EVALUATING THE IMPACTS OF CLIMATE CHANGE AND VARIABILITY ON
GRAZING DAIRY PRODUCTION
By
Maria Melissa Rojas-Downing

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Biosystems Engineering â Doctor of Philosophy
2017

ABSTRACT
EVALUATING THE IMPACTS OF CLIMATE CHANGE AND VARIABILITY ON
GRAZING DAIRY PRODUCTION
By
Maria Melissa Rojas-Downing
The global demand for milk production is expected to double by 2050, mostly due to
population growth and income rise in developing countries. Meanwhile, increases in milk
production can accelerate negative impacts that contribute to land degradation, climate change,
water shortages and pollution. Therefore, it is important to meet the increasing demand for milk
production under sustainable practices. Meanwhile, milk production is also impacted by climate
change and variability. The rise of carbon dioxide, temperature increases, and increasing
precipitation variation are the main factors that will impact forage quality and growth. Therefore,
among livestock systems, the grazing system is likely to be the most impacted by climate change
and variability because of its dependency on forage quality and quantity. However, the level of
impacts and measures to mitigate and adapt to these impacts are not very well known.
To address these knowledge gaps, we developed a study based on the following research
objectives: 1) understand the global impacts of climate change on livestock production, the
contribution of livestock production to climate change, and specific adaptation and mitigation
strategies for the sector; 2) study the adoption measure (pasture diversification) to identify the
most climate resilient pasture composition for a representative grazing dairy farm in Michigan;
and 3) identify the most sustainable milk production for a representative grazing dairy farm in
Michigan by considering economic, water, energy, and carbon footprints.
In order to address the first objective, a literature review was performed, which showed
that livestock production will be limited by climate change and variability and competition for

water, land, and food security. Meanwhile, the livestock sector is a major contributor to
greenhouse gas (GHG) emissions, driving further climate change. Consequently, the livestock
sector will be a key player in the mitigation of GHG emissions and improving global food
security. Therefore, in the transition to sustainable livestock production, there is a need for
assessing the use of adaptation and mitigation measures tailored to the location and livestock
production system.
To address objectives two and three, a representative farm was developed based on
grazing dairy farm practices surveys and incorporated into the Integrated Farm System Model
(IFSM). For the pasture compositions, four cool-season grass species with the two legumes were
evaluated, which resulted in 48 pasture compositions.
For objective two, the effectiveness and resiliency of the pasture compositions to climate
change impacts were evaluated based on economic and resource use criteria. Results showed that
the increase in precipitation and temperature of the most intensive climate scenario would
significantly improve farm net return per cow and whole farm profit. Perennial ryegrass with red
clover was identified as the most resilient pasture composition to improve farm economics and
resource use under climate change.
Under objective three, a multi-criterion decision making method was used to calculate a
new footprint (food footprint) to assess sustainability of a grazing dairy farm. Using IFSM,
carbon, water, energy, and economic impacts of the representative farm in 10 locations in
Michigan were calculated. Using food footprint, the most sustainable milk production level was
identified in each location. The results of this analysis are promising since it encourages a high
level of milk production (8,618 kg/cow/year) while promoting the most sustainable approach for
grazing farm management.

To my husband, who with all his love, support, sacrifice, encouragement and faith made the
accomplishment of this thesis possible. With all my love this is for you.

iv

ACKNOWLEDGMENTS

I would like to express my sincere gratitude to my advisor Dr. Amirpouyan
Nejadhashemi who believed in my capability and gave me the opportunity to obtain my PhD.
Besides being an excellent mentor, he became a friend for life. I am thankful for all the support,
patience, motivation and knowledge during my PhD program. Dr. Nejadhashemiâs, along with
Dr. Harriganâs guidance and encouragement, has made me the researcher I am today. For this, I
will be forever grateful.
To my committee members Dr. Timothy Harrigan, Dr. Joseph Messina, and Dr. David
Lusch, I would like to extend my sincere thanks for their guidance, advice, insightful comments
and support throughout the duration of my research. I would also like to thank Dr. Santiago
Utsumi, and Dr. Alan Rotz for their technical knowledge and support during the development of
my research.
I would never have been able to finish my research without the help from the dairy
farmers, extension specialists and Dr. Edwin Martinez (USDA-NRCS) who provided very useful
information for my research. Furthermore, I would like to thank the Biosystems and Agricultural
Engineering department, the Environmental Science Policy Program, and Bill & Rita Stout for
providing financial support.
I would like to also thank my lab mates for all their help, support and encouragement, and
for making my time here filled with great memories. And to the administrators and staff, which
besides helping me with the administrative side of the process they made me feel part of the
BAE family.

v

And my profound gratitude to my family who morally supported me and believed in me.
To my dear friends for becoming my family and always giving me good advice and
encouragement. My moments with you will be unforgettable. And last but not least, my greatest
appreciation is for my husband for his support, advice, motivation, love, faith and patience in
every moment of my graduate study. This study and degree would not have been possible
without all your help.

vi

TABLE OF CONTENTS
LIST OF TABLES .......................................................................................................................... x
LIST OF FIGURES ....................................................................................................................... xi
KEY TO ABBREVIATIONS ....................................................................................................... xv
1 INTRODUCTION ....................................................................................................................... 1
2 LITERATURE REVIEW ............................................................................................................ 4
2.1 Overview ............................................................................................................................... 4
2.2 Topical Review...................................................................................................................... 5
2.2.1 Impact of Climate Change on Livestock ........................................................................ 5
2.2.1.1 Quantity and quality of feeds .................................................................................. 8
2.2.1.2 Water ....................................................................................................................... 9
2.2.1.3 Livestock diseases ................................................................................................. 10
2.2.1.4 Heat stress ............................................................................................................. 12
2.2.1.5 Biodiversity ........................................................................................................... 15
2.2.1.6 Food security ......................................................................................................... 17
2.2.2 Impact of Livestock on Climate Change ...................................................................... 18
2.2.2.1 GHG emissions ..................................................................................................... 21
2.2.2.2 Land use ................................................................................................................ 24
2.2.2.3 Feed production..................................................................................................... 26
2.2.2.4 Animal production ................................................................................................ 28
2.2.2.4.1 Emissions by Species and Commodities ....................................................... 29
2.2.2.5 Manure .................................................................................................................. 30
2.2.2.6 Processing and transport ....................................................................................... 31
2.2.3 Summarizing Adaptation and Mitigation Practices ...................................................... 32
2.2.3.1 Adaptation measures ............................................................................................. 33
2.2.3.1.1 Livestock production and management systems ........................................... 33
2.2.3.1.2 Breeding strategies ........................................................................................ 34
2.2.3.1.3 Farmersâ perception and adaptive capacity ................................................... 35
2.2.3.2 Mitigation measures .............................................................................................. 35
2.2.3.2.1 Carbon sequestration ..................................................................................... 36
2.2.3.2.2 Enteric fermentation ...................................................................................... 37
2.2.3.2.3 Manure management ..................................................................................... 38
2.2.3.2.4 Fertilizer management ................................................................................... 39
2.2.3.2.5 Shifting human dietary trends........................................................................ 40
2.3 Summary ............................................................................................................................. 40
3 INTRODUCTION TO METHODOLOGY AND RESULTS ................................................... 43
4 PASTURE DIVERSIFICATION TO COMBAT CLIMATE CHANGE IMPACTS ON
GRAZING DAIRY PRODUCTION ......................................................................................... 45
4.1 Introduction ......................................................................................................................... 45
vii

4.2 Materials and Methods ........................................................................................................ 50
4.2.1 Study Area .................................................................................................................... 50
4.2.2 Data Collection ............................................................................................................. 52
4.2.2.1 Grazing Farm Survey ............................................................................................ 52
4.2.2.2 Representative Grazing Dairy Farm Description .................................................. 53
4.2.2.3 Climate Data ......................................................................................................... 55
4.2.3 Modelling Process ........................................................................................................ 59
4.2.3.1 Integrated Farm System Model (IFSM) ................................................................ 59
4.2.3.2 Selection of pasture compositions......................................................................... 60
4.2.3.3 Simulation of pasture compositions and climate scenarios in IFSM .................... 62
4.2.3.4 Identifying the impact of pasture compositions and climate change on economic
and resource use of a livestock grazing system .................................................... 63
4.2.3.5 Sensitivity analysis to identify the most resilient pasture composition ................ 64
4.2.3.6 Economic and resource use sensitivity under future climate change scenarios .... 65
4.3 Results and Discussion ........................................................................................................ 68
4.3.1 Identification of the best pasture composition based on resource use and economics . 68
4.3.1.1 Net return per cow over milk production .............................................................. 68
4.3.1.2 Whole Farm profit ................................................................................................. 70
4.3.1.3 Grazed forage consumed ....................................................................................... 71
4.3.1.4 Harvested forage production ................................................................................. 72
4.3.2 Impact of pasture compositions and climate change on economic and resource use of a
livestock grazing system .............................................................................................. 75
4.3.3 Economic and resource use sensitivity under future climate change scenarios ........... 80
4.3.3.1 Surface Analysis ................................................................................................... 80
4.3.3.2 Overall Sensitivity Assessment ............................................................................. 85
4.4 Conclusions ......................................................................................................................... 89
5 FOOD FOOTPRINT AS A MEASURE OF SUSTAINABILITY FOR GRAZING DAIRY
FARMSâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...92
5.1 Introduction ......................................................................................................................... 92
5.2 Materials and Methods ........................................................................................................ 95
5.2.1 Study Area .................................................................................................................... 95
5.2.2 Modeling Process .......................................................................................................... 98
5.2.2.1 Data Processing ................................................................................................... 101
5.2.2.1.1 Grazing Farm Survey and Review .............................................................. 101
5.2.2.1.2 Selection of pasture compositions ............................................................... 102
5.2.2.1.3 Selection of Target Milk Productions .......................................................... 103
5.2.3 Model Simulations ...................................................................................................... 104
5.2.3.1 Integrated Farm System Model (IFSM) .............................................................. 104
5.2.3.1.1 Scenario analysis of the impacts of climate variabilities, pasture
compositions, and target milk productions on sustainability grazing dairy
farms .......................................................................................................... 105
5.2.4 Footprint Analysis ...................................................................................................... 106
5.2.4.1 Food Footprint..................................................................................................... 109
5.2.5 Spatial Variability of Sustainable Pasture Composition and Milk Production ........... 112
5.2.6 The Social Component of Grazing Dairy Farms ........................................................ 113
viii

5.3 Results and Discussion ...................................................................................................... 114
5.3.1 Identification of the Most Sustainable Milk Production ............................................. 114
5.3.2 Identification of the Most Appropriate Pasture Composition..................................... 116
5.3.3 Spatial Variability of Sustainable Milk Productions and Associated Pasture
Compositions .............................................................................................................. 119
5.4 Conclusions ....................................................................................................................... 124
6 CONCLUSIONS ...................................................................................................................... 126
7 FUTURE RESEARCH RECOMMENDATIONS................................................................... 131
APPENDIX ................................................................................................................................. 133
REFERENCES ........................................................................................................................... 156

ix

LIST OF TABLES

Table 1. Summary of climate models used in NEX-GDDP and their average temperature and
precipitation projections for the study area for the period of 2040 to 2060 (IPCC, 2013;
PCMDI, 2015)âŚâŚâŚâŚâŚâŚâŚ...âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...âŚâŚâŚâŚâŚâŚâŚâŚ.âŚ..57
Table 2. Cool-season grasses characteristics (Barnes et al., 2007; Cassida et al., 2014)âŚâŚâŚâŚ.62
Table 3. Summary of the top four pasture compositions with the highest net return per cow over
milk productionâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ.âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...âŚ...70
Table 4. Top four pasture compositions with the highest whole farm profitâŚâŚâŚ..âŚâŚ..âŚâŚ.71
Table 5. CP and NDF for alfalfa and the pasture compositions with the highest quality.âŚ...âŚ..75
Table 6. Statistical comparison of net return, whole farm profit, grazed forage consumed, and
harvested forage produced by different grass and legume species (horizontal
comparison)âŚ...............................................................................................................78
Table 7. Statistical comparison of net return, whole farm profit, grazed forage consumed, and
harvested forage produced by different grass and legume species (vertical
comparison)âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ.....79
Table 8. Statistical comparison of net return, whole farm profit, grazed forage consumed and,
harvested forage produced by different climate scenarios (horizontal
comparison)âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ.80
Table 9. Average and overall sensitivity indices for all criteria, grass species, and legume
speciesâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ....88
Table 10. Average daily temperature and precipitation for the weather stations within the study
areaâŚâŚâŚâŚâŚâŚâŚ...âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ.98
Table 11. Representative grazing dairy farm descriptionâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ..101
Table S1. Major survey components (Rojas-Downing et al., 2017a)âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ..134
Table S2. Key survey data collected (Rojas-Downing et al., 2017a)âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ..136
Table S3. Statistical comparison of net return, whole farm profit, grazed forage consumed and,
harvested forage produced by different grass species and legume percentages in the
pasture composition (horizontal comparison)âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ.137
Table S4. Statistical comparison of net return, whole farm profit, grazed forage consumed and,
harvested forage produced by different grass species and legume percentages in the
pasture composition (vertical comparison)âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ.139
x

LIST OF FIGURES

Figure 1. Impacts of Climate Change on LivestockâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ..7
Figure 2. Impact of livestock on Climate ChangeâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...20
Figure 3. Contribution of livestock to the total GHG anthropogenic emissionsâŚâŚâŚâŚâŚ.âŚ.....21
Figure 4. Global GHG emissions from the livestock sector (modified from Gerber et al.,
2013)..............................................................................................................................23
Figure 5. Total global livestock GHG emission by specie and product (Adapted from Gerber et al.,
2013)âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ..âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ30
Figure 6. Study area and the location surveyed dairy farmsâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...âŚâŚ...51
Figure 7. Modeling process of the representative grazing dairy farmâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ....63
Figure 8. Sensitivity index areaâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ..67
Figure 9. Calculating the fraction of temperature to precipitationâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ.67
Figure 10. Net return per cow over milk production sensitivity index for perennial ryegrass and
red clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S=
sensitivity indexâŚâŚ..âŚâŚâŚ..âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...83
Figure 11. Whole farm profit sensitivity index for perennial ryegrass and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚ..................................................................................................âŚâŚ.84
Figure 12. Grazed forage consumed sensitivity index for perennial ryegrass and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚ.....................................................................................................................85
Figure 13. Harvested forage produced sensitivity index for perennial ryegrass and red clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...85
Figure 14. State of Michigan and its watershed boundariesâŚâŚâŚâŚâŚ.âŚâŚâŚâŚâŚ..âŚ..âŚâŚ...97
Figure 15. The conceptual framework of the modeling processâŚâŚâŚâŚâŚâŚâŚâŚ.âŚâŚâŚâŚ...100
Figure 16. Schematic representation of the integrated scenario analysisâŚâŚâŚâŚâŚâŚâŚâŚâŚ..106

xi

Figure 17. Food footprint for different combinations of red clover with a) Kentucky bluegrass, b)
orchardgrass, c) perennial ryegrass, and d) tall fescue for low (1,361 to 3,629
kg/cow/year), medium (4,082 to 5,897 kg/cow/year), and high (6,350 to 8,618
kg/cow/year) milk productionâŚ..âŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ..116
Figure 18. Histogram map for legume variations with grass speciesâŚâŚ....âŚâŚâŚâŚâŚâŚâŚâŚ117
Figure 19. Histogram map for grass species for different variations of legumeâŚâŚâŚâŚâŚâŚ....118
Figure 20. Histogram map of all pasture compositionsâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ.âŚ119
Figure 21. Spatial variability of most appropriate pasture compositions in MichiganâŚ..........âŚ121
Figure 22. Spatial variability of most sustainable milk productions in MichiganâŚâŚâŚ..âŚâŚ...122
Figure 23. Spatial variability of most sustainable milk productions and associated pasture
compositions in Michigan (Cluster analysis)âŚâŚâŚâŚâŚâŚâŚâŚ...âŚâŚâŚâŚâŚâŚ.123
Figure S1. Net return per cow over milk production sensitivity index of Perennial ryegrass and
white clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S=
sensitivity indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ141
Figure S2. Net return per cow over milk production sensitivity index of Tall Fescue and red clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...141
Figure S3. Net return per cow over milk production sensitivity index of Tall Fescue and white
clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...142
Figure S4. Net return per cow over milk production sensitivity index of Kentucky bluegrass and
red clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S=
sensitivity indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ.142
Figure S5. Net return per cow over milk production sensitivity index of Kentucky bluegrass and
white clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S=
sensitivity indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ.143
Figure S6. Net return per cow over milk production sensitivity index of Orchardgrass and red
clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...143
Figure S7. Net return per cow over milk production sensitivity index of Orchardgrass and white
clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ...144
Figure S8. Whole farm profit sensitivity index of Perennial ryegrass and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚâŚ144
xii

Figure S9. Whole farm profit sensitivity index of Tall Fescue and red clover pasture composition.
P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚâŚâŚâŚâŚâŚâŚ..145
Figure S10. Whole farm profit sensitivity index of Tall Fescue and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ145
Figure S11. Whole farm profit sensitivity index of Kentucky bluegrass and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ.146
Figure S12. Whole farm profit sensitivity index of Kentucky bluegrass and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ.146
Figure S13. Whole farm profit sensitivity index of Orchardgrass and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ147
Figure S14. Whole farm profit sensitivity index of Orchardgrass and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ147
Figure S15. Grazed forage consumed sensitivity index of Perennial ryegrass and white clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ..148
Figure S16. Grazed forage consumed sensitivity index of Tall Fescue and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ148
Figure S17. Grazed forage consumed sensitivity index of Tall Fescue and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ.149
Figure S18. Grazed forage consumed sensitivity index of Kentucky bluegrass and red clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ149
Figure S19. Grazed forage consumed sensitivity index of Kentucky bluegrass and white clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ150
Figure S20. Grazed forage consumed sensitivity index of Orchardgrass and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ.150
Figure S21. Grazed forage consumed sensitivity index of Orchardgrass and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ.151
Figure S22. Harvested forage produced sensitivity index of Perennial ryegrass and white clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ151
Figure S23. Harvested forage produced sensitivity index of Tall Fescue and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ.152
xiii

Figure S24. Harvested forage produced sensitivity index of Tall Fescue and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ.152
Figure S25. Harvested forage produced sensitivity index of Kentucky bluegrass and red clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ153
Figure S26. Harvested forage produced sensitivity index of Kentucky bluegrass and white clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
indexâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ153
Figure S27. Harvested forage produced sensitivity index of Orchardgrass and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ.154
Figure S28. Harvested forage produced sensitivity index of Orchardgrass and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity indexâŚ.154
Figure S29. Food footprint for different combinations of white clover with a) Kentucky bluegrass,
b) orchardgrass, c) perennial ryegrass, and d) tall fescue for low (1,361 to 3,629
kg/cow/year), medium (4,082 to 5,897 kg/cow/year), and high (6,350 to 8,618
kg/cow/year) milk productionâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚâŚ.155

xiv

KEY TO ABBREVIATIONS

ANPP: Net Primary Productivity
ARS: Agricultural Research Service
BCC: Beijing Climate Center
BCSD: Bias-Correction Spatial Disaggregation
BOM: Bureau of Meteorology
CCCma: Canadian Centre for Climate Modelling and Analysis
CGIAR: Consultative Group for International Agricultural Research
CH4: Methane
CMIP5: Coupled Model Intercomparison Project Phase 5
CNRM-CERFACS: Centre National de Recherches Meteorologiques - Centre Europeen de
Recherche et Formation Avancees en Calcul Scientifique
CO2: Carbon Dioxide
CP: Crude Protein
CSIRO: Commonwealth Scientific and Industrial Research Organization
CSIRO-QCCCE: Commonwealth Scientific and Industrial Research - Queensland Climate
Change Centre of Excellence
DM: Dry Matter
DNA: Deoxyribonucleic Acid
xv

FAO: Food and Agriculture Organization of the United Nations
FCPM: Fat and Protein Corrected Milk
GCESS: College of Global Change and Earth System Science
GCM: General Circulation Model
GFDL: Geophysical Fluid Dynamics Laboratory
GHG: Greenhouse Gas
GIS: Geographic Information System
GLEAM: Global Livestock Environmental Assessment Model
GMFD: Global Meteorological Forcing Dataset
GRAZPLAN: Decision Support Software for Grazing Enterprises
GWP: Global Warming Potential
HUC: Hydrologic Unit Codes
IFSM: Integrated Farm System Model
INM: Institute for Numerical Mathematics
IPCC: Intergovernmental Panel on Climate Change
IPSL: Institute Pierre-Simon Laplace
LCA: Life Cycle Assessment
MATLAB: Matrix Laboratory

xvi

MCDM: Multi-Criterion Decision-Making
MPI-M: Max Planck Institute for Meteorology
MRI: Meteorological Research Institute
MSU: Michigan State University
N2O: Nitrous Oxide
NASA: National Aeronautics and Space Administration
NCAR: National Center for Atmospheric Research
NCC: Norwegian Climate Centre
NCDC: National Climatic Data Center
NDF: Neutral Detergent Fiber
NEFA: Non-Esterified Fatty Acids
NEX-GDDP: Earth Exchange Global Daily Downscaled Projections
NFIFO: Net Farm Income from Operations
NOAA: National Oceanic and Atmospheric Administration
NSF-DOE-NCAR: National Science Foundation - Department of Energy - National Center for
Atmospheric Research
RCP: Representative Concentration Pathways
SD: Standard deviation
SWAT: Soil and Water Assessment Tool

xvii

USDA: U.S. Department of Agriculture
VIKOR: Multi-Criteria Optimization and Compromise Solution
WXGEN: Stochastic Weather Generator Model

xviii

1

INTRODUCTION

In the last 40 years, livestock production and demand have rapidly expanded and are
expected to continue, mainly due to population growth and improvement in the worldwide
standard of living (FAO, 2008; FAO, 2014). Livestock is a major sector of the agricultural
industry, accounting for 40% of the global value of agricultural outputs (FAO, 2017). In the
United States, the annual economic contribution from the livestock sector is worth $346 billion
(United Soybean Board, 2013). In addition, this sector contributes to the livelihoods of 70% of
the world's lowest income rural population and employs close to 1.1 billion people (Hurst et al.,
2005; FAO, 2014). Livestock products provide 17% of global kilocalorie consumption and 33%
of global protein consumption making them an important agricultural commodity for global food
security (Rosegrant et al., 2009). Therefore, the growth of the livestock sector provides
opportunities for agricultural development, food security, and poverty reduction (FAO, 2017).
However, it will be fundamental for the sector to meet the growing demand in a sustainable
manner under stressors, such as climate change.
Climate change is a threat to livestock production because of the impacts on quality and
quantity of feed crop and forage, water availability, animal and milk production, livestock
diseases, animal reproduction, and biodiversity (Chapman et al., 2012; IFAD, 2010; Polley et al.,
2013; Thornton et al., 2009; Henry et al., 2012; Nardone et al., 2010; Reynolds et al., 2010).
Among livestock systems, grazing is likely to be the most impacted by climate change because
of its dependency on feed quality and availability. Meanwhile, adaptation measures can increase
the resilience of livestock production to climate change. Therefore, there is a need to identify
successful adaptation measures that can be implemented to reduce the negative impacts of

1

climate change. Among adaption measures, plant diversification is one that has high probability
of selection by grazing farm producers since it only requires changing the seeding practices.
Most of the grazing livestock systems are for beef and dairy production and the demand
is expected to increase 44% and 55%, respectively, by 2030 (HavlĂ­k et al., 2014). In addition, the
United States is the largest cowâs milk producer in the world with 41% of its production
concentrated in the Great Lakes region (USDA-NASS 2014c; FAO 2016a). Therefore, the goal
of the first study was to identify the most resilient pasture composition for a representative
grazing dairy farm in order to minimize the negative impacts of climate change. The study
assesses the resiliency of the pasture compositions by evaluating the economic and resource use
(feed production and consumption) of the farm. The specific objectives were:
â˘

Identify the best pasture composition under current and future climate scenarios
concerning farm economics and resource use

â˘

Assess economic and resource use impacts of pasture compositions under future climate
scenarios

â˘

Evaluate the resiliency of pasture compositions to climate change
Meanwhile, livestock production is the largest user of resources in the world (FAO,

2017). Livestock feed production currently accounts for 80% of all agricultural land (FAO,
2017). In addition, this sector accounts for about 8% of global human water use (Nardone et al.,
2010). The livestock sector is also responsible for several environmental issues such as land and
habitat degradation, water and air pollution, and biodiversity destruction (Bellarby et al., 2013;
Reynolds et al., 2010; Steinfeld et al., 2006; Thornton and Gerber, 2010). Moreover, the
livestock sector is one of the largest contributors to anthropogenic GHG emissions (14.5% of
global anthropogenic GHG emissions), which are the major cause of global climate change

2

(Gerber et al., 2013). Therefore, there is a need to identify mitigation measures that will improve
the sustainability of livestock productions. The US Department of Agriculture (USDA) defines
sustainable agriculture as a site-specific plant-animal production system that in the long-term
will fulfill human feeding needs (e.g. milk), improve the environment (e.g. GHG emissions), use
resources efficiently (e.g. water, energy), maintain financial management (e.g. farm economy),
and enhance the life quality of both farmers and society (e.g. social wellbeing) (Gold, 2007).
Different indicators, such as footprints, have been used to measure sustainability (ÄuÄek et al.,
2012). However, the majority of studies have used carbon footprint as the only measure of
sustainability (Adom et al., 2012; de LĂŠis et al., 2015; FAO, 2010b; FlysjĂś et al, 2011; Mc
Geough et al., 2012; OâBrien et al., 2014). A comprehensive sustainability study should also
consider other aspects, such as water and energy requirements and economic feasibility that may
be impacted by the proposed management practices (OECD, 2004). Therefore, the goal of the
second study was to identify the most sustainable milk production system for a representative
grazing dairy farm under different climate conditions within the State of Michigan. The specific
objectives were:
â˘

Identify the most sustainable milk production system in different locations

â˘

Identify the pasture composition for each targeted milk production in different locations

â˘

Understand the spatial variability of sustainable milk productions and associated pasture
compositions for different locations

3

2
2.1

LITERATURE REVIEW

Overview
Human population is expected to increase from 7.2 to 9.6 billion by 2050 (UN, 2013). This

represents a population increase of 33%, but as the global standard of living increases, demand
for agricultural products will increase by about 70% in the same period (FAO, 2009a).
Meanwhile, total global cultivated land area has not changed since 1991 (O'Mara, 2012),
reflecting increased productivity and intensification efforts.
Livestock products are an important agricultural commodity for global food security because
they provide 17% of global kilocalorie consumption and 33% of global protein consumption
(Rosegrant et al., 2009). The livestock sector contributes to the livelihoods of one billion of the
poorest population in the world and employs close to 1.1 billion people (Hurst et al, 2005). There
is a growing demand for livestock products, and its rapid growth in developing countries has
been deemed the âlivestock revolutionâ (Thornton, 2010; Wright et al., 2012). Worldwide milk
production is expected to increase from 664 million tonnes (in 2006) to 1077 million tonnes (by
2050), and meat production will double from 258 to 455 million tonnes (Alexandratos and
Bruinsma, 2012). Livestock production is likely to be adversely affected by climate change,
competition for land and water, and food security at a time when it is most needed (Thornton,
2010).
Global climate change is primarily caused by GHG that result in warming of the atmosphere
(IPCC, 2013). The livestock sector contributes 14.5% of global GHG emissions (Gerber et al.,
2013), and thus may increase land degradation, air and water pollution, and declines in
biodiversity (Bellarby et al., 2013; Reynolds et al., 2010; Steinfeld et al., 2006; Thorton and
Gerber, 2010). At the same time, climate change will affect livestock production through

4

competition for natural resources, quantity and quality of feeds, livestock diseases, heat stress
and biodiversity loss. Therefore, the challenge is to maintain a balance between productivity,
household food security, and environmental preservation (Wright et al., 2012).
There is growing interest in understanding the interaction of climate change and agricultural
production and it is motivating a significant amount of research (Aydinalp and Cresser, 2008).
There is still limited research regarding the impacts of climate change on livestock production
(IPCC, 2014). This literature reviews the impacts of climate change on livestock production and
food security, and the livestock sectorâs contribution to climate change.
2.2

Topical Review

2.2.1 Impact of Climate Change on Livestock
Despite uncertainties in climate variability, the IPCC Fifth Assessment Report identified the
âlikely rangeâ of increase in global average surface temperature by 2100, which is between
0.3Â°C and 4.8Â°C (IPCC, 2013). The potential impacts on livestock include changes in production
and quality of feed crop and forage (Chapman et al., 2012; IFAD, 2010; Polley et al., 2013;
Thornton et al., 2009), water availability (Henry et al., 2012; Nardone et al., 2010; Thornton et
al., 2009), animal growth and milk production (Henry et al., 2012; Nardone et al., 2010;
Thornton et al., 2009), diseases (Nardone et al., 2010; Thornton et al., 2009), reproduction
(Nardone et al., 2010), and biodiversity (Reynolds et al., 2010). These impacts are primarily due
to an increase in temperature and atmospheric carbon dioxide (CO2) concentration, precipitation
variation, and a combination of these factors (Aydinalp and Cresser, 2008; Henry et al., 2012;
IFAD, 2010; Nardone et al., 2010; Polley et al., 2013; Reynolds et al., 2010; Thornton et al.,
2009). The impacts of climate change on livestock production factors are presented in Figure 1.
Temperature affects most of the critical factors for livestock production, such as water

5

availability, animal production, reproduction and health. Forage quantity and quality are affected
by a combination of increases in temperature, CO2 and precipitation variation. Livestock diseases
are mainly affected by an increase in temperature and precipitation variation.

6

Figure 1. Impacts of Climate Change on Livestock

7

2.2.1.1 Quantity and quality of feeds
Quantity and quality of feed will be affected mainly due to an increase in atmospheric CO2
levels and temperature (Chapman et al., 2012). The effects of climate change on quantity and
quality of feeds are dependent on location, livestock system, and species (IFAD, 2010). Some of
the impacts on feed crops and forage are:
â˘

Increase of CO2 concentration will result in herbage growth changes, with greater effect
on C3 species and less on grain yields (Chapman et al., 2012; Hatfield and Prueger, 2011;
Thornton et al., 2009, 2015). The effects of CO2 will be positive due to inducing partial
closure of stomata, reducing transpiration, and improving some plantsâ water-use
efficiency (Rotter and van de Geijn, 1999; Wand et al., 1999).

â˘

C4 species (which account for less than 1% of plants on Earth) are found in warm
environments, and have higher water-use efficiency than C3 plants. Temperature
increases to 30-35Â°C could increase herbage growth, with larger effects on C4 species.
However, the effects may vary depending on the location, production system used, and
plant species (Hatfield and Prueger, 2011; IFAD, 2010; Thornton et al., 2009; Thornton
and Herrero, 2010).

â˘

Changes in temperature and CO2 levels will affect the composition of pastures by altering
the species competition dynamics due to changes in optimal growth rates (IFAD, 2010;
Thornton et al., 2008; Thornton et al., 2009, 2015). Plant competition is influenced by
seasonal shifts in water availability (Polley et al., 2013). Primary productivity in pastures
may be increased due to changes in species composition if temperature, precipitation, and
concurrent nitrogen deposition increase (IPCC, 2007).

8

â˘

Quality of feed crops and forage may be affected by increased temperatures and dry
conditions due to variations in concentrations of water-soluble carbohydrates and
nitrogen. Temperature increases may increase lignin and cell wall components in plants
(Polley et al., 2013; Sanz-Saez et al., 2012), which reduce digestibility and degradation
rates (IFAD, 2010; Polley et al., 2013), leading to a decrease in nutrient availability for
livestock (Thornton et al., 2009). However, as CO2 concentration rises forage quality will
improve more in C3 plants than C4 plants. C3 plants also have greater crude protein
content and digestibility than C4 plants (Polley et al., 2013; Thornton et al., 2009; Wand
et al., 1999).

Impacts on forage quantity and quality depend on the region and length of growing season
(Polley et al., 2013; Thornton et al., 2009). An increase of 2Â°C will produce negative impacts on
pasture and livestock production in arid and semiarid regions and positive impacts in humid
temperate regions. The length of growing season is also an important factor for forage quality
and quantity because it determines the duration and periods of available forage. A decrease in
forage quality can increase methane emissions per unit of gross energy consumed (Benchaar et
al., 2001). Therefore, if forage quality declines, it may need to be offset by decreasing forage
intake and replacing it with grain to prevent elevated methane emissions by livestock (Polley et
al., 2013).
2.2.1.2 Water
Global agriculture uses 70% of fresh water resources, making it the worldâs largest consumer
(Thornton et al., 2009). However, global water demand is moving towards increased competition
due to water scarcity and depletion, where 64% of the worldâs population may live under waterstressful conditions by 2025 (Rosegrant et al., 2002).

9

Water availability issues will influence the livestock sector, which uses water for animal
drinking, feed crops, and product processes (Thornton et al., 2009). The livestock sector accounts
for about 8% of global human water use and an increase in temperature may increase animal
water consumption by a factor of two to three (Nardone et al., 2010). To address this issue, there
is a need to produce crops and raise animals in livestock systems that demand less water
(Nardone et al., 2010) or in locations with water abundance.
As sea level rises, more saltwater will be introduced into coastal freshwater aquifers (Karl et
al., 2009). Salination adds to chemical and biological contaminants and high concentrations of
heavy metals already found in waterbodies worldwide and may influence livestock production
(Nardone et al., 2010). Water salination could affect animal metabolism, fertility, and digestion.
Chemical contaminants and heavy metals could impair cardiovascular, excretory, skeletal,
nervous and respiratory systems, and impair hygienic quality of production (Nardone et al.,
2010).
There is a lack of research related to implications of reduced water availability for land-based
livestock systems due to climate change (Thornton et al., 2009). Therefore, it is important to
consider water availability and appropriate mitigation strategies in the context of sustainable
livestock production.
2.2.1.3 Livestock diseases
The effects of climate change on livestock diseases depend on the geographical region, land
use type, disease characteristics, and animal susceptibility (Thornton et al., 2009). Animal health
can be affected directly or indirectly by climate change, especially rising temperatures (Nardone
et al., 2010). The direct effects are related to the increase of temperature, which increases the
potential for morbidity and death. The indirect effects are related to the impacts of climate

10

change on microbial communities (pathogens or parasites), spreading of vector-borne diseases,
food-borne diseases, host resistance, and feed and water scarcity (Nardone et al., 2010; Thornton
et al., 2009; Tubiello et al., 2008).
Temperature increases could accelerate the growth of pathogens and/or parasites that live
part of their life cycle outside of their host, which negatively affects livestock (Harvell et al.,
2002; Karl et al., 2009; Patz et al., 2000). Climate change may induce shifts in disease spreading,
outbreaks of severe disease, or even introduce new diseases, which may affect livestock that are
not usually exposed to these type of diseases (Thornton et al., 2009). Evaluating disease
dynamics and livestock adaptation will be important to maintain their resilience. Global warming
and changes in precipitation affect the quantity and spread of vector-borne pests such as flies,
ticks, and mosquitoes (Thornton et al., 2009). In addition, disease transmission between hosts
will be more likely to happen in warmer conditions (Thornton et al., 2009). For example, White
et al. (2003) simulated the impacts of climate change on Australian livestock, finding that
livestock lost about 18% of their weight due to increased tick infestations. Wittmann et al. (2001)
also used a model to simulate the response of Culicoides imicola in Iberia, which is the main
vector of the bluetongue virus that affects mainly sheep and sometimes cattle, goat, and deer.
They reported that the vector would spread extensively with a 2Â°C increase in global mean
temperature. However, these predicted spreads may be prevented by disease surveillance and
technologies, such as DNA fingerprinting, genome sequencing, tests for understanding
resistance, antiviral medications, cross-breeding, and more (Perry and Sones, 2009; Thornton,
2010). Because of the dependence of diseases on animal exposure and multiple interactions of
factors, there are no methods to confirm actual disease risk (Randolph, 2008).

11

2.2.1.4 Heat stress
All animals have a thermal comfort zone, which is a range of ambient environmental
temperatures that are beneficial to physiological functions (FAO, 1986). During the day,
livestock keep a body temperature within a range of Âą0.5Â°C (Henry et al., 2012). When
temperature increases more than the upper critical temperature of the range (varies by species
type), the animals begin to suffer heat stress (FAO, 1986). Animals have developed a phenotypic
response to a single source of stress such as heat called acclimation (Fregley, 1996). Acclimation
results in reduced feed intake, increased water intake, and altered physiological functions such as
reproductive and productive efficiency and a change in respiration rate (Lacetera et al., 2003;
Nardone et al., 2010).
Heat stress on livestock is dependent on temperature, humidity, species, genetic potential, life
stage, and nutritional status. Livestock in higher latitudes will be more affected by the increase of
temperatures than livestock located in lower latitudes, because livestock in lower latitudes are
usually better adapted to high temperatures and droughts (Thornton et al., 2009). Confined
livestock production systems that have more control over climate exposure will be less affected
by climate change (Rotter and van de Geijn, 1999).
Heat stress decreases forage intake, milk production, the efficiency of feed conversion, and
performance (Haun, 1997; McDowell, 1968; Wyman et al., 1962). Warm and humid conditions
cause heat stress, which affects behavior and metabolic variations on livestock or even mortality.
Heat stress impacts on livestock can be categorized into feed nutrient utilization, feed intake,
animal production, reproduction, health, and mortality. The following presents these in more
detail:

12

Feed nutrient utilization and feed intake: Livestock have several nutrient requirements
including energy, protein, minerals, and vitamins, which are dependent on the region and type of
animal (Thornton et al., 2010). Failure to meet the dietary needs of cattle during heat stress
affects metabolic and digestive functions (Mader et al., 2003). Sodium and potassium
deficiencies under heat stress may induce metabolic alkalosis in dairy cattle, increasing
respiration rates (Chase, 2012).
Most of the research concerning feed intake in livestock animals has been focused on
cattle. Thermal livestock stress decreases feed intake (Mader and Davis, 2004; Thornton et al.
2009; Wyman et al., 1962) and efficiency of feed conversion (McDowell, 1968), especially for
livestock that are fed large amounts of high quality feeds (Haun, 1997). In the case of cattle, feed
intake reduction leads to a negative energy balance and reduced weight gain (Lacetera et al.,
1996; Lacetera et al., 2003). Reduction of water intake may also decrease sweating and feed
intake (Henry et al., 2012).
Animal production: One of the major causes of decreased production in the dairy and
beef industry is heat stress (Nardone et al., 2010) and significant economic losses have been
related to this. The United States livestock industry has an annual economic loss between 1.69
and 2.36 billion US dollars due to heat stress, of which 50% occurs in the dairy industry (StPierre et al., 2003). High-producing dairy cows generate more metabolic heat than lowproducing dairy cows. Therefore, high-producing dairy cows are more sensitive to heat stress.
Consequently, when metabolic heat production increases in conjunction with heat stress milk
production declines (Berman, 2005; Kadzere et al., 2002). Heat stress also affects ewe, goat, and
buffalo milk production (Finocchiaro et al., 2005; Nardone et al., 2010; Olsson and Dahlborn,

13

1989). However, less attention has been given to these animals because of their adaptability to
warm conditions and lower demand for their milk (Nardone et al., 2010).
In the case of meat production, beef cattle with high weights, thick coats, and darker
colors are more vulnerable to warming (Nardone et al., 2010). Global warming may reduce body
size, carcass weight, and fat thickness in ruminants (Mitloehner et al., 2001; Nardone, 2000). The
same is true in pig production, where larger pigs will have more reduction in growth, carcass
weight, and feed intake (Nardone et al., 2010). Pigletsâ survival may be reduced because of a
reduction of sows feed intake during suckling periods with temperatures greater than 25Â°C,
which reduces the milk yield of the sow (Lucas et al., 2000).
The poultry industry may also be compromised by low production at temperatures higher
than 30Â°C (Esminger et al., 1990). Heat stress on birds will reduce body weight gain, feed intake
and carcass weight, and protein and muscle calorie content (Tankson et al., 2001). Heat stress on
hens will reduce reproduction efficiency and consequently egg production because of reduced
feed intake and interruption of ovulation (Nardone et al., 2010; Novero et al., 1991). Egg quality,
such as egg weight and shell weight and thickness may also be negatively affected under hotter
conditions (Mashaly et al., 2004).
Reproduction: Reproduction efficiency of both livestock sexes may be affected by heat
stress. In cows and pigs, it affects oocyte growth and quality (Barati et al., 2008; Ronchi et al.,
2001), impairment of embryo development, and pregnancy rate (Hansen, 2007; Nardone et al.,
2010; Wolfenson et al., 2000). Cow fertility may be compromised by increased energy deficits
and heat stress (De Rensis and Scaramuzzi, 2003; King et al., 2006). Heat stress has also been
associated with lower sperm concentration and quality in bulls, pigs, and poultry (Karaca et al.,
2002; Kunavongkrita et al., 2005; Mathevon et al., 1998).

14

Health: Nardone et al. (2010) presented several livestock health problems related to
climate change. Prolonged high temperature may affect metabolic rate (Webster, 1991),
endocrine status (Johnson, 1980), oxidative status (Bernabucci et al., 2002), glucose, protein and
lipid metabolism, liver functionality (reduced cholesterol and albumin) (Bernabucci et al., 2006;
Ronchi et al., 1999), non-esterified fatty acids (NEFA) (Ronchi et al., 1999), saliva production,
and salivary HCO3- content. In addition, greater energy deficits affect cow fitness and longevity
(King et al., 2006).
Mortality: Warm and humid conditions that cause heat stress can affect livestock
mortality. Howden et al. (2008) reported that increases in temperature between 1 and 5Â°C might
induce high mortality in grazing cattle. As a mitigation measure, they recommend sprinklers,
shade, or similar management practices to cool the animals. Sirohi and Michaelowa (2007)
linked livestock mortality to several heat waves between 1994 and 2006 in the United States and
northern Europe.
More information is needed concerning the nature and extent of how heat stress affects feed
nutrient utilization and feed intake, animal production, reproduction, and health. With greater
knowledge related to nutritional and metabolism processes of livestock, management practices
could be adapted to increase animal performance.
2.2.1.5 Biodiversity
Biodiversity refers to a variety of genes, organisms, and ecosystems found within a specific
environment (Swingland, 2001) and contributes to human well-being (MEA, 2005). Populations
that are decreasing in genetic biodiversity are at risk, and one of the direct drivers of this
biodiversity loss is climate change (UNEP, 2012). Climate change may eliminate 15% to 37% of
all species in the world (Thomas et al., 2004). Temperature increases have affected species

15

reproduction, migration, mortality, and distribution (Steinfeld et al., 2006). The
Intergovernmental Panel on Climate Change Fifth Assessment Report states that an increase of 2
to 3Â°C above pre-industrial levels may result in 20 to 30% of biodiversity loss of plants and
animals (IPCC, 2014). By 2000, 16% of livestock breeds (ass, water buffalo, cattle, goat, pig,
sheep, and horse) were lost (Thornton et al., 2009). In addition, the FAO (2007) has stated that
from 7,616 livestock breeds reported, 20% were at risk, and almost one breed per month was
being extinguished. Cattle had the highest amount of extinct breeds (N=209) of all species
evaluated. The livestock species that had the highest percentages of risk of breed elimination
were chicken (33% of breeds), pigs (18% of breeds), and cattle (16% of breeds). However, the
breeds at risk depends on the region. Developing regions had between 7% and 10% of
mammalian species at risk (not restricted to livestock), but between 60% and 70% of mammalian
species are classified as of unknown risk. Conversely, in developed regions, where the livestock
industry is very specialized and based on a small number of breeds, the mammalian species at
risk were between 20% and 28% (FAO, 2007). Thornton et al. (2009) states that this biodiversity
loss is mainly because of the practices used in livestock production that emphasize yield and
economic returns and marginalization of traditional production systems where other
considerations are also important (such as ability to withstand extremes).
Livestock and plants will be highly affected by climate change and biodiversity loss. These
breeds and species cannot be replaced naturally; therefore, future work that studies the inherent
genetic capabilities of different breeds and identifies those that can better adapt to climate
conditions is vital.

16

2.2.1.6 Food security
About 842 million people (one in eight people worldwide) went hungry between 2011 and
2013, not receiving enough food to maintain an active and healthy life (FAO et al., 2013).
Livestock contributes greatly to food security because: (1) they are suppliers of global calories,
proteins, and essential micronutrients, (2) they are produced in areas that have difficulty growing
crops, (3) most of the feed for livestock is not appropriate for human consumption, and (4) they
provide manure for crop production (FAO, 2011). However, there are also concerns that
livestock production is detrimental to food security. First, the use of grains as feed in livestock
production is a worldwide concern because they are produced for animal feed and not for human
consumption. For example, in 2002, one-third of the global cereal harvest was used as livestock
feed (Steinfeld et al., 2006). The bulk of the livestock feed comes from grasses and legume
forage that grows on land not suitable to agriculture (O'Mara, 2012), and in many countries
livestock do not receive cereal supplements. In such areas, livestock are a positive contributor to
food security. The debate occurs in areas where cattle are pastured in areas perfectly suitable for
agriculture, or where they are fed substantial cereal supplements. Second, climate change, mostly
via an increase in temperature, may decrease intake of digestible nutrients. Therefore, livestock
production may decrease through declining forage quality and quantity and/or by reducing
animal feed intake. These two factors affect livestock production because animals will use the
available nutrients to first maintain their physiological needs, then for growth or milk production,
and finally for reproduction (Hatfield et al, 2008). Third, climate change also affects nutritional
content of livestock products because of potential increases in pathogens and diseases in their
food and effects on the animals themselves (Harvell et al., 2002; Karl et al., 2009; Patz et al.,
2000). As new pathogens and diseases emerge and spread, pesticide and veterinary medicine use

17

will change, consequently changing the principal transfer process of environmental contaminants
to food (Lake et al., 2012; van der Spiegel et al., 2012).
Sustainable livestock production needs more research, extension, and demonstration.
Livestock are an important contributor to food security, but it is important to maintain an
efficient conversion of natural resources to human food to sustain a neutral food balance (FAO,
2011). This can be accomplished through efficient production of protein from livestock (FAO,
2013). However, climate change will influence this conversion by affecting the nutritional
content of livestock products (Harvell et al., 2002; Karl et al., 2009; Patz et al., 2000) and
reducing livestock production (Hatfield et al, 2008). Currently, the livestock sectorâs best
approach to contribute to food security is by addressing the primacy of food balance (FAO,
2013; FAO, 2011).
2.2.2 Impact of Livestock on Climate Change
Livestock contribute 14.5% of the total annual anthropogenic GHG emissions globally
(Gerber et al., 2013). Livestock influence climate through land use change, feed production,
animal production, manure, and processing and transport (Figure 2). Feed production and
manure emit CO2, nitrous oxide (N2O), and methane (CH4), which consequently affects climate
change. Animal production increases CH4 emissions. Processing and transport of animal
products and land use change contributes to the increase of CO2 emissions (Figure 2).
The livestock sector is often associated with negative environmental impacts such as land
degradation, air and water pollution, and biodiversity destruction (Bellarby et al., 2013; Reynolds
et al., 2010; Steinfeld et al., 2006; Thorton and Gerber, 2010). Increases in livestock production
are expected to originate from a declining natural resource base, which will cause further

18

environmental damage without proper natural resources management (Thornton and Herrero,
2010).

19

Figure 2. Impact of livestock on Climate Change

20

2.2.2.1 GHG emissions
The primary livestock GHG emissions are CO2, CH4, and N2O. CH4 contributes the most to
anthropogenic GHG emissions (44%), followed by N2O (29%) and CO2 (27%) (Gerber et al.,
2013). Globally livestock contribute 44% of anthropogenic CH4, 53% of anthropogenic N2O and
5% of anthropogenic CO2 emissions (Figure 3). Higher concentrations of these gases, can be
explained by lower efficiency and productivity of livestock system due to excess loss of
nutrients, energy, and organic matter (Gerber et al., 2013).

Figure 3. Contribution of livestock to the total GHG anthropogenic emissions
The global warming potential (GWP) of CH4 and N2O over a 100-year time horizon differs
between the IPCC (2006), IPCC (2013), and the United Nations Framework Convention on
Climate Change (UNFCC, 2014). Current assessments have estimated higher GWP than what
21

was previously thought. IPCC (2006) is most commonly used in the literature for estimating
livestock GHG emissions, and reported that the warming potential of CH4 is 25 CO2-eq, while
N2O is 298 CO2-eq. However, the latest IPCC (2013) reported a warming potential for CH4 of 34
CO2-eq, while N2O is 310 CO2-eq. The UNFCCC (2014) stated that CH4 emissions have a
warming potential of 21 CO2-eq, and N2O is 310 CO2-eq over a 100-year time horizon.
The worldâs transportation sector emits around 5656 Tg CO2-eq yr-1 and the livestock sector
emits 7100 Tg CO2-eq yr-1 (DSI MSU, 2013; Gerber et al., 2013). Emissions from livestock
production contribute more GHG to the atmosphere than the entire global transportation sector.
The livestock sector contributes directly and indirectly to GHG emissions, including through
animal physiology, animal housing, manure storage, manure treatments, land application, and
chemical fertilizers (Casey et al., 2006; Monteny et al., 2001). Direct emissions from animal
sources include enteric fermentation, respiration, and excretions (Jungbluth et al., 2001). Indirect
emissions refer to emissions derived from feed crops, manure application, farm operations,
livestock products processing, transportation, and land use allocation for livestock production
(e.g. deforestation, desertification, carbon released from cultivated soils) (IPCC, 1997; Mosier et
al., 1998). In the livestock sector, indirect emissions play a greater role in the release of carbon to
the atmosphere than direct emissions (Steinfeld et al., 2006).
The livestock sectorâs contribution of 14.5% of total anthropogenic GHG emissions was
evaluated by Gerber et al. (2013) using a global livestock environmental assessment model
(GLEAM). GLEAM performs an analysis of the emissions of global livestock production along
supply chains. The main elements of the livestock supply chains that are analyzed by GLEAM
are: herd, feed, manure, animalsâ energy requirement, feed intake, production, and emissions,
allocation of the total emissions at the farmgate (physical farm boundaries) to co-products and

22

services (emission per kg of product), and post-farmgate emissions (transport and processing).
GLEAM also considers land use change as the conversion of forest to pasture or arable land for
crops.
The contributors of the 14.5% of livestock GHG emissions are presented in Figure 4. Enteric
fermentation is the largest contributor of the sectorâs emissions with 39.1%, followed by manure
management, application, and direct deposit with 25.9%, feed production with 21.1%, land use
change with 9.2%, post-farmgate with 2.9%, and direct and indirect energy with 1.8% (Gerber et
al., 2013). However, contribution to GHG emissions varies depending on the type of farming
system and region. For example, intensification of any of the three major livestock production
systems and expansion of industrialized (or landless) systems will increase CO2 emissions due to
greater fossil fuel use and less solar energy utilized by photosynthesis (Steinfeld et al., 2006).
Gerber et al., (2013) also estimated livestock GHG emissions by region, finding that, Asia
produce the highest, followed by Latin America and the Caribbean, Europe, North America,
Africa, and Oceania.

Figure 4. Global GHG emissions from the livestock sector (modified from Gerber et al., 2013)
23

2.2.2.2

Land use

Forests and natural habitats have been steadily converted to pasture and cropland since the
1850s (Goldewijk and Battjes, 1997). Agriculture lands cover about 38.5% of global total land
area, which consists of 28.4% arable land and 68.4% permanent meadows and pasture (DSI
MSU, 2015). Pasturelands have expanded by a factor of six since 1800, now covering 35 million
km2 (White et al., 2000). Agricultural land use change is related to two concepts: profit per unit
of land and opportunity cost (Steinfield et al., 2006a). Profit per unit of land refers to the
willingness of farmers to manage a specific land use. Profit will vary depending on several
factors, such as the landâs biophysical characteristics and price, access to markets, inputs, and
services. The opportunity cost concept compares the social and economic cost of different ways
to use the same land area. Opportunity costs include private production costs and ecosystem
service costs. Therefore, when non-marketable ecosystem services have no associated cost, land
use decisions are based on private profit per unit of land (Steinfeld et al., 2006).
The increasing demand for livestock products has significantly changed the natural
landscape. Land degradation is the deterioration of physical, chemical, and biological properties
of soil. Land degradation has been recognized as one of the drivers of land conversion from
forest to croplands and pastures because producers exhaust their soil resources and thus search
for more suitable land (Steinfeld et al., 2006). However, Asner et al., (2004) stated that due to
climate and soil characteristics, the expansion of pasture into marginal areas is limited; therefore,
they could only expand into areas with agro-ecological potential.
Land use change affects the natural carbon cycle, which consequently releases high amounts
of carbon into the atmosphere, increasing GHG emissions. Natural habitats, mainly forests,
sequester more carbon in soil and vegetation than croplands and pasturelands. Soil and terrestrial

24

vegetation sequester up to 40% of global anthropogenic CO2 emissions (The Royal Society,
2001). Furthermore, pasturelands contain more carbon than croplands. Croplands sequester 6%
of global carbon, while tropical savannas and temperate pasturelands together sequester 27%
(IPCC, 2000). However, soils sequester the most carbon in the terrestrial carbon cycle, and
double that of vegetation (Steinfeld et al., 2006). Sundquist (1993) estimated that 1,100 to 1,600
billion tonnes of carbon is stored in soils. However, soil carbon can be lost through burning,
volatilization, erosion, land use change, and agricultural management practices (Bolin et al.,
1982). Therefore, when a forest is converted to cropland and pasture by logging or burning, high
amounts of carbon are released into the atmosphere (Steinfeld et al., 2006).
Latin America has converted the most land from forest to pasture and croplands, and
livestock ranching is one of the drivers of this change (Wassenaar et al., 2007). In the past 40
years, forested areas in Central America decreased by almost 40%, coinciding with an increase
of pasturelands and cattle herds. In addition, crops used to feed livestock also affect land use
change. In 2004-2005 alone, soybean expansion replaced 1.2 million hectares of rainforest
(LEAD, 2014).
Deforestation, cultivated soils, and land degradation due to livestock production are the main
source of CO2 emissions. From total livestock GHG emissions, 9.2% is attributed to land use
change, where 6% is due to pasture expansion and 3.2% is due to feed crop expansion (Gerber et
al., 2013). However, land use change produces other emissions in addition to CO2. Land
conversion from forest to pastureland may also reduce CH4 oxidation by soil microorganisms,
resulting in pasturelands acting as net sources of CH4 when soil compaction from cattle hooves
limits gas diffusion (Mosier et al., 2004).

25

Regarding the other two contributors of livestock-land use GHG emissions, Steinfeld et al.
(2006) estimated that livestock-related cultivated soils produce around 28 million tonnes of CO2
per year and livestock-induced desertification of pastures produce 100 million tonnes of CO2 per
year. Reducing pasturelands as a mitigation strategy does not suggest an increase in soil carbon
stocks because the relationship between pasturelands and soil carbon sequestration is complex
due to their dependence on the environment, society, and economy (IFAD, 2010). Studies
suggest that pasturelands can either increase or decrease GHG emissions depending on the
grazing management and history, climate and ecosystem (IFAD, 2010; Henderson et al., 2015).
Therefore, grazing management that can increase carbon sequestration are: i) not exceeding
pastureland carrying capacity by having an effective stocking rate, ii) rotational grazing, and iii)
excluding degraded pasturelands from livestock grazing (IFAD, 2010; Tennigkeit and Wilkes,
2008).
2.2.2.3

Feed production

The use of manure and synthetic fertilizers for forage and feed crop production, processing of
feed, and transport of feed are the most important contributors of GHG emissions related to the
livestock sector (IFAD, 2010; Thornton and Herrero, 2010). These make up 45% of global
livestock anthropogenic GHG emissions, consisting primarily as CO2, N2O and NH4 (Gerber et
al., 2013).
The livestock sector contributes significantly to GHG emissions through the production of
nitrogenous fertilizers used to produce crops for animal feed (Steinfeld et al., 2006). Oil, coal,
and natural gas are used in fertilizer manufacture. By considering the amount of fertilizer use,
packaging, transport, and application in the livestock sector, the manufacturing process of
fertilizers contributes more than 40 million tonnes of CO2 annually (Steinfeld et al., 2006). Forty

26

percent of all the nitrogen up-take by crops comes from synthetic fertilizers (Smil, 2001).
Ammonia volatilization loss from synthetic nitrogen fertilizer is an indirect contributor to GHG
emissions. Between 4 to 5 million tonnes of mineral fertilizer is used for livestock feed
production. The average loss due to ammonia volatilization from mineral fertilizer is about 14%.
Therefore, it was estimated that the livestock sector contributes 3.1 million tonnes of global
ammonia volatilization from mineral fertilizers per year (Steinfeld et al., 2006).
N2O is another contributor to GHG emissions. Fertilizer use, agricultural nitrogen fixation,
and atmospheric nitrogen deposition generally increase N2O emissions (Bouwman, 1996). By
following the same assumptions as ammonia, with a 1% average of N2O-N (tonnes of nitrogen in
nitrous oxide form) loss rate from mineral fertilizer, the livestock sector contributes 0.2 million
tonnes N2O-N of global N2O emission from mineral fertilizer per year (Steinfeld et al., 2006). In
addition, leguminous feed crops for livestock account for additional N2O emissions. Steinfeld et
al. (2006) estimated their contribution by considering the global area of soybeans used for
livestock feed and doubling it to include alfalfa and clover, because there are no estimates of
worldwide alfalfa and clover production. Therefore, the contribution of leguminous feed crops is
more than 0.5 million tonnes per year of N2O-N emissions. By adding both contributors (mineral
fertilizer and leguminous feed crops), the total N2O-N emissions are 0.7 million tonnes per year
(Steinfeld et al., 2006). As growth in fertilizer and manure use continues, a 35-60% increase of
N2O emissions (0.9 to 1.1 million tonnes per year of total N2O-N emissions) is expected by 2030
(Bruinsma, 2003).
On-farm fossil fuel use in livestock production produces 50% more CO2 emissions than
manufacturing N fertilizers for feed. The livestock sector includes direct and indirect (e.g.
electricity) on-farm fossil fuel use, which is used for machinery operations, irrigation, heating,

27

cooling, ventilation, production of herbicides and pesticides, and more. More than half of fossilfuel use is attributed to feed production. By assuming CO2 emissions from on-farm fossil fuel
use are double that of manufacturing N fertilizers, and adding emissions related to livestock
rearing, on-farm fossil fuels account for 90 million tonnes of CO2 per year (Steinfeld et al.,
2006).
2.2.2.4

Animal production

In general, livestock respiration is not counted as a net source of CO2 emissions because they
are part of the global biological system cycle. The vegetation consumed by the animal originates
from the conversion of atmospheric CO2 to organic compounds or biomass. Therefore, under the
Kyoto Protocol it is assumed that the consumed amounts of CO2 in vegetative form are
equivalent to those emitted by the livestock. Conversely, the animal is a carbon sink because a
fraction of the carbon consumed is absorbed in the live tissue of the animal (UNFCCC, 1998)
and products such as milk.
Livestock contributes 44% of the worldâs anthropogenic CH4 emissions through their normal
digestive processes (enteric fermentation) and manure management (Gerber et al., 2013). Enteric
fermentation and manure account for 80% of 52 agricultural emission sources (Steinfeld et al.,
2006). During the animalsâ digestive process, enteric fermentation converts the feed consumed
into digestible feed. Enteric fermentation releases a CH4 by-product through exhalation
(Beauchemin et al., 2009). Therefore, this by-product is considered an energy loss (Gerber et al.,
2013). Feed composition and feed intake can vary enteric fermentation and hence methane
emissions. Increasing the concentrate (high energy feeds containing cereal grains and oil meals)
proportion in the animal diet can reduce methane emissions from the animal (Dourmad et al.,
2008; Yan et al., 2000).

28

Methane emissions vary depending on production systems and regional characteristics (e.g.
climate and landscape) (Gerber et al., 2013). The enteric fermentation produced by ruminant
livestock (e.g. cattle, sheep, and goat) emits globally between 87 and 94 Tg of methane annually
(IPCC, 2013). Mixed crop-livestock systems account for 64% of global enteric fermentation
methane emissions; grazing systems account for 35%, and industrial 1% (Steinfeld et al., 2006).
The high percentage from mixed crop-livestock systems reflects that two-thirds of total livestock
animals are present in those systems (Steinfeld et al., 2006). The countries that contribute the
most methane emissions related to livestock production are India, China, Brazil, and the United
States (IPCC, 2013; Olivier and Janssens-Maenhout, 2012). India, with the largest livestock
population in the world, emitted 11.8 Tg of CH4 in 2003, from which 91% derives from enteric
fermentation and 9% from manure management (Chhabra et al., 2013).
In Africa, methane emissions are expected to increase due to increases in livestock
populations. Herrero et al. (2008) estimated that African cattle, goats, and sheep, which produced
about 7.8 million tonnes of methane in 2000, are likely to increase to 11.1 million tonnes by
2030. If this linear relationship between methane emissions and livestock population continues,
global methane emissions from livestock production may increase 60% by 2030 (Bruinsma,
2003). However, changing feeding practices and manure management could moderate methane
emissions (Thornton and Herrero, 2010).
2.2.2.4.1 Emissions by Species and Commodities
Animals that contribute the most to livestock GHG emissions are beef and dairy cattle,
accounting for 65% of the total livestock GHG emissions (Gerber et al., 2013). Pigs, poultry,
buffaloes, and small ruminants contribute about 7 to 10%. If GHG emissions are estimated based
on commodities, beef cattle contribute the most with 41% of the sectorâs emission, followed by

29

dairy cattle (20%), swine (9%), buffalo (8%), poultry (8%), and small ruminant (6%) (Gerber et
al., 2013).
Enteric fermentation is the largest source of GHG emissions from cattle, buffalo, and small
ruminants, comprising between 43% and 63% of the livestock sector emissions (Figure 5).
However, for pigs and chickens the largest source of emissions is due to feed production
(between 25% and 27%), which includes fertilizer production, machinery use, and feed
transportation. Enteric fermentation from pigs is much lower than in ruminants because their
digestive process does not produce as much methane as a by-product (Gerber et al., 2013).

Figure 5. Total global livestock GHG emission by specie and product (Adapted from Gerber et
al., 2013)
2.2.2.5 Manure
Livestock manure releases CH4 and N2O gas. The decomposition of the organic materials
found in manure under anaerobic conditions releases methane (EPA, 1999). Liquid manure
found in lagoons or holding tanks releases more methane than dry manure (Burke, 2001).

30

Manure methane emissions are a function of air temperature, moisture, pH, storage time, and
animal diet (EPA, 1999; IFAD, 2010). Steinfeld et al. (2006) estimated global methane
emissions from manure decomposition of 17.5 million tonnes of CH4 per year. Pig manure
comprises almost half of global manure-related methane emissions. At the country level, China
has the highest global methane manure-related emissions, primarily due to pig manure (Steinfeld
et al., 2006).
N2O emissions from manure storage are dependent on environmental conditions, handling
systems, and duration of waste management. Manure must be handled aerobically and then
anaerobically to release N2O emissions, which is more likely to occur in dry waste-handling
systems. Steinfeld et al. (2006) reported that N2O emissions from stored manure are equivalent to
10 million tonnes N per year.
Nitrous oxide soil emissions from manure application are the largest source of global N2O
emissions (Steinfeld et al., 2006). Nitrogen emissions from applied or deposited manure are
dependent on soil infiltration, organic carbon amount, pH, soil temperature, precipitation, and the
plant/crop uptake rate (Mosier et al., 2004). Steinfeld et al. (2006) estimated that 1.7 million
tonnes of manure soil N2O are released per year. N2O emissions from applied manure are 40%
higher in mixed crop-livestock systems than the N2O emissions from excreted manure deposited
on pasture systems. Industrial production systems have 90% less N2O emissions than mixed
crop-livestock systems (Steinfeld et al., 2006).
2.2.2.6

Processing and transport

Energy costs of processing animals and their products combined with global livestock
production from âmarket-oriented intensive systemsâ can be used to obtain global processing
emissions. However, the source of the energy and its variation in the world is uncertain (WEC,

31

2015). Energy use depends on the type of livestock system and if they are small or large scale.
More than half of the energy used in confinement systems is for feed production, including seed,
herbicides, pesticides, and machinery. Substantial energy is also used for heating, cooling, and
ventilation systems (USDA-NRCS, 2006).
Nonetheless, some approximate estimations of energy use in the livestock sector have been
developed. Based on a study performed in Minnesota by Sainz (2003) of energy use for
processing, Steinfeld et al. (2006) estimated that the United States produces a âfew millionâ
tonnes of CO2 emissions related to total animal product and feeding processing. Following the
same trend, they estimated that the world produces âseveral tens of millionsâ tonnes CO2
emissions in animal-product processing.
Transportation of livestock products to retailers and transport of feed to livestock farms
contribute to GHG emissions. Long distance shipping is the most significant GHG emitter in this
category. For example, high volumes of soybean are transported long distances to be used as
feed (Steinfeld et al., 2006). Annual CO2 emissions in meat transportation based on FAO
statistics for 2001-2003 was estimated to be between 800-850 thousand tonnes of CO2 (Steinfeld
et al., 2006). These estimations lead to a greater understanding of the contribution of processing
and transportation to the livestock sector GHG burden. However, more research is needed to
obtain approximate estimations of CO2 emissions related to processing and transport of livestock
products.
2.2.3

Summarizing Adaptation and Mitigation Practices
There are several climate change adaptation and mitigation recommendations that can be

made based on the above discussion. Adaptation strategies can improve the resilience of crop
and livestock productivity to climate change (USDA, 2013). Mitigation measures could

32

significantly reduce the impact of livestock on climate change (Dickie et al., 2014). Adaptation
and mitigation can make significant impacts if they become part of national and regional policies
(FAO, 2009b).
2.2.3.1

Adaptation measures
Adaptation measures involve production and management system modifications,

breeding strategies, institutional and policy changes, science and technology advances, and
changing farmersâ perception and adaptive capacity (IFAD, 2010; Rowlinson et al., 2008;
USDA, 2013). Research is needed on assessments for implementing these adaptation measures
and tailoring them based on location and livestock system. This could be accomplished with GIS
and remote sensing technologies applicable at broad and local scales (Thornton et al., 2008).
2.2.3.1.1 Livestock production and management systems
An adaptation such as the modification of production and management systems involves
diversification of livestock animals and crops, integration of livestock systems with forestry and
crop production, and changing the timing and locations of farm operations (IFAD, 2010).
Diversification of livestock and crop varieties can increase drought and heat wave
tolerance, and may increase livestock production when animals are exposed to temperature and
precipitation stresses. In addition, this diversity of crops and livestock animals is effective in
fighting against climate change-related diseases and pest outbreaks (Batima et al., 2005; IFAD,
2010; Kurukulasuriya and Rosenthal, 2003).
Agroforestry (establishing trees alongside crops and pastures in a mix) as a land
management approach can help maintain the balance between agricultural production,
environmental protection and carbon sequestration to offset emissions from the sector.

33

Agroforestry may increase productivity and improve quality of air, soil, and water, biodiversity,
pests and diseases, and improves nutrient cycling (Jose, 2009; Smith et al., 2012).
Changes in mixed crop-livestock systems are an adaptation measure that could improve
food security (Herrero et al., 2010; Wani et al., 2009). This type of agricultural system is already
in practice in two-thirds of world, producing more than half of the milk, meat, and crops such as
cereal, rice and sorghum (Herrero et al., 2012). Changes in mixed crop-livestock systems can
improve efficiency by producing more food on less land using fewer resources, such as water
(Herrero et al., 2012; Steinfeld et al., 2006).
Shifting locations of livestock and crop production could reduce soil erosion and improve
moisture and nutrient retention (Kurukulasuriya and Rosenthal, 2003). Another adaptive measure
could be adjusting crop rotations and changing timing of management operations (e.g. grazing,
planting, spraying, irrigating). This measure can be adapted to changes in duration of growing
seasons, heat waves and precipitation variability (Batima et al., 2005; IFAD, 2010;
Kurukulasuriya and Rosenthal, 2003).
2.2.3.1.2 Breeding strategies
Changes in breeding strategies can help animals increase their tolerance to heat stress and
diseases and improve their reproduction and growth development (Henry et al., 2012;
Rowlinson, 2008). Therefore, the challenge is in increasing livestock production while
maintaining the valuable adaptations offered by breeding strategies, all of which will require
additional research (Thornton et al., 2008). In addition, policy measures that improve adaptive
capacity by facilitating implementation of adaptation strategies will be crucial (USDA, 2013).
For example, developing international gene banks could improve breeding programs and serve as
an insurance policy, such as has been done for plants with the In-Trust plant collections in the

34

CGIAR gene banks (Thornton et al., 2008). This would be a major breakthrough that requires
significant investment and international collaboration to succeed.
2.2.3.1.3 Farmersâ perception and adaptive capacity
One of the limiting factors for these changes to succeed is the disposition and capability
of farmers to recognize the problem and adopt climate change adaptation and mitigation
measures (Jones et al., 2013). Because of this, it is important to collect information about
farmersâ perceptions to mitigation and adaptation measures. One approach for collecting
information about farmersâ perceptions that has been used for mitigation and adaptation research
is qualitative; using open-ended survey questions or group discussion at workshops to understand
individual and group opinions (Barnes et al., 2013). By understanding farmersâ perceptions and
including them in rural policy development, there is a greater chance of accomplishing food
security and environmental conservation objectives (Barnes et al., 2013; Oliver et al., 2012).
Risk perception within farmer decision-making can be increased through education,
family farm succession, and social interaction among farmers and farming communities. Barnes
et al. (2013) applied a latent class clustering approach (which uses statistical methodology to
construct results) to evaluate the heterogeneity of dairy farmersâ risk perception of climate
change. Their results show that the drivers of risk perception due to climate change are family
members and the influence of succession planning. They recommended increasing the social
capital of farming communities to promote acceptance of communication strategies for climate
change adaptation and mitigation measures.
2.2.3.2

Mitigation measures
There is potential to reduce livestock sector GHG emissions through the implementation

of different technologies and practices. However, they are not widely used (Gerber et al., 2013).

35

Some of the technical options for mitigating the impact of livestock on climate change are
carbon sequestration, improving diets to reduce enteric fermentation, improving manure
management, and more efficient use of fertilizers (Steinfeld et al. 2006; Thornton and Gerber,
2010; UNFCCC, 2008). Mitigation measures need public policy support to be effective (Dickie
et al., 2014).
2.2.3.2.1 Carbon sequestration
Carbon sequestration can be achieved through decreasing deforestation rates, reversing of
deforestation by replanting (Carvalho et al., 2004), targeting for higher-yielding crops with better
climate change adapted varieties, and improvement of land and water management (Steinfeld et
al., 2006). A beef sector study performed in Brazil estimated a reduction of up to 25% of GHG
emissions related to grazing land use and land use change, accomplished by improving animal
and herd efficiency (Gerber et al., 2013).
Soil organic carbon can be restored in cultivated soils through conservation tillage,
erosion reduction, soil acidity management, double-cropping, crop rotations, higher crop
residues, mulching and more (Paustian et al., 1997; Steinfeld et al., 2006). Improving pasture
management can also lead to carbon sequestration by incorporating trees, improving plant
species, legume interseeding, introducing earthworms, and fertilization (Conant et al., 2001). In
addition, grass productivity and soil carbon sequestration could be improved by increasing
grazing pressure in grasslands that have a lower amount of grazing animals than the livestock
carrying capacity (Holland et al., 1992). Improving grazing land management could sequester
around 0.15 gigatonnes CO2-eq yr-1 globally (Henderson et al., 2015).

36

2.2.3.2.2 Enteric fermentation
Enteric fermentation is a source of methane emissions that can be reduced through
practices such as improvement of animal nutrition and genetics (EPA, 1999). Examples of
practices for mitigating enteric fermentation are: increasing dietary fat content (Beauchemin et
al., 2008; Martin et al., 2010), providing higher quality forage (Hristov et al., 2013), increasing
protein content (ICF International, 2013), providing supplements (e.g. bovine somatotropin, feed
antibiotics) (Boadi et al., 2004), and the use of antimethanogens (vaccines to suppress methane
emissions) (EPA, 2013). However, there is high uncertainty in the efficacy of these practices
because various studies have demonstrated that the initial reductions of enteric fermentation
achieved are only temporary (ICF International, 2013).
A one percent increase of dietary fat can decrease enteric methane emissions between 4
to 5% (Beauchemin et al., 2008; Martin et al., 2010). However, rumens need to limit fat content
to 8% of dry matter to avoid a decrease in livestock performance (Hales et al., 2012). Providing
higher quality forage also results in a reduction of methane emissions because it increases
digestibility (Hristov et al., 2013). An increase of protein content of feed can also improve
digestibility and reduce overall methane emissions per unit of product (ICF International, 2013).
Providing supplements, such as feed antibiotics, which tend to increase weight gain and
reduce feed intake per metric ton of meat produced, can reduce enteric fermentation (Boadi et al.,
2004). In the case of milk, bovine somatotropin (a bovine growth hormone) increases production.
An increase in milk production leads to less animals needed to produce the same amount of milk
and less emissions produced (EPA, 2013). Antimethanogen vaccines are another practice that
directly reduces methane emissions in the rumen. However, this is a new technology with limited
research on emission reduction efficiency and animal health (EPA, 2013).

37

2.2.3.2.3 Manure management
Most methane emissions from manure management are related to storage and anaerobic
treatment. Although manure deposited on pasture can produce nitrous oxide emissions, the
mitigation measures are often difficult to apply because of the manure dispersion on pasture
(Dickie et al., 2014). Therefore, most mitigation practices involve shortening storage duration,
improving timing and application of manure, used of anaerobic digesters, covering the storage,
using a solids separator, and changing the animal diets (ICF International, 2013).
Anaerobic digestion can reduce methane emissions while producing biogas (Gerber et al.,
2008). Anaerobic digesters are lagoons or tanks that maintain manure under anaerobic conditions
to capture biogas and combust it for producing energy or flaring. This process reduces the
potential of GHG emissions by converting methane into CO2 (ICF International, 2013).
Unfortunately, anaerobic digesters are costly for producers; the best approach for implementing
digesters is through policies that create enough incentive for adaptation (Dickie et al., 2014).
Similar to digesters, the covering of ponds, tanks or lagoons reduces emissions by capturing and
destroying methane (ICF International, 2013).
Other storage and handling practices can also reduce GHG emissions. Such practices
include reducing storage time, improving housing and waste management systems to handle
manure, and removing bedding from manure by using a solids separator (Dickie et al., 2014).
The solids separator is mostly used in confinement systems to remove solids from manure
streams that are entering the treatment or storage systems. By removing the solids from manure
streams methane emissions are reduced, the time between storage system cleaning is increased,
and crust formation is prevented (ICF International, 2013). These practices, compared to

38

anaerobic digesters, are usually low-cost and low-tech. However, they require more time and
effort from the producer (Dickie et al., 2014).
Adjusting animal diets can also be used as a mitigation measure, by changing the volume
and composition of manure. GHG emissions can be reduced by balancing dietary proteins and
feed supplements. If protein intake is reduced, the nitrogen excreted by animals can also be
reduced. Supplements such as tannins are also known to have the potential to reduce emissions.
Tannins are able to displace the nitrogen excretion from urine to feces to produce an overall
reduction in emissions (Dickie et al., 2014; Hess et al., 2006).
2.2.3.2.4 Fertilizer management
Fertilizer application on animal feed crops increases nitrous oxide emissions (Bouwman,
1996). Therefore, mitigation measures such as increasing nitrogen use efficiency, plant breeding
and genetic modifications (Dickie et al., 2014), using organic fertilizers (Denef et al., 2011),
regular soil testing, using technologically advanced fertilizers, and combining legumes with
grasses in pasture areas may decrease GHG emissions in feed production (Dickie et al., 2014).
Nitrogen use efficiency can be improved by applying the required amount that the crop
will absorb and when it needs the nutrients, and placing it where the plant can easily reach it.
Regular soil testing can be a part of a nutrient management plan depending on the region and
crop, and improve efficiency of nitrogen use (Dickie et al., 2014). Plant breeding and genetic
modifications can reduce the use of fertilizers by increasing a cropâs nitrogen uptake (Dickie et
al., 2014). Increasing the use of organic fertilizers would also decrease emissions because
organic fertilizers do not produce as much nitrogen oxide as synthetic fertilizers (Denef et al.,
2011). Furthermore, fertilizer technology has improved through regulating the release of
nutrients from the fertilizer and inhibiting nitrification to slow the degradation of the fertilizer

39

and maintain the nutrients available for the plant. However, these technologically advanced
fertilizers are more costly than the other practices mentioned above (Dickie et al., 2014). In the
case of pasturelands, the use of synthetic nitrogen can be reduced by combining legumes with
grasses. Legumes fix nitrogen through Rhizobium bacteria; therefore, the need for
supplementary nitrogen is reduced (USDA-NRCS, 2007).
2.2.3.2.5 Shifting human dietary trends
Most studies are focused on reducing GHG emissions on the supply-side of the livestock
production system. However, less research has focused on the demand section related to
consumption of livestock products. This mitigation measure goes along with the policy related to
a reduction in meat consumption, which may significantly reduce GHG emissions. Because beef
accounts for a large portion of GHG emissions from the livestock sector and it is the least
resource-efficient animal protein producer (Stehfest et al., 2009), the mitigation potential is high
for the beef component of the livestock sector. Research to understand why populations feel
compelled to increase animal protein consumption when they rise above the poverty line is
needed, as well as why those at the top of the economic ladder are compelled to improve their
diets by reducing meat consumption and returning to a more vegetarian diet. This conundrum
stands at the center of the challenge faced by the state of knowledge and policies surrounding the
livestock sector.
2.3

Summary
Climate change will affect livestock production and consequently food security.

Livestock production will be negatively impacted (due to diseases, water availability, etc.),
especially in arid and semiarid regions. In addition, climate change will affect the nutritional
content of livestock products, which are one of the suppliers of global calories, proteins and

40

essential micronutrients. Conversely, livestock production also influences climate change.
Deforestation due to expansion of pasturelands and croplands for livestock production
contributes 9.2% of total livestock GHG emissions (Gerber et al., 2013). However, the feed
production stage contributes the greatest fraction (almost half) of GHG emissions across the
complete livestock production process. It is expected that this stage will further increase its
contribution due to intensification of livestock production. Meanwhile, enteric fermentation is
the largest GHG contributor in the animal production stage. Therefore, if livestock numbers
continue to increase and feeding practices are not changed, global emissions due to livestock
production will continue to increase.
Climate change adaptation, mitigation practices, and policy frameworks are critical to protect
livestock production. Among the reviewed studies, diversification of livestock animals (within
species), using different crop varieties, and shifting to mixed crop-livestock systems seem to be
the most promising adaptation measures. By diversifying animal and crop varieties, the tolerance
to climate variability (e.g. drought, heat waves) and to diseases and pest outbreaks will be
improved. In addition, shifting to mixed crop-livestock systems can improve efficiency by
increasing production with the use of fewer resources. On the mitigation side, improvement of
animal nutrition and genetics are important because enteric fermentation is a major GHG emitter
in livestock production. However, the efficacy of these practices in reducing emissions is
uncertain and more research is needed concerning effective mitigation practices related to enteric
fermentation.
If we want effective adaptation and mitigation measures to address climate change and
livestock production, these measures should be scaled up through policy. For example,
understanding farmersâ perceptions and including them in policy development can improve food

41

security and environmental conservation by promoting widespread practice adoption. In addition,
a comprehensive view of costs, time, and effort required from the producer needs to be included
to the policy framework to maintain sustainable production systems.
Interactions between climate change and livestock production are still not well
understood, despite the amount of research performed. First, most studies were performed at a
continental or regional level; to assess the most vulnerable areas, local studies will be critical.
Second, most of the research concerning livestock production focuses on cattle; more studies
must be performed on non-ruminants. Third, there is a gap in research related to water
availability for livestock production. With projected increases in drought and declining water
quality in many places on earth, it is critical to identify the regions with the best conditions for
livestock production and improve the conditions for those that do not. Fourth, climate change
may induce livestock diseases (e.g. outbreaks of severe diseases or new diseases), affecting
animals that are not usually exposed to those diseases; there is a need to evaluate the dynamics of
those diseases on livestock and how animals adapt to them. Fifth, there is a gap in knowledge
related to the nutritional and metabolic processes of livestock; addressing this can improve
management practices that increase animal performance. Sixth, because breeds and species are
not a renewable resource, it is important to identify breeds with inherent genetic capabilities to
adapt to climate change. Seventh, there are large uncertainties in using only energy costs of
processing and transporting animals and their products to estimate GHG emissions at this stage
of the livestock production process. Therefore, a better approximate estimation of CO2 emissions
related to processing and transport of livestock products is needed. Eight, there is a lack of
information related to the current use of adaptation and mitigation measures defined by location
and livestock system.

42

3

INTRODUCTION TO METHODOLOGY AND RESULTS

This dissertation consists of two research papers that have been published and submitted
to scientific journals. The first study focuses on the development of a sensitivity analysis for
evaluating the impacts of climate change on grazing dairy systems, while identifying an efficient
adaptation measure. The second study builds upon the first study by introducing a new measure
called food footprint to evaluate sustainable practices on grazing dairy systems by considering
water, energy, carbon, and economic footprints.
The first paper, entitled âPasture diversification to combat climate change impacts on
grazing dairy productionâ, introduces a new concept for evaluating the resiliency of pasture
diversification as an adaptation measure on grazing systems to reduce the negative impacts of
climate change. To accomplish this, a representative farm was developed based on grazing dairy
farm practices surveys within the Michigan Lower Peninsula. Using a farm system model, the
impacts of current and future climate scenarios were evaluated at the farm level. For each climate
model scenario, the adaptation measure was tested for 48 different pasture compositions based
on the four typical grass and two legume species within the study area. Statistical analysis was
performed to understand the effectiveness of the adaptation measures based on economic and
resource use criteria. Finally, in order to understand the resiliency of each pasture composition, a
sensitivity analysis was performed based on each criterion.
The second paper, entitled âFood Footprint as a Measure of Sustainability for Grazing
Dairy Farmsâ uses the same representative grazing dairy farm and 48 pasture compositions
introduced in the first study to evaluate the sustainability of the farm by considering water,
energy, carbon, and economic footprints in 10 different locations in Michigan. The farm model
IFSM was employed to estimate milk production and required information to calculate energy,

43

water, carbon, and economic footprints for different combinations of pasture compositions, target
milk productions, and locations scenarios. A multi-criterion decision making method was used to
calculate the new sustainability measure called food footprint. By obtaining the food footprint
and performing the cluster analysis, the most sustainable milk production and appropriate pasture
composition for each location was identified.

44

4

PASTURE DIVERSIFICATION TO COMBAT CLIMATE CHANGE IMPACTS ON
GRAZING DAIRY PRODUCTION

4.1

Introduction
Climate change refers to long-term changes in patterns of temperature, precipitation,

wind, and other climatological factors (EPA, 2016). The Intergovernmental Panel on Climate
Change (IPCC) Fifth Assessment Report identified the âlikely rangeâ of increase in global
average surface temperature between 0.3Â°C and 4.8Â°C by the end of this century (IPCC, 2013).
Increase in temperature will have a domino effect on hydrological cycle elements that ultimately
increase the frequency of extreme events such as heat waves, droughts, and floods (Melillo et al.,
2014).
The global impacts of climate change on societies and nature are already evident and are
expected to increase in frequency during this century and beyond (Melillo et al., 2014). In the
agricultural sector, significant change in both production and crop/animal health can be expected.
Climate change will influence temperature and precipitation patterns, which can result in
changes in production levels, shifts in production length and time, and severe damages due to
extended drought or flood. Furthermore, each change can affect agricultural products price and
producersâ income. Climate change may also impact animal and crop health by changing the
quantity and spread of different pests and diseases (FAO, 2010a; Hatfield et al., 2014). Several
studies have indicated that climate change could reduce agriculture production mainly in areas
that currently suffer from high levels of food insecurity (FAO, 2010a; Hatfield et al., 2014;
Adhikari et al., 2015). Meanwhile, increased agricultural production is needed to support the
projected population growth from 7.2 to 9.6 billion by 2050 (FAO, 2009a).

45

Livestock is one of the major agricultural subsectors and the demand for livestock
products is expected to double by 2050 (Garnett, 2009). At the same time, livestock production
will be impacted by climate change. The impacts of climate change on livestock production
include variations in: quality and quantity of feed crop and forage (Calvosa et al., 2009;
Thornton et al., 2009; Chapman et al., 2012; Polley et al., 2013), availability of water (Thornton
et al., 2009; Nardone et al., 2010; Henry et al., 2012), quantity and quality of livestock products
(Thornton et al., 2009; Nardone et al., 2010; Henry et al., 2012), plant and animal diseases
(Thornton et al., 2009; Nardone et al., 2010), and animal reproduction (Nardone et al., 2010).
Among livestock systems, grazing is likely to be more affected by climate change than
confinement systems because: a) pasture composition is one of the main components of a grazing
system (Heckman et al., 2007) and b) feed quality and availability on pastureland will be
determined by the climate conditionsâ impacts on the growing season (Hatfield et al., 2014).
Meanwhile, few studies have been performed on the impacts of climate change on grazing
livestock systems (Gauly et al., 2013; IPCC, 2014) due to the fact that these impacts are usually
of a local concern.
Graux et al. (2013) used the Pasture Simulation model (PaSim) to evaluate climate
change impacts on the agronomic and environmental services at 12 grassland sites in France.
They evaluated the interaction of intensive and extensive managements on permanent and sown
grasslands under past (1970-1999) and future climate scenarios (2020-2049 and 2070-2099).
Results showed that by the end of the century all sites will shift to more arid climates resulting in
an increased interannual and seasonal variability of production and decreased milk production.
However, the role of economics or adaptation options, which may offset the impacts of climate
change on the production were not considered in this study. HĂśglind et al., (2013) evaluated the

46

impacts of climate change on the yield of two grass species in 14 locations in Northern Europe
(Iceland, Scandinavia, Baltic countries) by using local-scale near-future climate scenarios (20402065). The two selected species were (1) timothy, which is the most important grass species of
the region due to its ability to survive winter and (2) perennial ryegrass, which is better fitted for
warmer conditions that are expected due to climate change. Results showed an increase in yield
for both grass species under future climate scenarios. However, the risk for frost damage is still
in place, which makes it difficult to justify using perennial ryegrass as an alternative to timothy
for this region. Shrestha et al. (2015) used grass and economic models to determine the regional
impacts of climate change on the yield and family farm income in Ireland. Concerning yield,
results showed 56% increase for perennial ryegrass under future climate scenarios (2061-2090)
due to increased rainfall, temperature and CO2 concentration. Meanwhile, the overall income
will decrease due to an increase in production costs. However, for farms using concentrate feed
for livestock, income can be improved by substituting the concentrate feed with grass feed.
Moore and Ghahramani (2013) used the GRAZPLAN simulation model to evaluate the impact of
climate change on the net primary productivity (ANPP) of five livestock enterprises in 25
locations in southern Australia. Results indicated that ANPP will decrease in future scenarios
(2030, 2050, and 2070) with larger reductions occurring in low-rainfall locations. The average
reductions of ANPP were 9% in 2030, 7% in 2050 and 14% in 2070. Kalaugher et al. (2013)
used a theoretical integrated research methodology to evaluate common adaptation strategies for
the dairy sector in New Zealand. Results showed that the integrated bottom-up social research
and top-down biophysical modeling have the potential of providing a more complete adaptation
strategy for the farming system.

47

In this study, we are mainly focused on the impacts of climate change on grazing
livestock systems. Most of the grazing livestock systems are for beef and dairy production. The
United States is the second largest milk producer of the world and 41% of its production is
concentrated in the Great Lakes region (USDA-NASS, 2014c; FAO, 2016a). Milk production,
under a grazing system, is dependent on pasture intake (Muller, 2004). Therefore, a reduction of
pasture yield, which consequently reduces pasture intake, can have a significant impact on milk
production. Meanwhile, in the past few decades more farmers are transitioning from dairy
confinement systems to grazing systems, especially in the Great Lakes Region, because of the
increase in profitability compared with confinement operations (Nott, 2003; Kriegl and McNair,
2005). A study by Kriegl and McNair (2005) showed that grazing systems produced $887 net
farm income from operations (NFIFO) per cow, compared to confined systems which produced
$640 NFIFO per cow (Kriegl and McNair, 2005). The dairy grazing systems requires less feed,
startup cost, labor, equipment, and fuel costs compare to confinement systems and; therefore, are
more appealing to smallholder farmers (Aschmann and Cropper, 2007). Meanwhile, it is
important to point out that there is no general agreement on the definition of smallholder dairy
farmers in the US; however, in this study, a farm with less than a 100-lactating-cow was
considered a small farm.
In order to reduce the impact of climate change on livestock production, adaptation
strategies should be implemented. Adaptation strategies can increase the resilience of agricultural
production to climate change (Walthall et al., 2012). Adaptation strategies for livestock
production involve: 1) diversification of plant and animal species (Kurukulasuriya and
Rosenthal, 2003; Batima et al., 2005; Calvosa et al., 2009), 2) incorporation of forestry and crop
productions to the livestock systems (Steinfeld et al., 2006; Wani et al., 2009; Jose, 2009;

48

Herrero et al., 2010; Herrero et al., 2012; Smith et al., 2012), 3) adjustment of the timing for
farm operations (Kurukulasuriya and Rosenthal, 2003; Batima et al., 2005; Calvosa et al., 2009;),
4) shifting locations of livestock and crop production (Kurukulasuriya and Rosenthal, 2003) and,
5) improving breeding strategies (Rowlinson, 2008; Henry et al., 2012).
In this study, diversification of plant species will be the main adaptation strategy
explored. This strategy was selected due to its high probability of selection by producers since it
only requires changing the seeding practices. In this study, we will identify the most resilient
composition for a representative grazing dairy farm in Michigan in order to minimize the
negative impacts of climate change. The study assesses the effectiveness of pasture compositions
by evaluating: a) economic and b) resource use of the grazing dairy farms. In the economic
category, we evaluated the net return per cow over milk production and the whole farm profit. In
the resource use category, we evaluated grazed forage consumption and the harvested forage
production. In order to achieve the goal of this study three objectives were sought: 1) identify the
best pasture composition under current and future climate scenarios concerning farm economics
and resource use, 2) assess economic and resources use impacts of pasture compositions under
future climate scenarios, and 3) evaluate the resiliency of pasture compositions to climate
change. Considering the number of combination of pasture species, climate scenarios, and the
extend of analysis from economic and resource use standpoints, this study is the most
comprehensive study that has ever been performed on grazing dairy farms in the Great Lakes
region. Studies such as this are important to guide policy makers and stakeholders to address the
negative impacts of climate change in the largest milk-producing region of the United States.

49

4.2

Materials and Methods
The goal of this study is to evaluate and propose an adaptation strategy based on pasture

composition for common grazing dairy farms in Michigan under future climate change scenarios.
A representative farm was developed based on a livestock practices survey. The farms where
selected based on Michigan pastureland distribution and United States Department of
Agriculture (USDA) - National Agricultural Statistics Service (NASS).
In order to evaluate the impact of climate change, the representative farm was tested
using both current and future climate scenarios. The livestock model, IFSM, was used to
simulate the representative grazing dairy farm performance. Two climate conditions were
evaluated: a) Current: 21 climate models for the period of 1985 to 2005 and b) Future: 42
climate models for the period of 2040 to 2060. The climate conditions were obtained from the
National Aeronautics and Space Administration (NASA) Earth Exchange Global Daily
Downscaled Projections (NEX-GDDP).
For each climate model scenario, the adaptation strategy was tested for 48 different
pasture compositions for a total of 2064 runs. The 48 pasture compositions were based on the
four most common grass species and the two most common legume species.
4.2.1 Study Area
Five of the top 10 milk-producing states are in the Great Lakes region, including the state
of Michigan (USDA-NASS, 2014c). In recent years, there has been an increase in the transition
from confinement dairy systems to grazing dairy systems (Nott, 2003) within this region. This
study will focus on grazing dairy farms in the state of Michigan.
Two criteria were defined in selecting the study area for the representative grazing dairy
farm: a) pastureland distribution and b) number of dairy farms at a county level. The pastureland

50

distribution was obtained from the USDA Crop Data Layer (USDA-NASS, 2014b). In general,
pasturelands are mainly concentrated in the Southern portion of Lower Peninsula of Michigan.
The number of dairy farms at the county level was obtained from USDA-NASS (2014a) and the
majority was found in the Southern portion of Lower Peninsula of Michigan (Figure 6). The area
of concentrated pasturelands and dairy farms is aligned with the agricultural region of Michigan
(Nejadhashemi et al., 2012). Therefore, this region was selected as the study area.

Figure 6. Study area and the location surveyed dairy farms
51

4.2.2 Data Collection
4.2.2.1 Grazing Farm Survey
In order to capture different grazing farm practices, ground surveys were performed in
the spring and summer of 2013 throughout the study area (Rojas-Downing et al., 2017a). The
surveyed farms were selected considering regional distribution, size, and management style.
MSU extension advice was also sought for identifying the representative grazing dairy farms in
the region. Ultimately, five different farms were surveyed in the western, eastern, and central
part of the Michigan agricultural region (Figure 6). Direct field observations and structured
questionnaires based on IFSM input data requirements where used to obtain the information
during a single-visit. To maximize survey compliance, farmers were reminded to return answers
to the questions that they were not able to respond to during the in-person interview. The
information collected on the surveys by Rojas-Downing et al., (2017a) was used to develop a
representative grazing dairy farm in Michigan. The selected data for the representative farm was
reviewed by MSU extension and specialists in order to confirm the representativeness of the
farm and validating the input data that will later use in modeling exercise. This information
included farm management, pasture area, machinery operations, grazing strategy, equipment,
structures, number and type of animals, and associated prices for several farm inputs and outputs.
A summary of survey information is provided in Table S1 and S2 in the Supplementary
Materials.
The five surveyed grazing dairy farms represent different farm size and
management operations that are common in the study area. The farm sizes range between 83 ha
to 146 ha. Most of the farms have alfalfa areas for forage production and a grazing area with a

52

variety of pasture compositions. Based on this information and inputs from extension specialists,
the typical grass and legume species and compositions were identified.
The surveyed farms had between 90 and 750 lactating cows. The cattle frame for four of
the farms were small Holsteins and one farm had New Zealand Friesian. The target milk
production ranges from 6,804 to 10,866 kg/cow/year, while the average milk production in
Michigan is 9,500 kg/cow/year. Most of the cattle feed of the surveyed farms had high relative
forage to grain ratios, with pasture as the main feed and grain as an energy supplement.
All of the surveyed farms have a rotational grazing system with a stocking rate that
ranges between 0.4 to 2.1 animals per acre. The grazing strategy is similar for the five surveyed
farms with a labor for grazing management of 10 hours per week and all animals are grazed 22
hours per day. Regarding animal facilities, all farms had a milking center, cow housing, heifer
housing, feed facility, short-term manure storage, and forage storage.
4.2.2.2 Representative Grazing Dairy Farm Description
Based on the data obtained from the statewide survey, the representative grazing farm
was established in IFSM. The simulated farm has small frame Holsteins (454 kg), where 100
were lactating cows and 80 were replacement stock. The grazing area was 54 ha with a stand life
of 10 years and additional 27 ha forage production area that includes alfalfa, with a stand life of
5 years. The tillage and planting operations were performed between mid-October and midApril, using a chisel plow and a seedbed conditioner. Eighty percent of the manure collected in
the milking facilities was applied to the alfalfa area and the remaining twenty percent was
applied to the grazing area. Beside the collected manure, no additional fertilizers were applied on
the grass and alfalfa areas.

53

As part of the grazing strategy, all cows where grazed for the entire day during the
growing season (April to October), with a labor for grazing management of 20 hours per week.
The grazing area varied during spring (27 ha), summer (49 ha), and fall (49 ha) to account for
area harvested for hay or silage. Alfalfa was harvested as silage and hay four times a year
beginning around the end of the May while grass was harvested once a year as silage around the
same period.
The annual target milk production was 7,252 kg per cow corrected to 4.2% fat content.
All animals were fed with a rotation that provided a relatively high forage-to-grain ratio.
Additional supplements were given to animals besides the feed produced on the farm. Soybean
meal 48% was given as crude protein supplement, distiller grain was given as undegradable
protein and grain was given as the energy supplement.
The animal facilities included a bedded pack barn for cow housing, calf hutches and dry
lot for heifer housing, a short-term storage premix as a feed facility, and a double four parlor as a
milking center. The manure was collected in the facilities throughout the year using scrapers
with slurry pumps and stored in a top-loaded concrete tank for a four-month period.
The economic analysis included all major operations based on 2013 representative prices.
Any resources grown on or brought onto the farm were considered as costs, while any sale of
milk, forage, and animals were considered as income. All income and costs for producing milk,
feed and animals occurred in the same year. Therefore, the annual accounting considers all
resources used during a one-year production (Rotz et al., 2014). Important prices include diesel
fuel at $0.98/L, electricity at $0.13/KWh, labor wage at $16/h, new forage stand at $494/ha, high
quality hay at $265/tonne DM, hay at $198/tonne DM, and milk at $0.36/kg.

54

4.2.2.3 Climate Data
Three sets of climatological data were used to evaluate the impact of climate change on
the representative grazing dairy farm. These included: a) one historical data set based on
observed daily records obtained from a weather station and b) two future climate scenarios based
two representative concentration pathways obtained from NASA NEX-GDDP. The three sets of
climatological data include daily precipitation and minimum and maximum temperature.
The historical data based on observed daily records (1985 to 2005) were obtained from
the National Climatic Data Center (NCDC) for station number 725390 located at the Lansing
Capital City Airport. The station was selected since it represents the annual average precipitation
(720 mm) and temperature (8.6Â°C) of the study area.
The future climate conditions from NASA NEX-GDDP are global downscaled climate
scenarios that were obtained from the General Circulation Model (GCM) simulations under the
Coupled Model Intercomparison Project Phase 5 (CMIP5). These climate scenarios provide
worldwide projected changes on a regional level in response to two different GHG emissions
simulated by 21 models (Thrasher et al., 2012). The two different GHG emissions scenarios, also
known as Representative Concentration Pathways (RCP) include current and future climate
conditions. The current climate conditions are simulated by 21 climate models for the period
from 1985 to 2005. The future climate conditions are simulated by 42 climate projections
derived from 21 climate models and two RCP scenarios (RCP 4.5 and 8.5) for the period of 2040
to 2060 (Table 1). The annual average precipitation and temperature were increase by 63.8 mm
and 2.2 Â°C for RCP 4.5 and 74.1 mm and 2.8 Â°C for RCP 8.5, respectively.
The NASA NEX-GDDP dataset provides worldwide high resolution (0.25 degrees x 0.25
degrees) and bias-corrected climate scenarios. The Bias-Correction Spatial Disaggregation

55

(BCSD) method was used for generating the NASA NEX-GDDP dataset (Thrasher et al. 2012).
BCSD is a statistical downscaling logarithm that addresses the GCM limitations, such as coarse
resolution grids and local bias projections (Wood et al., 2002; Wood et al., 2004; Maurer et al.,
2008; Thrasher et al., 2012). The BSCD algorithm adjusts the offset of climate trends by
applying a linear shift based on the observational climate data from weather stations of the
Global Meteorological Forcing Dataset (GMFD) (Thrasher et al., 2012). However, since the
location of the study area does not overlap with one of the GMFD stations, additional
downscaling using the Delta method (Woznicki and Nejadhashemi, 2014) was performed
between the 21 current climate models and the historical data based on observed daily records.
In addition to daily temperature and precipitation, solar radiation data are also a required
parameter for IFSM but not available through NASA NEX-GDDP data set. Therefore, solar
radiation was estimated for the climate scenarios using a weather generator. The weather
generator used in this study is called WXGEN (Sharpley and Williams, 1990; Wallis and
Griffiths, 1995). WXGEN is a stochastic weather generator model embedded in the Soil and
Water Assessment Tool (SWAT) to fill-in gaps of weather data (Jeong et al., 2015). WXGEN is
based on monthly statistics estimated from daily weather data (Jeong et al., 2015). Therefore,
daily temperature and precipitation data were used as inputs in WXGEN to estimate daily solar
radiation data for the location of study. WXGEN has been extensively used in other studies in
different parts of the world and it has proven to be a powerful tool (Wallis and Griffiths, 1995;
Harmel et al., 2000; Sourkova and Pona, 2002; Birhanu et al., 2007; Lee and Kim, 2013; Baker
and Miller, 2013).

56

Table 1. Summary of climate models used in NEX-GDDP and their average temperature and precipitation projections for the study
area for the period of 2040 to 2060 (IPCC, 2013; PCMDI, 2015)
RCP 4.5
RCP 8.5
Model
Average
Average
Average
Average
Modeling Center/ID
Model Name
number
Temperature Precipitation Temperature Precipitation
(Â°C)
(mm)
(Â°C)
(mm)
1
Institute for Numerical
INM-CM4
9.7
793.2
10.1
783.4
Mathematics/ INM
2
Beijing Climate Center, China
BCC-CSM1.1
11.0
734.4
11.4
797.9
Meteorological Administration/
BCC
3
NorESM1-M
11.0
720.1
11.6
781.2
Norwegian Climate Centre/ NCC
4

Meteorological Research Institute
/MRI

MRI-CGCM3

10.1

808.5

10.5

863.4

5

Max Planck Institute for
Meteorology/ MPI-M

MPI-ESM-MR

10.1

866.2

11.0

801.5

MPI-ESM-LR
MIROC5

10.2
11.3

737.0
771.7

11.3
12.0

751.5
839.1

MIROC-ESM

11.6

800.1

12.9

817.6

MIROC-ESMCHEM
IPSL-CM5A-MR

12.1

811.0

13.0

785.9

10.7

714.0

11.5

726.1

IPSL-CM5A-LR

10.7

770.6

11.4

759.4

GFDL-ESM2M

9.7

774.8

10.1

759.2

GFDL-ESM2G

10.1

823.5

10.6

885.4

GFDL-CM3

11.5

914.6

12.1

912.2

CanESM2

11.2

795.0

11.9

767.2

6
7
8
9
10
11

Atmosphere and Ocean Research
Institute (The University of
Tokyo), National Institute for
Environmental Studies, and Japan
Agency for Marine-Earth Science
and Technology /MIROC
Institute Pierre-Simon
Laplace/IPSL

12
13

Geophysical Fluid Dynamics
Laboratory/NOAA GFDL

14
15

Canadian Centre for Climate
Modelling and Analysis/ CCCma

57

Table 1 (contâd)
16
Commonwealth Scientific and
Industrial Research Organization
in collaboration with the
Queensland Climate Change
Centre of Excellence/ CSIROQCCCE
17
Centre National de Recherches
Meteorologiques / Centre
Europeen de Recherche et
Formation Avancees en Calcul
Scientifique/CNRM-CERFACS
18
National Science Foundation,
Department of Energy, National
Center for Atmospheric Research/
NSF-DOE-NCAR
19
National Center for Atmospheric
Research/NCAR
20
College of Global Change and
Earth System Science, Beijing
Normal University/ GCESS
21
CSIRO (Commonwealth Scientific
and Industrial Research
Organization, Australia), and
BOM (Bureau of Meteorology,
Australia)/ CSIRO-BOM

CSIRO-Mk3.6.0

10.8

851.5

11.1

858.2

CNRM-CM5

10.6

816.2

11.3

772.4

CESM1-BGC

10.7

772.6

11.5

818.5

CCSM4

11.3

669.2

11.4

742.6

BNU-ESM

11.3

808.5

11.7

815.1

ACCESS1.0

11.2

706.3

12.2

638.0

58

4.2.3 Modelling Process
4.2.3.1 Integrated Farm System Model (IFSM)
The IFSM is a whole-farm simulation model developed by the USDA-Agricultural
Research Service (ARS) for over 35 years. IFSM can evaluate the resource use and the
environmental and economic sustainability of livestock and crop production systems (USDAARS, 2014). The model was developed and validated for temperate regions of the northern
United States (such as Michigan) and Southern Canada (USDA-ARS, 2014). However, the
model has been used in many regions where long-term weather data is available (Wachendorf
and Golinski, 2006; Bryant and Snow, 2010). IFSM model performance has been extensively
validated with real farming practices (Rotz, 2004; Rojas-Downing et al., 2017a, b). In addition,
the model accuracy has been evaluated under both dry and wet years with different pasture
mixtures (Corson et al., 2007).
IFSM is distinguished from other similar models because it is a process-based model
(USDA-ARS, 2014). Therefore, it is able to represent the interactions between farm components
and major biological and physical process throughout the simulation period (USDA-ARS, 2014).
For a livestock system, the major components simulated in IFSM include feed production and
disappearance, grazing management, animal performance, feed storage, manure handling, tillage,
planting and harvest operations, and costs and returns. IFSM comprises nine major submodels
that are: crop and soil, grazing, machinery, tillage and planting, crop harvest, crop storage, herd
and feeding, manure handling, and economic analysis.
Several assumptions were considered during the IFSM model development, such as
disregarding market considerations (e.g. the amount of livestock production does not affect
commodity prices) and that the production and accounting period is one year (e.g. at the end of

59

the year, crop inventories are sold and feed shortages are purchased). In addition, the impact of
climate variabilities on farm performance can be evaluated if all parameters remain constant.
This assumption allows evaluation of climate change impacts on farm performances.
The required input parameters for the IFSM model include: soil type, crop areas and
characteristics, grazing management, equipment and structures used, harvest, tillage, numbers of
animals at various ages, herd feeding, manure handling strategies, and prices for various farm
inputs and outputs (Table S1). IFSM simulation is daily, but the outputs can be presented as
long-term annual averages and standard deviations. Meanwhile, some outputs can be presented
in higher temporal resolution including annual (e.g. production costs), monthly (e.g. grazed
forage yield), weekly (e.g. labor information), and daily basis (e.g. forage crop yield and quality)
(USDA-ARS, 2014).
4.2.3.2 Selection of pasture compositions
The representative grazing dairy farm was developed based on the survey information
collected from the five farms, a literature review, and expert opinion. The representative farm
simulated in IFSM had two variables: a) pasture composition and b) climate condition.
The pasture composition varies depending on the grazing management goal and the
location. Producers select the species of the pasture composition based on: biomass production,
seasonal biomass distribution, animal palatability and nutrition, and persistency over time
(Sullivan et al., 2000). In Michigan, the most common functional groups are cool-season grasses
and legumes. Based on the information collected from the survey, the most common cool-season
grasses are orchardgrass (Dactylis glomerata) and perennial ryegrass (Lolium multiflorum). The
most common legumes are white clover (Trifolium repens) and red clover (Trifolium pratense)
and the percentage in the pasture composition varies between 0% and 50%. However, the most

60

common pasture composition in Michigan is 60% orchardgrass and 40% white clover. Based on
the aforementioned information, these four pasture species were selected.
The two typical grass species used in Michigan have different characteristics that made
them favorable (Table 2). For the orchardgrass, the overall yield is high, and it is resistant to the
cold climate of Michigan (winter hardiness). In addition, orchardgrass is easy to establish due to
fast germination (seedling vigor), while it makes âexcellentâ hay production. Meanwhile,
perennial ryegrass is an âexcellentâ pasture crop and is easy to manage. In addition, it can be
quickly established (excellent seedling vigor) but cannot survive a cold weather; and therefore, is
mostly planted in the southern part of the state, which is, in general, warmer than the rest of the
state. In regard to adaptability of the aforementioned grass species to climate change,
orchardgrass has âfairâ resistance to flooding but a âgoodâ resistance to droughts while perennial
ryegrass has âpoorâ drought resistance but a âgoodâ tolerance to flooding.
In addition to the above species, alternative varieties were evaluated since the typical
grass species in Michigan have âgoodâ or âpoorâ adaptability to extreme events (Table 2). The
two other alternative cool-season grasses in Michigan are Kentucky bluegrass (Poa pratensis)
and Tall Fescue (Lolium arundinaceum). Both species are âexcellentâ for pasture and have
resistance to the cold Michigan environment (winter hardiness). In addition, tall fescue can be
characterized as a âvery goodâ drought and flood resistant species, while Kentucky bluegrass
cannot tolerate drought (poor) but is considered a âgoodâ flood resistant grass.

61

Table 2. Cool-season grasses characteristics (Barnes et al., 2007; Cassida et al., 2014)
Grass
Forage yield Seedling Pasture Drought
Flood
Winter
for Michigan
vigor
resistant resistant hardiness
(Kg/ha)
Orchardgrass 7,173 -10,088
G
F
G
F
E
Perennial
4,035 -6,725
E
E
P
G
F
Ryegrass
Kentucky
P
E
P
G
E
Bluegrass
Tall Fescue
5,380-10,760
F
E
VG
VG
E
E = Excellent, VG = Very Good, G = Good, F = Fair, P = Poor

Hay

E
P
P
E

4.2.3.3 Simulation of pasture compositions and climate scenarios in IFSM
Figure 7 illustrates the overall modeling process of pasture compositions and climate
scenarios for the representative grazing dairy farm. All possible combinations of the four coolseason grass species (Orchardgrass, Perennial ryegrass, Kentucky bluegrass and, Tall Fescue)
with the two legumes (white clover and red clover) where simulated (Figure 7). The percentage
of legume in the pasture composition varies between 0% and 50%, in 10% increments. Overall,
48 different pasture compositions where simulated. Each of the 48 pasture combinations were
simulated under the current condition and under the 42 climate models of the future condition of
NASA NEX-GDDP (21 climate models of RCP 4.5 and 21 climate models of RCP 8.5).
Consequently, 2064 scenarios where simulated in IFSM (Figure 7). The representative grazing
dairy farm information from Table S1 was used as the main input to IFSM with the variables of
pasture composition and climate conditions (Figure 7).

62

Figure 7. Modeling process of the representative grazing dairy farm

4.2.3.4 Identifying the impact of pasture compositions and climate change on economic and
resource use of a livestock grazing system
Statistical analysis was performed to understand the significance of grass and legume
species compositions concerning the economic and resource use criteria and future climate
scenarios (RCP 4.5 and RCP 8.5). The Fisherâs Least Significant Difference test (Fisher, 1935)
was used to perform a multiple pairwise comparison among grass and legume species at a 5%
63

significance level (Îą=0.05) using MATLAB version 8.2.0.701 (R2013b). The pairwise
comparison analysis was performed to identify the impact of: (1) legumes percentages with four
different grass species in the pasture composition, (2) grass species at different percentages of
legumes in the pasture composition, (3) two legumes for each grass species, (4) four grass
species for each legume species, and (5) climate scenarios (RCP 4.5 and RCP 8.5).
4.2.3.5 Sensitivity analysis to identify the most resilient pasture composition
In order to understand the resiliency of each pasture composition under the current and
future climate scenarios (RCP 4.5 and RCP 8.5), a sensitivity analysis was performed. Equation
(1) was used to calculate the relative sensitivity index (Luo and Zhang, 2009; Woznicki and
Nejadhashemi, 2012):

đ(đ,đ) =

(đŚđ âđŚ0 )đĽ0
(đĽđ âđĽ0 )đŚ0

(1)

where, S(T,P) is the relative sensitivity index based on average temperature (T) or precipitation
(P), xo is the 20-year average temperature or precipitation based on current climate scenario, xi is
the 20-year average temperature or precipitation for the ith future climate model/scenario, yo is
the criteria value (the economic and resource use) under the current climate scenario, and yi is the
criteria value under the ith future climate model/scenario. Greater magnitude of the relative
sensitivity index indicates higher sensitivity of the pasture composition to the temperature or
precipitation variations under future climate scenarios.
A minimum of 50 model simulations are required for each pasture composition (Luo and
Zhang, 2009). Therefore, for each pasture composition, the percentage of legume was randomly
altered within the selected legume range. Since the legume percentage ranges between 0% and
50% and IFSM only accepts integer values, all legume percentages were simulated within the
range. Therefore, the two legumes (Red Clover and White Clover) were varied 51 times (1%
64

intervals) for each grass species (Perennial ryegrass, Tall Fescue, Kentucky bluegrass, and
Orchardgrass) and one current and 42 future climate model/scenario combinations, resulting in
17,544 simulations.
After calculating the relative sensitivity index for each criterion under future climate
scenarios, the Euclidean norm was calculated using Equation (2):
đ = âđđ2 + đđ2

(2)

where, S is the relative sensitivity index. The Euclidean norm provides the sensitivity of the
pasture composition to both temperature and precipitation.
4.2.3.6 Economic and resource use sensitivity under future climate change scenarios
Second-order response surface models (Bradley, 2007) were fitted to different sensitivity
indices. The second-order response surface model analyzes the factor level combinations
(precipitation and temperature) that provide the maximum and minimum sensitivity index in
each criterion with different pasture compositions.
To present the fitted response surface model results, three dimensional response surface plots
and two dimensional contour plots were created (Supplementary material, Figures S1 to S28).
Regarding the response surface plots, the z-axis represents the sensitivity index (response
variable), the y-axis the temperature (explanatory variable), and the x-axis the precipitation
(explanatory variable). Contour lines were added at the base of the surface plots to better display
the structure of the response surface. Regarding the contour plots, x- and y-axes represented the
explanatory variables (temperature and precipitation, respectively) while the contour lines
represented the levels of the response variable (sensitivity index).
Two factors were evaluated in both the response surfaces and contour plots. First, the area
with the highest sensitivity index, and second, the fraction representing the change of

65

temperature in Celsius for every 100 mm of precipitation. The fraction represents the importance
of temperature to precipitation on the sensitivity indices between the different pasture
compositions.
The change of the sensitivity index was presented with a color gradient as the background
combined with the contour lines. Each color in Figure 8 represents the area of a certain range of
sensitivity index from white-pink (high sensitivity) to pink-yellow (medium sensitivity) to
yellow-green (low sensitivity). The contour plots provided a clear identification of the most
sensitive area based on temperature and precipitation for all the pasture compositions. The
fraction of the change of temperature in Celsius for every 100 mm of precipitation was obtained
by drawing two lines from the highest value on the graph. One line is drawn parallel to the x-axis
(A) and the second line connects the inflexion points of the concave contour lines (B) as shown
in Figure 9. The vertical difference between the lines at the far right of the graph, named âCâ,
represents the difference in temperature for the entire range of the precipitation. By using the rule
of three, where the nominator is temperature and the denominator is precipitation, the fraction of
the change of temperature in Celsius for every 100 mm of precipitation was calculated.

66

Figure 8. Sensitivity index area

Figure 9. Calculating the fraction of temperature to precipitation

67

4.3

Results and Discussion

4.3.1 Identification of the best pasture composition based on resource use and economics
4.3.1.1 Net return per cow over milk production
The net return per cow for milk production is determined by IFSM as the sum of all
revenues minus the sum of all costs related to milk production (Rotz et al., 2014). The revenues
include incomes from feed production and milk sales while the costs include purchased feed and
activities associated with crop production such as equipment, fuel, electricity, facility, labor,
seeds, fertilizers, and chemicals.
Among evaluated pasture composition scenarios, the one with 50% perennial ryegrass
and 50% red clover was identified with the highest net return per cow over milk production for
both the current and future climate scenarios. In fact, we saw improvement in net return under
future climate scenarios. For example, on average the highest net return per cow ($2,149/cow-yr)
was observed under the RCP 8.5 climate scenario, followed by the RCP 4.5 climate scenario
($2,123/cow-yr), and then the current scenario ($2,106/cow-yr), respectively. This is mainly due
to increases in the annual average temperature and precipitation in the region as the current
average temperature is expected to rise from 8.6ËC to 10.8ËC under RCP 4.5 and to 11.5ËC under
RCP 8.5 by the middle of this century. Similarly, the current annual average precipitation is
expected to increase from 720 mm to 784 mm and 794 mm for RCP 4.5 and 8.5 climate
scenarios, respectively. This resulted in increased performances in perennial ryegrass, as the
ryegrass does not generally perform well under low rainfall and high soil temperature conditions
(MacFarlane, 1990).
In general, the differences between the best-case scenario (50% perennial ryegrass and
50% red clover) and other top ranked pasture compositions (different grass species with 50 %

68

legume) were small and less than 10 dollars (Table 3). Therefore, it can be concluded that the
percentage of legume is the main factor, besides climate scenario, influencing the net return per
cow over milk production. Meanwhile, among the evaluated legume species and beside the
overall profit, red clover is a preferred choice since it has a better potential for hay production
(Lacefield, 2013).
Looking more in depth into the costs and incomes associated with milk production
reveled similar trends between current and future climate scenarios. In general, within the same
climate scenario, the higher ranked pasture composition scenarios had higher incomes and costs
and vice versa. Meanwhile, for all climate scenarios, a pasture composition with 50% red clover
combined with perennial ryegrass had the highest net return per cow over milk production,
followed by tall fescue, Kentucky bluegrass, and orchardgrass. For example under RCP 8.5
climate scenario, the highest costs and incomes were associated with the most profitable scenario
(50% perennial ryegrass and 50% red clover) in which the average annual costs for equipment,
fuel and electricity, feed and machinery facilities, and seed, fertilizer and chemical are $39,686,
$5,653, $11,294, and $11,326 respectively. However, the perennial ryegrass also presented the
lowest purchased feed cost ($66,336/yr) and the highest income from feed sales ($29,173/yr).
Labor is considered an income for the representative farm, because the majority of dairy farmers
in Michigan (97%) are family owned, and thus family members tend to perform most of the
labor on the farm (United Dairy Industry of Michigan, 2016). Therefore, by subtracting the costs
($134,295/yr) from the feed sales ($29,173/yr), labor incomes ($27,188/yr) and milk incomes
($292,850/yr), the perennial ryegrass had the highest net return per cow over milk production
($2,149/yr) compared to tall fescue ($2,143/yr), Kentucky bluegrass ($2,142/yr), and
orchardgrass ($2,139/yr). Comparing to the current scenario, the average net return per cow for

69

the perennial ryegrass, tall fescue, Kentucky bluegrass, and orchardgrass were improved by
2.04%, 1.85%, 1.90%, and 1.86% under RCP 8.5 scenarios, respectively.
Table 3. Summary of the top four pasture compositions with the highest net return per cow over
milk production
Pasture composition
Net return per cow ($/cow-year)
Current
Future Scenario
Grass
Legume
Scenario
RCP 4.5
RCP 8.5
Mean (SD*) Mean (SD*) Mean (SD*)
50% Perennial Ryegrass
50% Red Clover
2,106 (809) 2,123 (431) 2,149 (473)
50% Tall Fescue
50% Red Clover
2,104 (816) 2,118 (437) 2,143 (474)
50% Kentucky Bluegrass 50% Red Clover
2,102 (815) 2,117 (433) 2,142 (474)
50% Orchardgrass
50% Red Clover
2,100 (817) 2,114 (441) 2,139 (472)
*SD: Standard Deviation
4.3.1.2 Whole Farm profit
The whole farm profit is obtained from IFSM as a return to management and unpaid
factors. IFSM calculates this by subtracting costs for feed, manure handling, animal housing,
animal care, and milking from the incomes of milk, excess feed and animal sales (Rotz et al.
2014).
The pasture composition that had the highest whole farm profit was 50% perennial
ryegrass and 50% red clover for all climate scenarios. The climate scenario that had the highest
whole farm profit was RCP 8.5 ($174,672/yr), followed by RCP 4.5 ($172,059/yr) and then the
current scenario ($170,406/yr), respectively.
In all climate scenarios, the top four pasture compositions with the highest whole farm
profit are all the four grass species combined with 50 % legume (Table 4). These top four pasture
compositions have a similar whole farm profit, and major differences are dependent on the
percentage of legume. Therefore, similar to the net return per cow, the percentage of legume is
the main factor affecting farm profit.

70

Overall, the higher ranked pasture composition scenarios had the highest incomes and
costs, and vice versa in all climate scenarios. This trend is the same as for the criterion of net
return per cow, where a pasture composition with 50% red clover combined with perennial
ryegrass has the highest whole farm profit and orchardgrass has the lowest. For example, under
RCP 8.5 the average annual costs associated with this scenario are equipment ($47,121/yr),
facilities ($47,228/yr), energy ($13,091/yr), seed, fertilizer and chemical ($11,326/yr), animal
purchase and livestock expense ($48,200/yr), milk hauling and marketing fees ($12,801/yr) and
property tax ($4,435/yr). However, the perennial ryegrass presented the lowest net purchased
feed and bedding cost ($38,018/yr). Therefore, by subtracting the costs ($222,220/yr) from the
labor ($59,601/yr), animal sales ($44,442/yr) and milk incomes ($292,850/yr), the perennial
ryegrass had the highest whole farm profit of $174,672/yr compared to tall fescue ($174,087/yr),
Kentucky bluegrass ($173,978/yr), and orchardgrass ($173,678/yr).
Table 4. Top four pasture compositions with the highest whole farm profit
Pasture composition
Whole farm profit ($/year)
Future Scenario
Current
RCP 4.5
RCP 8.5
Grass
Legume
Scenario
Mean
Mean
Mean (SD*)
(SD*)
(SD*)
50% Perennial Ryegrass
50% Red Clover
170,406
172,059
174,672
(8223)
(5375)
(5577)
50% Tall Fescue
50% Red Clover
170,173
171,540
174,087
(8432)
(5402)
(5573)
50% Kentucky Bluegrass 50% Red Clover
170,006
171,457
173,978
(8368)
(5401)
(5573)
50% Orchardgrass
50% Red Clover
169,735
171,157
173,678
(8420)
(5348)
(5607)
*SD: Standard Deviation
4.3.1.3 Grazed forage consumed
The grazed forage consumed is the amount of pasture consumed by herds. The
availability of pasture is predicted by IFSM and is based on the season, pasture species

71

characteristics, soil type, and weather conditions. In general, 60% of the pasture production is
kept for herd consumption and the remaining 40% is used for pasture regrowth (Rotz et al.,
2014).
For the grazed forage consumed criterion, the top four pasture compositions had the same
value for each climate scenario (current = 186 tonne DM/yr; RCP 4.5 = 185 tonne DM/yr;
RCP8.5 = 184 tonne DM/yr). In general and regardless of grass species, a combination of 30%,
40% and 50% of red clover produced between 184 (RCP 8.5) to 186 (current) tonne of dry
matter (DM) per year in all climate scenarios. Therefore, to maintain a high grazed forage
production for future scenarios, it is important to maintain at least 30% red clover in the pasture
composition. This finding is also supported by previous studies that demonstrated that at least
30% legume is required to improve pasture productivity (Rayburn, 2007). This can be explained
due to the fact that the red clover on average is able to produce two-thirds more forage than
white clover and tolerates wetter soils (Griffin, 2004), which is the case in the study area.
In general, the grazed forage consumption decreases with the future scenarios because of
an increase in average rainfall and temperature. Cool-season grasses tolerate a cooler and wetter
environment. The optimum growth of cool-season grasses is between 10 ËC and 21 ËC and the
growth decreases to zero as the average temperature increases to 32 ËC (Rayburn, 2007).
Therefore, an increase in temperature in the future scenarios reduces the amount of grass forage
produced.
4.3.1.4 Harvested forage production
To meet animal nutrients needs and maintain the desired production level, harvested
forage is provided to livestock in addition to pasture and supplemental feeds. Therefore, the
source of the harvested forage production was a combination of harvested pasture composition

72

and alfalfa to produce enough hay and silage to meet the animals feeding requirements. The
pasture composition that had the highest annual harvested forage production was 50% perennial
ryegrass and 50% red clover for the current and future climate scenarios. However, the RCP 8.5
climate scenario presented the highest harvested forage production (337 tonne DM), followed by
the current scenario (330 tonne DM) and RCP 4.5 (326 tonne DM). This trend is similar to the
forage yield. Alfalfa yield in the current scenario, RCP 4.5 and RCP 8.5 is 11.2 tonne/ha, 11.2
tonne/ha and 11.5 tonne/ha, respectively. Grass yield in the current scenario, RCP 4.5 and RCP
8.5 is 5.5 tonne/ha, 5.0 tonne/ha and 5.3 tonne/ha, respectively. The reduction of harvested
forage production of grass in the future scenarios is related to the increase in temperature, which
lowers the cool-season grass performance (MacFarlane, 1990; Rayburn, 2007). Despite the
decrease of grass yield in future scenarios, the highest harvested forage production was observed
for RCP 8.5 (337 tonne DM), mostly due to an increase in the alfalfa yield.
The top four pasture compositions with the highest harvested forage production have
50% red clover for all climate scenarios. This is related to the higher potential of red clover for
hay production compared to white clover (Lacefield, 2013). The forage production ranges from
327 to 330 tonne DM under current climate scenario. However, it is expected to decrease to 321
to 326 tonne DM under RCP 4.5 but increase to 332 to 337 tonne DM under RCP 8.5 scenario.
The harvested forage is classified into high and low quality based on nutrient availability
and concentration. The forage quality is influenced by several factors such as maturity of the
plant at the time of harvesting, plant species, climate, soil fertility, disease and insect pests, and
harvesting and storage losses (Rayburn, 2007). The high quality forage production follows the
same trend as the harvested forage production. The pasture composition that had the highest
annual high quality forage production contains 50% perennial ryegrass and 50% red clover for

73

all climate scenarios. The current scenario presented the highest high quality forage production
(233 tonne DM), followed by RCP 4.5 (136 tonne DM), and RCP 8.5 (128 tonne DM).
Neutral Detergent Fiber (NDF) and Crude Protein (CP) are widely used to determine
forage quality. The NDF is the portion of the forage that is insoluble in neutral detergent, which
contains the primary components of the plant cell wall. An increase in NDF decreases dry matter
intake. Legumes with an NDF lower than 40% and grasses with less than 50% NDF will be
considered high quality forage (Rayburn, 2007; Rayburn, 2008). The CP is the amount of
nitrogen and amino acids in the forage. The amino acids are used by the animals for growth and
milk production. A forage that has higher CP will usually be considered high quality (Rayburn,
2007; Rayburn, 2008). Based on the aforementioned criteria, the forage quality of alfalfa
decreases under future scenarios. Alfalfa CP decreases from 20.3% in the current scenarios to
19.0% and 18.7% in RCP 4.5 and RCP 8.5, respectively, which reduces the forage quality (Table
5). Meanwhile, the NDF level increases from 41.9% (current) to 47.6% (RCP 8.5) under the
future scenario (Table 5), which reduces the forage quality. However, a different trend was
observed with forage quality in regard to pasture composition. The pasture composition that
presented the highest CP in all climate scenarios was 50% perennial ryegrass and 50% red
clover. Red clover is commonly added to the forage to increase the protein level (Griffin, 2004).
The CP for this pasture composition increases from 31.7% (current) to about 33% for the future
scenarios (Table 5). This represents an improvement in forage quality under future climate
scenarios. The pasture composition that presented the highest NDF in all climate scenarios was
50% Perennial ryegrass and 50% White Clover. The NDF increases from 32.6% in the current
scenario to about 35% for the future scenarios (Table 5). This represents a decline in forage
quality under future climate scenarios.

74

Table 5. CP and NDF for alfalfa and the pasture compositions with the highest quality
Forage
Forage
Future scenario
Current
quality
RCP
RCP
scenario
factors
4.5
8.5
Alfalfa
CP a (%)
20.3
19.0
18.7
NDF b (%)
41.9
47.0
47.6
50% Perennial Ryegrass and 50% Red Clover
CP (%)
31.7
32.9
33.0
50% Perennial Ryegrass and 50% White Clover NDF (%)
32.6
34.6
34.8
a
Crude Protein (CP): higher CP, higher forage quality
b
Neutral Detergent Fiber (NDF): Legumes with an NDF lower than 40% and grasses with less
than 50% NDF are considered high quality forage.
4.3.2 Impact of pasture compositions and climate change on economic and resource use of a
livestock grazing system
In this section, a pairwise comparison analysis was performed to identify the significant
difference between grass and legume species in the pasture composition under two climate
scenarios (RCP 4.5 and RCP 8.5).
Comparison of the four types of grass species, while the legume percentages were held
constant in the pasture composition (Table S3), did not present any significant difference (Îą=
0.05) in the three criteria of (1) net return over milk production, (2) whole farm profit, and (3)
harvested forage production. Therefore, the implementation of perennial ryegrass, tall fescue,
Kentucky bluegrass, or orchardgrass with either red clover or white clover did not have
significant impact on the profits or forage production of the farm. However, for the grazed forage
consumed criterion, perennial ryegrass had significantly lower grazed forage consumed than the
other three grass species when legumes were not considered. However, under RCP 4.5 and with
10% red clover added to the pasture composition, orchardgrass was found to have significantly
lower grazed forage consumed than ryegrass. Therefore, while perennial ryegrass had less grazed
forage consumed than orchardgrass with no legume, the situation is reverse when 10% of red
clover was incorporated into the perennial ryegrass. Overall, the grass species did not have

75

significantly difference in any of the four criteria evaluated under the two climate scenarios
(Table 6).
The comparison of the percentage of the two legume species is presented in Table S4.
Overall, as the percentage of legume increased in the pasture composition the value of all four
criteria increased. This result can be explained by the importance of legume benefits including: i)
being a source of nitrogen due to their ability of fixing nitrogen through Rhizobium bacteria
(Aschmann and Cropper, 2007); ii) improvement of forage quality in the pasture (Aschmann and
Cropper, 2007); and iii) reduction of the summer slump of grass-only pasture (Aschmann and
Cropper, 2007). The following sections discuss the significant level of legume percentages that
affect economic and resource use criteria (Table S4):
The economic criteria: Under RCP 4.5 and 8.5 climate scenarios, a 10% increases in legume
has significant impacts on economic criteria (net return per cow and the whole farm profit) up to
20% for the red clover and up to 30% for the white clover. Above these points, no significant
economic impacts were observed.
The resource use criteria: Regarding the grazed forage consumed and red clover, no
significant impact was observed for the legume increase above 30% regardless of grass species;
however, for white clover this number was increased to 40%. Regarding the harvested forage
production, a 10% increase in legume has significant impacts on this criterion up to 20% for the
red clover; however, this range is 20% to 30% for the white clover.
Overall, the presence of red clover in the pasture composition resulted in significantly higher
values for the four criteria compared to the white clover (Table 7). These results can be
explained by the higher biomass yield of red clover in comparison to white clover (Griffin,
2004), which consequently increases the forage production and thus the farm profits.

76

Finally, the impacts of the climate scenarios on economic and resource use of a dairy grazing
system are presented in Table 8. Overall, it can be concluded that increases in precipitation and
temperature under more intensive climate scenario (RCP 8.5) can significantly improve the net
return per cow and whole farm profit while having no significant impact on resource use criteria.

77

Table 6. Statistical comparison of net return, whole farm profit, grazed forage consumed, and harvested forage produced by different grass and
legume species (horizontal comparison)*
RCP 4.5
RCP 8.5
Criteria
Legume
Perennial
Tall
Kentucky
Orchardgrass
Perennial
Tall
Kentucky
Orchardgrass
Ryegrass
Fescue
Bluegrass
Ryegrass
Fescue
Bluegrass
Red
Net
Clover
1,983a
1,978a
1,976a
1,973a
2,009a
2,003a
2,002a
1,999a
Return
White
($/cow/yr)
Clover
1,843a
1,841a
1,838a
1,837a
1,874a
1,871a
1,869a
1,868a
Whole
Red
Farm
Clover
158,168a
157,661a
157,480a
157,225a
160,785a
160,205a
160,055a
159,790a
Profit
White
($/yr)
Clover
144,297a
144,118a
143,848a
143,726a
147,357a
147,133a
146,911a
146,781a
Grazed
Red
Forage
Clover
156a
155a
155a
155a
155a
155a
155a
155a
Consumed White
(tonne/yr) Clover
133a
132a
132a
131a
136a
136a
136a
135a
Harvested Red
Clover
298a
296a
295a
294a
308a
305a
305a
303a
Forage
Produced White
(tonne/yr) Clover
263a
263a
262a
262a
270a
270a
269a
269a
* Values followed by the same letters are not significantly different (Îą=0.05).

78

Table 7. Statistical comparison of net return, whole farm profit, grazed forage consumed, and harvested forage produced by different grass and
legume species (vertical comparison)*
RCP 4.5
RCP 8.5
Criteria
Legume
Perennial
Tall
Kentucky
Orchardgrass
Perennial
Tall
Kentucky
Orchardgrass
Ryegrass
Fescue
Bluegrass
Ryegrass
Fescue
Bluegrass
Red
Net
Clover
1,983a
1,978a
1,976a
1,973a
2,009a
2,003a
2,.002a
1,999a
Return
White
($/cow/yr)
Clover
1,843b
1,841b
1,838b
1,837b
1,874b
1,871b
1,869b
1,868b
Whole
Red
Farm
Clover
158,168a
157,661a
157,480a
157,225a
160,785a
160,205a
160,055a
159,790a
Profit
White
($/yr)
Clover
144,297b
144,118b
143,848b
143,726b
147,357b
147,133b
146,911b
146,781b
Grazed
Red
Forage
Clover
156a
155a
155a
155a
155a
155a
155a
155a
Consumed White
(tonne/yr) Clover
133b
132b
132b
131b
136b
136b
136b
135b
Harvested Red
Clover
298a
296a
295a
294a
308a
305a
305a
303a
Forage
Produced White
(tonne/yr) Clover
263b
263b
262b
262b
270b
270b
269b
269b
* Values followed by the same letters are not significantly different (Îą=0.05).

79

Table 8. Statistical comparison of net return, whole farm profit, grazed forage consumed and,
harvested forage produced by different climate scenarios (horizontal comparison)*
Criteria
RCP 4.5
RCP 8.5
Net Return ($/cow/yr)
1,909a
1,937b
Whole Farm Profit ($/yr)
150,815a
153,627b
a
Grazed Forage Consumed (tonne/yr)
143
145a
Harvested Forage Produced (tonne/yr) 279a
287a
* Values followed by the same letters are not significantly different (Îą=0.05).
4.3.3 Economic and resource use sensitivity under future climate change scenarios
4.3.3.1 Surface Analysis
A sensitivity analysis was performed to understand the resiliency of the different pasture
compositions under the future climate scenarios. In general, the greater the magnitude of the
sensitivity index the higher the sensitivity and the lower the resilience of the pasture composition
to the temperature and precipitation variations under future climate scenarios. From 42 future
climate scenarios (21 climate models for each RCP 4.5 and RCP 8.5), only one climate scenario
under RCP 4.5 (NorESM1-M) resulted in very high sensitivity indexes for all the pasture
combinations. This is due to the minimal increase (0.1 mm) in precipitation under this scenario
compared to current condition. This makes the ratio very large in Equation (1) since the
denominator is much smaller than the numerator. Therefore, the climate scenario of NorESM1M was considered as an anomaly since the results deviated significantly from the results of other
scenarios and was removed from further analysis.
In order to understand the sensitivity index pattern for each criterion, a second-order response
surface model was used. Three-dimensional response surface plots and two-dimensional contour
plots were created for each pasture composition and criterion, resulting in 32 surface plots and 32
contour plots (Figures 10 to 13 and S1 to S28). In these figures, the precipitation (x-axis) varies
between 631 mm and 922 mm and the temperature (y-axis) range varies between 9.6ËC and
13ËC. The two factors evaluated in these plots are: (1) the area with the highest sensitivity index
80

and (2) the importance of temperature to precipitation between two or more different pasture
compositions.
The following sections discuss the results of the sensitivity analysis in the context of
economic and resource use:
The economic criteria: Based on the results presented in Figures 10, 11 and S1 to S14, the
area with a sensitivity index greater than 0.8 was identified as area in which the combination of
temperature, precipitation, and pasture compositions resulted in high sensitivity or low
resiliency. The area with the highest sensitivity index was on average larger for the whole farm
profit than the net return per cow over milk production. The whole farm profit was more
sensitive to ranges of average temperature and precipitation between 9.6ËC and 12.1ËC and 631
mm and 785 mm, respectively (e.g. Figure 11). Meanwhile, the net return per cow over milk
production was more sensitive to ranges of average temperature and precipitation between 9.6ËC
and 11.7ËC and 631 mm and 741 mm, respectively (e.g. Figure 10). This can be explained due to
the fact that the net return per cow was estimated based only on the milk production, while the
whole farm profit includes additional factors such as manure handling, animal housing, and
animal care and sales, which makes the criteria more sensitive to climatological variabilities.
Regarding the importance of temperature to precipitation on sensitivity indices, both
economic criteria had similar results. The pasture composition that was more sensitive to
temperature was the perennial ryegrass with red clover with the fraction above 0.5ËC/100 mm.
Meanwhile, the pasture composition that was less sensitive to temperature was the orchardgrass
with red clover with the fraction below 0.3 ËC/100. However, for the grass species combinations
with different legumes, the results were mixed. For example, perennial ryegrass is more sensitive
to temperature changes when is mixed with red clover than white clover for the economic

81

criteria. However, in the case of the Kentucky bluegrass and orchardgrass mixtures with
legumes, more sensitivity to temperature was observed for white clover than red clover. In the
case of tall fescue, both red and white clover had the same sensitivity to temperature.
The resource use criteria: For both resource use criteria, the area with a sensitivity index
greater than 6.0 was identified as the area in which the combination of temperature, precipitation,
and pasture compositions resulted in high sensitivity or low resiliency (Figures 12, 13 and S15 to
S28). The high sensitive area is usually larger for the grazed forage than the harvested forage.
The area for the grazed forage consumption ranges between 9.6ËC and 11.6ËC and 631 mm and
738 mm for average temperature and precipitation, respectively (e.g. Figure 12). Meanwhile, the
area for the harvested forage production ranges between 9.6ËC and 11.3ËC and 631 mm and 686
mm for average temperature and precipitation, respectively (e.g. Figure 13). The grazed forage
only includes the consumption of the different pasture compositions, while the harvested forage
production involves mostly alfalfa harvest. Therefore, the decrease of the sensitivity index area
of the harvested forage production can be explained by the alfalfaâs excellent heat and drought
tolerance compared to other grass and legume species in this study, which range between poor
and good tolerance (Cassida et al., 2014).
Regarding the importance of temperature to precipitation on sensitivity indices, the results
are mixed for the resource use criteria. In the case of grazed forage consumption, the Kentucky
bluegrass or orchardgrass combined with white clover were more sensitive to temperature
(1.1ËC/100 mm) than perennial ryegrass, tall fescue, and orchardgrass mixed with red clover (0.9
ËC/100 mm). In general, the combinations of all grass species with red clover were less sensitive
to temperature than white clover. In the case of harvested forage production, the pasture
composition compromising perennial ryegrass with red clover was more sensitive to temperature

82

(1.1ËC/100 mm). However, in general, all combinations of grass species with red clover were
more sensitive to temperature and had higher production rates than the combinations with white
clover. Meanwhile, the pasture composition that was the least sensitive to temperature changes
(0.7ËC/100 mm) was the tall fescue with white clover.
Overall, the area identified with the highest sensitivity index for all criteria is located within
the temperature range of 9.6ËC and 12.1ËC, and the precipitation range of 631 mm and 785 mm.
Therefore, this area will be highly impacted by climate change regardless the pasture
compositions or the economics and resource use criteria. In general, the mixture of perennial
ryegrass and red clover had the highest sensitivity to temperature to precipitation ratio in the
majority of criteria (net return per cow over milk production, whole farm profit and harvested
forage production), while the mixture of orchardgrass with red clover had the least sensitivity to
temperature to precipitation ratio in three criteria, which are net return per cow over milk
production, whole farm profit, and grazed forage consumption.

Figure 10. Net return per cow over milk production sensitivity index for perennial ryegrass and
red clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index
83

Figure 11. Whole farm profit sensitivity index for perennial ryegrass and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure 12. Grazed forage consumed sensitivity index for perennial ryegrass and red clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

84

Figure 13. Harvested forage produced sensitivity index for perennial ryegrass and red clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

4.3.3.2 Overall Sensitivity Assessment
The overall sensitivity assessment was performed based on the results of 1,312 sensitivity
indices (Section 4.2.3.5). This allows obtaining the absolute highest and the lowest sensitivity
index for each criterion along with the most and the least resilient pasture composition under
future climate scenarios.
Overall, the resource use criteria had a greater sensitivity indexes range (0.0 to 18.2) than the
economic criteria (0.0 to 3.4). The criterion with highest sensitivity indexes was the grazed
forage consumed, which ranged between 0.5 and 18.2. The pasture composition that had the
maximum sensitivity index from the grazed forage consumed criterion was orchardgrass with
white clover. The criterion with lowest sensitivity indexes was the net return per cow over milk
production, which ranged between 0.0 and 2.7, where perennial ryegrass combined with red
clover was the pasture composition with the maximum sensitivity index. While the least resilient
pasture composition was orchardgrass mixed with white clover.
85

The overall most resilient pasture composition and most and least resilient grass and legume
species are listed in Table 9. In this table, each cell with a white background represents the
average sensitivity indexes of 42 future climate scenarios. In addition, the Euclidean operator
norm was used to estimate the overall sensitivity index for each criterion, grass species, and
legume species (Table 9, cells with gray background). The following sections discuss the results
of Table 9 in the context of economic and resource use for each criterion:
The economic criteria: The results of the sensitivity analysis from the average condition
(background with white cells) revealed that the trend in both economic criteria were the same,
with perennial ryegrass presenting the highest sensitivity index, followed by tall fescue,
Kentucky bluegrass, and orchardgrass. In addition, white clover presented higher sensitivity
indexes than red clover. Overall, the whole farm profit criterion presented higher sensitivity
indexes than the net return per cow over milk production, which supports previous findings from
the second-order response surface model analysis.
The resource use criteria: In contrast with the economic criteria, resource use criteria
followed the opposite trend, where Orchardgrass presented the highest sensitivity index,
followed by Kentucky bluegrass, tall fescue, and perennial ryegrass (Table 9). Regarding the
legume species, in the grazed forage consumed criterion, white clover had higher sensitivity
indexes than red clover. However, the opposite trend was observed by the harvested forage
produced criterion. Overall, the grazed forage consumed criterion presented higher sensitivity
indices than the harvested forage produced, which is aligned with the discussion in Section
4.3.3.1.
Overall, the grazed forage consumed criterion presents the overall highest sensitivity index of
all criteria and pasture compositions. Meanwhile, net return per cow over milk production

86

presents the overall lowest sensitivity index. Regarding grass species, orchardgrass had the
overall highest sensitivity index for both red clover (2.87) and white clover (3.54), while
perennial ryegrass had the overall lowest sensitivity index (red clover = 2.71 and white clover =
3.35). Concerning legumes species, white clover had an overall higher sensitivity index (6.91)
than red clover (5.60). Therefore, the most resilient pasture composition under future climate
scenarios was perennial ryegrass with red clover and the least resilient pasture composition was
orchardgrass with white clover.

87

Table 9. Average and overall sensitivity indices for all criteria, grass species, and legume species

Perennial
Ryegrass
(R)
Tall Fescue
(T)
Kentucky
Bluegrass
(K)
Orchardgrass
(O)

Net Return
per Cow
(NR)

Whole
Farm
Profit
(WP)

Red Clover
Grazed
Forage
Consumed
(GF)

Harvested
Forage
Produced
(HF)

0.47

0.58

1.96

0.43

0.54

0.42
0.40

White Clover
Grazed
Harvested
Forage
Forage
Consumed Produced
(GF)
(HF)

Overall**
(SG )

Net
Return
per Cow
(NR)

Whole
Farm
Profit
(WP)

1.72

2.71

0.53

0.67

2.79

1.65

3.35

2.00

1.84

2.80

0.52

0.67

2.90

1.65

3.44

0.53

2.00

1.88

2.82

0.52

0.67

2.92

1.70

3.48

0.50

2.02

1.94

2.87

0.52

0.67

2.99

1.69

3.54

1.05

1.34

5.80

3.34

6.91

*

Overall (SC)
0.86
1.08
3.98
3.69
5.60
2
2
2
2
*đđ = âđ + đ + đž + đ , where SC overall sensitivity index for a criterion
** đđş = âđđ 2 + đđ2 + đşđš 2 + đťđš 2 , where SG= overall sensitivity index for a grass

88

Overall**
(SG )

4.4

Conclusions
The overall goal was to identify the best pasture composition for a representative grazing

dairy farm in Lower Peninsula of Michigan to reduce the impacts of climate change. Three
objectives were pursued: 1) identify the best pasture composition concerning the economic and
resource use criteria, 2) assess economic and resources use impacts of pasture compositions
under future climate scenarios, and 3) evaluate the resiliency of pasture compositions under
current and future climate scenarios.
Regarding the selection of the best pasture composition, concerning the economic and
resource use criteria, 50% perennial ryegrass and 50% red clover was identified as the best
option under current and future climate scenarios. This is mainly due to increases in the future
annual average temperature and precipitation in the region under future climate scenarios, which
increases the performance of perennial ryegrass. In general, and under all climate scenarios, the
top pick is always a pasture composition with 50% red clover combined with perennial ryegrass,
followed by tall fescue, Kentucky bluegrass, and orchardgrass. However, the grazed forage
consumed criterion was an exception where the top four pasture compositions had the same
value for each climate scenario.
In order to assess economic and resources use impacts of pasture compositions under future
climate scenarios, a pairwise comparison analysis was performed. The comparison of the four
types of grass species, while legume percentage was held constant, showed no significant
differences. However, when both grass and legume species were considered, it was revealed that
the presence of red clover could significantly improve outcomes under all climate scenarios. In
general, increasing the percentage of the two legume species resulted in improvement of
indicators for all criteria. Finally, evaluation of climate scenarios showed that under intensive

89

climate scenario (RCP 8.5), the net return per cow and whole farm profit were significantly
improved in comparison with RCP 4.5, but no significant impacts were observed for the resource
use criteria.
In order to evaluate the resiliency (capacity to adapt to external stresses) of pasture
compositions under the future climate scenarios, a sensitivity analysis was performed. The
surface and contour plots were used to identify the most sensitive conditions for the simulated
livestock grazing system. This condition was identified between 9.6ËC and 12.1ËC of temperature
and 631 mm and 785 mm of precipitation. Regarding the sensitivity of pasture compositions to
the ratio of temperature to precipitation, the pasture composed of perennial ryegrass and red
clover had the highest sensitivity to temperature in three criteria: net return per cow over milk
production, whole farm profit, and harvested forage production. Furthermore, the pasture
composed of orchardgrass with red clover had the least sensitivity to temperature in three
criteria: net return per cow over milk production, whole farm profit, and grazed forage
consumption.
The results of the overall sensitivity analysis revealed that the grazed forage consumed
criterion is the most sensitive criterion under future climate scenarios. In addition, the most
resilient pasture composition was perennial ryegrass with red clover while the least resilient
pasture composition was orchardgrass with white clover. Overall, the pasture composition of
50% perennial ryegrass and 50% red clover was identified as the best pasture composition for all
criteria while being the most resilient composition.
Future studies should consider expanding of this work in different climatological regions to
better understand the impacts of climate change on grazing livestock systems. However, the
concept of sensitivity analysis that was introduced in this study has global implications and can

90

help with understanding the resiliency of the different pasture compositions under the future
climate scenarios. This can provide the scientific basis for decision-making while providing
knowledge to policy makers and stakeholders to reduce the negative impacts of climate change.
Meanwhile, the concept of pasture diversification introduced in this study as an adaption
measure has global implications. This study found that 1) the percentage of legume is the main
factor, besides climate, influencing the criteria. Overall, as the percentage of legume increased in
the pasture composition, the value of all four criteria increased. This means that adding more
legume to the composition can be beneficial for grazing dairy farms; 2) grass species combined
with 50% legume was identified as the best composition to reduce the negative impacts of
climate change while improving the farm economics; however, providing more than 50% legume
may increase risk of illness to dry cows; and 3) to maintain a high grazed forage production for
future scenarios, it is important to maintain at least 30% legume in the pasture composition.

91

5

5.1

FOOD FOOTPRINT AS A MEASURE OF SUSTAINABILITY FOR GRAZING
DAIRY FARMS

Introduction
Greenhouse gas emissions are the major cause of global climate change, leading to an

increase in global average temperature (IPCC, 2014). Anthropogenic GHG emissions are the
dominant cause of the increase in global warming since mid-20th century, resulting in the highest
atmospheric concentrations of carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O), in
the last 800,000 years (IPCC, 2014). Among anthropogenic activities, livestock productions play
a major role by contributing 14.5% of global anthropogenic GHG emissions. In perspective, the
livestock sector is a larger contributor to GHG emission than the transportation sector (14%)
(Gerber et al., 2013; IPCC, 2014). The GHG emission from the livestock sector is due to
activities including enteric fermentation (39%), manure management, application, and direct
deposit (26%), feed production (21%), land use change (9%), post-farm gate (3%), and direct
and indirect energy usage (2%) (Gerber et al., 2013). Furthermore, emissions from the livestock
sector accounts for 5%, 44%, and 53% of the total human emissions for CO2, CH4, and N2O,
respectively (Gerber et al., 2013). In addition to contributing to climate change, the livestock
sector negatively affects the environment through land degradation, water pollution, water
shortage, and biodiversity destruction (Steinfeld et al., 2006; Rojas-Downing et al., 2017c).
Meanwhile, livestock product demand (mostly dairy and meat) is also expected to double by
2050 (Garnett, 2009). Therefore, there is a need to fulfill the demand in a sustainable manner
(Thornton, 2010).
In this study, sustainability refers to the ability to progress by considering the environment,
economic feasibility, and socially acceptable practices to meet the needs of the present and future

92

generations. Therefore, sustainability requires consideration of environmental, economic, and
social components of a system while finding the balance point of this system (OECD, 2004). To
study these components, different indicators have been used that are known as sustainability
measures (footprints) and usually calculated through life cycle assessment (LCA) techniques
(ÄuÄek et al., 2012). The use of footprints has become so popular that a simple search of this
word in the Google Scholar resulted in more than a million hits. Footprint refers to the
quantitative measurement of the human activities burdens on sustainability (Hoekstra, 2008;
ÄuÄek et al., 2012). Each type of footprint has its own definition, units of measurement, and
tools for its estimation. Footprints can be categorized as environmental (e.g., carbon, water, and
energy footprints), social (e.g., human rights, corruption, poverty, health, job footprints),
economic (e.g., financial, economic footprint), and composite footprints (e.g., ecological
footprint) (ÄuÄek et al., 2012).
A single footprint or a combination of them have been used to measure sustainability in
the livestock sector. For example, FAO (2010) used a carbon footprint obtained from LCA to
examine the global dairy farming sustainability. Through this analysis, methane was identified as
the largest contributor to GHG emissions from the dairy sector. However, using a single footprint
is not always the right approach for measuring the overall sustainability, especially for the area
of food production (Finkbeiner, 2009). Dyer et al. (2014) performed a LCA where carbon
footprint along with the impacts on non-carbon footprint, such as water, energy, and diseases,
were considered to measure the environmental impacts of livestock production in Canada. An
environmental impact matrix and a network diagram were used to consider the non-carbon
footprint. Next, the crop areas and non-carbon footprint elements were associated to different
types of livestock using the area-reference index. Results suggested that a sustainable solution

93

that reduces the current level of carbon footprint cannot be met. However, an increase in grassfed beef while using grains for the non-ruminants can reduce CO2 loss from the soil (Dyer et al.,
2014). Vasilaki et al. (2016) also used a combination of water and carbon footprints to assess the
environmental impact of different types of yogurt productions in a dairy plant. They used a LCAbased water and carbon footprint âcradle to gateâ approach. The results showed that the major
CO2-eq contributors (over 80%) are raw milk and milk-based ingredients while cleaning
operations account for 70% of the water usage. However, the carbon and water footprint can still
be reduced by minimizing waste during the processing stage. Cecchini et al. (2016) performed
environmental and economic sustainability analysis of five dairy farms in Italy. The LCA-based
carbon footprint was used to perform an environmental assessment while the economic
assessment was done considering all costs associated the production process. To identify the
relationship between the environmental and economic impacts, a regression analysis was
conducted. The results showed improvement in the overall sustainability outcomes can be
achieved when considering interrelationships between carbon footprint and economic aspects.
In general, only a limited number of studies have considered multiple footprints (usually two
environmental footprints) to assess the environmental impact or sustainability, while the majority
of studies only used carbon footprint as a measure of sustainability (Adom et al., 2012; de LĂŠis et
al., 2015; FAO, 2010b; FlysjĂś et al, 2011; Mc Geough et al., 2012; OâBrien et al., 2014).
However, a comprehensive sustainability study should consider different aspects that are
impacted by a proposed activity such as water and energy requirements and economic feasibility.
This study is unique since for the first-time carbon, energy, water, and economic footprints are
considered to assess the sustainability of grazing dairy farms. Here, we propose to use multicriterion decision making to determine the balanced options between conflicting criteria

94

(footprints). A new measure named food footprint was introduced to mimic the overall impacts
of grazing dairy farms and helps with identifying the most sustainable production systems. Using
this new measure, the most sustainable milk production can be identified for a location of
interest. The specific objectives of this study were to: (1) identify the most sustainable milk
production in different locations; (2) identify the pasture composition for each targeted milk
production in different locations; and (3) understand the spatial variability of milk productions
and associated pasture compositions for different locations.
5.2

Materials and Methods

5.2.1 Study Area
The United States is the number one cowsâ milk producer in the world by producing on
average about 91 billion kilograms of liquid milk annually (FAO, 2016b). Meanwhile, the state
of Michigan is one of the top five milk-producing states in the USA which is why it has been
considered for this study (USDA-NASS, 2014c). Among dairy systems, the grazing dairy is
more likely to be impacted by climate variability because of dependency on forage quality and
quantity (Hatfield et al., 2014; Heckman et al., 2007). Therefore, to understand the impacts of
climate variability on grazing systems in Michigan, the study area was divided into several
regions. The National Oceanic and Atmospheric Administration (NOAA) Climate Division
Boundaries were initially considered for this purpose. However, these boundaries do not
represent areas of climatological homogeneity. Instead, the division boundaries were defined
initially (between 1909 and 1949) based on drainage basins or major crops, and then in the late
1950s into structured boundaries that would coincide with the political boundaries, most which
do not relate to climate (Guttman and Quayle, 1996).

95

Due to limitations of the NOAA Climate Division Boundaries, an alternative approach
was considered in which Michigan is divided into 10 regions based on the watershed boundaries.
This approach is aligned with the natural division of the land surface that were mainly influenced
by geology and climate factors. Considering a watershed as a subunit for pasture study is also
useful since the soil types and distributions within a watershed are determined by the geology,
climate, and vegetation within a watershed (OWEB, 1999). In addition, the vegetation patterns
are influenced by the local climate, which determines the timing and amount of precipitation and
solar energy that falls into the watershed (OWEB, 1999). The U.S. Geological Survey used a
national standard hierarchical system based on surface hydrologic features to delineate the
watersheds in United States and classify them into hydrologic unit codes (HUC) (USGS, 2017).
For this study, the six-digit HUC was used. The study area was divided into 10 regions (Figure
14) and data from a weather station located close to the centroid of the watershed was used
(Viessman and Lewis, 2003). However, this location does not represent the spatial climate
variability within the watershed but rather climate variability for a single point.

96

1

3
2
5

4
6

7
8
9
10

Figure 14. State of Michigan and its watershed boundaries
Twenty years of daily weather data for the period of 1985 to 2005 were obtained from the
NCDC. The maximum temperature, minimum temperature and precipitation range from 10.6 ËC
to 15.3 ËC, -1.5 ËC to 4.0 ËC, and 757.3 mm to 904.8 mm, respectively within the study area
(Table 10).

97

Table 10. Average daily temperature and precipitation for the weather stations within the study
area
Average Max.
Average Min.
Average
Temperature
Temperature
Precipitation
Region
Station number
(ËC)
SD*
(ËC)
SD*
(mm)
SD*
1
USC00200089
10.6
12.8
-1.5
11.6
824.8
5.6
2
USC00204090
12.0
12.4
-0.6
11.7
764.4
5.6
3
USC00205816
10.8
11.8
0.0
10.4
857.2
7.2
4
USW00014817
12.3
11.7
0.9
10.5
891.5
6.2
5
USC00208417
12.3
12.3
-1.4
10.9
780.5
5.3
6
USC00207820
13.3
11.5
1.2
10.1
761.4
5.8
7
USC00207222
14.1
12.2
3.2
10.2
823.2
6.1
8
USC00206658
14.3
11.7
4.0
10.0
757.3
5.7
9
USC00200552
14.9
11.4
3.7
10.0
904.8
6.3
10
USC00200032
15.3
11.5
3.4
9.8
901.4
6.5
*SD: Standard deviation
Besides daily temperature and precipitation, solar radiation was also required for a farm
system modeling. In order to obtain daily solar radiation for each location, a widely used
stochastic weather generator called WXGEN was used (Sharpley and Williams, 1990; Wallis and
Griffiths, 1995). WXGEN used the daily temperature and precipitation data to calculate the solar
radiation (Jeong et al., 2015).
5.2.2 Modeling Process
The modeling process involves data processing, performing model simulations, and
footprint analysis to identify the most sustainable milk production and appropriate pasture
composition for each location within the study area (Figure 15). Under the data processing,
typical farm management, size, pasture compositions, and target milk productions were collected
through a survey from producers and extension specialists within the study area. However, for
cases that actual data cannot be obtained from farm surveys (e.g., manure characteristics,
structures initial costs, investment in fences and watering systems, animal fiber intake capacity),
reported data from literature were used to create a representative farm for the study area (RojasDowning, 2013). In addition, long-term (20 years) weather data for each location were obtained
98

to understand the climate variabilities, 48 pasture composition, and 32 target milk productions
were identified. The aforementioned information provided the inputs to IFSM. IFSM can
estimate milk production and required information to calculate energy, water, carbon, and
economic footprints for different combinations of pasture compositions, target milk productions,
and location (watershed) scenarios. A multi-criterion decision making method called VIKOR
was used to determine the balanced options between different footprints. The results from
VIKOR analysis was used to determine the food footprint representing the overall impacts of a
grazing dairy farm management practice on energy, water, carbon, and farm economics. By
obtaining the food footprint, the most sustainable milk production and appropriate pasture
composition of each location was identified.

99

Figure 15. The conceptual framework of the modeling process

100

5.2.2.1 Data Processing
5.2.2.1.1 Grazing Farm Survey and Review
Since it is not practical to perform modeling analysis for each individual farm within the
study area, a representative farm was developed. Five different grazing dairy farms throughout
the study area were surveyed to collect data regarding the typical farm management, size, pasture
compositions, and target milk productions (Rojas-Downing et al., 2017a). The surveys, which
included field observations, were performed in the spring and summer of 2013 and the farms
were selected considering location, size, and farming practice. The survey questionnaires were
structured based on IFSM data requirements and performed during a single-visit. The survey
questionnaires cover a variety of topics including farm management, pasture area, alfalfa area,
machinery operations, grazing strategy, equipment, structures, number and type of animals, and
prices for several farm inputs and outputs (Supplementary Materials, Table S1). A summary of
the key survey data collected is provided in Table S2. In the case that the required data for the
IFSM could not be obtained from the survey, additional means including literature review and
discussions with Michigan State University extension specialists were used to obtain the data. A
summary of the key representative grazing dairy farm data is provided in Table 11.

Table 11. Representative grazing dairy farm description
Main IFSM inputs
Value/Description
Cows
Lactating cows
100
Replacement stock
80
Cattle
Small frame Holsteins (454 kg)
Feed
Relative forage-to-grain ratio
High
Crude protein supplement
Soybean meal 48%
Undergradable protein
Distiller grain
Facilities
Cow housing
Bedded pack barn
101

Table 11 (contâd)
Heifer housing
Feed facility
Milking center
Manure
Manure collection
Manure storage
Manure applied to the grazing
area
Manure applied to the alfalfa
area
Forage areas
Grazing area
Alfalfa area
Fertilizers application
Grazing strategy
Animals grazed
Time on pasture
Labor for grazing management
Tillage and planting
Operations time
Machinery
Silage and hay harvest

Calf hutches and dry lot
Short-term storage premix
Double four parlor
Scrapers with slurry pumps
Top-loaded concrete tank for a four-month period
20%
80%
54 ha
27 ha
No
All cows
Full days during the grazing season (April to October)
20 hours per week
Mid-October and mid-April
Chisel plow and a seedbed conditioner
Alfalfa harvested as silage and hay four times a year;
grass harvested once a year as silage

Economics*
Diesel fuel
$0.98/L
Electricity
$0.13/Kwh
Labor wage
$16/h
New forage stand
$494/ha
High quality hay
$265/tonne DM
Hay
$198/tonne DM
Milk
$0.36/kg
Machinery economic life
10 years
Structures economic life
20 years
Real interest rate
6%
*All major operations based on 2013 representative prices

5.2.2.1.2 Selection of pasture compositions
The pasture composition varies depending on the grazing management goal (e.g.,
biomass production, animal nutrition), seasonal biomass distribution, animal palatability, and
102

persistency over time (Sullivan et al., 2000). Based on survey data, literature review, and inputs
from extension specialists, the typical grass and legume species and compositions of the study
area were identified. The most typical pasture composition in Michigan includes 60%
orchardgrass and 40% white clover. However, the typical functional groups (functional features
of an ecosystem service) are cool-season grasses, such as orchardgrass (Dactylis glomerata) and
perennial ryegrass (Lolium multiflorum) and legumes, such as white clover (Trifolium repens)
and red clover (Trifolium pratense). The percentage of legume in the pasture composition varies
between 0% and 50% within the study area. Orchardgrass, white clover, and red clover are
resistant to cold weathers (winter hardiness), while perennial ryegrass is tolerant to wet soils.
Perennial ryegrass and white clover tolerate heavy grazing, while orchardgrass and red clover
can be used as a hay crop or grazed by animals (Barnes et al., 2007; Ogle and St. John, 2008;
USDA-NRCS, 2002). Meanwhile, all of the aforementioned pasture species are high in nutritive
value when harvested appropriately.
Due to global warming and change in climate trends, two additional species were also
evaluated with relatively better adaptability to extreme events than current typical pasture
species. These species are Kentucky bluegrass (Poa pratensis) and tall fescue (Lolium
arundinaceum). Both species have good winter hardiness in Michigan and moderate tolerance of
wet soils, while tall fescue is also tolerant of drought. Kentucky bluegrass is suitable for pastures,
while tall fescue can be used for either hay or animal grazing (Barnes et al., 2007).
5.2.2.1.3 Selection of Target Milk Productions
Milk production, under any dairy system, depends on the animal genetic constitution,
type of management, diet, and environmental factors (EPA, 2015). However, under a grazing
system, milk production is mainly dependent on pasture intake while pasture quantity and quality

103

is primarily dependent on climate variability (Hatfield et al., 2014; Heckman et al., 2007; Muller,
2004). Regarding the animal genotype, most of the U.S. dairy cows (90%) are black and white
Holsteins because they can produce large amount of milk, butterfat, and protein (EPA, 2015).
However, producers with grazing systems prefer small framed Holsteins because they are more
efficient in converting grass to milk (Flanders and Gillespie, 2015).
Based on the farm surveys, the target milk production ranges between 6,804 to 10,866
kg/cow/year. However, in this study we used a wider range (1,361 to 15,876 kg/cow/year) that
allowed the model to examine the impacts of introducing new pasture compositions on milk
productions. The range was simulated in steps of 454 kg/cow/year, resulting in 32 scenarios for
the target milk productions.
5.2.3 Model Simulations
5.2.3.1 Integrated Farm System Model (IFSM)
IFSM is a whole-farm simulation model developed by the U.S. Department of
Agriculture-Agricultural Research Service (USDA-ARS). IFSM integrates livestock and crop
production systems to evaluate resource use, economics, and environmental sustainability of the
farm practices (USDA-ARS, 2016). IFSM has been extensively used in around the world,
although the model was developed for temperate regions of the northern United States (e.g.
Michigan) and Southern Canada (Bryant and Snow, 2008; USDA-ARS, 2016; Wachendorf and
Golinski, 2006). The model has been validated with various farm practices and pasture
compositions under different climate conditions (Corson et al., 2007; Rojas-Downing et al.,
2017b; Rotz, 2004).
IFSM is a process-based model that integrates the farm components with major biological
and physical process over a period of time (USDA-ARS, 2016). The main input data for IFSM

104

include: soil characteristics, crop areas and type, grazing management, equipment and structures,
harvest, tillage, types of animals, feed, manure handling, and prices related to the whole-farm.
Using this information, IFSM simulates feed production and consumption, grazing management,
animal performance, feed storage, manure management, tillage, planting, and harvest operations,
and economics (Rotz et al., 2014).
5.2.3.1.1 Scenario analysis of the impacts of climate variabilities, pasture compositions, and
target milk productions on sustainability grazing dairy farms
Figure 16 presents the integrated scenario analysis used in this study to evaluate the
impacts of climate variabilities, pasture compositions, and target milk productions on
sustainability of grazing dairy farms. For each of the 10 locations (Figure 14), the sustainability
of a representative grazing dairy farm was evaluated considering different pasture compositions
and target milk productions. The pasture compositions include all possible combinations of the
four cool-season grass species with two legumes by varying the legume percentage in the
composition between 0% and 50%, at 10% increments. These combinations resulted in 48
different pasture compositions. In addition, for each pasture composition, each of the 32 target
milk productions, ranging from 1,361 to 15,876 kg/cow/yr, were simulated. The combination of
10 locations, 48 pasture compositions, and 32 target milk productions, resulted in 15,360
different scenarios that where simulated in IFSM.

105

Figure 16. Schematic representation of the integrated scenario analysis
5.2.4 Footprint Analysis
Beside economic footprint, IFSM can directly be used to determine carbon, water, and
energy footprints of the representative grazing dairy farm. However, in order to calculate the
economic footprint, the IFSM model outputs are still necessary to account for the whole farm
profits and the milk productions. All footprints are ultimately normalized to the total amount of
milk production. Milk production is estimated as the maximum amount of milk that can be
produced per cow in a year. The maximum milk production is calculated based on the desired
target milk production, amount of feed, forage nutritional value, and type of concentrates fed
(Rotz, et al., 2014). The milk production was converted to the fat and protein corrected milk
(FCPM) using milk fat content of 4.2% and milk protein content of 3.3% (Rojas-Downing et al.,
2017a). Below, more detailed information is provided for each footprint used in this study:
Carbon footprint: Carbon footprint is one of the key environmental indicators that presents the
amount of GHG emissions over the lifecycle of a product or process (BSI, 2008). The unit of
measurement is tonnes of CO2 equivalents (t CO2-eq), which is based on the GWP indicators

106

(EC, 2007). Several studies have applied carbon footprint to assess the sustainability of the dairy
sector in regard to: allocating GHG emissions between meat and milk (Mc Geough et al., 2012);
comparing emissions of different production systems (e.g., confinement, grazing) (de LĂŠis et al.,
2015; FlysjĂś et al, 2011; OâBrien et al., 2014; Zehetmeier et al., 2014); and evaluating different
feeds (e.g. grain, forage) (Adom e t al., 2012). IFSM determines the carbon footprint as the net
emission of CO2, CH4, and N2O while considering all sources and sinks of CO2. In addition, a
carbon balance is enforced, so a portion of the CO2 assimilated in the feed production (grazed
and harvested forage) is exported through feed (harvested forage), milk, and animal productions.
Finally, the net emission is determined through a partial life cycle assessment by assuming the
GWP of 298 CO2-eq/kg for CH4 and 298 CO2-eq/kg for N2O (IPCC, 2001). In this study, the unit
CO2-eq per kg of FCPM is used for the carbon footprint since it was normalized to the total
amount of milk production.
Water footprint: Water footprint refers to the volume of water consumed per unit of time or per
functional unit in a LCA (Galli et al., 2011, 2012). Water footprint provides a better
understanding of water management and usage in a system. Therefore, it has been frequently
applied to assess: techniques to improve water usage on dairy farms (Palhares and Pezzopane,
2015; Murphy et al., 2017; Zonderland-Thomassen and Ledgard, 2012), water availability for
large-scale production systems (Huang et al., 2014), water scarcity (Owusu-Sekyere et al., 2016),
and different types of dairy products (Vasilaki et al., 2016). IFSM determines the water footprint
as the total amount of water used in the dairy farm system. The unit of measurement is kg of
water per kg of FCPM. The major use of water in the grazing dairy farm include the production
of feed crops, drinking water for animals, animal cooling, and cleaning of the parlor and holding
areas on the farm (Rotz, et al., 2014).

107

Energy footprint: Energy footprint is defined as the amount of energy required (fossil- and
renewable-based energy) per functional unit (Sobhani et al., 2012). Both the units of energy (e.g.,
Megajoule (MJ)) and global area (e.g., global hectare (gha)) can be used to report energy
footprint (Hermann et al., 2011). Energy footprint can provide valuable information about the
energy usage for different types of dairy systems, which ultimately can be used to identify
different energy saving strategies (Pagani et al., 2016). IFSM determines the energy footprint as
the total energy required to produce feed and milk while excluding the solar energy used to grow
feed crops. This includes all fuel and electricity used for milking, ventilation, and lighting, as
well as energy requirements for resources used on the farm (e.g., tractor, equipment used for feed
production, feeding, and manure handling). The unit of measurement is MJ per kg of FCPM.
Economic footprint: ÄuÄek et al. (2012) defined the economic footprint as the net economic
impact of a product or process, where the unit of measurement is currency. Meanwhile, this
footprint is mainly used by organizations, institutions, companies, and universities and its
application in agriculture sector is scarce (ÄuÄek et al., 2012; Clayton-Matthews and Watanabe,
2012). The accounting period in the IFSM model for performing the economic analysis is one
year. The model determines all costs and benefits associated with productions (e.g., milk, feed).
The economic analysis also includes fixed (e.g., equipment and structures) and variable (e.g.,
labor, resources) production costs. However, no interaction between the farm and the
surrounding markets is allowed. As a result, the efficiency of the technical and economic
production of the farm for a set of relative prices can be evaluated (Rotz, et al., 2014). The
economic footprint was determined by considering the whole-farm profit per kg of milk
production ($ per kg of FCPM). The whole farm profit is calculated by subtracting incomes from
the total costs. Incomes include, milk, animals and feed sales, and saving through usage of

108

family labor. Labor was considered as income for this study because 97% of dairy farmers in this
region are family owned, where most of the labor is performed by the family members (United
Dairy Industry of Michigan, 2016). The costs include feed, manure handling, animal housing,
animal care, and milking.
5.2.4.1 Food Footprint
In this study, the food footprint measure was used to mimic the overall impacts of grazing
dairy farms and helps with identifying the most sustainable production systems. This new
measure was calculated using a multi-criterion decision-making (MCDM) in which lower values
represent a more sustainable scenario. MCDM help with determining the best overall solution
considering multiple criteria (Zanakis et al., 1998).
The MCDM method used in this study is called multi-criteria optimization and
compromise solution (VIKOR). VIKOR was selected since it can solve complex decisionmaking problems with conflicting criteria (benefit and cost criteria) and non-commensurable
(different units) criteria, which is the case in this study (Opricovic and Tzeng, 2004). VIKOR
employs an aggregated function representing âcloseness to the idealâ, which originated in the
compromise programming method and determines a compromise solution, providing a maximum
âgroup utilityâ for the âmajorityâ and a minimum of an individual regret for the âopponentâ
(Opricovic and Tzeng, 2007; Opricovic, 2009). As a result, the highest ranked alternative by the
VIKOR method is the closest to the ideal solution (Opricovic, 2011).
In this study, there are four criteria entailing three cost criteria (water, energy, and carbon
footprint), and one benefit criterion (economic footprint) in relation to environmental and
economic factors. Thus, the VIKOR optimization method can provide efficient ranking regarding
the selection of the best scenario (alternative).

109

A generic form of a MCDM matrix for a set of conflicting criteria {đ1 , đ2 , âŚ , đđ , âŚ , đđ }
and a set of possible alternatives {đ´1 , đ´2 , âŚ , đ´đ , âŚ , đ´đ } can be constructed with the performance
ratings of đđđ . The weight vector, which refers to the relative importance of criteria, is modeled as
đ = [đ¤1 , đ¤2 , âŚ , đ¤đ , âŚ , đ¤đ ] where âđđ=1 đ¤đ = 1.

Alternatives

Criteria (factors)
đ1

đ2

âŚ

đđ

đ´1

đ11

đ12

âŚ

đ1đ

đ´2

đ21

đ22

âŚ

đ2đ

âŚ

âŚ

âŚ

âŚ

âŚ

đ´đ

đđ1

đđ2

âŚ

đ2đ

The main goal of the VIKOR optimization method is to determine the best and worst values
of alternatives in accordance with the cost/benefit criteria. There are named positive-ideal solution
(đđâ ) and the negative-ideal solution (đđâ ), respectively. Moreover, the compromise rankings are
performed by comparing the measure of closeness to the ideal alternative (Opricovic and Tzeng,
2007). The multi-criteria measure for compromise ranking is developed from a LP- metric
proposed by Yu (1973) used as an aggregating function in a compromise programming method.
Development of the VIKOR method started with the following form of LP- metric (Equation 1):
1/đ
â
(đđ â đđđ )
{â[đ¤đ . â
]}
(đđ â đđâ )đ
đ=1
đ

đżđ,đ =

;

1â¤đâ¤â

(1)

where, đżđ=1,đ (also known as đđ ) computes the sum of deviations and evaluates the maximum
group utility or majority (concordance), whereas đżđ=â,đ (also known as đđ ) measures the
maximum deviations and signifies the minimum individual regret of the opponent (discordance)

110

(Opricovic and Tzeng, 2004). The detailed procedure of the compromise ranking algorithm of the
VIKOR optimization method are described below.
Step1. The best and worst values of all the criterion functions (đđâ and đđâ ) are determined according
to the beneďŹt or cost criteria.
Step2. The đđ and đđ values are determined for all alternatives using the following relations.
đ

đđ = â đ¤đ . (
đ=1

|đđâ â đđđ |
|đđâ â đđâ |

đđ = max [đ¤đ . (
đ=1,2,..,đ

)

|đđâ â đđđ |
|đđâ â đđâ |

(2)
)]

(3)

Step3. The minimum and maximum values of đđ and đđ are estimated using equation 4:
đ â = min đđ , đ â = max đđ , đ â = min đđ , đ â = max đđ
đ

đ

đ

đ

(4)

Step4. Computing the values of đđ for all alternatives (in this case, i = 4) using Equation 5. đđ is a
measure that simultaneously takes đđ , đđ , and their best and worst values into consideration.
đđ = đ.

(đđ â đ â )
(đđ â đ â )
+
(1
â
đ).
(đ â â đ â )
(đ â â đ â )

(5)

where, the parameter of đ is initiated as strategy weight for majority of criteria (or the maximum
group utility) and (1 â đ) is the weight of individual regret. The compromise can be selected with
consensus (đ=0.5), voting by majority (đ>0.5), or veto (đ<0.5). Usually, the value of đ is taken
as 0.5. This means that the decision maker considers the same weights for the effect of đđ (50%)
and đđ (50%). However, any value between 0 and 1 can be considered for đ. In this study, we
considered consensus that are commonly used for a compromise solution (Opricovic and Tzeng,
2007; Sayadi et al., 2009).
Step5. The alternatives are ranked in separate lists based on ascending đ, đ, and đ values.

111

Step6. The following conditions are checked. If both conditions are satisfied, alternative đ´â˛ that
has minimum đ value, is proposed as a compromise solution.
Condition 1: It is known as âacceptable advantageâ condition in decision making, denoting
that the difference of the first two ranks of alternatives by đ (Aâ˛ : the first rank, đ´": the
second rank) satisfies Equation 6.
đ(đ´") â đ(đ´â˛) âĽ đˇđ

(6)

where, đˇđ = 1â(#đđ đđđĄđđđđđĄđđŁđđ  â 1).
Condition 2: It is known as âacceptable stabilityâ condition in decision-making, meaning
that đ´â˛ alternative must also be the best ranked alternative by the đ or/and đ measures.
If one of these two conditions is not satisfied, then a set of compromise solutions are recommended
as follows:
â˘

Alternatives đ´â˛ and đ´" are proposed if only Condition 2 is not satisfied, or

â˘

Alternatives {đ´â˛ , đ´", âŚ , đ´(đ) } are proposed if Condition 1 is not satisfied. Where, đ´(đ)
is determined by the Equation 7 for maximum M:
đ(đ´(đ) ) â đ(đ´â˛ ) < đˇđ

(7)

5.2.5 Spatial Variability of Sustainable Pasture Composition and Milk Production
In order to identify the spatial variability of the most sustainable milk production and
appropriate pasture composition in the 10 locations of the study area, a cluster analysis was
performed. A cluster analysis involves classifying, without supervision, patterns into groups by
acquiring insight into data (e.g., identifying irregularities) (Celebi et al., 2013; Woznicki et al.,
2015). The most widely used clustering algorithm is called k-means (Jain, 2010). This algorithm
is selected here due to its versatility (e.g., algorithm can be modified), simplicity of application,
112

and invariance to data ordering (Celebi et al., 2013). The procedure in this algorithm starts with
random selection of centers for arbitrary clusters, follows by assigning each data point to the
nearest center, and finalizes by reassigning the means (Celebi et al., 2013).
5.2.6 The Social Component of Grazing Dairy Farms
The ultimate goal of sustainability assessment is to provide valuable information to
producers as they are making management decisions. Meanwhile, even the most sustainable
solution may be rejected by producersâ due to negative social implications (OECD, 2004). One
way to include the social implication of a management practice is to calculate the social
footprint. However, this measure has not been introduced in the concept of the grazing dairy
farm and because of variability in management practices and preferences of dairy farmers, it is
very difficult to achieve a standardization (ÄuÄek et al., 2012). Therefore, the social measure of
sustainably was not included in determining the food footprint, but should be considered when
making management recommendation that are both sustainable and socially acceptable.
In general, and in order to incorporate the social aspect of a management scenario, an
extensive stakeholder feedback is required (von Keyserlingk et al., 2013). For this study, this
information was collected from extension specialists. Critical public issues related to grazing
dairies include food safety, animal health and welfare, landscape quality, and labor (van Calker
et al., 2005; von Keyserlingk et al., 2013). The public increasingly views access to pasture as a
key component of acceptable welfare and health for dairy cattle (von Keyserlingk et al., 2013;
Schuppli et al., 2014), and often views dairy products produced from pastured cattle as safer and
more desirable (von Keyserlingk et al., 2013). This should help grazing dairies, but unfortunately
there is a disconnect between what consumers say they value and how much they are willing to
pay for it at the moment of food purchase (von Keyserlingk et al., 2013). Many specialty dairy

113

processors in the Great Lakes region are currently saturated with milk, which limits opportunities
for new grazing dairies to get access to price premiums related to specialty product labeling.
Labor management was cited by von Kreyserlingk et al. (2013) as one of the greatest
immediate threats to sustainability of the US dairy industry. While this is partly an economic
consideration, there is also a strong social aspect related to perceived desirability of this type of
work. Immigrant laborers are often recruited, many of them undocumented (Martin, 2002) and
poor working conditions are a concern. Potential loss of the undocumented immigrant labor pool
due to proposed changes to immigration enforcement policies presents a severe challenge to
conventional dairy systems (von Keyserlingk et al., 2013). Grazing dairy systems may possibly
avoid part of this labor challenge because farms are generally smaller than large confinement
dairies and may be able to hire family members to accomplish most work. However, when hired
labor is needed, grazing dairies may be less able to compete with large dairies for competitive
wage rates in a shrinking labor pool.
5.3

Results and Discussion

5.3.1 Identification of the Most Sustainable Milk Production
The goal of this section is to identify the most sustainable milk production for the study
area considering the food footprint measure in which the lower value represents a more
sustainable milk production. As it was discussed in the Materials and Methods section, the range
between 1,361 and 15,876 kg/cow/year of target milk production was evaluated in this study.
However, after simulating the IFSM model for all locations, the maximum milk production was
8,618 kg/cow/year. The number looks reasonable since a previous study showed that the typical
target milk production level for grazing cows is about 7,700 kg/cow/year (Conner, et al., 2007).

114

Box and whisker plots (Figures 17 and S29) present the variability of the food footprint
for three ranges of milk production: low (1,361 to 3,629 kg/cow/year), medium (4,082 to 5,897
kg/cow/year), and high (6,350 to 8,618 kg/cow/year). Figure 17 shows the pasture compositions
of the four grass species with red clover while figure S29 shows the pasture compositions of the
four grass species with white clover considering all percentages of legume (0% to 50%).
However, both Figures 17 and S29 present a similar trend, where the high range of milk
production presents lower values of food footprint, while having less variability among the
pasture compositions and locations. Since all the footprints under study (energy, water, carbon,
and economic) were weighted equally, the low variability under the high range of milk
production represents the limited improvement in sustainability that is being gained by
increasing the target milk production from 6,350 to 8,618 kg/cow/year. Conversely, the low
range of milk production presents higher values of food footprint, while variability is much
greater among the pasture compositions and locations. This suggests that the most sustainable
range of milk production for any of the pasture compositions for this study is between 6,350 to
8,618 kg/cow/year, while having the least variability.
Moreover, the most appropriate pasture composition for each range of milk production
was identified. The most appropriate pasture composition for the high range of milk production
for all 10 locations is perennial ryegrass combined with red clover. This composition had the
lowest median food footprint (0.0473). Perennial ryegrass was also identified as the most
appropriate grass for the pasture composition in the medium range of milk production. However,
the most appropriate composition was achieved when it is combined with white clover (median
food footprint =0.2037). In the case of the low range of milk production, the most appropriate
pasture composition was identified as tall fescue with white clover, with a median value of

115

0.7069. Therefore, by identifying the most sustainable range of milk production and its
respective pasture composition, the sustainability of the grazing dairy farm can be considerably
improved.

Figure 17. Food footprint for different combinations of red clover with a) Kentucky bluegrass,
b) orchardgrass, c) perennial ryegrass, and d) tall fescue for low (1,361 to 3,629 kg/cow/year),
medium (4,082 to 5,897 kg/cow/year), and high (6,350 to 8,618 kg/cow/year) milk production
5.3.2 Identification of the Most Appropriate Pasture Composition
To identify the most appropriate pasture composition for the study area, all compositions
were ranked from 1 to 48 based on their food footprint score in which lower scores ranked
higher. Figures 18 to 20 shows a summary histogram map of the pasture compositions ranking
based on the results from all locations within the study area. The y-axis represents the pasture
116

compositions, were each row signifies steps of 10% legume ranging from 0% to 50%. The x-axis
represents the pasture compositions food footprint ranking from 1 to 48, were 1 indicates the
most appropriate pasture composition and 48 the least appropriate. The color bar at the right side
of histogram map shows how frequently a certain rank for a specific pasture composition has
been obtained. The yellow color indicates the maximum amount of counts in which a pasture
composition obtained that rank. As the color spectrum moves from yellow to dark blue (topdown), it indicates that the pasture composition obtained that rank less frequently than the other
ranks. The dark blue color indicates that the pasture composition did not obtained that rank. As
an example, on row one in Figure 18 (grass with 0% red clover), rank 47 happened more
frequently than any other ranking. In addition, this composition is one of the least appropriate
compared to any other grass and red clover mixtures (10% to 50%).
In general, having white clover in the pasture composition helps with improving
sustainability as the grass and white clover composition ranked higher than the grass and red
clover composition (Figure 18). In addition, the milk production becomes less sustainable when
no legume is added (Figure 18, rows 1 and 7). The best rankings for both red clover and white
clover are when 10% of these legumes are combined with a grass (Figure 18, rows 2 and 8).
However, white clover resulted in higher sustainability in milk production with 20%, 30%, and
even 40% in the composition than red clover (Figure 18, rows 9, 10, and 11, respectively).

Figure 18. Histogram map for legume variations with grass species
117

As presented in Figure 19, the least appropriate pasture compositions are Kentucky
bluegrass with no legume (row 1), orchardgrass with 0% legume (row 7), and perennial ryegrass
with 0% and 50% legume (rows 13 and 18, respectively). Among all grasses, the presence of tall
fescue resulted in the most sustainable milk production especially when it is combined with 10%
of legume (Figure 19, row 20). However, milk production becomes less sustainable with an
increase of legume percentage in the pasture composition with tall fescue.

Figure 19. Histogram map for grass species for different variations of legume
Comparing all compositions of pastures, Figure 20 shows that as legume percentage
increases (between 10% to 50%) the overall grazing dairy farm practice becomes less
sustainable. However, when grasses are not combined with legume, the pastures are ranked the
lowest. This indicates when a legume is not present in the pasture composition, the farm is less
sustainable. In summary, the least appropriate pasture composition was perennial ryegrass with
no legume (Figure 20, row 25) and the most appropriate pasture composition is tall fescue with
10% white clover (Figure 20, row 20). In addition, tall fescue with white clover in general

118

presents the best rankings (Figure 20, rows 44 to 48), making it the most sustainable milk
production.

Figure 20. Histogram map of all pasture compositions
5.3.3 Spatial Variability of Sustainable Milk Productions and Associated Pasture
Compositions
In order to understand climate variability of grazing dairy farms, 10 locations were
identified within the study area. For each location, the most sustainable milk production and
associated pasture composition was identified and presented in Figures 21, 22, and 23. Figure 21
shows the most appropriate composition of grass and legume for each location. Tall fescue with
white clover was identified as the most suitable pasture composition in eight locations,
positioned mostly in the Lower Peninsula of Michigan. Meanwhile, the most appropriate pasture
compositions of the locations 1 and 3 (the Upper Peninsula) are perennial ryegrass with white

119

clover and tall fescue with red clover, respectively. Nine of the locations identified white clover
as the most appropriate legume in the pasture composition, which can be related to the high
tolerance for heavy grazing and its high quality, especially for pastures (Ogle and St. John,
2008). Whereas, red clover has more potential for hay production as it can be quickly depleted if
grazed continuously (USDA-NRCS, 2002). In addition, tall fescue was identified in nine of the
locations as the most appropriate grass in the pasture composition, which can be related to its
high tolerability of extreme weather conditions (drought and flood) (Cassida et al., 2014).
The legume percentages of the pasture compositions for each location were also
identified (Figure 23). Ten percent legume in the pasture compositions was identified as the most
suitable legume percentage in nine locations and 20% in the remaining location. In another study
performed in the Lower Peninsula of Michigan, 50% legume in the pasture composition was
identified as the best scenario for forage production and economics of a grazing dairy farm
(Rojas-Downing et al., 2017b). However, Rojas-Downing et al. (2017b) did not considered
carbon generation and energy usage in their study. This highlights the importance of performing
a comprehensive sustainability study to make more informed decisions.

120

Figure 21. Spatial variability of most appropriate pasture compositions in Michigan
Regarding the most sustainable milk production, 8,618 kg/cow/year was identified as the
most sustainable production for all locations (Figure 22), which is the highest produced among
all the simulated scenarios. This result is promising since it shows that sustainability is
achievable even at high levels of milk production, while having a balanced use of resources.
121

Finally, a cluster analysis was performed to develop the overall sustainability map for
the location in which both pasture composition and milk production were considered (Figure 23).
The study area was divided into three clusters, where 90% tall fescue combined with 10% white
clover and 8,618 kg/cow/year were identified as the most appropriate practices for the majority
of the locations. These maps can be used by policymakers to develop a strategy for grazing dairy
farms that involves reducing water, energy, and environmental footprints, while maximizing the
profit and production of dairy farmers.

Figure 22. Spatial variability of most sustainable milk productions in Michigan
122

Figure 23. Spatial variability of most sustainable milk productions and associated pasture
compositions in Michigan (Cluster analysis)

123

5.4

Conclusions
The overall goal of this study was to identify the most sustainable milk production for

grazing dairy farms in Michigan by using a new measure named food footprint that was
calculated using a multi-criterion decision making method.
Regarding the most sustainable milk production, the result of this study showed that the
high range (6,350 to 8,618 kg/cow/year) is the most sustainable for any combination of the
pasture compositions and locations, while having the least variability in terms of food footprint.
These results are promising since it encourages high levels of milk production while promoting
the most sustainable approach for the grazing farm management. However, in order to achieve
these levels of milk production (6,350 to 8,618 kg/cow/year), one should consider the perennial
ryegrass with red clover composition (Figure 17).
Regardless of the level of milk production, the results of the multi-criterion decision
making procedure (food footprint estimation) provide three major findings: 1) white clover in the
pasture composition helps with improving sustainability more than with red cover; 2) farm
sustainability improves with maintaining low percentages of legume (10% and 20%) in the
pasture mixture; 3) among all grasses, tall fescue resulted in the most appropriate pasture
composition for any target milk production and location, especially when is combined with 10%
of white clover.
The cluster analysis helped with identifying the most sustainable grazing farm practice
within the study area. In general, the highest level of milk production (8,618 kg/cow/year) was
identified as the most sustainable for all the 10 locations. However, the most appropriate pasture
composition varied among the locations. For most locations (locations 2, 4, 5, 6, 7, 9, 10), 90%

124

tall fescue combined with 10% white clover was identified as the most suitable pasture
composition.
The results of this study can be further customized to address the needs of both policy
makers and stakeholders including identification of: a) the most sustainable milk production for a
specific pasture composition, b) the most appropriate pasture composition for a specific milk
production level, and c) the most suitable legume percentage in the pasture composition for a
specific milk production. In addition, the concept of the food footprint, which refers to the
impacts of a food product on carbon, energy, water, and economics, can be expanded to other
agricultural sectors including livestock and crops. However, the social aspects were not
considered into final recommendation due to locality of these aspects (e.g. variability in
management practices, labor, and preferences of dairy farmers) in each location. Future work
should consider the locality of social aspects into the decision making. This will ultimately help
policy makers with implementation of sustainable farm management practices that have a higher
probability of adaptation by farmers.

125

6

CONCLUSIONS

This research introduced new concepts for evaluating the sustainability of a grazing dairy
system in Michigan and impacts of climate change/variability on this system while identifying
the most appropriate adaptation and mitigation strategies. These new concepts where tested in
the state of Michigan. In the first study, a sensitivity analysis was performed to measure the
resiliency of an adaptation measure to mitigate the impacts of future climate scenarios. In the
second study, a new measure called food footprint was introduced as a measure which
sustainable farm practices can be evaluated considering water, energy, carbon, and economic
footprints. Using this new measure, we identified the most sustainable milk production and
appropriate pasture composition for a study area. Overall, summary of the major findings of first
and second study are presented in Table 12.
The results of the first study provided four major findings concerning the use of legume in
the pasture composition: 1) an increased ratio of legume to grass species in a pasture
composition, can be beneficial to both farm economy and resource use; 2) grass species
combined with 50% legume was identified as the best composition to reduce likely impacts of
climate change while improving the farm economics; 3) to maintain a high-grazed forage
production under future climate scenarios, it is important to incorporate at least 30% legume in
the pasture composition; and 4) the presence of red clover in a pasture composition could
significantly improve the farm economics and production under all future climate scenarios. In
addition, the sensitivity analysis applied in this study have policy implications for understanding
the resiliency of the different pasture compositions. Overall, the pasture composition of 50%
perennial ryegrass and 50% red clover was identified as the best pasture composition while being
the most resilient to climate change. Summary of the major findings are presented in Table 13.

126

The food footprint measure that was introduced in the second study provided four major
findings: 1) the high range of target milk production (6,350 to 8,618 kg/cow/year) is the most
sustainable practice regardless of the pasture compositions; 2) white clover in the pasture
composition helps improve sustainability more than with red cover; 3) farm sustainability is
improved by maintaining low percentages of legume (10% to 20%) in the pasture mixture; and
4) among all grasses, tall fescue was considered as the best choice in the pasture composition for
any level of sustainable target milk production. The food footprint that was introduced in this
study can be adopted in other regions to improve sustainability by reducing water, energy, and
environmental impacts of grazing dairy farms, while maximizing the farm profit and
productions. Summary of the major findings are presented in Table 14.
Table 12. Overall findings
Study time period
1985-2005 / 2040-2060

Criteria
Economic

1985-2005 / 2040-2060

Resource Use

1985-2005 / 2040-2060

Resilience to climate change

1985-2005

Milk production
sustainability

127

Finding
50% perennial ryegrass and
50% red clover was identified
as the best pasture
composition
50% perennial ryegrass and
50% red clover was identified
as the best pasture
composition
Perennial ryegrass and red
clover had the overall lowest
sensitivity index
90% tall fescue combined
with 10% white clover and
8,618 kg/cow/year were
identified as the most
appropriate practices for most
of the locations

Table 13. Summary of findings of first study
Analysis
Criteria
Criterion
Ranking
Economic
Net Return
per cow over
milk
production

Whole Farm
profit
Resource
Use

Grazed
forage
consumed
Harvested
forage
production

Statistical

Economic

Net Return
per cow over
milk
production
Whole Farm
profit

Resource
Use

Grazed
forage
consumed

Harvested
forage
production

Sensitivity

Economic

Net Return
per cow over

Current
50% perennial
ryegrass and 50%
red clover

50% perennial
ryegrass and 50%
red clover

Grass with 50%
red clover
50% perennial
ryegrass and 50%
red clover

-The
implementation of
any grass, with
either red clover or
white clover did
not have
significant impact
- As the
percentage of
legume increased
in the pasture
composition the
criteria values
increased
- The presence of
red clover in the
pasture
composition
resulted in
significantly
higher criteria
values compared
to the white clover
-Perennial
ryegrass presented

128

RCP 4.5
50%
perennial
ryegrass
and 50%
red clover
50%
perennial
ryegrass
and 50%
red clover
Grass with
50% red
clover
50%
perennial
ryegrass
and 50%
red clover

RCP 8.5
50%
perennial
ryegrass
and 50%
red clover
50%
perennial
ryegrass
and 50%
red clover
Grass with
50% red
clover
50%
perennial
ryegrass
and 50%
red clover

Remains
same as
current

Improves

Remains
same as
current
Remains
same as
current

Remains
same as
current

Improves
Remains
same as
current

Remains
same as
current

Table 13 (contâd)
milk
production
Whole Farm
profit
Resource
Use
Grazed
forage
consumed

Harvested
forage
production

Overall
sensitivity

higher sensitivity
indexes
-White clover
presented higher
sensitivity indexes
-Orchardgrass
presented the
highest sensitivity
indexes
-White clover
presented higher
sensitivity indexes
- Orchardgrass
presented the
highest sensitivity
indexes
-Red clover
presented higher
sensitivity indexes
-Perennial
ryegrass and red
clover had the
overall lowest
sensitivity index

129

Table 14. Summary of findings of second study
Analysis

Identification of the most sustainable milk
production

Identification of the most appropriate pasture
composition

Spatial variability of sustainable milk
productions and associated pasture
compositions

Finding
-The most sustainable range of milk
production for any of the pasture
compositions is between 6,350 to 8,618
kg/cow/year
- The most appropriate pasture composition
for the high range of milk production for all
10 locations is perennial ryegrass combined
with red clover
- White clover in the pasture composition
helps with improving sustainability
- Tall fescue in the pasture composition helps
with improving sustainability
- When a legume is not present in the pasture
composition, the farm is less sustainable
- As legume percentage increases (between
10% to 50%) the grazing farm practice
becomes less sustainable
- The most appropriate pasture composition is
tall fescue with 10% white clover
- Majority of locations identified white clover
as the most appropriate legume in the pasture
composition
- Tall fescue was identified in majority of
locations as the most appropriate grass in the
pasture composition
- Ten percent of legume in the pasture
compositions was identified as the most
suitable legume percentage in the majority of
locations
-8,618 kg/cow/year was identified as the most
sustainable production for all locations
-90% tall fescue combined with 10% white
clover and 8,618 kg/cow/year were identified
as the most appropriate practices for the
majority of locations

130

7

FUTURE RESEARCH RECOMMENDATIONS

This research introduced new measures for evaluating the resiliency and sustainability of
grazing dairy systems under current climate variabilities and future climate scenarios. However,
additional research should be performed on the applicability and limitations of these methods.
The following are suggestions for future research:
â˘

It is important to consider the local and regional agri-environmental policies and crop
insurance subsidies into the selection of the most resilient pasture compositions. This will
help identify the most feasible grazing management scenario(s) for a specific location/region.

â˘

The food footprint analysis should also consider additional factors such as farmer preferences
to account for acceptability of the recommended approach.

â˘

The spatial distribution and variability of climate and soil textures should be considered in
support of a statewide analysis.

â˘

The methods developed under this study can be incorporated within an online decision
support tool to help dairy farmers with sustainability and resiliency evaluation of their farms.

â˘

Remotely sensed data can play a major role for the farm model parametrization. This will
reduce time for data collection/surveys while improving the accuracy of the input datasets.

â˘

Future work should consider the local social norms concerning grazing dairy farms. This will
ultimately help policy makers with implementation of sustainable farm management
practices that have a higher probability of adoption by farmers.

â˘

This study was innovative in that for the first time a comprehensive indicator was defined to
measure the level of sustainability in the context of milk production in a grazing farm. Future
studies can use this method to help policy makers in defining the acceptable level for
sustainable farming using the food footprint indicator. A threshold value can be identified for
131

different regions based on the stakeholder preferences concerning energy and water usage,
GHG emissions, and economic returns while considering regional limitations.

132

APPENDIX

133

Table S1. Major survey components (Rojas-Downing et al., 2017a)
Farm
Survey
Management
Survey Item
Category
Component
Crop
Grazed Forage
â˘ Area
â˘ Stand life
â˘ Manure/Fertilizer applied and rates
â˘ Initial sward dry matter and composition
Harvested Forage â˘ Area
â˘ Stand life
â˘ Manure/Fertilizer applied and rates
Grazing

Farm
Operations

Management

â˘
â˘

Strategy

â˘
â˘
â˘
â˘

Machinery

â˘
â˘
â˘
â˘
â˘
â˘
â˘

Harvest/ Feeding
Tillage/ Planting
Miscellaneous
Sequence and dates of operations
Preferred harvest schedule
Harvest type and operations
Harvest parameters
o Maximum moisture content
o Neutral Detergent Fiber

Storage

â˘

Animal Facilities

â˘
â˘
â˘

Type, capacity and cost of storage for:
o High and low quality forage
o Dry hay
Milking Center
Cow and Heifer housing
Feed Facility

Herd

â˘
â˘

Tillage/planting
Harvest

Structures

Animals

â˘
â˘
â˘

Spring, Summer, and Fall grazing area
Grazing management costs
o Investment in fence and watering systems
o Annual seeds and chemicals cost
Labor
Animals grazed
Time on pasture
Grazing period

Type
Number of lactating and young stock over/under one
year animals
Target milk production
Milking and animal handling labor
Annual livestock expenses
134

Table S1 (contâd)

Economic

Herd feeding

â˘
â˘

Manure

â˘
â˘
â˘
â˘
â˘

Feeding method for grain, silage and hay
Ration constituents
o Minimum dry hay
o Relative forage: grain
o Crude protein supplement
o Undegradable protein supplement
o Energy supplement
Collection and use method
Manure characteristics
Average hauling distance
Storage period, type, and initial cost
Type and amount of bedding use

Buying prices

â˘
â˘
â˘
â˘
â˘
â˘
â˘
â˘
â˘
â˘

Seed and chemicals per acre
Supplements
Hay
Minerals and vitamins
Bedding
High quality hay
Milk
Cull cow
Bred heifer
Suckling calf

Selling prices

135

Table S2. Key survey data collected (Rojas-Downing et al., 2017a)
Survey item
Description

Range

Farm size

-

83 to 146 ha

Harvested forage crops

Mostly alfalfa

-

Grazing forage crops

Variety of pasture compositions
including mostly perennial ryegrass,
orchardgrass, white clover, and red
clover

-

Lactating cows

-

90 to 750 cows

Cattle frame

Small Holsteins and New Zealand
Friesian

-

-

6,804 to 10,866
kg/cow/year

Cattle feed

High relative forage to grain ratios,
with pasture as the main feed and
grain as an energy supplement

-

Grazing system

Rotational

-

Stocking rate

-

1.0 - 5.2 animals/ha

Labor for grazing management

-

6 to 10 hours/week

Grazing time

22 hours per day

-

Animal facilities

Milking center, cow housing, heifer
housing, feed facility, short-term
manure storage, and forage storage

-

Target milk production

136

Table S3. Statistical comparison of net return, whole farm profit, grazed forage consumed and, harvested forage produced by different
grass species and legume percentages in the pasture composition (horizontal comparison)*
Criteria

Legume

Red
Clover
Net
Return
($/cow/yr)
White
Clover

Red
Clover
Whole
Farm
Profit
($/yr)
White
Clover

Legume
%
0
10
20
30
40
50
0
10
20
30
40
50
0
10
20
30
40
50
0
10
20
30
40
50

RCP 4.5
Perennial
Kentucky
Ryegrass Tall Fescue Bluegrass
1,497a
1,501a
1,499a
2,013a
2,006a
2,003a
2,063a
2,056a
2,053a
2,091a
2,084a
2,082a
2,110a
2,102a
2,101a
2,123a
2,118a
2,117a
1,497a
1,501a
1,499a
a
a
1,761
1,754
1,751a
1,894a
1,888a
1,885a
1,949a
1,947a
1,944a
1,971a
1,971a
1,969a
1,986a
1,986a
1,983a
110,455a
110,917a
110,679a
160,966a
160,236a
159,990a
166,019a
165,247a
165,022a
a
a
168,797
168,085
167,924a
170,712a
169,939a
169,809a
172,059a
171,540a
171,457a
110,455a
110,917a
110,679a
136,448a
135,693a
135,474a
149,326a
148,680a
148,382a
154,577a
154,400a
154,088a
156,710a
156,780a
156,500a
a
a
158,266
158,238
157,964a

137

Orchardgrass
1,501a
1,999a
2,049a
2,078a
2,099a
2,114a
1,501a
1,747a
1,881a
1,943a
1,968a
1,983a
110,915a
159,589a
164,557a
167,540a
169,592a
171,157a
110,915a
135,054a
148,074a
153,980a
156,403a
157,930a

Perennial
Ryegrass
1,514a
2,043a
2,092a
2,118a
2,136a
2,149a
1,514a
1,810a
1,930a
1,979a
1,997a
2,013a
112,244a
164,024a
168,908a
171,553a
173,310a
174,672a
112,244a
141,314a
152,817a
157,520a
159,344a
160,906a

RCP 8.5
Tall
Kentucky
Fescue
Bluegrass
1,519a
1,517a
2,034a
2,032a
2,083a
2,082a
2,110a
2,109a
2,129a
2,128a
2,143a
2,142a
1,519a
1,517a
a
1,803
1,801a
1,923a
1,921a
1,976a
1,973a
1,996a
1,993a
2,012a
2,009a
112,693a
112,503a
163,058a
162,890a
167,995a
167,840a
a
170,744
170,589a
172,652a
172,531a
174,087a
173,978a
112,693a
112,503a
140,634a
140,422a
152,196a
151,954a
157,249a
157,030a
159,221a
158,969a
a
160,806
160,590a

Orchardgrass
1,519a
2,028a
2,077a
2,105a
2,125a
2,139a
1,519a
1,797a
1,917a
1,972a
1,993a
2,009a
112,681a
162,459a
167,432a
170,230a
172,258a
173,678a
112,681a
140,001a
151,618a
156,933a
158,943a
160,512a

Table S3 (contâd)

Red
Clover
Grazed
Forage
Consumed
(tonne/yr)
White
Clover

0

23a

24b

24b

24b

22a

24b

23b

24b

10
20
30
40
50
0
10
20
30
40
50

173a
182a
185a
185a
185a
23a
118a
148a
161a
170a
175a

172ab
182a
184a
185a
185a
24b
116a
147a
160a
169a
174a

171ab
182a
184a
185a
185a
24b
116a
146a
160a
169a
174a

171b
181a
184a
185a
185a
24b
114a
145a
159a
168a
174a

174a
182a
184a
184a
184a
22a
130a
153a
164a
171a
176a

173a
182a
184a
184a
184a
24b
128a
152a
163a
171a
176a

173a
182a
184a
184a
184a
23b
128a
152a
163a
171a
176a

172a
181a
184a
184a
184a
24b
127a
151a
162a
171a
175a

246a
278a
296a
307a
315a
321a
246a
253a
259a
266a
271a
276a

250a
295a
312a
323a
330a
337a
250a
260a
268a
275a
281a
286a

251a
291a
308a
320a
327a
334a
251a
260a
268a
275a
280a
286a

250a
290a
307a
319a
327a
334a
250a
259a
267a
273a
279a
285a

251a
288a
306a
317a
326a
332a
251a
259a
267a
273a
279a
285a

0
245a
246a
245a
a
a
10
284
281
280a
Red
20
301a
298a
298a
a
a
Clover
30
312
309
308a
a
a
Harvested
40
319
317
317a
a
a
Forage
50
326
323
323a
Produced
0
245a
246a
245a
a
a
(tonne/yr)
10
254
254
253a
a
a
20
260
260
259a
White
Clover
30
267a
267a
266a
a
a
40
272
272
271a
a
a
50
277
277
276a
* Values followed by the same letters are not significantly different (Îą=0.05).

138

Table S4. Statistical comparison of net return, whole farm profit, grazed forage consumed and, harvested forage produced by different
grass species and legume percentages in the pasture composition (vertical comparison)*
Criteria

Legume

Red
Clover
Net
Return
($/cow/yr)
White
Clover

Red
Clover
Whole
Farm
Profit
($/yr)
White
Clover

Legume
%
0
10
20
30
40
50
0
10
20
30
40
50
0
10
20
30
40
50
0
10
20
30
40
50

Perennial
Ryegrass
1,497a
2,013b
2,063c
2,091cd
2,110de
2,123e
1,497a
1,761b
1,894c
1,949d
1,971de
1,986e
110,455a
160,966b
166,019c
168,797cd
170,712de
172,059e
110,455a
136,448b
149,326c
154,577d
156,710de
158,266e

RCP 4.5
Tall
Kentucky
Fescue
Bluegrass
1,501a
1,499a
2,006b
2,003b
2,056c
2,053c
2,084cd
2,082d
2,102de
2,101de
e
2,118
2,117e
1,501a
1,499a
1,754b
1,751b
1,888c
1,885c
1,947d
1,944d
1,971de
1,969de
1,986e
1,983e
110,917a
110,679a
b
160,236
159,990b
165,247c
165,022c
168,085cd
167,924d
169,939de
169,809de
171,540e
171,457e
110,917a
110,679a
135,693b
135,474b
148,680c
148,382c
d
154,400
154,088d
156,780de
156,500de
158,238e
157,964e

139

Orchardgrass
1,501a
1,999b
2,049c
2,078d
2,099de
2,114e
1,501a
1,747b
1,881c
1,943d
1,968de
1,983e
110,915a
159,589b
164,557c
167,540d
169,592de
171,157e
110,915a
135,054b
148,074c
153,980d
156,403de
157,930e

Perennial
Ryegrass
1,514a
2,043b
2,092c
2,118cd
2,136d
2,149d
1,514a
1,810b
1,930c
1,979d
1,997d
2,013d
112,244a
164,024b
168,908c
171,553cd
173,310d
174,672d
112,244a
141,314b
152,817c
157,520d
159,344d
160,906d

RCP 8.5
Tall
Kentucky
Fescue
Bluegrass
1,519a
1,517a
2,034b
2,032b
2,083c
2,082c
2,110cd
2,109cd
2,129d
2,128d
d
2,143
2,142d
1,519a
1,517a
1,803b
1,801b
1,923c
1,921c
1,976d
1,973d
1,996d
1,993d
2,012d
2,009d
112,693a
112,503a
b
163,058
162,890b
167,995c
167,840c
170,744cd
170,589cd
172,652d
172,531d
174,087d
173,978d
112,693a
112,503a
140,634b
140,422b
152,196c
151,954c
d
157,249
157,030d
159,221d
158,969d
160,806d
160,590d

Orchardgrass
1,519a
2,028b
2,077c
2,105cd
2,125d
2,139d
1,519a
1,797b
1,917c
1,972d
1,993d
2,009d
112,681a
162,459b
167,432c
170,230cd
172,258d
173,678d
112,681a
140,001b
151,618c
156,933d
158,943d
160,512d

Table S4 (contâd)
0

23a

24a

24a

10
173b
172b
171b
c
c
Red
20
182
182
182c
Clover
30
185d
184d
184d
d
d
Grazed
40
185
185
185d
d
d
Forage
50
185
185
185d
Consumed
0
23a
24a
24a
(tonne/yr)
b
b
10
118
116
116b
c
c
20
148
147
146c
White
d
d
Clover
30
161
160
160d
40
170e
169e
169e
e
e
50
175
174
174e
0
245a
246a
245a
b
b
10
284
281
280b
c
c
Red
20
301
298
298c
cd
cd
Clover
30
312
309
308cd
40
319d
317de
317de
Harvested
d
e
Forage
50
326
323
323e
a
a
Produced
0
245
246
245a
(tonne/yr)
10
254ab
254ab
253ab
bc
bc
20
260
260
259bc
White
cd
cd
Clover
30
267
267
266cd
40
272cd
272cd
271cd
d
d
50
277
277
276d
* Values followed by the same letters are not significantly different (Îą=0.05).

140

24a

22a

24a

23a

24a

171b
181c
184d
185d
185d
24a
114b
145c
159d
168e
174e
246a
278b
296c
307cd
315de
321e
246a
253ab
259bc
266cd
271cd
276d

174b
182c
184d
184d
184d
22a
130b
153c
164d
171e
176e
250a
295b
312c
323cd
330d
337d
250a
260ab
268bc
275bcd
281cd
286d

173b
182c
184d
184d
184d
24a
128b
152c
163d
171e
176e
251a
291b
308c
320cd
327d
334d
251a
260ab
268bc
275bcd
280cd
286d

173b
182c
184d
184d
184d
23a
128b
152c
163d
171e
176e
250a
290b
307c
319cd
327d
334d
250a
259ab
267bc
273bcd
279cd
285d

172b
181c
184d
184d
184d
24a
127b
151c
162d
171e
175e
251a
288b
306c
317cd
326d
332d
251a
259ab
267abc
273bcd
279cd
285d

Figure S1. Net return per cow over milk production sensitivity index of Perennial ryegrass and
white clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
index

Figure S2. Net return per cow over milk production sensitivity index of Tall Fescue and red
clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

141

Figure S3. Net return per cow over milk production sensitivity index of Tall Fescue and white
clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S4. Net return per cow over milk production sensitivity index of Kentucky bluegrass and
red clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

142

Figure S5. Net return per cow over milk production sensitivity index of Kentucky bluegrass and
white clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity
index

Figure S6. Net return per cow over milk production sensitivity index of Orchardgrass and red
clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

143

Figure S7. Net return per cow over milk production sensitivity index of Orchardgrass and white
clover pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S8. Whole farm profit sensitivity index of Perennial ryegrass and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

144

Figure S9. Whole farm profit sensitivity index of Tall Fescue and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S10. Whole farm profit sensitivity index of Tall Fescue and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

145

Figure S11. Whole farm profit sensitivity index of Kentucky bluegrass and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S12. Whole farm profit sensitivity index of Kentucky bluegrass and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

146

Figure S13. Whole farm profit sensitivity index of Orchardgrass and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S14. Whole farm profit sensitivity index of Orchardgrass and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

147

Figure S15. Grazed forage consumed sensitivity index of Perennial ryegrass and white clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S16. Grazed forage consumed sensitivity index of Tall Fescue and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

148

Figure S17. Grazed forage consumed sensitivity index of Tall Fescue and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S18. Grazed forage consumed sensitivity index of Kentucky bluegrass and red clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

149

Figure S19. Grazed forage consumed sensitivity index of Kentucky bluegrass and white clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S20. Grazed forage consumed sensitivity index of Orchardgrass and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

150

Figure S21. Grazed forage consumed sensitivity index of Orchardgrass and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S22. Harvested forage produced sensitivity index of Perennial ryegrass and white clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

151

Figure S23. Harvested forage produced sensitivity index of Tall Fescue and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S24. Harvested forage produced sensitivity index of Tall Fescue and white clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

152

Figure S25. Harvested forage produced sensitivity index of Kentucky bluegrass and red clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S26. Harvested forage produced sensitivity index of Kentucky bluegrass and white clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

153

Figure S27. Harvested forage produced sensitivity index of Orchardgrass and red clover pasture
composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

Figure S28. Harvested forage produced sensitivity index of Orchardgrass and white clover
pasture composition. P= precipitation (mm), T= temperature (ÂşC), S= sensitivity index

154

Figure S29. Food footprint for different combinations of white clover with a) Kentucky
bluegrass, b) orchardgrass, c) perennial ryegrass, and d) tall fescue for low (1,361 to 3,629
kg/cow/year), medium (4,082 to 5,897 kg/cow/year), and high (6,350 to 8,618 kg/cow/year)
milk production

155

REFERENCES

156

REFERENCES

Adhikari, U., Nejadhashemi, A.P., Woznicki, S.A., 2015. Climate change and eastern Africa: a
review of impact on major crops. Food Energy Secur. 4, 110â132.
Adom, F., Maes, A., Workman, C., Clayton-Nierderman, Z., Thoma, G., Shonnard, D., 2012.
Regional carbon footprint analysis of dairy feeds for milk production in the USA. Int. J.
Life Cycle Assess. 17, 520-534.
Alexandratos, N., Bruinsma, J., 2012. World agriculture towards 2030/2050: the 2012 revision.
ESA Working paper No. 12-03. FAO, Rome.
Aschmann, S., Cropper, J.B., 2007. Profitable grazing-based dairy systems. Range and Pasture
Technical Note No. 1, Natural Resources Conservation Service, U.S. Department of
Agriculture.
Asner, G.P., Elmore, A.J., Olander, L.P., Martin, R.E., Harris, A.T., 2004. Grazing systems,
ecosystems responses, and global change. Annu. Rev. Env. & Resour. 29, 261-299.
Aydinalp, C., Cresser, M.S., 2008. The effects of climate change on agriculture. Agric. &
Environ. Sci. 5, 672-676.
Baker, T.J., Miller, S.N., 2013. Using the Soil and Water Assessment Tool (SWAT) to assess
land use impact on water resources in an East African watershed. J. Hydrol. 486, 100â
111.
Barati, F., Agung, B., Wongsrikeao, P., Taniguchi, M., Nagai, T., Otoi, T., 2008. Meiotic
competence and DNA damage of porcine oocytes exposed to an elevated temperature.
Theriogenology. 69, 767-772.
Barnes, A.P., 2013. Heterogeneity in climate change risk perception amongst dairy farmers: A
latent class clustering analysis. Appl. Geogr. 41, 105-115.
Barnes, R., Nelson, C., Moore, K., Collins, M., 2007. Forages: an introduction to grassland
agriculture, 6th ed. Iowa State University Press, Ames, Iowa.
Batima, P., Bat, B., Tserendash, L., Bayarbaatar, S., Shiirev-Adya, S., Tuvaansuren, G.,
Natsagdorj, L., Chuluun, T., 2005. Adaptation to Climate Change. Vol. 90, ADMON
Publishing, Ulaanbaatar.
Beauchemin, K.A., Eriksen, L., Norgaard, P., Rode, L.M., 2008. Nutritional management for
enteric methane abatement: a review. Aust. J. Exp. Agr. 48, 21â27.

157

Beauchemin, K.A., McAllister, T.A., McGinn, S.M., 2009. Dietary mitigation of enteric methane
from cattle. CAB Reviews: Perspectives in Agriculture, Veterinary Science, Nutrition
and Natural Resources. 4(35), 1-8.
Bellarby, J., Tirado, R., Leip, A., Weiss, F., Lesschen, J.P., Smith, P., 2013. Livestock
greenhouse gas emissions and mitigation potential in Europe. Glob. Change Biol. 19, 318.
Benchaar, C., Pomar, C., Chiquette, J., 2001. Evaluation of dietary strategies to reduce methane
production in ruminants: a modeling approach. Can. J. of Anim. Sci. 81, 563-574.
Berman, A.J., 2005. Estimates of heat stress relief needs for Holstein dairy cows. J. Anim. Sci.
83, 1377-1384.
Bernabucci, U., Lacetera, N., Basirico, L., Ronchi, B., Morera, P., Seren, E., Nardone, A., 2006.
Hot season and BCS affect leptin secretion of periparturient dairy cows. Dairy Sci., 89:
348-349.
Bernabucci, U., Lacetera, N., Ronchi, B., Nardone, A., 2002. Markers of oxidative status in
plasma and erythrocytes of transition dairy cows during hot season. J. Dairy Sci. 85,
2173-2179.
Birhanu, B.Z., Ndomba, P.M., Mtalo, F.W., 2007. Application of SWAT model for mountainous
catchment, in: Lake Abaya Research Symposium (LARS) - Catchment and Lake
Research. University of Siegen and Arba Minch University, Siegen.
Boadi, D., Benchaar, C., Chiquette, J., MassĂŠ, D., 2004. Mitigation strategies to reduce enteric
methane emissions from dairy cows: Update review. Can. J. of Anim. Sci. 84, 319â335.
Bolin, B., Crutzen, P.J., Vitousek, P.M., Woodmansee, R.G., Goldberg, E.D., Cook, R.B., 1982.
SCOPE 21- The Major Biogeochemical Cycles and Their Interactions[WWW
Document]. Scientific Committee On Problems of the Environment (SCOPE). URL
http://www.scopenvironment.org/downloadpubs/scope21/chapter01.html
Bouwman, A.F., 1996. Direct emission of nitrous oxide from agricultural soils . Nutr. Cycl.
Agroecosys. 46(1), 53-70.
Bradley, N., 2007. The Response Surface Methodology. Indiana University South Bend.
Bruinsma, J., 2003. World Agriculture: Towards 2015/2030: An FAO Perspective. Earthscan,
London.
Bryant, J.R., Snow, V.O., 2008. Modelling pastoral farm agroâecosystems: A review. New Zeal.
J. Agric. Res. 51, 349â363.
BSI (British Standard Institute), 2008. Guide to PAS 2050. How to Assess the Carbon Footprint
of Goods and Services. British Standards, London, UK.

158

Burke, D., 2001. Dairy Waste Anaerobic Digestion Handbook [WWW Document].
Environmental Energy Company, Washington. URL
http://makingenergy.com/newsite/publications/Dairy%20Waste%20Handbook%20doc.pd
f
Calvosa, C., Chuluunbaatar, D., Fara, K., Mwandundu, S., Firmian, I., Benhammouche, A. and
Oehninger, E., 2009. Livestock and climate change: Livestock Thematic Papers Tools for
project design. IFAD (International Fund for Agricultural Development), Rome, Italy.
Carvalho, G., Moutinho, P., Nepstad, D., Mattos, L., Santilli, M., 2004. An Amazon perspective
on the forestâclimate connection: Opportunity for climate mitigation, conservation and
development?. Environ. Dev. & Sustainability. 6(1â2), 163â174.
Casey, K.D., Bicudo, J.R., Schmidt, D.R., Singh, A., Gay, S.W, Gates, R.S., Jacobson, L.D.,
Hoff, S.J., 2006. Air quality and emissions from livestock and poultry production/waste
management systems, in: Rice, J.M., Caldwell, D.F., Humenik, F.J. (Eds.), Animal
Agriculture and the Environment: National Center for Manure and Waste Management
White Papers. American Society of Agricultural and Biological Engineers, St. Joseph,
Mich, pp. 40.
Cassida, K., Paling, J., Kapp, C., 2014. 2014 Michigan forage variety test report. Michigan State
University Extension, Forage Factsheet #15â01. East Lansing, Michigan.
Cecchini, L., Torquati, B., Paffarini, C., Barbanera, M., Foschini, D., Chiorri, M., 2016. The
Milk Supply Chain in Italyâs Umbria Region: Environmental and Economic
Sustainability. Sustainability 8, 728.
Celebi, M.E., Kingravi, H.A., Vela, P.A., 2013. A comparative study of efficient initialization
methods for the k-means clustering algorithm. Expert Syst. Appl. 40, 200â210.
Chapman, S.C., Chakraborty, S., Dreccer, M.F., Howden, S.M., 2012. Plant adaptation to climate
change: opportunities and priorities in breeding. Crop & Pasture Sci. 63, 251-268.
Chase, L.E., 2012. Climate Change Impacts on Dairy Cattle. Climate Change and Agriculture:
Promoting Practical and Profitable Responses [WWW Document]. URL
http://www.climateandfarming.org/pdfs/FactSheets/III.3Cattle.pdf
Chhabra, A., Manjunath, K.R., Panigrahy, S., Parihar, J.S., 2013. Greenhouse gas emissions
from Indian livestock. Climatic Change. 117, 329â344.
Clayton-Matthews, A., Watanabe, P., 2012. Massachusetts Immigrants by the Numbers , Second
EditionâŻ: Demographic Characteristics and Economic Footprint.
Conant, R.T., Paustian, K., Elliott, E.T., 2001. Grassland management and conversion into
grassland: Effects on soil carbon. Ecol. Appl. 11, 343â355.

159

Conner, D., Heller, M.C., Cocciarelli, S., Hamm, M.W., 2007. Opportunities in Grazing Dairy
Farms: Assessing Future Options. A white paper of the C.S. Mott Group for Sustainable
Food Systems at Michigan State University, East Lansing, MI.
Corson, M.S., Alan Rotz, C., Howard Skinner, R., Sanderson, M.A., 2007. Adaptation and
evaluation of the integrated farm system model to simulate temperate multiple-species
pastures. Agric. Syst. 94, 502â508.
ÄuÄek, L., KlemeĹĄ, J.J., Kravanja, Z., 2012. A Review of Footprint analysis tools for monitoring
impacts on sustainability. J. Clean. Prod. 34, 9â20.
de LĂŠis, C.M., Cherubini, E., Ruviaro, C.F., PrudĂŞncio da Silva, V., do Nascimento Lampert, V.,
Spies, A., Soares, S.R., 2015. Carbon footprint of milk production in Brazil: a
comparative case study. Int. J. Life Cycle Assess. 20, 46â60.
De Rensis, F., Scaramuzzi, R.J., 2003. Heat stress and seasonal effects on reproduction in the
dairy cow: a review. Theriogenology. 60, 1139-1151.
Denef, K., Archibeque, S., Paustian, K., 2011. Greenhouse Gas Emissions from U.S. Agriculture
and Forestry: A Review of Emission Sources, Controlling Factors, and Mitigation
Potential: Interim report to USDA under Contract #GS-23F-8182H[WWW Document].
URL
http://www.usda.gov/oce/climate_change/techguide/Denef_et_al_2011_Review_of_revie
ws_v1.0.pdf.
Dickie, A., Streck, C., Roe, S., Zurek, M., Haupt, F., Dolginow, A., 2014. Strategies for
Mitigating Climate Change in Agriculture: Abridged Report. Climate Focus and
California Environmental Associates, prifadred with the support of the Climate and Land
Use Alliance [WWW Document]. URL www.agriculturalmitigation.org
Dourmad, J., Rigolot, C., van der Werf, H., 2008. Emission of Greenhouse Gas: Developing
management and animal farming systems to assist mitigation, in: Rowlinson, P., Steel,
M., Nefzaoui, A., Livestock and Global Climate Change conference proceeding,
Cambridge University Press, Tunisia, pp. 36-39.
DSI MSU, 2015. Decision Support and Informatics. Michigan State University. URL
http://dsiweb.cse.msu.edu/
Dyer, J.A., VergĂŠ, X.P.C., Desjardins, R.L., McConkey, B.G., 2014. Assessment of the Carbon
and Non-Carbon Footprint Interactions of Livestock Production in Eastern and Western
Canada. Agroecol. Sustain. Food Syst. 38, 541â572.
EC (European Commission), 2007. Carbon Footprint: What It Is and How to Measure It [WWW
Document]. URL lca.jrc.ec.europa.eu/Carbon_footprint.pdf
EPA (U.S. Environmental Protection Agency), 1999. Livestock Manure Management [WWW
Document]. URL http://www.epa.gov/methane/reports/05-manure.pdf

160

EPA(U.S. Environmental Protection Agency), 2013. Global mitigation of non-COâ greenhouse
gases [WWW Document]. URL
http://www.epa.gov/climatechange/Downloads/EPAactivities/MAC_Report_2013.pdf
EPA (U.S. Environmental Protection Agency), 2015. Ag 101: Dairy Production [WWW
Document]. URL https://www.epa.gov/sites/production/files/201507/documents/ag_101_agriculture_us_epa_0.pdf
EPA (US Environmental Protection Agency), 2016. Climate change: basic information. In: Clim.
Chang. Div. URL https://www3.epa.gov/climatechange/basics/
Esminger, M.E., Oldfield, J.E., Heinemann, W.W., 1990. Feeds and nutrition: formerly Feeds &
nutrition, complete, second ed., Ensminger Publishing Company, Clovis, CA.
FAO (Food and Agriculture Organisation of the United Nations), 2010a. Climate-Smart
Agriculture: Policies, practices and financing for food security, adaptation and mitigation.
Food and Agriculture Organization of the United Nations, Rome.
FAO (Food and Agriculture Organization of the United Nations), 1986. Farm structures in
tropical climates: Animal environmental requirements [WWW Document]. FAO, Rome.
URL http://www.fao.org/docrep/s1250e/s1250e10.htm
FAO (Food and Agriculture Organization of the United Nations), 2007. The State of the World's
Animal Genetic Resources for Food and Agriculture: in brief, edited by Barbara
Rischkowsky & Dafydd Pilling, Rome.
FAO (Food and Agriculture Organization of the United Nations), 2009a. Global agriculture
towards 2050. High Level Expert Forum Issues Paper. FAO, Rome.
FAO (Food and Agriculture Organization of the United Nations), 2009b. The State of Food and
Agriculture: Livestock in the Balance[WWW Document]. FAO, Rome. URL
http://www.fao.org/docrep/012/i0680e/i0680e.pdf
FAO (Food and Agriculture Organization of the United Nations), 2011. World livestock 2011:
Livestock in food security[WWW Document]. FAO, Rome. URL
http://www.fao.org/docrep/014/i2373e/i2373e.pdf
FAO (Food and Agriculture Organization of the United Nations), 2013. Climate-Smart
Agriculture: Sourcebook [WWW Document]. FAO, Rome. URL http://www.fao.org/3/ai3325e.pdf.
FAO (Food and Agriculture Organization of the United Nations), 2016a. Milk production. In:
Dairy Prod [WWW Document]. FAO, Rome. URL http://www.fao.org/agriculture/dairygateway/milk-production/en/#.VtCK0_krKUk.
FAO (Food and Agriculture Organization of the United Nations), IFAD (International Fund for
Agricultural Development), WFP (World Food Programme), 2013. The State of Food
Insecurity in the World: The multiple dimensions of food security. FAO, Rome.
161

FAO (Food and Agriculture Organisation of the United Nations), 2008. The Global Livestock
Sector â a Growth Engine - NR fact sheet. Rome.
FAO (Food and Agriculture Organisation of the United Nations), 2010b. Greenhouse Gas
Emissions from the Dairy Sector A Life Cycle Assessment. Animal Production and
Health Division, Rome.
FAO (Food and Agriculture Organisation of the United Nations), 2014. Livestock and Animal
Production [WWW Document]. FAOâs Anim. Prod. Heal. Div. URL
http://www.fao.org/ag/againfo/themes/en/animal_production.html
FAO (Food and Agriculture Organisation of the United Nations), 2016b. FAOstat, Statistical
databases-Livestock Primary. Food and Agriculture Organization of the United Nations,
Rome.
FAO (Food and Agriculture Organisation of the United Nations), 2017. Animal production
[WWW Document]. Food Agric. Organ. United Nations. URL
http://www.fao.org/animal-production/en/
Finkbeiner, M., 2009. Carbon footprintingâopportunities and threats. Int J Life Cycle Assess
14, 91â94.
Finocchiaro, R., van Kaam, J., Portolano, B., Misztal, I., 2005. Effect of heat stress on
production of Mediterranean dairy sheep. J. Dairy Sci. 88, 1855-1864.
Fisher, R.A., 1935. The design of experiments. Oliver and Boyd, Edinburgh.
Flanders, F.B., Gillespie, J.R., 2015. Modern Livestock & Poultry Production, 9th ed. Cengage
Learning.
FlysjĂś, A., Henriksson, M., Cederberg, C., Ledgard, S., Englund, J.-E., 2011. The impact of
various parameters on the carbon footprint of milk production in New Zealand and
Sweden. Agric. Syst. 104, 459â469.
Fregley, M. J., 1996. Adaptations: some general characteristics, in: Fregley, M. J., Blatteis, C.
M., Handbook of physiology, Section 4: Environmental physiology, Oxford University
Press, pp. 3-15.
Galli, A., Wiedmann, T., Ercin, E., Knoblauch, D., Ewing, B., Giljum, S., 2011. Integrating
Ecological, Carbon and Water Footprint: defining the âFootprint Familyâ and its
Application in Tracking Human Pressure on the Planet. Technical Document, Surrey,
UK.
Galli, A., Wiedmann, T., Ercin, E., Knoblauch, D., Ewing, B., Giljum, S., 2012. Integrating
ecological, carbon and water footprint into a âFootprint Familyâ of indicators: definition
and role in tracking human pressure on the planet. Ecological Indicators 16, 100e112.
Garnett, T., 2009. Livestock-related greenhouse gas emissions: impacts and options for policy
162

makers. Environ. Sci. Policy 12, 491â503.
Gauly, M., Bollwein, H., Breves, G., BrĂź Gemann, K., DĂ¤ Nicke, S., Das-, G., Demeler, J.,
Hansen, H., Isselstein, J., KĂś Nig, S., LohĂś Lter, M., Martinsohn, M., Meyer, U.,
Potthoff, M., Sanker, C., SchrĂś Der, B., Wrage, N., Meibaum, B., Von SamsonHimmelstjerna, G., Stinshoff, H., Wrenzycki, C., 2013. Future consequences and
challenges for dairy cow production systems arising from climate change in Central
Europe â a review. Anim. Anim. Consort. 7, 843â859.
Gerber, P.J., Carsjens, G.J., Pak-uthai, T., Robinson, T.P., 2008. Decision support for spatially
targeted livestock policies: diverse examples from Uganda and Thailand. Agric. Syst. 96,
37-51.
Gerber, P.J., Steinfeld, H., Henderson, B., Mottet, A., Opio, C., Dijkman, J., Falcucci, A.,
Tempio, G., 2013. Tackling climate change through livestock: A global assessment of
emissions and mitigation opportunities. FAO, Rome.
Gold, M. V., 2007. Sustainable Agriculture: Definitions and Terms [WWW Document]. USDA
Spec. Ref. briefsâŻ; 99-02. URL https://www.nal.usda.gov/afsic/sustainable-agriculturedefinitions-and-terms#top
Goldewijk, K., Battjes, J.J., 1997. A hundred year (1980-1990) database for integrated
environmental assessments. National Institute of Public Health and the Environment,
Bilthoven, The Netherlands.
Graux, A.-I., Bellocchi, G., Lardy, R., Soussana, J.-F., 2013. Ensemble modelling of climate
change risks and opportunities for managed grasslands in France. Agric. For. Meteorol.
170, 114â131.
Griffin, T.S., 2004. Forage Facts: Growing Forage Legumes in Maine [WWW Document].
Coop. Ext. Publ. Univ. Maine. URL https://extension.umaine.edu/publications/2261e/
Guttman, N.B., Quayle, R.G., 1996. A Historical Perspective of U.S. Climate Divisions. Bull.
Am. Meteorol. Soc. 77, 293â303.
Hales, K.E., Cole, N.A., MacDonald, J.C., 2012. Effects of corn processing method and dietary
inclusion of wet distillers grain with solubles on energy metabolism and enteric methane
emissions of finishing cattle. J. Anim. Sci. 90(9), 3174-3185.
Hansen, P.J., 2007. Exploitation of genetic and physiological determinants of embryonic
resistance to elevated temperature to improve embryonic survival in dairy cattle during
heat stress. Theriogenology. 68(4), S242-S249.
Harmel, R.D., Richardson, C.W., King, K.W., 2000. Hydrologic response of a small watershed
model to generated precipitation. Trans. ASAE 43, 1483â1488.

163

Harvell, C.D., Mitchell, C.E., Ward, J.R., Altizer, S., Dobson, A.P., Ostfeld, R.S., Samuel, M.D.,
2002. Climate warming and disease risks for terrestrial and marine biota. Sci. 296:21582162.
Hatfield, J., Takle, G., Grotjahn, R., Holden, P., Izaurralde, R.C., Mader, T., Marshall, E., and
Liverman, D., 2014. Ch. 6: Agriculture. Climate Change Impacts in the United States:
The Third National Climate Assessment, J. M. Melillo, Terese (T.C.) Richmond, and G.
W. Yohe, Eds., U.S. Global Change Research Program, 150-174.
Hatfield, J.L., Boote, K., Fay, P., Hahn, L., Izaurralde, C., Kimball, B.A., Mader, T., Morgan, J.,
Ort, D., Polley, W., Thomson, A., Wolfe, D., 2008. Agriculture, in: Walsh, M., The
effects of climate change on agriculture, land resources, water resources, and biodiversity
in the United States. A Report by the U.S. Climate Change Science Program and the
Subcommittee on Global Change Research. Washington, DC., USA, pp. 362.
Hatfield, J.L., Prueger, J.H., 2011. Agroecology: implications for plant response to climate
change, in: Yadav, S.S., Redden, R.J., Hatfield, J.L., Lotze-Campen, H., Hall, A.E., Crop
adaptation to climate change, Wiley-Blackwell, Chichester, UK, pp. 27-43.
Haun, G.L., 1997. Dynamic responses of cattle to thermal heat loads. J. Anim. Sci., 77, 10-20.
HavlĂ­k, P., Valin, H., Herrero, M., Obersteiner, M., Schmid, E., Rufino, M.C., Mosnier, A.,
Thornton, P.K., BĂśttcher, H., Conant, R.T., Frank, S., Fritz, S., Fuss, S., Kraxner, F.,
Notenbaert, A., 2014. Climate change mitigation through livestock system transitions.
Proc. Natl. Acad. Sci. U. S. A. 111, 3709â14.
Heckman, E., Hinds, S., Johnson, K., Perkins, J., Shelton, V., Zupancic, R., 2007. Managementintensive grazing in Indiana. USDA-NRCS Extension bulletin AY-328, Indiana.
Henderson, B.B, Gerber, P.J., Hilinski, T.E., Falcucci, A., Ojima, D.S., Salvatore, M., Conant,
R.T., 2015. Greenhouse gas mitigation potential of the worldâs grazing lands: Modeling
soil carbon and nitrogen fluxes of mitigation practices. Agriculture, Ecosystems and
Environment, 207: 91â100.
Henry, B., Charmley, E., Eckard, R., Gaughan, J.B., Hegarty, R., 2012. Livestock production in
a changing climate: adaptation and mitigation research in Australia. Crop and Pasture
Science, 63: 191-202.
Hermann, B.G., Debeer, L., De Wilde, B., Blok, K., Patel, M.K., 2011. To compost or not to
compost: Carbon and energy footprints of biodegradable materialsâ waste treatment.
Polym. Degrad. Stab. 96, 1159â1171.
Herrero, M., Thornton, P.K., Kruska, R., Reid, R.S., 2008. Systems dynamics and the spatial
distribution of methane emissions from African domestic ruminants to 2030. Agr.
Ecosyst. Environ. 126, 122-137.
Herrero, M., Thornton, P.K., Notenbaert, A., Msangi, S., Wood, S., Kruska, R., Dixon, J.,
Bossio, D., van de Steeg, J., Ade Freeman, H., Li, X., ParthasarathyRao, P., 2012.
164

Drivers of change in cropâlivestock systems and their potential impacts on agroecosystems services and human wellbeing to 2030: A study commissioned by the CGIAR
Systemwide Livestock Programme. International Livestock Research Institute, Nairobi,
Kenya.
Herrero, M., Thornton, P.K., Notenbaert, A.M., Wood, S., Msangi, S., Freeman, H. A., Bossio,
D., Dixon, J., Peters, M., van de Steeg, J., Lynam, J., ParthasarathyRao, P., Macmillan S.,
Gerard, B., McDermott, J., Sere, C., Rosegrant, M., 2010. Smart Investments in
Sustainable Food Production: Revisiting Mixed Crop-Livestock Systems. Sci. 327, 822825.
Hess, H.D., Tiemann, T.T., Noto, F., Carulla, J.E., Kreuzer, M., 2006. Strategic use of tannins as
means to limit methane emission from ruminant livestock. International Conference on
Greenhouse Gases and Animal Agriculture, Elsevier International Congress Series,
Zurich, Switzerland, pp. 1293, 164-167.
Hoekstra, A.Y., 2008. Water Neutral: Reducing and Offsetting the Impacts of Water Footprints.
UNESCO-IHE, Delft, the Netherlands. Value of Water Research Report Series No. 28.
HĂśglind M, Thorsen SM, Semenov MA (2013) Assessing uncertainties in impact of climate
change on grass production in northern Europe using ensembles of global climate
models. Agric For Meteorol 170:103â113.
Holland, E.A., Parton, W.J., Detling, J.K., Coppock, D.L., 1992. Physiological responses of plant
populations to herbivory and their consequences for ecosystem nutrient flow. Am. Nat.
140(4), 85â706.
Howden, S.M., Crimp, S.J., Stokes, C.J., 2008. Climate change and Australian livestock systems:
impacts, research and policy issues. Aust. J. Exp. Agr. 48, 780-788.
Hristov, A.N., Oh, J., Firkins, J.L., Dijksta, J., Kebreab, E., Waghorn, G., Makkar, H.P.S.,
Adesogan, A.T., Yang, W., Lee, C., Gerber, P.J., Henderson, B., Tricarico, J.M., 2013.
Mitigation of methane and nitrous oxide emissions from animal operations: I. A review
of enteric methane mitigation options. American Society of Animal Science, Vol 91.
Huang, J., Xu, C.-C., Ridoutt, B.G., Liu, J.-J., Zhang, H.-L., Chen, F., Li, Y., 2014. Water
availability footprint of milk and milk products from large-scale dairy production systems
in Northeast China. J. Clean. Prod. 79, 91â97.
Hurst, P., Termine, P., Karl, M., 2005. Agricultural Workers and their Contribution to
Sustainable Agriculture and Rural Development. FAO, Rome.
ftp://ftp.fao.org/docrep/fao/008/af164e/af164e00.pdf
ICF International, 2013. Greenhouse Gas Mitigation Options and Costs for Agricultural Land
and Animal Production within the United States[WWW Document]. URL
http://www.usda.gov/oce/climate_change/mitigation_technologies/GHG_Mitigation_Opti
ons.pdf
165

IFAD (International Fund for Agricultural Development), 2010. Livestock and climate change
[WWW Document]. URL http://www.ifad.org/lrkm/events/cops/papers/climate.pdf
IPCC (Intergovermental Panel on Climate Change), 1997. IPCC/OECD/IEA Programme on
National Greenhouse Gas Inventories [WWW Document]. URL Intergovernmental panel
on climate change. http://www.ipcc-nggip.iges.or.jp/public/mtdocs/pdfiles/rockhamp.pdf.
IPCC (Intergovermental Panel on Climate Change), 2000. Land use, land use change, and
forestry. Cambridge University Press, Cambridge.
IPCC (Intergovermental Panel on Climate Change), 2001. Climate change 2001: The scientific
basis, in: Houghton, J.T., Ding, Y., Griggs, D.J., Noguer, M., van der Linden, P.J., Dai,
X., Maskell, K., and Johnson, C.A. (Eds.), Contribution of Working Group I to the third
assessment report of the Intergovernmental Panel on Climate Change, Cambridge
University Press, Cambridge, United Kingdom and New York, NY, USA.
IPCC (Intergovermental Panel on Climate Change), 2006. 2006 IPCC Guidelines for National
Greenhouse Gas Inventories, Prepared by the National Greenhouse Gas Inventories
Programme, Eggleston H.S., Buendia L., Miwa K., Ngara T. and Tanabe K. (eds).
Published: IGES, Japan.
IPCC (Intergovermental Panel on Climate Change), 2007. Climate Change 2007: Synthesis
Report, in: Pachauri, R.K and Reisinger, A. (Eds.), Contribution of Working Groups I, II
and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change. IPCC, Geneva, Switzerland, 104 pp.
IPCC (Intergovermental Panel on Climate Change), 2013. Climate Change 2013: The Physical
Science Basis, in: Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K.,
Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M. (Eds.), Contribution of
Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change, Cambridge University Press, Cambridge, United Kingdom and New
York, NY, USA, 1535 pp.
IPCC (Intergovermental Panel on Climate Change), 2014. Climate Change 2014: Impacts,
Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects, in: Field, C.B.,
Barros, V.R., Dokken, D.J., Mach, K.J., Mastrandrea, M.D., Bilir, T.E., Chatterjee, M.,
Ebi, K.L., Estrada, Y.O., Genova, R.C., Girma, B., Kissel, E.S., Levy, A.N.,
MacCracken, S., Mastrandrea, P.R., White, L.L. (Eds.), Contribution of Working Group
II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change,
Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA,
1132 pp.
Jain, A.K., 2010. Data clustering: 50 years beyond K-means, Pattern Recognition Letters.
Jeong, J., Williams, J.R., Rossi, C.G., Taylor, R.A., Wang, X., Fox, W.E., 2015. Development of
the Spatial Rainfall Generator (SRGEN) for the Agricultural Policy/Environmental
Extender Model. JAWRA J. Am. Water Resour. Assoc. 51, 154â167.
166

Johnson, H.D., 1980. Depressed chemical thermogenesis and hormonal function in heat, in:
Yousef, M.K. (Ed.), Environmental Physiology:Ageing, Heat and Altitude, Elsevier
North Holland Inc, New York, USA, pp. 3-9.
Jones, A.K., Jones, D.L., Edwards-Jones, G., Cross, P., 2013. Informing decision making in
agricultural greenhouse gas mitigation policy: A BestâWorst Scaling survey of expert
and farmer opinion in the sheep industry. Environ. Sci. & Policy. 29, 46-56.
Jose, S., 2009. Agroforestry for ecosystem services and environmental benefits: An overview.
Agroforestry Syst. 76, 1â10.
Jungbluth, T., Hartung, E., Brose, G., 2001. Greenhouse gas emissions from animal houses and
manure stores. Nutr. Cycl. Agroecosyst, 60: 133â145.
Kadzere, C.T., Murphy, M.R., Silanikove, N., Maltz, E., 2002. Heat stress in lactating dairy
cows: a review. Liv. Prod. Sci. 77, 59-91.
Kalaugher, E., Bornman, J.F., Clark, A., Beukes, P., 2013. An integrated biophysical and socioeconomic framework for analysis of climate change adaptation strategies: The case of a
New Zealand dairy farming system. Environ. Model. Softw. 39, 176â187.
Karaca, A.G., Parker, H.M., Yeatman, J.B., McDaniel, C.D., 2002. Role of seminal plasma in
heat stress infertility of broiler breeder males. Poult. Sci. 81, 1904-1909.
Karl, T.R., Melillo, J.M., Peterson, T.C., 2009. Global climate change impacts in the United
States. U.S. Global Change Research Programme. Cambridge University Press.
King, J.M., Parsons, D.J., Turnpenny, J.R., Nyangaga, J., Bakari, P., Wathes, C.M., 2006.
Modellig energy metabolism of Friesians in Kenya smallholdings shows how heat stress
and energy deficit constrain milk yield and cow replacement rate. Anim. Sci. 82, 705716.
Kriegl, T., McNair, R., 2005. Pastures of plenty: financial performance of Wisconsin grazing
dairy farms. University of Wisconsin-Madison.
Kunavongkrita, A., Suriyasomboonb, A., Lundeheimc, N., Learda, T.W., Einarsson, S., 2005.
Management and sperm production of boars under differing environmental conditions.
Theriogenology. 63, 657-667.
Kurukulasuriya, P., Rosenthal, S., 2003. Climate change and agriculture: a review of impacts and
adaptations. Climate Change Series Paper No. 91, World Bank, Washington DC.
Lacefield, G., 2013. Red and White Clover. University of Kentucky Cooperative Extension
Service, Kentucky.
Lacetera, N., Bernabucci, U., Ronchi, B., Nardone, A., 1996. Body condition score, metabolic
status and milk production of early lactating dairy cows exposed to warm environment.
Riv. Agr. Subtrop. Trop. 90, 43â55.
167

Lacetera, N., Bernabucci, U., Ronchi, B., Nardone, A., 2003. Physiological and productive
consequences of heat stress: The case of dairy ruminants. Proc. of the Symposium on
Interaction between Climate and Animal Production: EAAP Technical Serie , pp. 7, 45â
60.
Lake, I.R., Hooper, L., Abdelhamid, A., Bentham, G., Boxall, A.B., Draper, A., FairweatherTait, S., Hulme, M., Hunter, P.R., Nichols, G., Waldron, K.W., 2012. Climate Change
and Food Security: Health Impacts in Developed Countries. Environ. Health Persp. 120,
1520-1526.
LEAD (Livestock Environment and Development), 2014. Livestock role in deforestation [WWW
Document]. Food and Agriculture Organization of the United Nations. URL
http://www.fao.org/agriculture/lead/themes0/deforestation/en/
Lucas, E.M., Randall, J.M., Meneses, J.F., 2000. Potential for evaporative cooling during heat
stress periods in pig production in Portugal. J. Agr. Eng. Res. 76, 363-371.
Luo, Y., Zhang, M., 2009. Management-oriented sensitivity analysis for pesticide transport in
watershed-scale water quality modeling using SWAT. Environ. Pollut. 157, 3370â3378.
MacFarlane, A.W., 1990. Field experience with new pasture cultivars in Canterbury. Proc. New
Zeal. Grassl. Assoc. 52, 139â143.
Mader, T.L., 2003. Environmental stress in confined beef cattle. J. Anim. Sci. 81, 110-119.
Mader, T.L., Davis, M.S., 2004. Effect of management strategies on reducing heat stress of
feedlot cattle: Feed and water intake. J. Anim. Sci. 82, 3077-3087.
Martin, C., Morgavi, D.P., Doreau, M., 2010. Methane mitigation in ruminants: from microbe to
the farm scale. Anim. 4(3):351â365.
Martin, P., 2002. Guest Workers: New Solution, New Problem? A study conducted for the Pew
Hispanic Center [WWW Document], Washington DC. URL
http://www.pewhispanic.org/files/reports/7.pdf
Mashaly, M.M., Hendricks 3rd, G.L., Kalama, M.A., Gehad, A.E., Abbas, A.O., Patterson, P.H.,
2004. Effect of heat stress on production parameters and immune responses of
commercial laying hens. Poult. Sci. 83, 889-894.
Mathevon, M., Buhr, M.M., Dekkers, J.M., 1998. Environmental, management, and genetic
factors affecting semen production in Holstein bulls. J. Dairy Sci. 81, 3321-3330.
Mc Geough, E.J., Little, S.M., Janzen, H.H., McAllister, T.A., McGinn, S.M., Beauchemin,
K.A., 2012. Life-cycle assessment of greenhouse gas emissions from dairy production in
Eastern Canada: A case study. J. Dairy Sci. 95, 5164â5175.
McDowell, R.E., 1968. Climate versus man and his animals. Nature. 218, 641-645.

168

MEA (Millenium Ecosystem Assessment), 2005. Ecosystems and human well-being:
biodiversity synthesis [WWW Document]. World Resources Institute, Washington, DC.
URL http://www.unep.org/maweb/documents/document.354.aspx.pdf
Melillo, J.M., Richmond, T., Yohe, G.W., 2014. Climate Change Impacts in the United States
Climate Change Impacts in the United States: The Third National Climate Assessment.
U.S. Global Change Research Program.
Mitloehner, F.M., Morrow, J.L., Dailey, J.W., Wilson, S.C., Galyean, M.L., Miller, M.F.,
McGlone, J.J., 2001. Shade and water misting effects on behaviour, physiology,
performance, and carcass traits of heat-stressed fedlot cattle. J. Anim. Sci. 79, 2327-2335.
Monteny, G.J., Groenestein, C.M., Hilhorst, M.A., 2001. Interactions and coupling between
emissions of methane and nitrous oxide from animal husbandry. Nutr. Cycl. Agroecosys.
60, 123â132.
Moore, A.D., Ghahramani, A., 2013. Climate change and broadacre livestock production across
southern Australia. 1. Impacts of climate change on pasture and livestock productivity,
and on sustainable levels of profitability. Glob. Chang. Biol. 19, 1440â1455.
Mosier, A., Kroeze, C., Nevison, C., Oenema, O., Seitzinger, S., Van Cleemput, O., 1998.
Closing the global N2O budget: Nitrous oxide emissions through the agricultural nitrogen
cycle: OEDC/IPCC/IEA phase II development of IPCC guideline for national greenhouse
gas methodology. Nutr. Cycl. Agroecosys. 52, 225â248.
Mosier, A., Wassmann, R., Verchot, L., King, J., Palm, C., 2004. Methane and nitrogen oxide
fluxes in tropical agricultural soils: sources, sinks and mechanisms. Environ. Dev. &
Sustainability. 6, 11-49.
Muller, L.D., 2004. Managing to get more milk and profit from pasture. PennState Extension,
Pennsylvania State University, Pennsylvania.
Murphy, E., De Boer, I.J.M., Van Middelaar, C.E., Holden, N.M., Shalloo, L., Curran, T.P.,
Upton, J., 2017. Water footprinting of dairy farming in Ireland. J. Clean. Prod. 140, 547â
555.
Nardone, A., 2000. Weather conditions and genetics of breeding systems in the Mediterranean
area. XXXX International Sysmposium of Societa Italiana per il Progresso della
Zootecnia, Ragusa, Italy, pp. 67-92.
Nardone, A., Ronchi, B., Lacetera, N., Ranieri, M.S., Bernabucci, U., 2010. Effects of climate
change on animal production and sustainability of livestock systems. Livest. Sci. 130, 5769.
Nejadhashemi, A.P., Wardynski, B.J., Munoz, J.D., 2012. Large-Scale Hydrologic Modeling of
the Michigan and Wisconsin Agricultural Regions to Study Impacts of Land Use
Changes. Trans. ASABE 55, 821â838.

169

Nott, S., 2003. Evolution of dairy grazing in the 1990âs. Department of Agricultural Economics
Staff Paper No. 2003-07, Michigan State University, East Lansing, Michigan.
Novero, R.P., Beck, M.M., Gleaves, E.W., Johnson, A.L., Deshazer, J.A., 1991. Plasma
progesterone, luteinizing hormone concentrations, and granulosa cell responsiveness in
heat-stressed hens. Poult. Sci. 70, 2335-2339.
OâBrien, D., Capper, J.L., Garnsworthy, P.C., Grainger, C., Shalloo, L., 2014. A case study of
the carbon footprint of milk from high-performing confinement and grass-based dairy
farms. J. Dairy Sci. 97.
OECD (Organisation for Economic Co-operation and Development), 2004. Measuring
Sustainable Development: Integrated Economic, Environmental and Social Frameworks.
Paris, France.
Ogle, D., St. John, L., 2008. Plant Guide for white clover (Trifolium repens L.). Aberdeen, Idaho
83210.
Oliver, D.M., Fish, R.D., Winter, M., Hodgson, C.J., Heathwaite, A.L., Chadwick, D.R., 2012.
Valuing local knowledge as a source of expert data: farmer engagement and the design of
decision support systems. Environ. Modell. Softw. 36, 76â85.
Olivier, J. G., Janssens-Maenhout, G., 2012. Part III: Greenhouse gas emissions, in: International
Energy Agency (IEA), CO2 Emissions from Fuel Combustion, 2012 Ed, Paris, France,
pp. III.1âIII.51.
Olsson, K., Dahlborn, K., 1989. Fluid balance during heat stress in lactating goats. Q. J. Exp.
Physiol. 74, 645-659.
O'Mara, F. P., 2012. The role of grasslands in food security and climate change. Ann. BotLondon. 110, 1263â 1270.
Opricovic, S., 2009. A Compromise Solution in Water Resources Planning. Water Resour.
Manag. 23, 1549â1561.
Opricovic, S., 2011. Fuzzy VIKOR with an application to water resources planning. Expert Syst.
Appl. 38, 12983â12990.
Opricovic, S., Tzeng, G.-H., 2004. Compromise solution by MCDM methods: A comparative
analysis of VIKOR and TOPSIS. Eur. J. Oper. Res. 156, 445â455.
Opricovic, S., Tzeng, G.-H., 2007. Extended VIKOR method in comparison with outranking
methods. Eur. J. Oper. Res. 178, 514â529.
OWEB (Oregon Watershed Enhancement Board), 1999. Oregon Watershed Assessment Manual.
Oregon.
Owusu-Sekyere, E., Scheepers, M., Jordaan, H., 2016. Water Footprint of Milk Produced and
170

Processed in South Africa: Implications for Policy-Makers and Stakeholders along the
Dairy Value Chain. Water 8, 322.
Pagani, M., Vittuari, M., Johnson, T.G., De Menna, F., 2016. An assessment of the energy
footprint of dairy farms in Missouri and Emilia-Romagna. Agric. Syst. 145, 116â126.
Palhares, J.C.P., Pezzopane, J.R.M., 2015. Water footprint accounting and scarcity indicators of
conventional and organic dairy production systems. J. Clean. Prod. 93, 299â307.
Patz, J.A., Graczyk, T.K., Geller, N., Vittor, A.Y., 2000. Effects of environmental change on
emerging parasitic diseases. Int. J. Parasitol. 30, 1395-1405.
Paustian, K., Andren, O., Janzen, H.H., Lal, R., Smith, P., Tian, G., Tiessen, H., Van Noordwijk,
M., Woomer, P. L., 1997. Agricultural soils as a sink to mitigate carbon dioxide
emissions. Soil Use Manage. 13(4), 230-244.
PCMDI (Program for Climate Model Diagnosis and Intercomparison), 2015. CMIP5 Coupled
Model Intercomparison Project. In: WCRP (World Clim. Res. Program). URL
http://cmip-pcmdi.llnl.gov/cmip5/availability.html.
Perry, B., Sones, K., 2009. Global livestock disease dynamics over the last quarter century:
drivers, impacts and implications. FAO, Rome.
Polley, H.W., Briske, D.D., Morgan, J.A., Wolter, K., Bailey, D.W., Brown, J.R., 2013. Climate
change and North American rangelands: Trends, projections, and implications.
Rangeland Ecol. Manage. 66, 493-511.
Randolph, S.E., 2008. Dynamics of tick-borne disease systems: minor role of recent climate
change. Rev. Sci. Tech. Oie. 27, 367-381.
Rayburn, E.B., 2007. Forage utilization for pasture-based livestock production. Natural
Resource, Agriculture, and Engineering Service (NRAES), New York.
Rayburn, E.B., 2008. Animal production systems for pasture-based livestock production. Natural
Resource, Agriculture, and Engineering Service (NRAES), New York.
Reynolds, C., Crompton, L., Mills, J., 2010. Livestock and climate change impacts in the
developing world. Outlook Agr. 39, 245-248.
Rojas-Downing, M.M., 2013. Resource Use and Conservation and Environmental Impacts in the
Transition from Confinement to Pasture-Based Dairies. Michigan State University, East
Lansing, Michigan, Thesis.
Rojas-Downing, M.M., Harrigan, T., Nejadhashemi, A.P., 2017a. Resource use and economic
impacts in the transition from small confinement to pasture-based dairies. Agric. Syst.
153, 157â171.
Rojas-Downing, M.M., Nejadhashemi, A.P., Abouali, M., Daneshvar, F., Al Masraf, S.A.D.,
171

Herman, M.R., Harrigan, T., Zhang, Z., 2017b. Pasture diversification to combat climate
change impacts on grazing dairy production. Mitig. Adapt. Strateg. Glob. Chang. 1â27.
Rojas-Downing, M.M., Nejadhashemi, A.P., Harrigan, T., Woznicki, S.A., 2017c. Climate
change and livestock: Impacts, adaptation, and mitigation. Clim. Risk Manag. 16: 145163.
Ronchi, B., Bernabucci, U., Lacetera, N., Verini Supplizi, A., Nardone, A., 1999. Distinct and
common effects of heat stress and restricted feeding on metabolic status in Holstein
heifers. Zootec. Nutr. Anim. 25, 71-80.
Ronchi, B., Stradaioli, G., Verini Supplizi, A., Bernabucci, U., Lacetera, N., Accorsi, P. A.,
Nardone, A., Seren, E., 2001. Influence of heat stress and feed restriction on plasma
progesterone, estradiol-17Î˛ LH, FSH, prolactin and cortisol in Holstein heifers. Livest.
Prod. Sci. 68, 231-241.
Rosegrant, M. W., Cai, X., Cline, S. A., 2002. Global water outlook to 2025: Averting and
impending crisis. International Water Management Institute (IWMI), 2020 Vision for
Food, Agriculture, and the Environment, International Food Policy Research Institute
(IFPRI). Washington, D.C, Colombo, Sri Lanka.
Rosegrant, M.W., Fernandez, M., Sinha, A., 2009. Looking into the future for agriculture and
AKST, in: International Assessment of Agricultural Knowledge, Science and Technology
for Development (IAASTD), McIntyre, B.D., Herren, H.R., Wakhungu, J., Watson, R.T.
(Eds.), Agriculture at a crossroads, Island Press, Washington, DC, pp. 307-376.
Rotter, R., van de Geijn, S.C., 1999. Climate change effects on plant growth, crop yield and
livestock. Climatic Change. 43, 651-681.
Rotz, C.A., 2004. The Integrated Farm System Model: A Tool for Developing more
Economically and Environmentally Sustainable Farming Systems for the Northeast, in:
NABEC. 2004 Northeast Agricultural and Biological Engineering Conference Sponsored
by ASAE. ASABE, St. Joseph, Mich., pp. 4â22.
Rotz, C.A., Corson, M.S., Chianese, D.S., Montes, F., Hafner, S.D., Bonifacio, H.F., Coiner,
C.U., 2014. The Integrated Farm System Model â Reference Manual - Version 4.2.
Pasture Systems and Watershed Management Research Unit, Agricultural Research
Service, United States Department of Agriculture.
Rowlinson, P., 2008. Adapting livestock production systems to climate change: temperate zones,
in: Rowlinson, P., Steel, M., Nefzaoui, A., Livestock and Global Climate Change
conference proceeding, Cambridge University Press, Tunisia, pp. 61-63.
Rowlinson, P., Steele, M., Nefzaoui, A., 2008. Livestock and Global Climate Change:
Adaptation I and II, in: Rowlinson, P., Steel, M., Nefzaoui, A., Livestock and Global
Climate Change conference proceeding, Cambridge University Press, Tunisia, pp. 56-85.

172

Sainz, R., 2003. Framework for calculating fossil fuel use in livestock systems. Environment and
Development Initiative Report.
ftp://ftp.fao.org/docrep/nonfao/lead/x6100e/X6100E00.pdf
Sanz-Saez, A., Erice, G., Aguirreolea, J., MuĂąoz, F., Sanchez-Diaz, M., Irigoyen, J.J., 2012.
Alfalfa forage digestibility, quality and yield under future climate change scenarios vary
with Sinorhizobium meliloti strain. Plant Physiol. 169, 782-788.
Sayadi, M.K., Heydari, M., Shahanaghi, K., 2009. Extension of VIKOR method for decision
making problem with interval numbers. Appl. Math. Model. 33, 2257â2262.
Schuppli, C.A., von Keyserlingk, M.A.G., Weary, D.M., 2014. Access to pasture for dairy cows:
Responses from an online engagement. J. Anim. Sci. 92, 5185â5192.
Sharpley, A.N., Williams, J.R., 1990. EPIC-erosion/productivity impact calculator: 1. Model
documentation, USDA-ARS Technical Bulletin 1768.
Shrestha, S., Abdalla, M., Hennessy, T., Forristal, D., Jones, M.B., 2015. Irish farms under
climate change â is there a regional variation on farm responses? J. Agric. Sci. 153, 385â
398.
Sirohi, S., Michaelowa, A., 2007. Sufferer and cause: Indian livestock and climate change.
Climatic Change. 85, 285-298.
Smil, V., 2001. Enriching the Earth: Fritz Haber, Carl Bosch, and the Transformation of World
Food Production. MIT Press, Cambridge.
Smith, J.P., Pearce, B.D., Wolfe, M.S., 2012. Reconciling productivity with protection of the
environment: Is temperate agroforestry the answer? . Renew. Agr. Food Syst. 28(1), 80â
92.
Sobhani, R., Abahusayn, M., Gabelich, C.J., Rosso, D., 2012. Energy Footprint analysis of
brackish groundwater desalination with zero liquid discharge in inland areas of the
Arabian Peninsula. Desalination 291, 106â116.
Stehfest, E., Bouwman, L., van Vuuren, D.P., den Elzen, M., Eickhout, B., Kabat, P., 2009.
Climate benefits of changing diet. Climatic Change, 95(1-2), 83-102.
Steinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M., Haan, C., 2006. Livestock's
long shadow: environmental issues and options. FAO, Rome.
St-Pierre, N.R., Cobanov, B., Schnitkey, G., 2003. Economic losses from heat stress by U.S.
livestock industries. J. Dairy. Sci. 86, E52-E77.
Sullivan, K., DeClue, R., Emmick, D., 2000. Prescribed grazing and feeding management for
lactating dairy cows. USDA-NRCS, Syracuse.
Sundquist, B., 1993. The global carbon dioxide budget. Sci. 259, 934-941.
173

Swingland, I.A., 2001. Biodiversity, definition of. Encyclopedia of Biodiversity, Volume 1: 377391.
Tankson, J.D., Vizzier-Thaxton, Y., Thaxton, J.P., May, J.D., Cameron , J.A., 2001. Stress and
nutritional quality of broilers. Poult. Sci. 80, 1384-1389.
Tennigkeit, T., Wilkies, A., 2008. Carbon finance in Rangelands [WWW Document]. World
Agroforestry Centre (ICRAF). URL
http://www.fao.org/fileadmin/templates/agphome/scpi/cgwg/carbon_finance.pdf
The Royal Society, 2001. The role of land carbon sinks in mitigating global climate change.
Policy document 10/01. The Royal Society, London.
Thomas, C.D., Cameron, A., Green, R.E., Bakkenes, M., Beaumont, L.J., Collingham, Y.C.,
Erasmus, B.F.N., de Siqueira, M.F., Grainger, A., Hannah, L., Hughes, L., Huntley, B.,
van Jaarsveld, A.S., Midgley, G.F., Miles, L., Ortega-Huerta, M.A., Peterson, A.T.,
Phillips, O.L., Williams, S.E., 2004. Extinction risk from Climate change. Nature. 427,
145-148.
Thornton P.K., Herrero M., Freeman A., Mwai O., Rege E., Jones P., McDermott J., 2008.
Vulnerability, Climate Change and Livestock: Research Opportunities and Challenges for
Poverty Alleviation. International Livestock Research Institute (ILRI), Kenya.
Thornton, P.K., 2010. Livestock production: recent trends, future prospects. Philos. T. Roy. Soc.
B. 365, 2853â2867.
Thornton, P.K., Boone, R.B., Ramirez-Villegas J., 2015. Climate change impacts on livestock.
CGIAR Resrach program on Climate Change, Agriculture and Food Security (CCAFS),
Working Paper No. 120.
Thornton, P.K., Gerber, P.J., 2010. Climate change and the growth of the livestock sector in
developing countries. Mitig. Adapt. Strateg. Glob. Change. 15, 169-184.
Thornton, P.K., Herrero, M., 2010. The Inter-linkages between rapid growth in livestock
production, climate change, and the impacts on water resources, land use, and
deforestation. World Bank Policy Research Working Paper, WPS 5178. World Bank,
Washington, DC.
Thornton, P.K., Van de Steeg, J., Notenbaert, A., Herrero, M., 2009. The impacts of climate
change on livestock and livestock systems in developing countries: A review of what we
know and what we need to know. Agric. Syst. 101, 113-127.
Thrasher, B., Maurer, E.P., McKellar, C., Duffy, P.B., 2012. Technical Note: Bias correcting
climate model simulated daily temperature extremes with quantile mapping. Hydrol.
Earth Syst. Sci. 16, 3309â3314.

174

Tubiello, F., Schmidhuber, J., Howden, M., Neofotis, P.G., Park, S., Fernandes, E., Thapa, D.,
2008. Climate Change Response Strategies for Agriculture: Challenges and Opportunities
for the 21st Century. The World Bank, Washington, DC.
UN (United Nations), 2013. World population projected to reach 9.6 billion by 2050 [WWW
Document]. United Nations Department of Economic and Social Affairs. URL
http://www.un.org/en/development/desa/news/population/un-report-world-populationprojected-to-reach-9-6-billion-by-2050.html
UNEP (United Nations Environment Programme), 2012. Global Environment Outlook 5:
Chapter 5 [WWW Document]. URL
http://www.unep.org/geo/pdfs/geo5/GEO5_report_C5.pdf
UNFCCC (United Nations Framework Convention on Climate Change), 1998. Kyoto Protocol to
the United Nations Framework Convention on Climate Change adopted at COP3 in
Kyoto, Japan, on 11 December 1997 [WWW Document]. URL
http://unfccc.int/resource/docs/convkp/kpeng.pdf
UNFCCC (United Nations Framework Convention on Climate Change), 2008. Challenges and
opportunities for mitigation in the agricultural sector: technical paper [WWW
Document]. URL http://unfccc.int/resource/docs/2008/tp/08.pdf
UNFCCC (United Nations Framework Convention on Climate Change), 2014. Global Warming
Potentials [WWW Document]. URL http://unfccc.int/ghg_data/items/3825.php
United Dairy Industry of Michigan, 2016. Dairy Facts [WWW Document]. URL
http://www.milkmeansmore.org/print/36.
United Soybean Board, 2013. Animal Agriculture Economic Analysis: 2002 - 2012. Alexandria,
VA.
USDA (United States Department of Agriculture), 2013. Climate change and agriculture in the
United States: Effects and adaptation [WWW Document]. USDA technical bulletin,
Washington, DC. URL
http://www.usda.gov/oce/climate_change/effects_2012/CC%20and%20Agriculture%20R
eport%20%2802-04-2013%29b.pdf
USDA-ARS (United States Department of Agriculture- Agricultural Research Service), 2014.
Integrated Farm System Model. In: Pasture Syst [WWW Document]. Watershed Mgt.
Res. Unit.. URL http://www.ars.usda.gov/Main/docs.htm?docid=8519.
USDA-ARS (United States Department of Agriculture- Agricultural Research Service), 2016.
Integrated Farm System Model [WWW Document]. URL
https://www.ars.usda.gov/northeast-area/up-pa/pswmru/docs/integrated-farm-systemmodel/
USDA-NASS (United States Department of Agriculture-National Agricultural Statistics
Service), 2014a. 2012 Census of Agriculture- Michigan State and County Data.
175

USDA-NASS (United States Department of Agriculture-National Agricultural Statistics Service),
2014b. Cropland Data Layer. In: Michigan Pasturel.. URL
https://nassgeodata.gmu.edu/CropScape/
USDA-NASS (United States Department of Agriculture-National Agricultural Statistics
Service), 2014c. Milk cows and production: Final estimates 2008-2012 [WWW
Document]. URL
http://usda.mannlib.cornell.edu/usda/nass/milkcowest//2010s/2014/milkcowest_Milk
Cows and Production - Final Estimates, 2008-12.pdf
USDA-NRCS (United States Department of Agriculture- Natural Resources Conservation
Service), 2006. Conservation Practices that Save: Energy Conservation in Confined
Animal Operations [WWW Document]. URL
http://www.nrcs.usda.gov/Internet/FSE_DOCUMENTS/nrcs143_023625.pdf
USDA-NRCS (United States Department of Agriculture- Natural Resources Conservation
Service), 2007. Legumes for Resource Conservation Programs [WWW Document]. URL
http://www.nrcs.usda.gov/Internet/FSE_DOCUMENTS/stelprdb1042287.pdf
USDA-NRCS (United States Department of Agriculture- Natural Resources Conservation
Service), 2002. Plant Fact Sheet for Red Clover (Trifolium pratense L.). Washington,
DC.
USGS (United States Geological Survey), 2017. Hydrologic Unit Maps [WWW Document].
URL https://water.usgs.gov/GIS/huc.html
van Calker, K.J., Berentsen, P.B.M., Giesen, G.W.J., Huirne, R.B.M., 2005. Identifying and
ranking attributes that determine sustainability in Dutch dairy farming. Agric. Human
Values 22, 53â63.
van der Spiegel, M., van der Fels-Klerx, H.J., Marvin, H.P., 2012. Effects of climate change on
food safety hazards in the dairy production chain. Food Res. Int. 46, 201â208.
Vasilaki, V., Katsou, E., PonsĂĄ, S., ColĂłn, J., 2016. Water and carbon footprint of selected dairy
products: A case study in Catalonia. J. Clean. Prod. 139, 504â516.
Viessman Jr., W., Lewis, G.L., 2003. Introduction to hydrology, 5th ed. Pearson Education Inc.,
Upper Saddle River, NJ.
von Keyserlingk, M., Martin, N., Kebreab, E., Knowlton, K., Grant, R., Stephenson, M., Sniffen,
C., Harner III, J., Wright, A., Smith, S., 2013. Invited review: Sustainability of the US
dairy industry. J. Dairy Sci. 96, 5405â5425.
Wachendorf, M., Golinski, P., 2006. Towards Sustainable intensive dairy farming in Europe.
Grassl Sci Eur 11, 159â174.
Wallis, T.W.R., Griffiths, J.F., 1995. An assessment of the weather generator (WXGEN) used in
the erosion/productivity impact calculator (EPIC). Agric. For. Meteorol. 73, 115â133.
176

Walthall, C., Hatfield, J., Backlund, P., 2012. Climate Change and Agriculture in the United
States: Effects and Adaptation. USDA Technical Bulletin 1935. Washington, DC.
Wand, S.J., Midgley, G.F., Jones, M.H., Curtis, P.S., 1999. Responses of wild C4 and C3 grass
(Poaceae) species to elevated atmospheric CO2 concentrations: a meta-analytic test of
current theories and perceptions. Glob. Change Biol. 5, 723-141.
Wani, S.P., Rockstrom, J., Oweis, T., 2009. Rainfed Agriculture: Unlocking the Potential. CAB
International, Wallingford, UK.
Wassenaar, T., Gerber, P., Verburg, P.H., Rosales, M., Ibrahim, M., Steinfeld, H., 2007.
Projecting land use changes in the Neotropics: The geography of pasture expansion into
forest. Glob. Environ. Change. 17, 86â104.
Webster, A.J., 1991. Metabolic responses of farm animals to high temperature, in: Ronchi, B.,
Nardone, A., Boyazoglu, J. (Eds.), Animal husbandry in warm climates, EAAP
Publication, pp. 15-22.
WEC (World Energy Council), 2015. 2015 World Energy Issues Monitor [WWW Document].
URL http://www.worldenergy.org/wp-content/uploads/2015/01/2015-World-EnergyIssues-Monitor.pdf
White, N., Sutherst, R.W., Hall, N., Whish-Wilson, P., 2003. The vulnerability of the Australian
beef industry to impacts of the cattle tick (Boophilus microplus) under climate change.
Climatic Change. 61, 157-190.
White, R.P., Murray, S., Rohweder, M., 2000. Pilot analysis of global ecosystems: grassland
ecosystems [WWW Document]. World Resources Institute, Washington, DC. URL
http://www.wri.org/sites/default/files/pdf/page_grasslands.pdf
Wittmann, E.J., Mellor, P.S., Baylis, M., 2001. Using climate data to map the potential
distribution of Culicoides imicola (Diptera: Ceratopogonidae) in Europe. Rev. Sci. Tech,
20: 731-740.
Wolfenson, D., Roth, Z., Meidan, R., 2000. Impaired reproduction in heat-stressed cattle: basic
and applied aspects. Anim. Rep. Sci. 60(1-3), 535-547.
Woznicki, S.A., Nejadhashemi, A.P., 2012. Sensitivity Analysis of Best Management Practices
Under Climate Change Scenarios1. JAWRA J. Am. Water Resour. Assoc. 48, 90â112.
Woznicki, S.A., Nejadhashemi, A.P., Ross, D.M., Zhang, Z., Wang, L., Esfahanian, A.-H., 2015.
Ecohydrological model parameter selection for stream health evaluation. Sci. Total
Environ. 511, 341â353.
Woznicki, S.A., Pouyan Nejadhashemi, A., 2014. Assessing uncertainty in best management
practice effectiveness under future climate scenarios. Hydrol. Process. 28, 2550â2566.

177

Wright, I.A., Tarawali, S., Blummel, M., Gerard, B., Teufel, N., Herrero, M., 2012. Integrating
crops and livestock in subtropical agricultural systems. J. Sci. Food Agr. 92, 1010-1015.
Wyman, O., Johnson, H.D., Merilan, C.P., Berry, I.L., 1962. Effect of ad libitum and force
feeding of two rations on lactating dairy cows subject to temperature stress. J. Dairy Sci.
45, 1472-1478.
Yan, T., Agnew, R.E., Gordon, F.J., Porter, M.G., 2000. Prediction of methane energy output in
dairy and beef cattle offered grass silage-based diets. Livest. Prod. Sci. 64, 253-263.
Yu, P.L., 1973. A Class of Solutions for Group Decision Problems. Manage. Sci. 19, 936â946.
Zanakis, S.H., Solomon, A., Wishart, N., Dublish, S., 1998. Multi-attribute decision making: A
simulation comparison of select methods. Eur. J. Oper. Res. 107, 507â529.
Zehetmeier, M., Gandorfer, M., Hoffmann, H., Mu ller, U.K., De Boer, I.J.M., Heibenhuber, A.,
2014. The impact of uncertainties on predicted greenhouse gas emissions of dairy cow
production systems. J. Clean. Prod. 73.
Zonderland-Thomassen, M.A., Ledgard, S.F., 2012. Water footprinting â A comparison of
methods using New Zealand dairy farming as a case study. Agric. Syst. 110, 30â40.

178