(Advances in agronomy 117) donald l sparks (eds ) advances in agronomy 117 academic press, elsevier (2012)

ADVANCES IN AGRONOMY
Advisory Board

PAUL M. BERTSCH

RONALD L. PHILLIPS

University of Kentucky

University of Minnesota

KATE M. SCOW

LARRY P. WILDING

University of California, Davis

Texas A&M University

Emeritus Advisory Board Members

JOHN S. BOYER

KENNETH J. FREY

University of Delaware

Iowa State University

EUGENE J. KAMPRATH

MARTIN ALEXANDER

North Carolina State, University

Cornell University

Prepared in cooperation with the
American Society of Agronomy, Crop Science Society of America, and Soil Science
Society of America Book and Multimedia Publishing Committee
DAVID D. BALTENSPERGER, CHAIR
LISA K. AL-AMOODI
WARREN A. DICK
HARI B. KRISHNAN
SALLY D. LOGSDON

CRAIG A. ROBERTS
MARY C. SAVIN
APRIL L. ULERY


VOLUME ONE HUNDRED SEVENTEEN

ADVANCES IN
AGRONOMY

EDITED BY
DONALD L. SPARKS
Department of Plant and Soil Sciences
University of Delaware
Newark, Delaware, USA

AMSTERDAM • BOSTON • HEIDELBERG • LONDON
NEW YORK • OXFORD • PARIS • SAN DIEGO
SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO
Academic Press is an imprint of Elsevier


Academic Press is an imprint of Elsevier
525 B Street, Suite 1900, San Diego, CA 92101-4495, USA
225 Wyman Street, Waltham, MA 02451, USA
32 Jamestown Road, London, NW1 7BY, UK
The Boulevard, Langford Lane, Kidlington, Oxford, OX51GB, UK
Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands
First edition 2012
Copyright Ó 2012 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system or transmitted in
any form or by any means electronic, mechanical, photocopying, recording or otherwise
without the prior written permission of the publisher
Permissions may be sought directly from Elsevier's Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email:
Alternatively you can submit your request online by visiting the
Elsevier web site at and selecting Obtaining permission
to use Elsevier material
Notice
No responsibility is assumed by the publisher for any injury and/or damage to persons or
property as a matter of products liability, negligence or otherwise, or from any use or
operation of any methods, products, instructions or ideas contained in the material herein.
Because of rapid advances in the medical sciences, in particular, independent verification of
diagnoses and drug dosages should be made
ISBN: 978-0-12-394278-4
ISSN: 0065-2113 (series)

For information on all Academic Press publications
visit our website at store.elsevier.com
Printed and bound in USA
12 13 14 15
10 9

8 7

6 5

4 3

2

1


CONTRIBUTORS
Numbers in parantheses indicate the pages on which the authors’ contributions begin.

Virupax C. Baligar (117)
USDA-ARS – Beltsville Agricultural Research Center, Beltsville, MD, USA.
Catherine Bonin (1)
The Ohio State University, School of Environment and Natural Resources, Carbon
Management and Sequestration Center, Columbus, OH, USA.
Rufus L. Chaney (51)
Senior Research Agronomist, USDA-Agricultural Research Service, Environmental
Management and Byproducts Utilization Lab, BARC-West Beltsville, MD, USA.
Bhagirath S. Chauhan (315)
International Rice Research Institute (IRRI), Los Ba~

nos, Philippines.
Zhenli He (117)
Indian River Research and Education Center, Institute of Food and Agricultural Sciences,
University of Florida, Fort Pierce, FL, USA.
Shanying He (117)
Indian River Research and Education Center, Institute of Food and Agricultural Sciences,
University of Florida, Fort Pierce, FL, USA; College of Environmental Science and
Engineering, Zhejiang Gongshang University, Hangzhou, China.
Ram A. Jat* (191)
International Crops Research Institute for the Semi-Arid Tropics, Patancheru, Andhra
Pradesh, India.
Mangi L. Jat (315)
International Maize and Wheat Improvement Center (CIMMYT), New Delhi, India.
Rattan Lal (1)
The Ohio State University, School of Environment and Natural Resources, Carbon
Management and Sequestration Center, Columbus, OH, USA.
Gulshan Mahajan (315)
Punjab Agricultural University, Ludhiana, Punjab, India.
S. P. Milroy (275)
CSIRO Plant Industry, Wembley, Western Australia, Australia.
M. L. Poole (275)
CSIRO Plant Industry, Wembley, Western Australia, Australia.
* Current address: Directorate of Groundnut Research, Junagdh, Gujarat, India

ix


x

Contributors

M. M. Roper (275)
CSIRO Plant Industry, Wembley, Western Australia, Australia.
Kanwar L. Sahrawat (191)
International Crops Research Institute for the Semi-Arid Tropics, Patancheru, Andhra
Pradesh, India.
Virender Sardana (315)
Punjab Agricultural University, Ludhiana, Punjab, India.
Jagadish Timsina (315)
International Rice Research Institute (IRRI), Los Baños, Philippines.
Suhas P. Wani (191)
International Crops Research Institute for the Semi-Arid Tropics, Patancheru, Andhra
Pradesh, India.
Xiaoe Yang (117)
Ministry of Education Key Laboratory of Environment Remediation and Ecosystem Health,
College of Environmental and Resources Science, Zhejiang University, Zijingang Campus,
Hangzhou, China.


PREFACE
Volume 117 contains six excellent reviews that focus on some of the most
prominent global issues of our time- energy, environment, and food
production. Chapter 1 is a timely and comprehensive review of the agronomic and ecological implications of biofuel production that includes
impacts on land use, soil erosion and water quality, nutrient cycling, and
biodiversity. Chapter 2 addresses food safety issues related to mineral and
organic fertilizers with emphasis on the presence of trace metals. Chapter 3
reviews the mechanisms of nickel uptake and hyperaccumulation by plants,
and impacts related to soil remediation. Chapter 4 deals with conservation
agriculture in the semi-arid tropics, with respect to environmental,
economic, and ecological opportunities and challenges. Chapter 5 is an

overview of the use of green and brown manures in dryland wheat
production systems in Mediterranean-type environments. Chapter 6 is
focused on the benefits, disadvantages, and strategies for enhancing the
productivity and sustainability of rice-wheat cropping systems in the IndoGangetic Plains region of the Indian subcontinent.
I am most grateful to the authors for their first-rate reviews.
DONALD L. SPARKS
Newark, Delaware, USA

xi


C H A P T E R O N E

Agronomic and Ecological
Implications of Biofuels
Catherine Bonin and Rattan Lal
Contents
1. Introduction
2. Ecosystem Functions and Services
3. Land Use Change
3.1. Greenhouse gas emissions from land conversion
3.2. Using land enrolled in the Conservation Reserve Program
3.3. Biofuels and restoration of degraded lands
4. Soil Erosion and Water Quality
4.1. Use of agricultural residues
5. Nitrogen Cycling
5.1. Nitrogen and litter/residue management
5.2. Nitrogen uptake and biomass removal
5.3. Gaseous emissions and volatilization
6. Human Impacts on Biodiversity

6.1. Biodiversity in agroecosystems
6.2. Diverse perennial grasslands
6.3. Effects on wildlife
6.4. Diversity at the landscape level
6.5. Pests and biocontrol
7. Biofuels and the Soil Carbon Budget
7.1. Land/soil preparation
7.2. Soil carbon budget
8. Invasive Potential of Bioenergy Crop Species
8.1. Invasive species as feedstock
9. Food versus Fuel
10. Conclusions
11. Future Challenges
Acknowledgments
References

2
6
8
9
10
12
13
16
17
19
20
21
22
22

23
24
25
26
28
28
29
30
32
32
34
36
37
37

The Ohio State University, School of Environment and Natural Resources, Carbon Management
and Sequestration Center, Columbus, OH, USA
Advances in Agronomy, Volume 117
ISSN 0065-2113,
DOI: />
Ó 2012 Elsevier Inc.
All rights reserved.

1


2

Catherine Bonin and Rattan Lal

Abstract
Biofuels can be alternative energy sources which simultaneously reduce dependence on fossil fuels and mitigate climate change by reducing greenhouse gas
(GHG) emissions. In the US, over 50 billion liters of ethanol produced in 2010 is
mandated to increase to 136 billion liters by 2022. Globally, approximately
33.3 million ha (Mha) of land under production of biofuels in 2008 may increase
to as much as 82 Mha by 2020. Whereas data on the energy efficiency and GHG
balances for biofuels are available, information on agronomic and ecological
consequences of large-scale production of bioenergy crops is sparse. Thus,
this paper describes the potential effects that bioenergy production may
have on ecosystems. Conversion of land to biofuel crops may have significant
impacts on ecosystem services such soil and water quality, GHG emissions,
wildlife habitat, net primary productivity, and biological control, and plant
diversity at both the landscape and the regional levels. Production of exotic
species for feedstock may increase the risk of escape from agriculture and invasion into natural ecosystems. Several feedstocks, while suitable on the basis of
energy and GHG assessments, may have negative ecosystem impacts (i.e.,
increased N export in the Gulf of Mexico). Bioenergy feedstock may compete
with food crops for land, water, and nutrient resources, resulting in higher prices for food as well as potential increases in malnutrition and food insecurity.
Biofuels can be a sustainable and renewable source of energy, but assessments
must include ecological impacts, economic costs, and energetic efficiencies.

1. Introduction
Biofuels are widely considered as renewable and sustainable alternatives to fossil fuels. They are touted as an energy production system that
can have both a positive energy balance while offsetting greenhouse gas
(GHG) emissions through carbon (C) sequestration. The US set an auspicious goal of supplying the equivalent of 30% of the nation's petroleum use
from biomass, requiring 900 M Mg (1 Mg ¼ 106 g ¼ 1 metric ton) to 1
billion Mg of feedstock, and predict that the generation of this extra
biomass will be based on increases in crop yields and changes in land use
(Perlack et al., 2005; Somerville, 2006). In terms of bioethanol production,
corn (Zea mays L.) grain is presently the most common source in the US,
with over 50 billion liters produced by 189 plants in 2010 (Fig. 1;

Renewable Fuels Association, 2011).
However, corn grain will likely be only a temporary feedstock option
due to land limitations: even if all corn produced in the US were used
for ethanol production, it would only supply 12% of US gasoline needs
(Hill et al., 2006). In addition, the Energy Independence and Security
Act of 2007 (EISA), which mandates that 136.3 billion liters of
renewable fuels be produced annually by 2022, has capped contributions


3

Agronomic and Ecological Implications of Biofuels

50000

200

Ethanol Produced
Ethanol Plants

150

40000
30000

100

20000

Ethanol Plants

Ethanol Produced (10 6 L)

60000

50

10000
0
1975

0

1980

1985

1990

1995

2000

2005

2010

Year

Figure 1 United States ethanol production and ethanol plants. Data modified from the

Renewable Fuels Association (2011).

from corn grain at 56.8 billion liters (Sissine, 2007). Therefore, a variety of
other bioenergy feedstocks are also being promoted for the development of
second generation biofuels such as perennial grasses, woody species, and
agricultural residues (Table 1).
In the decade ending in 2010, bioenergy feedstocks have undergone
intense scrutiny and evaluation to determine the net energy yields and
GHG balances through the use of life cycle assessment (LCA). Even though
the primary goal of biofuels is to provide energy with a low C footprint,
LCAs show that not all biofuels are created equally in terms of energy
and GHG fluxes (Adler et al., 2007; Davis et al., 2009). These assessments
suggest that the varying results in energy and GHG balances may be caused
by differences in species attributes, crop production practices, land use
changes, and conversion technologies (Fargione et al., 2008; Huang et al.,
2009).
Corn grain is a feedstock within the first generation of biofuels, which are
fuels derived from plant sugars, starches and oils (Soetaert and Vandamme,
2009). Other first generation feedstocks include sugarcane (Saccharum
officinarum L.), oil palm (Elaeis guineensis Jacq.), and soybean (Glycine max
(L.) Merr.). Primary feedstock options vary by country: the US uses corn
grain, Brazil relies on sugarcane, the European Union uses wheat (Triticum
aestivum L.) and sugar beet (Beta vulgaris L.), while both China and Canada
use wheat and corn grain (Balat and Balat, 2009). Examination of nine
first generation feedstocks based on ecological, energy, and GHG emission
parameters suggest that tropical species such as sugarcane and oil palm may
be a better option than temperate species such as corn and wheat in terms


Table 1

Biofuel feedstock production and ecological summary

a

Stand persis- Fertilizer
tence (years) need

Stress
tolerance

Ecological
concerns

3

>20

low

>20

invasive potential in
western US

3

low

drought and

flood
drought

3

>20?

low

1e2

>10?

medium

drought and
flood

invasive

2e3

>10?

medium

saline soils

invasive

1

e

high

1
3þ

e
>20

high
low/
medium

1e2

5

high

risk of soil erosion and N
leaching

3

25

2e3

50

medium/
high
medium

plantations may cause
deforestation
plantations may cause
deforestation

may lower water quality
flood

drought

may lower soil quality

Oil palm yield in fresh fruit bunch weight (FFB); crude oil conversion ratio of 0.18 kg oil per 1 kg FFB (Papong et al., 2010).
Sources: Achten et al. (2007); Achten et al. (2008); Angelini et al. (2009); Djomo et al. (2011); Fillion et al, (2009); Hartemink (2008); Heaton et al. (2008);
Hoskinson et al. (2007); Lewandowski et al. (2003); Linderson et al. (2007); Openshaw (2000); Papong et al. (2010); Parrish and Fike (2005); Tilman et al.
(2006); USDA-NASS (2011); Whan et al. (1976).

Catherine Bonin and Rattan Lal

Switchgrass
15
(Panicum virgatum)
Native grass mix

3.6 (degraded)
6.0 (fertile)
29.6
Miscanthus
(Miscanthus Â
giganteus)
Reed canarygrass 9.0
(Phalaris
arundinacea)
Giant reed
37.7
(Arundo donax)
Corn (grain)
9.3
(Zea mays)
Corn (stover)
5.1
11.7
Woody species
(Salix and
Populus spp.)
82
Sugarcane
(Saccharum
officinarum)
Oil palm (Elaeis
18.7a
guineensis)
Jatropha ( Jatropha 5.0
curcas)

Time to
establish
(years)

4

Bioenergy species

Yield
(Mg haL1)


Agronomic and Ecological Implications of Biofuels

5

of sustainability (de Vries et al., 2010). Corn, while relatively high-yielding
and with a high ethanol productivity, also requires large amounts in inputs
(i.e., fertilizers) which reduce its net energy ratio (NER, energy
outputs:energy inputs) to about 1.7 (Liska et al., 2009). Most temperate
annual biofuel feedstock (i.e., cereal grains, sugar beet) have NERs that
range between 1 and 4 (Venturi and Venturi, 2003). In contrast,
sugarcane, which can yield over 80 Mg haÀ1, has an NER of 9.2
(Hartemink, 2008). Although sugarcane's energy production indicates it to
be a desirable biofuel feedstock, as with other first generation choices, its
use as a fuel source directly affects food availability and brings it into
competition with other food crops for land and other resources (Pimentel
and Patzek, 2008).
Second generation feedstocks are suggested as a solution to the problems

caused by first generation options and as a way to reduce competition with
food crops for limited resources. Second generation, or lignocellulosic feedstocks, include perennial grasses, trees, and agricultural and forestry residues
or wastes. Because the entire plant can be used instead of just the grains,
sugars, or fats, energy yields per hectare can be much larger. In addition,
grass and tree feedstocks may be grown on marginal lands with fewer
inputs, which can reduce GHG emissions and energy requirements
(Debolt et al., 2009). Switchgrass (Panicum virgatum L.) may have an
NER as large as 13.1 and can reduce GHG emissions by 95% when
compared to gasoline, while miscanthus (Miscanthus x giganteus) can
produce 2.6 times more ethanol per hectare than corn grain can and has
a potential NER of 12e66 (Heaton et al., 2008; Schmer et al., 2008;
Venturi and Venturi, 2003). Tree species such as willows (Salix spp.) and
poplars (Populus spp.) can produce 11.5 Mg haÀ1 or more of biomass
each year and may have NERs of over 20 when burned for electricity
(Djomo et al., 2011). Residues and waste products, as byproducts of
agricultural and anthropogenic activities, would not require any
additional lands or resources for production. The NER for corn stover of
approximately 2.2 is lower than that of some other second-generation
feedstocks; however, it is important to note that the corn grain may be
used for other purposes (Luo et al., 2009). Nonetheless, excessive residue
removal may reduce agronomic yield and degrade soil quality by
accelerating soil erosion and depleting soil C levels (Blanco-Canqui et al.,
2006; Lindstrom, 1986). Thus, residue removal rates must be balanced
with agronomic and environmental requirements. Other secondgeneration feedstocks include jatropha ( Jatropha curcas L.), which is
a drought resistant tree that produces non-edible oils from its seeds that
could reclaim eroded or degraded land (Openshaw, 2000). It can be
high-yielding under the appropriate environmental conditions, but more
work needs to be done to determine the potential of this plant
(Trabucco et al., 2010). A third generation of biofuels, microalgae and

6

Catherine Bonin and Rattan Lal

cyanobacteria, are being developed for the production of biodiesel (Sayre,
2010). Microalgae are very productive and would be land efficient,
although the input costs for water and energy could be high, reducing
the NER to close to or less than one (Batan et al., 2010; Clarens et al.,
2011).
While much research has focused on the energy and GHG aspects of
biofuels, other questions on the sustainability of biofuel production in terms
of ecosystem functioning and services are less commonly addressed. There
may be significant repercussions to many ecosystem properties and services
if there is a large-scale land-use change to bioenergy crop feedstock production. However, the ecological and agronomic implications of large-scale
bioenergy cropping systems are not fully understood. Both feedstock
species and management practices will likely affect ecosystem properties
differently. Future growing demands will increase costs and reduce the
availability of vital resources such as arable land, water, nutrients, and pesticide inputs. If bioenergy systems are to be sustainable alternatives to fossil
fuels, there must be an equilibrium between energy yields and the impacts
on soil quality, wildlife habitat, nutrient cycling, and water cycling (Clarens
et al., 2010; Lardon et al., 2009).
Although all three generations of biofuels have advantages and disadvantages in terms of energy balances, GHG emissions, net primary productivity (NPP), and environmental quality, this paper focuses primarily on first
and second-generation feedstocks. The principal objective of this paper is to
discuss the potential consequences that biofuels have on agricultural and
ecological processes and services, and identify gaps in scientific knowledge
that must be addressed in order to understand the full ecological implications of biofuels. Readers are referred to other reviews that discuss specific
issues related to energy and GHG balances for biofuels (Adler et al., 2007;
Davis et al., 2009; Hill et al., 2006).

2. Ecosystem Functions and Services
Ecosystem functions and services are related, but different.
Ecosystem functions are described as the processes within an ecosystem,
while ecosystem services are the benefits that humans receive from
ecosystem functions (Costanza et al., 1997). A single ecosystem function
may provide more than one service: for example, the function “soil
retention” may both maintain farmland and also prevent soil erosion
(de Groot et al., 2002). Examples of ecosystem services include water
and climate regulation, soil formation, nutrient cycling, and food
production (Table 2). These ecosystem services have a global value of
an estimated $33 trillion yrÀ1 (Costanza et al., 1997), but many of these


Agronomic and Ecological Implications of Biofuels

7

Table 2 Ecosystem services and related processes, modified from the Millennium
Ecosystem Assessment (2005) and de Groot et al. (2002).
Type

Service

Supporting

Soil formation

Regulating

Provisioning

Cultural

Process

Rock weathering; accumulation of
organic matter
Primary production Transfer of solar energy into plant and
animal biomass
Nutrient cycling
Storage and recycling of nutrients
Water cycling
Regulation of runoff and water
discharges
Climate
Interactions of biotic and abiotic
elements as well as land cover
Air quality
Regulation of pollutants
Flood
Regulation of runoff and water
discharges
Disease
Regulation of disease populations
through biotic interactions
Waste treatment
Removal of wastes by organisms
Pest control
Regulation of pests by predators and
biotic interactions

Pollination
Movement of floral gametes by biota
Food
Transfer of solar energy into plant and
animal biomass
Clean water
Filtering and storage of fresh water
Raw materials:
Transfer of solar energy into plant and
wood, fiber, etc.
animal biomass
Fuel
Transfer of solar energy into plant and
animal biomass
Genetic resources Genetic variability of wild resources
Biochemicals,
Variation and uses for medicinal plants
pharmaceuticals
Ornamentals
Variation in flora for aesthetic uses
Recreational,
Variation in landscape for human
ecotourism
relaxation
Spiritual
Variation in landscape for spiritual and
religious needs
Educational
Variation in landscape for educational
purposes

ecosystem services are at risk. Fifteen of the 24 ecosystem services
evaluated by the Millennium Ecosystem Assessment (2005) are being
degraded or are in decline. Human modification of lands through
farming and urbanization may affect food production, water and air


8

Catherine Bonin and Rattan Lal

quality, climate control, and the availability of natural resources (Foley
et al., 2005; Hooper et al., 2005; Tilman et al., 2001). In addition,
decreasing biodiversity may further impact ecosystem service declines
(Loreau et al., 2001).

3. Land Use Change
Humans have been altering the appearance of the landscape for thousands of years and bioenergy cropping systems are likely to accelerate the
change (Meyer and Turner, 1992; Vitousek et al., 1997). As human populations have increased, so has alteration of the landscape (Fig. 2). In 1700,
only 5% of land was under urban settlements, with nearly 95% of land
existing in natural or semi-natural states, but by 2000 over one-half of
global land was dominated by humans, to form a new biome, the
“anthrome,” one dominated by anthropogenic activities (Ellis et al.,
2010). Between 1850 and 1990, the global agricultural land area
quadrupled to 1360 Mha, with land use change (LUC) for agricultural

a)
Natural Vegetation

Population (millions)

b)

1700

1850

1990

Forest/Shrubland
Grassland
Desert
Agriculture

10000
8000
6000
4000
2000
0
1700 1750 1800 1850 1900 1950 2000 2050 2100

Year

Figure 2 a) Changes in land cover proportion from natural vegetation through 1990.
Total land area is estimated at 134.1 million km2, and b) increase in world population
(in millions) from 1700 through 2011, and predictions (dashed line) through 2100. Data
from Goldewijk (2005), Goldewijk (2001), and the United Nations, Department of
Economic and Social Affairs, Population Division (2011).

Agronomic and Ecological Implications of Biofuels

9

activities releasing approximately 105 Pg C (petagram ¼ 1015 g ¼ 1 GT) by
vegetation and soil C losses (Houghton, 1999).
The amount of land in the US planted to corn will remain at recordhigh levels of 37 Mha due to high demand for ethanol and corn exports
and approximately 35e40% of the US corn crop will be used for ethanol
production (USDA, 2010). Increases in ethanol production may come from
either increases in feedstock yields or increases in land under feedstock
production. An estimated 30% of the predicted increase in future corn
production may be due to higher yields, which would be a preferred
method of ethanol production increases, as it would not cause increased
competition for land resources (Keeney and Hertel, 2009). However,
yield gains can only increase production to a certain point, and once
reached, future increases in feedstock production may result in LUC. In
2008, 33.3 Mha were under biofuel production worldwide, but by 2020
the land area is expected to increase to as much as 82 Mha (Fargione
et al., 2010). In the US, cropland, pasture, and idle land area is expected
to decrease, while area for perennial crops will increase by as much as
20.2 Mha (Perlack et al., 2005). Conversion of a large area of land to
feedstock production will have repercussions on GHG emissions as well
as a variety of ecosystem processes such as C and N cycling, water
quality, and wildlife habitat.

3.1. Greenhouse gas emissions from land conversion
The impact that biofuels have on GHG emissions has been widely
addressed by several papers (Adler et al., 2007; Hoefnagels et al., 2010; Whitaker et al., 2010) and is briefly outlined here in terms of how LUC affects
emissions. Production of bioenergy feedstocks may affect LUC directly or

indirectly. Direct LUC emissions stem from the conversion of land to feedstock plantations. Indirect emissions occur as a result of additional land
conversion that occurs from the land uses or ecosystems displaced by the
bioenergy plantation (Melillo et al., 2009; Searchinger et al., 2008).
Increased bioethanol demand in the US will affect the allocation of US
land area under other crops and ecosystems, but will also indirectly cause
changes in land uses around the globe (Keeney and Hertel, 2009). Direct
LUC impacts on GHG emissions and their corresponding C debt may
be calculated by estimating loss of soil C stock due to land conversion
(Fargione et al., 2008), but calculating indirect effects is more
challenging. Indirect LUC is a large source of uncertainty when assessing
the impacts of bioenergy crops, and due to this uncertainty, current
predictions may be underestimating indirect emissions by as much as
140% (Plevin et al., 2010).
Direct LUC can have significant impacts on GHG emissions, as land
conversion results in a large initial release of GHG (i.e., a “C debt”) that


10

Catherine Bonin and Rattan Lal

may take decades or centuries to recover (Fargione et al., 2008). Without
taking LUC into account, some corn and cellulosic ethanol production
systems show predicted 20% and 70% reductions in GHG compared to
gasoline, but after factoring in LUC, these two ethanol systems may
actually release 93% and 50% more GHGs (Searchinger et al., 2008), but
these calculations have been debated (Mathews and Tan, 2009).
Although biofuel feedstock may sequester C and reduce emissions, the
initial costs of land conversion may make these systems more costly in C
terms for a significant period of time, particularly when converting lands

not currently under annual crop production. Set-aside lands, such as
those enrolled in the Conservation Reserve Program (CRP), may
sequester more C than would their conversion to corn for ethanol for as
long as four decades, while conversion to cellulosic biofuels may not
have this effect (Piñeiro et al., 2009). As such, perennial feedstocks
grown on marginal or abandoned land may be preferred over the
conversion of natural ecosystems because they may incur a lower C debt
(Fargione et al., 2008).

3.2. Using land enrolled in the Conservation Reserve Program
As a response to increased demands for biofuels, increases in feedstock
production may come from a variety of sources, including yield gains
and increases in land area under production. Some of this area will include
marginal and CRP lands. While models predict that much of the land used
for perennial grass feedstocks will come from cropland, up to 31% may
come from previous CRP land (McLaughlin et al., 2002). Land under
CRP is typically less productive, more erodible, and has steeper slopes
than cropland (Secchi et al., 2009). The primary objective of the CRP is
to protect lands from soil and water erosion mainly through grassland
conversion, but wildlife have also benefited from this program (Best
et al., 1997; Reynolds et al., 2001). As corn prices rise due in part to
biofuel demands, producers will likely begin to crop on marginal land,
increasing water erosion, N and P losses due to sediment loss, and
decreasing soil C stocks (Secchi et al., 2009). Conversion of CRP land to
bioenergy feedstock production may also create a large C debt if annual
crops are cultivated and displace the perennial species currently
established on the CRP land: approximately 11 Mg C haÀ1 may be
released in the first year through the conversion of CRP land to annual
crop production for biofuels, when accounting for increased GHG
emissions and depletions of the soil organic carbon (SOC) stocks, which

would take decades to repay (Gelfand et al., 2011).
Whereas corn production, particularly as a continuous corn crop, has
deleterious impacts when grown on previous CRP lands, perennial grass
feedstocks have been suggested for growth on CRP and marginal lands.


Agronomic and Ecological Implications of Biofuels

11

Native grasses are one of the most common choices for vegetation in CRP
lands, with 2.7 Mha of CRP land under native grass plantings in 2009
(CRP, 2009). These long-lived species have deep roots that can stabilize
soil and reduce water runoff, potentially furthering the goals of the
CRP. Several species of native grasses, in particular switchgrass, are
proposed as CRP species as well as biofuel feedstock. Native, warmseason grasses such as switchgrass can be suitable for both CRP and
biofuel production because they are high-yielding under low fertilizer
inputs, drought tolerant, can improve soil stability, and also grow under
a wide range of conditions (Lewandowski et al., 2003). It would be ideal
if these lands could simultaneously be managed for biomass production
and also achieve the CRP objectives (Tilman et al., 2006).
The Food, Conservation, and Energy Act of 2008 permits haying and
biomass harvesting on CRP lands, provided that land is managed to protect
the soil and wildlife objectives of the program (U.S. Congress, 2008). There
have been conflicting results as to whether bioenergy cropland can provide
the same services as CRP and natural grasslands, depending on the
feedstock and land use management (Fargione et al., 2009). For example,
monocultures may have lower wildlife habitat quality than diverse
mixtures (McCoy et al., 2001). Standing biomass remaining in the field
after harvest may be beneficial for control of soil erosion and

water runoff, and also provide habitat for wildlife. To meet both CRP
and bioenergy needs, switchgrass biomass yields may be maximized with
low N fertilizer inputs and harvest after a killing frost (Mulkey
et al., 2006). Furthermore, perennial grass feedstock production on
CRP land may mitigate GHG emissions by capturing a minimum of
2.3 Mg CO2e haÀ1 yrÀ1 (Gelfand et al., 2011). However, if grasses are
fertilized, increased N2O emissions may cause a net release of GHG, even
though SOC stocks increase (Robertson et al., 2011b).
Harvest management is important for both maintaining high yields and
achieving CRP goals. Increased harvest frequency from one cut per year to
two or three cuts may lower yields of perennial grasses (Cuomo et al.,
1996). Even repeated annual single-cut harvests on CRP lands without
additional fertilizers or organic amendments may lower total yields over
time as a result of nutrient loss through plant biomass removal without
replacement (Mulkey et al., 2006; Venuto and Daniel, 2010). Cases
where there have been long-term high yields on marginal lands with low
inputs, such as the diversity experiment by Tilman et al. (2006), may be
suspect in a bioenergy cropping system because of management choices,
as Tilman et al.'s plots were burned rather than the biomass removed.
Stand persistence and productivity may impact wildlife habitat quality as
well. While CRP and marginal lands have the potential to provide
bioenergy without interfering with food crop production, they must be
carefully monitored to ensure plant vigor and persistence.


12

Catherine Bonin and Rattan Lal

3.3. Biofuels and restoration of degraded lands

The CRP enrolls marginal cropland, but lands may be degraded for other
reasons. Degraded or marginal lands have low productivity and environmental quality due to a history of intensive use and disturbances such as
farming, mining, or erosion. These lands are generally infertile and have
low SOC stocks. Bioenergy feedstock such as perennial grasses and trees
have been proposed for use on marginal lands as a method to improve
soil quality, enhance SOC stocks, and improve soil fertility (BlancoCanqui, 2010). Perennial species are already used in riparian buffers to
capture agricultural runoff. Bioenergy feedstock present opportunities to
address both ecological and energy problems. “Extremophile energy
crops,” species capable of high productivity under stresses such as salinity,
drought, or extreme temperatures, have also been suggested for use on
marginal lands, as they are typically efficient in their use of nutrients and
water (Bressan et al., 2011).
Several biofuel feedstock options are tolerant of acid soils and various
toxic compounds, and may aid in phytoremediation of contaminated soils
(Blanco-Canqui, 2010; Peterson et al., 1998; Rockwood et al., 2004).
Short-rotation woody crops (SRWC) may be able to take up nutrients
from wastewater, simultaneously reducing fertilizer needs and improving
water quality (Adegbidi et al., 2001). Poplars grown as a vegetative filter
strip with landfill leachate take up 159 kg of leachate kgÀ1 aboveground
poplar biomass, can treat 338 kg N haÀ1 yrÀ1, and may yield nearly
12 Mg haÀ1 woody biomass (Licht and Isebrands, 2005). Based on an
LCA comparing bioremediation with willows against a traditional
excavation-and-fill method for a landfill, remediation using plants may
have fewer negative environmental impacts (Suer and Andersson-Sköld,
2011). Giant reed (Arundo donax L.) can tolerate arsenic and heavy metals
and could be used in phytoremediation of contaminated soils (Mirza
et al., 2010; Papazoglou et al., 2005).
Highly productive perennial species tolerant of poor soil conditions may
also aid in revegetating degraded mine soils. Surface mine soils typically
contain high amounts of rock fragments, low nutrient concentrations,

and low SOC concentrations (Akala and Lal, 2000; Bendfeldt et al.,
2001; Haering et al., 2004). In addition, these soils may also be compacted
and acidic, all of which can make the establishment of persistent vegetative
cover challenging (Haering et al., 2004). However, these degraded soils
must be reclaimed and revegetated in order to limit soil erosion and
restore soil conditions, a process required by the Surface Mining Control
and Reclamation Act (SMCRA) of 1977. Switchgrass and reed
canarygrass (Phalaris arundinacea L.) can be dominant species on surface
mine soils in Virginia, producing 16.5 Mg haÀ1 and 5.0 Mg haÀ1,
respectively (Evanylo et al., 2005). Likewise, hybrid poplar has also been


Agronomic and Ecological Implications of Biofuels

13

recommended to reforest reclaimed mine soils as a result of high survival
and growth rates (Casselman et al., 2006). If biofuels can be grown on
contaminated or degraded land, they may be able to produce biomass for
energy without competing with food crops for land and may
simultaneously improve the condition of degraded or contaminated sites.
While growing perennial feedstock on degraded lands may be beneficial
for soil and water quality, the ecological consequences of using degraded
lands to produce bioenergy is uncertain. Bioenergy production from abandoned agricultural lands could provide as much as 8% of global energy
needs (Campbell et al., 2008), and including bioenergy production from
other marginal lands would further increase this value. However, yields
are generally lower on marginal lands, requiring higher selling prices for
producers to break even (Mooney et al., 2009). As with CRP land,
feedstock production on other marginal lands may require greater inputs
to produce sufficient yields and may result in greater erosion and

nutrient runoff (Dominguez-Faus et al., 2009). Harvesting degraded lands
for bioenergy could also have serious implications for conservation issues.
While degraded lands may have lower diversity than natural systems, in
many situations they are more diverse than plantations and agricultural
lands (Plieninger and Gaertner, 2011). Plant and animal species diversity
may be reduced if degraded lands are used for bioenergy crops instead of
restored for nature conservation.

4. Soil Erosion and Water Quality
Land use choices and water quality are closely linked. Large-scale land
use changes from annual to perennial crops will affect sediment losses,
nutrient runoff, and water yields (Schilling et al., 2008). Soil erosion and
nutrient runoff into bodies of water may have serious effects on soil
quality and plant productivity, as well as on ecosystem and human
health. The impacts that bioenergy cropping systems have on soil erosion
and water quality could be either positive or negative, depending on the
species and management practices. Fertilized annual cropping systems are
expected to have more negative effects on soil and water than low-input
perennial systems. Increased corn production for ethanol may increase
nutrient runoff due to greater use of fertilizers, particularly in large corngrowing regions such as the Mississippi River Basin. An estimated 80%
of the increased production of corn will occur in the Mississippi River
Basin, and could increase N and P loads by 37% and 25%, respectively
(Simpson et al., 2008). With a scenario of 136 billion liters of corn-grain
ethanol by 2022, dissolved inorganic N export into the Gulf of Mexico


14

Catherine Bonin and Rattan Lal

would increase by 34%, dramatically increasing the risk of hypoxia and
contributing to the “Dead Zone” in the Gulf of Mexico (Donner and
Kucharik, 2008).
Perennial species provide an opportunity to reduce the negative soil and
water impacts that annual bioenergy crops present. Riparian buffer strips
established with grasses or trees reduce nutrient runoff, sediment loss, filter
pesticides lost from crop fields, stabilize stream banks, and reduce bank
erosion (Lyons et al., 2000; Udawatta et al., 2002). Perennial grasses such
as switchgrass have lower N requirements, are more efficient at using N,
may reduce runoff, and also improve soil quality (Parrish and Fike,
2005). Compared to a corn-soybean rotation system, switchgrass
plantings can reduce N and P runoff by 50e90%, while delaying harvest
until winter could prevent additional nutrient runoff (Simpson et al.,
2008). Long term switchgrass stands fertilized at 68 kg N haÀ1 may have
N losses through surface and ground water of under 2 kg haÀ1, which is
approximately 3% of the N applied (Sarkar et al., 2011). Fertilizer
application rates may affect the size of reduction of NO3eN in surface
runoff, with greater reductions as application amount decreases: when
224 kg N haÀ1 is applied, there is a 16% reduction in NO3eN runoff
compared to the baseline cropping system, but with no fertilization, this
reduction increases to 65% (Nelson et al., 2006). Large-scale changes in
land use to bioenergy plantations of either corn or perennial crops would
significantly affect whether N or P losses would increase or decrease.
Compared to a scenario where land is used for large-scale corn
production, large-scale switchgrass bioenergy plantings would result in
a 57% decrease in NO3eN losses and nearly a 98% decrease in P losses
(Schilling et al., 2008). Woody species grown for bioenergy also have the
potential to significantly reduce N and P losses from runoff (Thornton
et al., 1998).
Along with reducing surface runoff nutrient losses, biofuels may reduce

nitrate leaching. Nitrates reduce subsurface water quality and when present
in drinking water may be harmful to human health, particularly young children (Townsend et al., 2003). Under miscanthus, nitrate losses decreased
after the establishment year and, when fertilized with 60 kg N haÀ1 or
less, did not exceed the nitrate limit of 10 mg N lÀ1 set by the EPA
(Christian and Riche, 1998; EPA, 2011). In contrast, the 10 mg N lÀ1
limit is often exceeded in a cornesoybean cropping system, especially as
fertilizer application rates increase (Jaynes et al., 2001). In one four-year
study, nitrate losses under switchgrass and miscanthus were only 4e8% of
those losses under cornesoybean rotations (McIsaac et al., 2010), while
nitrate loss reductions of up to 98% have been found when comparing
CRP land to continuous corn systems (Randall et al., 1997). The first
year of plantation establishment generally has the highest nitrate losses,
and by the second or third year, losses drop sharply as stands establish


Agronomic and Ecological Implications of Biofuels

15

(Christian and Riche, 1998; McIsaac et al., 2010). Perennial grasses may
have reduced nitrate losses for several reasons, including requiring lower
N applications, having a longer growing season and having a larger root
system, both of which should increase N uptake. Diverse perennial
stands, with a potential for higher root biomass and more efficient
resource use, may also reduce nitrate leaching (Scherer-Lorenzen et al.,
2003).
Sediment loss from agricultural land and entry into surface waters is
a major problem. Sediments can increase water turbidity, reducing plant
photosynthetic activity and productivity, and can also negatively affect
fish and other aquatic invertebrates: even low levels of turbidity may

decrease primary productivity by as much as 13% (Ryan, 1991). In
sugarcane production systems, residue is frequently burned on the fields
to facilitate harvest and transport, a practice that releases GHG into the
atmosphere and leaves bare ground at risk of soil erosion rates of
17e505 Mg haÀ1 yrÀ1 (Hartemink, 2008). Development of new
harvesting equipment so that burning is not required can retain this soil
and allow the soil to store an additional 1500 kg C haÀ1 yrÀ1 (Galdos
et al., 2010). Sediment loads are predicted to increase with the
expansion of annual row crops and be reduced for perennial grasses, but
all feedstock plantations may generate sediment losses if grown on
marginal or steep lands (Love and Nejadhashemi, 2011). Although
sediment and runoff from switchgrass may initially be higher during the
establishment year (Thornton et al., 1998), switchgrass may reduce
sediment yield by more than 99%, edge-of-field erosion by 98%, and
surface runoff by 55%, when compared to cropping systems (Nelson
et al., 2006). In contrast, SRWC may reduce sediment losses by as
much as 85%, even in the first year of establishment (Thornton et al.,
1998).
The conversion of large areas of land to bioenergy plantations may
impact landscape hydrology. Changes in water yield would alter the
amount of potential nutrient runoff, as increases in water yields may also
increase N and P runoff losses. Shifts to large-scale production of corn
for ethanol under a large area conversion from CRP grassland to corn
would reduce evapotranspiration, increasing water flow by as much as
8%, particularly during spring and late fall (Schilling et al., 2008). In
contrast, perennial species have higher amounts of evapotranspiration
stemming from a longer growing period and higher biomass production
that would reduce the water yield. Under a large area conversion to
switchgrass, water yields could decrease by as much as 28%, when
compared to current cropping conditions (Schilling et al., 2008). Soils

under CRP land have lower plant available water (PAW) within the top
1.5e2 m, when compared to row crops due to greater water use by the
perennial species (Randall et al., 1997).


16

Catherine Bonin and Rattan Lal

Although conversion from annual to perennial crops may alter
hydrology, each perennial feedstock species may behave differently. The
differences in root distribution and water capture can affect changes in
soil water: soil under miscanthus, with most roots in the top 0.35 m, tends
to have higher summer soil moisture at lower depths than soil under
switchgrass or giant reed (Monti and Zatta, 2009). Just as with perennial
grasses, SRWC plantations would use more water than annual crops,
lowering soil moisture and potentially reducing water yield and the size
of floods if planted on a large scale (Perry et al., 2001). Compared to soil
under switchgrass, soil under miscanthus has lower soil moisture later
during the growing season and in combination with greater
evapotranspiration, may potentially reduce surface flows by 32% if
planted over a large area (McIsaac et al., 2010). Modeling simulations
suggest that miscanthus may be planted at under 10% of US crop area
without impacts on hydrology; however, given that energy efficiency is
maximized by concentrating biofuel plantations, miscanthus land
coverage would likely exceed 25% or even 50% in areas surrounding
refineries or energy plants (VanLoocke et al., 2010). Reductions in water
yield may provide benefits such as reduced flooding, but may also have
adverse effects, such as later recharges of soil moisture and extended
periods of low water flow.

4.1. Use of agricultural residues
Agricultural residues have many uses; i.e., as an input of organic matter
(OM) to soils, as feed for animals, feedstock for bioenergy, and as raw materials for industry. Using corn stover for cellulosic ethanol may reduce SOC
stocks, soil quality, crop yields, and soil faunal activities due to the loss of
OM and surface residues (Blanco-Canqui and Lal, 2007; Karlen et al.,
1994; Karlen et al., 2009; Lal, 2009). The data in Figs. 3 and 4 show the
relationship between agricultural residues and SOC stocks or soil erosion,
respectively: as residue is removed, SOC generally decreases and soil
erosion increases. In addition, decreases in crop residues may also reduce
crop yields: for each 1 Mg haÀ1 of residue that is removed, grain yield
declines by 0.10 Mg haÀ1 while future residue production decreases by
0.30 Mg haÀ1 (Wilhelm et al., 1986). Agricultural residues are important
as a method for controlling nutrient runoff and soil erosion. Removal of
50% of crop residues can double rates of soil erosion, while higher rates
of removal can greatly increase N and P sediment losses (Blanco-Canqui
et al., 2009). Although complete removal of residues is detrimental to soil
and water quality, it is possible that lower levels of residue removal,
estimated at 25% or less, may be acceptable for maintaining soil and
water quality (Blanco-Canqui and Lal, 2007; Blanco-Canqui et al., 2009).
The benefits of additional ethanol yield from agricultural byproducts must


17

Agronomic and Ecological Implications of Biofuels

Relative Amount of SOC

1.10

r² = 0.520

1.00
0.90
0.80
0.70
0.60
0

25

50

75
100 125 150
Residue Present ( )

175

200

Figure 3 Relative amount of soil organic C (SOC) compared to residue remaining on
fields. The SOC is relative based on SOC amounts when 100% of residue is left on the
fields. Data from Blanco-Canqui et al. (2006), Blanco-Canqui and Lal (2007), and
Maskina et al. (1993).
8.0

Relative Soil Erosion

7.0
6.0
5.0
4.0
3.0
2.0

r² = 0.4596

1.0
0.0
0

50

100

150

200

Residue Present (%)

Figure 4 Relative amount of soil erosion compared to residue remaining on fields.
Soil erosion is relative based on erosion amounts when 100% of residue is left on the
fields. Data from Blanco-Canqui et al. (2009), Lindstrom (1986), and Powers et al.
(2011).

be carefully weighed against the costs of increasing fertilizer inputs and
potential loss in soil quality.

5. Nitrogen Cycling
As discussed above, N losses through runoff and nitrate leaching are
typically lower under perennial species with established root systems and
a longer period of standing vegetative cover. However, nutrient cycling


18

Catherine Bonin and Rattan Lal

also includes the internal translocation of nutrients across seasons and the
plant litter dynamics. Internal cycling, plant material decomposition, and
mineralization by microorganisms provide readily accessible sources of N
for plants, while some N turns into stable humus compounds that provide
a slow release of N (Clark, 1977). The schematic in Fig. 5 outlines the
general movement of N in traditional cropping systems and a perennial
biofuel system. While N cycling in traditional systems such as corn and
wheat is more understood, knowledge of the movement of N through
a perennial bioenergy system is incomplete and requires more work.
Understanding the total system balance for N and other nutrients is
important to optimize fertilizer applications to ensure adequate plant
nutrient supply and to minimize nutrient loss. Ideally, biofuel species
would be productive under low nutrient inputs and would further
reduce their nutrient needs through three factors: nutrient-poor harvested
tissue, internal movement of nutrients to roots, and litter decomposition
(Tom, 1994).


Agronomic and Ecological Implications of Biofuels

19

Figure 5 Movement of nitrogen within a) traditional cropping and b) perennial biofuel systems. Percentages represent the amount of N added to the system that is
removed by each component. Note: for part b, percentages do not add up to 100 due to
values taken from multiple sources, uncertainty in calculations, and uncalculated N
translocation and storage in roots. Data from Adler et al. (2007); Bransby et al. (1998);
Liebig et al. (2006); Marshall et al. (1999); McLaughlin et al. (2002), and Smil (1999).

5.1. Nitrogen and litter/residue management
Leaves, agricultural residues, and other sources of plant litter are an important part of nutrient cycling. In a mixed-grass prairie, litter may comprise up
to 60% of the aboveground biomass and contain approximately 50% and
60% of aboveground C and N stocks (Schuman et al., 1999). Litter
decomposition rates can be affected by its quality, as well as litter N and
lignin concentrations (Fog, 1988; Melillo et al., 1982). Litter rich in N
and low in lignin content tend to decompose more quickly and may
make N available for plant uptake more rapidly (Melillo et al., 1982).
Leaf litter impacts nutrient cycling through incorporation into the soil,
through microorganism activities, and by leaching of surface litter (Clark,
1977). As willow stands establish, N release from leaf litter may reach
over 100 kg N haÀ1, while internal recycling may be over 50 kg N haÀ1,
which will significantly reduce N fertilizer needs (Tom, 1994).
Nitrogen cycling within an ecosystem may also be measured by using
15
N-enriched fertilizers and tracking it within plant parts and soil. In a