations which
receive more GDD have the potential to achieve higher yield totals (Swanson
and Wilhelm, 1996). Given that GDD accumulations were dissimilar at the three
sites, temperature could have had a signiﬁcant effect on yield differences.
Amngnjc Yielg

Differences among sites included climate, soil type, shading, crop
demands, and logistics required differing harvest dates (Table 3.4). Canola, on

average, contains 40% oil, making the seed a very attractive food and energy

108

 

source for birds, in particular the American Goldﬁnch (Carduelis tn'stis L.). In
2008, swathing canola before it reached physiological maturity resulted in 5%
higher yields (S. Smith and K. Thelen, unpublished data). Canola was swathed
when 60-75% of the seeds were black and contained 30-40% moisture as
recommended by Oklahoma State University Extension (2007). To protect
sunﬂower crop yield in both years, plots were hand harvested and dried when
they had reached fully physiological maturity (R9) but had not yet dried to

acceptable mechanical harvest moisture percentages or storage moistures of

hatmm -
I‘ll . J-

9.5%.

Statistical analysis determined that blocking within locations was not a
signiﬁcant factor (p > 0.05). Therefore, crop yield totals from all plots per crop
were averaged to determine a yield mean for each location. MSU, LC and UP

yield means differed (p < 0.05) in both years (Table 3.8).

Table 3.8. Mean crop yields of bioenergy crops in 2007 and 2008 at MSU (n=4),
LC (n=6), and UP (n=6). Means within a year and crop followed by a common
letter are not different at (p < 0.05) using Fisher’s Protected Least Signiﬁcant
Difference test.

 

 

 

 

Crop Yield (Mg Ital)
2007 2008

MSU LC UP MSU LC UP
Corn 4.87 a 7.56 a 3.85 b 9.89 b 11.39 a 5.53 c
Soybean 1.83 a 2.17 a 1.16 b 2.10 a 1.74 b 0.75 c
Sunﬂower 2.54 a 2.43 a 1: 2.70 b 3.83 a II:
Canola + 0.36 b 1.26 a + 1.38 b 1.87 a
Switcﬁhgrass 0.61 b 4.41 a # 7.39 b 11.71 a 8.46 b

 

1'Yield data not computed due to wildlife damage; samples retained for quality analysis.
:I:Crop was not planted.

#Crop was not harvested.

109

 

In 2007, grain yield of corn and soybean at MSU and LC did not differ yet
UP yielded less than both locations (p<0.05). MSU yielded 21% while LC yielded
49% more com grain than UP. Sunﬂower yield was essentially the same MSU
and LC. Canola yielded 71% more seed at UP compared to LC. LC yielded 86%

more switchgrass biomass than MSU.

In 2008, MSU signiﬁcantly yielded 17% more soybean than LC and 64%
more than UP. UP soybean yield was substantially less than LC yield by 57%.
MSU corn grain yield was 13% lower than LC yield, yet was signiﬁcantly higher
than UP. UP yielded 51% less corn grain than LC. LC signiﬁcantly yielded 30%
more sunﬂower than MSU. Canola yield was 26% higher at UP compared to LC.
Canola yield at MSU was destroyed by wildlife. Switchgrass at LC signiﬁcantly
yielded 37% more biomass than MSU and 28% more than UP. Switchgrass
stand establishment at LC was favorable. Lower weed density at LC lead to
higher biomass yield as switchgrass did not have to compete with weed species
for resources. In 2008, yields overall were higher at LC as compared to MSU

and UP as LC yielded signiﬁcantly more com, sunﬂower, and switchgrass.

Lower soybean yield (Table 3.8) was inversely related to latitude (Figure

3.1) with MSU yielding 2.10, LC 1.74, and UP 0.75 Mg ha'l.

110

15.00
10.00

Mg ha‘1

0.00

3.00
2.00

Mg ha'1

0.00

4.00

Mg ha'1

0.00

2.00

1.00

Mg ha'1

0.00

10.00

5.00

Mg ha'1

0.00

 

 

5.00 -

 

 

 

 

 

 

 

 

 

 

 

 

1.00 -

 

 

 

 

 

 

 

2.00 -

 

 

 

 

 

 

 

 

 

 

 

 

I
42 44 T 46
Soybean
42 I. 44 46
Sunflower
I if 1
42 7': 44
Canola
f
44 46
Switchgrass

 

 

 

l'—-I

 

 

 

—_

 

42

46

Figure 3.1a-e. Crop yield response and standard errors (0:0.05) at three

latitudes 42°N (MSU), 44°N (LC), and 46°N (UP).

111

 

Canola yield responded positively to northern, cooler environmental conditions as
UP returned signiﬁcantly higher yields (Table 3.8). These results indicate canola
is a candidate biodiesel crop to be planted in northern climates while soybean
should be planted in southern climates. Switchgrass returned higher biomass
yield at the two northern sites, but showed elasticity across latitudes. Corn yield
is limited by latitude and is not considered an optimal bioenergy choice for
northern climates. Sunﬂower seems to be a candidate for bioenergy production
based on yield; however the high market value of sunﬂower could make it an

unproﬁtable feedstock source of biofuel (Nelson et al., 1994).

Marginality of land for production of biofuel is not limiting. Latitude effects
are crop speciﬁc. If a given crop is planted at a given latitude, marginal land yield
is suitable for bioenergy feedstock production. The MSU site is part of a
traditional cropping system not limited by marginal soil conditions or latitude.
Favorable conditions at MSU should support optimum yield. In 2007, MSU
recorded yields which were consistently higher than UP. Yields were
comparable between MSU and LC with LC returning higher switchgrass biomass
yields. MSU had the highest soybean yields while UP had the highest mnola.
Based on two years of yield data, marginal land supports bioenergy crop

production.

Corn Grain Ethanol Yield

Gravimetric fermentation of corn grain is industry protocol for ethanol

production. Ethanol yields were averaged over 2 years for MSU, LC, and UP

112

:3

 

(Figure 3.2). Ethanol yields at MSU, LC, and UP of 0.453, 0.433, and 0.398 L
ethanol/kg corn were all signiﬁcantly different. MSU and LC yields were

consistent with industry average yields of 0.417 L/kg (Bothast and Schlicher,

 

 

 

 

2004)
Corn Grain Ethanol Yield
0.500
a b
0.400 —
E
E. 0.300 —
g 0.200 -
3
0.100 r
0.000 -
MSU LC UP

 

 

 

Figure 3.2. Two-year average corn grain ethanol yields at MSU, LC, and UP,
2007 and 2008 data combined; means followed by a differing letter are
statistically signiﬁcant at a p < 0.05 using Fisher’s Protected Least Signiﬁcant
Difference test.

Factors that inﬂuence the ethanol yield potential of corn grain are difﬁcult
to use as yield predictors because they are not well understood. Dein et al.
(2002) found that the starch content of corn did not correlate well with ethanol
yields. They suggested that ethanol yield may be affected by climatic conditions
and locations. Reicks et al. (2009) found that, when the corn grain ﬁlling period
was shortened by a killing frost, ethanol yield was signiﬁcantly reduced.

Dissimilar ethanol yields can be explained by differing latitude (Figure 3.1),

113

temperature (Table 3.6), and GDD unit accumulations (T able 3.7). Corn suited
for the same maturity zone was used at all experimental locaﬂons which could
have affected physiological maturity and consequently ethanol yield at northern
locations. LC and UP may not have accumulated enough GDD (Table 3.7) to
reach maturity.
Egtty Acig Methvl ﬁters Yielg

Biodiesel, which consists of fatty acid methyl esters (FAME), is derived

from triglycerides through a process termed transesteriﬁcation. Following

1:1—1:75;.) ..
‘ _ ____—q

triglyceride extraction, ﬁve FAME compounds are of particular interest for
biodiesel production, including palmitic C1620, stearic C18:0, oleic C18:1, linoleic
C182, and linolenic C18:3 fatty acids. FAME proﬁles will vary among crops and
growing conditions. FAME levels were determined using gas chromatography
with a ﬂame ionization detector (CG-F ID), with combined two-year results

indicating minimal signiﬁcant differences between MSU, LC, and UP (Table 3.9).

114

Table 3.9. Results of fatty acid methyl ester (FAME) analysis of oilseed crops at
MSU, LC, and UP, 2007 and 2008 dates combined (mean percent of total oil
content by weight); % FAME followed by a differing letter within a fatty acid are
statistically signiﬁcant at p < 0.05 using Fisher’s Protected Least Signiﬁcant
Difference test.

% fatty acid methyl esters/ml crop oil
n Palmitic Stearic Oleic Linoleic Linolenic

 

Soybean
MSU 32 11.68 b 5.33 a 24.63 b 50.92 a 7.44 b

 

 

LC 48 13.16 a 5.37 a 26.19 a 47.93 b 7.35 b

UP 48 10.24 c 4.52 b 25.57 a 51.08 a 8.59
Canola

MSU 32 5.11 b 2.27 a 60.65 b 22.60 a 9.37

LC 48 5.88 a 1.93 b 63.52 a 19.81 b 8.86 3

UP 48 5.08 b 2.32 a 63.93 a 20.19 b 8.49 b
Sunﬂower

MSU 32 5.18 a 4.74 a 77.36 a 12.63 a

LC 48 4.66 a 4.95 a 81.07 a 9.24 b

 

1' Sunﬂower oil does not contain linolenic FAME.

In soybean, the palmitic, stearic, linoleic, and linolenic FAME percentages
differed signiﬁcantly between MSU, LC, and UP, and data were consistent with
those determined by Goering et al. (1982). Palmitic acid is the most common
saturated fatty acid found in plants (Goering et al.,1982; Kincs, 1985). Because
they are highly saturated, palmitic and stearic acids have high viscosity, causing
gelling and coagulation problems in diesel engines (Pryde, 1983; Ma and Hanna,
1999); therefore, low levels of palmitic and stearic acid are desirable. Linoleic
and linolenic FAME are unsaturated omega-6 fatty acids. Biodiesel with high
levels of linoleic and linolenic fatty acids, due to their unsaturation, has reduced

storage duration. Given these factors, optimum biodiesel is obtained by

115

i
it

.1 Imam mm.

 

optimizing oleic fatty acid levels. MSU soybean produced 1.6 and 0.9% less
oleic acid than LC and UP, respectively. In soybean, levels of palmitic and
stearic differed as latitude increased, with LC yielding the highest levels and UP
recorded the lowest. UP yielded the highest levels of linoleic and linolenic fatty

acids (Table 3.9).

Canola oil fatty acid proﬁles were affected signiﬁcantly across marginal

locations. lzquierdo et al. (2002) found that the oleic: linoleic ratio depends on

 

the temperature during the early stages of oil synthesis in oilseeds. Differences
in oleic acid levels were signiﬁcant: MSU contained the lowest levels with 3.3
and 2.9% less than UP and LC, respectively. Northern locations of LC and UP
yielded lower levels of linoleic and linolenic fatty acids compared to MSU (Table
3.9). Mixed results across latitudes were found when comparing palmitic and
stearic acid levels. Due the higher levels of oleic and reduced levels of palmitic,
stearic, linoleic, and linolenic acids, canola oil resulted from northern locations

possessed optimum biodiesel quality.

Sunﬂower, grown at MSU and LC, yielded approximately equal oleic and

stearic fatty acid levels (Table 3.9). Lower levels of linoleic and linolenic fatty

 

acids were produced at LC. Due to the precedence of oleic acid yielding
optimum biodiesel, MSU and LC are both considered acceptable for producing

sunﬂower for conversion to high quality biodiesel.

Determining optimum oilseed crops for location-speciﬁc management

must include consideration of latitude before any confounding marginal soil

116

features. Southern locations in Michigan, based on FAME analysis, are
acceptable for sunﬂower biodiesel production. Northern locations should focus
primarily on canola, since oleic acid levels are acceptable and crop yield is
competitive. Soybean is well suited to southern locations, based on crop yield
and FAME proﬁles. Soil and agronomic factors including differences in marginal
land characteristics such as water holding capacity, fertility levels, and GDD

accumulation, can affect fatty acid proﬁles in oilseed crops (lzquierdo et al., r'l- '

.m—

3.171.. .._

2002). The differing latitude (Figure 3.1), temperature (Table 3.6) and GDD
accumulation (Table 3.7) of the three research locations resulting in differing

FAME levels (Table 3.9).

Total Oil Yield

Two-year average total oil quantity for the oilseed crops differed between
MSU (n=8), LC (n=12), and UP (n=12) for soybean and canola, but showed no

signiﬁcance for sunﬂower (Figure 3.3).

117

 

 

Oil Quantities

Percent Oll Content

 

Soybean Canola Sunﬂower

 

 

 

Figure 3.3. Mean total oil content (% by weight) of soybean, canola, and
sunﬂower of oilseed crops at MSU, LC, and UP, 2007 and 2008 dates combined;
means followed by a differing letter are statistically signiﬁcant while * represents
no statistical signiﬁcance at p < 0.05 using Fisher’s Protected Least Signiﬁcant
Difference test.

Total oil yield of soybean appeared to increase as latitude decreased,
ranging from 17.8 to 21.5% of seed dry mass, which is consistent with data
reported by Dardanelli et al. (2006). Oil content of soybean differed signiﬁcantly
between locations. Soybean at MSU yielded 21.5% while LC and UP yielded
20.0 and 17.8%, respectively. MSU did not statistically differ from LC, although
MSU soybean yielded 7% more oil. Oil content did not differ between LC and
UP, but MSU soybean yielded signiﬁcantly more oil, by 17%, than UP soybean.

Sunﬂower seed accumulates oil in the ﬁnal stage of reproductive growth,
and oil content can differ with environmental conditions including soil type,
shading, agronomic practices, and precipitation (Zheljazkov et al. 2008). Seed
oil concentration is important for growers and processors. For example, sunﬂower

118

crushing plants in North Dakota and Kansas offer a 2% price premium for every
1% oil above 40% (National Sunﬂower Association 2007). No location
differences were observed, with sunﬂower at MSU yielding 57.6 and LC yielding
55.5% oil. However, these percentages may be slight overestimates because a
small fraction of seed coats could not be removed from the samples and was
included in the oil extraction process.

Jackson (2000) found that seed oil content of canola was reduced by
increasing applications of N fertilizer. At both MSU and ROSE, canola was
treated with the same, low-input fertilizer regime with no additional N
amendments. Canola total oil content can also be reduced by low levels of
precipitation during critical oilseed ﬁlling and oil accumulation periods (Chen et al.
2005). Canola at MSU (36.7%) contained signiﬁcantly less oil compared to LC
(39.0%) and UP (45.3%); however, all data fell within the range reported by
Sosulskl et al. (1988). Northem, marginal sites appear to be well suited for
canola production based on crop yield (Table 3.8), FAME (Table 3.9) and total oil

content (Figure 3.3).

Differences in latitude (Figure 3.1), precipitation (Table 3.5), temperature
(Table 3.6) and GDD accumulation (Table 3.7) between MSU, LC, and UP
affected total oil content. Crop oil content did not appear to be affected by fertility
levels. Soybean beneﬁted from southern locations; conversely, canola beneﬁted
from northern locations. Sunﬂower oil content did not differ between MSU and
LC. Marginal land sites produced oilseed crops which were comparable to the

conventional agronomic site in grain yield as well as total oil content.

119

Total Cmstalline Cellulose Yield

Single year means of crystalline cellulose content of switchgrass differed
signiﬁcantly between MSU, LC, and UP and results were not year dependent

(Figure 3.4).

 

Crystalline Cellulose
400 a
350 b "
300 —————’
250 ~—

 

 

hASU

200 - .1
' am

150 -— '
100 -— .
50 ---—T

Its/ms

IIJP

 

 

   

2007 2008

 

 

 

Figure 3.4. Average crystalline cellulose content of switchgrass at MSU, LC, and
UP in pg cellulose/mg switchgrass; means followed by a differing letter are
statistically signiﬁcant at p < 0.05 using Fisher’s Protected Least Signiﬁcant
Difference test.

Switchgrass grown at MSU yielded higher levels of crystalline cellulose
compared to LC and UP in 2007 and 2008. Stand age may inﬂuence cellulose
content (Gomide, 1969). MSU yielded 12% less cellulose in 2007 than in 2008.
Crystalline cellulose levels in all second-year switchgrass stands, LC and UP in
2008 and MSU 2007, were not signiﬁcantly different (Figure 3.4). In 2007, ﬁrst-
year stands at LC produced switchgrass with the lowest crystalline cellulose

content at 277.8 pglmg, while switchgrass at MSU in 2008 was the highest

120

yielding at 373.1 pglmg. Marginal land characteristics did not appear to have an

effect on differences in cellulose content.

Switchgrass is a photoperiod-sensitive plant. Switchgrass cellulose
concentration and composition can be inﬂuenced by many environmental factors
(Moser and Vogel 1995). Temperature is the environmental factor that most
strongly affects plant cell walls, with high temperatures causing increased rate of
reproductive maturation. Temperature and water stress can affect morphological
development by resulting in higher cell wall concentrations, with increases
occurring in both polysaccharide and lignin fractions (Buxton and Fales, 1994).
Temperature and GDD accumulation were dissimilar at the three locah'ons and
could have inﬂuenced total cellulose content.

Patton and Gieseker (1942) reported differencesin cellulose and lignin
content depending upon maturity as well as grass species. Switchgrass grown at
MSU was established one year prior to LC and UP plots. A more mature stand
of switchgrass is more competitive with annual weed species, and the higher
purity of the switchgrass stand likely resulted in higher crystalline cellulose
content. In addition, different sampling practices were employed in the two
years. Taking into account stand age and sampling practices, total cellulose
content at MSU, LC, and UP was similar when equal maturity switchgrass

samples were compared.

121

Total Fuel Yield

Total fuel yield from each crop and location was determined by total crop

yield and quantity of fuel produced (Table 3.10).

Table 3.10. Total fuel yield from crops grown on marginal sites (LC and UP)
compared to a traditional cropping system (MSU).

 

 

 

Crop Fuel Yield (kg ha'l)1
2007 2008

MSU LC UP MSU LC UP
Corn 2200 3300 1600 4500 5000 2100
Soybean 380 390 210 460 380 130
Sunﬂower 1500 1400 1 1500 2100 1
Canola + 140 510 + 630 940
Switchgrass2 200 1200 # 2800 3600 2600

 

 

tFuel yield not computed; wildlife damage.
:I:Fuel yield not computed; crop was not planted.

#Fuel yield not computed; crop was not harvested.
1Fuel yield assumes 100% recovery

2Total crystalline cellulose reported in kg ha'1 due to various cellulosic ethanol conversion
practices

In both years, corn at MSU and LC yielded higher more ethanol than oilseed
crops. At UP in 2008, switchgrass following the establishment year yielded more
fuel than corn. Sunﬂower yielded the most fuel of all oilseed crops at MSU and
LC, while canola was best suited at UP. However, biodiesel levels were not
comparable to those of ethanol produced from corn and switchgrass. Biodiesel
produced from oilseeds is more energy-dense than ethanol from grain or

biomass due to high concentrations of available carbon (Shapouri et al., 1995;

122

Hill et al., 2006). Based on preliminary results, in locations where soil and
climatic conditions are marginal, a perennial biomass crop may yield more fuel

than intensively managed annual crops.

Economic Analysis

Producers planning to implement bioenergy cropping rotations in their
farming operations will make decisions based ultimately on optimizing economic
returns. Under marginal land conditions, positive economic outcomes will be
possible if the input costs are limited.

Table 3.11. Economic return utilizing marginal land (LC and UP) compared to
traditional (MSU) cropping system acreage.

 

 

 

 

Crop Economics
2007 . 2008

MSU LC UP MSU LC UP
Corn - + - + + +
Soybean - + - + + -
Sunflower + + X + + X
Canola X + + X + +
Switchgrass - - - + + +

 

- Represents an economic net loss
+ Represents an economic net gain
X Indicates the crop was not harvested

Returns from corn, soybean, and sunﬂower at MSU will provide a ﬁnancial
gain. Canola returns suffered fromwildlife damage. Sunﬂower, at MSU and LC,
was proﬁtable in both years. Corn and soybean were consistently proﬁtable at
LC; at MSU, these two crops produced an estimated net loss in 2007 probably
because of weeds limiting available water. At LC, all crops returned economic

gain with exception of switchgrass in the establishment year. UP did not

123

 

 

consistently provide economic gain for soybean and corn. Although, switchgrass
does not currently have market value as a bioenergy crop, however $41.50 Mg’1
(dry weight basis) price projections were adapted from Sokhansanj et al. (2009).
Production and economic analysis of a perennial crop lends to high inputs in the
establishment year and decreasing inputs in subsequent years. Switchgrass,
following the establishment year, indicated a proﬁt. Agronomic and management
practices need to be reﬁned for all bioenergy crops to be proﬁtable under "'
marginal conditions. Under ideal ﬁeld conditions, all crops have the potential to
be proﬁtable. More years of research are needed to allow modeling for economic
analysis.

Determining the optimum bioenergy crop to be grown for proﬁt on
marginal land is not clear-cut. Currently, cellulosic ethanol is not commercially
produced in the US. McLuughlin and Kszos (2005) have reported that the
economic return of ethanol production from lignocellulosic sources surpasses
that of annual crops. Under current industry practices, and given environmental
conditions at LC and UP during 2007 and 2008, the greatest economic return
was obtained from canola and corn. Once cellulosic production technologies are
reﬁned, switchgrass production on marginal sites may prove to return the highest

economic gains.

CONCLUSIONS
The LC and UP marginal land sites provided bioenergy crop yield and
quality similar to that from traditionally cropped land, indicating the potential for

similar marginal sites across the Midwest to also be viable options for bioenergy

124

production. Fuel quantity and quality including ethanol, total oil, crystalline
cellulose, and FAME produced from marginal sites showed minimal differences
when compared to areas with traditionally high fertility. Preliminary economic
analysis supports potential investment in economic endeavors on marginal
acreage. Perennial cellulosic crops such as switchgrass hold great promise for
production on marginal land sites where water availability and soil fertility will limit
yield of annual crops. Increasing available feedstock for energy production on
alternative production sites eases stress on market demand for food crops.
However, expansion of cropping systems into marginal land needs to be
approached with caution to avoid disrupting environmentally sensitive acreage

and ecosystems.

125

runs-£21

 

APPENDICES

Appendix 3.1. Economic analysis of MSU in 2007 utilizing actual yields.

 

 

‘ r,“

 

S ha ‘1
MSU 2007 Canola Corn Soybean Sunﬂower Switchgrass
TOTAL REVENUE 169 968 581 4623 28
CASH EXPENSES
Seed 71 287 83 37 92
Fertilizer 19 70 19 70 19
Herbicides 43 30 30 56 6
Energy 316 386 319 386 782
Machinery and Repairs 217 219 178 213 536
Labor 25 25 25 25 25
TOTAL CASH EXPENSES 691 1016 654 787 1459
REVENUE ABOVE CASH EXPENSES -522 -48 -73 3836 -1431

 

Appendix 3.2. Economic analysis of LC in 2007 utilizing actual yields.

 

 

 

S be ‘1
LC 2007 Canola Corn Soybean Sunﬂower Switchgrass
TOTAL REVENUE 14636 1538 783 4419 164
CASH EXPENSES '
Seed 71 287 83 37 92
Fertilizer 19 70 19 70 19
Herbicides 43 30 30 56 6
Energy 316 386 319 386 782
Machinery and Repairs 217 219 178 213 S36
Labor 25 25 25 25 25
TOTAL CASH EXPENSES 691 1016 654 787 1459
REVENUE ABOVE CASH EXPENSES 13945 523 130 3633 -1295

 

126

Appendix 3.3. Economic analysis of UP in 2007 utilizing actual yields.

 

 

 

5 ha '1
UP 2007 Canola Corn Soybean Switchgrrass
TOTAL REVENUE 52 769 263 110
CASH EXPENSES
Seed 71 287 83 92
Fertilizer 19 70 19 19
Herbicides 43 30 30 6
Energy 316 386 319 782
Machinery and Repairs 217 219 178 536
Labor 25 25 25 25
TOTAL CASH EXPENSES 691 1016 654 1459
REVENUE ABOVE CASH EXPENSES —639 -247 -391 -1349

 

Appendix 3.4. Economic analysis of MSU in 2008 utilizing actual yields.

 

 

 

5 ha ’1
MSU 2008 Canola Corn Soybean Sunﬂower Switchgrass
TOTAL REVENUE 23 1548 676 1380 338
CASH EXPENSES
Seed 71 287 83 37 0
Fertilizer 19 70 19 70 19
Herbicides 43 30 30 56 9
Energy 316 386 319 386 36
Machinery and Repairs 217 219 178 213 140
Labor 25 25 25 25 25
TOTAL CASH EXPENSES 691 1016 654 787 229
REVENUE ABOVE CASH EXPENSES -668 532 22 593 109

 

127

Appendix 3.5. Economic analysis of LC in 2008 utilizing actual yields.

 

 

 

5 ha '1
LC 2008 Canola Corn Soybean Sunﬂower Switchgrass
TOTAL REVENUE 571 1785 563 7623 533
CASH EXPENSES
Seed 71 287 83 37 0
Fertilizer 19 70 19 70 19
Herbicides 43 30 30 56 9
Energy 316 386 319 386 36
Machinery and Repairs 217 219 178 213 140
Labor 25 25 25 25 25
TOTAL CASH EXPENSES 691 1016 ' 654 787 229
REVENUE ABOVE CASH EXPENSES -120 769 -91 6836 304

 

Appendix 3.6. Economic analysis of UP in 2008 utilizing actual yields.

 

 

 

$ ha '1
UP 2008 Canola Corn Soybean Switchgrass
TOTAL REVENUE 775 1331 97 351
CASH EXPENSES
Seed 71 287 83 _ 0
Fertilizer 19 70 19 19
Herbicides 43 30 30 9
Energy 316 386 319 36
Machinery and Repairs 217 219 178 140
Labor 25 25 25 25
TOTAL CASH EXPENSES 691 1016 654 229
REVENUE ABOVE CASH EXPENSES 84 315 -557 122

 

128

 

LITERATURE CITED

Albersheim, P., D. Nevins, P. English, and A. Karr. 1967. A method for the
analysis of sugars In plant cell wall polysaccharides by gas-liquid
chromatography. Carbohydr. Res. 5:340-345.

AOCS. 2000. AM 2-93: Determination of oil content in oilseeds. In Ofﬁcial
methods and recommended practices of the American Oil Chemists’ Society.
Champaign.

Bothast, R. and M. Schlicher. 2004. Biotechnological process for conversion of
corn into ethanol. Appl. Microbiol. Biotechnol. 67: 19-25.

J.R. Wilcox, Editors, Soybeans: Improvement Production and Uses (2nd edn.),

Burton, J. 1987. Quantitative genetics: results relevant to soybean breeding In: N
ASA, CSSA, and SSSA, Madison, WI (1987) Agron. Monogr. 16. 5‘

Buxton, D. and S. Fales. 1994. Plant environment and quality. p. 155-199. In G.
C. Fahey, Jr., M. Collins, D. R. Mertens, and L. E. Moser (eds). Forage Quality,
Evaluation, and Utilization. ASA-CSSA-SSSA, Madison, WI.

Calms, John. 2007. Corn for the starving or ethanol for fuel?: An ethical
dilemma for members of the automobile culture. Sci. 8 Soc. 5(2) 117-126.

Chen, C., G. Jackson, K. Neill, D. Wichman, G. Johnson, and D. Johnson. 2005.
Determining the feasibility of early seeding canola in the northern Great Plains.
Agron. J. 97:1252—1262.

Dardanelli, J., M. Balzarini, M. Martinez, M. Cuniberti, S. Resnik, S. Ramunda, R.
Herrero, and H. Baigorri. 2006. Soybean maturity groups, environments, and

their interaction deﬁne mega-environments for seed composition in Argentina.
Crop Sci 46: 1 939-1 947.

Dein, B, R. Bothast, L. lten, L. Barrios, and S. Eckhoff. 2002. Fate of Bt protein
and inﬂuence of corn hybrid on ethanol production. Cereal Chem. 79:582—585.

Dzantor, E., T., Chekol, and L., Vough. 2000. Feasibility of using forage
grasses and legumes for phytoremediation of organic pollutants. J. Environ.
Sci. Health. Part A 35:1645—1 661.

Goering, C., A. Schwab, M. Daugherty, E. Pryde, and A. Heakin. 1982. Fuel
properties of eleven oils. Trans. ASAE 25: 1472-1483.

129

 

Gomide, J., C. Noller, G. Mott, J. Conrad and D. Hill. 1969. Effect of plant age
and nitrogen fertilization on the chemical composition and in vitro cellulose
digestibility of tropical grasses. Agron. J. 61 :1 16-120.

Hammond, E.G. 1991. Organization of rapid analysis of lipids in many individual
plants. Springer-Verlag, Berlin.

Hill, J., E. Nelson, D. Tilman, S. Polasky, and D. Tiffany. 2006. Environmental,
economic, and energetic costs and beneﬁts of biodiesel and ethanol biofuels.
Proc. Natl. Acad. Sci. 103 (30): 11206-11210.

lzquierdo, N., L. Aguirrezabal, F. Andrade and V. Pereyra. 2002. Night
temperature affects fatty acid composition in sunﬂower oil depending on the
hybrid and the phenological stage. Field Crops Research 77:115—126.

Jackson, G. 2000. Effects of nitrogen and sulfur on canola yield and nutrient
uptake. Agron. J. 92:644—649.

Jiang, P. and K. Thelen. 2004. Effect of soil and topographic properties on crop
yield in a North-central com-soybean cropping system. Agron. J. 96:252-258.

Kincs, F. 1985. Meat fat formulation. JAOCS 62: 815-818.

Kimak, H., E. Dogan, M. Alpaslan, S. Celik, E. Boydak, and O. Copur. 2008.
Drought Stress Imposed at Different Reproductive Stages Inﬂuences Growth,
Yield and Seed Composition of Soybean. Philippine Agricultural Scientist 91 :3
261-268.

Lemuz, C.R., B.S. Dien, M.E. Tumbleson, V. Singh and K.D. Rausch. 2004.
Ethanol yield determination for dry grind corn processing. Proceedings corn
utilization and technology conference. pp61.

Ma, F. and M. Hanna. 1999. Biodiesel production: a review. Bioresource
Technology 70: 1-15.

Matthaus, B. and L. Bruhl. 2001. Comparison of different methods for the
determination of the oil content in oilseeds. JAOCS 78:1 95-102.

McLaughlin, S. and L. Kszos. 2005. Development of switchgrass (Panicum
virgatum L.) as a bioenergy feedstock in the United States. Biomass and
Bioenergy. 282515-535.

130

i i

I?”

 

Morrison, M., N. McLaughlin, E. Cober, and G. Butler. 2006. When is short-
season soybean most susceptible to water stress? Canadian Journal of Plant
Science. 86:5 Special Issue: Sp. lss. SI 1327-1331.

Moser, L. and K. Vogel. 1995. Switchgrass, big bluestem, and indiangrass.
Chapter 32. pp. 409—420. In R .F. Barnes, D. A. Miller, and C. J. Nelson (eds).
Forages, 5th ed. Vol.l: An introduction to grassland agriculture. Iowa State Univ.
Press, Ames, IA.

National Agricultural Statistics Service, United States Department of Agriculture.
Agricultural prices. October, 2008. Available at http:/lusda.mannlib.comell.edu
lusda/nass/AgriPric/l20003/2008/AgriPric—10-31-2008.pdf (veriﬁed 23 Feb. 2008). ,4

National Sunﬂower Association. 2007. Available at http:/lwww.sunﬂowemsa.org
(veriﬁed 23 Jan. 2009).

 

Natural Resources Conservation Service. 2009. Land Capability Classes.
Available at http:/lwww.mn.nrcs.usda.gov/technical/nri/ﬁndings/cropland_lcc.htm
(veriﬁed 07 Feb. 2009).

Nelson, R., S. Howell, and J. Weber. 1994. Potential feedstock supply and
costs for biodiesel production. Presented at the Sixth National Bioenergy
Conference, Reno/Sparks, Nevada. October 2-6. Available at
http:/lwww.biodiesel.org/resources/reportsdatabase/reports/genl1 9941006_gen-
290.pdf (veriﬁed 13 Feb. 2008).

Oklahoma State University Extension. 2007. Canola crop production. Available
at http:/[www.canola.okstate.edu/cropproduction/harvesting/index.htm (veriﬁed
07 Jan. 2008).

Patton, A. and L. Gieseker. 1942. Seasonal changes in the lignin and cellulose
content of some Montana Grasses. J. Anim Sci. 1:22-26.

Pryde, E. 1983. Vegetable oil as diesel fuel: Overview. JAOCS 60: 1557-1558.

Reicks, G., H. Woodard, and A. Bly. 2009. Improving the Fermentation
Characteristics of Com through Agronomic and Processing Practices. Agron. J.
101:201—206.

Rondanini, D., R. Savin and A.J. Hall. 2003. Dynamics of fruit growth and oil
quality of sunﬂower (Helianthus annuus L.) exposed to brief intervals of high
temperature during grain ﬁlling. Field Crops Res. 83:79—90.

[SAS] Statistical Analysis Systems. 2009. SAS/STAT User's Guide. Release
9.1.3. Cary, NC: Statistical Analysis Systems Institute.

131

Schmer, M., K. Vogel, R. Mitchell, and R. Perrin. 2008. Net energy of cellulosic
ethanol from switchgrass. PNAS 105 (2): 464-469.

Selvendran, R. and O'Neill, M. 1987. Isolation and analysis of cell walls from
plant material. In Methods of Biochemical Analysis, G. David, ed. John Wiley 8
Sons, pp. 25—153.

Shapouri, H., J. Dufﬁeld, and M. Graboski. 1995. Estimating the net energy
balance of corn ethanol. United States Department of Agriculture
Agricultural Economic Report Number 721.

Sokhansanj, S., M. Sudhagar, A. Turhollow, A. Kumar, D. Bransby, L. Lynd, and
M. Laser. 2009. Large-scale production, harvest and logistics of switchgrass
(Panicum virgatum L.) - current technology and envisioning a mature technology.
Biofuels, Bioprod. Bioref. 3: 1 24—1 41 .

Sosulskl, K., F. Sosulskl, and E. Coxworth. 1988. Carbohydrase Hydrolysis of
Canola to Enhance Oil Extraction with Hexane. Journal of the American Oil
Chemists' Society 65:3.

Stein, D. 2009. Crop enterprise budgets. Available at
http:/lwww.msu.edu/user/steind (veriﬁed 25 Feb 2009).

Summers, C., D. Haviland, and L. Godfrey. 2008. UC IPM Pest Management
Guildlines: Sugarbeet. UC ANR Publication 3469.

Swanson, S.P., Wilhelm, W.W., 1996. Planting date and residue rate effects on
growth, partitioning, and yield of corn. Agron. J. 88.

United States Department of Agriculture, Economic Research Service. 2003. Ag
Economic Report No. 823.

United States Department of Agriculture, Farm Service Agency. 1999. The
Conservation Reserve Program, 18th Signup. Washington, DC.

United States Department of Agriculture, National Agriculture Statistical Service.
- 1999. Census of Agriculture - 1997. Washington, DC.

Updegraf, D. 1969. Semimicro determination of cellulose in biological materials.
Anal. Biochem. 32:420 - 424.

Walsh, M., D. De La Torre Ugarte, H. Shapouri, and S. Slinsky. 2003.
Bioenergy crop production in the United States. Environmental and Resource
Economics 24: 313-333.

132

“52:21:82 .7. r2};

 

 

Wang, S., W. lngledew, K. Thomas, K. Sosulski, and F. Sosulski. 1999.
Optimization of fermentation temperature and mash speciﬁc gravity for fuel
alcohol production. Cereal Chemistry 76282-86.

York, W.S., Darvill, A.G., McNeil, T., Stevenson, T.T., and Albersheim, P. (1985)
Isolation and characterization of plant cell walls and cell wall components. Meth.
Enzymol. 11823-40.

Zheljazkov, V., B. Vick, M. Ebelhar, N. Buehring, B. Baldwin, T. Astatkie, and J.
Miller. 2008. Yield, Oil Content, and Composition of Sunﬂower Grown at
Multiple Locations in Mississippi. Agron. J. 100:635—642.

133

Chapter 4. Green Peach Aphid (Mus persicae Sulzer) Response to
Increased BioEnergy Crop Production

ABSTRACT

Increasing alternative crop production for biofuels could create agronomic
pest related issues for established cropping systems. Canola (Brassica napus L.
var. napus) is a promising oilseed crop anticipated to be used for biodiesel
production due to its high oil content, and in particular, monounsaturated oleic
acid. There is concern the insect pest green peach aphid (Myzus persicae
Sulzer) (GPA) due to its polyphagy and particularly due to its ability to serve as a
disease vector for the viruses alfalfa mosaic virus, beet necrotic yellow vein, beet
western yellows virus, cucumber mosaic virus, impatiens necrotic spot virus, and
tomato spotted wilt virus will negatively impact other established cropping
systems. Canola ﬁelds in Michigan as well as greenhouse studies were sampled
to determine if canola would serve as a host to GPA which can transmit various
viruses. In 2007, 2008 and 2009, GPA presence was determined by visual
observation and canola tissue samples were analyzed for the six listed viruses.
Visual observations did not detect GPA. Furthermore, canola tissue samples
were negative for the presence of all viruses regardless of location or fertility
regime. Additionally, a greenhouse experimental control with aphid cages also
detected no virus presence. These results suggest that with appropriate scouting
for GPA and potentially vectored viruses, canola production could be expanded

in the Great Lakes Region with minimal risk to other crop species.

134

INTRODUCTION

The public demand for renewable fuels and energy efﬁcient crops has
once again spurred interest in using canola (Brassica napus L.) for biodiesel
production in Michigan (Copeland, 1998). Canola was utilized during World War
II and the 1973 Energy Crisis, as an oilseed crop in biodiesel production
(Ramadhas et al., 2004). Canola seed content averages 40% oil and 23%
protein (Oplinger et. al, 1989) while its immediate counterpart soybean (Glycine
max (L.) Merrill) is inversely composed of 20% oil and 40% protein. Canola was
the ﬁrst crop, due to high oil content, used for transesteriﬁcation in the production
of biodiesel (Mattson et al., 2007). Soybean has been utilized for biodiesel
production in the US. due to its availability, ease of management and production
in the Midwest, as well as the nitrogen ﬁxation by legumes (Mattson et al., 2007).
Conversely, canola oil is better suited than soybean oil for production of quality
biodiesel as 60% of the oil is comprised of monounsaturated oleic acid and
demonstrates better stability and winter operability (Puppan, 2002). Canola is
the biggest source of feedstock for biodiesel production in the world and has
become more attractive as rapeseed breeders have improved yield levels to 2.9
ton oil ha", reported in Northern Germany (Puppan, 2002). Rapeseed, including
canola, contributes 84% of the raw material in the world used for production of :
biodiesel while soybean and sunﬂower only account for 1 and 13%, respectively
(Korbitz, 2000).

Renewable fuels, including biodiesel which can be produced from canola,

have potential to reduce green house gas emissions (GHG) and global warming

135

potential (GWP) (Grifﬁn and Lave, 2006; Sheehan et al., 1998). Fronning et al.
(2008) found fertilization regimes incorporating compost and manure
amendments in bioenergy crops can further improve their carbon footprint.
Bioenergy cropping systems which employ manure and compost may be better
suited for reduction of GHG, but may provide a suitable environment for
detrimental insect pests and plant pathogens due to high nitrogen levels
(Lampkin, 1990).

Green peach aphid (Myzus persicae Sulzer) (GPA) has been noted as a

 

worldwide pest in canola production, thriving in temperate regions, and is
considered the most important vector of plant vimses, as its distribution is
worldwide (Capinera, 2005). GPA is only one of 4,400 species of aphids in the
world and can survive in environments in which the primary host plants (Prunus
sp.) are absent. GPA are polyphagous (Weber, 1985), having the ability to feed
on a wide range of plants, and can survive on secondary hosts including
important agronomic crops such as Brassica, sugarbeet, sugar cane, tobacco,
tomato, and potato (van Emden et al., 1969; Blackman and Eastop, 2000).
Research conducted by Buntin and Raymer (1994) indicated GPA is a serious
pest of winter canola grown in Georgia. GPA mouthparts have evolved which
allow for ingestion of a liquid diet (Auclair, 1963), primarily phloem where most
viruses are typically found (Loebenstein, 2001). However, most plant damage
does not occur by GPA inﬂicting cellular damage with piercing sucking
mouthparts but rather by vector transmission (Moran and Thompson, 2001).

Buntin and Raymer (1994) found aphid injury can include reduced plant height,

136

 

winter survival, and delayed plant development. In the same three year study in
Georgia, they found GPA reduced oilseed yields of canola ranging from 9.9-
34.8%. Pilorge et al. (1997) has documented GPA causing signiﬁcant yield
losses in canola in France and Germany.

GPA has the potential to spread numerous infectious and yield limiting
viruses. There is concern the insect pest green peach aphid (GPA) due to its
polyphagy and particularly due to its ability to serve as a disease vector for the
viruses alfalfa mosaic virus, beet necrotic yellow vein, beet western yellows virus,
cucumber mosaic virus, impatiens necrotic spot virus, and tomato spotted wilt
virus will negatively impact other established cropping systems.

Alfalfa mosaic virus (AMV) is a virus which primarily infects alfalfa
however, it can be prevalent in other leguminous crops. Symptoms include
yellow mosaic or calico pattering on leaves and stunted plant growth (Koike and
Davis, 2007). The virus is spread by aphids and potentially GPA to other crops
including canola, chickpea, and faba bean (Schwinghamer, 2002). Studies have
indicated a reduction in fresh weight biomass yield from 5 to 30% in alfalfa with
reduction associated with impairment of plant physiology, decreased capacity of
nodulation, and an increased susceptibility to winterkill (Tu and Holmes, 1980).

Beet necrotic yellow vein (BNWV) is a virus transmitted in the soil by the
widely distributed fungus Polymyxa betae (Keskin, 1964). BNWV is responsible
for a sugar beet disease named rhizomania which consists of a rootlet
proliferation which induces a wine-glass-like shaped tap root with reduced sugar

contents (Duffus, 1986). Duffus also stated Rhizomania is the most damaging

137

 

sugar beet virus worldwide. The presence of GPA does not determine infection,
but will increase symptoms of the disease.

Beet western yellows virus (BWYV) can limit quality and total yield in
canola, sugar beets, and potatoes (Copeland, 1998). Symptoms include
yellowing of plant tissue resembling that of nutrient deﬁciency (Koike and Davis,
2007). Infection of canola by BWYV in combination with direct feeding damage
by GPA has the potential to reduce yield as much as 46% under maximum
infestation conditions. In the same study, seed oil content was also reduced by
as much as three percent (Jones et al., 1997). GPA is considered the most
important vector of BWYV (Hertel et al., 2004). Although BWYV is not known to
occur in Michigan, the possibility of its occurrence is still of great concern to
those in the Michigan sugarbeet industry (Copeland, 1998). BWYV has negative
implications for canola production. and is increasingly present as GPA
populations increase.

Cucumber mosaic vims (CMV) has a broad host range. The disease
affects a number of important vegetables and omamentals including tomato,
pepper, cucumber, melons, squash, spinach, celery, beets, and petunia
(Tisserat, 2006). Plants infected with CMV will develop a slight yellowing and
mottling of the older leaves. Expanding leaves typically become twisted, curl
downward, and develop a shoestring appearance. Symptoms of infection include
stunting and reduced yield and quality of crop. The disease is spread by aphids

and potentially GPA (Nameth, 1995).

138

Impatiens necrotic spot virus (INSV) can infect over 300 species of plants
in the US. Symptoms of INSV infection can include stunting, ring spots, brown
or purple spots on leaves or stems, stem browning (cankers), ﬂower breaking
and death. INSV is typically spread by the western ﬂower thlip (Windham et al.,
1998) however should be monitored for spread via GPA.

Tomato spotted wilt virus (T SWV) typically does not infect canola,
however it is a concern in other horticultural crops. Plant symptoms can include
many brown necrotic spots with the margins of leaves wilting and becoming
yellow (Koike and Davis, 2007). TSWV is typically spread by thrips however
should be monitored for spread via GPA.

GPA populations may be dependent on fertilizer regimes manipulated by
differing management strategies. Soluble nitrogen levels have been shown to
positively inﬂuence aphid growth and reproduction (Nevo and Coll, 2001).
Signiﬁcantly higher numbers of aphids were recorded in cabbage fertilized with
manure when compared to non-fertilized cabbage plants, which is believed to be
primarily due to nitrogen availability (Duchovskiene and Raudonis, 2008).
Yardim and Edwards (2003) found increased fertilizer use, either organic or
synthetic, led to an increase in aphid populations. Interestingly, they also
observed aphid numbers were lower in tomatoﬁelds fertilized with manure as an
organic fertilizer when compared to synthetic fertilizers. They concluded manure
amendments may have the ability to reduce aphid pest infestations over the long

term.

139

GPA populations can reach very high densities, particularly on young plant
tissue which under given conditions, can result in water stress, wilting, and
reduced growth rate of plants (Petitt and Smilowilz, 1982). They found prolonged
aphid infestation can cause appreciable reductions in potato yield and biomass
accumulation. Additionally, early season infestation can be particularly damaging
to potato in particular, even if aphids are subsequently removed. Blemishes on
plant tissue resulting from aphid feeding usually take the form of yellow spots and
leaf distortions, which are common only on the primary host.

Cellular damage is not the only repercussion of aphids. Honeydew is a
complex mixture of nutrients including many free amino acids and amides
secreted from the terminal opening of the aphid gut (Way, 1963). Contamination
of harvestable plant material can occur on foliage crops infested with aphids and
honeydew, also causing loss. In Arkansas, mild winters allow ovenlvintering
survival of green peach aphid on spinach, thereby leading to contamination
problems with honeydew (McLeod, 1991). Nevertheless, compared with other
aphid species, GPA does not produce high amounts of honeydew which can
support the growth of sooty mold, Ascomycete fungi.

Weber (1985) found GPA tolerates a wide array of different compounds in
its polyphagy on over 400 plant species which includes more than 50 families.
Additionally, over a three year period, GPA exhibited genetic variability in host
plants with high adaptability to sugar beets as well as potatoes. In 1982,
Radcliffe concluded GPA serves as the primary vector of three sugarbeet and

nine potato viruses and is the single most signiﬁcant pest of potato in Minnesota.

140

Narrower row spacing and in-row plant spacing of host crops also increases
aphid impact (San/var, 2008).

When increasing the total acreage of canola to meet energy needs, it is
necessary to consider the negative implications that may affect other cropping
systems. GPA is recognized as a pest of sugarbeet (Beta vulgaris L.),
(Summers, 2005), potato (Solanum tuberosum L.) (Radcliffe, 1982), and canola
(Brassica napus L. var. napus) (NDSU, 2009). Sugarbeets account for 25% of
the world’s supply of raw sugar (Rush et al., 2006). In the US, sugarbeets are
grown on 441,000 hectares with average annual yields of 27 million tons.
Sugarbeet is an established and substantial crop in Michigan, which is the fourth
leading producer of sugarbeets in the nation providing 110 million dollars in
revenue (United States Department of Agriculture, National Agriculture Statistics
Service, 2008). Potatoes are also an important cropping system in the US. In
2007, 428,000 hectares were devoted to potato production accruing 3.9 billion
dollars. The state of Michigan, though not a leading state in terms of potato
production, planted 14,400 hectares grossing 156 million dollars. Canola
acreage in the US. is currently 409,000 hectares with a national yield average of
1,697 kg ha'1 totaling 4 billion kilograms. Revenue generated from canola in the
US. totaled 276 million dollars. Approximately 94% of US. canola production is
concentrated in North Dakota, Minnesota, and Montana which also have
established sugarbeet and potato industries. Revenue generated from canola
grown in Michigan is insigniﬁcant on a national outlook (United States

Department of Agriculture, National Agriculture Statistics Service, 2008).

141

Understanding the disease triangle is important when considering the
affects of plant pests such as GPA on canola. The disease triangle is a
conceptual model which relates the connection between the presence of a
susceptible host, a disease causing organism, and ideal environmental or
weather conditions. If any one of these three factors are missing, no plant

diseases will be found and GPA presence will be insigniﬁcant.

US. food and energy independence could be attained by efﬁcient crop
selection and management for bioenergy production which could include
cultivation of canola for conversion to biodiesel. In this study, we sought to utilize
existing canola ﬁelds under producer management in addition to various
university canola trials in the state of Michigan to determine by visual and
laboratory analysis if GPA would serve as a detrimental insect pest to canola and

subsequently other established cropping industries.
MATERIALS AND METHODS

Canola leaf tissue was collected from ﬁve locations across the lower
peninsula of Michigan in 2007 and 2008. Sampling locations in 2007 included
Missaukee County, 44° 18’N, 85° 12’W, Ingham County, 42° 41’N, 89° 29’W
treated with and without anaerobic swine manure applications, Oakland County,
42° 43’N, 83° 37’W, and two locations in Leelanau County, 44° 59’N, 85° 39’W.
All locations were managed with fertilizer amendments except for the Ingham
County location which employed differing fertility regimes. In 2008, county

locations remained the same other than ﬁelds in Leelanau County which were

142

not sampled. Aphid presence was determined by visual observations in addition
to tissue sampling on 2 July 2007 and 30 June 2008. Time of sample collection
occurred when pod, stem, and leaf tissue were green in color and occurred
before the plant had reached morphological maturity using Munsell color charts
(Munsell, 1977). Sampling was delayed as long as possible to ensure adequate
time for infection. Sampling was conducted following a randomized sampling
strategy. The sampled leaves consisted of one leaf from the newest growth
centrally located on each canola plant. Composite samples for each site were

comprised of 20 representative leaves.

To supplement ﬁeld conditions, canola was also grown in a glass-glazed
greenhouse at Michigan State University, Ingham County, 42° 41’N, 89° 29’W in
2009. Canola was protected from pests using aphid cages. Canola grown
served as an experimental control to ensure no GPA interference. Leaf samples
were collected 35 d after planting, before canola had bolted. One composite
sample was collected from the greenhouse study and was comprised of 20

representative leaves.

Leaf tissue samples were processed by Agdia Incorporated (Elkhart, IN)
which offers services including crop speciﬁc analysis to detect the presence of
plant pathogens. The testing service does not have a protocol for canola viruses,
however analysis used for sugarbeet tissue samples were employed. The
sugarbeet protocol tested for the presence of six common viruses including:

AMV, BNYW, BWYV, CMV, INSV, and TSWV. Tissue sample results for virus

143

presence were compared against a positive control to determine a positive or

negative prognosis.
RESULTS AND DISCUSSION

Throughout the growing season and greenhouse growth period, insect
scouting indicated no presence of GPA. Rosettes, leaves, and stems did not
appear to have damaging physical symptoms. Additionally, no visual
conﬁrmation of GPA or subsequent honeydew was found on canola plants

throughout the growing season.

Viral analyses for AMV, BNYVV, BWW, CMV, INSV, and TSWV were all
negative compared to the positive control. Tissue sample analysis returned
negative results at all locations regardless of management practices, fertility

regime, or year.

The absence of GPA in canola cropping systems, regardless of
management practices, suggests the promise of canola to become a leading
feedstock for biodiesel production. Potentially, GPA could serve as a vector for
several detrimental viruses in canola, sugarbeet, and potato, however viruses

were not detected in this study.
CONCLUSIONS

GPA is a polyghagous insect pest which feeds on over 400 species of
plants encompassing 50 families (Weber, 1985). The ability to shift feeding

patterns to various plant species could potentially be troublesome as canola

144

acreage is increased for bioenergy feedstocks. Canola leaf tissue analyses did
not show a presence of any tested vimses. This research found GPA was not a

pest of canola as ﬁeld scouting did not determine GPA presence.

Research needs to be conducted to observe interactions between
potential bioenergy crops and their affect on traditional agronomic crops. As the
biofuel economy gains strength, research will be needed on evaluating
comparable crop diseases and relationships with traditional cropping systems.
The presence of GPA due to potential increases of canola production for
bioenergy production is going to continue to be a point of emphasis which will
need to be monitored. These results suggest that canola acreage could be
expanded in the Great Lakes Region with minimal risk to other crop species

provided scouting for GPA and potentially vectored viruses is implemented.

145

 

LITERATURE CITED
Auclair, J. 1963. Aphid feeding and nutrition. Annu. Rev. Entomol. 8:439-490.

Blackman, R. and Eastop, V. 2000. Aphids on the world’s crops: an
identiﬁcation and information guide. London, John Wiley and Sons Ltd.

Buntin, D. and P. Raymer. 1994. Pest Status of Aphids and Other Insects in
Winter Canola in Georgia. Journal of Economic Entomology 87: 4 1097-1104.

Capinera, J. 2005. Myzus persicae (Sulzer): green peach aphid. Available at
http:/lentomology.ifas.uﬂ.edu/creatures/veg/aphid/green_peach_aphid.htm
(veriﬁed 16 March 2009).

Copeland, L. 1998. Nontraditional ﬁeld crops: canola/rapeseed and lupins.
Michigan State University Extension Ag Experiment Station Special Reports -
SR559201.

Duchovskiene, L. and L. Raudonis. 2008. Seasonal abundance of Brevicoryne
brassicae L. and Diaeretiella rapae (M’lntosh) under different cabbage
growing systems. 54 (4): 260-264.

Duffus, J. 1986. Rhizomania (Beet Necrotic Yellow Vein) Pages 29-30, In:
Compendium of Beef Diseases and Insects. APS Press, St. Paul, MN.

Fronning, B., K. Thelen, and D. Min. 2008. Use of manure, compost, and cover
crops to supplant crop residue carbon in corn stover removed cropping systems.
Agron. J. 1002 1703-1710.

Grifﬁn, W. and L. Lave. 2006. Cellulosic ethanol in an oil and carbon
constrained world. P. 21-32. In A high growth strategy for ethanol. The Aspen
Inst, Washington, DC.

Hertel, K., M. Schwinghamer, and R. Bambach. 2004. Virus diseases in
canola and mustard. Agnote DPI 495, lst edition.

Jones, R., B. Coutts, and J. Hawkes. 2007. Yield-limiting potential of beet
western yellows virus in Brassica napus. Australian Journal of Ag. Research.
58(8): 788-801.

Keskin, B. 1964. Polymyxa batae n.sp., a parasite in the roots of Beta vulgaris
toumefort, particularly during the early growth of the sugarbeet. .Arch Mikrobiol
1 9: 348-374.

Koike, S. and R. Davis. 2007. UC IPM Pest Management Guidelines. UC ANR
Publication 3450. Available at http:/Iwww.ipm.ucdavis.edu/PMG/r441101411.
html (veriﬁed 01 May 2009).

146

Korbitz, W. 2000. World-wide trends in production and marketing of biodiesel,
Proc. ALTENER Seminar, University of Technology in Graz, Graz.

Lampkin, N. 1990. Organic Farming. Farming Press Books, Ipswich, UK.

Loebenstein, G., P. Berger, A. Brunt, and R. Lawson. 2001. Virus and Virus-like
Diseases of Potatoes and Production of Seed-potatoes. Published by Springer.

Mattson, J., W. Wilson, C. Duchsherer. 2007. Structure of the Canola and
Biodiesel Industries. Agribusiness 8 Applied Economics Report No. 606.
Available at http:/Iageconsearch.umn.edu/bitstream/7644/1/aer606.pdf (veriﬁed
23 March 2009).

McLeod, P. 1991. Inﬂuence of temperature on translaminar and systemic
toxicities of aphicides for green peach aphid (Homoptera: Aphididae)
suppression on spinach. Journal of Economic Entomology. 84:1558-1561.

Moran, P. and G. Thompson. 2001. Molecular responses to aphid feeding in
Arabidopsis in relation to plant defense pathways. Plant Physiol. 125 1074-1085.

Munsell, AH. 1977. Munsell color charts for plant tissues. 2nd ed. Macbeth Div.
of Kollmorgan Corp., Baltimore, MD.

Nameth, S. 1995. Mosaic Virus Diseases of Vine Crops. Ohio State University
Extension Bulletin HYG-3109-95. ’

Nevo, E. and M. Coll. 2001. Effect of nitrogen fertilization on Aphis gossypii
(Homoptera: Aphididae): variation in size, color, and reproduction. Journal of
Economic Entomology 94(1):27-32.

North Dakota State University Extension. 2009. Aphids in canola. Available at
http:/Iwww.ag.ndsu.edu/procrop/rpslaphidsOB.htm (veriﬁed 03 March 2009).

Oplinger, E., L. Hardman, E. Gritton, J. Doll, and K. Kelling. 1989. Alternative
ﬁeld crops manual, canola (rapeseed). Ext. Bull. Nov. 1989. Univ. of Wisconsin,
Madison, WI.

Petitt, F. and Z. Smilowitz. 1982. Green peach aphid feeding damage to potato
in various plant growth stages [Myzus persicae]. Journal of Economic
Entomology. 75(3): 431-435.

Pilorge, E., C. Maisonneuve, and Y. Balanger. 1997. Les ravageurs du colza
d’hiver. Les points techniques du CETIOM. Edition CETIOM, Paris, France.

Puppan, D. 2002. Environmental evaluation of biofuels. Periodica Polytechnica
Ser. Soc. Man. Sci. 10:1 95—116.

147

 

Radcliffe, E. 1982. Insect pests of potato. 'Ann Rev. Entomol. 27: 173-204.

Ramadhas, A., S. Jayaraj, and C. Muraleedharan. 2004. Use of vegetable oils
as I.C. engine fuels - a review. Renewable Energy. 29:5 727-742.

Rush, C., H. Liu, and R. Lewellen. 2006. The continuing saga of Rhizomania of
sugar beets in the United States. Plant Disease 90: 4-15.

San/var, S. 2008. Plant spacing - a non polluting toot for aphid (Hemiptera:
Aphididae) management in canola, Brassica napus. Journal of Entomological
Society of Iran. 27 (2): 13—22.

Schwinghamer, M. 2002. Virus situation in chickpea, faba bean and canola.
Available at http:/Iwww.grdc.com.au/director/events/onfarmtrials?item_id=
4829BBOCA9AC5A1090F7A53B484E59058pageNumber=2 (veriﬁed 01 May
2009)

Sheehan, J., V. Camobreco, J. Dufﬁeld, M. Graboski, and H. Shapouri. An
Overview of Biodiesel and Petroleum Diesel Life Cycles. National Renewable
Energy Laboratory, US. Department of Energy. Golden, CO. May 1998.

Summers, C., D. Haviland, and L. Godfrey. 2005. UC IPM Pest Management
Guidelines: Sugarbeet. UC ANR Publication 3469. Revised. University of
California [Online]. Available at
http:/Iwww.ipm.ucdavis.edu/PMG/r735300211.html (Veriﬁed 06 Jan. 2008).

Tisserat, N. 2006. Virus diseases of tomato. Kansas State Research and
Extension Publication L-723.

Tu, J. and T. Holmes. 1980. Effect of alfalfa mosaic virus infection or nodulation,
forage yield, forage protein, and over-wintering of alfalfa. Phytopath. 2.97: 1-9.

United States Department of Agriculture, National Agriculture Statistics Service.
2008. Sugarbeets: National statistics. Available at http:/Mww.nass.usda.gov
/QuickStats/index2.jsp (veriﬁed 04 May 2009).

van Emden, H., Eastop, V., Hughes, R., and Way, M. 1969. The ecology of
Myzus persicae. Annual Review of Entomology. 14: 197-270.

Way, M. 1963. Mututalism between ants and honeydew producing Homoptera.
Annu. Rev. Entomol. 82307-344.

Weber, G. 1985. Genetic variability in host plant adaptation of the green peach
aphid, Myzus persicae. Entomol. exp. appl. 38: 49-56.

148

Windham, A.‘, F. Hale, J. Yanes. 1998. Impatiens Necrotic Spot Virus

A Serious Pathogen of Floral Crops. The University of Tennessee Extension
Bulletin SP370A.

Yardim, E. and C. Edwards. 2003. Effects of organic and synthetic fertilizer
sources on pest and predatory insects associated with tomatoes.
Phytoparasitica. 31: 324-329.

149

 

mm“ Illlll Illllllllllllllllll llllllllllllss
3 1293 03062 6315