Citation
Towards regional sustainability assessment utilizing community based participatory research, sustainability indicators, and future scenario modeling

Material Information

Title:
Towards regional sustainability assessment utilizing community based participatory research, sustainability indicators, and future scenario modeling
Creator:
Dubinsky, Jonathan ( author )
Language:
English
Physical Description:
1 electronic file (59 pages) : ;

Thesis/Dissertation Information

Degree:
Doctorate ( Doctor of philosophy)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
College of Engineering and Applied Sciences, CU Denver
Degree Disciplines:
Engineering and applied science
Committee Chair:
Karunanithi, Arunprakash
Committee Members:
Janson, Bruce
Cabezas, Heriberto
Hopton, Matthew
Main, Deborah

Subjects

Subjects / Keywords:
Greenhouse gas mitigation -- Colorado ( lcsh )
Water use -- Colorado ( lcsh )
Renewable energy sources ( lcsh )
Sustainable agriculture -- Colorado ( lcsh )
Greenhouse gas mitigation ( fast )
Renewable energy sources ( fast )
Sustainable agriculture ( fast )
Water use ( fast )
Colorado ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Decision making with regards to sustainability at a regional/local level is increasingly recognized as an important issue. This research focused on the rural agricultural region of San Luis Valley (SLV) in southern Colorado and builds on an Environmental Protection Agency sustainability study in the same region. The goal of this research was to select and calculate relevant sustainability indictors, and then use those baseline results to model potential future scenarios. A region-specific greenhouse gas (GHG) accounting indicator and a consumptive water use indicator were developed and calculated in SLV over the period of 1980 - 2010. In addition to sustainability modeling, this research leveraged the principles of Community Based Participatory research (CBPR) to engage local stakeholders throughout the process to ensure that the research was relevant to the region. ( ,,, )
Review:
From a carbon emissions perspective, the baseline assessment showed that on a per gross domestic/regional product (GDP) basis, SLV carbon emissions are almost twice that of the national average, indicating that, with all things being equal, agricultural economies contribute disproportionately more towards GHG emissions. A future scenario analysis revealed that SLV can reduced GHG emissions by ~5% through crop changes and alterations to the crop rotation regime. Another scenario showed that SLV, with its significant solar resource, has the potential to offset much or all of its GHG emissions by increasing solar development and utilizing renewable energy carbon credits.
Review:
The baseline water use assessment of the region showed that SLV meets 85% of the crop water demand through irrigation, whereas the global average is only 19%. This highlights the heavy reliance the economy has on its ground and surface water resources in a region where agriculture is the number one economic driver. When comparing livestock water use (including feed crops grown to support local livestock) to export crops in the region (i.e., potatoes, alfalfa, small grains) we see that livestock consume significantly less groundwater than the export crops on a per ton of product basis, revealing the critical role livestock play in this arid region with a depleting aquifer. Results from the future scenario modeling showed that irrigation water use could be reduced by ~10% through realistic shifts in the crop regime while keeping land fallowing to a minimum.
Review:
This research was successful in terms of engaging a rural region around issues of sustainability. Through training and knowledge transfer from researchers to the community, SLV now has the ability to use the sustainability indicator models developed in this research. This work is highly relevant to policymakers and planners, and will provide the community with some necessary tools to make policy choices for sustainable growth in the region.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: Adobe Reader.
Restriction:
Embargo ended 05/16/2019
Statement of Responsibility:
by Jonathan Dubinsky.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
on10137 ( NOTIS )
1013720387 ( OCLC )
on1013720387
Classification:
LD1193.E553 2016d D93 ( lcc )

Downloads

This item has the following downloads:


Full Text
TOWARDS REGIONAL SUSTAINABILITY ASSESSMENT
UTILIZING COMMUNITY BASED PARTICIPATORY RESEARCH, SUSTAINABILITY INDICATORS, AND FUTURE SCENARIO MODELING
By
JONATHAN DUBINSKY B.S., University of Kansas, 2004 M.Eng., University of Colorado Denver, 2013
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Engineering and Applied Science
2016


This thesis for the Doctor of Philosophy degree by Jonathan Dubinsky has been approved for the Engineering and Applied Science Program
by
Arunprakash Karunanithi, Advisor Bruce Janson, Chair Heriberto Cabezas Matthew Hopton Deborah Main
December 17, 2016


Dubinsky, Jonathan (PhD, Engineering and Applied Science)
Towards Regional Sustainability Assessment Utilizing Community Based Participatory Research, Sustainability Indicators, and Future Scenario Modeling
Thesis directed by Associate Professor Arunprakash Karunanithi.
ABSTRACT
Decision making with regards to sustainability at a regional/local level is increasingly recognized as an important issue. This research focused on the rural agricultural region of San Luis Valley (SLV) in southern Colorado and builds on an Environmental Protection Agency sustainability study in the same region. The goal of this research was to select and calculate relevant sustainability indictors, and then use those baseline results to model potential future scenarios. A region-specific greenhouse gas (GHG) accounting indicator and a consumptive water use indicator were developed and calculated in SLV over the period of 1980 - 2010. In addition to sustainability modeling, this research leveraged the principles of Community Based Participatory research (CBPR) to engage local stakeholders throughout the process to ensure that the research was relevant to the region.
From a carbon emissions perspective, the baseline assessment showed that on a per gross domestic/regional product (GDP) basis, SLV carbon emissions are almost twice that of the national average, indicating that, with all things being equal, agricultural economies contribute disproportionately more towards GHG emissions. A future scenario analysis revealed that SLV can reduced GHG emissions by ~5% through crop changes and alterations to the


crop rotation regime. Another scenario showed that SLV, with its significant solar resource, has the potential to offset much or all of its GHG emissions by increasing solar development and utilizing renewable energy carbon credits.
The baseline water use assessment of the region showed that SLV meets 85% of the crop water demand through irrigation, whereas the global average is only 19%. This highlights the heavy reliance the economy has on its ground and surface water resources in a region where agriculture is the number one economic driver. When comparing livestock water use (including feed crops grown to support local livestock) to export crops in the region (i.e., potatoes, alfalfa, small grains) we see that livestock consume significantly less groundwater than the export crops on a per ton of product basis, revealing the critical role livestock play in this arid region with a depleting aquifer. Results from the future scenario modeling showed that irrigation water use could be reduced by ~10% through realistic shifts in the crop regime while keeping land fallowing to a minimum.
This research was successful in terms of engaging a rural region around issues of sustainability. Through training and knowledge transfer from researchers to the community, SLV now has the ability to use the sustainability indicator models developed in this research. This work is highly relevant to policymakers and planners, and will provide the community with some necessary tools to make policy choices for sustainable growth in the region.
IV


The form and content of this abstract are approved. I recommend its publication.
Approved: Arunprakash Karunanithi
v


DEDICATION
I dedicate this work to my loving and supportive sweetheart, Elizabeth. She has encouraged me to always pursue my dreams with confidence and grace.
VI


ACKNOWLEDGMENTS
First I thank my advisor, Dr. Arunprakash Karunantihi, for his countless hours of time and counsel spent with me throughout this process. He contributed intellectually to the research and analysis and he mentored me through personal and professional challenges. I thank him for his financial support and his trust in me to accomplish this large task.
I also thank and acknowledge my collaborators on this research. Thank you, Dr. Matthew Hopton and Dr. Matthew Heberling, from the U.S. Environmental Protection Agency for your countless discussions and insights throughout this process. To my PhD committee: Dr. Bruce Janson, Dr. Deborah Main, Dr. Heriberto Cabezas, and Dr. Mathew Hopton. Thank you for your commitment to this work and for your helpful comments, constructive criticism, patience and support.
I acknowledge the members of the Community Advisory Board (CAB), without whom this research would not be possible. A very special thank you to Richard Sparks, Sheldon Rockey, Cleave Simpson, Patrick O’Neill, and George Whitten, who were always on call and ready and willing to discuss agricultural land use in detail. Also a special thank you to Christine Canaly, John Stump, and Mary Hoffman for attending so many meetings, asking so many good questions, and being so committed to this work.
VII


Thank you to all members of the CAB (listed below in alphabetical order):
Tony Aloia, Colorado Fish and Wildlife Michael Armenta, Conejos County Clean Water
Fred Bunch, Chief of Resources Management at Great Sand Dunes National Park
Christine Canaly, Director of the San Luis Valley Ecosystem Council Nathan Coombs, Director of the Conejos Conservancy District Claudia Ebel, San Luis Valley Local Foods Coalition Regenarldo Garcia, Rocky Mountain Prevention Research Center Andrea Guajardo, Director of Conejos County Clean Water Mary Hoffman, Adams State University Community Partnerships Jared Beeton, Adams State University Karla Shriver, Rio Grande County Commissioner Jim Meitz, Sustainability Park (SEED)
Patrick O’Neill, Owner of Soil Health Services
Lawrence Pacheco, Costilla County Commissioner
Sheldon Rockey, Rockey Farms
Leroy Salazar, Water Users Solution Sub-Committee
Cleave Simpson, Manager of the Rio Grande Water Conservation District
Richard Sparks, Natural Resource Conservation Service
John Stump, Former head of the SLV Development Resources Group
George and Julie Whitten, Saguache County Ranchers
VIII


I acknowledge other collaborators in this research. To Debbi Main, Elizabeth Baker-Jennings, and Tamara Chernomordik for greatly contributing to the CBPR portion of the research. Thank you to Rio de la Vista, James Heath, and Dr. Willem Schreuder for providing region specific context and data for the Consumptive Water Use Chapter.
Thank you to Brian Lewendowski from University of Colorado Boulder for working with me on the Implan model and data. A special thank you to Matthew Stermer, Mark Easter, and the entire Comet-farm team at Colorado State University.
I recognize the U.S. Environmental Protection Agency’s office of Research and Development, National Risk Management Research Laboratory, Sustainable Technology Division for developing the foundation of knowledge in this research and for providing funding to make this effort possible (cooperative agreement number 83522701).
Finally, I would like to acknowledge someone that worked on and discussed issues of sustainability long before it was a popular topic. Buckminster Fuller’s writings, which were given to me by my father at age 12, informed me of our responsibilities as crewmembers aboard Spaceship Earth. He sounded the alarm over a generation ago, and provided us with prophetic insights of how we can improve the lives of all human kind while simultaneously protecting and enhancing the very planet (spaceship) that makes it all possible.
IX


TABLE OF CONTENTS
CHAPTER
I INTRODUCTION.................................................................1
Background...................................................................1
Sustainability Science....................................................2
Measuring Sustainability..................................................3
Rural Regions.............................................................4
Previous Work................................................................5
Ecological Footprint Analysis.............................................6
Emergy Analysis...........................................................6
Green Net Regional Product................................................7
Fisher Information........................................................7
Present Work.................................................................8
Arrangement of the Dissertation..........................................13
II GREENHOUSE GAS ACCOUNTING..................................................17
Abstract....................................................................17
Introduction................................................................18
Methodology.................................................................21
Energy and Buildings Sector..............................................22
Agriculture and Land Use Sector..........................................24
Transportation Sector....................................................27
Waste Sector.............................................................28
Solar Credit.............................................................30
Results and Discussion......................................................32
x


Total GHG emissions in SLV...............................................32
Upstream Emissions.......................................................33
Carbon Credit............................................................34
Emissions by Category....................................................35
Comparison...............................................................36
Uncertainty..............................................................40
III CONSUMPTIVE WATER USE.....................................................42
Abstract...................................................................42
Introduction...............................................................43
Methods....................................................................49
Crop Water Use of SLV....................................................51
Livestock Water Use of SLV...............................................54
Municipal and Industrial Water Use of SLV................................57
Results and Discussion......................................................57
Total Consumptive Water Use of SLV.......................................57
Crop Consumptive Water Use of SLV........................................59
Livestock Consumptive Water Use of SLV...................................61
Agricultural Products....................................................63
Blue Water Scarcity......................................................64
Looking Forward..........................................................67
IV COMMUNITY ENGAGEMENT......................................................70
Abstract...................................................................70
Introduction...............................................................71
Background to Participatory Research.....................................71
XI


Community Engagement and Sustainability Research
73
Case of San Luis Valley.....................................................74
Goals.......................................................................75
Methodology...................................................................76
Defining the Research Question..............................................77
Stakeholder Identification..................................................78
Community Advisory Board....................................................80
Co-develop Sustainability Indicator Baseline................................82
Selecting Scenarios.........................................................83
Results.......................................................................85
Future Scenarios............................................................85
Training and Knowledge Transfer.............................................92
Discussion....................................................................92
Best Practices and Lessons Learned..........................................92
Limitations.................................................................94
Conclusion....................................................................95
V FUTURE SCENARIO MODELING......................................................96
Abstract......................................................................96
Introduction..................................................................97
Methods......................................................................102
Crop Regime Change Scenario................................................102
Solar Energy Development Scenario..........................................Ill
Results / Discussion.........................................................116
Results from the Crop Regime Change Scenario...............................116
Results from the Solar Energy Development Scenario.........................122
xii


Summary of Findings
127
VI Conclusion............................................................130
Summary of Results......................................................130
Chapter II: Greenhouse Gas Accounting................................130
Chapter III: Consumptive Water Use Analysis..........................131
Chapter IV: Community Engagement.....................................132
Chapter V: Future Scenario Modeling..................................133
Future Work.............................................................135
Economic Analysis....................................................135
Sustainability Metrics and Future Scenario Modeling..................136
Maintaining Relationships............................................142
Beyond SLV...........................................................142
References................................................................144
APPENDIX..................................................................160
A - Greenhouse Gas (GHG) Emissions Indicator............................160
B - Consumptive Water Use (CWU) Indicator...............................190
C - Future Scenario Modeling............................................212
xiii


LIST OF TABLES
Table 1: Material flows, emission factors, and total GHG emissions in the San
Luis Valley...........................................................31
Table 2: Comparing the SLV inventory with GHG inventories at various spatial
scales................................................................39
Table 3: A summary of water sources considered in this study..............49
Table 4: Cultivated area and crop production in SLV (2000-2010)............53
Table 5: Livestock herd in SLV (2000 - 2010) for each animal type.........54
Table 6: Consumptive Water Content (CWC) of feed in the SLV (2000-2010) (1 Ac-ft. = 1233.5 m3)...............................................55
Table 7: Comparison of major agricultural products in SLV (1 Ac-ft. = 1233.5 m3).
........................................................................64
Table 8: The future scenarios selected by the CAB for modeling with the
sustainability indicators...............................................86
Table 9: Crop rotations under the crop regime change scenario................91
Table 10: Comparison between the baseline crop rotational acreage in SLV and the crop rotational acreage under the crop regime change scenario.......104
Table 11: Inputs used to assess the baseline GHG emissions in the region using the COMET-Farm model, and the outputs produced for each of the crop
rotations assessed
107


Table 12: Inputs used to assess the GHG emissions of the crop regime change future scenario using the COMET-Farm model, and the outputs produced for
each of the crop rotations assessed..............................108
Table 13: Data used for estimating electricity needed per volume of groundwater use. Data from Xcel energy were only available from 2006 to 2010, so this
period was used to establish the average.........................109
Table 14: Summary of the solar development pathways explored under the solar
energy development scenario.......................................112
Table 15: Annual crop acreage in SLV (avg. 2000 - 2010; acres) as well as crop
acreage under the fully implemented crop regime change scenario..117
Table 16: Changes in GHG emissions from soil in SLV comparing the baseline
scenario and the crop regime change scenario......................119
Table 17: Results on changes in irrigation consumptive water use from the crop
regime change scenario............................................120
Table 18: Results from the GHG emissions indicator when modeling the two solar
development pathways in the solar energy development scenario.....124
Table 19: Results from the consumptive water use indicator when modeling the two solar development pathways in the solar energy development scenario
..................................................................127
Table 20: Data and data sources used for the GHG emissions indicator.Error!
Bookmark not defined.
xv


Table 21: Model inputs for the GHG emissions indicator (1980-1989). Data sources are presented in Table 20. Blank cells in this table show data were unavailable for that year through the data provider. The model linearly interpolates data gaps......................Error! Bookmark not defined.
Table 22: Model inputs for the GHG emissions indicator (1990-1999). Data sources are presented in Table 20. Blank cells in this table show data were unavailable for that year through the data provider. The model linearly interpolates data gaps......................Error! Bookmark not defined.
Table 23: Model inputs for the GHG emissions indicator (2000-2009). Data sources are presented in Table 20. Blank cells in this table show data were unavailable for that year through the data provider. The model linearly interpolates data gaps......................Error! Bookmark not defined.
Table 24: Model inputs for the GHG emissions indicator (2010-2017). Data sources are presented in Table 20. Blank cells in this table show data were unavailable for that year through the data provider. The model linearly interpolates data gaps......................Error! Bookmark not defined.
Table 25: Model outputs from the GHG emissions indicator (1980-1989).Error!
Bookmark not defined.
Table 26: Model outputs from the GHG emissions indicator (1990-1999).Error!
Bookmark not defined.
Table 27: Model outputs from the GHG emissions indicator (2000-2009).Error!
Bookmark not defined.
XVI


Table 28: Model outputs from the GHG emissions indicator (2010-2016).Error!
Bookmark not defined.
Table 29: Data and data sources used for the CWU (CWU) indicator.....Error!
Bookmark not defined.
Table 30: Model inputs for the CWU indicator (1980-1989). Data sources for all data are presented in Table 29. Pay close attention as the units change throughout the table.........................Error! Bookmark not defined.
Table 31: Model inputs for the CWU indicator (1990-1999). Data sources for all data are presented in Table 29. Pay close attention as the units change throughout the table.........................Error! Bookmark not defined.
Table 32: Model inputs for the CWU indicator (2000-2009). Data sources for all data are presented in Table 29. Pay close attention as the units change throughout the table.........................Error! Bookmark not defined.
Table 33: Model inputs for the CWU indicator (2010-2014). Data sources for all data are presented in Table 29. Pay close attention as the units change throughout the table.........................Error! Bookmark not defined.
Table 34: Model outputs for the CWU indicator (1980-1990).Error! Bookmark
not defined.
Table 35: Model outputs for the CWU indicator (1991-2001).Error! Bookmark
not defined.
Table 36: Model outputs for the CWU indicator (2002-2010).Error! Bookmark
not defined.
XVII


xviii


LIST OF FIGURES
Figure 1: The San Luis Valley. Whole counties were always considered for data collection even though parts of the county may fall outside of the Upper Rio
Grande river basin geographic boundary................................9
Figure 2: Average monthly precipitation (1980 - 2010) during the growing season
(May to August) in SLV (PRISM Climate Group 2016)*....................10
Figure 3: Community engagement timeline in the SLV during the course of this
research..............................................................15
Figure 4: The GHG accounting for each of the major emission categories in San
Luis Valley. This includes both direct emissions and upstream emissions. .33 Figure 5: The breakdown of GHG emission contributions of the four major
categories averaged between 2006 and 2012.............................36
Figure 6: Change in storage of the unconfined aquifer in SLV from 1975-2016
(Davis Engineering 2016).................................................47
Figure 7: Geospatial representation of crop production in SLV................52
Figure 8: Average annual water allocation in SLV from 2000-2010 (1 Ac-ft. =
1233.5 m3)...............................................................58
Figure 9: The crop water use allocated between CWUgreen-crops, CWUSUrface-crops, and CWUground-crops as well as the consumptive water content of each crop
(2000-2010)*....................................................................60
Figure 10: Livestock consumptive water use (CWUnvestock) for SLV allocated
between CWUgreen-crops> CWUsurface-crops, and CWUground-crops OS Well 3S the
XIX


consumptive use content of each crop. Cattle represent 98% of all
CWUiivestock in the region*.................................................62
Figure 11: Monthly water availability and monthly water use (2000-2010) (1 Ac-ft.
= 1233.5 m3)................................................................65
Figure 12: A summary of the key aspects of a Community Based Participatory Research project (Israel et al. 1998).......................................73
Figure 13: Framework for community engagement in San Luis Valley. The number in the circle indicates the order of each step. * The scenario
modeling methods and results are presented in Chapter V...............77
Figure 14: Interview guide used to spark discussion when informally engaging
with residents of SLV..................................................78
Figure 15: Regional map of San Luis Valley with geographic distribution of the
Community Advisory Board members*......................................81
Figure 16: San Luis Valley with aquifer boundaries mapped in yellow and
irrigation wells in green (Colorado Geological Survey 2016)...........88
Figure 17: San Luis Valley with groundwater pumping response areas outlined in purple (Sub districts fall within the boundary of response areas) (CDWR
2015d).................................................................89
Figure 18: The four IPCC emissions scenarios, presented in the 1990 summary to policymakers (IPCC Working Group I 1990), expressed in both radioactive forcing and equivalent carbon dioxide concentrations*......................98
xx


Figure 19: San Luis Valley with groundwater pumping response areas/sub
districts outlined in purple (CDWR 2015d).............................102
Figure 20: The impact on the GHG emissions indicator from the crop regime change scenario*......................................................118
Figure 21: The average agricultural (crops and livestock) consumptive water use (black line) shown with the average water budget (average water inflows
minus compact deliveries) (pink line) in San Luis Valley*...........121
Figure 22: The impact on the GHG emissions indicator when the community
response solar development pathway is implemented*..................125
Figure 23: The impact on the GHG emissions indicator when the DOE/BLM solar development pathway is implemented*..................................126
Figure 24: Surface soil texture distribution among all irrigated lands in SLV. Error! Bookmark not defined.
Figure 25: Surface soil texture distribution among all alfalfa lands in SLV. ...Error! Bookmark not defined.
Figure 26: Surface soil texture distribution among the alfalfa lands sample in
SLV............................................Error! Bookmark not defined.
Figure 27: Surface soil texture distribution among all potato lands in SLV. ...Error! Bookmark not defined.
Figure 28: Surface soil texture distribution among the potato lands sample in SLV.................................................Error! Bookmark not defined.
XXI


Figure 29: Surface soil texture distribution among
Figure 30: Surface soil texture distribution among
SLV...........................................
Figure 31: Surface soil texture distribution among
SLV...........................................
Figure 32: Surface soil texture distribution among sample in SLV......................................
all the small grain lands in SLV. Error! Bookmark not defined, the small grain lands sample in Error! Bookmark not defined, all meadow/pasture lands in Error! Bookmark not defined, the meadow/pasture lands Error! Bookmark not defined.
XXII


CHAPTER I
INTRODUCTION
Background
Humanity’s future depends on the ability of the built environment to work within the constraints of the planet’s natural systems. For the past century, society has been in a period of rapid growth, made possible in large part to the discovery and wide use of fossil fuels (Smil 2010). In his final work, Howard T. Odum, likened the evolution of society to that of a forest (Odum and Odum 2001). He noted that during primary and secondary succession, the health of a forest is measured by growth, however in a mature climax-forest, health is measured not by growth, but by the forest’s ability to maintain a high level of biodiversity and to quickly rebound from natural disasters. Inspirational ideas from the natural world such as this have given rise to the field of sustainability science (Kates et al. 2001) and resilience thinking (Gunderson and Holling 1996). Resilience is defined as the ability of a system to respond to a stress while maintaining system identity and primary functions (Walker et al. 2004). Indicators such as Gross Domestic Product (GDP) only tell us how well our societal system is growing. Developing measures necessary to evaluate society in this climax-oriented stage, Odum claims, is humanity’s greatest challenge. We must identify and make use of new measures that provide meaningful information for all stages of human/earth co-existence (England 1998; Costanza et al. 1998; Lawn 2003).
1


Sustainability Science
This research contributes to the growing field of sustainability science. The definition of sustainability, as articulated by the United Nations World Commission on Environment and Development, is: “Development that meets the needs of the present generation without compromising the future generations ability to meet their own needs” (Brundtland 1985; United Nations Division for Sustainable Development 1992). Planning and policy making that incorporates sustainable growth principles at a regional/local level is increasingly recognized as an important practice (Ali 2013; Graymore et al. 2010). Entire frameworks have been developed to help infrastructure designers (e.g., planners and engineers), policymakers, and individual users understand the complex nature of modern society, and to provide a forum for communication across disciplines (Ramaswami et al. 2012; Ramos 2010; Dovers 1995; Fiksel et al. 2012). Fostering open lines of communication between various actors in society which normally do not interact, is a relatively new endeavor that has been triggered by a need/demand for sustainability (Ramaswami et al. 2012). These open lines of communication are critical to assist local government, who must often balance the competing interests of their constituents, in fostering economic growth while also managing for sustainability.
Flowever, policymakers do not always have tools necessary to measure the sustainability implications of their actions. Tools for measuring economic growth are well known and commonly used, and policymakers can easily look back over the previous years to see how activities in their region have correlated
2


to changes in GDP and Net Domestic Product (NDP). These trends can be used to project how policy changes may affect those measures into the future. It is important that the analysis not end there. Decision-makers must also understand how policies fit into the context of the three pillars of sustainability (Janou~ 2012). The pillars are the economic, social, and environmental components of a human system, and all three must be sustainable for the system to be considered sustainable (United Nations 2002). It is no longer sufficient to simply measure the economy—there is a need to assess the environmental resources on which those economies are based.
Measuring Sustainability
It is difficult to measure how policies and activities in a region will affect sustainability. To address this issue, much work has been done over the last decade in the development of sustainability metrics and indicators (Morse 2015; Zidansek et al. 2014; Ness et al. 2007; Ramaswami et al. 2008). Sustainability indicators measure one characteristic of a system (e.g., C02 emissions (IPCC 2006; WRI 2004; U.S. EPA 2013a) or biodiversity (Tasser et al. 2008)), whereas a metric combines many indicators (or variables) through aggregation (e.g., Green Net Regional Product (GNRP) (Hamilton et al. 1997; Kahn 1995)). Sustainability measures can be computed over time to gain an understanding of how resource usage is changing. There are numerous types of sustainability metrics and indicators that now exist as tools for policymakers (Mori and Christodoulou 2012; Mayer 2008). A number of geographic regions have
3


recognized the importance of measuring sustainability, and have developed action plans based on sustainability metrics and indicators (Gallucci 2013; Reed et al. 2006; Obama 2013).
Rural Regions
Much of the research into sustainability metrics and indicators has been focused on countries, but increasingly, focus has turned to cities and urban areas, understandably, because urban population centers are experiencing some of the fastest growth across the globe (Kennedy et al. 2010; Hillman and Ramaswami 2010). This research, however, does not focus on the consumption-heavy urban regions, but rather on the surrounding hinterlands. Consumption-based metrics and indicators at the city-scale must be coupled with studies that look at the production-based regions that help support cities to get a full picture of the specific system in question. According to Amory Lovins (Lovins and Hawken 2007), “Natural Capital” is the fundamental basis for all economic growth. This is because natural resources and ecological systems provide vital life-support services and resources to society. It is often the case that city-scale assessments will use state or national level data because local/region specific studies are lacking (Ramaswami et al. 2008). The reliance on natural resources (i.e., land, water, etc.) for economic growth is clearly evident in agricultural regions. A need exists for analysis of these regions from a sustainability perspective to both improve city-scale assessments as well as inform policymakers in rural areas.
4


Previous Work
The United States Environmental Protection Agency (EPA) Office of Research and Development created a methodology for computing sustainability metrics at a regional scale. They began their effort to quantify sustainability using a 7 county region as a pilot study. The region was encompassed by the San Luis Valley (SLV), Upper Rio Grande river basin (SLV), and the Great Sand Dunes National Park and Preserve (U.S. EPA 2010). This region was selected due to its limited population, distinct physical and political boundaries, the strong support of the local population, and remote geographic location, whereby tracking inputs and outputs was expected to be straightforward (U.S. EPA 2010). The purpose of EPA’s research was to identify metrics for sustainability to capture fundamental aspects of the system and apply those metrics to a geographic region over time to assess whether the system is moving toward or away from sustainability (Eason and Cabezas 2012; Heberling and Hopton 2012).
The four elements of the system EPA selected to measure were the human burden on the environment, economic well-being, energy flows, and system order. These properties were selected because they represent one or more component of the three pillars of sustainability. EPA used data specific to the region, or scaled to the region when necessary, and computed the four sustainability metrics over time. EPA published a number of journal articles and a report presenting the results of the research (Hopton and White 2012; Heberling et al. 2012; Campbell and Garmestani 2012; Eason and Cabezas
5


2012; U.S. EPA 2010). The four metrics used to assess sustainability in the
region are discussed below.
Ecological Footprint Analysis
To measure environmental burden, EPA conducted an Ecological Footprint Analysis (EFA) (Hopton and White 2012). EFA is a consumption-based metric that calculates the amount of productive land needed to support the population in a given area (Wackernagel and Rees 1996). EFA’s strength is that it is a relatively straightforward methodology with easy to understand results. EPA’s simplified approach to EFA used 35 variables for the 26-year period (1980-2005). Data were collected at the smallest geographic level available (i.e., municipality, county, region, state, country).
Emergy Analysis
To measure energy flows, EPA conducted an Emergy Analysis (Campbell and Garmestani 2012). Emergy Analysis is a method to track the quantity of embodied solar energy (unit = “solar emjoules” or “emergy”) in various items and economic activities/services (Tilley 2004; Baral and Bakshi 2010). It incorporates the accumulation of stored energy in a system as well as the movement of energy in the form of goods and services across boundaries. As a simplified example, energy from the sun reaches earth and is stored in photosynthetic organisms (both living and in the form of fossil fuels (i.e., coal, oil, and natural gas)). These energy reserves are mined and utilized for work. Work performed, which requires a certain quantity of coal, could also be thought of in terms of the
6


quantity of solar radiation needed to produce that quantity of coal (Odum 1971). This tracking of energy flows can be done with fossil fuels as well as with other goods and services in the economy. The accounting of these flows of energy, and human and natural resources needed to provide these goods/services to the economy, is an Emergy Analysis (Tilley 2004). The EPA’s Emergy Analysis considered emergy flows in the form of renewable and non-renewable energy, import and export of goods and services, and other key materials exchanges (e.g., agricultural chemicals, building materials etc.).
Green Net Regional Product
The idea that sustainability can be measured by GDP alone is now understood to be short sighted, and Green Net Regional Product (GNPR) adds to GDP by capturing the externalities associated with natural resource depletion through alternative-valuation methods (Pezzey et al. 2006; Mota et al. 2010; Heberling et al. 2012). GNRP is an environmental economic metric that adjusts the NRP to account for natural resource depletion in dollars. This process assumes that man-made-capital cannot simply be substituted for natural capital, but that the loss of natural capital reduces the utility for future generations (Lovins and Hawken 2007).
Fisher Information
Fisher Information (FI) comes from information theory, and measures the order or predictability of data (Cabezas and Karunanithi 2008; Cabezas and Fath 2002). It is a method of quantifying the information content in data that correlates
7


with order in the system dynamics. This is important because order is a marker of well-functioning natural and human systems. FI, as a sustainability metric, requires tracking of multiple important system variables simultaneously over time. To do this, a MATLAB program was developed by EPA (Eason and Cabezas 2012) to analyze data and detect whether the system is stable. EPA included 53 variables that were each classified into one of six categories: Consumption, Demographic, Energy, Environment, Land, and Production.
Present Work
This present research, which was funded by EPA Office of Research and Development, builds on the previous work performed in SLV. It utilizes the same geographic/political boundaries as the previous EPA study and refers to the region simply as SLV (Figure 1). For the purpose of data collection, the complete counties of Saguache, Rio Grande, Mineral, Hinsdale, Alamosa, Conejos, and Costilla were considered. SLV is roughly 21,000 km2 in area, it has an average elevation at the valley floor of 2,300 m (7,600 ft.), and the high desert receives an average of 222 mm (9 in.) of precipitation annually (Figure 2) (PRISM Climate Group 2016).
8


SAGUACHE
• Moffat
J'hinsdale
Cjenter
MINERAL
ALAMOSA
RIO GRANDE
Alamosa
# Towns Highways
L _i Counb«s >
Bureau of Land Management Bureau of Reclamation Indian Reservation
Military Reservation and Corps of Engineers National Grasslands National Park Service Other Federal Prwate State
State, County. City; Widlife, Park , and Outdoor Recreation Areas US Fish and WSdiife Service US Forest Service
t'-/’wA h- 4=£85j—V 1 ,a Jarth
CONEJOS w / N 1 -f
ivjaiiaag>a #Sail LUIS r~ff— r
© 1 Antonito1, T *
Figure 1: The San Luis Valley. Whole counties were always considered for data collection even though parts of the county may fall outside of the Upper Rio Grande river basin geographic boundary.
9


Figure 2: Average monthly precipitation (1980 - 2010) during the growing season (May to August) in SLV (PRISM Climate Group 2016)*.
*The units are in inches of precipitation (1 inch of rain in an area is enough to evenly cover the ground in that area with a layer of water 1 inch deep). (1 inch = 25.4 mm)
Agriculture is the major economic sector in the region. It is an ideal location for growing potatoes, barley (much of the barley for Coors Brewing Company ® beer is grown here), and alfalfa. An interesting note, the region was the first place in North America to grow quinoa for commercial use due to its Andean-like climate (Thier 2010). Though the 7 county region spans more than
10


21,000 km2, the population consists of only 50,000 people (U.S. Census Bureau 2012). SLV has the oldest town in the state (San Luis) (History Colorado 2015), the highest percentage of Hispanic residents (47%) in the state (SLV Development Resources Group 2013), and is home to some of the poorest counties in Colorado (U.S. Census Bureau 2011). Juxtaposed to the poverty there is prosperity for some potato, grain, and alfalfa farmers who are able to utilize the underlying aquifer to irrigate crops for export. More details on the region are presented in subsequent chapters.
In this research, two specific sustainability indicators were added to the suite of metrics developed by EPA: a greenhouse gas (GHG) accounting indicator and a consumptive water use indicator. GHG emissions are an indicator of global climate change, which is considered one of the most pressing human rights, security, and environmental issues of our time (Gosling and Arnell 2016; Barnett 2003; Douglas et al. 2012; Stallworthy 2017). Three of the EPA metrics (EFA, GNRP, and FI) relied on scaled down state level GHG emissions for their analysis because region specific data were lacking. The agricultural sector alone accounts for >10% of global anthropogenic GHG emissions (IPCC 2014). GHG accounting in rural agricultural regions is key for developing region specific mitigation strategies to reduce global GHG emissions (see Chapter II). Water use accounting is also critical for agricultural regions. In these regions the economy rests on water availability, and in the arid SLV, understanding water use and ways to reduce water consumption in critical for sustainability (Gibson
11


and et.al. 2015). Drought and overuse has threatened the regions underlying aquifer, which is relied on for the majority of the irrigation water for crops (CDWR 2015a) (see Chapter III). Carbon and water are also tied together for reasons including: 1) increased drought and water shortages have been linked to climate change (Malcolm et al. 2012; Calzadilla et al. 2013), 2) pumping water (including irrigation water) requires energy which in turn results in more GHG emissions from electricity production (Hussey and Pittock 2012; Rothausen and Conway 2011), and 3) improving soil health leads to greater water holding capacity of the soil (which reduces irrigation requirements (Altieri 1999)), and increasing soil carbon, which is a marker of soil health, can reduce atmospheric GHGs (Powlson et al. 2011).
The GHG emissions and water use indictors developed for the region were done in collaboration with local experts in SLV. In addition, this research proposes a framework for engaging rural agricultural communities around issues of sustainability, which includes a methodology for calculating sustainability indicators, over time, and developing some likely future scenarios using the indicators (see Chapter IV). A goal of this research is to empower the community to make informed decisions and develop with them tools for sustainable management. The only way to know where you are going is to understand where you have been (Santayana 1905; Ives and Boatwright 1999). This is true in life for a single person and it speaks truth to planning and policymaking for our society and its relationship with natural systems that support society. The
12


following sections provide an outline describing the content of this dissertation.
Arrangement of the Dissertation
Chapter I (current chapter) introduces the work and provides the reasoning, need, and purpose for doing it.
Chapter II provides an overview of the need for and importance of GHG accounting and details the methods and results of the GHG accounting indicator model. Climate change, which has been exacerbated by human induced GHG emissions, may disproportionately impact the agricultural sector (Malcolm et al. 2012). Here, a region-specific GHG accounting indicator is presented for agricultural regions, an often-overlooked area.
Chapter III provides an overview of the importance of consumptive water use accounting and details the methods and results of the consumptive water use indicator model. Issues surrounding water use are particularly critical to SLV because of limited rainfall and a depleting aquifer. Moreover, the state has overappropriated water from rivers and streams, as well as groundwater resources (Poppleton 2013; Fryar 2008). Because water is such an important resource to this agricultural economy, it is crucial to have a clear measure of how different practices affect water resources in this region. The SLV community has been grappling with sustainable water management for decades, and this research presents a new perspective on water consumption in the region.
Chapter IV details the experience of engagement with the local community over the course of the research. During the process of developing the GHG
13


accounting and consumptive water use indicators, there were data that were not readily available. To find these data, a number of in-person visits were made to the region. The purpose was to meet with people from local government, farmers and ranchers, and others stakeholders who possess information and perspectives needed for this research. Through meeting with local stakeholders, it became clear that developing a formal community engagement process would be key for the work. This resulted in the development of a Community Advisory Board (CAB). In addition to satisfying specific data needs, the CAB allowed for a forum to transition the resulting tools to the community, which was requested by the EPA (the grantor of this research). To transition the methodology and tools, a group of stakeholders attended a series of seminars where they were taught how to compute and use the original four metrics and additional GHG and consumptive water use indicator models. A timeline of community engagement (Figure 3) is below, and the details of the process are outlined in Chapter IV.
14


Community Engagement Timeline
Figure 3: Community engagement timeline in the SLV during the course of this research.
Chapter V describes the future scenario modeling phase of the research. The ability for regions to develop plans and implement actions toward sustainability hinges on a clear historic accounting and baseline scenario. Once a region-specific baseline was established in SLV, which included updating the four metrics developed by EPA from 2005 to 2010 as well as developing the GHG accounting indicator (Chapter II) and the consumptive water use indicator (Chapter III), the research moved into the future scenario modeling phase. The future scenarios selected for proof of concept modeling were conceptualized with the CAB. The future scenarios were modeled using the GHG accounting and consumptive water use indicators to produce a picture of several potential futures in the region. These proof of concept future scenarios were developed to a) advance the science of sustainability modeling and b) add to a foundation of knowledge for the local community.
15


The final chapter, Chapter VI, summarizes some of the major findings and outcomes from the research. A concise review of each of the chapter’s main results is presented and discussed. In addition, areas for potential future work beyond the scope of this research are offered.
16


CHAPTER II
GREENHOUSE GAS ACCOUNTING
Greenhouse Gas Accounting of Rural Agrarian Regions: The Case of
San Luis Valley
Abstract
Rural regions, with a dominant agricultural economic base, have a vastly different greenhouse gas emissions profile than urban regions and hence require a unique accounting method. This chapter presents a greenhouse gas (GHG) inventorying methodology tailored specifically for rural agricultural regions. The methodology was applied to San Luis Valley (SLV) in south central Colorado with intent to establish a clear emissions baseline and to analyze, in fine detail, the regions emissions profile. The results show that SLV has an annual per capita emission rate of 30.5 MT C02 equivalents (C02e) whereas the average for the United States is about 21 MT C02e. The higher per capita emissions can be attributed to the production of agricultural goods and services that are exported rather than consumed. Because per capita emissions might not paint an accurate picture for export based economies, data were recalibrated on a per dollar GDP basis. On this basis, SLV emissions are almost twice that of the national average, indicating that, with all things being equal, agricultural activities contribute disproportionately towards GHG emissions. Through a detailed quantitative analysis, the results show that SLV, with its significant solar resource, has the potential to offset much or all of their emissions. The findings
17


from this work offer useful insights for local stakeholders to develop plans and implement policies towards GHG mitigation.
Introduction
Awareness in the global scientific community and in the general public about greenhouse gas (GHG) inventories and their benefits for policymakers in addressing climate change has greatly increased in the last decade (Obama 2013, 2009). In addition, communities and governments play an important role in reducing GHG emissions as they have broad influence over activities that result in significant direct and indirect GHG emissions within their boundary and jurisdiction. GHG inventories create a baseline that can be used to identify sectors, sources, and activities responsible for GHG emissions, assess relative contributions of emission sources, establish local climate action plans and policies, quantify benefits of activities that reduce emissions, and foster informed communication with stakeholders (U.S. EPA 2015a). Previous efforts to baseline communities, such as those led by the Intergovernmental Panel on Climate Change (IPCC), had primarily focused at the national level. However, with the leadership of organizations such as the International Council for Local Environmental Initiatives (ICLEI) and the World Resources Institute (WRI) there has been a push to look more specifically at cities and urban areas because of their dense population and high consumption (Ramaswami et al. 2008; IPCC 2006; Arikan et al. 2012). This work is one such attempt to conduct a GHG baseline analysis on a more local level, but focused on a rural agrarian region.
18


Typical city-scale GHG inventories account for residential, commercial/industrial energy use (primarily in-boundary buildings/facilities), and transportation. In addition, hybrid methods account for cross-boundary contributions associated with urban material consumption (e.g., cement, fuel, food, etc.) and transportation (surface and air). Different accounting approaches, ranging from pure geographic production-based accounting, pure consumption-based accounting to hybrid geographic - plus key infrastructure supply chain accounting have been developed at the city scale (Ramaswami et al. 2008; Lenzen and Crawford 2009; Hillman and Ramaswami 2010; Dodman 2009). In addition to accounting for in-boundary emissions allocation, issues related to trans-boundary emissions and life cycle supply chain emissions have been addressed (Ramaswami et al. 2008). These inventories have been used for future planning in cities addressing water, energy, and material needs of urban communities.
Whereas these studies have been influential for cities and nations in making plans and implementing actions towards reducing carbon emissions, very little work has been done on explicit accounting of GHG emissions for rural regions. According to the IPCC, land use accounts for more than 24% of the world’s overall GHG emissions, and agricultural production is the largest contributor to that sector (IPCC 2013). The U.S. Department of Agriculture (USDA) also recently published a report that shows that the agricultural sector will be one of the hardest affected by a changing climate (Malcolm et al. 2012).
19


Moreover, rural agricultural regions, although small in population, have a disproportionate influence on global GHG emissions. This is a key reason why engaging with an agricultural community on GHG inventories is so important.
This rural GHG inventory examined the agricultural region known as San Luis Valley (SLV) in south central Colorado. The U.S. Environmental Protection Agency selected this region as a pilot study to develop and test sustainability metrics partly because of its isolated geography and well defined agricultural economy (Heberling and Hopton 2012; Heberling et al. 2012; Hopton and White 2012; Campbell and Garmestani 2012; Eason and Cabezas 2012). The SLV is a 100-mile long and 60-mile wide (21,000 km2) upland agricultural valley surrounded by the 14,000-foot (4,400 m) peaks of the Sangre de Christos to the East and the expansive San Juan wilderness area to the West. The valley floor sits at 7,500 ft. (2,300 m) and is an ideal location for growing potatoes (it is the second largest potato producing area in the United States next to Idaho) and barley (the majority of Coors Brewing Company ® barley is grown here).
The sphere of influence of local government is typically limited to its geographic jurisdiction. Therefore, urban policymakers have limited ability to direct meaningful improvements to their emission profile in categories such as food, because they are consumers of food and not the producers. Therefore, engaging rural food producing regions through local carbon accounting metrics is key for fostering policy change. Because the emission profiles in agricultural regions are widely different than those in urban regions, there is a great need to
20


develop context specific accounting approaches for rural regions that capture activities such as agricultural energy demand, livestock raising, and soil GHG fluxes. Further, in view of the wide variability in agricultural practices (e.g., tilling practices, crop rotations, manure management, groundwater vs. surface water use), and soil and climatic conditions, these inventories need to be based on local context specific, bottom-up emission factors derived from regional data (Fraser et al. 2006). With a robust and descriptive region specific baseline GFIG accounting, local decision-makers can begin the process of developing plans and implementing actions towards GHG emissions reductions.
Methodology
This study presents a novel method of rural agricultural region GHG accounting and data reporting. The methodology for developing and conducting this inventory uses the IPCC 2006 release of GHG inventory for nations as well as other GHG inventories in the field for structure and inclusions (IPCC 2006; WRI 2004; Ramaswami et al. 2008). The GHGs considered in this study are C02, CH4, and N20 and are presented as carbon dioxide equivalents (C02e) based on the IPCC 2013 report (IPCC 2013). The sectors of interest, consistent with the IPCC, are Energy and Buildings, Transportation, Agriculture, Land Use, and Waste. Because manufacturing makes up only about 1% of the economy in the region (SLV Development Resources Group 2013), the Industrial Processes and Product Use sector was excluded from this study. In addition to in-boundary emissions associated with the sectors described above, we also looked at
21


upstream life cycle emissions when relevant. Also, producing an inventory that was reproducible not just by the scientific community, but by our local community partners in the region was essential. Six months were spent performing a stakeholder analysis based on principles found in Community Based Participatory Research (CBPR) that led to the formation of a Community Advisory Board (CAB) (Israel et al. 1998). The CAB, which was composed of stakeholders from the water user community (farmers and ranchers), local government, conservation groups, federal land managers and others from across the region, provided insights into the unique nature of the area as well as aided in data collection and provided critical review. Moreover, because of the explicit goal of transferring this indicator model and methodology to the community, it was attempted, whenever possible, to use publicly available data such as census data (U.S. Census Bureau), National Agricultural Statistics (NASS), Colorado Department of Transportation (CDOT), and others that rely as little as possible on complex computer models for the inventory. The information below shows how data were collected and the emission factors that were selected for each of the sectors. Each sector’s material flows, emission factors, and sources are summarized in Table 1.
Energy and Buildings Sector
Electricity: Annual electricity consumption in SLV was obtained from the two utility companies - Xcel Energy (Xcel) and San Luis Valley Rural Electric Cooperative (SLV-REC) - who supply electricity to the region. Total consumption
22


was 222,023 MWh for Xcel and 211,993 MWh for SLV-REC in 2012. Further,
each company provided the residential and commercial breakdown of the electricity consumption, and the amount of electricity used for agricultural irrigation. The life cycle GHG emission factors from Xcel and SLV-REC included both direct emissions from power plants as well as upstream emissions from the mining and handling of the raw resource, namely coal and natural gas. The direct emission factor for Tri State Generation and Transmission (Tri State), from whom SLV-REC purchases 99% of its electricity (Waudby 2015), was based on emission factors of individual power plants that provide electricity to Tri State (Tri-State 2015). These power plant specific emission factors were obtained from the EPA 2012 eGRID dataset (U.S. EPA 2012a). The electricity grid mix for SLV is 60% coal, 25% natural gas, and 15% renewable. The direct emissions of SLV-REC grid mix were determined to be 0.76 kgC02e/kWh, whereas the direct emission factor for Xcel was determined to be 0.75 kg C02e/kWh (Xcel 2013). Emissions upstream from the power plant due to mining and transport of coal and natural gas was estimated as 25% and 6% of the total emissions, respectively (Spath et al. 1999; Spath and Mann 2000; Heath et al. 2014; Whitaker et al. 2012). A combined average emission factor for the region was calculated as 0.83 kgC02e/KWh.
Natural Gas and Propane: Community-wide natural gas usage for residential and commercial purposes was directly obtained from Xcel, which is the sole natural gas supplier to SLV. The annual residential and commercial
23


natural gas consumption was 762 million MJ and 646 million MJ, respectively. Annual propane consumption was estimated for the region based on the number and square footage of homes in the region and the number of homes using propane for heat (U.S. EPA 2010). In addition, this estimated value was verified through contact with the major propane providers in the region. The life cycle emission factor for the use of natural gas and propane in a furnace, which includes both direct and upstream emissions, was 9.35 kg C02e/gallon and 7.03 kg C02e/gallon, respectively (Register 2009; Howarth et al. 2011; Heath et al. 2014). In 2012, SLV emissions from natural gas and propane were 119,000 MT C02e.
Agriculture and Land Use Sector
Livestock Emissions: Methane, a potent GHG, is produced as a byproduct of enteric fermentation (digestion) in livestock and the amount of methane that is released depends on the type of digestive tract, age, and weight of the animal as well as the quality and quantity of the feed consumed (IPCC 2006). SLV has significant livestock operations consisting mainly of cattle ranching with relatively small sheep and hog operations. For 2012, NASS reports head of cattle in the region as 83,000, head of sheep as 12,700, and head of hogs as 220 (U.S. Department of Agriculture 2015). IPCC suggests a regionalized tier 2 approach for calculating the emission factor for cattle and a generalized tier 1 approach for sheep, goats, and swine (IPCC 2006). To reflect the variation in emission rates among cattle, instead of using the North American
24


average of 53 kg CH4/head as reported by IPCC, a region specific emission factor was developed by first categorizing the herd into IPCC suggested subgroups. Then using local data for climate, feeding situation, age, size, and cattle subgroup and the IPCC model, a herd specific emission factor of 50 kg CH4/head was calculated (Whitten 2015). Following IPCC tier 1 approach, the global industrialized average emission factors of 5 kg CH4/head and 1 kgCH4/head for sheep and hogs, respectively, were used. For manure management, an IPCC tier 1 approach for each of the three animal types was used (IPCC 2006). Cattle manure management practices in the region produce 2 kg Chiphead, sheep management produces 0.2 kg CH4 / head, and hogs produce 13 kg CH4 / head according to the IPCC tier 1 approach. In 2012 livestock in the SLV were responsible for 145,000 MTC02e.
Soil Nitrous Oxide Emissions: The use of fertilizers and managed soils is a major contributor to GHG emissions globally. This section describes the estimation of N20 emissions from managed soil due to agricultural inputs of nitrogen (synthetic N fertilizers; N deposited by grazing animals, and decomposing crop residue). Because actual data were not available, average quantities of nitrogen applied per acre by crop type and crop rotation was estimated through in-depth interviews with two key agriculture consultants who work with local farmers in the Valley (Oniel 2015; Dillion 2015). Based on their inputs, the following average rate of nitrogen application was used in this study: 180 Ibs./acre for small grains after a potato crop, 80 Ibs./acre for small grains
25


after an alfalfa crop, 220 lbs./acre for continuous small grains, 11 Ibs./acre for alfalfa, and 185 Ibs./acre for potatoes after a small grain crop. Local data on N inputs through grazing animals were based on animal head counts and their feeding situation (IPCC 2006). The IPCC guidelines were followed to estimate GHG emissions from managed soils, and both direct N20 emissions from nitrification and denitrification of N inputs, as well as indirect N20 emissions due to N volatilization, leaching, and run off were accounted. Upstream emissions associated with the production of nitrogen fertilizer were obtained from Blonk Consultants (Zeist et al. 2012), and were found to be 4.0 kg C02e/kg N applied based on the North America average fertilizer mix. In 2012 there was an estimated total of 35 million lbs. of nitrogen applied, which in turn was responsible for 140,000 MT C02e emissions.
Agricultural Machinery: The on-farm fuel use from tractors, farm trucks, and processing equipment was estimated using data reported by NASS on dollars spent on agricultural fuels in the region (U.S. Department of Agriculture 2015) with all fuel consumption assumed to be diesel and consumed on the farm. The cost per gallon for diesel was based on the reported price for western states. All fuel was road tax-exempt red diesel, so the Colorado road tax was subtracted from the regular fuel price (U.S. EIA 2016). The life cycle emission factor for diesel fuel was obtained from Argonne National Laboratory’s Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation Model (GREET) (Argonne National Laboratory 2015). This model reports the pump-to-wheels
26


(PTW) emission factor (direct) for diesel as 9.18 kgC02e/gallon, and the wells-to-pump (WTP) emissions (upstream) as 2.3 kgC02e for both gasoline and diesel. In 2012, the Valley consumed 6.7 million gallons of farm diesel, which accounts for an annual emission of 77,000 MTC02e.
Transportation Sector
On-Road Cars and Trucks: Direct tailpipe emission from on-road transportation was calculated by multiplying vehicle miles travelled (VMT) in the region by the emission factors for gasoline and diesel fuel. The total VMT in the region was collected using Colorado Department of Transportation (CDOT) static highway reports (CDOT 2012), which provided annual VMT data by county divided up into two classes, lightweight personal vehicles (class 1-3) and larger commercial vehicles (class 4-13). However, these publicly available data capture only major roadways in the region, and do not include traffic through the minor county roads, which can be significant for rural regions. In order to include the smaller roads, region specific data were obtained directly from CDOT based on their traffic model to estimate “off grid” VMTs in the seven county region. All travel within the region’s boundary was allocated to the region because there was not a valid way of disassociating traffic that was just passing through, and based on the mountainous and isolated geography of the region, it was assume that pass through trips are minimal. The total lightweight personal vehicle traffic in the region in 2012 was estimated at 1.4 million VMT per day and the normalized personal VMT in the region yielded 30 VMT/person/day which is very close to the
27


national average of ~28 VMT/person/day (U.S. DOT 2012). The total commercial large truck (class 4 through 13) traffic for the region in 2012 was around 140,000 VMT per day. In order to estimate the weighted average fuel economy for the personal vehicle fleet in the region, vehicle registration data were obtained from the San Luis Valley Development Resources Group (SLVDRG) (SLV Development Resources Group 2013). The average fuel economy for lightweight personal vehicles (class 1-3) was estimated as 25.6 miles/gallon. For commercial large trucks (class 4-13) the national average fuel economy of 7.3 miles/gallon was used in this study (U.S. DOT 2015). The GREET model was again used to derive the life cycle emissions factors for gasoline and diesel. The pump-to-wheels (PTW) emission factor (direct) for gasoline was 8.71 kgC02e/gallon and for diesel it was 9.18 kgC02e/gallon. The wells-to-pump (WTP) emissions (upstream) were 2.3 kgC02e for both gasoline and diesel (Argonne National Laboratory 2015). The on-road emissions from lightweight personal vehicles in 2012 were estimated as 230,000 MTC02e and the emissions from commercial large trucks were 79,000 MTC02e.
Waste Sector
Landfill emissions: There are two major solid waste facilities in the region: SLV Regional Landfill and Saguache County Landfill and Recycling Center. Data of municipal solid waste (MSW) generated in the region were obtained from the Colorado Department of Public Health and Environment (CDPHE) and were reported as 63 thousand tons (1.4 tons per capita) in 2012 (CDPHE 2015).
28


Landfill emissions were estimated using EPA’s Waste Reduction Model (WARM) (U.S. EPA 2015b). SLV regional, the larger of the two sites, provided their waste composition broken into three categories: household (87%), construction (11%), and other (2%). WARM required further composition detail for which the U.S. average data from the EPA was used (U.S. EPA 2013b). The transportation portion of the analysis was omitted from the WARM model because those emissions would have been accounted for in the transportation section of this GHG accounting. The results from WARM show landfill emissions of 56,000 MTC02e that can be attributed to solid waste generated.
Waste Water Treatment: The quantity of emissions from wastewater depends on the treatment type and, in SLV, was either rural septic or centralized aerobic treatment. Data needed were the number of people living in the region and the percent of those people living in cities and in rural areas. It was assumed that all city population used centralized aerobic treatment and all rural population used anaerobic septic treatment. This is a critical distinction because anaerobic treatment produces higher CH4 emissions according to IPCC. In 2012, of the nearly 47,000 people living in the Valley, 50% were in cities and 50% resided in rural areas (U.S. Census Bureau 2012). Using these data, an emission factor of 0.24 MT CH4/MT BOD was developed for the Valley. Biological Oxygen Demand (BOD) in the wastewater was estimated using the average BOD concentration per person for North America (85g/person/day)
29


(IPCC 2006). The total treatment demand in SLV was estimated at 1,482 MT
BOD and the total associated annual emissions were 12,000 MTC02e.
Solar Credit
Data on utility scale solar energy production were obtained for this study and allocated in the form of carbon credits to the region. The total solar capacity in SLV in 2012 was 87 MW, which produced 198,363 MWh of electricity (U.S. EIA 2015). The upstream emission factors for the production of the power plants were 0.04 kgC02e/KWh for photovoltaic (PV) and 0.02 kgC02e/KWh for concentrated solar power (CSP) (Burkhardt et al. 2011; Hsu et al. 2012; Kim et al. 2012). A carbon credit was allocated to the region for the total amount of utility-scale solar produced based on the avoided emissions.
30


Table 1: Material flows, emission factors, and total GHG emissions in the San Luis Valley.
Regional material or energy flow (MFA) Data year MFA data source GHG emission factor (use phase) GHG emission factor (upstream) EF data source Total GHG emitted = MFA x EF (MTC02e)
Energy and Buildings
Electricity 434,016 MWh 2012 (Dallinger 2015; Waudby 2015) 0.76 kgC02e/kWh 0.075 kgC02e/kWh (Heath et al. 2014; Xcel 2013) 362,000
Solar Credit 198,363 MWh 2012 (U.S. EIA 2015) -0.83 kgC02e/KWh 0.04kgC02e/ KWh (PV) 0.02 kgC02e/ KWh (CSP) (Burkhardt et al. 2011; Hsu et al. 2012) -161,000
Natural Gas 1,220 MMJ 2012 (Dallinger 2015) 0.05 kgC02e/MJ 0.014 kg C02e/MJ (Heath et al. 2014) 80,000
Propane 4.9 million gal 2012 (U.S. EPA 2010) 5.63 kgC02e / gallon 2.15 kgC02e/gallon (U.S. EPA 2010; Frischknecht and Rebitzer 2005; Ecoinvent v2.0 2007) 39,000
Agriculture and Land Use Sector
(Oniel
Soil GHG emissions from crops 13,009 MT synthetic N fertilizer 2010 2015; Dillion 2015; CDWR 2015b) IPCC tier 1 methodology 4.00 kgC02e / kg N (IPCC 2006; Zeist et al. 2012) 210,000
Soil GHG emissions from livestock 3,662 MT N applied by livestock 2012 (U.S. Departme nt of Agriculture 2015) IPCC tier 1 (IPCC 2006) 132,000
methodology
Enteric 83,144 cattle 12,711 2012 (U.S. Departme nt of Agriculture 2015) 50 kgCH4 /cattle 5 kgCH4 /sheep 1 kgCH4 /hog (IPCC 2006) 140,000
Fermentation sheep 221 hogs
Manure 83,144 cattle 12,711 2012 NASS(U.S Departme 2 kgCH4 /cattle 5 kgCH4 /sheep 1 kgCH4 /hog (IPCC 2006) 5,600
Management sheep 221 hogs nt of Agriculture 2015)
Agricultural Machinery 6.7 million gal 2012 (U.S. Departme nt of Agriculture 2015) 9.18 kg C02e / gallon diesel 2.3 kgC02e / gallon diesel (Argonne National Laboratory 2015) 78,000
Transportation Sector
Personal Vehicles 17.9 million gal gasoline 3.1 million 2012 (CDOT 2012) 8.71 kg C02e / gal gasoline 9.18 kg C02e / 2.3 kgC02e / gal gasoline or diesel (Argonne National Laboratory 234,000
31


gal diesel gal diesel 2015)
Large Trucks 6.9 million gal diesel 2012 79,000
Waste Sector
Landfills 63,414 tons 2012 (CDPHE 0.89 MTC02e / (U.S. EPA 2015b, 56,000
2015) ton MSW
2013b)
745 MT BOD (U.S.
rural septic Census 0.24 MT CH4 /
Waste Water 737 MT BOD 2012 Bureau MT BOD (IPCC 2006) 12,000
centralized 2012)
Results and Discussion Total GHG emissions in SLV
A consolidated summary of all material and energy flows, emission factors, and data sources used in this study can be seen in Table 1. For the year 2012, it was estimated that 1.27 million metric tons of carbon dioxide equivalents (MMT C02e) emissions could be attributed to SLV, which includes a 0.16 million metric ton carbon credit for the utility-scale solar in the region. The region hosts four utility-scale solar facilities with a combined total of 87 MW of capacity. SLV has a per capita emission of roughly 30 MT C02e whereas the average per capita emissions for the United States is about 21 MT C02e (U.S. EPA 2013a). The higher per capita emissions in SLV reflect that a significant amount of economic activity in the region is associated with production of agricultural goods and services that are exported rather than consumed. Electricity is the highest contributor to total emissions at 0.36 MMT C02e, followed by passenger vehicles at 0.23 MMT C02e, and soil GHG emissions (crops) at 0.21 MMT C02e (Figure 4). Soil GHG emissions, typically not accounted in city-scale GHG inventories,
32


from agricultural activities and livestock raising contributes ~20% of total GHG emissions. Similarly, methane emissions from livestock contribute ~10% to the total GHG emissions of the SLV.
4.00EKIS
Upstream 14% ^
3.00EK)S
2.00EH)S
Direct
86%
O
l.OOEtOS
O.OOE+OO
y
f J/• o'
s* y y y y // ,y y y y ,y
* ^ ^ y %°
oro
y
-2.00EM)S e>
/ *

(S
* Direct • Upstream Solar Credit
Figure 4: The GHG accounting for each of the major emission categories in San Luis Valley. This includes both direct emissions and upstream emissions.
Upstream Emissions
Direct emissions within the Valley account for about 86% of total emissions, and the remaining 14% are upstream emissions that occur outside of the geographic boundary (Figure 4). However, note that with regards to electricity, the GHG emissions from power plants are categorized under direct emissions but the power plants that service the region are located outside of the
33


geographical boundaries. The inclusion of life cycle analysis of upstream emissions coupled with direct use emissions can provide a more comprehensive emission profile for agricultural regions. Upstream emissions from fertilizer, which accounts for 3.6% of the SLV’s total emissions inventory, are the largest of any of the upstream emissions. Excluding upstream emissions may cause policymakers to focus on other sectors, when in fact the way soils are managed and synthetic fertilizer usage could be a more impactful place to focus. If an agricultural inventory does not include upstream emissions from fertilizer they may miss a crucial piece of the picture.
Carbon Credit
The carbon credit was based on a robust accounting of electricity sources including analyzing the utility-scale solar energy and attributing carbon credits based on avoided emissions. This exercise can provide clarity on the current and potential ability of the region to offset its emissions with renewable energy sources. The utility scale solar production in the region produced 198,000 MWh in 2012, which is 44% of the region’s electricity consumption and is equal to 0.16 million metric tons of avoided C02e emissions (Figure 4). The current credit in this study reduced the overall GHG emissions from the region by ~9%. With these types of data, policymakers can begin assessing future scenarios based on potential emissions reduction targets. The Bureau of Land Management (BLM) in SLV is currently proposing four Solar Energy Zones (SEZ) in the region. This is part of a larger federal initiative called the Solar Energy Program, which
34


proposes that BLM land should relax the barriers to private solar development on public land (DOE/BLM 2012). A 50% build out of the proposed BLM solar energy plan (700 MW) would see a 1.5 MMMT C02e credit, which was more than the total emissions from the region in 2012.
Emissions by Category
Activities responsible for GHG emissions were classified into the four broad categories (Figure 5). The associated calculation involved dividing electricity between agriculture, residential and commercial. In addition, agriculture and livestock soil GHG emissions, methane emissions from livestock and manure management, and agricultural machinery emissions were allocated to the agriculture category and passenger vehicles and large truck transport emissions were assigned to transportation category. The residential category included energy for buildings, wastewater treatment, and landfill emissions. The results show that agriculture is the largest contributor of GHG emissions in the region (~47%) followed by transportation (~20%), residential (~20%), and commercial (~12%). Of the agricultural category, ~43% is associated with soil, while GHG emissions and electricity for irrigation contributes ~15% of agricultural emissions. Passenger vehicle transport (class 1-3) is responsible for the majority (~ 74.5%) of transportation emissions and large trucks account for the remaining (~25.5%) emissions. Note that the truck transportation includes truck transport for agricultural products. Building energy (electricity, natural gas, and propane) consumption dominates residential sector emissions (~75%).
35


Airplanes Trains Agricultural
Trucks (Class—^ Q.3%^ 0.0% Electricity
passenger vehicles (class
1-3)
16.4%
Electricity
commercial
9.0%
Natural Gas commercial 2.6%
waste water 0.8%
landfill
3.9%
electricity
9.0%
Residential
3.0%
Propane Upstream 9 2% 2.7% 3.6%
Figure 5: The breakdown of GHG emission contributions of the four major categories averaged between 2006 and 2012.
Comparison
These results are now compared with GHG inventories related to different
geographical scales (city, state, and country). Because the four scales
examined, regional hinterland (SLV), city (Denver), state (Colorado), and
national, all have very different population densities and economies, it allowed for
an interesting comparison in terms of emissions per GDP in addition to emission
per capita. The SLV and Denver GHG emissions data include upstream
emissions, whereas the inventories of Colorado and USA do not (Table 2).
Further, because there are certain methodological differences in each of these
studies, fossil energy use (use phase) is also presented for comparison
purposes. The results show that SLV has the lowest per capita emissions
36


amongst the four spatial scales, when considering only fossil energy use (16 MT C02e/capita) (Table 2). This is because SLV has less commercial and industrial activity in comparison to urban areas. In Denver, for example, energy used for commercial buildings is about 9.7 MT C02e/capita while the corresponding number in SLV is 3.2 MT C02e/capita. This difference (6.5 MT C02e/capita) is partially offset by agricultural energy consumption (pumping energy and fuel for machinery) in SLV, which was about 3.4 MT C02e/capita. On the other hand, SLV has the highest per capita emissions when we consider the total GHG emissions (26 MT C02e/capita). This is attributed to the significant level of nonenergy related agricultural emissions (such soil N20 emissions) associated with SLV, which is not present in Denver.
However, per capita comparisons across diverse locales (such as rural region vs. urban region) provide very limited insights, because these different locales provide widely varying types of functions (food production vs. services). Therefore, looking at emissions per dollar GDP as opposed to emissions per capita would be more meaningful. SLV has the highest per GDP emissions by far of all the areas at 670-780 kg C02e/$ GDP. When considering only fossil energy use, SLV has a lower per capita GHG emissions compared to Denver and Colorado by 17% and 18% respectively. However, if we consider GHG emissions/$ GDP as the comparison we see that SLV’s emissions/$ GDP exceed Denver’s and Colorado’s by 30% and 6%, respectively. Therefore, when comparing across different spatial scales, we conclude SLV (an agricultural
37


driven economy) is much less efficient, on a GHG emission basis, at producing capital than Denver (a consumer based economy). This is not surprising because the type of economic activities in SLV and Denver are vastly different. For example, >50% of the emissions from SLV are due to exported agricultural products, and cities (like Denver) are predominantly consumption-centric and have service oriented economies. Note that the above analysis is based on energy use alone and does not include emissions from agricultural land use (e.g., soil GHG emissions, livestock methane emissions etc.). This type of analysis can assist state-level decision-makers to prioritize resources available for GHG mitigation purposes.
38


Table 2: Comparing the SLV inventory with GHG inventories at various spatial scales.
San Luis Valley Denver(Ramaswami et al. 2008) Colorado (Arnold et al. 2014) USA (U.S. EPA 2013a)
Total Inventory (MMT C02e) 1.2-1.4* 11 -15 130 6,500
Fossil Energy Use (MMT C02e) 0.7 11 97 5,100
Total Inventory (MT C02e/capita) 26 - 30* 19-25 26 21
Fossil Energy Use (MT C02e/capita) 16 19 19 16
Total inventory (kg C02e/$ GDP) 670 - 780* 210-280 470 400
Fossil Energy Use (kg C02e/$ GDP) 370 200 350 310
Data for GDP of the San Luis Valley and Denver were obtained from IMPLAN, working with the Business Resource System of the University of Colorado Boulder (IMPLAN 2012), while data for GDP of Colorado and the U.S. came from the Bureau of Economic Analysis (BEA 2012). *The high and low estimate (i.e., range) of the total inventory for SLV and Denver are showing results with and without upstream emissions.
Overall it was found that emissions from electricity production dominate
the GHG emissions profile of all the locales, at roughly 25% of their GHG
emissions (Arnold et al. 2014; U.S. EPA 2013a). This information may lead
policymakers to wrongly conclude that all these vastly different regions would
benefit from the same types of policies to reduce emissions from electricity. For
example a state mandate to insulate homes or increase efficiency as a means of
reducing electricity usage would make sense in urban areas where buildings
account for ~50% of the emissions (Hillman and Ramaswami 2010; Ramaswami
et al. 2008), but an entirely different set of policies might be better in rural areas,
39


such as SLV. Improvements to agricultural practices (i.e., agro-ecological practices; low input farming (Altieri 1999)) and rotation cropping patterns (such as cover cropping) could have a much larger impact than trying to reduce energy consumption in the residential or commercial sectors. Therefore, region specific inventories with highly disaggregated data as presented in this work are key for local policymakers.
Uncertainty
This study utilizes large amount of data that have underlying uncertainty associated with them. There exists uncertainty in these raw data (material and energy flows) such as VMTs, fertilizer application rates, and liquid fuels consumed, as well as in the emission factors associated with both direct emissions like wastewater treatment, landfills, natural gas burning, and upstream emissions like fuel refining. Please note this specific example of uncertainty that emerged during data disaggregation regarding electricity consumption for irrigation. Though all electricity consumption in the region was accurately accounted for, the irrigation component provided by Xcel Energy was most likely an underestimate due to the restrictions on the type of data they were allowed to provide under privacy laws. Thus, the irrigation portion of GHG emissions is likely a conservative estimate. Though a robust accounting was presented, it is important to understand the results of the study within the overall context of possible data uncertainty. Unfortunately, information on this uncertainty is rarely available for such data.
40


Finally, it would be pertinent to note that GHG accounting can also be based on regional consumption (usually referred to as carbon footprint) (Peters 2008; Larsen and Hertwich 2009; Davis et al. 2010; Wiedmann 2009). As discussed earlier, the type of accounting (consumption or production based) used should depend on the utility of the inventory to provide data and insight towards mitigation strategies, which in turn is related to the sphere of influence of the relevant local governments. Therefore, for regions that are predominantly consumption oriented (e.g., cities) carbon footprinting might be appropriate because they can help identify consumption related mitigation strategies, whereas for production centric regions (e.g., SLV), a hybrid approach such as the one proposed here will offer the most value.
41


CHAPTER III
CONSUMPTIVE WATER USE
Consumptive Water Use Analysis of Upper Rio Grande Basin in
Southern Colorado
Abstract
Water resource management and governance at the river basin scale is critical for the sustainable development of rural agrarian systems in the West. This research applies a consumptive water use analysis, inspired by the Water Footprint methodology, to the Upper Rio Grande Basin (RGB), or San Luis Valley (SLV), in south-central Colorado. The region is characterized by water stress, high desert conditions, declining land health, and a depleting water table. Region-specific data and models were utilized to analyze the consumptive water use of SLV. The study reveals that, on average, SLV experiences three months of water shortage per year due to the unsustainable extraction of groundwater (GW). Results show that agriculture accounts for 77% of overall water consumption and it relies heavily on an aquifer (about 50% of agricultural consumption) that is being depleted over time. It was found that, even though potato cultivation provides the most efficient conversion of groundwater resources into economic value (m3 GW/$) in this region, it relies predominantly (81%) on the aquifer for its water supply. However, cattle, another important agricultural commodity produced in the region, provide good economic value, but rely significantly less on the aquifer (30%) for water needs. The results from this chapter are timely to the region’s community, which is currently in the process of
42


developing strategies for sustainable water management.
Introduction
Water and energy use are two important resource-use issues of the century as there are limits to the extent that humanity can continue to increase its appropriation of fossil fuel resources and freshwater from the natural environment (Ridoutt and Pfister 2010). Further, water and energy interdependencies and supply constraints have been recognized by researchers to pose significant risks that can potentially and unintentionally shift overall impacts geographically and temporally (Fulton and Cooley 2015; Hussey and Pittock 2012). In this context, proper quantification of fresh water resources needed for the environment as well as human-made products (e.g., agricultural produce) and processes (e.g., power plant cooling) across nations and sub-regions is the first step towards addressing water use issues (Hussey and Pittock 2012; Hoekstra et al. 2011). Water Footprint (WF) (Chapagain and Hoekstra 2004; Hoekstra et al. 2011; Mekonnen and Hoekstra 2011a) is an approach that provides a useful framework to quantify fresh water usage along the entire supply chain (Hoekstra and Hung 2002). WF has been used to assess both direct consumptive use and indirect embodied water content of individual products such as milk and clothes (Chapagain and Hoekstra 2004), as well as for populations of nations and planet (Gerbens-Leenes et al. 2013; Mekonnen and Hoekstra 2011b). However, this chapter, which pulls certain concepts from WF, focuses only on consumptive water use of a regional system with the aim to provide data and analysis for both
43


consumers of water from the river-basin as well as water managers (Zeng et al. 2012; Dumont et al. 2013; Dong et al. 2013; Appasamy et al. 1999). Note that this analysis is based on water consumption as opposed to the more conventional water withdrawal measure, in that a consumptive use analysis recognizes the interconnectedness of the hydrologic cycle and appropriately allocates evaporation, return flows, irrigation inefficiencies, and other withdrawn but “unused” water (Hoekstra et al. 2011).
SLV is a high altitude agricultural valley in south-central Colorado and encompasses roughly 21,000 km2 (8,000 square miles). It has an average elevation at the valley floor of 2,300 m (7,600 ft.) above sea level with an average annual precipitation of 222 mm (9 in) (PRISM Climate Group 2016). It is a snowmelt driven hydrologic system, with the majority of water into the SLV coming from river and stream flows from the surrounding mountain ranges. Hay production and pasturelands utilize some of the snowmelt, but much of it flows into the massive aquifer that underlies the entire valley, a portion of which is pumped back for irrigation. It is also important to note that not all of the water that flows into SLV is “owned” by the region. Two major interstate compacts, the Rio Grande Compact (1938) (Hinderlinder et al. 1938) and the Amended Costilla Creek Compact (1963) (Whitten and Reynolds 1963), require SLV to allocate water to downstream states based on the annual snowmelt. This precious water supply is what allows this high desert region to be one of the nation’s key producers of potatoes, alfalfa, and barley. The region’s limited population,
44


isolated location, and the fact that its natural watershed boundaries coincide
somewhat with its political boundaries are factors that make this region an ideal study area (U.S. EPA 2010).
The impacts of groundwater pumping on surface water flows in the region have been recognized, and the extreme drought of 2002 accelerated the situation to the forefront. The Colorado Division of Water Resources (CDWR), the department responsible for administering water rights, monitoring water use and stream flows, and issuing permits for wells and other water infrastructure projects in the state, developed a robust groundwater (CDWR 2015b) model which has shown that continuous and increasing depletion of the aquifer (as monitored since the late 1970s by Davis Engineering (see Figure 6)) has affected surface water flows over the last ten years and impacted agriculture which relies on those flows. This region-specific model provides a complete water balance and describes consumptive use in the region that includes return flows, crop shortages, as well as consumption by native vegetation (CDWR 2016a). Because agriculture is the major economic activity in the region, the community has recognized the importance of working towards sustainable management of the region’s water resources on which its economy relies heavily (SLV Development Resources Group 2013). The community recognized the importance of water and demanded legislation from the state that requires sustainable water management, namely Senate Bill 04-222 (Entz et al. 2004) signed in 2004, which is currently being implemented. As part of this bill, CDWR
45


has promulgated well rules and regulations and submitted them to the Colorado Water Court in the Fall of 2015 (CDWR 2015a). The purpose of this legislation is to ensure that the prior appropriation system is upheld, given the now apparent impacts of groundwater pumping on surface water flows. The priority system, or prior appropriation system, is based on the “first in time, first in right” concept of Western U.S. water law, which appropriates surface water rights to the first person to divert the water for beneficial use (CDWR 2016b). SLV’s surface water was fully appropriated by 1900, but since groundwater at that time was thought to be delinked from the surface water system, the state continued to issue well permits for groundwater pumping until the 1970s. The new legislation requires repayment of injurious depletions to senior surface water rights holders due to groundwater pumping (CDWR 2015a).
Another major component of the well rules and regulations, and in some regard the most interesting due to the major drop in aquifer levels that took place during the 2002 drought (Figure 6), is the requirement that the confined aquifer be maintained at the average historic levels similar to the period of 1978 to 2000. This legislation may set a precedent for future sustainable aquifer management and related legislation across the country. The State Engineer’s Office and the CDWR’s Division 3 Engineer, who administer SLV, are responsible for managing and enforcing the new groundwater rules. Also allowed under the legislation is the formation of groundwater management sub-districts of the Rio Grande Water Conservation District (RGWCD). These entities are tasked with managing their
46


water use, replacing injurious depletions to the senior surface water rights holders through replacement water, financial water saving incentive programs, and restoring and maintaining a sustainable aquifer within the parameters of the legislation (CDWR 2015a).
Change in Unconfined Aquifer Storage
400.000
200.000 0
-200,000
-400,000
l—
LU
LU
Lu -600,000
ec
u
<
-800,000
-1,000,000
-1,200,000
-1,400,000
-1,600,000
Figure 6: Change in storage of the unconfined aquifer in SLV from 1975-2016 (Davis Engineering 2016).
Simultaneously, Governor John Hickenlooper issued an Executive Order
in 2013 requiring the Colorado Water Conservation Board (CWCB) to develop a
statewide water plan. Each of the state’s nine river basins host a conservation
47


roundtable composed of water users and stakeholders in each region. The roundtables were tasked to develop Basin Implementation Plans (BIP). The BIPs lay out projects, goals, and priorities for the basin’s future. The Rio Grande BIP (which encompassed the region of SLV) was published in the summer of 2015 and focuses on achieving a balance of competing water needs through cooperative management of water resources (Gibson and et.al. 2015). The BIP outlines detailed plans and allocated funds to help manage water resources. The BIP presents ways to improve soil health and increase soil water holding capacity, improve stream flow forecasting, irrigation improvements, head gate and ditch restoration projects, and highlights the Rio Grande headwaters restoration project. It also emphasizes the importance of maintaining a healthy stream corridor to achieve efficient compact deliveries at the state line. BIPs from across the state informed the Colorado Water Plan, which was delivered to the Governor in November of 2015, and will influence water use in the state. (www.Coloradowaterplan.org).
This unique regional context required the development of SLV-specific methods, models, and data to understand and address the problem. In an attempt to fill information and data gaps, a robust and descriptive baseline consumptive water use analysis (CWU) is presented. This model will be available as a resource to assist local decision-makers who are continuing to develop plans and implement actions towards sustainable water use in SLV.
48


Methods
This analysis tracks the origin (source) of water, whether effective rain (green water) or rivers and streams (blue water) (Falkenmark 1991; Food and Agriculture Organization of the United Nations (FAO) 1993). In the present study, blue water is also divided into surface and groundwater components due to the unique hydrologic features of SLV. Water consumption in SLV is characterized into the following three categories: 1) crops, 2) livestock, and 3) municipal/industrial use; and the following three water sources were considered: 1) effective precipitation (green water), 2) blue surface water runoff (blue water surface/snowmelt), and 3) blue groundwater from the aquifer (blue water ground) (Table 3). The grey water component, which represents water needs for pollutant discharge and dilution, was not considered in this study due to lack of region specific data and community interest.
Table 3: A summary of water sources considered in this study.
Water Source Description
Green Water Green water is the quantity of effective rainfall (ER), meaning rain that fell on crops during the growing season, consumed by the basin annually.
Blue Water Surface Blue water from surface water measures the quantity of snowmelt runoff from rivers and streams that is diverted and consumptively used in the region. Re-charge water is considered groundwater in this analysis.
Blue Water Ground Blue water from groundwater is a measure of the quantity of aquifer water that is pumped and consumptively used or directly sub-irrigated through roots in crop fields.
Because this research focuses on the water consumption patterns of the region, only water originating within the basin was considered and any upstream-
49


embodied water of imported goods and services was excluded. In this way, the analysis differs from the traditional Water Footprint methodology.
The CWU calculations adapted for SLV are shown below:
CWUSLV = CWUgreen + CWUbiue
CWUgreen CWUgreen_Cr0pS + CWUgreen_UVeSt0Ck
cwuMue CWUsurface + cwuground
CWUsurface CWUsurface cr0pS + CWUsurfacejivestock + CWUsurface m&i
CWUground CWUground Cr0pS T CUWground livestock T CWUgr0Urld_M&.I
A unique aspect of the presented regional CWU assessment is that first total CWU in SLV was estimated using a local groundwater model, and then the Consumptive Water Content (CWC) of each crop was derived from this overall CWU, making the results highly specific to SLV. The CWC represents the quantity of water needed to produce a unit of a good or service (i.e., quantity of crop) (Hoekstra et al. 2011; Liu et al. 2009). This level of regional specificity was possible due to the close collaboration of our research team with CDWR, and the use of regional data from CDWR’s HydroBase that contains real-time, historic, and geographic data related to water resources in Colorado. The Rio Grande Decision Support System (RGDSS) groundwater model was used (CDWR 2016c, 2016a), which is specifically calibrated for this region, and allows for a robust,
50


region-specific accounting that does not rely on global averages (which is typical of consumptive water use studies).
Crop Water Use of SLV
Calculations of crop water consumption were performed within the Colorado crop consumptive use model (StateCU) component of RGDSS and considered all of the area in Water Division 3 (which closely matches the boundary of SLV) (CDWR 2015c). StateCU, as applied in RGDSS, models the water budget of a ditch service area by calculating the quantity of water consumed by different crops using a modified Blaney-Criddle method with locally calibrated crop coefficients (CDWR 2015c). In addition, local data related to temperature, precipitation, rooting depths, growing season start and end dates, available water content of soils, irrigated acres, cropping patterns, surface and groundwater diversions, ditch conveyance and irrigation application efficiencies were supplied as inputs into the StateCU model. Data were developed and maintained within HydroBase (CDWR 2016a) as part of RGDSS efforts.
The four major crops cultivated in the region: alfalfa, potatoes, small grains (mainly barley) and meadows/pasture grass were assessed on a monthly time step for years 2000 to 2010. They were assessed for potential evapotranspiration, effective precipitation (EF), and the remaining irrigation water requirement (IWR).
51


Legend
â–¡ Water Districts j Counties
2010 Irrigated Lands CROP TYPES
ALFALFA NEW ALFALFA GRASS PASTURE j POTATOES | SMALL GRAINS H COVER CROP | VEGETABLES | BLUEGRASS FALL WHEAT
~I I
18 Miles
Figure 7: Geospatial representation of crop production in SLV.
Much like the global crop consumptive water use model (CROPWAT), which many studies use to estimate crop water consumption, the StateCU model is used to estimate the effective precipitation and the amount of irrigation water consumed by the crops (Chapagain and Hoekstra 2004; Zeng et al. 2012; Dumont et al. 2013). Using a local model (StateCU) coupled with a local groundwater model (RGDSS) is unique and critical because there may be times when precipitation and irrigation water supplies are insufficient to meet the entire crop demands, and for these instances there is a shortage that may be met if the water table is within the crop’s root zone. The RGDSS groundwater model simulates the water table and the associated crop demand shortage, and if the water table is within the crop-rooting zone, the groundwater model automatically
52


meets the crop water demand through sub irrigation (direct groundwater consumption from crop roots). Further, there may be times when water supplies fall short of meeting the crop demands, which may affect crop yields. Crop yield data were obtained at the county level through the National Agricultural Statistics Service (NASS) (U.S. Department of Agriculture 2015) (Table 4). *
Table 4: Cultivated area and crop production in SLV (2000-2010).
Crop Cultivated area Cultivated area Yield Production Market Value
(thousand acres) (thousand ha) (tons / acre) (thousand tons) $ / ton
Small Grains 94.1 37.6 2.9 267.4 191.3
Potatoes 66.5 26.6 18.7 1,237.8 189.3
Alfalfa 141.3 56.9 3.6 502.3 133.3
Meadows and pasture hay 211.2 84.5 1.7 165.2 +’ 119.2
The cultivated area values come from the RGDSS and utilize aerial imagery, satellite imagery, field verification, NASS, and Rio Grande Water Conservation District (RGWCD) data to evaluate irrigated lands and crop types. By multiplying the cultivated area by crop yield data from NASS, the total production values were developed. The market value is the annual average value on price received in the state of Colorado from 2005-2010 (U.S. Department of Agriculture 2015).
* The cultivated area for meadows and pasture hay is an aggregated value from RGDSS. Some of this land is used for wildlife habitat/wetlands in addition to cattle grazing and haying. The native grasses that are hayed could be sold and exported from SLV or used for livestock within SLV in the current year or subsequent years. The production value given in the table for pasture hay includes only the baled hay exported from SLV as reported by NASS (which is an underestimate of the total production in the region).
Due to the difference in the evaluation methods of this research and StateCU, the results were post-processed to quantify the individual crop consumptive use amounts. Specifically, StateCU evaluates crop use by ditch service area, whereas this research presents region wide average crop consumptive use. The analysis utilizes information from the Alamosa climate
53


station as an approximation for the remainder of Water Division 3 to evaluate crop demands, and then normalizes the crop demands back to the Water Division 3 StateCU model results. By combining the post-processing results from the StateCU model and crop yield data (U.S. Department of Agriculture 2015), CWCgreen, CWCSUrface, and in combination with the RGDSS groundwater model, CWCground, was determined for each of the major crops in the region.
Livestock Water Use of SLV
CWU of livestock (CWUnvetock) was calculated in SLV by multiplying the CWC (m3/ton live weight) of each livestock (CWCnvestock) type by the production quantity in the SLV (Table 5).
Table 5: Livestock herd in SLV (2000 - 2010) for each animal type.
Livestock Amount Production Market Value
(head) (ton live weight) $ / ton (adult) $ / ton (juvenile)
Cattle 83,144 23,464 2,040 2,420
Sheep/Goats 12,711 639 708 2,087
Hogs 221 24 996 996
The head count and production values were collected from NASS, and represent the counties of SLV (U.S. Department of Agriculture 2015). The market values are the average value on price received in the state of Colorado from 2005-2010 (U.S. Department of Agriculture 2015).
The three major water uses needed for calculating the CWCnvestock are drinking water requirement (DWR), service water requirement (SWR), and feed water requirement (FWR) (Zeng et al. 2012; Mekonnen and Hoekstra 2012). SWR is a measure of the water used for keeping and maintaining the animals (e.g. washing, feed mixing etc.). The DWR and SWR were estimated using global annual averages from Mekonnen’s and Hoekstra’s (Mekonnen and
54


Hoekstra 2012) analysis of livestock water footprints for grazing animals. In addition, CWCnvestock also included the water embodied in the feed that the livestock eat (FWR), as this water also originates in SLV. The FWR, which accounts for 99% of the grazing livestock CWU, was calculated based on local CWC of the feed crops grown in the region (Table 6) coupled with the amount and feed conversion efficiency (FCE) of each type of feed consumed by the livestock.
Table 6: Consumptive Water Content (CWC) of feed in the SLV (2000-2010) (1 Ac-ft. = 1233.5 m3).
CWC - green m3/ton CWC - surface m3/ton CWC - ground m3/ton CWC -Total m3/ton
Irrigated Pasture Grass 214.4 542.7 450.4 1,351.5
Alfalfa 135.8 202.6 531.9 870.3
Dry land Pasture Grass 1,266.7 1,266.7
The average moisture content of the various feeds were collected from the local extension office (Reynolds 2015). It was assumed that 60% of the baled feed was from irrigated grasslands and 40% was from locally grown alfalfa. The CWC of the feed in the region was calculated using yield data for hay alfalfa and other hay grass from NASS. To account for the grazing that takes place during spring and fall on the irrigated pastures, an additional yield of 0.5 tons per acre was added to the yield of hay grass (excluding alfalfa). The CWC of upland dry land pasture was averaged as 1.5 tons/acre in a wet year (800 m3 or 8 inches precipitation) and 0.15 tons/acre in a dry year (300 m3 or 3in precipitation) (Whitten 2015; Sparks 2015).
FCE is a measure of the quantity of feed needed to produce a certain quantity of output.
FWR = lFCEf * VWCf
55


/ = feed type
DMI
FCE = —-----:--;---—
live animal weight
DMI = dry matter intake
A daily dry matter intake (DMI) of 3% of the body weight for adult cattle and 1.5% for calves and grazing sheep, in accordance with Mekonnen and Hoekstra (Mekonnen and Hoekstra 2012) and confirmed by the local extension office and a local rancher, was used (Reynolds 2015; Whitten 2015). Total forage consumption values were based on the assumption that all feed requirements needed by livestock were satisfied through local resources, as is the practice in the region. Public grazing lands in the region have the capacity to support a herd of cattle and sheep spending 3 months of the year on dry land pasture (NFS 2015; BLM 1991; Page 2016), 5 months grazing on private irrigated pasture and meadows, and 4 months eating baled hay grass and/or hay alfalfa (Reynolds 2015; Whitten 2015). Total livestock headcount was obtained from NASS. Because the FWR of hogs originates from outside the region and their population is small (Table 5), they were not included in this study. Typically, ranchers in SLV do not finish cattle or sheep locally but tend to export live animals; therefore, the unit of ton live animal was used for livestock production instead of the more conventional unit of kg meat.
56


Municipal and Industrial Water Use of SLV
CWU of the municipal and industrial sector (CWUm&i) was calculated based on input data used in the RGDSS groundwater model on a monthly time step for 2000 - 2010. The value for CWUm&i was based on the total pumping and return flows for all non-irrigation high capacity wells within the RGDSS model domain, which covers the entire region. In addition to groundwater pumping, this sector also includes CWU from reservoir evaporation of surface water. It is important to note that the manufacturing and industrial sector in SLV represents only ~1% of the overall economy (SLV Development Resources Group 2013).
Results and Discussion
Total Consumptive Water Use of SLV
An examination of the results shows that SLV’s consumptive water use is dominated by agricultural activities. Crops account for 63% of CWU of the region, followed by livestock at 13%, and municipal and industrial use accounting for just 3%. In addition, non-agricultural native vegetation accounts for about 21% of the overall CWU in SLV (Figure 8). The total CWU (in million m3)was separated into green water, surface water, and groundwater (Figure 8). By breaking up the CWUbiue into two components: CWUgr0und and CWUSUrface, analysis could be done to see how the various water use types draw upon specific water sources (ground vs. surface water).
57


Alfalfa

Meadows / Pasture
I
Cattle
Small Grains Potatoes M&l Reservoir evap Sheep Cover Crops
â–  Surface â– Ground *-Green
Figure 8: Average annual water allocation in SLV from 2000-2010 (1 Ac-ft. = 1233.5 m3).
Alfalfa is the largest overall user of water followed by meadows/pasture and cattle (Figure 8). It is important to note that meadows/pasture are useful for much more than just grazing cattle, as demonstrated by the fact that even after allocating the water needed for raising forage for cattle and sheep, the remaining meadows are still one of the largest overall contributors to CWU in SLV. This relatively large water use is due in part to the fact that the majority of irrigation water used for meadows/pasture is surface water that is diverted for agricultural use during early spring and summer, in accordance with prior appropriation system. In many locations, surface water is diverted from rivers or streams onto
58
Ac-Ft


the meadows to grow grass where ranchers graze cattle, store hay for the winter, and export haylage (baled grasses). In addition to providing grazing land, meadows sustain a vast and complex network of wetlands, native habitat, and species diversity.
Crop Consumptive Water Use of SLV
Overall, 57% of crop water requirement is met though groundwater, 27% is met though surface water, and 16% is from effective precipitation during the growing season (Figure 9). Alfalfa is the largest consumer of groundwater in SLV, whereas meadows/pasture is the largest consumer of surface and green water. The blue water portion (BWP) combines surface water and groundwater needed for all crops in the region and accounts for about 85% of the CWUCroPs-This BWP for crops is much higher than the global average of 19% (Liu et al. 2009), indicating that the region is much more reliant on irrigation in comparison to other regions. In addition, the result that 57% of the CWUcrops comes from groundwater resources has a profound and direct bearing on the long-term sustainability of the aquifer (a key requirement of the new well rules and regulations).
59


Consumptive Water Content (CWC)
Figure 9: The crop water use allocated between CWUgreen-crops, CWUSUrface-crops, and CWUground-crops as well as the consumptive water content of each crop (2000-2010)*.
* This figure should be read inside-out with water demand by source type shown in the inner ring and the different crops demanding that type of water represented in the outer ring.
60


Livestock Consumptive Water Use of SLV
The livestock consumptive water use aspect considers water needed for feed crops, drinking water, and service water. In SLV, 98% of the water needs for livestock can be attributed to FWR. Overall, 40% of the livestock water needs are met by green water and 30% by ground and surface water (Figure 10). Livestock are able to utilize the sparse green water that falls in the region through dry land pasture grazing in the summer months. This, along with meadows and pasture grazing, is why we see such a large green water percentage for CWUiivestock- The majority of the surface and groundwater portion of the CWUiivestock is attributed to raising irrigated pasture for grazing and winter hay feed.
61


Alfalfa
1.5%
Consumptive Water Content (CWC)
SLV Sheep Herd SLV Cattle Herd
0 20 40 60 80 100
Hundreds
m3 / ton live weight
â–  Surface
â–  Ground Green
Figure 10: Livestock consumptive water use (CWUnvestock) for SLV allocated
between CWUgreen-crops> CWUsurface-crops, and CWUground-crops OS Well 3S the
consumptive use content of each crop. Cattle represent 98% of all CWUnvestock in the region*.
* This figure should be read inside-out with water demand by source type shown in the inner ring and the different feed crops demanding that type of water represented in the outer ring.
62


Agricultural Products
In this research, groundwater intensity is defined as a measure of the quantity of groundwater needed per ton, or per dollar revenue, of agricultural products (Table 7). The dollar value for agricultural products sold in Colorado was obtained from USDA (U.S. Department of Agriculture 2015) using the previous 5-year average and was used to calculate groundwater intensity in dollars/m3 (Table 4 and Table 5).
When revenue is used as the index, potatoes have the highest efficiency in converting groundwater into dollars, followed by livestock (i.e., sheep and cattle). Efficient conversion of groundwater resources to dollars captures two major issues related to the region, economy and a depleting aquifer. Potatoes are efficient at producing high yields (17 tons / acres) and hence have a low groundwater use per dollar revenue. However, this is offset significantly by the fact that they require the highest groundwater portion (81%) per ton of crop (Table 7). This is in contrast to sheep and cattle, which are second in efficient conversion of groundwater resources into dollars, however have the lowest groundwater portion (~30%) per ton of any of the agricultural products in the region. Further, the current consumption of blue water for crops and livestock (almost all of which is exported) results in a net displacement of 1,120 million m3/year (908 thousand ac-ft.) in the form of virtual water flows (calculated as area under the ‘blue water CWU crops and feed’ curve; Figure 11) of surface and groundwater.
63


Table 7: Comparison of major agricultural products in SLV (1 Ac-ft. = 1233.5 m3).
CWU Content - SLV CWU Valley Wide Surface Water Portion (SWP) Groundwater Portion (GRD-WP) Green Water Portion (GRD- WP) Groundwat er Use / $ Revenue
(m3 f1) (Million m3) % % % (m3 GW $'1)
Alfalfa 871 417 23.1 61.3 15.5 2.61
Potatoes 100 124 4.5 81.7 13.9 0.41
Small Grains 759 201 13.7 74.1 12.2 2.26
Live Cattle 9,530 238 30.6 29.8 39.7 1.69
Live Sheep 6,535 5.9 30.1 30.4 39.5 1.54
Blue Water Scarcity
Blue water scarcity (BWS) (Hoekstra et al. 2011; Zeng et al. 2012) was used as a measure to understand water sustainability of the region. BWS occurs anytime the CWUbiue exceeds the blue water availability (WAbiUe) (Figure 11).
WAMue = Runoff — EFR — Interstate Compact Deliveries EFR = Environmental Flow Requirments
Data on monthly average runoff from the mountains into SLV was used to develop a natural runoff curve for the region (Figure 11). Data were composed of gauged and ungauged inflows and provided by RGDSS. The modified runoff curve (Figure 11) accounts for farmers using the aquifer as a reservoir, which is a local practice, and was based on monthly recharge data from RGDSS. Farmers direct their surface water allocations into “recharge pits” at the corners of their fields so water can infiltrate into the aquifer. To estimate the WAb|Ue curve, both the interstate compact water that is owed to downstream states, and the
64


environmental flow requirements (EFR) were subtracted from the total runoff (Figure 11). Because region specific data were not available for EFR, the value 80% of the total runoff was used to be allocated to stream and ecosystem services as proposed by Floekstra (Floekstra et al. 2011; Zeng et al. 2012).
500
450
400
350
300
250
200
150
100
50
0
400
"Total Runoff
-Blue Water Availability
Blue Water CU Crops and Feed ----Modified Runoff w/ Recharge
Figure 11: Monthly water availability and monthly water use (2000-2010) (1 Ac-ft. = 1233.5 m3).
SLV experiences blue water stress, which is the difference between CWUbiue and WAbiue, during seven months of the year (Figure 11). This resource overutilization causes severe damage to the ecosystem of the region (Kang et al. 2007; Zeng et al. 2012). In addition, SLV has been experiencing water shortage because of aquifer mining during two months of the year. Water shortage occurs
65


anytime CWUbiue exceeds the total runoff. The bulk of runoff occurs in May and June, but the spike in CWUbiue occurs between June and August. This local water shortage can be seen whenever the “modified runoff w/recharge” curve drops below the CWUbiue curve (Figure 11). Even with the use of re-charge pits, SLV has been experiencing water shortage, mainly in July and August, of about 41 million m3, annually. To make matters worse, it is predicted that the natural runoff in spring will begin to shift from June to May due to the effects of climate change (Gibson and et.al. 2015; Dagmar and Vaddey 2013). Once CWUbiue exceeds runoff, it does not take long to see the effects. Water shortage due to dry conditions and over-pumping has caused a major decline in the aquifer (Davis Engineering 2016).
The aquifer is likely to be further affected by changes in climate; some of these effects are apparent today. The aquifer is recharged by surface water runoff, which infiltrates to the groundwater through rivers and streams, rim recharge, flood irrigation, and state issued recharge decrease. A drying regime in the surrounding mountains, which is predicted by climatologists, will reduce snow pack and subsequently limit the aquifers’ ability to recover from the annual groundwater pumping (Dagmar and Vaddey 2013). Senate Bill 04-222 and the groundwater rules and regulations, requires water users in the region to maintain aquifer levels similar to the average levels between the years 1978 and 2000. This is already a difficult task and will be even more challenging under a changing climate (Hurd and Coonrod 2007; Malcolm et al. 2012). It is predicted
66


that stream flows at the upper gauges will decrease by approximately one-third between now and 2100 (Dagmar and Vaddey 2013; Gibson and et.al. 2015).
Looking Forward
SLV is a microcosm of many dry regions across the globe where water demand exceeds supply, and drought due to a changing climate is adding stress to the already over appropriated systems (El-Beltagy and Madkour 2012). This CWU analysis developed for SLV can provide valuable information and perhaps management models for other areas facing similar challenges.
This is a complex system and solely increasing efficiency may not be the answer. Along the riparian corridors, diverted water used for growing meadows and pasture sustains a variety of wetland types that are invaluable for waterfowl, wildlife, recreation, and other ecosystem services (Mitsch and Gosselink 1993). These seasonal wetlands account for a substantial portion of the wetlands in the region (Gibson and et.al. 2015). While flooding land during the growing season may not appear to be an efficient use of water, it provides an important service, as these lands provide habitat for native species, as well as grazing land for livestock (Gibson and et.al. 2015; Niemuth et al. 2004; Mitsch and Gosselink 1993). In addition, it is important to sustain a “wet sponge” along the river corridor, so that Rio Grande Compact water deliveries can be effectively managed. If river corridor zones were permanently dried by change of use, it would cause a substantial challenge to the State’s ability to meet compact requirements. This situation was experienced during the 2002 drought and
67


subsequent years (Blankenbuehler 2016). On the other hand, through this analysis, we see that native vegetation consumes significant amounts of water, which affects the region’s ability to maintain a sustainable aquifer. It will be important for future regional water use analyses to allocate water towards ecosystem services and environmental flow requirements. Much work has been done in the area of quantifying ecosystem services and these methodologies could be applied to river basins such as SLV (Costanza et al. 1998, 2014, 2011; Costanza 2014).
Maintaining a sustainable aquifer is of utmost importance in the region. Managing the use/loss while maintaining the agricultural heritage and economy of the region is important. However, the aquifer is continually depleted by extraction of groundwater for agricultural purposes. The results indicate that agricultural products, such as sheep and cattle, offer an opportunity for the region to move towards sustainable water management, because they do not rely as heavily on groundwater, and utilize the scarce green water through dry land grazing. This opportunity is somewhat constrained due to space limitations for dryland pasture grazing and the potential for ongoing drought. Therefore, alternative-grazing practices such as cover crop grazing will need to be employed if the number of livestock were to increase.
A variety of soil health management practices, implemented by an active group of farmers and ranchers, are showing promising results (Rockey 2014; Scully 2014). These practices include the use of cover crops, compost
68


application, alternative pest and weed control practices, reduced tillage, grazing animals on green manure, and others. These practices are part of a larger strategy to improve soil health (as measured by increased organic matter in the soils, water-holding capacity, soil fertility, and other properties) and reduce the need for pumped water to maintain soil moisture for growing crops. Soil health is a field of growing interest in the community and holds promise for reducing pumping while sustaining viable agricultural production. It is our hope that this region-specific CWU assessment framework can aid this and other similar regions in their pursuit for sustainable water resource management. The proposed framework could be replicated in other basins, provided region specific data and models are available.
69


CHAPTER IV
COMMUNITY ENGAGEMENT
Engaging a Rural Agricultural Community in Sustainability Indicators and Future Scenario Identification: Case of the San Luis Valley
Abstract
This chapter describes the application of community-based participatory research (CBPR) to engage representative community stakeholders in a rural agricultural region in identifying future sustainability scenarios for modeling with sustainability indicators. Over the course of two years, researchers and the Community Advisory Board identified, deliberated, and, based on their input, conceptually framed future scenarios using sustainability indicators for modeling and action. The suggested scenarios (for modeling) that emerged through this engagement were centered on solar energy development in the region, and changes to the cropping regime in San Luis Valley (SLV) in southern Colorado. SLV is a unique geographically isolated agricultural region that has been looked at both by the EPA and the state of Colorado as an ideal location for implementing sustainability measures. Importantly, in line with CBPR principles and best practices, community stakeholders remained committed and engaged throughout the research process. Through training and knowledge transfer from researchers to the community, SLV now has the capacity to use local data to update region-specific greenhouse gas emissions and consumptive water use indicator models. This engagement was successful both in terms of its usefulness in steering the research direction as well as its impact on community
70


stakeholders involved in this project. Based on this experience, we recommend this participatory approach to researchers seeking to improve the relevance and impact of region-specific sustainability analyses.
Introduction
Background to Participatory Research
Working with a community that is the subject of a research project can lead to the integration of technical and local knowledge, thereby increasing the chances that the research will result in relevant design, adoption of the findings, and increased local capacity (Reed 2008). This means researchers can be a catalyst to the process of social action, but the community acted upon must also be involved. The practice of community-based participatory research (CBPR), some say, can trace its origins to Popular Education whose founder was Paulo Freire.
...concrete reality consists not only of concrete facts and (physical) things, but also includes the ways in which the people involved with these facts perceive them... concrete reality is the connection between subjectivity and objectivity, never objectivity isolated from subjectivity”
(Freire 1982).
Freire’s ideas awakened a movement, both in the scientific community and the social justice community, to the idea that social change is rooted in education and empowerment. These early studies and interventions (Whyte 1991; Borda and Orland Rahman 1991) led to the development of the field of Participatory
71


Action Research, which puts the communities and their perspectives at the center of research. In the 1990’s, many of these concepts were integrated into public health research to engage communities affected by complex health issues. Much like issues of poverty and oppression, complex health issues (such as HIV/AIDS) need the community involved and informed in order to make real change (Petrow et al. 1990; Brown 1991; McKinlay 1993). It is not enough to simply present facts. A collection of best practices were developed (Figure 12) to encourage a partnership between researchers and community members to benefit both parties (researchers and communities), and allow for “a more balanced set of political, social, economic, and cultural priorities” (Israel et al. 1998). For researchers, community participation offers, first and foremost, invaluable access to “primary resources,” in the form of local expertise. The community, in turn, can often gain access to more technical resources from research partners. Through participatory research, researchers and community members can influence, learn from, and foster trust in one another, allowing a
72


well-conceptualized, comprehensive study that is useful for the community.
1. The community involved comprises its own unit of identity.
2. Research builds on the strengths and resources within the community (takes advantage of social capital).
3. Incorporates collaborative partnerships in all phases of research.
4. Integrates knowledge and subsequent action.
5. Facilitates a co-learning and empowering process that allows for inequalities in knowledge and background.
6. Utilizes a cyclical and iterative process, and is not merely a study never to be revisited.
7. Views issues from the perspectives of all involved.
8. Shares findings and knowledge gained with all partners and stakeholders (transference).
Figure 12: A summary of the key aspects of a Community Based Participatory Research project (Israel et al. 1998).
Community Engagement and Sustainability Research
More recently, CBPR and other community engagement frameworks have been used in the field of environmental management and sustainability research (Reed 2008). Sustainability Visioning, the process of creating discourse around positive visions about our societies’ future to stimulate change, has been investigated and implemented (Iwaniec and Wiek 2014). Similar methods have been used to engage cities and municipalities around climate action planning using future scenario modeling (Sheppard et al. 2011; Shaw et al. 2009; Burch 2010) and community engagement frameworks (Ramaswami et al. 2014). Community engagement and future scenario modeling has also been successful at the national level to approach issues of water uncertainty in a changing climate, and alternative water governance techniques in a location facing ongoing
73


water management problems (Kuzdas et al. 2016; Kuzdas and Wiek 2014). Involving the community in the development of sustainability indicators and sustainability goals/targets allows community members and stakeholders to provide input on their goals and priorities, providing valuable data for future management decisions. Furthermore, as the community integrates into the process, they become more invested in the outcome (Fraser et al. 2006).
Case of San Luis Valley
This research takes lessons learned from applications of community-based sustainability research and applies them to the agricultural region of San Luis Valley (SLV) in Southern Colorado. Sustainability assessment and planning has a long history in SLV. For over a decade the community has been working with the state to develop a plan to comprehensively understand, and sustainably manage, the arid agricultural region’s vast aquifer system (Gibson and et.al. 2015). As part of this effort, the Colorado Division of Water Resources (CDWR) has developed a robust groundwater model that is used for setting water use targets and monitoring progress on aquifer management (CDWR 2015d). The region was also the subject of an Environmental Protection Agency (EPA) pilot study on regional sustainability metrics (Hopton et al. 2010). EPA considered SLV an ideal study location because of its distinct hydrologic boundaries, limited population, large amount of publically owned land, and interest expressed by government agencies and the local population (U.S. EPA 2010). The study developed and calculated four metrics for sustainability in the region over time.
74


The results were presented in a suite of publications in the Journal of Environmental Management (Heberling et al. 2012; Hopton and White 2012; Eason and Cabezas 2012; Campbell and Garmestani 2012; Heberling and Hopton 2012). The present study builds on the work of EPA by adding two new sustainability indictors, a greenhouse gas (GHG) emissions model (Dubinsky and Karunanithi 2017a) and a consumptive water use model (Dubinsky and Karunanithi 2017b), to the suite of metrics. This chapter outlines the community engagement portion of the research, which involved local stakeholders in regional sustainability indicator development and future scenario selection for modeling with the indicators.
Goals
The overall goal of engagement with the community in this case was threefold:
1. Utilize local knowledge and expertise to better understand the baseline results and future trajectories of the GHG emission and consumptive water use indicator models.
2. Collaborate with community representatives to develop realistic future scenarios for modeling with the two sustainability indicators.
3. Transfer the baseline sustainability indicators and future scenario model results to decision-makers in the region.
The following sections outline the involvement with the community in SLV from first contact; through formation of, and collaboration with, a Community
75


Advisory Board, and culminating with knowledge transfer. The research describes why having the community involved produced relevant outcomes, and presents best practices/learned along the way. Without these steps involving the community, the research might be generic, less relevant, and less useful to the region. Outcomes from this work suggest that community involvement not only fostered access to critical information and data sources for the researchers, but also helped shape the way the community understands sustainability in their region.
Methodology
A graphical representation of the community engagement framework that guided the work conducted on the SLV from 2013-2016 is presented in Figure 13. Because the research involved human subjects, it required review and approval by the University of Colorado Denver’s Institutional Review Board (IRB) in November of 2013. The IRB declared the research as “exempt” status, meaning minimal risk to the human subjects involved.
76


Research
Question
• EPA identified issue
• Collect data/indicator baseline
• Define scenarios/ scenario modeling
Knowledge
Transfer
• Gather a subset of the CAB for training
• Hold training sessions on the indicator models
7


Composed of stakeholders Meets three times per year Involved throughout the process ^
Use expert input to refine assumptions

4
Define
Scenarios
Relevant and desirable Feasible Measurable
Data collection phase Ongoing and iterative with expert input
it
*

Figure 13: Framework for community engagement in San Luis Valley. The number in the circle indicates the order of each step. * The scenario modeling methods and results are presented in Chapter V.
Defining the Research Question
Based on initial discussions with research collaborators, the primary research question that guided this work was identified as: What scenarios and/or issues do the community of SLV identify as important to the sustainability of the region? This question guided the research from stakeholder identification through knowledge transfer (steps 2 through 7; Figure 13), and fulfilled the goal of defining and conceptualizing future scenarios for modeling.
77


Stakeholder Identification
Identifying stakeholders requires gaining an understanding of individuals, groups, and organizations affected by decisions faced while shaping the research, and the results from the outcome of the research (Freeman 1984; Reed et al. 2009; Rowe and Frewer 2000). Stakeholder identification began with multiple trips to SLV (the research office was in Denver, ~230 miles away) to attend community meetings and social forums related to the topic of sustainability in the region. These visits presented an opportunity to meet people and begin conversations about the research topic. Over the course of 5 multi-day trips to the region, 40 impromptu one-on-one interviews were conducted (Figure 14).
Interview Guide
Thank you for taking the time to speak with me today. My research looks at regional sustainability assessment for use by decision makers and community groups. I am conducting interviews in the valley to get an understanding of people's views of sustainability in the Valley.
I will keep all comments confidential and anonymous and it is by choice that you are participating today. You do not have to participate if you do not want to.
1. What do you do here in the valley and for how long? In what part of the valley do you reside and how old are you?
2. What do you consider to be the valleys most valuable assets or attributes? What do you love most about the valley?
3. Do you think that the valley is utilizing its resources wisely?
4. What does sustainability mean to you for the San Luis Valley in terms of Economics, environment, and social/culture?
5. If you have 100% to give out for prioritizing in the Valley how much would you give to each: Economics, Environment, Social/Culture.
6. If you can think back to the "glory years" of the Valley when would that be and what was it like?
7. What future scenarios for the valley would you be interested in me considering in this study? What ideas other then CREP and fallowing do you consider possible for the San Luis Valley to achieve sustainability?
8. Do you plan to participate in a sub district?
9. Who in the Valley do you think I should talk to about these questions?
Figure 14: Interview guide used to spark discussion when informally engaging with residents of SLV.
These one-on-one interviews took place at community forums and acted
as a conversation starter, but were not used quantitatively. The engagement
process was evolving in real time and the most beneficial outcome of the one-on-
78


Full Text

PAGE 1

TOWARDS REGIONAL SUS TAINABILITY ASSESSME NT UTILIZING COMMUNITY BASED PARTICIPATORY RESEARCH, S USTAINABILITY INDICA TORS , AND FUTURE SCENARI O MODELING By JONATHAN DUBINSKY B.S., University of Kansas, 2004 M.Eng., University of Colorado Denver , 2013 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Engineering and Applied Science 20 16

PAGE 2

ii This thesis for the Doctor of Philosophy degree by Jonathan Dubinsky has been approved for the Engineering and Applied Science Program by Arunprakash Karunanithi, Advisor Bruce Janson , Chair Heri berto Cabezas Matthew Hopton Deborah Main December 17 , 2016

PAGE 3

iii Dubinsky, Jonathan (PhD, Engineering and Applied Science) Towards R egional S ustainability A ssessment U tilizing Community Based Participatory Research, Sustainability Indicators , and Future Scenario Modeling Thesis directed by Associate Professor Arunprakash Karunanithi . ABSTRACT Decision making with regards to sustainability at a regional/local level is increasingly recognized as an important issue. This research focused on the rura l agricultural region of San Luis Valley (SLV) in southern Colorado and builds on an Environmental Protection Agency sustainability study in the same region . The goal of this research was to select and calculate relevant su stainability indictors, and then use those baseline results to model potential future scenarios . A region specific greenhouse g as (GHG) accounting indicator and a consumptive water u se indicator were developed and calculated in SLV over the period of 1980 Ð 2010. In addition to sustainability modeling, t his research leveraged the principles of Community Based Participatory research (CBPR) to engage local stakeholders throughout the process to ensure that the research was relevant to the region . From a carbon emissions perspectiv e, the baseline assessment showed that on a per gross domestic/regional p roduct (GDP) basis, SLV carbon emissions are almost twice that of the national average, indicating that, with all things being equal, agricultural economies contribute disproportionat ely more towards GHG emissions. A future scenario analysis revealed that SLV can reduced GHG emissions by ~5% through crop changes and alterations to the

PAGE 4

iv crop rotation regime. Another scenario showed that SLV, with its significant solar resource, has the potential to offset much or all of its GHG emissions by increasing solar development and utilizing renewable energy carbon credits . T he baseline water use assessment of the region showed that SLV meets 85% of the crop water demand through irrigation , where as the global average is only 19% . This highlight s the heavy reliance the economy has on its ground and surface water resources in a region where agriculture is the number one economic driver . When comparing livestock water use (including feed crop s grown to support local livestock ) to export crops in the region (i.e. , potatoes, alfalfa, small grains) we see that livestock consume significantly less groundwater than the export crops on a per ton of product basis, revealing the critical role livestoc k play in this arid region with a depleting aquifer . Results from the f uture scenario modeling show ed that irrigation water use could be reduced by ~10% through realistic shift s in the crop regime while keeping land fallowing to a minimum. This research was successful in terms of engaging a rural region around issues of sustainability . Through training and knowledge transfer from researchers to the community, SLV now has the ability to use the sustainability indicator models developed in this research . This work is highly relevant to policymakers and planners , and will provide the community with some necessary tools to make policy choices for sustainable growth in the region.

PAGE 5

v The form and content of this abstract are approved. I recommend its publicati on. Approved: Arunprakash Karunanithi

PAGE 6

vi DEDICATION I dedicate this work to my loving and supportive sweetheart, Elizabeth. She has encouraged me to always pursue my dreams with confidence and grace.

PAGE 7

vii ACKNOWLEDGMENTS F irst I thank my advisor, Dr. Arunprakash Karunantihi, for his countless hours of time and counsel spent with me throughout this process. He contributed intellectually to the research and analysis and he mentored me through personal and professional challenges. I thank him for his financial support and his trust in me to accomplish this large task. I also thank and acknowledge my collaborators on this research. Thank you, Dr. Mat thew Hopton and Dr. Ma t t hew Heberling, from the U.S. Environmental Protection Agency for your countless discussions and insights throughout this process. To my PhD commit tee: Dr. Bruce Janson, Dr. Deborah Main, Dr. Heriberto Cabezas, and Dr. Mat hew Hopton. Thank you for your commitment to this work and fo r your helpful comments, constructive criticism , patience and support. I acknowledge the members of the Community Advisory Board (CAB), without whom this research would not be possible. A very special thank you to Richard Sparks, Sheldon Rockey, Cleave Si mpson, Patrick O'Neill, and George Whitten, who were always on call and ready and willing to discuss agricultural land use in detail. Also a special thank you to Christine Canaly, John Stump, and Mary Hoffman for attending so many meetings, asking so many good questions, and being so committed to this work.

PAGE 8

viii Thank you to all members of the CAB (listed below in alphabetical order): Tony Aloia, Colorado Fish and Wildlife Michael Armenta, Conejos County Clean Water Fred Bunch, Chief of Resources Management a t Great Sand Dunes National Park Christine Canaly, Director of the San Luis Valley Ecosystem Council Nathan Coombs, Director of the Conejos Conservancy District Claudia Ebel, San Luis Valley Local Foods Coalition Regenarldo Garcia, Rocky Mountain Preventio n Research Center Andrea Guajardo, Director of Conejos County Clean Water Mary Hoffman, Adams State University Community Partnerships Jared Beeton, Adams State University Karla Shriver, Rio Grande County Commissioner Jim Meitz, Sustainability Park (SEED) P atrick O'Neill, Owner of Soil Health Services Lawrence Pacheco, Costilla County Commissioner Sheldon Rockey, Rockey Farms Leroy Salazar, Water Users Solution Sub Committee Cleave Simpson, Man a ger of the Rio Grande Water Conservation District Richard Sparks , Natural Resource Conservation Service John Stump, Former head of the SLV Development Resources Group George and Julie Whitten, Saguache County Ranchers

PAGE 9

ix I acknowledge other collaborators in this research. To Debbi Main, Elizabeth Baker Jennings , and Tamara Chernomordik for greatly contributing to the CBPR portion of the research. Thank you to Rio de la Vista , James Heath , and Dr. Willem Schreuder for providing region specific context and data for the Consumptive Wa ter Use C hapter . Thank you to B rian Lewendowski from University of Colorado Boulder for working with me on the Implan model and data. A special thank you to Matthew Stermer, Mark Easter, and the entire Comet farm team at Colorado State Universi ty. I recognize the U.S. Environmental Pro tection Agency's office of Research and Development, National Risk Management Research Laboratory, Sustainable Technology Division for developing the foundation of knowledge in this research and for providing funding to make this effort possible (cooperati ve agreement number 83522701 ) . Finally , I would like to acknowledge someone that worked on and discussed issues of sustainability long before it was a popular topic. Buckminster Fuller' s writings, which were given to me by my father at age 12, informed me of our responsibilit ies as crewmembers aboard Spaceship E arth. He sounded the alarm over a generation ago, and provided us with prophetic insights of how we can improve the lives of al l human kind while simultaneously protecting and enhancing the very planet (spaceship) that makes it all possible .

PAGE 10

x TABLE OF CONTENTS !"#$%&' ( )( )*%'+,-!%)+* ( ................................ ................................ ................................ ...................... ( / ( 0123456789 ( ................................ ................................ ................................ ................................ ........ ( / ! "#$%&'(&)'*'%+!",'-(,! ................................ ................................ ................................ ................................ .... ! / ! 0-&$#1'(2!"#$%&'(&)'*'%+ ! ................................ ................................ ................................ ............................. ! 3 ! 4#1&*!4-2'5($ ! ................................ ................................ ................................ ................................ .................... ! 6 ! $5:;<67=(>653 ( ................................ ................................ ................................ ................................ ... ( ? ! 7,5*52',&*!855%91'(%!:(&*+$'$ ! ................................ ................................ ................................ .................... ! ; ! 7<-12+!:(&*+$'$ ! ................................ ................................ ................................ ................................ .............. ! ; ! =1--(!>-%!4-2'5(&*!?15@#,% ! ................................ ................................ ................................ ....................... ! A ! 8'$B-1!C(D51<&%'5( ! ................................ ................................ ................................ ................................ .......... ! A ! $5:=:8@(>653 ( ................................ ................................ ................................ ................................ ..... ( A ! :11&(2-<-(%!5D!%B-!E'$$-1%&%'5( ! ................................ ................................ ................................ ........... ! F3 ! ))( B'&&*"+-C&(B#C(#!!+-*%)*B ( ................................ ................................ ................. ( /D ! #E=@512@ ( ................................ ................................ ................................ ................................ .............. ( /D ! )8@56972@<68 ( ................................ ................................ ................................ ................................ ..... ( /A ! F:@G696H64I ( ................................ ................................ ................................ ................................ ..... ( J/ ! 7(-12+!&(@!G#'*@'(2$!"-,%51 ! ................................ ................................ ................................ ................... ! // ! :21',#*%#1-!&(@!H&(@!I$-!"-,%51 ! ................................ ................................ ................................ ........... ! /6 ! J1&($951%&%'5(!"-,%51 ! ................................ ................................ ................................ ................................ ! /A ! K&$%-!"-,%51 ! ................................ ................................ ................................ ................................ ................... ! /L ! "5*&1!M1-@'% ! ................................ ................................ ................................ ................................ ..................... ! 3N ! ':=7H@=(189(,<=27==<68 ( ................................ ................................ ................................ ................. ( KJ !

PAGE 11

xi J5%&*!=O =!-<'$$'5($!'(!"HP ! ................................ ................................ ................................ ..................... ! 3/ ! I9$%1-&#%&'(-C& ( ................................ ................................ ......................... ( MJ ! #E=@512@ ( ................................ ................................ ................................ ................................ .............. ( MJ ! )8@56972@<68 ( ................................ ................................ ................................ ................................ ..... ( MK ! F:@G69= ( ................................ ................................ ................................ ................................ ............. ( MN ! M159!K&%-1!I$-!5D!"HP ! ................................ ................................ ................................ ............................... ! QF ! H'R-$%5,S!K&%-1!I$-!5D!"HP ! ................................ ................................ ................................ ..................... ! Q6 ! 0#(','9&*!&(@!C(@#$%1'&*!K&%-1!I$-!5D!"HP ! ................................ ................................ ...................... ! QA ! ':=7H@=(189(,<=27==<68 ( ................................ ................................ ................................ ................. ( ?D ! J5%&*!M5($#<9%'R-!K&%-1!I$-!5D!"HP ! ................................ ................................ ................................ .. ! QA ! M159!M5($#<9% 'R-!K&%-1!I$-!5D!"HP ! ................................ ................................ ................................ .. ! QT ! H'R-$%5,S!M5($#<9%'R-!K&%-1!I$-!5D!"HP ! ................................ ................................ ......................... ! ;F ! :21',#*%#1&*!?15@#,%$ ! ................................ ................................ ................................ ................................ . ! ;3 ! G*#-!K&%-1!",&1,'%+ ! ................................ ................................ ................................ ................................ ..... ! ;6 ! H55S'(2!851U&1@ ! ................................ ................................ ................................ ................................ .......... ! ;A ! )L( !+FF-*)%O(&*B#B&F&*% ( ................................ ................................ ......................... ( DP ! #E=@512@ ( ................................ ................................ ................................ ................................ .............. ( DP ! )8@56972@<68 ( ................................ ................................ ................................ ................................ ..... ( D/ ! G&,S215#(@!%5!?&1%','9&%51+!4-$-&1,B ! ................................ ................................ ............................... ! AF !

PAGE 12

xii M5<<#('%+!7(2&2-<-(%!&(@!"#$%&'(&)'*'%+!4-$-&1,B ! ................................ ................................ ! A3 ! M&$-!5D!"&(!H#'$!P&**-+ ! ................................ ................................ ................................ ............................... ! A6 ! =5&*$ ! ................................ ................................ ................................ ................................ ................................ ... ! AQ ! F:@G696H64I ( ................................ ................................ ................................ ................................ ..... ( DQ ! E-D'('(2!%B-!4-$-&1,B!V#-$%'5( ! ................................ ................................ ................................ ............. ! AA ! "%&S-B5*@-1!C@-(%'D',&%'5( ! ................................ ................................ ................................ ........................ ! AL ! M5<<#('%+!:@R'$51+!G5&1@ ! ................................ ................................ ................................ .................... ! LN ! M5 W @-R-*59!"#$%&'(&)'*'%+!C(@',&%51!G&$-*'(! ................................ ................................ .................. ! L/ ! "-*-,%'(2!",-(&1'5$ ! ................................ ................................ ................................ ................................ ...... ! L3 ! ':=7H@= ( ................................ ................................ ................................ ................................ ................ ( A? ! 8#%#1-!",-(&1'5$ ! ................................ ................................ ................................ ................................ ........... ! LQ ! J1&'('(2!&(@!X(5U*-@2-!J1&($D-1 ! ................................ ................................ ................................ ........ ! T/ ! ,<=27==<68 ( ................................ ................................ ................................ ................................ ......... ( NJ ! G-$%!?1&,%',-$!&(@!H-$$5($!H-&1(-@ ! ................................ ................................ ................................ .... ! T/ ! H'<'%&%'5($ ! ................................ ................................ ................................ ................................ ...................... ! T6 ! !682H7=<68 ( ................................ ................................ ................................ ................................ ......... ( N? ! L( R-%-'&(C!&*#')+(F+,&S)*B ( ................................ ................................ ..................... ( NQ ! #E=@512@ ( ................................ ................................ ................................ ................................ .............. ( NQ ! )8@56972@<68 ( ................................ ................................ ................................ ................................ ..... ( ND ! F:@G69= ( ................................ ................................ ................................ ................................ .......... ( /PJ ! M159!4-2'<-!MB&(2-!",-(&1'5 ! ................................ ................................ ................................ .............. ! FN/ ! "5* &1!7(-12+!E-R-*59<-(%!",-(&1'5 ! ................................ ................................ ................................ . ! FFF ! ':=7H@=(T(,<=27==<68 ( ................................ ................................ ................................ .................... ( //Q ! 4-$#*%$!D15
PAGE 13

xiii C7UU15I(6V(R<89<84= ( ................................ ................................ ................................ .................. ( /JD ! L)( !682H7=<68 ( ................................ ................................ ................................ ........................ ( /KP ! C7UU15I(6V(':=7H@= ( ................................ ................................ ................................ .................... ( /KP ! MB&9%-1!CCY!=1--(B5#$-!=&$!:,,5#(%'(2 ! ................................ ................................ ........................... ! F3N ! MB&9%-1!CCCY!M5($#<9%'R-!K&%-1!I$-!:(&*+$'$ ! ................................ ................................ ............... ! F3F ! MB&9%-1!CPY!M5<<#('%+!7(2&2-<-(% ! ................................ ................................ ................................ ! F3/ ! MB&9%-1!PY!8#%#1-!",-(&1'5!05@-*'(2 ! ................................ ................................ ............................... ! F33 ! R7@75:(>653 ( ................................ ................................ ................................ ................................ .. ( /K? ! 7,5(5<',!:(&*+$'$ ! ................................ ................................ ................................ ................................ ..... ! F3Q ! "#$%&'(&)'* '%+!0-%1',$!&(@!8#%#1-!",-(&1'5!05@-*'(2 ! ................................ ............................... ! F3; ! 0&'(%&'('(2!4-*&%'5($B'9$ ! ................................ ................................ ................................ ..................... ! F6/ ! G-+5(@!"HP ! ................................ ................................ ................................ ................................ ................... ! F6/ ! ':V:5:82:= ( ................................ ................................ ................................ ............................. ( /MM ! #$$&*,)W ( ................................ ................................ ................................ .............................. ( /QP ! #( X ( B5::8G67=:(B1=(YB"BZ(&U<==<68=()89<21@65 ( ................................ ............................... ( /QP ! 0( X ( !68=7U[@<;:(>1@:5(-=:(Y!>-Z()89<21@65 ( ................................ ................................ .... ( /NP ! !( \ ( R7@75:(C2:815<6(F69:H<84 ( ................................ ................................ ................................ ... ( J/J !

PAGE 14

xiv LIST OF TABLES Table 1: Material flows, emission factors, and total GHG emissions in the San Luis Valley. ................................ ................................ ................................ ... 31 ! Table 2: Comparing the SLV inventory with GHG inventories at various spatial scales. ................................ ................................ ................................ .......... 39 ! Table 3: A summary of water sources considered in this study. .......................... 49 ! Table 4: Cultivated area and crop production in SLV (2000 2010). ..................... 53 ! Table 5: Livestock herd in SLV (2000 Ð 2010) for each animal type. .................. 54 ! Table 6: Consumptive Water Content (CWC) of feed in the SLV (2000 2010) (1 Ac ft. = 1233.5 m 3 ). ................................ ................................ ....................... 55 ! Table 7: Comparison of major agricultural products in SLV (1 Ac ft. = 1233.5 m 3 ). ................................ ................................ ................................ ...................... 64 ! Table 8: The future scenarios selected by the CAB for modeling with the sustainability indicators. ................................ ................................ ................ 86 ! Table 9: Crop rotations under the crop regime change scenario. ........................ 91 ! Table 10: Comparison between the baseline crop rotational acreage in SLV and the crop rotat ional acreage under the crop regime change scenario. ........ 104 ! Table 11: Inputs used to assess the baseline GHG emissions in the region using the COMET Farm model, and the outputs produced for each of the crop rotations assessed. ................................ ................................ ..................... 107 !

PAGE 15

xv Table 12: Inputs used to assess the GHG emissions of the crop regime change future scenario using the COMET Farm model, and the outputs produced for each of the crop rotations assessed. ................................ .......................... 108 ! Table 13: Data used for estimating electricity needed per volume of g roundwater use. Data from Xcel energy were only available from 2006 to 2010, so this period was used to establish the average. ................................ ................. 109 ! Table 14: Summary of the solar development pathways explored under the solar energy development scenario. ................................ ................................ .... 112 ! Table 15: Annual crop acreage in SLV (avg. 2000 Ð 2010; acres) as well as crop acre age under the fully implemented crop regime change scenario. ......... 117 ! Table 16: Changes in GHG emissions from soil in SLV comparing the baseline scenario and the crop regime change scenario. ................................ ......... 119 ! Table 17: Results on changes in irrigation consumptive water use from the crop regime change scenario. ................................ ................................ ............ 120 ! Table 18: Results from the GHG emissions indicator when modeling the two solar development pathways in the solar energy development scenario. ........... 124 ! Table 19: Results from the consumptive water use indicator when modeling the two solar development pathways in the solar energy development scenario ................................ ................................ ................................ .................... 127 ! Table 20: Data and data sources used for the GHG emissions indicator. ..... Error! Bookmark not defined. !

PAGE 16

xvi Table 21: Model inputs for the GHG emissions indi cator (1980 1989). Data sources are presented in Table 20. Blank cells in this table show data were unavailable for that year through the data provider. The model linearly interpolates data gaps. ................................ .. Error! Bookmark not defined. ! Table 22: Model inputs for the GHG emissions indicator (1990 1999). Data sources are presented in Table 20. Blank cells in this table show data were unavailable for that year through the data provider. The model lin early interpolates data gaps. ................................ .. Error! Bookmark not defined. ! Table 23: Model inputs for the GHG emissions indicator (2000 2009). Data sources are presented in Table 20. Blank cells in this table show data were unavailable for that year through the data provider. The model linearly interpolates data gaps. ................................ .. Error! Bookmark not defined. ! Table 24: Model inputs for the GHG emissions indicator (2010 2017). Data sources are presented in Table 20. Blank cells in this table show data were unavailable for that year through the data provider. The model linearly interpolates data gaps. ................................ .. Error! Bookmark not defined. ! Table 25: Model outputs from the GHG emissions indicator (1980 1989). .... Error! Bookmark not defined. ! Table 26: Model outputs from the GH G emissions indicator (1990 1999) ..... E rror! Bookmark not defined. ! Table 27: Model outputs from the GHG emissions indicator (2000 2009). .... Error! Bookmark not defined. !

PAGE 17

xvii Table 28: Model outputs from the GHG emissions indicator (2010 2016). .... Error! Bookmark not defined. ! Table 29: Data and data sources used for the CWU (CWU) indicator ........... Error! Bookmark not defined. ! Table 30: Model inputs for the CWU indicator (1980 1989). Data sources for all data are presented in Table 29. Pay close attention as the units change throughout the table. ................................ ...... Error! Bookmark not defined. ! Table 31: Model inputs for the CWU indicator (1990 1999). Data sources for all data are presented in Table 29. Pay close a ttention as the units change throughout the table. ................................ ...... Error! Bookmark not defined. ! Table 32: Model inputs for the CWU indicator (2000 2009). Data sources for all data are presented in Table 29. Pay close attention as the units change throughout the table. ................................ ...... Error! Bookmark not defined. ! Table 33: Model inputs for the CWU indicator (2010 2014). Data sources for all data are pr esented in Table 29. Pay close attention as the units change throughout the table. ................................ ...... Error! Bookmark not defined. ! Table 34: Model outputs for the CWU indicator (1980 1990). .... Error! Bookmark not defined. ! Table 35: Model outputs for the CWU indicator (1991 2001). .... Error ! Bookmark not defined. ! Tab le 36: Model outputs for the CWU indicator (2002 2010). .... Error! Bookmark not defined. !

PAGE 18

xviii

PAGE 19

xix LIST OF FIGURES Figure 1: The San Luis Valley. Whole counties were always considered for data collection even though parts of the county may fall outside of the Upper Rio Grande river basin geographic boundary. ................................ ...................... 9 ! Figure 2: Average month ly precipitation (1980 Ð 2010) during the growing season (May to August) in SLV (PRISM Climate Group 2016)*. ............................... 10 ! Figure 3: Community engagement timeline in the SLV during the course of this research. ................................ ................................ ................................ ....... 15 ! Figure 4: The GHG accounting for each of the major emission categories in San Luis Valley. This includes both direct emissi ons and upstream emissions. . 33 ! Figure 5: The breakdown of GHG emission contributions of the four major categories averaged between 2006 and 2012. ................................ ............. 36 ! Figure 6: Change in storage of the unconfined aquifer in SLV from 1975 2016 (Davis Engineering 2016). ................................ ................................ ............ 47 ! Figure 7: Geospatial representation of crop production in SLV. ......................... 52 ! Figure 8: Average annual water allocation in SLV from 2000 2010 (1 Ac ft. = 1233.5 m 3 ). ................................ ................................ ................................ ... 58 ! Figure 9: The crop water use allocated between CWU green crops , CWU surface crops , and CWU ground crops as well as the consumptive water content of each crop (2000 2010)*. ................................ ................................ ................................ 60 ! Figure 10: Livestock consumptive water use (CWU livestock ) for SLV allocated between CWU green crops , CWU surface crops , and CWU ground crops as well as the

PAGE 20

xx consumptive use content of e ach crop. Cattle represent 98% of all CWU livestock in the region*. ................................ ................................ ............. 62 ! Figure 11: Monthly water availability and monthly water use (2000 2010) (1 Ac ft. = 1233.5 m 3 ). ................................ ................................ ................................ 65 ! Figure 12: A summary of the key aspects of a Community Based Participatory Research project (Israel et al. 1998). ................................ ............................ 73 ! Figure 13: Framework for community engagement in San Luis Valley. The number in the circle indicates the order of each step. * The scenario modeling methods and results are presented in Chapter V. ......................... 77 ! Figure 14: Interview guide used to spark discussion when informally engaging with residents of SLV. ................................ ................................ ................... 78 ! Figure 15: Regional map of San Luis Valley with geographic distribution of the Community Advisory Board members*. ................................ ........................ 81 ! Figure 16: San Luis Valley with aquifer boundaries mapped in yellow and irrigation wells in green (Colorado Geological Survey 2016). ....................... 88 ! Figure 17: San Luis Valley with groundwater pumping response areas outlined in purple (Sub districts fall within the boundary of response areas) (CDWR 2015d). ................................ ................................ ................................ .......... 89 ! Figure 18: The four IPCC emissions scenarios, presented in the 1990 summary to policymakers (IPCC Working Group I 1990), expressed in both radioactive forcing and equivalent carbon dioxide concentrations*. ............................... 98 !

PAGE 21

xxi Figure 19: San Luis Valley with groundwater pumping response areas/sub districts outlined in purple (CDWR 2015d). ................................ ................. 102 ! Figure 20: The impact on the GHG emissions indicator from the crop regime change scenario*. ................................ ................................ ....................... 118 ! Figure 21: The average agricultural (crops and livestock) consumpti ve water use (black line) shown with the average water budget (average water inflows minus compact deliveries) (pink line) in San Luis Valley*. .......................... 121 ! Figure 22: The impact on the GHG emissions indicator when the community response solar development pathway is implemented*. ............................. 125 ! Figure 23: The impact on the GHG emissions indica tor when the DOE/BLM solar development pathway is implemented*. ................................ ..................... 126 ! Figure 24: Surface soil texture distribution among all irrigated lands in SLV. Error! Bookmark not defined. ! Figure 25: Surface soil texture distribution among all alfalfa lands in SLV. ... Error! Bookmark not defined. ! F igure 26: Surface soil texture distribution among the alfalfa lands sample in SLV. ................................ ............................... Error! Bookmark not defined. ! Figure 27: Surface soil texture distribution among all potato lands in SLV. ... Error! Bookmark not defined. ! Figure 28: Surface soil texture distribution among the potato lands sample in SLV. ................................ ............................... Error! Bookmark not defined. !

PAGE 22

xxii Figure 29: Surface soil texture distribution among all the small grain lands in SLV. ................................ ................................ ....... Error! Bookmark not defined. ! Figure 30: Surface soil texture distribution among the small grain lands sample in SLV. ................................ ............................... Error! Bookmark not defined. ! Figure 31: Surface soil texture distribution among all meadow/pasture lands in SLV. ................................ ............................... Error! Bookmark not defined. ! Figure 32: Surface soil texture distribution among the meadow/pasture lands sample in SLV. ................................ ............... Error! Bookmark not defined. !

PAGE 23

1 CHAPTER I INTRODUCTION Background Human ity's future depends on the ability of the built environment to work within the constraints of the planet's natural systems. For the past century , society has been in a period of rapid growth, made possible in large part to the discovery and wide use of fossil fuels (Smil 2010) . In his final work, Howard T. Odum, likened the evolution of society to that of a forest (Odum and Odum 2001) . He noted that during primary and secondary succession, the health of a forest is measured by growth, however in a mature climax forest, health is measured not by growth, but by the forest's ability to maintain a high level of biodiversity and to quickly rebound from natural disasters . Inspirational ideas from the natural world such as this have given rise to the field of sustainability science (Kates et al. 2001) and resilience thinking (Gunderson and Holling 1996) . Resilience is defined as the ability of a system to respond to a stress while maintaining system identity and primary functions (Walker et al. 2004) . Indicators such as G ross D omestic P roduct (GDP) only tell us how well our societal system is growing. D eveloping m easures necessary to evaluate society in this climax oriented stage , Odum claims , is humanity's greatest challenge. We must identify and make use of new measures that provide meaningful information for all stages of human/ earth co existence (England 1998; Costanza et al. 1998; Lawn 2003) .

PAGE 24

2 Sustainability Science This research contri butes to the growing field of sustainability science . The definition of sustainability, as articulated by the United Nations World Commission on Environment and Development , is: "Development that meets the needs of the present generation without compromising the future generations ability to meet their own needs" (Brundtland 1985; United Nations Division for Sustainable Development 1992) . Planning and policy making that incorporate s sustainable growth princi ples at a regio nal/local level is increasingly recognized as an important practice (Ali 2013; Graymore et al. 2010) . Entire frameworks have been developed to help infrastructure designers ( e.g., planners and engineers), policymakers, and individual users understand the complex nature of modern society, and to provide a forum for communication a cross disciplines (Ramaswami et al. 2012; Ramos 2010; Dovers 1995; Fiksel et al. 2012) . Fostering open lines of communication between v arious actors in society which normally do not interact, is a relatively new endeavor that has been triggered by a need/demand for sustainability (Ramaswami et al. 201 2) . These open lines of communication are critical to assist l ocal government , who must often balance the competing int erests of their constituents , in foster ing economic growth while also managing for sustainability . However, policymakers do not always have tools necessary to measure the sustainability implications of their actions. Tools for measuring economic growth are well known and commonly used, and policymakers can easily look back over the previous years to se e how activities in their region have correlated

PAGE 25

3 to changes in GDP and Net Domestic Product (NDP). The se trends can be used to project how policy changes may affect those measures into the future. It is important that the analysis not end t here. Decisio n makers must also understand how policies fit into the context of the three pillars of sustainability (Janou 2012) . The pillars are the economic, social , and environmental components of a human system , and all three must be sustainable for the system to be considered sustainable (United Nations 2002) . It is no longer sufficient to simply measure the economy Ñ there is a need to assess the environmental resources on which th ose econom ies are based. Measuring Sustainability It is difficult to measure how policies and activities in a region will affect sustainability. To address this issue, much work has been done over the last decade in the development of sustainability metrics and indicato rs (Morse 2015; Zidansek et al. 2014; Ness et al. 20 07; Ramaswami et al. 2008) . Sustainability i ndicator s measure one characteristic of a system (e.g., CO 2 emissions (IPCC 2006; WRI 2004; U.S. EPA 2013a) or biodiversity (Tasser et al. 2008) ), where as a metric combines many indicators (or variables) through aggregation (e.g., G reen N et R egional P roduct (GNRP) (Hamilton et al. 1997; Kahn 1995) ) . Sustainability me asures can be computed over time to gain an unders tanding of how resource usage is changing . There are numerous types of sustainability metrics and indicators that now exist as tools for policymakers (Mori and Christodoulou 2012; Mayer 2008) . A number of geographic regions have

PAGE 26

4 recognized the importance of measuring sustainability, and have developed action plans based on sustainability metrics and indicators (Gallucci 2013; Reed et al. 2006; Obama 2013) . Rural Regions Much of the research into sustainability metrics and indicators has been focused on countries, but increasingly , focus has turned to cities and urban areas, understandably , because urban population centers are experiencing some of the fastest growth across the globe (Kennedy et al. 2010; Hillman and Ramaswami 2010) . This research , however, does not focus on the consumption heavy urban regions, but rather on the surrounding hinterlands . Consumption based metrics and indicato rs at the city scale must be coupled with studies that look at the production based regions that help support cities to get a full picture of the specific system in question . According to Amory Lovins (Lovins and Hawken 2007) , " N atural Capital " is the fundamental basis for all economic growth . This is because natural resources and ecological systems provide vital life support services and resources to society . It is often the case that city scale assessments will use state or national level data because local/ region specific studies are lacking (Ramaswami et al. 2008) . The reliance on natural resource s (i.e. , land, water , etc.) for economic growth is clearly evident in agricultural regions. A need exists for analysis of these regions from a su stainability perspective to both improve city scale assessments as well as inform policymakers in rural areas.

PAGE 27

5 Previous Work The United States Environmental Protection Agency (EPA) Office of Research and Development created a methodology for computing sust ainability metrics at a regional scale. They began their effort to quantify sustainability using a 7 county region as a pilot study . T he region was encompassed by the San Luis Valley (SLV), Upper Rio Grande river basin ( SLV ) , and the Great Sand Dunes Nat ional Park and Preserve (U.S. EPA 2010) . This region was selected due to its limited population, distinct physical and political boundaries, the strong support of the local population, and remote geographic location, whereby tracking inputs and outputs was expected to be straightforward (U.S. EPA 2010) . The purpose of EPA's research was to identify metrics for sustainability to capture fundament al aspects of the system and apply those metrics to a geographic region over time to assess whether the system is moving toward or away from sustainability (Eason and Cabezas 2012; Heberling and Hopton 2012) . The four elements of the system EPA selected to measure were the human burden on the environment, economic well being , energy flows, and system order. These properties were selected becau se they represent one or more component of the three pillars of sustainability . E PA used data specific to the region, or scaled to the region when necessary, and computed the four sustainability metrics over time . EPA published a number of journal articles and a report presenting the results of the research (Ho pton and White 2012; Heberling et al. 2012; Campbell and Garmestani 2012; Eason and Cabezas

PAGE 28

6 2012; U.S. EPA 2010) . The four metrics used to assess sustainability in the region are discussed below. Ecological Footprint Analysis To measure environmental b urden , EPA conducted an Ecological Footprint Analysis (EFA) (Hopton and White 2012) . EFA is a consumption based metric that calculates the amount of produc tive land needed to support the population in a given area (Wackernagel and Rees 1996) . EFA's strength is that it i s a relatively straightforward methodology with easy to understand results. EPA's simplified approach to EFA used 35 variables for the 26 year period (1980 2005). Data were collected at the smallest geographic level available (i.e., municipality, county, region, state, country) . Emergy Analysis To measure energy flows , EPA conducted an Emergy Analysis (Campbell and Garmestani 2012) . Emergy Analy sis is a method to track the quantity of embodied solar energy (unit = "solar emjoules" or "emergy") in various items and economic activit ies /services (Tilley 2004; Baral and Bakshi 2010) . It incorporates the accumulation of stored e n ergy in a system as well as the movement of e n ergy in the form of good s and services across boundaries. As a simplified example , energy from the sun reac hes earth and is store d in photosynthetic organisms (both living and in the form of fossil fuels ( i.e. , coal, oil, and natural gas ) ) . These energy reserves are mined and utilized for work. Work performed, which requires a certain quantity of coal, could also be thought of in terms of the

PAGE 29

7 quantity of solar radiation needed to produce that quantity of coal (Odum 1971) . This tracking of energy flows c an be done with fossil fuels as well as with other goods and services in the economy. The accounting of these flows of energy, and human and natural resources needed to p rovide these goods/services to the economy, is an Emergy Analysis (Tilley 2004 ) . The EPA 's Emergy Analysis considered e mergy flows in the form of renewable and non renewable energy, import and export of goods and services, and other key materials exchanges (e.g., agricultural chemicals, building materials etc.) . Green Net R egion al Product The idea that sustainability can be measured by GDP alone is now understood to be short sighted , and Green Net Regional Product (GNPR) adds to GDP by capturing the externalities associated with natural resource depletion through alternative valu ation methods (Pezzey et al. 2006; Mota et al. 2010; Heberling et al. 2012) . GNRP is an environmental economic metric that adjusts th e NRP to account for natural resource depletion in dollars . This process assumes that man made capital cannot simply be substituted for natural capital, but that the loss of natural capital reduces the utility for future generations (Lovins and Hawken 2007) . Fisher Information Fisher Information (FI) comes from information theory, and measures the order or predictability of data (Cabezas and Karunanithi 2008; Cabezas and Fath 2002) . It is a method of quantifying the information content in data that correlates

PAGE 30

8 with order in the system dynamics . This is important because order is a marker of well functioning natural and human systems . FI, as a sustainability metric, requires tracking of multiple important system variables simultaneously over time. To do this, a MATLAB program was developed by EPA (Eason and Cabezas 2012) to analyze data and detect whether the system is stable. EPA included 53 variables that were each classified into one of six categories: Consumption, Demographic, Energy, Environment, Land, and Production. Present Work This present research, which was fun ded by EPA Office of Research and Development , builds on the previous work performed in SLV. It utilizes the same geographic/political boundaries as the previous EPA study and refers to the region simply as SLV ( Figure 1 ). For the purpose of data collection , the complete counties of Saguache, Rio Grande, Mineral, Hinsdale, Alamosa, Conejos, and Costilla were considered . SLV is roughly 21,000 km 2 in area, it has an average elevation at the valley floor of 2,300 m (7,600 ft.) , and the high desert receives an average of 222 mm (9 in . ) of precipitation annually ( Figure 2 ) (PRISM Climate Group 2016) .

PAGE 31

9 Figure 1 : The San Luis Valley . Whole counties were always considered for data collection even though parts of the county may fall outside of the Upper Rio Grande river b asin geographic boundary.

PAGE 32

10 Figure 2 : Average monthly precipitation (1980 Ð 2010 ) during the growing season (May to August) in SLV (PRISM Climate Group 2016) * . * The units are in inches of precipitation (1 inch of rain in an area is enough to evenly cover the ground in that area with a layer of water 1 inch deep). (1 inch = 25.4 mm) Agriculture is the major economic sector in the region . I t is an ideal location for growing potatoes, barley (much of the barley for Coors Brewing Company ¨ beer is grown here), and alfalfa . An interesting note, the region was the first place in North Am erica to grow quinoa for commercial use due to its Andean like climate (Thier 2010) . Though the 7 county region spans more than Precipitation (in.) Precipitation (in.) Precipitation (in.) Precipitation (in.)

PAGE 33

11 21,000 km 2 , the population consists of only 50,000 people (U.S. Census Bureau 2012) . SLV has the oldest town in the state (San Luis) (History Colorado 2015) , t he highest percentage of Hispanic residents (47%) in the state (SLV Development Resources Group 2013) , and is home to some of the poorest counties in Colorado (U.S. Census Bureau 2 011) . J uxtaposed to the poverty there is prosperity for some potato, grain, and alfalfa farmers who are able to utilize the underlying aquifer to irrigate crops for export. More details on the region are presented in subsequent chapters. In this resea rch, t wo specific sustainability indicators were added to the suite of metrics developed by EPA: a g reenhouse g as (GHG) accounting indicator and a consumptive water use indicator. GHG emissions are an indicator of global climate change, which is considere d one of the most pressing human rights, security, and environmental issues of our time (Gosling and Arnell 2016; Barnett 2003; Douglas et al. 2012; Stallworthy 2017) . T hree of the EPA metrics (EFA, GNRP, and FI) relied on scaled down state level GHG emissions for their analysis because region specific data were lacking . The agricultural sector alone accounts for >10% of global anthropogenic GHG emissions (IPCC 2014) . GHG accounting in rural agricultural regions is key for developing region specific mitigation strategies to re duce global GHG emissions (see C hapter II). Water use accounting is also critical for agricultural regions. In these regions the economy rests on water availability, and in the arid SLV , understanding water use and ways to reduce water consumption in critical for sustainability (Gibson

PAGE 34

12 and et.al. 2015) . Drought and overuse has threatened the regions underlying aquifer, which is relied on for the majority of the irrigation water for crops (CDWR 2015a) (see C hapter III). C arbon and water are also tied together for reasons including : 1) increased drought and water shortages have be en linked to climate change (Malcolm et al. 2012; Calzadilla et al. 2013) , 2) pumping water (including irrigation water) requires energy which in turn results in more GHG emis sions from electricity production (Hussey and Pittock 2012; Rothausen and Conway 2011) , and 3) improving soil health leads to greater water holding capacity of the soil (which reduces irrigation requirements (Altieri 1999) ), and increasing soil carbon , which is a marker of soil heal th, can reduce atmospheric GHGs (Powlson et al. 2011) . The GHG emissions and water use indictors developed for the region we re done in collaboration with local experts in SLV . In addition, this research proposes a framework for engaging rural ag ricultural communities around issues of sustainability, which includes a methodology for calculating sustainability indicators , over time, and developing some likely future scenarios using the indicators (see Chapter IV). A goal of this research is to emp ower the community to make informed decisions and develop with them tools for sustainable management. The only way to know where you are going is to understand where you have been (Santayana 1905; Ives and Boatwright 1999) . This is true in life for a single person and it speaks truth to planning and policymaking for our society and its relationship with natural system s that support society. The

PAGE 35

13 following sections provide an outline describing the content of this dissertatio n. Arrangement of the Dissertation Chapter I (current chapter) introduces the work and provides the reasoning, need, and purpose for doing it. Chapter II provides an overview of the need for and importance of GHG accounting and details the methods and results of the GHG accounting indicator model . Climate change, which has been e xacerbated by human induced GHG emissions, may disproportionately impact the agricultural sector (Malcolm et al. 2012) . Here , a region specific GH G accounting indicator i s presented for agricultural regions, an often overlooked area. Chapter III provides an overview of the importance of consumptive water use accounting and details the methods and results of the consumptive water use indicator model . Issues surrounding water use are particularly critical to SLV because of limited rainfall and a depleting aquifer . Moreover, the state has over appropriated water from rivers and streams , as well as groundwater resources (Poppleton 2013; Fryar 2008) . Because water is such an important reso urce to this agricultural economy, i t is crucial to have a clear measure of how different practices affect water resources in this region. The SLV community has been grappling with sustainable water management for decades, and this research presents a new perspective on wa ter consumption in the region. Chapter I V details the experience of engagement with the local community over the course of the research . During the process of developing the GHG

PAGE 36

14 accounting and consumptive water use indicators , there were data that were not readily available. To find these data , a number of in person visits were made to the region . The purpose was to meet with people from local government, farmers and ranchers, and others stakeholders who possess information and perspecti ve s needed for this research. Through meeting with local stakeholders , it became clear that developing a formal community engagement process would be key for the work. This resulted in the development of a Community Advisory Board (CAB). In addition to satisfying specific data needs , the CAB allowed for a forum to transition the resulting tools to the community, which was requested by the EPA (the grantor of this research) . To transition the methodology and tools, a group of stakeholders attended a seri es of seminars where they were taught how to compute and use the original four metrics and additional GHG and consumptive water use indicator models. A timeline of community engagement ( Figure 3 ) i s below, and the details of the process are outlined in Chapter IV .

PAGE 37

15 Figure 3 : Community engagement timeline in the SLV during the course of this research . Chapter V describes the future scenario m odeling phase of the research. The ability for regions to develop plans and implement actions toward sustainability hinges on a clear historic accounting and baseline scenario. Once a region specific baseline was established in SLV , which included updati ng the four metrics developed by EPA from 2005 to 2010 as well as developing the GHG accounting indicator (C hapter II) and the co nsumptive water use indicator (C hapter III), the research moved into the future scenario modeling phase. The future scenarios selected for proof of concept modeling were conceptualized with the CAB . The future scenarios were modeled using the GHG accounting and consumptive water use indicators to produce a picture of several potential futures in the region. These proof of concept future scenarios were developed to a) ad vance the science of sustainability modeling and b) add to a foundation of knowledge for the local community.

PAGE 38

16 The final chapter, Chapter VI , summarizes some of the major findings and outcomes from the research. A concise review of each of the chapter's main results is presented and discussed. In addition, areas for potential future work beyond the scope of this research are offered.

PAGE 39

17 CHAPTER II GREENHOUSE GAS ACCOU NTING Greenhouse Gas Accoun ting of Rural Agrarian Regions: The Case of San Luis Valley Abstract Rural regions, with a dominant agricultural economic base, have a vastly different greenhouse gas emissions profile than urban regions and hence require a unique accounting method. This chapter presents a greenhouse gas (GHG) inventorying methodology tailored specifically for rural agricultural regions. The methodology was applied to San Luis Valley (SLV) in south central Colorado with intent to establish a clear emissions baseline and to analyze, in fine detail, the regions emission s profile. The results show that SLV has an annual per capita emission rate of 30.5 MT CO 2 equivalents (CO 2 e) whereas the average for the United States is about 2 1 MT CO 2 e. The higher per capita emissions can be attributed to the production of agricultural goods and services that are exported rather than consumed. Because per capita emissions might not paint an accurate picture for export based economies , data we re recalibrated on a per dollar GDP basis. O n this basis, SLV emissions are almost twice that of the national average , indicating that, with all things being equal, agricultural activities contribute disproportionately towards GHG emissions. Through a d e tailed quantitative analysis, the results show that SLV, with its significant solar resource, has the potential to offset much or all of their emissions. The findings

PAGE 40

18 from this work offer useful insights for local stakeholders to develop plans and impleme nt policies towards GHG mitigation. Introduction Awareness in the global scientific community and in the general public about greenhouse g as (GHG) inventories and their benefits for policymakers in addressing climate change has greatly increased in the las t decade (Obama 2013, 2009) . In addition, communities and governments play an important ro le in reducing GHG emissions as they have broad influence over activities that result in significant direct and indirect GHG emissions within their boundary and jurisdiction. GHG inventories create a baseline that can be used to identify sectors, sources, and activities responsible for GHG emissions, assess relative contributions of emission sources, establish local climate action plans and policies, quantify benefits of activities that reduce emissions, and foster informed communication with stakeholders (U.S. EPA 2015a) . Previous efforts to baseline communities, such as those led by the Intergovernmental Panel on Climate Change (IPCC), had primarily focused at the national level. However, with the leadership of organizations such as the International Council for Local Environmental Initiatives (ICLEI) and the World Resources Institute (WRI) there has been a push to look more specifically at cities and urban areas because of their dense population and high consumption (Ramaswami et al. 2008; IPCC 2006; Arikan et al. 2012) . This work is o ne such attempt to conduct a GHG baseline analysis on a more local level , but focused on a rural agrarian region.

PAGE 41

19 Typical city scale GHG inventories account for residential, commercial/industrial energy use (primarily in boundary buildings/facilities), and transportation. In addition, hybrid methods account for cross boundary contributions associated with urban material consumption (e.g. , cement, fuel, food , etc.) and transportation (surface and air). Different accounting approaches, ranging from pure geo graphic production based accounting, pure consumption based accounting to hybrid geographic Ð plus key infrastructure supply chain accounting have been developed at the city scale (Ramaswami et al. 2008; Lenzen and Crawford 2009; Hillman and Ramaswami 2010; Dodman 2009) . In addition to accounting for in boundary emissions allocation, issues related to trans boundary emissions and life cycle supply c hain emissions have been addressed (Ramaswami et al. 2008) . These inventories have been used for future planning in cities addressing water, energy, and material needs of urban communities. Whereas these studies have been influential for cities and nations in making plans and implementing actions towards reducing carbon emissions, very little work has been done on explicit accounting of GHG emissions for rural regions. According to the IPCC, land use accounts for more tha n 24% of the world's overall GHG emissions , and agri cultural production is the largest contributor to that sector (IPCC 2013) . The U.S. Department of Agriculture (USDA) also recently published a report that shows that the agricultural sector will be one of the hardest affected by a changing climate (Malcolm et al. 2012) .

PAGE 42

20 Moreover, rural agricultural regions, although small in population , have a disproportionate influence on global GHG emissions. This is a key reason why engaging with an agricultural community on GHG inventories is so impor tant. This rural GHG inventory examined the agricultural region known as San Luis Valley ( SLV ) in south central Colorado . T he U.S. Environmental Protection Agency selected this region as a pilot study to develop and test sustainability metrics partly beca use of its isolated geography and well defined agricultural economy (Heberling and Hopton 2012; Heberling et al. 2012; Hopton and White 2012; Campbell and Garmestani 201 2; Eason and Cabezas 2012) . The SLV is a 100 mile long and 60 mile wide ( 21,000 km 2 ) upland agricultural valley surrounded by the 14,000 foot (4,400 m) peaks of the Sangre de Christos to the East and the expansive San Juan wilderness area to the West. The valley floor sits at 7,500 f t. ( 2,300 m ) and is an ideal location for growing potatoes (it is the second largest potato produc ing area in the United States next to Idaho) and barley (the majority of Coors Brewing Company ¨ barley is grown here). The sphere of influence of local government is typically limited to its geographic jurisdiction. Therefore , urban policymakers have limited ability to direct meaningful improvements to their emission profile in categories such as food, becau s e they are consum ers of food and not the producers. Therefore, engaging rural food producing regions through local carbon accounting metrics is key for fostering policy change. Because the emission profiles in agricultural regions are widely different than those in urban regions, there is a great need to

PAGE 43

21 develop context specific accounting approaches for rural regions that capture activities such as agricultural energy demand, livestock raising, and soil GHG fluxes. Further, in view of the wide variability in agricultura l practices (e.g. , tilling practices, crop rotations, manure management, groundwater vs. surface water use), and soil and climatic conditions, these inventories need to be based on local context specific, bottom up emission factors derived from regional da ta (Fraser et al. 2 006) . With a robust and descriptive region specific baseline GHG accounting, local decision maker s can begin the process of developing plans and implementing actions towards GHG emissions reductions. Methodology This study presents a novel method of ru ral agricultural region GHG accounting and data reporting. The methodology for developing and conducting this inventory uses the IPCC 2006 release of GHG inventory for nations as well as other GHG inventories in the field for structure and inclusions (IPCC 2006; WRI 2004; Ramaswami et al. 2008) . The GHGs considered in this study are CO 2 , CH 4 , and N 2 O and are presented as carbon dioxide equivalents (CO 2 e) based on the IPCC 2013 report (IPCC 2013) . The sectors of interest, consistent with the IPCC, are Energy and Buildings, Transportation, Agriculture, Land Use, and Waste . B ecause manufacturing makes up only about 1% of the economy in the region (SLV Development Resources Group 2013) , the Industrial Processes and Product Use sector was excluded from this study. In addition to in boundary emissions associated with the sectors described above , we also looked at

PAGE 44

22 upstream life cycle emissions when relevant. Also, producing an inventory that was reprodu cible not just by the scientific community, but by our local community partners in the region was essential . Six months were spent performing a stakeholder analysis based on principles found in Community Based Participatory Research (CBPR) that led to the formation of a Community Advisory Board (CAB) (Israel et al. 1998) . The CAB , which was composed of stakeholders from the water user community (farmers and ranchers), local government, conservation groups, federal land managers and others from across the region , provided insights in to the unique nature of the area as well as aided in data collection and provided critical review. Moreover, because of the explicit goal of transferring this indicator model and methodology to the community, it was attempted, whenever possible, to use pu blicly availabl e data such as census data (U . S . Census Bureau), National Agricultural Statistics (NASS), Colorado Department of Transportation (CDOT) , and others that rely as little as possible on complex computer models for the inventory. The information below shows how data were collected and the emission factors that were selected for each of the sectors. Each sector's material flows, emission factors, and sources are summarized in Table 1 . Energy and Buildings Sector Electricity: Annua l electricity consumption in SLV was obtained from the two utility companies Ð Xcel E nergy (Xcel) and San Luis Valley Rural Electric Cooperative (SLV REC) Ð who supply electricity to the region. Total consumption

PAGE 45

23 was 222,023 MWh for Xcel and 211,993 MWh for SLV REC in 2012. Further, each company provided the residential and commercial breakdown of the electricity consumption, and the amount of e lectricity used for agricultural irrigation. T he life cycle GHG emission factors from Xcel and SLV REC included both direct emissions from power plants as well as upstream emissions from the mining and handling of the raw resource, namely coal and natural gas. The direct emission factor for Tri State Generation and Transmission (Tri State), from whom SLV REC purchases 99% of its electricity (Waudby 2015) , was based on emission factors of individual power plants that p rovide electricity to Tri State (Tri State 2015) . These power plant specific emission factors were obtained from the EPA 2012 eGRID dataset (U.S. EPA 2012a) . The electricity grid mix for SLV is 60% coal, 25% natural gas, and 15% renewable. The direct emissions of SLV REC grid mix were determined to be 0 .76 kgCO 2 e/kWh , whereas the direct emission factor for Xcel was determined to be 0.75 kg CO 2 e/kWh (Xcel 2013) . Emissions upstream from the power plant due to mining and transport of coal and natural gas was estimated as 25% and 6% of the total emissions , respectively (Spath et al. 1999; Spath and Mann 2000; Heath et al. 2014; Whitaker et al. 2012) . A combined average emission factor for the region was calculated as 0.83 kgCO 2 e/KWh. Natural Gas and Propane: Community wide natural gas usage for residential and commercial purposes was directly ob tained from Xcel, which is the s ole natural gas supplier to SLV. The annual residential and commercial

PAGE 46

24 natural gas consumption was 762 million MJ and 646 million MJ, respectively. Annual propane consumption was estimated for the region based on the numbe r and square footage of homes in the region and the number of homes using propane for heat (U.S. EPA 2010) . In addition, this estimated value was verified through contact with the major propane providers in the region. The life cycle emission factor for the use of natural gas and propane in a furnace, which includes both direct a nd upstream emissions, was 9.35 kg CO 2 e/gallon and 7.03 kg CO 2 e/gallon, respectively (Register 2009; Howarth et al. 2011; Heath et al. 2014) . In 2012, SLV emissions from natural gas and prop ane were 119,000 MT CO 2 e . Agriculture and Land Use Sector Livestock Emissions: Methane, a potent GHG, is produced as a by product of enteric fermentation (digestion) in livestock and the amount of methane that is released depends on the type of digestive tract, age, and weight of the animal as well as the quality and quantity of the feed consumed (IPCC 2006) . SLV h as significant livestock operations consisting mainly of cattle ranching with relatively small sheep and hog operations. For 2012, NASS reports head of cattle in the region as 83 , 000, head of sheep as 12 , 700, and head of hogs as 220 (U.S. Department of Agriculture 2015) . IPCC suggests a regionalized tier 2 approach for calculating the emission factor for cattle and a generalized tier 1 approach for sheep, goats, and swine (IPCC 2006) . To reflect the variation in emission rates among cattle, instead of using the North American

PAGE 47

25 average of 53 kg CH 4 /head as reported by IPCC, a region specific emission factor was developed by first categorizing the her d into IPCC suggested subgroups. Then using local data for climate, feeding situation, age, size, and cattle subgroup and the IPCC model, a herd specific emission factor of 50 kg CH 4 /head was calculated (Whitten 2015) . Followi ng IPCC tier 1 approach, the global industrialized average emission factors of 5 kg CH 4 /head and 1 kgCH 4 /head for sheep and hogs, respectively, were used. For manur e management, an IPCC tier 1 approach for each of the three animal types was used (IPCC 2006) . Cattle manure management practices in the region produce 2 kg CH 4 /head, sheep management produces 0.2 kg CH 4 / head, and hogs produce 13 kg CH 4 / head accordi ng to the IPCC tier 1 approach. In 2012 livestock in t he SLV were responsible for 145,000 MTCO 2 e . Soil Nitrous Oxide Emissions: The use of fertilizers and managed soils is a major contributor to GHG emissi ons globally. This section describes the estimation of N 2 O emissions from managed soil due to agricultural inputs of nitrogen (synthetic N fertilizers; N deposited by grazing animals, and decomposing crop residue). Because actual data were not available, average quantities of nitrogen applied per acre by crop type and crop rotation was estimated through in depth interviews with two key agriculture consultants who work with local farmers in the Valley (Oniel 2015; Dillion 2015) . Based on their inputs, the following average rate of nitrogen application was used in this study: 180 lbs. /acre for small grains after a potato crop , 80 lbs. /acre for small grains

PAGE 48

26 after an alfalfa crop , 220 lbs. /acre for continuous small grains, 11 lbs. /acre for alfalfa, and 185 lbs. /acre for potatoes after a small grain crop . Local data on N inputs through grazing animals were based on animal head counts and their feeding situation (IPCC 2006) . The IPCC guidelines were followed to estimate GHG emissions from managed soils , and both direct N 2 O emissions from nitrification and denitrification of N inputs, as well as indirect N 2 O emissions due to N volatilization, leaching, and run off were accounted . Upstream emissions associated with the production of nitrogen fertilizer were obtained from Blonk Consultants (Zeist et al. 2012) , and were found to be 4.0 kg CO 2 e/kg N applied based on the North America average fertilizer mix . In 2012 there was an estimated total of 35 million lbs. of nitrogen applied, which in turn was responsible for 140,000 MT CO 2 e emissions . Agricultural Machinery: The on farm fuel use from tractors, farm trucks, and processing equipment was estimated using data reported by NASS on dollars spent on agricultural fuels in the region (U.S. Department of Agriculture 2015) with all fuel consumption assumed to be diesel and consumed on the farm. The cost per gallon for diesel was based on the repor ted price for western states. All fuel was road tax exempt red diesel, so the Colorado road tax was subt racted from the regular fuel price (U.S. EIA 2016) . The life cycle emission factor for diesel fuel was obtained from Argo nne National Laboratory's Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation Model ( GREET) (Argonne National L aboratory 2015) . This model reports the pump to wheels

PAGE 49

27 (PTW) emission factor (direct) for diesel as 9.18 kgCO 2 e/gallon, and the wells to pump (WTP) emissions (upstream) as 2.3 kgCO 2 e for both gasoline and diesel. In 2012 , the Valley consumed 6.7 milli on gallons of farm diesel, which accounts for an annual emission of 77,000 MTCO 2 e. Transportation Sector On Road Cars and Trucks: Direct tailpipe emission from on road transportation was calculated by multiplying vehicle miles travelled (VMT) in the region by the emission factors for gasoline and diesel fuel. The total VMT in the region was collected using Colorado Department of Transportation (CDOT) static highway reports (CDOT 2012) , which provided annual VMT data by county divided up into two classes, lightweight personal vehicles (class 1 3) and larger commercial vehicles (class 4 13). However, these publicly available data capture only major roadways in the reg ion, and do not include traffic through the minor county roads, which can be significant for rural regions. In order to include the smaller roads, region specific data were obtained directly from CDOT based on their traffic model to estimate "off grid" VMTs in the seven county region. All travel within the region's boundary was allocated to the region because there was not a valid way of disassociating traffic that was just passing t hrough, and based on the mountainous and isolated geography of the region , it was a ssume that p ass through trips are minimal. The total lightweight personal vehicle traffic in the region in 2012 was estimated at 1.4 million VMT per day and the normalized personal VMT in the region yielded 30 VMT/person/day which is very close to the

PAGE 50

28 national average of ~28 VMT/person/day (U.S. DOT 2012) . The total commercial large truck (class 4 through 13) traffic for the region in 2012 was around 140,000 VMT per day. In order to estimate the weighted ave rage fuel economy for the personal vehicle fleet in the region, vehicle registration data were obtained from the San Luis Valley Development Resources Group (SLVDRG) (SLV Development Resources Group 2013) . The average fuel economy for lightweight personal vehicles (class 1 3) was estimated as 25.6 miles/gallon. For co mmercial large trucks (class 4 13) the national average fuel economy of 7.3 miles/gallon was used in this study (U.S. DOT 2015) . The GREET model was again used to derive the life cycle emissions factors for gasoline and diesel. The pump to wheels (PTW) emission factor (direct) for gasoline was 8.71 kgCO 2 e/gallon and for diesel it was 9.18 kgCO 2 e/gallon. The wells to pump (WTP) e missions (upstream) were 2.3 kgCO 2 e for both gasoline and diesel (Argonne National Laboratory 2015) . The on road emissions from lightweight personal vehicles in 2012 were estimated as 230,000 MTCO 2 e and the emissions from commercial large trucks were 79,000 MTCO 2 e. Waste Sector Landfill emissions: There are two major solid waste facilities in the region: SLV Reg ional Landfill and Saguache County Landfill and Recycling Center. Data of municipal solid waste (MSW) generated in the region were obtained from the Colorado Department of Public Health and Environment (CDPHE) and were reported as 63 thousand tons (1.4 to ns per capita) in 2012 (CDPHE 2015) .

PAGE 51

29 Landfill emissions wer e estimated using EPA's Waste Reduction Model (WARM) (U.S. EPA 2015b) . SLV regional, the larger of the two sites, provided th eir waste composition broken into three categories: household (87%), construction (11%), and other (2%). WARM required further composition detail for which the U.S. average data from the EPA was used (U.S. EPA 2013b) . The transportation portion of the analysis was omitted from the WARM model because those emissions would have been accounted for in the transportation section of this GHG accounting. The results from WARM show landfill emissions of 56,000 MTCO 2 e that can be attributed to solid waste generated. Waste Water Treatment: The quantity of emi ssions from wastewater depends on the treatment type and , in SLV , was either rural septic or centralized aerobic treatment. Data needed were the number of people living in the region and the percent of those people living in cities and in rural areas. It was assumed that all city population used centralized aerobic treatment and all rural population used anaerobic septic treatment. This is a critical distinction because anaerobic treatment produces higher CH 4 emissions according to IPCC. In 2012, of the nearly 47,000 people living in the Valley, 50% were in cities and 50% reside d in rural areas (U.S. Census Bureau 2012) . Using these data , an emission factor of 0.24 MT CH 4 /MT BOD was developed for the Valley. Biological Oxygen Demand (BOD) in the wastewater was estimated using the average BOD concentration per person for North America (85g/person/day)

PAGE 52

30 (IPCC 2006) . The total treatment demand in SLV was estimated at 1,482 MT BOD and the total associa ted annual emissions were 12,000 MT CO 2 e. Solar Credit Data on utility scale solar energy production were obtained for this study and allocated in the form of carbon credits to the region. The total solar capacity in SLV in 2012 was 87 MW , wh ich produced 198,363 MWh of electricity (U.S. EIA 2015) . The upstream emission factors for the production of the power plants were 0.04 kgCO 2 e/KWh for photovoltaic (PV) and 0.02 kgCO 2 e/KWh for concentrated solar power (CSP) (Burkhardt et al. 2011; Hsu et al. 2012; Kim et al. 2012) . A carbon credit was allocated to the region for the total amount of utility scale solar produced b ased on the avoided emissions.

PAGE 53

31 Table 1 : Material flows, emission factors, and total GHG emission s in the San Luis Valley. Regional material or energy flow (MFA) Data year MFA data source GHG emission factor (use phase) GHG emission factor (upstream) EF data source Total GHG emitted = MFA x EF (MTCO 2 e) Energy and Buildings Electricity 434,016 MWh 2012 (Dallinger 2015; Waudby 2015) 0.7 6 kgCO 2 e/kWh 0.075 kgCO 2 e/kWh (Heath et al. 2014; Xcel 2013) 362,000 Solar Credit 198,363 MWh 2012 (U.S. EIA 2015) 0.83 kgCO 2 e/KWh 0.04kgCO 2 e/ KWh (PV) 0.02 kgCO 2 e/ KWh (CSP) (Burkhardt et al. 2011; Hsu et al. 2012) 161,000 Natural Gas 1,220 MMJ 2012 (Dallinger 2015) 0.05 kgCO 2 e/MJ 0.014 kg CO 2 e/MJ (Heath et al. 2014) 80,000 Propane 4.9 million gal 2012 (U.S. EPA 2010) 5.63 kgCO 2 e / gallon 2.15 kgCO 2 e/gallon (U.S. EPA 2010; Frischknecht and Rebitzer 2005; Ecoinvent v2.0 2007) 39,000 Agriculture and Land Use Sector Soil GHG emissions from crops 13,009 MT synthetic N fertilizer 2010 (Oniel 2015; Dillion 2015; CDWR 2015b) IPCC tier 1 methodology 4.00 kgCO 2 e / kg N (IPCC 2006; Zeist et al. 2012) 210,000 Soil GHG emissions from livestock 3,662 MT N applied by livestock 2012 (U.S. Departme nt of Agriculture 2015 ) IPCC tier 1 methodology ----------------(IPCC 2006) 132,000 Enteric Fermentation 83,144 cattle 12,711 sheep 221 hogs 2012 (U.S. Departme nt of Agriculture 2015) 50 kgCH 4 /cattle 5 kgCH 4 /sheep 1 kgCH 4 /hog ---------------(IPCC 2006) 140,000 Manure Management 83,144 cattle 12,711 sheep 221 hogs 2012 NASS (U.S . Departme nt of Agriculture 2015) 2 kgCH 4 /cattle 5 kgCH 4 /sheep 1 kgCH 4 /hog ---------------(IPCC 2006) 5,600 Agricultural Machinery 6.7 million gal 2012 (U.S. Departme nt of Agriculture 2015) 9.18 kg CO 2 e / gallon diesel 2.3 kgCO 2 e / gallon diesel (Argonne National Laboratory 2015) 78,000 Transportation Sector Personal Vehicles 17.9 million gal gasoline 3.1 million 2012 (CDOT 2012) 8.71 kg CO 2 e / gal gasoline 9.18 kg CO 2 e / 2.3 kgCO 2 e / gal gasoline or diesel (Argonne Na tional Laboratory 234,000

PAGE 54

32 gal diesel gal diesel 2015) Large Trucks 6.9 million gal diesel 2012 79,000 Waste Sector Landfills 63,414 tons 2012 (CDPHE 2015) 0.89 MTCO 2 e / ton MSW ---------------(U.S. EPA 2015b, 2013b) 56,000 Waste Water 745 MT BOD rural septic 737 MT BOD centralized 2012 (U.S. Census Bureau 2012) 0.24 MT CH4 / MT BOD ---------------(IPCC 2006) 12,000 Results and Discussion Total GHG emissions in SLV A consolidated summary of all material and energy flows, emission factors, and data sources used in this study can b e seen in Table 1 . For the year 2012, it was estimate d that 1.27 million metric tons of carbon dioxide equivalents (MMT CO 2 e) emissions could be attributed to SLV, which includes a 0.16 million metric ton carbon credit for the utility scale solar in the region. The region hosts four utility scale solar facilities with a combined total of 87 MW of capacity. SLV has a per capita emission of rou ghly 30 MT CO 2 e whereas the average per capita emissions for the United States is about 21 MT CO 2 e (U.S. EPA 2013a) . The higher per capita emissions in SLV reflect that a significant amount of economic activity in the region is associated with production of agricultural goods and services that are exported rather than consumed. Electricity is the highest contrib utor to total emissions at 0.36 MMT CO 2 e, followe d by passenger vehicles at 0.23 MMT CO 2 e, and soi l GHG emissions (crops) at 0.21 MMT CO 2 e ( Figure 4 ). Soil GHG emissions, typically not acco unted in city scale GHG inventories,

PAGE 55

33 from agricultural activities and livestock raising contributes ~20% of total GHG emissions. Similarly, methane emissions from livestock contribute ~10% to the total GHG emissions of the SLV. Figure 4 : The GHG accounting for each of the major emission categories in San Luis Valley . This includes both direct emissions and upstream emissions. Upstream Emissions Direct emissions within the Valley account for about 86% of total emi ssions, and the remaining 14% are upstream emissions that occur outside of the geographic boundary ( Figure 4 ). However, note that with regards to electricity , the GH G emissions from power plants are categorized under direct emissions but the power plants that service the region are located outside of the

PAGE 56

34 geographical boundaries. The inclusion of life cycle analysis of upstream emissions coupled with direct use emissi ons can provide a more comprehensive emission profile for agricultural regions. Upstream emissions from fertilizer, which accounts for 3.6% of the SLV's total emissions inventory, are the largest of any of the upstream emissions. Excluding upstream emiss ions may cause policymakers to focus on other sectors, when in fact the way soils are managed and synthetic fertilizer usage could be a more impactful place to focus. If an agricultural inventory does not include upstream emissions from fertilizer they ma y miss a crucial piece of the picture. Carbon Credit The carbon credit was based on a robust accounting of electricity sources including analyzing the utility scale solar energy and attributing carbon credits based on avoided emissions. This exercise can provide clarity on the current and potential ability of the region to offset its emissions with renewab le energy sources. The utility scale solar production in the region produced 198,000 MWh in 2012 , which is 44% of the region's electricity consumption and is equal to 0.16 million metric ton s of avoided CO 2 e emissions ( Figure 4 ). The current credit in this study reduced the overall GHG emissions from the region by ~9%. With these types of data, policymakers can begin assessing future scenarios based on potential emissions reduction targets. The Bureau of Land Management (BLM) in SLV is currently proposing four Solar Energy Zones (SEZ) in the region . This is part of a larger federal initiative called the Solar Energy Program, which

PAGE 57

35 proposes that BLM land should relax the barriers to private solar development on public land (DOE/BLM 2012) . A 50% build out of the proposed BLM solar energ y plan (700 MW) would see a 1.5 MMMT CO 2 e credit, which was more t han the total emissions from the region in 2012. Emissions by Category Activities responsible for GHG emissions were classified into the four broad categories ( Figure 5 ). The associated calculation involved dividing electricity between agriculture, residential and commercial. In addition, agriculture and livestock soil GHG emissions, methane emissions from livestock and manure management, and agricultural machinery emissions were allocated to the agriculture category and passenger vehicles and large truck transport emissions we re assigned to transportation category. The residential category included energy for buildings, wastewater treatment, and landfill emissions. The results show that agriculture is the largest contributor of GHG emissions in the region (~47%) followed by t ransportation (~20%), residential ( ~20% ), and commercial ( ~12% ) . Of the agricultural category, ~43% is associated with soil , while GHG emissions and electricity for irrigation contributes ~15% of agricultural emissions. Passenger vehicle transport (class 1 3) is responsible for the majority (~ 74.5%) of transportation emissions and large trucks account for the remaining (~25.5%) emissions. Note that the truck transportation includes truck transport for agricultural products. Building energy (electricity , natural gas , and propane) consumption dominates residential sector emissions (~75%).

PAGE 58

36 Figure 5 : The breakdown of GHG emission contributions of the four major categories averaged between 2006 and 2012 . Comparison These results a re now compare d with GHG inventories related to different geographical scales (city, state, and country). Bec ause the four scales examined, regional hinterland (SLV), city (Denver), state (Colorado), and national, all have very different population densit ies and economies, it allowed for an interesting comparison in terms of emissions per GDP in addition to emission per capita. The SLV and Denver GHG emissions data include upstream emissions , whereas the inventories of Colorado and USA do not ( Table 2 ). Furt her, because there are certain methodological differences in each of these studies, fossil energy use (use phase) is also presented for comparison purposes. The results show that SLV has the lowest per capita emissions

PAGE 59

37 amongst the four spatial scales , when consider ing only fossil energy use (16 MT CO 2 e/capita ) ( Table 2 ) . This is because SLV has le ss commercial and industrial activity in comparison to urban areas. I n Denver , f or example, energy used for commercial buildings is about 9.7 MT CO 2 e/capita while the corresponding number in SLV is 3.2 MT CO 2 e/capita. This difference (6.5 MT CO 2 e/capita) is partially offset by agricultural energy consumption (pumping energy and fuel for machinery) in SLV, which was about 3.4 MT CO 2 e/capita. On the other hand , SLV has the highest per capita emissions when we consid er the total GHG emissions (26 MT CO 2 e/ca pita). This is attributed to the significant level of non energy related agricultural emissions (such soil N 2 O emissions) associated with SLV, which is not present in Denver. However , per capita comparisons across diverse locales (such as rural region vs . urban region) provide very limited insights, because these different locales provide widely varying types of functions (food production vs . services). Therefore, looking at emissions per dollar GDP as opposed to emissions per capita would be more meaningf ul. SLV has the highest per GDP emission s by far of all the areas at 670 780 kg CO 2 e/$ GDP. When considering only fossil energy use, SLV has a lower per capita GHG emissions compared to Denver and Colorado by 17% and 18% respectively. However, if we con sider GHG emissions/$ GDP as the comparison we see that SLV's emissions/$ GDP exceed Denver's and Colorado's by 30% and 6%, respectively. Therefore, when comparing across different spatial scales , we conclude SLV (an agricultural

PAGE 60

38 driven economy) is much l ess efficient, on a GHG emission basis, at producing capital than Denver (a consumer based economy). This is not surprising because the type of economic activities in SLV and Denver are vastly different. For example , > 50% of the emissions from SLV are du e to exported agricultural products, and cities (like Denver) are predominantly consumption centric and have service oriented economies. Note that the above analysis is based on energy use alone and does not include emissions from agricultural land use (e .g., soil GHG emissions, livestock methane emissions etc.) . This type of analysis can assist state level decision maker s to prioritize resources available for GHG mitigation purposes.

PAGE 61

39 Table 2 : Compar ing the SLV inventory with GH G inventories a t various spatial scales . San Luis Valley Denver (Ramaswami et al. 2008) Colorado (Arnold et al. 2014) USA (U.S. EPA 2013a) Total Inventory (MMT CO 2 e) 1.2 Ð 1.4 * 11 15 13 0 6,500 Fossil Energy Use (MMT CO 2 e) 0.7 11 97 5,100 Total Inventory (MT CO 2 e/capita) 26 Ð 30 * 19 25 26 21 Fossil Energy Use (MT CO 2 e/capita) 16 19 19 16 Total inventory (kg CO 2 e/$ GDP) 670 Ð 780 * 210 280 47 0 40 0 Fossil Energy Use (kg CO 2 e/$ GDP) 370 20 0 350 31 0 D ata for GDP of the San Luis Valley and Denver were obtained from IMPLAN , working with the Business Resource System of the University of Colorado Boulder (IMP LAN 2012) , while data for GDP of Colorado and the U.S. came from the Bureau of Economic Analysis (BEA 2012) . * The high and low estimate (i.e., range) of the total inventory for SLV and Denver are showing results with and without upstream emissions. Overall it was found that emissions from electricity production dominate the GHG emissions profile of all the locales, at roughly 25% of their GHG em issions (Arnold et al. 2014; U.S. E PA 2013a) . This information may lead policymakers to wrongly conclude that all these vastly different regions would benefit from the same types of policies to reduce emissions from electricity. For example a state mandate to insulate homes or increase efficiency as a means of reducing electricity usage would make sense in urban areas where buildings account for ~50% of the emissions (Hillman and Ramaswami 2010; Ramaswami et al. 2008) , but an entirely different set of policies might be better in rural areas,

PAGE 62

40 such as SLV. Improvements to agricultural practices ( i.e. , agro ecological practices; low input farming (Altieri 1999) ) and rotation cropping patterns ( such as cover cropping) could have a much larger impact than trying to reduce energy consumption in the residential or commercial sectors. Therefore, region specific inventories with highly disaggregated data as presented in t his work are key for local policymakers. Uncertainty This study utilizes large amount of data that have underlying uncer tainty associated with them. T here exists uncertainty in these raw data (material and energy flows) such as VMTs, fertilizer applicatio n rates, and liquid fuels consumed, as well as in the emission factors associated with both direct emissions like wastewater treatment, landfills, natural gas burning, and upstream emissions like fuel refining. Please note this specific example of uncerta inty that emerged during data disaggregation regarding electri city consumption for irrigation. Though all electricity consumption in the region was accurately accounted for, the irrigation component provided by Xcel Energy was most likely an underestimate due to the restrictions on the type of data they were allowed to provide under privacy laws. Thus, the irrigation portion of GHG emissions is likely a conservative estimate. Though a robust accounting was presented , it is important to understand the res ults of the study within the overall context of possible data uncertainty. Unfortunately, information on this uncertainty is rarely available for such data.

PAGE 63

41 Finally, i t would be pertinent to note that GHG accounting can also be based on regional consumpti on (usually referred to as carbon footprint) (Peters 2008; Larsen and Hertwich 2009; Davis et al. 2010; Wiedmann 2009) . A s discussed earl ier, the type of accounting (consumption or production based) used should depend on the utility of the inventory to provide data and insight towards mitigation strategies, which in turn is related to the sphere of influence of the relevant local government s. Therefore, for regions that are predominantly consumption oriented (e.g., cities) carbon foot printing might be appropriate because they can help identify consumption related mitigation strategies , whereas for production centric regions (e.g., SLV) , a hybrid approach such as the one proposed here will offer the most value.

PAGE 64

42 CHAPTER III CONSUMPTIVE WATER US E Consumptive Water Use Analysis of Upper Rio Grande Basin in Southern Colorado Abstract Water resource management and governance at the river basin scale is critical for the sustainable development of rural agrarian systems in the West. This research applies a consumptive water use analysis, inspired by the Water Footprint methodology, to th e Upper Rio Grande Basin ( RGB ) , or San Luis Valley (SLV), in south central Colorado. The region is characterized by water stress, high desert conditions, declining land health, and a depleting water table. R egion specific data and models were utilized to analyze the consumptive water use of SLV . The study reveals that, on average, SLV experiences three months of water shortage per year due to the unsustainable extraction of groundwater (GW). R esults show that agriculture accounts for 77% of overall wate r consumption and it relies heavily on an aquifer (about 50% of agricultural consumption) that is being depleted over time. It was found that, even though potato cultivation provides the most efficient conversion of groundwater resources into economic val ue (m 3 GW/$) in this region, it relies predominantly (81%) on the aquifer for its water supply. However , cattle, another important agricultural commodity produced in the region, provi de good economic value, but rely significantly less on the aquifer (30%) for water needs. The results from this chapter are timely to the region's community, which is currently in the process of

PAGE 65

43 developing strategies for sustainable water management. Introduction Water and energy use ar e two important resource use issues of the century as there are limits to the extent that humanity can continue to increase its appropriation of fossil fuel resources and freshwater from the natural environment (Ridoutt and Pf ister 2010) . Further, water and energy interdependencies and supply constraints have been recognized by researchers to pose significant risks that can potentially and unintentionally shift overall impacts geographically and temporally (Fulton and Cooley 2015; Hussey and Pittock 2012) . In this context, proper quantification of fresh water resources needed for the environment as well as human made products (e.g., agricultural produce) and processes (e.g., power plant cooling) across nations and sub regions is the first step towards addressing water use issues (Hussey and Pittock 2012; Hoekstra et al. 2011) . Water Footprint (WF) (Chapagain and Hoekstra 2004; Hoekstra et al. 2011; Mekonnen and Hoek stra 2011a) is an approach that provides a useful framework to quantify fresh water usage along the entire supply chain (Hoekstra and Hung 2002) . WF has been used to assess both direct consump tive use and indirect embodied water content of individual products such as milk and clothes (Chapagain and Hoekstra 2004) , as well as for populations of nations and planet (Gerbens Leenes et al. 2013; Mekonne n and Hoekstra 2011b) . However, this chapter , which pulls certain concepts from WF, focuses only on consumptive water use of a regional system with the aim to provide data and analysis for both

PAGE 66

44 consumers of water from the r iver basin a s well as water m anagers (Zeng et al. 2012; Dumont et al. 2013; Dong et al. 2013; Appasamy et al. 1999) . Note that this analysis is based on water consumptio n as opposed to the more conventional water withdrawal m easure , in that a consumptive use analysis recognizes the interconnectedness of the hydrologic cycle and appropriately allocates evaporation, return flows, irrigation inefficiencies, and other withdra wn but "unused" water (Hoekstra et al. 2011) . SLV is a high altitude agricultural valley in s outh c entral Colorado and encompasses roughly 21,000 km 2 (8,000 square miles). It has an average elevation at the valley fl oor of 2,300 m (7,600 ft.) above sea level with an average annual precipitation of 222 mm (9 in) (PRISM Climate Group 2016) . It is a snowmelt driven hydrologic system, with the majority of water into the SLV coming from river and stream flows from the surrounding mountain ranges. Hay production and pasturelands utilize some of the snowmelt, but much of it flows into the massive aquifer that underlies the entire valley, a portion of which is pumped back for irrigation. It is also important to note that not all of t he water that flows into SLV is "owned" by the region. Two major interstate compacts, the Rio Grande Compact (1938) (Hinderlinder et al. 1938) and the Amend ed Costilla Creek Compact (1963) (Whitten and Reynolds 1963) , require SLV to allocate water to downstream states based on the annual sno wmelt. This precious water supply is what allows this high desert region to be one of the nation's key producers of potato es , alfalfa, and barley . The region's limited population,

PAGE 67

45 isolated location, and the fact that its natural watershed boundaries coin cide somewhat wi th its political boundaries are factors that make this region an ideal study area (U.S. EPA 2010) . The impacts of groundwater pumping on surface water flows in the region have been recognized , and the extreme drought of 2002 accelerated the situation to the forefront. The Colorado Division of Water Resources ( CDWR ), the department responsible for administering water rights, monitoring water use and stream flows, and issuing permits for wells and other water infrastructure projects in the state, developed a robust groundwater (CDWR 2015b) model which has shown that continuous and increasing depletion of the aquifer ( as monitored since the late 1970s by Davis Engineering (see Figure 6 ) ) has affected surface water flows over the last ten years and impacted agriculture which relies on those flows. This region specific model provides a complete water balance and describes consumptive use in the region that includes return flows, cr op shortages, a s well as consumption by native vegetation (CDWR 2016a) . Because agriculture is the major economi c activity in the region, the community has recognized the importance of working towards sustainable management of the region's water resources on which its economy relies heavily (SLV Development Resources Group 2013) . The community recognized the importance of water and demanded legislation from the state that requires sustainable water management, namely Senate Bill 04 222 (Entz et al. 2004) signed in 2004, which is currently being implemented. As part of this bill, CDWR

PAGE 68

46 has promulgated well rules and regulations and submitt ed them to the Colorado Water Court in the Fall of 2015 (CDWR 2015a) . The purpose of this legislation is to ensure that the prior appropriation system is upheld, given the now apparent impacts of groundwater pumping on surface water flows. The priority system, or prior appropriation system, is based on the "first in time , first in right" concept of W estern U.S. water law, which appropriates surface water rights to the first person to divert the water for beneficial use (CDWR 2016b) . SLV 's surface water was fully appropriated by 1900 , but since groundwater at that time was thought to be delinked from the surface water system, the state continued to issue well permits for groundwater pumping until the 1970s. The new legislation requires repayment of injurious depletions to senior surface water rights holders due to groundwater pumping (CDWR 2015a) . Another major component of the well rules and regulations, and in some regard the most interesting due to the major drop in aquifer levels that took place during the 2002 drought ( Figure 6 ) , is the requirement that the confined aquifer be maintained at the average historic levels similar to the period of 1978 to 2000. This legisl ation may set a precedent for future sustainable aquifer management and related legislation across the country. The State Engineer's Office and the CDWR's Division 3 Engineer, who administer SLV , are responsible for managing and enforcing the new groundwa ter rules. Also allowed under the legislation is the formation of groundwater management sub districts of the Rio Grande Water Conservation District (RGWCD). These entities are tasked with managing their

PAGE 69

47 water use, replacing injurious depletions to the s enior surface water rights holders through replacement water, financial water saving incentive programs, and restoring and maintaining a sustainable aquifer within the parameters of the legislation (CDWR 2015a) . Figure 6 : Change in storage of the unconfined aquifer in SLV from 1975 2016 (Davis Engineering 2016) . Simultaneously, Governor John Hickenlooper issued an Executive Order in 2013 requiring the Colorado Water Conservation Board (CWCB) to develop a statewide water plan. Each of the state's nine river basins host a conservation

PAGE 70

48 roundtable composed of water users and stakeholders in each region. The roundtables were tasked to develop Basin Impleme ntation Plans (BIP). The BIPs lay out projects, goals, and priorities for the basin ' s future. The Rio Grande BIP (which encompassed the region of SLV) was published in the summer of 2015 and focuses on achieving a balance of competing water needs through cooperative management of water resources (Gibson and et.al. 2015) . The BIP outlines detailed plans and allocated funds to help manage water resources. The BIP present s ways to improve soil health and increase soil water holding capacity, improv e stream flow forecasting, irrigation improvements, head ga te and ditch restoration projects, and highlights the Rio Grande headwaters restoration project. It also emphasizes the importance of maintaining a healthy stream corridor to achieve efficient compact deliveries at the state line. BIPs from across the st ate informed the Colorado Water Plan, which was delivered to the Governor in November of 2015 , and will influence water use in the state . (www.Coloradowaterplan.org). This unique regional context required the development of SLV specific methods, models, an d data to understand and address the problem. In an attempt to fill information and data gaps, a robust and descriptive baseline consumptive water use analysis ( CWU ) is presented . This model will be available as a resource to assist local decision maker s who are continuing to develop plans and implement actions towards sus tainable water use in SLV.

PAGE 71

49 Methods This analysis tracks the origin (source) of water , whether effective rain (green water) or rivers and streams (blue water) (Falkenmark 1991; Food and Agriculture Organization of the United Nations (FAO) 1993) . In the present study, blue water is also divided into surface and groundwater components due to the unique hydrologic features of SLV . W ater consumption in SLV is characterized into the following t hree categories: 1) crops, 2) livestock, and 3) municipal/i nd ustrial use; and the following three water sources were considered : 1) effective precipitation (green water), 2) blue surface water runoff (blue water surface/snowmelt), and 3) blue groundwater from the aquifer (blue water ground) ( Table 3 ). The grey water component, which represents water needs for pollutant discharge and dilution, was not considered in this study due to lack of region specific data and community interest. Table 3 : A summary of water sources considered in this study . Water Source Description Green Water Green water is the quantity of effective rainfall (ER), meaning rain that fell on crops during the growing season, consumed by the basin annually. Blue Water Surface Blue water from surface water measures the quantity of snowmelt runoff from rivers and streams that is diverted and consumptively used in the region. Re charge water is considered groundwater in this an alysis. Blue Water Ground Blue water from groundwater is a measure of the quantity of aquifer water that is pumped and consumptively used or directly sub irrigated through roots in crop fields. Because this research focuses on the water consump tion patterns of the region, only water originating within the basin was considered and any upstream

PAGE 72

50 embodied water of imported goods and services was excluded. In this way, the analysis differs from the traditional Water Footprint methodology. The CWU calculations adapted for SLV are shown below: !"# !"# ! !"# !"##$ ! !"# !"#$ !"# !"##$ ! !"# !"##$ ! !"#$% ! !"# !"##$ ! !"#$%&'() !"# !"#$ ! !"# !"#$%&' ! !"# !"#$%& !"# !"#$%&' ! !"# !"#$%&' ! !"#$% ! !"# !"#$%&' ! !"#$%&'() ! ! !"# !"#$%&' ! ! ! ! !"# !"#$%& ! !"# !"#$%& ! !"#$% ! !"# !"#$%& ! !"#$%&'() ! !"# !"#$%& ! ! ! ! A unique aspect of the presented regional CWU assessment is that first total CWU in SLV was estimate d using a local groundwater model, and then the Consumptive Water Content (CWC) of each crop was derived from this overall CWU , making the results highly specific to SLV . The CW C represents the qua ntity of water needed to produce a unit of a good or service (i.e., quantity of crop) (Hoekstra et al. 2011; Liu et al. 2009) . This level of regional specificity was possible due to the close collaboration of our research team with CDWR, and the use of regional data from CDWR's HydroBase that contains real time, historic, and geographic data related to water resources in Colorado. T he Rio Grande Decis ion Support System (RGDSS) groundwater model was used (CDWR 2016c, 2016a) , which is specifically calibrated for this region, and allow s for a robust,

PAGE 73

51 region specific accounting that does not rely on global averages (which is typical of consumptive water use studies). Crop Water Use of SLV Calculation s of crop water consumption were performed within the Colorado crop consumptive use model (StateCU) component of RGDSS and considered all of the area in Water Division 3 (which closely matches the b oundary of SLV ) (CDWR 2015c) . StateCU, as applied in RGDSS, models the water budget of a ditch service area by calculating the quantity of water consumed by different crops using a modified Blaney Criddle method with loca lly calibrated crop coefficients (CDWR 2015c) . In addition, local data related to temperature, precipitation, rooting depths, growing season start and end dates, available water content of soils, irrigated acres, cropping patterns, surface and groundwater diversions, ditch conveyance and irrigation application efficiencies were supplied as inputs into the StateCU model. Data were developed and maintained within Hy droBase (CDWR 2016a) as part of RGDSS efforts. The four major crops cultivated in the region: alfalfa, potatoes, small grains (mainly barley) and meadows/pasture grass were assessed on a monthly time step for years 2000 to 2010. They were assessed for potential evapotranspiration, effective precipitation (EF), and the remaining irri gation water requirement (IWR).

PAGE 74

52 Figure 7 : Geospatial representation of crop production in SLV . Much like the global crop consumptive water use model (CROPWAT), which many studies use to estimate crop water consumption, the StateCU model is used to estimate the e ffective precipitation and the amount of irrigation water consumed by the crops (Chapagain and Hoekstra 2004; Zeng et al. 2012; Dumont et al. 2013) . Using a local model (StateCU) coupled with a local groundwater model (RGDSS) is unique and critical becau se there may be times when precipitation and irrigation water supplies are insufficient to meet the entire crop demands, and for these instances there is a shortage that may be met if the water table is within the crop's root zone. The RGDSS groundwater m odel simulates the water table and the associated crop demand shortage, and if the water table is within the crop rooting zone, the groundwater model automatically

PAGE 75

53 meets the crop water demand through sub irrigation (direct groundwater consumption from crop roots) . Further, there may be times when water supplies fall short of meeting the crop demands, which may affect crop yields. Crop yield data were obtained at the county level through the National Agricultural Statistics Service (NASS) (U.S. Department of Agriculture 2015) ( Table 4 ). Table 4 : Cultivated area and crop production in SLV (2000 2010). Crop Cultivated area Cultivated area Yield Production Market Value (thousand acre s ) (thousand ha) (tons / acre ) (thousand tons) $ / ton Small Grains 94.1 37.6 2.9 267.4 191.3 Potatoes 66.5 26.6 18.7 1 , 237.8 189.3 Alfalfa 141.3 56.9 3.6 502.3 133.3 Meadows and pasture hay 211.2 84.5 1.7 165.2 + * 119.2 The cultivated area values come from the RGDSS and utilize aerial imagery, satellite imagery, field verification, NASS, and Rio Grande Water Conservation District (RGWCD) data to evaluate irrigated lands and crop types. By multiplying the cultivated area by c rop yield data from NASS , the total production values were developed. The market value is the annual average value on price received in the state of Colorado from 2005 2010 (U.S. Department of Agriculture 2015) . * The cultivated area for meadows and pasture hay is an aggregated value from RGDSS. Some of this land is used for wildlife habitat/wetlands in addition to cattle grazing and haying. The native grasses that are hayed could be sold and exported from SLV or used for livestock within SLV in the current year or subsequent years. The production value given in the table for pasture hay includes only the baled hay exported from SLV as reported by NASS (which is an underestimate of the total production in the region). Due to the difference in the evaluation methods of this research and StateCU, the results were post processed to quantify the individual crop consumptive use amounts. Specifically, StateCU evaluates crop use by ditch service area , whereas this research presents region wide average crop consumptive use. The analysis utilizes information from the Alamosa climate

PAGE 76

54 station as an approximation for the remainder of Water Division 3 to evaluate crop demands, and then normalizes the crop demands back to the Water Division 3 StateCU model results. By combining th e post processing results from the StateCU model and crop yield data (U.S. Department of Agriculture 2015) , CWC green , CWC surface , and in combination with the RGDSS groundwater model, CWC ground , was determined for each of the major crops in the region . Livestock Water Use of SLV CWU of live stock (C W U livetock ) was calculated in SLV by multiplying the CWC (m 3 /ton live weight) of each livestock (CWC livestock ) type by the production quantity in the SLV ( Table 5 ). Table 5 : Livestock herd in SLV (2000 Ð 2010) for each animal type . Livestock Amount Production Market Value (head) (ton live weight) $ / ton (adult) $ / ton (juvenile) Cattle 83,144 23,464 2 , 040 2 , 420 Sheep/Goats 12,711 639 708 2 , 087 Hogs 221 24 996 996 The head count and production values were collected from NASS , a nd represent the counties of SLV (U.S. Department of Agriculture 2015) . The market values are the average value on pr ice received in the state of Colorado from 2005 2010 (U.S. Department of Agriculture 2015) . The three major water uses needed for calculating the CWC livestock are drinking water requirement (DWR), service water requirement (SWR), and feed water requirement (FWR) (Zeng et al. 2012; Mekonnen and Hoekstra 2012) . SWR is a measure of the water used for keeping and maintaining the animals (e.g. washing, feed mixing etc.) . The DWR and SWR were estimated using global annual averages from Mekonnen 's and Hoekstra's (Mekonnen and

PAGE 77

55 Hoekstra 2012) analysis of livestock water footprints for grazi ng animals. In addition, CWC livestock also include d the water embodied in the feed that the livestock eat (FWR), as this water also originates in SLV . The FWR, which accounts for 99% of the grazing livestock CWU , was calculated based on local CWC of the feed crops grown in the region ( Table 6 ) coupled with the amount and feed conversion efficiency (FCE) of each type of f eed consumed by the livestock. Table 6 : Consumptive Water Content (CWC) of feed in the SLV (2000 2010) (1 Ac ft. = 1233.5 m 3 ) . CWC Ð green m 3 /ton CWC Ð surface m 3 /ton CWC Ð ground m 3 /ton CWC Ð Total m 3 /ton Irrigated Pasture Grass 214.4 542.7 450.4 1,351.5 Alfalfa 135.8 202.6 531.9 870.3 Dry land Pasture Grass 1,266.7 ------------1,266.7 The average moisture content of the various feeds were collected from the local extension office (Reynolds 2015) . It was assumed that 60% of the baled feed was from irrigated grasslands and 40% was from locally grown alfalfa. The CWC of the feed in the region was calculated using yield data for hay alfalfa and other hay grass from NASS. To account for the grazing that takes place during spring and fall on the irrigated pastures, an additional yield of 0.5 tons per acre was added to the yield of hay grass (excluding alfalfa). The CWC of upland dry land pasture was averaged as 1.5 tons/acre in a wet year (800 m 3 or 8 in ches precipitation) and 0.15 tons/acre in a dry year (300 m 3 or 3in precipitation) (Whitten 2015; Sparks 2015) . FCE is a measure of the quantity of feed needed to produce a certain quantity of output. !"# ! ! !"# ! ! !"# !

PAGE 78

56 ! ! !""# ! !"#$ !"# ! !"# !"#$ ! !"#$!% ! !"#$ ! ! !"# ! !"# ! !"##$% ! !"#$%& A daily dry matter intake (DMI) of 3% of the body weight for adult cattle and 1.5% for calves and grazing sheep , in accordance with Mekonnen and Hoekstra (Mekonnen and Hoekstra 2012) and confirmed by the local extension office and a local rancher , was used (Reynolds 2015; Whitten 2015) . Total forage consumption values were based on the assumption that all feed requirements needed by livestock were satisfied through local resources, as is the practice in the regi on. Public grazing lands in the region have the capacity to support a herd of cattle and sheep spending 3 months of the year on dry land pasture (NFS 2015; BLM 1991; Page 2016) , 5 months grazing on private irrigated pasture and meadows, and 4 months eating baled hay grass and/or hay alfalfa (Reyn olds 2015; Whitten 2015) . Total livestock headcount was obtained from NASS . B ecause the FWR of hogs originate s from outside the region and their population is small ( Table 5 ), they were not included in this study. Typically, ranchers in SLV do not finish cattle or sheep locally but tend to export live animals; therefore, the unit of ton live animal was used for livestock production instead of the more conventional unit of kg meat.

PAGE 79

57 Municipal and Industrial Water Use of SLV CWU of the municipal and industrial sector (C W U M&I ) was calculated based on input data used in the RGDSS groundwater model on a monthly time st ep for 2000 2010. The value for C W U M &I was based on the total pumping and return flows for all non irrigation high capacity wells within the RGDSS model domain, which covers the entire region. In addition to groundwater pumping, this sector also includ es CWU from reservoir evaporation of surface water. It is important to note that the manufacturing and industrial sector in SLV represents only ~1% of the overall economy (SLV Development Resources Group 2013) . Results and Discussion Total Consumptive Water Use of SLV An examination of the results show s that SLV 's consumptive water use is dominated by agricultural activities. Crops account for 63% of CWU of the region , followed by livestock at 13% , and municipal and industrial use accounting for just 3%. In addition, non agricultural native vegetation accounts for about 21% of the overall CWU in SLV ( Figure 8 ). The total CWU (in million m 3 ) was separated into green water, surface water, and groundwater ( Figure 8 ). By breaking up the C W U blue into two components: C W U ground and C W U surface , analys i s could be done to see how the various water use types draw upon specific water sources (ground vs. surface water).

PAGE 80

58 Figure 8 : Average annual water allocation in SLV from 2000 2010 (1 Ac ft. = 1233.5 m 3 ) . Alfalfa is the largest overall user of water followed by meadows / pasture and cattle ( Figure 8 ). It is important to note that meadows / pasture are useful for much more than just grazing cattle, as demonstrated by the fact that even after alloca ting the water needed for raising forage for cattle and sheep, the remaining meadows are still one of the largest overall contributor s to CWU in SLV . This relatively large water use is due in part to the fact that the majority of irrigation water used for meadows / pasture is surface water that is diverted for agricultural use during early spring and summer, in accordance with prior appropriation system. In many locations, surface water is diverted from rivers or streams onto

PAGE 81

59 the meadows to grow grass where ranchers graze cattle, store hay for the winter, and export haylage (baled grasses). In addition to providing grazing land, meadows sustain a vast and complex network of wetlands, native habitat, and species diversity. Crop Consumptive Water Use of SLV O verall, 57% of crop water requirement is met though groundwater , 27% is met though surface water, and 16% is from effective precipitation during the growing season ( Figure 9 ). Alfalfa is the largest consumer of groundwater in SLV, whereas meadows/pasture is the largest consumer of surface and green water. The blue water portion (BWP) combines surface water and groundwater needed for all crops in the region and account s for about 85% of the C W U crops . This BWP for crops is much higher than the global average of 19% (Liu et al. 2009) , indicating that the region is much mor e reliant on irrigation in comparison to other regions. In addition, the result that 57% of the C W U crops comes from groundwater resources has a profound and direct bearing on the long term sustainability of the aquifer (a key requirement of the new well rules and regulations).

PAGE 82

60 Figure 9 : The crop water use allocated between C W U green crops , C W U surface crops , and C W U ground crops as well as the consumptive water co ntent of each crop (2000 2010) * . * This figure should be read inside out with water demand by source type shown in the inner ring and the different crops demanding that type of water represented in the outer ring .

PAGE 83

61 Livestock Consumptive Water Use of SLV Th e livestock consumptive water use aspect considers water needed for feed crops, drinking water, and service water. In SLV , 98% of the water needs for livestock can be attributed to FWR. Overall, 40% of the livestock water needs are met by green water and 30% by ground and surface water ( Figure 10 ). Livestock are able to utilize the sparse green water that falls in the region through dry land pasture grazing in th e summer months. This, along with meadows and pasture grazing, is why we see such a large green water percentage for CWU livestock . The majority of the surface and groundwater portion of the C W U livestock is attributed to raising irrigated pasture for graz ing and winter hay feed.

PAGE 84

62 Figure 10 : Livestock consumptive water use (C W U livestock ) for SLV allocated between C W U green crops , C W U surface crops , and C W U ground crops as well as the consumptive use content of each crop. Cattle represent 98% of all CWU livestock in the region * . * This figure should be read inside out with water demand by source type shown in the inner ring and the different feed crops demanding that type of water represented in the outer ring.

PAGE 85

63 Agricultural Products In this research, g roundwater intensity is defined as a measure of the quantity of groundwater needed per ton , or per dollar revenue , of agricultural products ( Table 7 ). The dollar value for agricultural products sold in Colorado was obtained from USDA (U.S. Department of Agriculture 2015) using the previous 5 year average and was used to calculate groundwater intensity in dollars/m 3 ( Table 4 and Table 5 ) . W hen revenue is used as the index, potatoes have the highest efficiency in converting groundwater int o dollars , followed by livestock (i.e. , sheep and cattle ) . Efficient conversion of groundwater resources to dollars captures two major issues related to the region , economy and a depleting aquifer. Potatoes are efficient at producing high yields (17 tons / acres) and hence have a low groundwater use per dollar revenue. However, this is offset significantly by the fact that they require the highest groundwater portion (81%) per ton of crop ( Table 7 ). This is in contrast to sheep and cattle, which are second in efficient conversion of groundwater resources into dollars , however have the lowest groundwater portion (~30%) per ton of any of the agricultural products in the region. Further, the current consumption of blue water for crops and livestock (almost all of which is exported) results in a net displacement of 1,120 million m 3 /year (908 thousand ac ft. ) in the form of virtual water flows (calculated as area under the Ôblue water CWU crops and feed' curve ; Figure 11 ) of surface and groundwater .

PAGE 86

64 Table 7 : Comparison of major agricultural products in SLV (1 Ac ft. = 1233.5 m 3 ). CWU Content SLV CWU Valley Wide Surface Water Portion (SWP) Groundwater Portion (GRD WP) Green Water Portion (GRD WP) Groundwat er Use / $ Revenue (m 3 t 1 ) (Million m 3 ) % % % (m 3 GW $ 1 ) Alfalfa 871 417 23.1 61.3 15.5 2.61 Potatoes 100 124 4.5 81.7 13.9 0.41 Small Grains 759 201 13.7 74.1 12.2 2.26 Live Cattle 9 , 530 238 30.6 29.8 39.7 1.69 Live Sheep 6 , 535 5.9 30.1 30.4 39.5 1.54 Blue Water Scarcity Blue w ater s carcity (BWS) (Hoekstra et al. 2011; Zeng et al. 2012) was used as a measure to understand water sustainability of the region. BWS occurs anytime the C W U blue exceeds the b lue w ater a vailability (WA blue ) ( Figure 11 ). !" !"#$ ! !"#$%% ! !"# ! !"#$%&#'#$ ! !"#$%&' ! !"#$%"&$"' !"# ! !"#$%&"'(")*+ ! !"#$ ! !"#$%&'"()* Data on m onthly average runoff from the mountains into SLV was used to develop a natural runoff curve for the region ( Figure 11 ) . Data were composed of gauged and ungauged inflows and provided by RGDSS. The modified runoff curve ( Figure 11 ) accounts for farmers using the aquifer as a reservoir, which is a local practice, and wa s based on monthly recharge data from RGDSS . Farmers direct their surface water allocations into "recharge pits" at the corners of their fields so water can infiltrate into the aquifer. To estimate the WA blue curve, both the interstate compact water that is owed to downstream states, and the

PAGE 87

65 e nvironmental f low r equirements (EFR) were subtracted from the total runoff ( Figure 11 ) . Because region specific data were not available for EFR, the value 80% of the total runoff was used to be allocated to stream and ecosystem services as proposed by Hoekstra (Hoekstra et al. 2011; Zeng et al. 2012) . Figure 11 : Monthly water availability and monthly water use (2000 2010) (1 Ac ft. = 1233.5 m 3 ) . SLV experiences b lue w ater s tres s, which is the difference between C W U blue and WA blue , during seven months of the year ( Figure 11 ). This resource overutilization causes severe damage to the ecosystem of the region (Kang et al. 2007; Zeng et al. 2012) . In addition , SLV has been experiencing w ater s hortage because of aquifer mining during two months of the year. W ater s hortage occurs

PAGE 88

66 anytime C W U blue exceeds the total runoff. Th e bulk of runoff occurs in May and June, but the spike in C W U blue occurs between June and August. This local w ater s hortage can be seen whenever the "modified runoff w/recharge " curve drops below the C W U blue curve ( Figure 11 ) . Even with the use of re charge pits, SLV has been experiencing w ater s hortage, mainly in July and August, of about 41 million m 3 , annually. To make matters worse , it is predicted that the natural runoff in spring will begin to shift from June to May due to the effects of climate change (Gibson and et.al. 2015; Dagmar and Vaddey 2013) . Once C W U blue exceeds runoff, it does not take long to see the effects. Water s hortage due to dry conditions and over pumpi ng has caused a major decline in the aquifer (Davis Engineering 2016) . The aquifer is likely to be further aff ected by changes in climate ; some of these effects are apparent today. The aquifer is recharged by surface water runoff, which infiltrates to the groundwater through rivers and streams, rim recharge, flood irrigation, and state issued recharge decrease. A dry ing regime in the surround ing mountains , which is predicted by climatologists, will reduce snow pack and subsequently limit the aquifers' ability to recover from the annual groundwater pumping (Dagmar and Vaddey 2013) . Senate Bill 04 222 and the groundwater rules and regulations, requires water users in the region to maintain aquifer levels similar to the average levels between the years 1978 and 2000. This is already a difficult task and will be even more challenging under a changing climate (Hurd and Coonrod 2007; Malcolm et al. 2012) . It is predicted

PAGE 89

67 that stream flows at the upper gauges will decrease by approximately o ne third between now and 2100 (Dagmar and Vad dey 2013; Gibson and et.al. 2015) . Looking Forward SLV is a microcosm of many dry regions across the globe where water demand exceeds supply, and drought due to a changing climate is adding stress to the already over appropriated systems (El Beltagy and Madkour 2012) . This CWU analysis developed for SLV can provide valuable information and perhaps management models for other areas facing similar challenges. This is a complex system and solely increasing efficiency may not be the answer. Along the riparian corridors, diverted water used for growing meadows and pasture sustains a variety of wetland types that are invaluable for waterfowl, wildlife, recreation , and other ecosystem services (Mitsch and Gosselink 1993) . These seasonal wetlands account for a substantial portion of the wetlands in the region (Gibson and et.al. 2015) . While flooding land during the growing season may not appear to be an efficient use of water, it provides an important service , as these lands provide habitat for native species , as well as grazing land for livestock (Gibson and et.al. 2015; Niemuth et al. 2004; Mitsch and Gosselink 1993) . In addition, it is important to sustain a "wet sponge" along the river corridor, so that Rio Gr ande Compact water deliveries can be effectively managed. If river cor ridor zones were permanently dried by change of use, it would cause a substantial challenge to the State's ability to meet compact requirements . This situation was experienced during th e 2002 drought and

PAGE 90

68 subsequent years (Blankenbuehler 2016) . On the other hand, through this analysis, we see that native vegetation consumes significant amounts of water, which affects the region ' s ability to maintain a sustainable aquifer. It will be important for future regional water use analyses to allocate water towards ecosystem services and environmental flow requirements. Much work has been done in the area of quantifying ecosystem services and these methodologies could be applied to river basins such as SLV (Costanza et al. 1998, 2014, 2011; Costanza 2014) . Maintaining a sustainable aquifer is of utmo st importance in the region. Managing the use/loss while maintaining the agricultural heritage and economy of the region is important. However , the aquifer is continually depleted by extraction of groundwater for agricultural purposes. The results indica te that agricultural products, such as sheep and cattle, offer an opportunity for the region to move towards sustainable water management , because they do not rely as heavily on groundwater , and utilize the scarce green water through dry land grazing. This opportunity is somewhat constrained due to space limitations for dryland pasture grazing and the potential for ongoing drought . Therefore, a lternative grazing practices such as cover crop grazing will need to be employed if the number of livestock we re to increase. A variety of soil health management practices, implemented by an active group of farmers and ranchers, are showing promising results (Rockey 2014; Scully 2014) . These practices include the use of cover crops, compost

PAGE 91

69 application, alternative pest and weed contr ol practices, reduced tillage, grazing animals on green manure, and others . These practices are part of a larger strategy to improve soil health (as measured by increased organic matter in the soils, water holding capacity, soil fertility , and other prope rties ) and reduce the need for pumped water to maintain soil moisture for growing crops. Soil health is a field of growing interest in the community and holds promise for reducing pumping while sustaining viable agricultural production. It is our hope th at this region specific CWU assessment framework can aid this and other similar regions in their pursuit for sustainable water resource management. The proposed framework could be replicated in other basins , provided region specific data and models are av ailable.

PAGE 92

70 CHAPTER IV COMMUNITY ENGAGEMENT Engaging a Rural Agricultural Community in Sustainability Indicators and Future Scenario Identification: Case of the San Luis Valley Abstract This chapter describes the application of community based participatory research (CBPR) to engage representative community stakeholders in a rural agricultural region in identifying future sustainability scenarios for modeling with sustainability indicators . Over the course of two years, researchers and the Community Advisory B oard identified, deliberated, and , based on their input, conceptually framed future scenarios using sustainability indicators for modeling and action . The suggested scenarios (for modeling) tha t emerged through this engagement were centered on solar energy development in the region , and changes to the cropping regime in San Luis Valley (SLV) in southern Colorado . SLV is a unique geographically isolated agricultural region that has been looked at both by the EPA and the state of Colorado as an ideal location for implementing sustainability measures. Importantly, in line with CBPR principles and best practices, community stakeholders remain ed committed and engaged throughout the research process . Through training and knowledge transfer from researchers to the community, SLV now has the capacity to use l ocal data to update region specific greenhouse gas emissions and consumptive water use indicator models . This engagement was successful both in te rms of its usefulness in steering the research direction as well as its impact on community

PAGE 93

71 stakeholders involved in this project. Based on this experience, w e recommend this participatory approach to researchers seeking to improve the relevance and impac t of region specific sustainability analyses . Introduction Background to Participatory Research W orking with a community that is the subject of a research project can lead to the integration of technical and local knowledge, thereby increasing the chances that the research will result in relevant design, adoption of the findings, and increased local capacity (Reed 2008) . This means r esearchers can be a catalyst to the process of social action, but the community acted upon must also be involved . T he practice of community based participatory research (CBPR) , some say, can trace its origins to Popular Education whose founder was Paulo Freire. Éconcrete reality consists not only of concrete facts and (physical) things, but also includes the ways in which the people involved with these facts perceive them É conc rete reality is the connection between subjectivity and objectivity, never objectivity isolated from subjectivity " (Freire 1982) . Freire's ideas awakened a movement, both in the scientific community and the social justice community, to the idea that social change is rooted in education and empowerment. These early studies and interventions (Whyte 1991; Borda and Orland Rahman 1991) led to the development of the field of Participatory

PAGE 94

72 Action Research, which puts the communities and their perspectives at the center of research. In the 1990's, many of these concepts we re integrated into public health research to engage communities affected by complex health issues. Much like issues of poverty and oppression, complex health issues (such as HIV/AIDS) need the community involved and informed in order to make real change (Petrow et al. 1990; Brown 1991; McKinlay 1993) . It is not enou gh to simply present facts. A collection of best practices were developed ( Figure 12 ) to encourage a partnership between researchers and community members to benefit both parties (researchers and communities), and allow for "a more balanced set of political, social, economic, and cultural priorities " (Israel et al. 1998) . For researchers, community participation offers, first and foremost, inv aluable access to "primary resources," in the form of local expertise. The community, in turn, can often gain access to more technical resources from research partners . Through participatory research , researchers and community members can influence, lear n from, and foster trust in one another, allowing a

PAGE 95

73 well conceptualized, comprehensive study that is useful for the community . Figure 12 : A summary of the key aspects of a Community Based Participatory Research project (Israel et al. 1998) . Community Engagement and Sustainability Research More recently , CBPR and other community engagement frameworks have been used in the field of environmental management and sustainability research (Reed 2008) . S ustainability V isioning, the process of creating discourse around positive visions about our societies' future to stimulate change, has been investigated and implemented (Iwaniec and Wiek 2014) . Similar methods have been used to engage cities and municipalities around climate action plannin g using future scenario modeling (Sheppard et al. 2011; Shaw et al. 2009; Burch 2010) and community engagement frameworks (Ramaswami et al. 2014) . Community engagement and future scenario modeling has also been successful at the national level to approach issues of wat er uncertainty in a changing climate, and alternative water governance techniques in a location facing ongoing 1. The community involved comprises its own unit of identity. 2. Research builds on the strengths and resources within the community (takes advantage of social capital). 3. Incorporates collaborative partnerships in all phases of research. 4. Integrates knowledge and subsequent action. 5. Facilitates a co learning and empowering process that allows for inequalities in knowledge and background. 6. U tilizes a cyclical and iterative process, and is not merely a study never to be revisited. 7. Views issues from the perspectives of all involved. 8. Shares findings and knowledge gained with all partners and stakeholders (transference).

PAGE 96

74 water management problems (Kuzdas et al. 2016; Kuzdas and Wiek 2014) . Involving the community in the development of sustainability indicators and sustainability goals/ targets allows community members and stakeholders to provide input on their goals and priorities, providing valuable data for future management decisions. Furthermore, as the community integrates into the process, they become more invested in the outcome (Fraser et al. 2006) . Case of San Luis Valley This research takes lessons learned from applications of community based sustainability research and applies them to the agricultural region of San Luis Valley (SLV) in So uthern Colorado . Sustainability assessment and planning has a long history in SLV . For over a decade the community has been working with the state to develop a plan to comprehensively understand, and sustainably manage, the arid agricultural region's vast aquifer system (Gibson and et.al. 2015) . As part of this effort, the Colorado Division of Water Resources (CDWR) has developed a robust groundwater model that is used for setting water use targets and monitoring progress on aquifer management (CDWR 2015d) . The region was also the subject of an Environmental Protection Agency (EPA) pilot study on regional sustainability metrics (Hopton et al. 2010) . EPA considered SLV an ideal study location because of its distinct hydrologic boundaries, limited population, large amount of publically owned land, and interest expressed by government agencies and the local population (U.S. EPA 2010) . The study developed and calculated four metrics for sustainability in the region over time.

PAGE 97

75 The results were presented in a suite of publications in the Journal of Environmental Management (Heberling et al. 2012; Hopton and White 2012; Eason and Cabezas 2012; Campbell and Garmestani 2012; Heberling and Hopton 2012) . The present study builds on the work of EPA by adding two new su stainability indictors, a greenhouse gas (GHG) emissions model (Dubinsky and Karunanithi 2017a) and a consumptive water use model (Dubinsky and Karunanithi 2017b) , to the suite of metrics. This chapter outlines the community engageme nt portion of the research, which involved local stakeholders in regional sustainability indicator development and future scenario selection for modeling with the indicators. Goals T he overall goal of engagement with the community in this case was threefo ld: 1. Utilize local knowledge and expertise to better understand the baseline results and future trajectories of the GHG emission and consumptive water use indicator models. 2. Collaborate with community representatives to develop realistic future scenarios fo r modeling with the two sustainability indicators. 3. Transfer the baseline sustainability indicators and future scenario model results to decision maker s in the region. The following sections outline the involvement with the community in SLV from first contact; through formation of , and collaboration with , a Community

PAGE 98

76 Advisory Board , and culminating with knowledge transfer. The research describes why having the community involved produced relevant outcome s , and present s best practices / learne d along the way. Without these steps involving the community, the r esearch might be generic , less relevant , and less useful to the region . O utcomes from this work suggest that community involvement not only fostered access to critical information and dat a sources for the researchers , but also helped shape the way the community understands sustainability in their region. Methodology A graphical representation of the community engagement framework that guided th e work conducted on the SLV from 2013 2016 is presented in Figure 13 . Because the research involved human subjects, it required review and approval by the University of Colorado Denver's Institutional Review Boa rd (IRB) in November of 2013 . The IRB declared the research as "exempt" status, meaning minimal risk to the human subjects involved.

PAGE 99

77 Figure 13 : Framework for community engagement in San Luis Valley . The number in the circle indicates the ord er of each step. * The scenario mode ling methods and results are presented in C hapter V . Defining the Research Question Based on initial discussions with research collaborators , the primary research question that guided this work was ident ified as : What scenarios and/or issues do the community of SLV identify as important to the sustainability of the region? This question guided the research from stakeholder identification through knowledge transfer (steps 2 through 7; Figure 13 ), and fulfilled the goal of defining and conceptualizing future scenarios for modeling .

PAGE 100

78 Stakeholder Identification Identifying stakeholders requires gain ing an understanding of individuals, groups, and organizations affected by decisions faced while shaping the research, and the results f rom the outcome of the research (Freeman 1984; Reed et al. 2 009; Rowe and Frewer 2000) . Stakeholder identification began with multiple trips to SLV (the research office w as in Denver, ~230 miles away) to attend community meetings and social forums related to the topic of sustainability in the region. These visits presented an opportunity to meet people and begin conversations about the research topic . Over the course of 5 multi day trips to the region , 40 impromptu one on one interviews were conducted ( Figure 14 ). Figure 14 : Interview guide used to spark discussion when informally engaging with residents of SLV . These one on one interviews took place at community forums and acted as a conversation starter, but were not used quantitatively. The engagement process was evolving in real time and the most beneficial outcome of the one on

PAGE 101

79 one interviews turned out to b e a list of names of community leaders whom we were encouraged to meet. These community leaders were contacted and more in depth interviews were scheduled to increase understanding of the community context as it relates to sustainability . These early stages of the engagement process required time, commitment, and resources. As t h e process unfolded, it helped produce an understanding of the uniqueness of the region, led to key relationship building, and ultimately improved stakeholder parti cipation in the research because the individuals in the community saw the commitment of the researchers . Due to the complex and deliberative nature of the research question, the geographic distance between the research o ffice and the community, and limite d time and money, it was determined that the best way to engage with the community over the course of the project would be though a Community Advisory Board (CAB) composed of stakeholders identified through the interview process. In total , nine unique gro ups were identified as key stakeholders that should be engaged in the sustainability research. The stakeholder groups are listed below in alphabetical order. Individuals representing the various stakeholders were contacted via telephone and invited to jo in the CAB. 1. Alfalfa farmers 2. The c ultural and physical identity of the region 3. The e conomic stability of the region 4. Environmental/non governmental advocacy groups 5. Potato/grain farmers 6. Municipalities and local governments 7. National/state wildlife and parks ser vices 8. Ranchers

PAGE 102

80 9. Water conservation/ conservancy districts Community Advisory Board The establishment of local deliberative groups is an important component of CBPR, and requires many considerations (Reed et al. 2009) . Th e size of the group is one such consideration . F or example , a larger group can allow greater representation from the community, more expertise, and more knowledge transfer. Larger groups, however, may have a more difficult time reaching a consensus, oper ate less smoothly, and limit opportunity to develop interpersonal relationships (Reed 2008) . This SLV CAB was comprised of a diverse group of stakeholders, and represented identified key segments of the community as it related to sustainability . The size of the CAB was not fixed (new members were always welcome to attend meetings and participate) but averaged around 10 15 of mostly core members. This size was managea ble while still representing the core industries and issues of the region ( Figure 15 ).

PAGE 103

81 Figure 15 : Regional map of San Luis Valley with geographic distribution of the Community Advisory Board members *. * T he size of the flag is correlated to the number of board members from that location (image adapted from https://www.colorado.gov/pacific/townoflajara/slv). Another consideration in forming the CAB was geographic. In rural areas such as SLV that are large and have low population density, it was important to include community members from remote areas. This require d home visits, phone calls, and flexible scheduling of meetings to ensure that geographic isolation d id not result in exclusion of individuals or key stakeholders (Fraser et al. 2006; Reed et al. 2006) ( Figure 15 ). The CAB met five times ( July 2014 to Jul y 2016) to discuss the research project, inform the development of the sustainability indicator baseline s , identify

PAGE 104

82 issues relevant to the region, and conceptualize future scenarios for modeling. The meetings were loosely structured, and allowed time for research updates, stakeholder review of the two sustainability indicator models, discussion on future scenarios, and an open forum for board members to provide input and share relevant current happenings in the region . The meetings lasted ~2 hours and were audio recorded and transcribed for review , so that all information could be preserved . The CAB members' cumulativ e knowledge was important to understand results and their input on future scena rios shaped the direction of the research. Additionally, at each meeting board members were asked whether there were any significant stakeholder groups missing, and adjusted ou r membership accordingly by extending invitations through email and phone to new members to participate on the CAB . Co develop S ustainability I ndicator B aseline The two sustainability indicators selected for modeling on the region were greenhouse gas emi ssions (Dubinsky and Karunanithi 2017a) and consumptive water use (Dubinsky and Karunanithi 2017b) . The CAB did not participate in select ion of the indicators, as this occurred prior to the start of the engagement process. However, they were involved throughout data collection and evaluation phase. Because the CAB was composed of individuals with knowledge and information that would otherwise be unavailable to researchers, especially researchers not from the region, CAB meetings pre sented an opportunity for critical review of the sustainability indicator models. For example, it was easy for

PAGE 105

83 some members of the CAB who are heavily involved in the water issues of SLV, to notice and point out suspicious results from the consumptive wat er use model. Though this iterative process required more time and effort, it allowed us to revisit initial model assumptions, thereby shaping the indicators to be more accurate and better meet the needs of the community. Selecting S cenarios The main goal during CAB meetings was to conceptualize potential future scenarios for sustainability modeling in the SLV. Scenarios represent plausible futures that are useful to systematically explore the feasibility and efficacy of proposed policy decisions, bef ore decisions have actually been made (Swart et al. 2004; Kishita et al. 2016; Postma and Liebl 2005) . Communities and local governments can use clear, well defined scenarios coupled with sustainability indicators to help decision maker s understand the potential implications of their policy choices (Reed 2008; Fraser et al. 2006; Wilmsen 2008; Brugmann 1996) . The research team planned to utilize the two sustainability indicators to model some potential future scenarios of interest to the community. The scenario selection process required simplification of complex issues into analytically and intuitively tangible ones (e.g., scenarios that could be defined in models) (Trutnevyte et al. 2011) . The challenge becam e how a team of academics, industry representatives, and local decision makers refine broad conversations into meaningful scenarios, and identifying the guiding questions for the process.

PAGE 106

84 The three guiding questions used to collaboratively identify and ultimately select scenarios were: 1. Relevance and desirability to the region . Does the scenario fit in with what is already being discussed in the region, or is it a brand new idea? 2. Feasibility of the scenario . Does the scenario technically work or are there legal constraints, are there modeling constraints etc.? 3. The ability to quantitatively measure the outcomes . Can the GHG emissions model and the consumptive water use model capture the proposed changes to the system? Using these guiding p rinciples, a process similar to concept mapping (Trochim and Cabrera 2005) was employed to help structure the thoughts, opinions, and ideas of the stakeholders over three CAB meetings. Meetings were used to i dentify issu es of concern, suggest initia l potential scenarios, discuss with CAB, add or remove issues as warranted, and tweak scenarios. A white board was used to map ideas and the list evolved and was refined substantially over the brainstorming process. As an exa mple, initially the idea of geothermal resource utilization was proposed as a potential scenario for modeling. H owever , the deliberative process rev ea led that numerous companies have drilled exploratory wells, and the companies determined it was not economically practical in SLV . This local knowledge was key in selecting relevant future scenarios, which eventually were refined, using consensus among participants, and categorized under two broad themes , scenarios based on solar

PAGE 107

85 energy resourc e utilization in SLV and scenarios based on changes to the cropping regime in SLV . Results Future Scenarios During the deliberative brainstorming process, which took place over three CAB meetings, the CAB narrowed its focus to two main areas for future sc enario modeling Ñ solar energy development and changes to the cropping regime in the region ( Table 8 ). These scenarios are not intended as recommendations for the regi on, but as relevant, feasible, and measurable scenarios for proof of concept modeling using the two sustainability indicators developed in collaboration with EPA .

PAGE 108

86 Table 8 : The future scenarios selected by the CAB for modeling with the sustainability indicators. Description Solar Energy Development Scenario The solar energy development scenario is composed of three development pathways: 1) Business as Usual 2) Department of Energy/Bureau of Land Management strategy (1,000 additional MW capacity on public land) 3) Community response based on SLV/Solar Transmission Line Alternatives and Redundancy Recommendations (250 additional MW capacity on private farmland ) Crop Regime Change The crop regime change scenario is based on the goal of maintaining the agricultural heritage of the region while working toward a sustainable aquifer system, which is required by Senate Bill 04 222 (Entz et al. 2004) and the new groundwater rules and regulations (CDWR 2015d) . Crop regime changes explored in the scenario are: 1) Alfalfa land fallowing 2 ) Shorter alfalfa rotations in combination with small grain 3) Potato/grain land f allowing 4) Increases in green manure rotations among potato farmers Solar Energy Development Scenario: The SLV has the highest solar energy potential in the state of Colorado and one of the highest in the United States (SLV Development Resources Group 2013) . The CAB was able to articulate how the local community perceives solar resource development, including positive and negative aspects of alternative approaches. The Department of Energy in collaboration with the Bureau of Land Management (DOE/BLM) produced the Solar Developm ent Strategy for Western St ates, which identified SLV as a key location in the country for solar potential, and recommends large amounts of private investment on state and federal lands in SLV for solar energy infrastructure (DOE/BLM 2012) . This plan has received mixed reviews from the community . One reason for contr adictory perspectives is

PAGE 109

87 that all of the growth in solar energy development would be on undeveloped public land , which is perceived as an asset to the community. In addition, the perception is that b ecause this is federal land, the tax revenue will only h ave been seen at the state and federal levels, and the local municipalities will not benefit much from the economic growth . During these conversations, the existence of the SLV Solar/Transmission Working G roup , which produced the SLV/Solar Transmission Lin e Alternatives and Redundancy Recommendations working paper (STARR), became known to the research team (SLVEC 2010) . STARR is a document developed by the San Luis Valley Ecosystem Council in collaboration with ~200 concerned citizens as a response to the DOE/ BLM report. STARR proposes solar development, though not as much as the DOE/BLM pathway ( Table 8 ) , and suggests growth take s place mainly on private irrigated farmland . Because water usage is a prima ry concern in the region, displacing irrigated farmland with solar energy infrastructure has two perceived advantages over public land development , including : 1) reducing water consumption by removing irrigated lands from production , and 2) provid ing a new source of revenue for farmers as an incentive to no longer grow crops on their land. These two solar development pathways were selected for future scenario modeling because they closely met our requirements for being relevant, feasible, and measurable as described above.

PAGE 110

88 Crop Regime Change Scenario : The economy of SLV is dominated by the agricultural sector (SLV Development Resources Group 2013) . The agricultural industry relies heavily on the underlying aquifer , which spans much of the 8,000 sq km (3,200 sq mi) valley floor ( Figure 16 ) and is estimated to hold more than 2,466 billion m 3 of water (2 billion acre ft) (Colorado Geological Survey 2016) . However, overuse and climactic variation are causing the aquifer to deplete. Figure 16 : San Luis Valley with aquifer boundaries mapped in yellow and irrigation wells in green (Colorado Geological Survey 2016) . This over use o f groundwater has caused a 1.2 billion m 3 (1 million acre ft ) drop in the aquifer level following th e historic drought , which starte d in 2002 (Gibson and et.al. 2015) . Aquifer depletion, and its effect on the highly regulated surface water flows in the region , has resulted in state rules for groundwater use

PAGE 111

89 and aquifer sustainability targets in the region (CDWR 2015a) . As part of the water use rules and regulations SLV is now divided into response areas, which deli neate the areas of groun dwater pumping that a ffect individual stream systems ( Figure 17 ). Well owners that fall in response areas have the option to either opt in to self governing water sub districts tasked with developing plans and implementing actions to meet these new water use rules or manage their individual stream impacts and sustainability requirements on their own. Figure 17 : San Luis Valley with groundwater pumping response areas outlined in purple (S ub district s fall within the boundary of response area s ) (CDWR 2015d) . The crop regime change scenario developed with the CAB , specifically those heavily involved in the wa ter user community, consist of a combination of shifting of cropping patterns, fallowing of land, and increased soil health

PAGE 112

90 measures with the target of bringing water consumption in line with average water inflows ( Table 9 ).

PAGE 113

91 Table 9 : Crop rotations under the crop regime change scenario. Rotation Baseline scenario (avg. 2000 2010) Acres Future scenario (2025) Acres Potato/Grain 123,053 62,915 Potato / Green manure Cover Crop 9,862 45,000 Potato/Grain Land Fallowing ----25,000 Alfalfa/Grain (7 yr . alfalfa 1 yr. barley) 159,021 0 Alfalfa / Grain (5 yr . alfalfa 2 yr. barley) ----151,521 Alfalfa / Grain Land Fallowing ----7,500 Continuous grain 14,930 14,930 Wheat Winter All 1,226 1,226 Meadows or Pasture 211,198 211,198 Vegetables 4,804 4,804 Total irrigated acres 524,094 491, 594 The proposed crop rotational changes ( Table 9 ) are considered "likely" scenarios that would help SLV to reach the water use reduction target recommended by the members of the board representing in the water user community . Putting toget her a crop regime change scenario into a plausible plan would have been hit or miss without the expertise of the CAB . Once the crop changes were identified, many more offline meetings and phone conversations were held with individual members of the CAB. The meetings and discussions led to the specific details of the crop regime change scenario. Once an agreement was reached on the relevant, feasible, and measurable scenarios (solar development and crop regime change ) , the CAB and the research team worked to understand and clarify the underlying assumptions necessary to model these scenarios using the two sustainability

PAGE 114

92 indicators. This was a dynamic and iterative process that took place over many months. The results of the scenario modeling will be prese nted in C hapter V . Training and K nowledge T ransfer An important component of this project was to make research methods and their results available and usable by the community . This was accomplished by teaching people how to update data for calculating the regional sustainability indicators, and how to interpret the results . Given this need, a subs et of the CAB was identified to receive additional information and training . It was necessary to incorporate key members of the CAB in this new group to ensure a smooth transition and allow us to transfer knowledge in a n effective way. In order to do this , selected community members participated in three technical sessions describing the models, understanding the assumptions, and learning abo ut the various sources of data used . These sessions focused on how the Excel template s were built , how data were collected, and how to maintain and update the models . The group was given full access to all data, model calculations, and Excel spreadsheets via Google Drive . Discussion Best Practices and L essons L earned Throughout this process, best practices were identified when engaging a rural agricultural community around issues of sustainability.

PAGE 115

93 1. Involve the community as early as possible in the resear ch process. When input from the community help shapes the research, the community will be more engaged and invested throughout the process. This was shown with the development of the scenarios because the CAB was involved in the ir creation, they were interested in their outcome. This also promoted transparency around the underlying assumptions of the indicator models, as the members were a witness to their refinement. 2. Partner with existing groups and organizations in the region. These groups are not only a source of interested participants, but they help gain acceptance of an outsider working within the region. Some of these groups have been working on similar issues for years, have been vetted by the community, and are generally composed of well connected individuals. Furthermore, connecting with existing groups early on in the process helps to prevent duplicate efforts and promote trust . In cases where overlap did occur , those researchers were contacted to avoid conflicts and enhance results. 3. Communicate frequently (although not so frequently as to be an imposition) with CAB members, keeping them informed with regular status updates. If communication lapsed for more than a couple of months, members could potentially lose interest in the research . One way to stay connected was to regularly send updated results from the indicators via email and/or phone calls . Because Internet access was inconsistent

PAGE 116

94 throughout the region, phone calls were identified as the preferred method of communication for many members. 4. Openness and flexibility is key when engaging as a non resident from the region of study. This is true both in terms of schedules as well as in research direction. The less solidified the research directions are at the beginning, the greater potential the research will benefit the region and the researchers . Limitations The work summarized here follows the tradition of CBPR research, w hich recognizes that each research project presents specific considerations an d requires a tailored approach . For example, the CAB was not involved in choosing the two sustainability indicators (GHG emissions and consumptive water use ), but the y did work to critically evaluate the indicators during their development . The research team used its expertise and understanding of sustainability to identify two indicators capable of capturing relevant issues well before the community engagement process started. T he importance of the indicators and how and why they were chosen was explain ed to the CAB . Working with the CAB to identify the indicators could have been a more impactful approach, but was not possible given the constraints of the research timeline. The CAB was an effective way for a research team with limited funds and personne l to engage with a cross section of the community in the region. However, this approach came at the cost of limiting the number of voices in the

PAGE 117

95 conversation. Though the attempt was made to convene a CAB that represented a diverse constituency, there we re stakeholders who did not have access to the forum. These absent voices included representatives from a younger generation, schoolteachers and administrators, migrant farm workers, social justice advocacy groups, the aquaculture community and many other s. As demonstrated in this study, there are numerous benefits to including stakeholders in the research process. There are always limitations to a community based research project, but that does detract from the inherently grounded and meaningful results that emerge from research immersed in the community being studied. Conclusion The application of CBPR in SLV was successful in refining sustainability indicators and selecting future scenarios for modeling us ing the region specific indicators. Throughout the study , the community members were actively involved in the CAB . Their involvement provided local context of the issues and local knowledge about potential solutions , which allowed for the creation of rel evant, and realistically implementable scenarios. Because members of the CAB were involved throughout the process, it eased the transfer of information to members of SLV. Thus, we suggest similar studies would benefit greatly by using CBPR in sustainabil ity research in this geographic area, and others.

PAGE 118

96 CHAPTER V FUTURE SCENARIO MODE LING Future Scenario Modeling in San Luis Valley Utilizing Locally Calibrated Sustainability Indicators Abstract Scenario modeling can be used as a tool to guide decision maker s towards desired outcomes from policy decisions. When coupled with sustainability indicators, future scena rio modeling can inform stakeholders about a variety of social and environmental issues. This research engaged the rura l agricultural community of San Luis Valley around two relevant sustainability issues, water use and greenhouse gas (GHG) emissions. A local deliberative body, Community Advisory Board (CAB), worked with the research team to identify and formulate some likely futur e sc enarios to be used with two locally calibrated sustainability indicator models. The CAB identified some goals and targets and the scenarios they selected focused on crop regime changes and solar energy development. The results indicate that through speci fic shifting of cropping rotations and some minimal land fallowing , SLV could reduce water use and GHG emissions while increasing soil carbon and improving soil health . In addition, the solar energy development pathways investigated by this study show tha t the potential exists to offset most or all of the region's GHG emissions through renewable energy carbon credits with minimal new water use. This research is timely for policymakers, and adds to the growing field of sustainability science.

PAGE 119

97 Introduct ion The field of sustainability science has emerged to understand, explain, and present alternatives to humanity's struggle to co exist with the natural systems that support the planet (Spangenberg 2011; Clark and Dickson 20 03; Kates et al. 2001) . The discipline seeks to point the way toward a more sustainable society (Komiyama and Takeuchi 2006) . A sustainable society is one that meets the needs of the present generation without compromising future generations' abilities to meet their own needs (Brundtland 1985; United Nations Division for Sustainable Development 1992) . Two tools used in sustainability science are sustainability indicators and future scenario modeling. Sust ainability indicators help to simplify complex environmental and social ecological issues down to a measurement that can be used in targeting goals (Morse 2015; Rametsteiner et al. 2011; Ramos 2010; McCool and Stankey 2004) . An example of a sustainability indicator is greenhouse gas emissions, which is correlated with human induced climate change (IPCC 2014) . Future scenario m odeling is a tool which presents plausible futures that are useful when s ystematically exploring the feasibility and effi cacy of proposed policy choices before decisions have actually been made (Swart et al. 2004; Kis hita et al. 2016; Postma and Liebl 2005) . Communities and local governments can use clear, well defined scenarios coupled with sustainability indicators to help decision maker s understand the potential implications of their policy choices (Reed 2008; Fraser et al. 2006; Wilmsen 2008; Brugmann 1996) . An example of scenario modeling using a

PAGE 120

98 sustainability indi cator was presented by t he Intergovernmental Panel on Climate Change (IPCC) in the ir 1990 Policymakers Summary (Nakicenovic et al. 2000) . In the report they present four emissions scenarios based on potential policy choices ( Figure 18 ). Figure 18 : T h e four IPCC emissions scenarios, presented in the 1990 summary to policymakers (IPCC Working Group I 1990) , expressed in both radioactive forcing and equivalent carbon dioxide concentrations * . * In the Business as Usual Scenario (Scenario A) , the energy supply is coal intensive , and on the demand side only modest efficiency increases are achieved. In Scenario B the energy supply mix shifts towards lower carbon fuels, notably natural gas, and large efficiency increases are achieved. In Scenario C , a shift towards renewables and nu clear energy takes place in the second half of the 21 st century and CFCs are now phased out, and agricultural emissions limited. For Scenario D , a shift to renewables and nuclear energy reduces emissions of carbon dioxide initially, more or less stabilizin g emissions in the industrialized countries. The field of sustainability science has grown and there are now numerous examples of communities partnering with researchers to use future scenario

PAGE 121

99 modeling for sustainability analysis (Trutnevyte et al. 2011; Mistry et al. 2014; Marques et al. 2009; Ramos 2009) . Together, sustainability indicators and future scenario modeling can be used to track a region's progress on targets from the past to some potential futures. When sustainability scenarios are developed in conjunction with stakeholders, findings can be more meaningful and result in higher likelihood of adoption (Fraser et al. 2006; Ramaswami et al. 2014; Shaw et al. 2009; Kuzdas and Wiek 2014) . When collaborating with stakeholders on sustainability modeling, the first step is defining goals, which are broad qualitative statements about the objectives of the modeling. The second step is selection of indicators, which are quantitative measures to assess progress towards or away from the stated goals. The third step is identification of targets (i.e., a desired level of emissions, a reduction in resource consumption etc.) (Kates and Parris 2003) . In C hap ter IV, the methods used to form a deliberative body, the Community Advisory Board (CAB), and conceptualize future scenarios for modeling are described. This chapter presents the methods and findings from the future scenario modeling using the scen arios i dentified in C hapter IV and the sustain ability indicators outlined in Chapter II and C hapter III. Together with the CAB, the following goals were identified which met the research objectives and satisfied community interests : 1) reduce atmospheric GHG emi ssions , and 2) maintain availability of fresh water in the region. GHG emissions and

PAGE 122

100 consumptive water use were identified as appropriate indicators , capable of measuring progress toward the goals. Subsequently, the CAB identified two major scenario them es related to the stated goals: 1) Solar energy development in SLV with the goal of reducing GHG emissions and consumptive water use with no specific target . Climate change is driven, at least in part, by increasing concentrations of GHGs such as CO 2 , CH 4 , and N 2 O (IPCC 2014) . Scenarios with a goal of reducing human effects of climate change include some combination of reducing GHG emissions and/or promoting carbon sequestration (Kates and Pa rris 2003) . One way to reduce GHG emissions is by increasing renewable energy use/production, which lowers the GHG intensity of electricity. In SLV, there is an estimated 100 MW of solar energy production not including private homes (Barrows et al. 2016) . The s olar e nergy d evelopment scenario explored two possible pathways for increased solar energy development in SLV, and assessed im plications of those pathways using the sustainability indicators (i.e., GHG emissions and consumptive water use). 2) C hanges to the cropping regime with the goal of increasing groundwater availability by balancing water use with water inflows. Colorado Di vision of Water Resources (CDWR) issued new water use rules and regulations governing the SLV region (CDWR 2015d) . These rules are based on the findings of the Rio Grande Decision Support System (RGDSS) groundwater modeling team who have shown that groun dwater pumping in the region affects

PAGE 123

101 the natural stream flows, thereby impacting surface water rights holders. Surface water rights are protected under western water law, which is based on the doctrine of prior appropriation (Benson 2012) . The purpose of the groundwater legislation is twofold: 1) repay injurious depletions to surface water rights holder s caused by groundwater pumpin g , and 2) planning for and maintain ing a sustainable aquifer system. Stream depletion compensation caused by groundwater pumping and aquifer sustainability requirements are based on activities in modeled response areas in SLV ( Figure 19 ). The response areas delineate the extent of impact from groundwater pumping on the individual stream systems in the region. Owners of groundwater wells that are in a response area have the following two options: 1) opt in to self governing water sub districts (based on their response area) tasked with developing plans and implementing actions to meet these new water use rules , or 2) manage their individual stream impacts and sustainability requirements on their own.

PAGE 124

102 Figure 19 : San Luis Valley with groundwater pumping response areas/sub districts outlined in purple (CDWR 2015d) . Based on the CDWR model's annual response function analysis of stream impacts due to groundwater pumping, and estimates of conditions necessary (i.e., inflows, pumping amounts, recharge) to maintain a sustainable aquifer, each sub district (or individual well owner) will be required to reduce groundwater pumping and repay inj urious deplet ions accordingly. Methods Crop Regime Change Scenario The crop regime change scenario was based on the target of reduc ing pumping in the region as a whole by ~ 98 .5 million m 3 ( ~80,000 ac ft .) within the next 10 years, which is what is thought to be feasible/ needed to move towards a

PAGE 125

103 balance between water use and average annual stream inflows (i.e., discharge = recharge) . The scenario assumes the following changes to the crop rotations in an attempt to meet the groundwater use reduction target in SLV : 1) a change in rotational practice among the alfalfa growers from 7 years of alfalfa and 1 year of grain to 5 years of alfalfa and 2 years of grain ( alfalfa requires 24 in ches of water ; other small grain s require 18 in ches of water annually (1 inch of wat er is equal to 1 inch of water covering every acre of land growing a particular crop. The equivalent volume is 0.0254 m (1 inch) x 4046.8 m 2 (1 acre) = 102.8 m 3 per acre (27,156.9 gallons per acre)) 2) f allow ing of 7,500 acres of alfalfa /grain (7 yr./1 yr .) rotation land , 3) f allow ing of 25,000 acres of land in the Potato/grain (1 yr . /1 yr . ) rotation land area , and 4) shifting 35,000 acres of land from a potato / grain rotation (1 yr./1 yr.) to a potato /green manure cover crop rotation (1 yr./1 yr.). These changes mean po tato production in SLV reduc ing by 12,500 acres, small grain production reduc ing by 4,446 acres, and alfalfa production reduc ing by 33,123 acres on an annual basis ( Table 10 ).

PAGE 126

104 Table 10 : Comparison between the baseline crop rotational acreage in SLV and the crop rotational acreage under the crop regime change scenario. Rotation Baseline scenario (avg. 2000 2010) Acres Future scenario (2025) Acres Potato/Grain 123,053 62,915 Potato / Green manure Cover Crop 9,862 45,000 Potato/Grain Land Fallowing ----25,000 Alfalfa/Grain (7 yr. alfalfa 1 yr. barley) 159,021 0 Alfalfa / Grain (5 yr. alfalfa 2 yr. barley) ----151,521 Alfalfa / Grain Land Fallowing ----7,500 Continuous grain 14,930 14,930 Wheat Winter All 1,226 1,226 Meadows or Pasture 211,198 211,198 Vegetables 4,804 4,804 Total irrigated acres 524,094 491, 594 The sections that follow describe the methods used to evaluate the crop regime change scenario using the two sustainability indicators: GHG emissions and consumptive water use. GHG Emissions Related to the Crop Regime Change Scenario First, a business as usual (BAU) scenario was developed based on the findings in C hapter II , and the assumption of no other major changes to GHG emissions in the region from 2012 to 2025. Next, components of the GHG emissions indicator that would be affected by the crop regim e change scenario were identified . The components identified for evaluation were 1) changes to soil GHG emissio ns from agricultural activities , 2) shifts in electricity usage from changes to groundwater pumping , and 3) changes in fertilizer use on

PAGE 127

105 potato/ green manure rotational land. The following sections outline the procedures for estimating changes in these areas. !"#$%&'&%()#**#"+* % % Major crop rotations in the region were assessed for soil GHG emissions using COMET Farm (CSU 2016) , a spatially explicit carbon flux model developed by researchers at Colorado State University . COMET Farm was designed to assist growers in estimating the carbon emissions associated with their farming operations by evaluating different management practices/options for reducing GHG emissions and sequestering more carbon. COMET Farm is an online tool that uses geospatial data, Web Soil Survey data (Soil Survey Staff 2016) , and the Daily Century (DayCent) biogeochemical model (Del Grosso et al. 2001) . DayCent is the underlying process model COMET Farm uses to estimate GHG fluxes. It is a widely used biogeochemical model that draws upon empirically derived equations to simulate daily carbo n and nitrogen fluxes in agricultural soil systems (Metherell et al. 1993; Del Grosso et al. 2001, 2006) . A detailed overview of the sources, and sinks, of GHG emissions from cropland systems considered in the COMET Farm tool is provided in C hapter 3 (table 3 1) of U.S. Department of Agriculture's "Quantifying Greenhouse Gas Fluxes in Agriculture and Forestry: Methods for Entity Scale Inventory" (Eve et al. 2014) . Input variables needed for COMET Farm include average temperatures and precipitation , surface soi l texture class , land cover/use data such as vegetation type, cultivation/planting schedules, amount and timing of nutrient

PAGE 128

106 amendments , and N inputs from the atmosphere and soil type. The model then calculates water content, temperature by layer, plant pr oduction and allocation by net primary production (NPP), decomposition of organic matter, mineralization of nutrients, N emissions from nitrification and denitrification, and methane oxidation (Del Grosso et al. 2001, 2006) . The outputs generated from COMET Farm include N gas flux (N 2 O, NO x , N 2 ), CO 2 flux, soil organic C and N, H 2 O and NO 3 leaching. These outputs are further aggregated to generate overall GHG emissions with the unit CO 2 e. The analysis was constrained by computational capabilities and time because the geographic region under examination was quite large. Therefore , it was necessary to select a representative sample of parcels to use in the model. Sampling was done with ArcGIS (ESRI 2013) and consisted of selecting 20 (~120 acre) parcels from different parts of the region for each of the four major crop types (parcel sample maps are presented in appendix C). Each sample set had a similar so il texture distribution as the entire rotational cropping area (i.e., the potato/grain rotational land sample soil texture distribution matches the entire regional potato/grain rotation land soil texture distribution). It is important that the distributio n of soil texture samples match the regional distribution of soil texture because texture has the largest single effect on soil GHG emissions output in the COMET Farm model. The result is an emission factor expressed in units of MT CO 2 e per acre for each of the major crops in SLV ( Table 11 ).

PAGE 129

107 Table 11 : Inputs used to assess the baseline GHG emissions in the region using the COMET Farm model, and the outputs produced for each of the crop rotations assessed. BASELINE Inputs Outputs Rotation Crop Water (in./acre) Nitrogen (lbs./acre) Tillage (events) MT CO 2 e/acre Potato/Grain Potato 16 190 2 0.83 Potato/Grain Barley 18 170 2 Alfalfa/Grain (7 yrs. alfalfa 1 yrs. barley) Alfalfa 25 11 No till * 0.79 Alfalfa/Grain (7 yrs. alfalfa 1 yrs. barley) Barley 18 80 2 Continuous Grain Barley 18 205 2 0.53 Meadow / Pasture Pasture grass & meadows 14 0 No till 0.07 Once a baseline was established, the crop regime changes outlined above ( Table 10 ) were input in the COMET Farm model ( Table 12 ). It was assumed that the cropping regime changes would be implemented immediately, and COMET Farm tracked the impacts to GHG soil emissions over a 15 year period. Again, as with the baseline scenario each of the land use types in the f uture scenarios resulted in an emission factor in units of MT CO 2 e ( Table 12 ).

PAGE 130

108 Table 12 : Inputs used to assess the GHG emissions of the crop regime change future scenario using the COMET Farm model, and the outputs produced for each of the crop rotations assessed. FUTURE SCENARIOS Inputs Outputs From To Crop Water (in./acre) Nitrogen (lbs./acre) Tillage (events) MT CO 2 e/acre Potato/Grain Fallow Fallow 6 (3 yr. to establish vegetation ) 0 No till 0.24 Potato/Grain Potato Green Manure Potato 14 100 2 0.29 Potato/Grain Potato/Green manure Green Manure 6 0 2 Alfalfa/Grain (7 yr. alfalfa 1 yr. barley) Fallow Fallow 6 (3 yr. to establish vegetation ) 0 No till 0.24 Alfalfa/Grain (7 yr. alfalfa 1 yr. barley) Alfalfa/Grain (5 yr. alfalfa 2 yr. barley) Alfalfa 25 11 No till* 0.97 Alfalfa/Grain (7 yr. alfalfa 1 yr. barley) Alfalfa/Grain (5 yr. alfalfa 2 yr. barley) Barley 18 80 2 * The tillage events needed for an alfalfa/barley rotation were allocated to the barley rotation, hence no tillage events were allocated to the alfalfa rotation in the COMET Farm model analysis. ($,-./#-#.0%1*,%2/")%&/"3+456.,/%73)8#+9% % GHG emissions from groundwater pumping were calculated based on electricity usage. Real d ata on electricity consumption from agriculture in SLV were obtained from Xcel Energy (Dallinger 2015) . Real and modeled data on regional groundwater pumping from agriculture were provided by CDWR (CDWR 2015b) . These data were used toget her to estimate average electricity needed per volume of groundwater use ( Table 13 ). This resulted in a region wide average of 297 kW h per ac ft. of consumptive grou ndwater use.

PAGE 131

109 Table 13 : Data used for estimating electricity needed per volume of groundwater use. Data from Xcel energy were only available from 2006 to 2010, so this period was used to establish the average. 2006 2007 2008 2009 2010 Avg. Agricultural Electricity Consumption (MWh) 104,053 102,048 120,552 109,942 130,526 ---Agricultural G roundwater use (ac ft) 415,639 385,207 383,408 336,429 398,925 ---kWh per ac ft of Groundwater U se 250 265 314 327 327 297 &/,,+%:6+3/,% ;".6.#"+* % Increasing soil biodiversity through rotational cover cropping using green manure has long been recognized as an effective way to increase soil fertility and soil water holding capacity (Raimbault and Vyn 1991; Altieri 2002, 1999; Fageria 2007) . The owners of Rockey Farm, an organic seed potato operation in SLV, are advocates for water conservation and promote practices to maintain and improve soil health (Scully 2014; Greenberg et al. 2014; Mintz 2014) . Data from Rockey Farm was used to estimate fertilizer application and water use on a potato field in SLV that is in a gree n manure rotation (1 yr. potato /1 yr. green manure). For 10 years , Rockey Farm has collected data for irrigation and fertilizer demand on their farm, which has shifted from the traditional potato/grain rotation (1 yr./1 yr.) to a potato/green manur e rota tion (1 yr./1 yr.). Since the shift, they have seen a decrease in fertilizer application by ~45% (from 180 lbs . N to 100 lbs . N for each potato crop ) (Rockey 2016) . This is because the green

PAGE 132

110 manure contributes to increased fertility i n the soil. Synthetic f ertilizer use contributes to GHG emissions both during the manufacture and production phase and during the use and application phase. In addition, due to improved soil moisture holding capacity from the increase in soil carbon from the green manure, Rockey Farm has shown a 10% decrease in irrigation application requirements for potatoes grown after a green manure cover crop (Rockey 2014) . The new potato/green manure rotations in the future scenario were based on data for water use and fertilizer use from Rockey F arm and adjusted for a market potato crop. This was needed as Rockey Farm is a seed potato grower and only utilizes 90 days of the growing season where market potato growers, which represent the majority in the region, use 110 days . The following formula summarizes the components of the GHG emissions indicator used to analyze the crop regime change scenario. !"#"$% ! !"! ! !"#$$#%&$ ! !"# ! !"! ! ! !"#$ ! !"! ! ! ! !"#!$%& ! !"! ! ! ! ! ! !"#$%&%'"# ! !"! Consumptive Water Use Related to the C rop Regime Change Scenario To understand the potential effect of proposed changes to the cropping regime on water consumption, water use for each crop was multiplied by average total acreage in the region to establish a BAU scenario. All consumptive water use changes (due to proposed crop regime changes) were calculated by

PAGE 133

111 subtraction from the BAU scenario. The single change in consumptive water use considered in the analysis was based on changes in cultivated acreage, and the subsequent water use changes , w ith one exceptio n, which is water use of a potato crop in a rotation with green manure. In that case, the potato crop water use was reduced by 10% in the future scenario based on data from Rockey farms (Rockey 2014) . The reduction in water use was modeled to occur on all new potato/green manure rotational land. Solar Energy Development Scenario The SLV region was identified as a key location for solar energy resources in a 2012 Department of Energy and the Bureau of Land Management assessment of solar development on public lands in the western United State s (DOE/BLM 2012) . The report calls for a 20 year build out of an additional 1 , 000 MW of solar energy production in the region. This would involve relaxing regulations on development activity on public lands as well as require new transmission infrastructure for export of electricity . One of the solar scenario development pathways considered, referred to as the DOE/BLM pathway, was based on the recommendations from this report ( Table 14 ) . The sec ond solar development pathway considered, referred to as the community response pathway, was based on the SLV/Solar Transmission Line Alternatives and Redundancy Recommendations working paper (STARR), which was developed by a community group headed by the San Luis Valley Ecosystem Council (SLVEC 2010) . This development pathway arose as a response to the

PAGE 134

112 DOE/BLM proposal, and stipulates that future solar development takes place exclusively on private irrigated farmland. To further differentiate this pathway from the DOE/BLM pathway, the CAB decided tha t solar production in the community response pathway should not exceed current transmission infrastructure ( Table 14 ). This transmission infrastructure limit assumes the successful completion of the upgraded transmission lines proposed by Xcel E nergy (Sullivan 2015) . This planned upgrade is expected to increase the current maximum solar development capacity by 2 5 0 MW. Table 14 : Summary of the solar development pathways explored under the solar energy development scenario. Additional MW C apacity (2025) Land Type New Transmission I nfrastructure DOE/BLM 1,000 Public land 526 mi of 345 kv lines Community Response 250 Private irrigated alfalfa land 0* * Assumes Xcel Energy's plan to upgrade the transmission corridor over Poncha Pass from 2 69kv lines to 2 115 kv lines is realized (Sullivan 2015) . GHG Emissions Related to the Solar Energy Development Scenario A business as usual (BAU) scenario was developed based on the findings from C hapter II , and the assumption of no other major changes to GHG emissions in the region from 2012 to 2025. Next, components of the GHG emissions indicator that would be affected by the solar energy development scenario were identified . The following components we re identified for evaluation: 1) life cycle emissions from solar energy infrastructure (pa nels, transmissions

PAGE 135

113 lines etc.) , 2) solar energy credits /offsets from solar development , and 3) changes in GHG emissions from land changing from crop production to so lar energy production. The following sections outline the procedures for modeling these areas. <#2,%=0-$,%()#**#"+*%2/")%!"$6/%>+2/6*./3-.3/, % Though production of solar energy does not produce GHG emissions locally, there are secondary emissions (upstream emissions) associated with manufacture, use, and disposal of solar facilities. The secondary emission factors used for the manufacture, use, and disposal of solar panels were 40 kg CO 2 e per MWh for photovoltaic (PV) and 20 kg CO 2 e per Mwh for concentratin g solar power (CSP) (Hsu et al. 2012; Burkhardt et al. 2011) , i. e . , the two technologies currently in use. It was assumed that new solar development would occur as 75% PV and 25% CSP, following the current proportions of solar panels used in the region. In addition to solar facilities, transmission infrastructure also has life cycle GHG emissions associated wi th them. The estimate of 300 MT CO 2 e per mile for 345 kv line and 2,000 MT CO 2 e per substation with an assumed average lifespan of 35 years was used (Jorge et al. 2012a, 2012b) . Estimating a lifespan allowed for allocation of these emissions annually. !"$6/%(+,/90%=/,4#.* % A carbon credit/offset was assign ed to the region based on avoided grid electricity use and associated emissions. The proportion of fuels used to generate electricity for SLV is 60% coal, 25% natural gas, and 15% renewable

PAGE 136

114 and the current local emission factor is 0.83 kgCO 2 e/KWh (Dubinsky and Karunanithi 2017a) . It was assumed that the increase in solar development would phase in linearly and be f ully implemented in year 2025, and credits were assigned accordingly. !?#2.%2/")%-/"8%8/"43-.#"+%."%*"$6/ % Land that is converted from crop production to solar energy production was assessed for changes in soil GHG emissions and electricity usage (e.g., pumping of groundwater). The method used to estimate electricity usage from groundwater pumping (kWh/ac ft) is described in the crop regime change scenario section above. In addition, GHG emissions from land use (i.e. , tillage, nitrogen applicatio n, irrigation) at a rate of 0.79 MT CO 2 e/ acre ( Table 11 ) would be avoided if alfalfa were no longer grown where new solar development occurs (it was assumed that new solar development would take place exclusively on alfalfa land) . The following f ormula summarizes the components of the GHG emission indicator used to analyze the solar energy development scenario: !"#"$% ! !"! ! !"#$$#%&$ ! !"# ! !"! ! !"#$% ! !"#$%&'$()'($* ! !"! ! !"#$% ! !"!#$% ! !"#$%& ! ! ! ! ! !"#!$%& ! !"!

PAGE 137

115 Consumptive Water Use Related to the Solar Development Scenario In order to understand how the consumptive water use in SLV would chang e under both of the solar development scenario's development pathways, a BAU trend line was proj ected using the average water use in the region from 1980 Ð 2010 (see C hapter III) . Next, components of the consumptive water use indicator that would be affected by the solar energy development scenario were identified. The following components were identified for evaluation: 1) new water use from the operation of solar facilities , and 2) changes in water use from land shifting from alfalfa product ion to solar energy production. 1*,%6+4%@8,/6.#"+%"2%!"$6/%A6-#$#.#,* % The quantity of water consumption f rom use and operation of the solar facilities was based on a harmonization study from the National Renewable Energy Laboratory (NREL), which developed water consumption factors for PV and CSP per Mwh (PV 6 gal / Mwh and CSP 890 gal / Mwh) (Meldrum et al. 2013) . This water use factor was applied to each of the solar d evelopment pathways. !?#2.%2/")%-/"8%8/"43-.#"+%."%*"$6/%8/"43-.#"+ % A change in consumptive water use on land converted from cropland to solar production occurred only under the community response development pathway. This change in water use was calculat ed based on a shift away from alfalfa cultivation (average of 61 cm (24 in.) per acre) to land used for solar energy production (PV 6 gal / Mwh and CSP 890 gal / Mwh).

PAGE 138

116 The following formula summarizes the components of the consumptive water use indicator u sed to analyze the solar energy development scenario: !"#"$% ! !"#$% ! !"# ! !"# ! !"#$% ! !"# ! !"# ! !"# ! !"#$%&'!( ! !" ! !"#$% ! ! ! ! ! !"#$ ! !"#$%&'(#) ! !"#$ ! !" ! !"#$% ! !"#$%&'(#) ! !"#$ Results / Discussion Results from the Crop Regi me Change Scenario The crop regime change scenario outlined in the methods section was developed with guidance from key members of the water user community who participated as members of this research's CAB. The scenario reduces crop cultivation in the re gion by 6.3% ( Table 15 ). Though land fallowing is perceived in the region as a negative practice, this amount of land fallowing was acceptable to the experts on the CAB. The scenario reduces potato production by 19%, and roughly half of the remaining land under potato cultivation shifts from a potato/small grain rotation to a potato/green manure cover crop rotation, while the other half remains in a potato/small grai n rotation ( Table 15 ). These changes increase green manure/cover crop acreage by 1 7,500 acres (7,082 ha), which a 350% increase from the BAU scenario. Small grains production was reduced in the shift from a potato/grain rotation to a potato/green manure rotation (12,368 acres (5,005 ha)). However, simultaneous alfalfa production in SLV shifted from a 7 yr. alfalfa / 1 yr. small grain rotation to a 5 yr. alfalfa / 2 yr. small grain rotation. This change in the length of an alfalfa stand allowed for more acres to

PAGE 139

117 be in small grains production in any given year compared to the BAU. Therefore, total annual small grains production was only reduced by 4.7% a nnually when compared to the BAU scenario. The crop with the largest decrease in annual acreage under the crop regime change scenario was alfalfa , at ~23% of the total annual acreage (mainly due to shift from a 7 yr. a lfalfa crop to a 5 yr. crop). Table 15 : Annual crop acreage in SLV (avg. 2000 Ð 2010; acres) as well as crop acreage under the fully implemented crop regime change scenario. Crop !"# $ %&'()*+,$-)./ 0 $ 1222 3 12425 6*,7$*'/+8'$ &9)(/' $ :&'()*+, $ ;'*&'(<$69)(/' $ Potatoes All 66,457 53,957 1 9 Grains (spring wheat, barley) 94,126 89,680 4.7 Cover Crops/Green Manure 4,931 22,500 35 0 Hay Alfalfa 141352 108229 23 Wheat Winter All 1,226 1,226 0.0 Meadows or Pasture 211,198 211,198 0.0 Vegetables 4,804 4,804 0.0 Total irrigated acreage 524,093 491,593 6.3 Changes in GHG Emissions d ue to the Crop Regime Change Scenario The changes in crop production under the crop regime change scenario resulted in a net GHG emissions reduction of ~60,000 MT CO 2 e in 2025 from the proj ection of the BAU scenario (~5% regional GHG reduction) ( Figure 20 ).

PAGE 140

118 Figure 20 : The impact on the GHG emissions indicator from the crop regime change scenario * . * The last data point is in 2012 and a computed value was used for 2025. No values were computed between 2013 and 2024. The bulk of these reductions can be attributed to decreased nitrogen fertilizer use and avoided GHG emissions associated with fertilizer application. This reduction i n fertilizer emissions is seen most clearly in the shift from an annual potato/grain rotation to potato/green manure rotation ( Table 16 ). This change in rotation, and subsequent reduction in fertilizer use, resulted in a ~47,000 MT CO 2 e per year reduction in the region. The shift to a potato/green manure rotation also resulted in higher levels of carbon seque st ration in the soil, which contributed to lowering regional GHG emission by ~10,000 MT CO 2 e. However, the shift from a 7 year rotation of alfalfa and small grain to a 5 year rotation with small grain resulted in a net increase in GHG emissions of 15,000

PAGE 141

119 MT CO 2 e per year. This increase from the BAU scenario captured by COMET Farm is due to the shorter rotation, which re quires an increase in acres being tilled in SLV in a given year. This is because alfalfa is a perennial crop and is not tilled during the l ife of the stand. Tillage is associat ed with increased GHG emissions because soil disturbance promotes elevated levels of soil bacterial decomposition/respiration resulting in CO 2 loss to the atmosphere (Paustian et al. 2000; Reicosky 1997) . Table 16 : C hange s in GHG emission s from soil in SLV comparing the baseline scenario and the crop regime change scenario . Rotations Acres (avg. 2000 2010) Acres 2025 Soil GHG Emissions (avg. 2000 2010) Soil GHG Emissions 2025 Difference (MT CO 2 e) (MT CO 2 e) (MT CO 2 e) Potato / Grain 123,053 62,915 96,646 49,439 47,257 Potato / Green Manure 9,862 45,000 2,595 12,852 10,036 Alfalfa/Grain * 159,021 151,521 131,588 146,404 +15,087 Fallowed Alfalfa / Grain ---7,500 ---1,787 1,787 Fallowed Potato / Grain ---25,000 ---2,829 2,829 Continuous Grain 14,930 14,930 7,874 7,874 0 Meadow / Pasture 211,198 211,198 14949 14,949 0 Total 518,064 518,064 218,122 175,916 46,822 *Alfalfa rotation changed from 7 year alfalfa / 1 year small grain to a 5 year alfalfa / 2 years small grains . A smaller fraction of the reduction of GHG emissions from the crop regime change scenario can be attributed to the reduced electricity consumed from groundwater pumping in the region due to crop changes. Both the reduction in irrigated land from fallowing as well as the increase in water holding capacity of soil in a potato /green manure rotation resulted in a net decrease in emissions

PAGE 142

120 b ecause of a decrease in electricity usage. This resulted in a net decrease of ~12,500 MT CO 2 e ( Figure 20 ). Changes in Consumptive Water Use due to the Crop Regime Change Scenario The crop regime change scenario resulted in an overall decrease in consumptive water use of 103 million m 3 (83,000 a c ft) in the year 2025 ( Figure 21 ). The biggest effects were due to decreases in alfalfa acreage and shorter alfalfa rotations (from 7 yr. to 5 yr.). These changes a ccount for 1/3 of the reduction in consumptive water use. The change from a potato/small grain rotation to a potato/green manure rotation, and the fallowing of potato/grain land, accounts for another 28% of the reduction in water use. Finally, changes in small grain acreage as a result of shifting away from a potato/grain rotation accounts for the remaining (6%) of the water savings ( Table 17 ). Table 17 : Results on changes in irrigation consumptive water use from the crop regime change scenario. Million m 3 Thousand Ac ft Percent Change BAU Scenario Future Scenario BAU Scenario Future Scenario Alfalfa 323.3 249.6 261.3 202.0 6.4 Potato 106.0 66.5 86.0 53.9 3.5 Small Grain 182.7 176.6 148.1 143.2 0.5 Green manure 2.1 13.1 1.7 12.7 +1.0 Meadows / Pasture 384.1 384.8 311.4 312.0 0.0 Winter Wheat 0.1 0.2 0.1 0.2 0.0 Livestock * 140.3 140.3 113.8 113.8 0.0 Total 1137.7 1030.8 922.3 835.6 9.4 *Livestock includes irrigation water use from feed crops.

PAGE 143

121 These reductions in water use meet the target set of 98 .5 million m 3 ( ~80,000 ac ft .) of water use reductions, which moves the region closer to balancing water use with average water inflows ( Figure 21 ). Water use greater than the average inflow of water (975 million m 3 annually) strains the system by depleting the aquifer , which in turn impacts surface water flows (this is discussed further in Chapter III). Scenarios such as this one can allow decision maker s to see which crop regime changes will have the best possible outcome from a su stainable water management perspective. Figure 21 : The average agricultural (crops and livestock) consumptive water use (black line) shown with the average water budget (average water inflows minus compact deliveries) (pink line ) in San Luis Valley * . * The colored wedges represent water use reduction from the crop regime change scenario . The last data point is in 2010 and a computed value was used for 2025. No values were computed between 2011 and 2024.

PAGE 144

122 Results from the Solar Energy Development Scenario Two solar development pathways in SLV were selected for future scenario modeling. The first was based on the DOE/BLM plan to lease public land for 1 , 000 MW of new solar facilities in SLV. The second pathway was based on a community response to the DOE/BLM plan and recommends that new solar facilities in SLV take place exclusively on ir rigated farmland. In addition, the CAB elected to limit new solar development in the community response pathway to not exceed current transm ission infrastructure. Therefore, this pathway would only increase solar production in the region by 250 MW on 810 ha (~2,000 acres) of irrigated alfalfa land. Changes in GHG Emissions due to the Solar Energy Development Scenario In the analysis, SLV was assessed for GHG emission reductions based on avoided grid electricity use and associated emissions. For every MWh of new solar production in the region, SLV was credited 0.83 MT CO 2 e , which is the direct and upstream emission factor associated with elec tricity in SLV (Dubinsky and Karunanithi 2017a) . When fully implemen ted in 2025, SLV would receive 425,000 MT CO 2 e credits annually under the c ommunity response pathway, and 2.1 million MT in CO 2 e credits annually under the DOE/BLM pathway ( Table 18 ). This GHG emissions credit/ offset had the single largest impact on the emissions profi le of SLV ( Figure 22 and Figure 23 ). In addition to credits for solar energy production, the community response pathway saw a decrease in CO 2 e from shifting land from alfalfa production to solar energy production. The loss of 810

PAGE 145

123 ha ( 2,000 acres) of alfalfa land, and the subsequent reduction in irrigation, reduced electricity consumption and lowered GHG emissions by 1,700 MT CO 2 e ( Table 18 ) . In addition, the COMET Farm soil carbon fl ux model showed a reduction in GHG emissions of 1,500 MT CO 2 e when the 2,000 acres (810 ha) are no longer tilled or amended with nitrogen fertilizer ( Table 18 ). Addit ionally, new GHG emissions associated with manufacture, use, and disposal of solar panels as well as new emissions from the additional transmission infrastructure were needed to support solar production. Under the community response pathway, the quantity of solar development was constrained by the transmission infrastructure, allowing for an additional 250 MW of solar development in 2025 (Doyle 2015; SLVEC 2010) . Therefore, no additional upstream emissions for transmission infrastructure were assigned. However, the manufacture, use, and disposal emissions from the community response pathway resulted in an increase of 25,000 MTCO 2 e ( Table 18 ). Under the DOE/BLM pathway the region would increase in solar capacity by 1,000 M W on non irrigated public lands. The additional transmission infrastructure needed to support this increase would require four new >461 MW substations and 536 miles of 345 kv transmission lines (DOE/BLM 2012) . In this pathway, the upstream use and disposal emissions equate to 124,000 MT CO 2 e per year, and the new transmission infrastructure needed to support this scenario is responsible for 4,700 MT CO 2 e per year.

PAGE 146

124 When considering net GHG emissions from solar development, the region can expect a 400,000 MT CO 2 e decrease from the projected B AU scenario in 2025 when following the community response pathway ( Figure 22 ). When assessing the DOE/BLM pathway for net GHG emissions, the region can expect a 1.9 million MT CO 2 e decrease from the projected BAU scenario in 2025 ( Figure 23 ). Table 18 : Results from the GHG emissions indicator when modeling the two solar development pathways in the solar energy development scenario. Solar Credit (MT CO 2 e) Groundwater Pumping Changes (MT CO 2 e) Soil GHG emissions Changes (MT CO 2 e) Manufacture use and Disposal of Solar (MT CO 2 e) Solar Transmission Infrastructur e (MT CO 2 e) Percent Change from BAU (%) DOE/BLM ! "#$%&#"'( ) ---------+123,518 +4,737 168 % Community Response 424,524 ! $#&*$ ) 1,510 +25,840 -----33 %

PAGE 147

125 Figure 22 : The impact on the GHG emissions indicator when the community response solar development pathway is implemented * . * The last data point is in 2012 and a computed value was used for 2025. No values were computed between 2013 and 2024.

PAGE 148

126 Figure 23 : The impact on the GHG emissions indicator when the DOE/BLM solar development pathway is implemented * . * The last data point is in 2012 and a computed value was used for 2025. No values were computed between 2013 and 2024. Changes to Consum ptive Water Use due to the Solar Energy Development Scenario The decrease of 810 ha ( 2,000 acres) planted in alfalfa under the community response pathway resulted in a 5 million m 3 (4,000 ac ft) annual reduction in water use. This was due to the avoided irrigation from alfalfa crop production. There was a slight increase in water use, annually, from the operation of the solar facilities for both t he community response pathway ( 850,000 m 3 (700 ac ft) ) and for the DOE/BLM pathwa y ( 4.2 million m 3 (3,480 ac ft)). When considering all water use reductions, as well as new water use from the operation of the facilities, the community response pathway results in a 4.2

PAGE 149

127 million m 3 (3,400 ac ft) decrease in consumptive water use annually from the BAU scenario, and the DOE/BLM pathway would result in a ~4.2 million m 3 (3,480 ac ft) increase in water use in the region. Both of these changes in water use represent a 0.4% shift from the BAU ( Table 19 ). Table 19 : Results from the consumptive water use indicator when modeling the two solar development pathways in the solar energy development scenario New Water Use (m3) Water Use R eductions (m3) Net Water Use Change (m3) Percent Change from BAU (%) DOE/BLM (#"+"#,+* ) ------)(#"+"#,+* + 0.4% Community Response .'.#'*+ '#$%"#,'* ) 4,244,114 0.4% Summary of Findings In this study, two future scenarios were outlined and calculated to demonstrate how sustainability targets could be identified and goals achieved while meeting community needs and interests. Scenarios were based on stakeholder interest and proposed policy . The scenarios were modeled and potential outcomes were presented to the local community. The regional sustainability goals considered were to reduce water consumption in order to sustainably manage the regions aquifer with the target of balancing water use with average water inflow s and to reduce GHG emissions in the region with no specific target. The first scenario explored changes to the cropping regime in SLV. The cropping changes developed with local experts included small amounts of land fallowi ng (6% of BAU), shortened alfalfa rotations (from 7 yr. to 5 yr.), and a shift

PAGE 150

128 from a potato/small grain rotation to a potato/green manure cover crop rotation. If practices outlined in this scenario were fully implemented, the results indicate a ~10% decr ease in consumptive water use on an annual basis, and a ~5% decrease in region wide GHG emissions. These results did bring regional water use closer to average water inflows, but fell short of fully balancing the water budget. This means that unless wate r inflows increase in the future, water use at this level may continue to deplete the aquifer. The bulk of the GHG emissions reductions in this scenario were due to decreased nitrogen fertilizer usage and the emissions associated with nitrogen application . The second future scenario investigated solar energy development in SLV. Two pathways for solar development were explored. One pathway, based on research by DOE/BLM, would increase solar energy production in SLV by 1,000 MW utilizing public lands. The second solar development pathway was based on a community derived response to the DOE/BLM pathway that would limit solar development in the region to current transmission infrastructure capacity (an addition al 250 MW above current levels). Th e community r esponse pathway would place new solar energy production in SLV solely on private irrigated farmland with the goal of reducing water use from irrigation while providing alternative revenue to the farmers . If the DOE/BLM pathway were to occur, the analysis reveals region wide GHG emissions would be offset by 168% through a carbon credit for renewable energy production. The DOE/BLM solar development pathway would increase consumptive water use in the region by

PAGE 151

129 0.4% due to use and operation of the solar facil ities. If the community response pathway were to occur, 33% of the regions GHG emissions would be offset through renewable energy carbon credits, and consumptive water use would decrease by 0.4% due to land shifting from alfalfa production (a water intens ive crop) to solar energy production. The solar scenarios explored would have large implications for GHG emissions reductions while having a relatively minor impact on water use in the region. The future scenarios explored in this research were d eveloped in collaboration with the local community and the analysis was successful in terms of providing insight to the implications of potential futures in regards to two sustainability indicators (GHG emissions and consumptive water use) . This research was aime d specifically at addressing issues in SLV, but the research is relevant for policymakers in all regions who are interested in measuring and managing various aspects of sustainability.

PAGE 152

130 CHAPTER VI CONCLUSION Summary of Results This dissertation contribu tes to the growing body of work related to sustainability assessment at the regional scale. As discussed previously, d efining sustainability and having quantitative tools to measure its various aspects is critical for local decision maker s. This research contributes to the growing interest in the topic of sustainability assessment, and builds on the recent efforts of the EPA to develop regional sustainability metrics (U.S. EPA 2010) in SLV by adding two new measures, a greenhouse gas (GHG) emissions indicator (Chapter II) and a consumptive water use (CWU) indicator ( Chapter III). A Community Advisory Board (CAB) in SLV helped guide the research and ensure its value to the region (Chapter IV). Finally, scenarios developed with the CAB were modeled using the two sustainability indicators (Chapter V), and the results w ere presented to the local community. The following sections outline the major findings and outcomes from each of the chapters of the dissertation. Chapter II: Greenhouse Gas Accounting The results from the GHG emissions indicator demonstrated that acti vities in SLV (primarily agriculture related) produce more GHG emissions per dollar of economic growth compared to urban areas. Previous research on carbon emissions have often focused on cities or countries (Hillman and Ramaswami 2010; Ramaswami et al. 2008; U.S. EPA 2013a) , so the revelation that rural

PAGE 153

131 areas are more intensive in terms of GHG emissions (on per GDP basis) is critical and suggests more work should focus on analyzing and improving the emission profile of rural agricultural regions. The GHG emission analysis also showed that 14% of the region's emissions come from sources beyond the geographic boundaries (e.g., fuel processing, elect ricity production etc.), the largest being from the manufacture and production of fertilizer. Available nutrients in soil are key for agricultural production so they cannot be ignored or avoided, but could potentially be better managed by following recomm endations to improve soil health. Lastly, the GHG emissions accounting indicator revealed that harnessing part of the solar energy available in SLV, and offsetting electricity produced from fossil fuels, is key to reducing SLV's GHG emissions profile. Th e research demonstrated that SLV could become carbon neutral by increasing solar energy production. Chapter III: Consumptive Water Use Analysis The CWU indicator was developed with the purpose of providing new insights to the local community about the w ater use profile of SLV. In SLV, 85% of crop water demand is met through irrigation, whereas the global average is only 19%. This highlights the region's heavy reliance on irrigation water. For the first time, an analysis of water use from export crops w as separated from water use for local livestock (including the water used to grow feed). This analysis included categorizing all water use according to its source (i.e., rain water, groundwater, and surface water) (Falkenmark 1991; Food and Agriculture

PAGE 154

13 2 Organization of the United Nations (FAO) 1993; Hoekstra et al. 2011) . The results from the CWU assessment clearly demonstrate to stakeholders which specific agricultural activities draw on various water sources, and therefor allows water users to better understand the system. The CWU model revealed that livestock rely heavily on the surface water flows for feed, but they use significantly less groundwater than the export crops (i.e., potatoes, barley, alfalfa etc.) in the region on a per ton of product basis. This highlights the important niche that livestock hold in t he region's groundwater stressed agricultural economy. In addition, the analysis shows that SLV is experiencing water scarcity for seven months out of the year, and water shortage, resulting in aquifer depletion, for two months of the year. Ground and su rface water use must become equal to or less than the natural runoff, if the region is to stop depleting the aquifer and come into compliance with the states sustainability legislation. The CWU indicator is another tool that decision makers in SLV now hav e to inform water use policy decisions. Chapter IV: Community Engagement The community engagement portion of this research opened a dialogue between a rural agricultural community and a group of sustainability researchers at the public university. Among the key outcomes was the formation of a successful, involved, and interest ed CAB. Over the course o f two years, stakeholders offered critical assessment during the development of two regional sustainability indicators, and the CAB was involved in identifying future scenarios

PAGE 155

133 for modeling. The goal of involving the CAB in scena rio selection was to ensure that the scenario modeling be relevant to the region, feasible to actually implement, and measurable using the indicator models. Based on their interests and knowledge of the needs of the community, the CAB focused in on future scenarios involving changes to the crop regime as well as two different solar energy development pathways. In addition, three training seminars were held in the region in order to build capacity so that the community, if they choose, can continue to use l ocal data to update the region specific GHG emissions and consumptive water use indicator models in the future. The group has access to all source material and models needed to continue that work. Engaging with the local community enhanced the relevance of the research and results, and it confirmed that outreach beyond the comfort zone of the campus is going to be key to the advancement of sustainability science. Chapter V: Future Scenario Modeling The relevant, feasible, and measurable future scenarios developed in conjunction with the CAB (Chapter VI) were then used for future scenario modeling using the two sustainability indicators described in Chapters I and II. The future scenario modeling showed how changes to the crop regime in the form of rotati onal cropping adjustments as well as changes to local solar energy production would affect the GHG emissions and consumptive water use profiles in the region. According to the CAB the scenarios analyzed are plausible for SLV,

PAGE 156

134 and show the potential benefit s to policy makes that this type of analysis can offer. If practices outlined in the crop regime change scenario, which include shorter alfalfa rotations, a shift from a potato/grain rotation to potato/cover crop rotation, and a ~6% fallowing of irrigated crop land were fully implemented, the result would be a ~5% decrease in overall annual GHG emissions and a ~10% decrease in annual consumptive water use. This decrease in water use equates to 103 million m 3 (83,000 ac ft), which brings the region close to balancing average water use with average natural water inflows. Balancing water use with water inflows is needed to meet sustainability targets set forth by the state of Colorado. The solar energy development scenario compared business as usual to the fo llowing two alternative development pathways: 1) a DOE/BLM proposal to foster 1,000 MW of new solar energy development to occur on public land in SLV, and 2) a community response which suggests a 250 MW increase in solar energy production to occur on priva te irrigated farmland. The results show that solar energy has the potential of reducing/offsetting GHG emissions in SLV by 33% 168%. There would be a slight increase (<1%) in consumptive water use due to operation of the solar energy facilities, and a s light decrease (<1%) in water use due to displacement of farmland, but it is overall a relatively minor impact on regional water use overall. Finding ways to reduce the intensity of GHG emissions from agricultural activities, which significantly contribut e to global

PAGE 157

135 emissions, is a crucial step is combating climate change. Human induced climate change, and the associated cascading effects, is understood as a major threat to the social, economic, and environmental aspects of society and the planet. This s cenario offers an approach for mitigating agricultural GHG emissions. Future Work Economic Analysis Additional work can be done to understand how the future scenarios outlined in Chapter V impact the socio economic systems in SLV. Examining how the descri bed agricultural changes affect the socio economic stability of the region would add to the robustness of the work. We are currently in discussions with a research team at Colorado State University (CSU) that is working to develop a model to predict the e conomic effects of changes to the cropping regime specific to SLV. Using Implan, a county level economic input/output model (IMPLAN 2012) , the researc h team at CSU is able to predict how changes to the agricultural sector may cascade to other economic sectors (e.g., retail, manufacturing etc.). We believe that coupling EPA's sustainability metrics and the two new GHG emissions and consumptive water use sustainability indicators with this economic model could provide even more insight to help the community make sound management decisions.

PAGE 158

136 Sustainability Metrics and Future Scenario Modeling Conversations are ongoing with scientists at EPA to develop an ap proach for analyzing the effects that the future scenarios may have on the four sustainability metrics discussed in Chapter I (i.e., Emergy Analysis, Ecological Footprint Analysis, Green Net Regional Product, and Fisher Information). The GHG emissions and consumptive water use indicators, which were used to assess the scenarios, can inform policymakers about these two specific aspects of the system, but alone they do not specify whether the region is moving toward or away from sustainability. To do this, sustainability metrics must be used. Understanding how the future scenarios affect the metrics ensures that there are no inadvertent negative consequences in regards to sustainability resulting from implementing the changes to the region. What follows is the preliminary , but somewhat speculative, discussion on how the scenarios might affect each metric. The below section should be treated more as a qualitative brainstorm intended to initiate discussion around possible future research rather than concrete f acts or research findings. Further research is required to either confirm or reject these predictions. Crop Regime Change Scenario We expect that the crop regime change scenario might potentially affect the Emergy Analysis metric (Campbell and Garmestani 2012) in the following ways: 1) emergy inflows would decrease due to reduced demand for synthetic fertilizer imports as well as lowered electricity imports from reduced groundwater

PAGE 159

137 pumping, 2) Emergy exports would also decrease due to reduced agricultural production because of land fallowing an d rotational cropping adjustments, and 3) non renewable emergy usage due to groundwater extraction above the groundwater recharge rate will decrease because of the ~10% decrease in consumptive water use expected under the crop regime change scenario. The first two components discussed represents a relativity minor change to the system, however balancing groundwater use with the groundwater recharge rate could significantly improve the sustainability of SLV with regards to emergy. We anticipate that the not able component from the crop regime change scenario that might affect the Ecological Footprint metric (Hopton and White 2012) is decreased groundwater pumpi ng, which reduces electricity demand and therefore decreases the "e nergy land" footprint. Energy l and is needed in the metric for carbon sequestration from fossil fuel usage. This change would result in a slightly reduced Ecological Footprint, which is a net positive for the regio n. However, this reduction in energy l and is not expected to change the Ecologic Balance in SLV as a whole, so the scenario is not expected to shift the direction of sustainability. We expect that the crop regime change scenario might affect the Green Net Regional Product (GNRP) (Heberling et al. 2012) of SLV in the following ways: 1) decreased groundwater use is expected to increase groundwater storage, resulting in an increasing GNRP, however, effective measures fo r this analysis have not yet been identified, 2) increased green manure cover crop

PAGE 160

138 rotations are expected to decrease regional soil erosion, resulting in an increasing GNRP, whereas shorter alfalfa rotations (which require more tillage) will increase soil erosion, resulting in a slightly decreasing GNRP, 3) reduced electricity usage from groundwater pumping results in lowered GHG emissions which equates to an increasing GNRP for the region, and 4) land fallowing will result in a reduction in economic activi ty, which could lead to a reduced NRP. Taking these components together, the crop regime change scenario is expected to be a net positive for SLV regarding GNRP, however more careful analysis is needed to effectively capture and assess the changes. The Fi sher Information metric (Eason and Cabezas 2012) , which measures system order, relies on data from the other three metrics and is therefore expected to change accordingly. Nothing from the crop regime change scenario indicates that there would be a major shift in system order. All the preliminary quali tative descriptions above suggest that the crop regime change scenario will likely move the region, however slightly, in the direction of sustainability in regard to the four metrics proposed by EPA. However, this needs to be more carefully examined and co nfirmed through rigorous and quantitative research. Solar Energy Development Scenario Initial analysis of the solar energy development scenario reveals that increasing renewable energy production, which reduces demand for fossil fuel based electricity sour ces, may be the single largest driver in the scenario

PAGE 161

139 affecting the sustainability metrics. Though not all of the solar electricity produced will be immediately consumed in the region, the assumption is that SLV should receive credit for the displaced non renewable energy. Also, because the amount of land needed for solar development is relatively small, the change in land use is not expected to result in a shift in direction of sustainability. An outline of how the components of each of the EPA sustainab ility metrics can be expected to shift based on the scenario is discussed below. The implementation of the solar energy development scenario could result in the following changes to the Emergy Analysis: 1) A decrease in emergy imports of electricity due to an increase in local renewable electricity production, 2) a one time increase in emergy imports for the materials/services needed to build the solar facilitates and transmission infrastructure, and 3) changes in emergy imports (i.e., fertilizer, agricultu ral chemicals, electricity for irrigation , barley production etc.) due to land shifting from crop production to solar energy production . Because electricity has a large transfo r mity ( 170 ,000 sej/J), decreasing the reliance on electricity imports has the p otential to improve the regional Emergy Balance and move SLV toward sustainability. It is anticipated this change in electricity flow may outweigh any of the other shifts in emergy discussed above. Therefore, increasing solar energy production in the reg ion is likely positive from an emergy perspective. Similar to the crop regime change scenario, the primary way the solar energy development scenario is likely to affect the Ecological Footprint metric is

PAGE 162

140 by decreasing the "energy land" needed for SLV. Nee d for energy land , which is virtual land allocated for sequestering carbon from fossil fuel emissions, will be reduced because the region will be using/producing more renewable (non carbon based) electricity. The metric will also capture the shift of land from "arable land" and "pasture land" to "built up" land due to solar development, but this can be expected to have a minimal impact on the metric since it is a relatively small amount of land. However, the change in energy land has the potential to resu lt in a major reduction in the regions Ecological Footprint; energy land accounts for roughly half of SLV's total Ecological Footprint. Other changes to land use from the scenario are likely insignificant and not expected to shift the Ecological Balance o f the region. The GNRP of SLV can be expected to shift as a result of pursuing the solar energy development scenario in two ways: 1) changes in the economy from the investment in solar energy infrastructure and production (affecting NRP component), and 2) a reduction in regional GHG emissions due to carbon credits (see Chapter II) for renewable electricity production (affecting the depreciation of natural capital). More work needs to be done to quantify the magnitude of these changes, however, both can be expected to be positive for SLV in terms of GNRP. The Fisher Information metric, which measures system order, relies on data from the other three metrics and is therefore expected to change

PAGE 163

141 accordingly. Nothing from the solar energy development scenario indicates a major shift in system order. In summary, the solar energy development scenario lowers GHG emissions by decreasing electricity imports from outside sources, as well as increases production of clean renewable energy, all on a relatively small la nd area. As indicated above, these changes are likely to result in positive movement for all four of the sustainability metrics. Therefore, increasing solar energy production in the region is expected to be a net positive for sustainability in SLV , howeve r careful quantitative analysis is needed to confirm these predictions . Summary While these qualitative results are promising, more work needs to be done to develop methods to quantify the effects from the scenario modeling on the four sustainability metrics. Each metric will need to be carefully analyzed for primary as well as cascading affects resulting from the implementation of the scenarios. There is, however, currently no evidence that the future scenarios put forth by the CAB, which target wat er use reductions and GHG emissions reductions, would have a negative effect on the regions overall sustainability. In fact, the preliminary results indicate that both scenarios push the region in the direction of sustainability.

PAGE 164

142 Maintaining Relationships It would be beneficial to both entities if the University of Colorado Denver and the CAB were to maintain an ongoing relationship. The CAB will need ongoing support from the University if they are to take ownership of the metrics and indicator models de veloped over the course of this research. They will need available correspondence when collecting data and analyzing the results for future years. It will be beneficial to the University research team to keep communication open to ensure the tools develo ped are being used appropriately, as well as maintaining a relationship for collaboration on potential future research. The University should build on the immense effort put forth in establishing the CAB and implementing this research. Beyond SLV The res earch outlined here could also b e applied to other regions. EPA has already applied the four sustainability metrics developed in SLV to the island of Puerto Rico (U.S. EPA 2012b) . Puerto Rico made an ideal study location due to its size and isolated geography (island ), high dependence on local/coastal resources, and its long history of being a subject scientific investigation. That being said, any interested region could potentially be the subject of a sustainability assessment. It requires a desire from stakeholder s and decision makers to participate and their willingness to dedicate time and resources to the effort. A key asset in this work in the future will be public universities, which are institutions designed to serve the public interest (American Academy of Arts &

PAGE 165

143 Sciences 2015) . They are strat egically positioned to partner with communities and municipalities around issues of sustainability. Research has shown that the owners of a problem are in a unique position to help solve it (Checkland 1981) , so partnering local community stakeholders with unbiased researchers is the best approach for assessing regional sustainability , and for those findings to lead to the implementation of sustainable policies/ practices.

PAGE 166

144 REFERENCES Ali, M. 2013. Sustainability Assessment: Context of Resource and Environmental Policy . Amsterdam; Boston: Elsevier Academic Press. Altieri, M.A. 1999. The ecological role of biod iversity in agroecosystems. Agriculture, Ecosystems and Environemt 74: 19 Ð 31. Altieri, M.A. 2002. Agroecology!: The Science of Natural Resource Management for Poor Farmers in Marginal Environments. Agriculture, Ecosystems & Environment 93: 1 Ð 24. American A cademy of Arts & Sciences. 2015. Public Research Universities!: Why They Matter . The Lincoln Project: Excellence and Access in Public Higher Education . https://www.amacad.org/LincolnProject. Accessed October 10, 2016. Appasamy, P., S. Chopde, A. Dixit, D. Gyawali, S. Janakarajan, D.S. Kumar, R.M. Mathur, et al. 1999. Rethinking The Mosaic: Investigations into Local Water Management . Jaipur. Argonne National Laboratory. 2015. GREET Model. Argonne, IL: U.S. Department of Energy. https://greet.es.anl.gov/greet /index.htm. Arikan, Y., R. Desai, P. Bhatia, and W.K. Fong. 2012. Global Protocal for Community Scale Greenhouse Gas Emissions (GPC) . C40 Cities Climate Leadership Group and ICLEI Local Governments for Sustainability, Sao Paulo. Arnold, S., J. Dileo, and T. Takushi. 2014. Colorado Greenhouse Gas Inventory Ñ 2014 Update Including Projections to 2020 & 2030 . Denver, CO. https://www.colorado.gov/pacific/sites/default/files/AP COGHGInventory2014Update.pdf. Baral, A. and B.R. Bakshi. 2 010. Thermodynamic Metrics for Aggregation of Natural Resources in Life Cycle Analysis: Insight via Application to some Transportation Fuels. Environmental Science & Technology 44(2): 800 Ð 7. http://www.ncbi.nlm.nih.gov/pubmed/20020741. Barnett, J. 2003. Se curity and climate change. Globa Environmental Change 13: 7 Ð 17. Barrows, C., T. Mai, S. Haase, J. Melius, M. Mooney, C. Barrows, T. Mai, S. Haase, J. Melius, and M. Mooney. 2016. Renewable Energy Deployment in Colorado and the West!: A Modeling Sensitivity and GIS Analysis; NREL/TP 6A20 65350. Golden, CO. BEA. 2012. National, Regional, and State Economic Accounts. Bureau of Economic Analysis . http://www.bea.gov. Accessed January 5, 2015. Benson, R.D. 2012. Alive But Irreleveant: The Prior Appropriation Doct rine in Today's Western Water Law. University of Colorado Law Review (675): 1 Ð 41. Blankenbuehler, P. 2016. After years of drought and overuse, the San Luis Valley aquifer refills. High Country News . http://www.hcn.org/articles/after years of drought and ove ruse a water basin refills in the san luis valley. Accessed

PAGE 167

145 March 15, 2017. BLM. 1991. San Luis Resource Area: Proposed Resource Management Plan and Final Environmental Impact Statement . Bureau of Land Management, Canon City. Borda, F. and M.A. Orland Rahman. 1991. Action and Knowledge: Breaking the Monopoly with Participatory Action Research . New York: Technol.Publ/Apex. Brown, R.E. 1991. Community Action for Health Promotion: A Strategy to Empower Individuals and Communities. International Journal of Health Services 21(3): 441 Ð 456. Brugmann, J. 1996. Planning for Sustainability at the Local government Level. Environmental Impact Assessment Review 9255(96): 363 Ð 379. Brundtland, G.H. 1985. World commission on environment and dev elopment. Environmental Policy and Law 14(1): 26 Ð 30. Burch, S. 2010. Transforming Barriers into Enablers of Action on Climate Change!: Insights from Three Municipal Case Studies in British Columbia , Canada. Global Environmental Change 20(2): 287 Ð 297. http ://dx.doi.org/10.1016/j.gloenvcha.2009.11.009. Burkhardt, J.J., G. a Heath, and C.S. Turchi. 2011. Life Cycle Assessment of a Parabolic Trough Concentrating Solar Power Plant and the Impacts of Key Design Alternatives. Environmental Science & Technology 45 (6): 2457 Ð 2464. Cabezas, H. and B.D. Fath. 2002. Towards a Theory of Sustainable Systems. Fluid Phase Equilibria 194 Ð 197: 3 Ð 14. Cabezas, H. and A.T. Karunanithi. 2008. Fisher Information, Entropy, and the Second and Third Laws of Thermodynamics. Industrial & Engineering Chemistry Research47 47(15): 5243 Ð 5249. Calzadilla, A., K. Rehdanz, R. Betts, P. Falloon, A. Wiltshire, and R.S.J. Tol. 2013. Climate Change Impacts on Global Agriculture. Climatic Change 120: 357 Ð 374. Campbell, D.E. and A.S. Garmestani. 201 2. An Energy Systems View of Sustainability: Emergy Evaluation of the San Luis Basin, Colorado. Journal of Environmental Management 95(1): 72 Ð 97. http://www.ncbi.nlm.nih.gov/pubmed/22115513. Accessed March 4, 2013. CDOT. 2012. Online Transportation Informa tion System (OTIS). http://dtdapps.coloradodot.info/Otis/Statistics. Accessed January 2, 2015. CDPHE. 2015. Statewide Landfill Disposal Statistics. Hazardous Materials and Waste Management Division . Denver, CO. https://www.colorado.gov/pacific/cdphe/swrepo rts. Accessed November 15, 2015. CDWR. 2015a. Statement of Basis and Purpose. Ground Water Well Rules and Regulations . http://water.state.co.us/groundwater/GWAdmin/UseAndMeasurement/Pages /RioGrandeRBRules.aspx. Accessed November 15, 2015. CDWR. 2015b. Rio Grande Decision Supprt System. Denver, CO: Colorado

PAGE 168

146 Division of Water Resources. http://cdss.state.co.us/basins/Pages/RioGrande.aspx. CDWR. 2015c. Colorado Consumtive Use Model (StateCU). http://cdss.state.co.us/software/Pages/StateCU.aspx. Accessed Novemb er 15, 2015. CDWR. 2015d. Rules Governing the Withdrawal of Groundwater in Water Division No. 3 (The Rio Grande Basin) and Establishing Criteria for the Beginning and End of the Irrigation Season in Water Division No. 3 for All Irrigation Water Rights. Den ver, CO. http://water.state.co.us/DWRIPub/Documents/FINAL Groundwater Rules for Division 3 September 23 2015 2.pdf. CDWR. 2016a. Rio Grande Desicion Support System Datasets. http://water.state.co.us/SurfaceWater/RulemakingAndAdvising/SLVAC/Page s/SLVRespons eFunctions.aspx. Accessed November 15, 2015. CDWR. 2016b. Prior Appropriation Law. Water Rights . http://water.state.co.us/surfacewater/swrights/pages/priorapprop.aspx. Accessed October 15, 2016. CDWR. 2016c. Colorado Desicion Support System. http://cdss.state.co.us/Pages/CDSSHome.aspx. Accessed November 15, 2015. Chapagain, A.K. and A.Y. Hoekstra. 2004. Water Footprint of Nations Volume 1!: Main Report . Vol. 1. Netherlands. Checkland, P. 1981. Systems Thinking, Systems Practice . Chichester [Su ssex];New York; J. Wiley. Clark, W.C. and N.M. Dickson. 2003. Sustainability Science!: The Emerging Research Program. Proceedings of the National Academy of Sciences 100(14): 8059 Ð 8061. Colorado Geological Survey. 2016. Mountainous Region Aquifers. Ground Water Atlas . http://coloradogeologicalsurvey.org/water/groundwater atlas/. Accessed September 12, 2016. Costanza, R. 2014. Time to leave GDP behind. Nature 505: 283 Ð 205. Costanza, R., R. De Groot, S. Farber, M. Grasso, B. Hannon, K. Limburg, S. Naeem, et a l. 1998. The Value of the World ' s Ecosystem Services and Natural Capital. Ecological Economics 387: 253 Ð 260. Costanza, R., R. De Groot, P. Sutton, S. Van Der Ploeg, S.J. Anderson, I. Kubiszewski, S. Farber, and R.K. Turner. 2014. Changes in the Global Va lue of Ecosystem Services. Global Environmental Change 26: 152 Ð 158. http://dx.doi.org/10.1016/j.gloenvcha.2014.04.002. Costanza, R., I. Kubiszewski, D. Ervin, R. Bluffstone, J. Boyd, D. Brown, H. Chang, et al. 2011. Valuing Ecological Systems and Services. Biology Reports 6(July): 1 Ð 6. CSU. 2016. COMET Farm (TM) Greenhouse Gas Assessment System. www.comet farm.com. Accesse d July 29, 2016. Dagmar, L. and S. Vaddey. 2013. West Wide Climate Risk Assessment: Upper

PAGE 169

147 Rio Grande Impact Assessment. U.S. Department of the Interior; Bureau of Reclemation, Albuquerque. Dallinger, C. 2015. Individual Communication with Xcel Energy. Denv er, CO: Xcel Energy; Curtis.Dallinger@xcelenergy.com. Davis, S.J., K. Caldeira, W.C. Clark, S.J. Davis, and K. Caldeira. 2010. Consumption Based Accounting of C02 Emissions. Proceedings of the National Academy of Sciences 107(12): 5687 Ð 5692. Davis Engineer ing. 2016. Change in Unconfined Aquifer Storage. 1314 Eleveth Street Alamosa, CO: Comissioned by the Rio Grande Water Concervation District. Dillion, M. 2015. Expert Interview with Colorado State Extension in the San Luis Valley. Alamosa, CO. Dodman, D. 20 09. Blaming Cities for Climate Change? An Analysis of Urban Greenhouse Gas Emissions Inventories. Environment and Urbanization 21(1): 185 Ð 201. DOE/BLM. 2012. Final Programmatic Environmental Impact Statement ( PEIS ) for Solar Energy Development in Six Sou thwestern States: Chapter 10 Colorado Proposed Solar Energy Zones . Vol. 3. Bureau of Land Management & Department of Energy. http://solareis.anl.gov. Dong, H., Y. Geng, J. Sarkis, T. Fujita, T. Okadera, and B. Xue. 2013. Regional water footprint evaluation in China: a case of Liaoning. The Science of the Total Environment 442: 215 Ð 24. http://www.ncbi.nlm.nih.gov/pubmed/23178781. Accessed November 26, 2013. Douglas, E.M., P.H. Kirshen, M. Paolisso, C. Watson, J. Wiggin, A. Enrici, and M. Ruth. 2012. Coastal Flooding, Climate Change and Environmental Justice!: Identifying Obstacles and Incentives for Adaptation in two Metropolitan Boston Massachusetts communities. Mitigation Adaptation Strategies for Global Climate Change 17: 537 Ð 562. Dovers, S.R. 1995. A Fram ework for Scaling and Framing Policy Problems in Sustainability. Ecological Economics 12: 93 Ð 106. Doyle, R. 2015. Interview. Alamosa, CO: Alamosa County Planning. Dubinsky, J. and A.T. Karunanithi. 2017a. Greenhouse Gas Accounting of Rural Agrarian Regions : The Case of San Luis Valley. ACS Sustainable Chemistry and Engineering 5(1): 261 Ð 268. Dubinsky, J. and A.T. Karunanithi. 2017b. Consumptive Water Use Analysis of Upper Rio Grande Basin in Southern Colorado. Environmental Science & Technology 51(8): 4452 Ð 4460. Dumont, A., G. Salmoral, and M.R. Llamas. 2013. The Water Footprint of a River Basin with a Special Focus on Groundwater: The case of Guadalquivir Basin (Spain). Water Resources and Industry 1 Ð 2: 60 Ð 76. http://linkinghub.elsevier.com/retrieve/pii/S22 1237171300005X. Accessed March 3, 2014. Eason, T. and H. Cabezas. 2012. Evaluating the Sustainability of a Regional

PAGE 170

148 System sing Fisher information in the San Luis Basin, Colorado. Journal of Environmental Management 94(1): 41 Ð 9. http://www.ncbi.nlm.nih.gov/pubmed/21930337. Accessed March 5, 2013. Ecoinvent v2.0. 2007. Life Cycle Inventory of Propane Gas . Dubendorf. El Beltagy, A. and M. Madkour. 2012. Impact of climate change on arid lands agriculture. Agriculture an d Food Security . BioMed Central Ltd. England, R.W. 1998. Measurement of Social Well Being: Alternatives to Gross Domestic Product. Ecological Economics . Elsevier B.V. Entz, Chlouber, Grossman, Philips, Taylor, and Teek. 2004. SENATE BILL 04 222 . Alamosa, C O. http://water.state.co.us/DWRIPub/San Luis Valley Advisory Committee/Tab C Senate Bill 04 222.pdf. ESRI. 2013. ArcGIS Desktop: Release 10.2. Redlands, CA: Environmental Systems Research Insitutute. Eve, M., D. Pape, M. Fligge, R. Steele, D. Man, M. Ril ey Gilbert, and S. Biggar. 2014. Quantifying Greenhouse Gas Fluxes in Agriculture and Forestry: Methods for Entity Scale Inventory. Technical Bulletin Number 1939. Office of the Cheif Economist. Washington D.C. Fageria, N.K. 2007. Green Manuring in Crop Pr oduction. Journal of Plant Nutrition (30): 691 Ð 719. Falkenmark, M. 1991. Regional Environmental Management: The Role of Man Water Land Interactions . World Bank, Environment and Development, Policy and Research Dvision, Washington D.C. Fiksel, J., T. Eason, and H. Frederickson. 2012. A Framework for Sustainability Indicators at EPA . EPA Office of Research and Development. Food and Agriculture Organization of the United Nations (FAO). 1993. Land and Water Integration and River Basin Management. In Proceedings of an FAO Informal Workshop, Land and Water Bulletin . Fraser, E.D.G., A.J. Dougill, W.E. Mabee, M. Reed, and P. McAlpine. 2006. Bottom up and top down: analysis of participatory processes for sustainability indicator identification as a pathway to communit y empowerment and sustainable environmental management. Journal of Environmental Management 78(2): 114 Ð 27. Freeman, R.E. 1984. Strategic Management: A Stakeholder Approach . New York. Freire, P. 1982. Creating Alternative Research Methods: Learning to Do it by Doing it. Creating Knowledge: a monopoly? Participatory research in Development. New Delhi. Frischknecht, R. and G. Rebitzer. 2005. The Ecoinvent Database System: A Comprehensive Web Based LCA Database. Journal of Cleaner Production 13(13 Ð 14): 1337 Ð 134 3. http://linkinghub.elsevier.com/retrieve/pii/S0959652605001253. Accessed July 14, 2014. Fryar, A.E. 2008. When the Rivers Run Dry: Water -The Defining Crisis of the Twenty First Century. Environmental and Engineering Geoscience 14(1):

PAGE 171

149 53 Ð 54. Fulton, J. a nd H. Cooley. 2015. The Water Footprint of California's Energy System, 1990 2012. Environmental Science & Technology 46(6): 2 Ð 18. Gallucci, M. 2013. 6 of the World's Most Extensive Climate Adaptation Plans. Inside Climate News . http://insideclimatenews.org /news/20130620/6 worlds most extensive climate adaptation plans. Gerbens Leenes, P.W., M.M. Mekonnen, and A.Y. Hoekstra. 2013. The Water Footprint of Poultry, Pork and Beef: A Comparative Study in Different Countries and Production Systems. Water Resources and Industry 1(2): 25 Ð 36. http://linkinghub.elsevier.com/retrieve/pii/S2212371713000024. Accessed January 21, 2014. Gibson, M. and et.al. 2015. Rio Grande Basin Implementation Plan . Alamosa, CO. www.riograndewaterplan.com. Gosling, S.N. and N.W. Arnell. 2 016. A Global Assessment of the Impact of Climate Change on Water Scarcity. Climatic Change 134: 371 Ð 385. Graymore, M.L.M., N.G. Sipe, and R.E. Rickson. 2010. Sustaining Human Carrying Capacity: A tool for regional sustainability assessment. Ecological Eco nomics 69(3): 459 Ð 468. http://linkinghub.elsevier.com/retrieve/pii/S0921800909003255. Accessed August 14, 2013. Greenberg, K., K. Kropp, D. Weingarten, and S. Wentzel Fisher. 2014. Sustaining Farming in the Arid West!: Stories of Young Farmers, Water and R esilience. National Young Farmers Coalition : 12 Ð 13. http://www.youngfarmers.org. Grosso, S.J. Del, W.J. Parton, A.R. Mosier, M.D. HArtman, D.S. Brenner, and D.S. Schimel. 2001. Simulated Interaction of Carbon Dynamics and Nitrogen Trace Gas Fluxes Using th e DAYCENT Model. In Modeling Cabon and Nitrogen Dynamics for Soil Management , 303 Ð 332. New York: Lewis Publishers. Grosso, S.J. Del, W.J. Parton, A.R. Mosier, M.K. Walsh, D.S. Ojima, and P.E. Thornton. 2006. DAYCENT National Scale Simulations of Nitrous Oxide Emissions from Cropped Soils in the United States. Journal of Environmental Quality 35: 1451 Ð 1460. Gunderson, L .H. and C.S. Holling. 1996. Barriers and Bridges to the Renewal of Ecosystems and Institutions . Ed. by S.S. Light. Environments . Vol. 23. Wilfrid Laurier University Environments. Hamilton, K., G. Atkinson, and D. Pearce. 1997. Genuine Savings as an Indic ator of Sustainability . World Bank. Heath, G. a., P. O'Donoughue, D.J. Arent, and M. Bazilian. 2014. Harmonization of Initial Estimates of Shale Gas Life Cycle Greenhouse Gas Emissions for Electric Power Generation. Proceedings of the National Academy of S ciences of the United States of America Early Acce: E3167 Ð E3176. http://www.ncbi.nlm.nih.gov/pubmed/25049378. Heberling, M.T. and M.E. Hopton. 2012. Introduction to the Special Collection of

PAGE 172

150 papers on the San Luis Basin Sustainability Metrics Project: A Me thodology for Evaluating Regional Sustainability. Journal of Environmental Management 111: 272 Ð 8. http://www.ncbi.nlm.nih.gov/pubmed/22560056. Accessed March 5, 2013. Heberling, M.T., J.J. Templeton, and S. Wu. 2012. Green Net Regional Product for the San Luis Basin, Colorado: An Economic Measure of Regional Sustainability. Journal of Environmental Management 111: 287 Ð 97. http://www.ncbi.nlm.nih.gov/pubmed/22483369. Accessed August 21, 2013. Hillman, T. and A. Ramaswami. 2010. Greenhouse Gas Emissions Footp rints and Energy Use Benchmarks for Eight U.S. Cities. Environmental Science & Technology 44(6): 1902 Ð 1910. Hinderlinder, M.C., T.M. McClure, and F.B. Clayton. 1938. Rio Grande River Compact. Santa Fe. http://cwcb.state.co.us/legal/Pages/InterstateCompacts .aspx. History Colorado. 2015. This week in Colorado History Colorado's oldest town, San Luis, is Established. http://historycolorado.org/blogs/hc/2015/04/08/this week in colorado history colorados oldest town san luis is established/. Hoekstra, A.Y., A.K. Chapagain, M.M. Aldaya, and M.M. Mekonnen. 2011. The Water Footprint Assessment Manual . Washington DC: earthscan. Hoekstra, A.Y. and P.Q. Hung. 2002. Virtual Water Trade; A Quantification of Virtual Water Flows Between Nations in Relation to International Trade . Research Report Series . Netherlands. Hopton, M.E., H. Cabezas, D. Campbell, T. Eason, A.S. Garmestani, M.T. Heberling, A.T. Karunanithi, J.J. Templeton, D. White, and M. Zanowick. 2010. Development of a Multidisciplinary Approach to Assess Regional Sustainability. International Journal of Sustainable Development & World Ecology 17(1): 48 Ð 56. Hopton, M.E. and D. White. 2012. A Simplified Ecological Footprint at a Regional Scale. Journal of Environmental Management 111: 279 Ð 86. http://www.ncbi.nlm.nih.gov/pubmed/22033065. Accessed March 5, 2013. Howarth, R.W., R. Santoro, and A. Ingraffea. 2011. Methane and the Greenhouse Gas Footprint of Natural Gas from Shale Formations. Climatic Change 106(4): 679 Ð 690. Hsu, D., P . O'Donoughue, V. Fthenakis, G. Heath, H. Kim, P. Sawyer, J. Choi, and D. Turney. 2012. Life Cycle Greenhouse Gas Emissions of Crystalline Silicon Photovoltaic Electricity Generation: Systematic Review and Harmonization. Journal of Industrial Ecology 16(S1 ): S112 Ð S135. Hurd, B.H. and J. Coonrod. 2007. Climate Change and its Implications for New Mexico's Water Resources and Economic Opportunities . Department of Agricultural Economics and Agricultural Business. Las Cruces, NM. Hussey, K. and J. Pittock. 2012. The Energy Water Nexus: Managing the Links between Energy and Water for a Sustainable Future. Ecology and Society 17(1): 1 Ð 8. http://hdl.handle.net/10535/8207. IMPLAN. 2012. Value Added Data for the 5 counties of San Luis Valley.

PAGE 173

151 Huntersville, NC: © Copyr ight 2015 IMPLAN Group LLC. IPCC. 2006. Guidlines for National Greenhouse Gas Inventories . Prepared by the Intergovernmental Panel on Climate Change's National Greenhouse Gas Inventories Programme. Hayama, Japan. http://www.ipcc nggip.iges.or.jp/public/200 6gl/. IPCC. 2013. Climate Change 2013: The Physical Science Basis. 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. IPCC. 2014. Climate Change 2014 Synthesis Report: Summary for Policymakers . IPCC Working Group I. 1990. Policymakers Summary . Ed. by G.J. Jenkins and J.J. Ephraums (eds.) J.T. Houghton. Cambridge, Great Britain, New York, NY, USA and Melbourne, Australia. Israel, B.A., A.J. Schulz, E.A. Parker, and A.B. Becker. 1998. Review of Community Based Research: Assessing Partnership Approaches to Improve Public Health. Annual Review of Public Health 19: 173 Ð 202. Ives, C. and H. Boatwright. 1999. Emerson. In Essays Before a Sonata: The Majority and Other Writings , 11 Ð 14. W. W. Norton & Company. Iwaniec, D. and A. Wiek. 2014. Advancing Sustainability Visioning Practice in Planning Ñ The General Plan Update in Phoenix , Arizona Advancing Sustainability Visioning Practi ce in Planning Ñ The General Plan Update in Phoenix , Arizona. Planning, Practice & Research 29(5): 543 Ð 568. Janou, S. 2012. How to Understand and Measure Environmental Sustainability: Indicators and Targets. Ecological Indicators 17: 4 Ð 13. Jorge, R.S., T .R. Hawkins, and E.G. Hertwich. 2012a. Life Cycle Assessment of Electricity Transmission and Distribution Ñ part 1!: Power Lines and Cables. International Journal of Life Cycle Assessment 17(9): 9 Ð 15. Jorge, R.S., T.R. Hawkins, and E.G. Hertwich. 2012b. Li fe Cycle Assessment of Electricity Transmission and Distribution Ñ part 2!: Transformers and Substation Equipment. International Journal of Life Cycle Assessment 17: 184 Ð 191. Kahn, J.R. 1995. The Economic Approach to Environmental and Natural Resources . Oa k Brook, IL: Dryden Pr. Kang, E., L. Lu, and Z. Xu. 2007. Vegetation and Carbon Sequestration and their Relation to Water Resource management in an Inland River Basin of Northwest China. Environemtal Management (85): 702 Ð 710. Kates, R.W., W.C. Clark, R. Cor ell, J.M. Hall, C.C. Jaeger, I. Lowe, J.J. McCarthy, et al. 2001. Sustainability Science. Science 292(5517): 641 LP 642. http://science.sciencemag.org/content/292/5517/641.abstract. Kates, R.W. and T.M. Parris. 2003. Long Term Trends and a Sustainability T ransition. Proceedings of the National Academy of Sciences 100(14). Kennedy, C., J. Steinberger, B. Gasson, Y. Hansen, T. Hillman, M. Havr‡nek, D. Pataki, A. Phdungsilp, A. Ramaswami, and G.V. Mendez. 2010.

PAGE 174

152 Methodology for Inventorying Greenhouse Gas Emiss ions from Global Cities. Energy Policy 38(9): 4828 Ð 4837. http://linkinghub.elsevier.com/retrieve/pii/S0301421509006387. Accessed November 18, 2013. Kim, H., V. Fthenakis, J. Choi, and D. Turney. 2012. Life Cycle Greenhouse Gas Emissions of Thin film Photov oltaic Electricity Generation: Systematic Review and Harmonization. Journal of Industrial Ecology 16(S1): S110 Ð S121. Kishita, Y., K. Hara, M. Uwasu, and Y. Umeda. 2016. Research Needs and Challenges Faced in Supporting Scenario Design in Sustainability Sci ence!: A Literature Review. Sustainability Science 11(2): 331 Ð 347. Komiyama, H. and K. Takeuchi. 2006. Sustainability Science!: Building a New Discipline. Sustainability Science 1(August 2005): 1 Ð 6. Kuzdas, C., B.P. Warner, A. Wiek, R. Vignola, M. Yglesias , and D.L. Childers. 2016. Sustainability Assessment of Water Governance Alternatives!: The case of Guanacaste Costa Rica. Sustainability Science 11(2): 231 Ð 247. Kuzdas, C. and A. Wiek. 2014. Governance Scenarios for Addressing Water Conflicts and Climate Change Impacts. Environmental Science and Policy 42: 181 Ð 196. http://dx.doi.org/10.1016/j.envsci.2014.06.007. Larsen, H.N. and E.G. Hertwich. 2009. The Case for Consumption Based Accounting of Greenhouse Gas Emissions to Promote Local Climate Action. Environmental Science and Policy 12: 791 Ð 798. Lawn, P.A. 2003. A theoretical foundation to support the Index of Sustainable Economic Welfare (ISEW), Genuine Progress Indicator (GPI), and other related indexes. Ecological Economics . Elsevier B.V. Lenzen, M. and R. Crawford. 2009. The path exchange method for hybrid LCA. Environmental Science and Technology 43(21): 8251 Ð 8256. Liu, J., a. J.B. Zehnder, and H. Yang. 2009. Global consumptive water use for crop production: The importance of green water and virtu al water. Water Resources Research 45(5): 1 Ð 15. Lovins, L.H. and P. Hawken. 2007. A Road Map for Natural Capitalism Key ideas from the Harvard Business Review article By Amory B . Harvard Business Review : 1 Ð 4. Malcolm, S., E. Marshall, M. Aillery, P. Heise y, M. Livingston, and K. Day rubenstein. 2012. Agricultural Adaptation to a Changing Climate. Economic and Environmental Impications Vary by U.S. Region . Marques, J.C., A. Basset, T. Brey, and M. Elliott. 2009. The Ecological Sustainability Trigon Ð A Proposed Conceptual Framework for Creating and Testing Management Scenarios. Marine Pollution Bulletin 58(12): 1773 Ð 1779. http://dx.doi.org/10.1016/j.marpolbu l.2009.08.020. Mayer, A.L. 2008. Strengths and Weaknesses of Common Sustainability Indices for Multidimensional Systems. Environment International 34(2): 277 Ð 91. http://www.ncbi.nlm.nih.gov/pubmed/17949813. Accessed August 27, 2014. McCool, S.F. and G.H. S tankey. 2004. Indicators of Sustainability!: Challenges

PAGE 175

153 and Opportunities at the Interface of Science and Policy. Environemtal Management 33(3): 294 Ð 305. McKinlay, J.B. 1993. The Promotion of Health Through Planned Sociopolitical Change: Challenges for Res earch and Policy. Social Science & Medicine 36(2): 109 Ð 117. http://www.sciencedirect.com/science/article/pii/027795369390202F. Mekonnen, M.M. and A.Y. Hoekstra. 2011a. National Water Footprint Accounts; The Green, Blue, and Grey Water Footprint of Producti on and Consumption . Vol. 1. The Netherlands. Mekonnen, M.M. and A.Y. Hoekstra. 2011b. National Water Footprint Accounts. Institute for Water Education 1(50): 1 Ð 50. Mekonnen, M.M. and A.Y. Hoekstra. 2012. A Global Assessment of the Water Footprint of Farm A nimal Products. Ecosystems 15(3): 401 Ð 415. http://www.springerlink.com/index/10.1007/s10021 011 9517 8. Accessed February 25, 2013. Meldrum, J., G. Heath, and J. Macknick. 2013. Life Cycle Water use for Electricity Generation!: A Review and Harmonization o f Literature Estimates. Environmental Research Letters 8. Metherell, A.K., L.A. Harding, V.C. Cole, and W.J. Parton. 1993. CENTURY: Soil Organic Matter Model Environment Technical Documentation. GPSR Technical Report (# 4): 1 Ð 70. Mintz, M. 2014. The Biotic Farmer. The Furrow (December): 22 Ð 24. JohnDeere.com/Furrow. Mistry, J., C. Tschirhart, C. Verwer, R. Glastra, J. Mistry, O. Davis, D. Jafferally, et al. 2014. ScienceDirect Our common future!? Cross scalar scenario analysis for social Ð ecological sustainab ility of the Guiana. Environmental Science and Policy 44: 126 Ð 148. Mitsch, W.J. and J.G. Gosselink. 1993. Wetlands . Second Edi. New York: Van Nostrand Reinhold. Mori, K. and A. Christodoulou. 2012. Review of Sustainability Indices and Indicators: Towards a New City Sustainability Index (CSI). Environmental Impact Assessment Review 32(1): 94 Ð 106. http://www.sciencedirect.com/science/article/pii/S0195925511000758. Morse, S. 2015. Developing Sustainability Indicators and Indices. Sustainable Development 23(2): 84 Ð 95. Mota, R.P., T. Domingos, and V. Martins. 2010. Analysis of Genuine Saving and Potential Green Net National Income: Portugal, 1990 2005. Ecological Economics 69(10): 1934 Ð 1942. http://dx.doi.org/10.1016/j.ecolecon.2010.04.026. Nakicenovic, N., O. Da vidson, G. Davis, A. Grubler, T. Kram, E.L. La Rovere, B. Metz, et al. 2000. Summary for Policymakers: Emissions Scenarios . https://www.ipcc.ch/pdf/special reports/spm/sres en.pdf. Ness, B., E. Urbel Piirsalu, S. Anderberg, and L. Olsson. 2007. Categorisin g tools for sustainability assessment. Ecological Economics 60(3): 498 Ð 508.

PAGE 176

154 NFS. 2015. Rio Grande National Forest Ð Assessment 8 Multiple Uses . Forestwide Planning Assessment . National Forest Service, Creede CO. Niemuth, N.D., M.A. Bozek, and N.F. Payne. 2 004. Chapter 8: Managment of Natural Palustrine Wetlands. In Wetland and Riparian Areas of the Intermountain West: Ecology and Managment , ed. by Mark C. McKinstry, Wayne A. Hubert, and Stanley H. Anderson. Austin: University of Texas Press. Obama, B. 2009. Federal Leadership in Environmental, Energy, and Economic Performance. U.S.A. https://www.fedcenter.gov/programs/eo13514/. Obama, B. 2013. The President's Climate Action Plan. Washington DC. https://www.whitehouse.gov/sites/default/files/image/president27sclimateacti onplan.pdf. Odum, H.T. 1971. Environment, Power, and ociety . New York: Wiley Interscience. Odum, H.T. and E.C. Odum. 2001. A Prosperous Way Down: Principles and Policies . Boulder, Colorado: University Press of Colorado. Oniel, P. 2015. Expert Interview with San Luis Valley farm consultants Agro Engineering. Alamosa, CO. Page, K. 2016. Colorado State Land Board Database Query. Alamosa, CO. Paustian, K., J. Six, E.T. Elliott, and H.W. Hunt. 2000. Management Options for Reducing CO2 Emissions from Agricultural Soils. Biogeochemistry 48: 147 Ð 163. Peters, G.P. 2008. From Production Based to Consumption Based National Emission Inventories. Ecological Economics 5(1). Petrow, S., P. Franks, and T.R. Wolfred. 1990. Ending the HIV Epidemic: Community Strategies in Disease Prevention and Health Promotion . Santa Cruz, CA: Network Publications. Pezzey, J.C.V., N. Hanley, K. Turner, and D. Tinch. 2006. Comparing Augmented Sustainability Measures fo r Scotland: Is There a Mismatch? Ecological Economics 57(1): 60 Ð 74. http://linkinghub.elsevier.com/retrieve/pii/S0921800905001242. Accessed November 26, 2013. Poppleton, J. 2013. The Resilient Rio Grande Basin. Headwaters . http://issuu.com/cfwe/docs/hw_31_ finalweb. Postma, T.J.B.M. and F. Liebl. 2005. How to Improve Scenario Analysis as a Strategic Management Tool? Technological Forecasting & Social Change 72: 161 Ð 173. Powlson, D.S., A.P. Whitmore, and K.W.T. Goulding. 2011. Soil Carbon Sequestration to Mit igate Climate Change!: A Critical Re Examination to Identify the True and the False. European Journal of Soil Science 62(February): 42 Ð 55. PRISM Climate Group. 2016. PRISM Gridded Climate Data. Oregan State University. http://prism.oregonstate.edu. Accesse d October 10, 2016. Raimbault, B.A. and T.J. Vyn. 1991. Crop Rotation and Tillage Effects on Corn

PAGE 177

155 Growth and Soil Structural Stability. Agronomy Journal 83(6): 979 Ð 985. Ramaswami, A., T. Hillman, B. Janson, M. Reiner, and G. Thomas. 2008. Policy Analysis A Demand Centered , Hybrid Life Cycle Methodology for City Scale Greenhouse Gas Inventories. Environmental Science & Technology 42(17): 6455 Ð 6461. Ramaswami, A., D. Main, M. Bernard, A. Chavez, A. Davis, G. Thomas, K. Schnoor, et al. 2014. Planning for Low Carbon Communities in U.S. cities!: A Participatory Process Model between Academic Institutions, Local Governments and Communities in Colorado. Carbon Managment 2(4): 397 Ð 411. Ramaswami, A., C. Weible, D. Main, T. Heikkila, S. Siddiki, A. Duvall, A. Pattis on, and M. Bernard. 2012. A Social Ecological Infrastructural Systems Framework for Interdisciplinary Study of Sustainable City Systems: An Integrative Curriculum Across Seven Major Disciplines. Journal of Industrial Ecology 16(6): 801 Ð 813. Rametsteiner, E ., H. PŸlzl, J. Alkan Olsson, and P. Frederiksen. 2011. Sustainability Indicator Development Ñ Science or Political Negotiation? Ecological Indicators 11(1): 61 Ð 70. http://linkinghub.elsevier.com/retrieve/pii/S1470160X09001046. Accessed August 4, 2014. Ramos , T.B. 2009. Development of Regional Sustainability Indicators and the Role of Academia in this Process: The Portuguese Practice. Journal of Cleaner Production 17(12): 1101 Ð 1115. http://linkinghub.elsevier.com/retrieve/pii/S0959652609000717. Accessed Augus t 18, 2014. Ramos, T.B. 2010. A Framework for Regional Sustaiability Assessment: Developing Indicators for a Portuguese Region. Sustainable Development (18): 211 Ð 219. Reed, M.S. 2008. Stakeholder participation for environmental management: A literature revi ew. Biological Conservation 141(10): 2417 Ð 2431. http://linkinghub.elsevier.com/retrieve/pii/S0006320708002693. Accessed July 9, 2014. Reed, M.S., E.D.G. Fraser, and A.J. Dougill. 2006. An Adaptive Learning Process for Developing and Applying Sustainability Indicators with Local Communities. Ecological Economics 59(4): 406 Ð 418. http://linkinghub.elsevier.com/retrieve/pii/S0921800905005161. Accessed July 24, 2014. Reed, M.S., A. Graves, N. Dandy, H. Posthumus, K. Hubacek, J. Morris, C. Prell, C.H. Quinn, and L.C. Stringer. 2009. Who's In and Why!? A Typology of Stakeholder Analysis Methods for Natural Resource Management. Journal of Environmental Management 90: 1933 Ð 1949. Register, F. 2009. Mandatory Reporting of Greenhouse Gases; Final Rule. Tables C 1 and C 2 at FR. http://www.epa.gov/ghgreporting/documents/pdf/2013/documents/memo

PAGE 178

156 2013 technical revisions.pdf. Reicosky, D.C. 1997. Tillage Induced CO2 Emission from Soil. Nutrient Cycling in Agroecosystems 49: 273 Ð 285. Reynolds, M. 2015. Expert Interview with C olorado State Extenision in the San Luis Valley. Alamosa, CO. Ridoutt, B.G. and S. Pfister. 2010. Reducing Humanity's Water Footprint. Environmental Science and Technology 44(16): 6019 Ð 6021. Rockey, B. 2014. Managing for Soil Health when Raising Potatoes. Center, CO: United States Department of Agriculture. http://www.conservationwebinars.net/webinars/managing for soil health when raising potatoes. Rockey, S. 2016. Co Owner of Rockey Farm; Personal Interview. Center, CO. Rothausen, S.G.S.A. and D. Conway. 2 011. Greenhouse Gas Emissions from Energy Use in the Water Sector. Nature Clim. Change 1(4): 210 Ð 219. http://dx.doi.org/10.1038/nclimate1147. Rowe, G. and L.J. Frewer. 2000. Public Participation Methods!: A Framework for Evaluation. Science, Technology, & Human Values 25(1): 3 Ð 29. Santayana, G. 1905. The Life of Reason: Volume 1 Reason in Common Sense . Scully, T. 2014. Thriving on Healthy Soil Rockey Farms. Acres 43(12): 1 Ð 4. http://www.acresusa.com/magazine. Shaw, A., S. Sheppard, S. Burch, D. Flanders, A. Wiek, J. Carmichael, J. Robinson, and S. Cohen. 2009. Making Local Futures Tangible Ñ Synthesizing, Downscaling, and Visualizing Climate Change Scenarios for Participatory Capacity Building. Global Environmental Change 19: 447 Ð 463. Sheppard, S.R.J., A. Sh aw, D. Flanders, S. Burch, A. Wiek, J. Carmichael, J. Robinson, and S. Cohen. 2011. Future Visioning of Local Climate Change: A Framework for Community Engagement and Planning with Scenarios and Visualisation. Futures 43(4): 400 Ð 412. http://dx.doi.org/10.1 016/j.futures.2011.01.009. SLV Development Resources Group. 2013. Comprehensive Economic Development Strategy (CEDS). http://www.slvdrg.org/2013ceds.php. Accessed January 10, 2015. SLVEC. 2010. SLV Solar/Transmission line Alternatives and Redundancy Recomm endations compiled by: The San Luis Valley Solar/Transmission Work Group in cooperation with the San Luis Valley Ecosystem Council and Citizens for San Luis Valley Water Protection Coalition. www.slvec.org. Smil, V. 2010. Energy Transitions: History, Requi rments, Prospects . Santa Barbara, CA: Praeger. Soil Survey Staff. 2016. Web Soil Survey. Natural Resources Conservation Service, United States Department of Agriculture . Natural Resources Conservation Service, United States Department of Agriculture. https ://websoilsurvey.sc.egov.usda.gov/. Accessed March 4, 2016. Spangenberg, J.H. 2011. Sustainability Science: A Review, an Analysis and some Empirical Lessons. Environmental Concervation 38(3): 275 Ð 287.

PAGE 179

157 Sparks, R. 2015. Expert Interview with NRCS in the San Luis Valley. Alamosa, CO. Spath, P.L. and M.K. Mann. 2000. Life Cycle Assessment of a Natural Gas Combined Cycle Power Generation System . National Renewable Energy Lab. Golden, CO. Spath, P.L., M.K. Mann, and D.R. Kerr. 1999. Life Cycle Assessment of Coa l fired Power Production . Golden, CO, September 1. http://www.osti.gov/servlets/purl/12100 SypAiE/webviewable/. Stallworthy, M. 2017. Environmental Justice Imperatives for an Era of Climate Change. Journal of Law and Society 36(1): 55 Ð 74. Sullivan, K. 2015 . BLM Finalizes Environmental Assessment for Xcel Powerline Project over Poncha Pass. Bureau of Land Management. Swart, R.J., P. Raskin, and J. Robinson. 2004. The Problem of the Future: Sustainability Science and Scenario Analysis. Global Environmental Ch ange 14: 137 Ð 146. Tasser, E., E. Sternbach, and U. Tappeiner. 2008. Biodiversity Indicators for Sustainability Monitoring at Municipality Level!: An Example of Implementation in an Alpine Region. Ecological Indicators 8: 204 Ð 223. Thier, D. 2010. No Quinoa: The Story of a Cursed Crop. The Atlantic . https://www.theatlantic.com/health/archive/2010/01/quinoa the story of a cursed crop/33638/. Tilley, D.R. 2004. Howard T. Odum's Contribution to the Laws of Energy. Ecological Modelling 178(1 Ð 2): 121 Ð 125. http://l inkinghub.elsevier.com/retrieve/pii/S0304380003005519. Accessed November 26, 2013. Tri State. 2015. Baseload Information from their Website. Tri State Generation and Transmission . http://www.tristategt.org/AboutUs/baseload resources.cfm. Trochim, W.M.K. an d D. Cabrera. 2005. The Complexity of Concept Mapping for Policy Analysis. Policy Analysis and Management 7(1): 11 Ð 22. Trutnevyte, E., M. Stauffacher, and R.W. Scholz. 2011. Supporting Energy Initiatives in Small Communities by Linking Visions with Energy Scenarios and Multi Criteria Assessment. Energy Policy 39(12): 7884 Ð 7895. http://dx.doi.org/10.1016/j.enpol.2011.09.038. U.S. Census Bureau. 2011. Selected Economic Characteristics 2006 2010 American Community Survey 5 Year Estimates. http://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml? pid=ACS_10_5YR_DP03&prodType=table. U.S. Census Bureau. 2012. Quick Stats. U.S. Census Bureau. http://www.census.gov. Accessed January 12, 2014. U.S. Department of Agriculture. 20 15. Quick Stats Tool. National Agricultural Statistics Service. http://www.nass.usda.gov/Quick_Stats/index.php. Accessed January 2, 2015. U.S. DOT. 2012. National Transportation Statistics. U.S. Department of

PAGE 180

158 Transportation . http://www.rita.dot.gov/bts/sit es/rita.dot.gov.bts/files/publications/national_tra nsportation_statistics/index.html. Accessed January 2, 2015. U.S. DOT. 2015. 1995 2012 Highway Statistics. Washington DC. www.fhwa.dot.gov/policyinformation/statistics.cfm. Accessed January 1, 2015. U.S. E IA. 2015. Energy Information Administration Electricity Data Browser. Energy Information Administration . http://www.eia.gov/electricity/data/browser/. U.S. EIA. 2016. Diesel Fuel Prices. U.S. Energy Information Administration. http://www.eia.gov/petroleum/ data.cfm#summary. Accessed May 15, 2016. U.S. EPA. 2010. San Luis Basin sustainability metrics project: a methodology for evaluating regional sustainability. Heberling, M.T. and M.E. Hopton, eds. Cincinnati, OH. U.S. EPA. 2012a. eGRID2012 Version 1.0. Envi ronmental Protection Agency . http://www.epa.gov/cleanenergy/energy resources/egrid/. U.S. EPA. 2012b. Puerto Rico Sustainable Communities Research Project. In Annual U.S. EPA, ORD, Caribbean Environmental Protection Division (CEPD) Meeting . Office of Resea rch and Development, National Risk Management Research Laboratory, Sustainable Technology Division. https://cfpub.epa.gov/si/si_public_record_Report.cfm?dirEntryID=246424. U.S. EPA. 2013a. Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990 Ð 2013 (Ap ril 2015) . Vol. 4. Washington DC. http://www.epa.gov/climatechange. U.S. EPA. 2013b. Municipal Solid Waste Facts and Figures Full Report. Environmental Protection Agency. http://www.epa.gov/solidwaste/nonhaz/municipal/msw99.htm. Accessed January 1, 2014. U .S. EPA. 2015a. Local Climate and Energy Program. Environmental Protection Agency . http://www.epa.gov/statelocalclimate/local/index.html. U.S. EPA. 2015b. WARM Model. Environmental Protection Agency. http://epa.gov/epawaste/conserve/tools/warm/index.html. Accessed January 1, 2014. United Nations. 2002. Report of the World Summit on Sustainable Development . Johannesburg. United Nations Division for Sustainable Development. 1992. United Nations Conference on Environment & Development. Rio de Janerio , Brazil. http://www.un.org/esa/sustdev/agenda21.htm. Accessed October 20, 2016. Wackernagel, M. and W. Rees. 1996. Our Ecological Footprint: Reducing Human Impact on the Earth . Gabriola Island, BC: New Society Publishers. Walker, B., Rian, C.S. Holling, S.R. Carpe nter, and A. Kinzig. 2004. Resilience, Adaptability and Transformability in Social Ð Ecological Systems. Ecology and Society 9(2). Waudby, J. 2015. Interview with CFO of Rural Electric of the San Luis Valley.

PAGE 181

159 Monte Vista CO. Whitaker, M., G. a. Heath, P. O'D onoughue, and M. Vorum. 2012. Life Cycle Greenhouse Gas Emissions of Coal Fired Electricity Generation: Systematic Review and Harmonization. Journal of Industrial Ecology 16(SUPPL.1). Whitten, G. 2015. Expert Interview with a Local Cattle Rancher. Saguache , CO. Whitten, J.E. and S.E. Reynolds. 1963. Amended Costilla Creek Compact. Santa Fe. http://cwcb.state.co.us/legal/Pages/InterstateCompacts.aspx. Whyte, W. 1991. Participatory Action Research . Newbury Park, CA: Sage Publictions. Wiedmann, T. 2009. A revi ew of Recent Multi Region Input Output Models used for Consumption Based Emission and Resource Accounting. Ecological Economics 69(2): 211 Ð 222. http://dx.doi.org/10.1016/j.ecolecon.2009.08.026. Wilmsen, C. 2008. Partnerships for Empowerment: Participatory Research for Community Based Natural Resource Management . Sterling, VA; London: Earthscan. WRI. 2004. The Greenhouse Gas Protocol. World Resources Institute and World Council for Sustatinable Development. Xcel. 2013. CO2 Emissions Reporting. Xcel Energy . https://www.xcelenergy.com/staticfiles/xe/Corporate/CRR2012/environment/ emission reduction/carbon dioxide.html. Zeist, W.J. Van, M. Marinussen, R. Broekema, E. Groen, A. Kool, M. Dolman, and H. Blonk. 2012. LCI data for the Calculation Tool Feedprint for G reenhouse Gas Emissions of Feed Production and Utilization . Blonk Consultants, Gravin Beatrixstraat 34 2805 PJ Gouda the Netherlands. www.blonkconsultants.nl. Zeng, Z., J. Liu, P.H. Koeneman, E. Zarate, and a. Y. Hoekstra. 2012. Assessing Water Footprint at River Basin Level: A Case Study for the Heihe River Basin in Northwest China. Hydrology and Earth System Sciences 16(8): 2771 Ð 2781. http://www.hydrol earth syst sci.net/16/2771/2012/. Accessed November 26, 2013. Zidansek, A., M. Limbek, and I. Slaus. 2014. Contemporary Crises and Sustainability Indicators. Journal of Sustainable Development of Energy 2(2): 100 Ð 107.

PAGE 182

160 APPENDIX A Ð Greenhouse Ga s (GHG) Emissions Indicator Table 20 : Data and data sources used for the GHG emissions indicator . Component Purpose/Notes Source Website Specific data source, link , and/ or contact person County/local data Population Used for all of the per/capita calculations Census Bureau https://www.census .gov https://www.census.gov/qui ckfacts/dashboard Income Used for all per/income calculations Bureau of Economic Analysis http://www.bea.gov http://www.bea.gov/iTable/i Table.cfm?reqid=70&step= 1&isuri=1&acrdn=5#reqid= 70&step=1&isuri=1 Natural gas consumption Residential and commercial emissions Xcel Energy https://www.xcelen ergy.com Curt Dallinger; Xcel Energy | Responsible By Nature; Director Gas Resource Planning; 1800 Larimer St Electricity consumption Residential, commercial, and agricultural emissions Xcel Energy (Public data request) Rural Electric of the SLV (Public data request) https://www.xcelen ergy.com http://www.slvrec.c om Curt Dallinger; Xcel Energy | Responsible By Nature; Director Gas Resource Planning; Curtis.Dallinger@xcelenerg y.com JoAn Waudby; San Luis Valley REC; Chief Financial Officer; jwaugh@slvrec.com; (719)852 6631 Local utility scale solar generation Emissions credits Energy Information Administration http://www.eia.gov http://www.eia.gov/electricit y/data/browser/?src=home f1# Solid waste generation Landfill emissions SLV D evelopment R esources G roup http://www.slvdrg.or g http://www.slvdrg.org/ceds/ CEDS2013/2013 %20P%20 Utilities.pdf Population in towns and rural areas Waste water emissions Census Bureau https://www.census .gov http://www.census.gov/quic kfacts/table/PST045215/00 SLV vehicle composition Vehicle emissions SLV Development Resources Group http://www.slvdrg.or g http://www.slvdrg.org/ceds/ CEDS2013/2013%20O%20 Transportation.pdf Head of livestock Livestock emissions N ational Agricultural Statistics https://www.nass.u sda.gov https://quickstats.nass.usda .gov Acres of Agricultural C olorado http://water.state.co http://cdss.state.co.us/Mod

PAGE 183

161 crops planted land use emissions D ivision of W ater Resources .us/Home/Pages/de fault.aspx eling/Pages/ConsumptiveU seStateCU.aspx Contact: James Heath Fuel consumed from agriculture Agricultural fuel emissions National Agricultural Statistics https://www.nass.u sda.gov https://quickstats.nass.usda .gov National level data U.S. Consumer Price Index (for 2002 dollars) Used for all per/$ benchmarking Bureau of Economic Analysis http://www.bea.gov http://www.bea.gov/iTable/i ndex_nipa.cfm BTU per train car mile Train emissions Association of American Railroads http://www.slvdrg.or g http://www.slvdrg.org/ceds/ CEDS2013/2013%20O%20 Transportation.pdf Modeled data Waste water generation Waste water emissions Intergovernme ntal Panel on Climate Change http://www.ipcc.ch http://www.ipcc nggip.iges.or.jp/public/2006 gl/vol5.html P ropane consumption Residential emissions E nvironmental P rotection A gency https://www.epa.go v/aboutepa/about national risk management research laboratory nrmrl https://cfpub.epa.gov/si/si_ public_record_report.cfm?d irEntryId=230098 p.g. 39 Daily vehicle miles traveled (VMT) Vehicle emissions C olorado D epartment of T ransportation https://www.codot.g ov http://dtdapps.coloradodot.i nfo/otis/TrafficData Average miles per gallon in the SLV Vehicle emissions SLV Development Resources Group http://www.slvdrg.or g http://www.slvdrg.org/ceds/ CEDS2013/2013%20O%20 Transportation.pdf Train car miles travelled Train emissions SLV Development Resources Group http://www.slvdrg.or g http://www.slvdrg.org/ceds/ CEDS2013/2013%20O%20 Transportation.pdf Airplane hours Airplane emissions F ederal A viation A dministration http://www.faa.gov http://www.gcr 1.com/5010w eb/

PAGE 184

162 Table 21 : Model inputs for the GHG emissions indicator (1980 1989) . Data sources are presented in Table 20 . Blank cells in this table show data were unavailable for that year through the data provider . The model linearly interpolates data gaps.

PAGE 185

163

PAGE 186

164

PAGE 187

165

PAGE 188

166

PAGE 189

167 Table 22 : Model inputs for the GHG emissions indicator (1990 1999). Data sources are presented in Table 20 . Blank cells in this table show data were unavailable for that year through the data provider. The model linearly interpolates data gaps.

PAGE 190

168

PAGE 191

169

PAGE 192

170

PAGE 193

171

PAGE 194

172 Table 23 : Model inputs for the GHG emissions indicator (2000 2009 ). Data sources are presented in Table 20 . Blank cells in this table show data were unavailable for that year through the data provider. The model linearly interpolates data gaps.

PAGE 195

173

PAGE 196

174

PAGE 197

175

PAGE 198

176

PAGE 199

177 Table 24 : Model inputs for the GHG emissions indicator (2010 2017 ). Data sources are presented in Table 20 . Blank cells in this table show data were unavailable for that year through the data provider. The model linearly interpolates data gaps.

PAGE 200

178

PAGE 201

179

PAGE 202

180

PAGE 203

181

PAGE 204

182 Table 25 : Model outputs from the GHG emissions indicator (1980 1989) .

PAGE 205

183

PAGE 206

184 Table 2 6 : Model outputs from the GHG emissions indicator (1990 1999)

PAGE 207

185

PAGE 208

186 Table 27 : Model outputs from the GHG emissions indicator (2000 2009) .

PAGE 209

187

PAGE 210

188 Table 28 : Model outputs from the GHG emissions indicator (2010 2016) .

PAGE 211

189

PAGE 212

190 B Ð Consumptive W ater U se (CWU) Indicator Table 29 : Da ta and data sources used for the CWU (C W U) indicator Component Purpose/Notes Source Website Specific data source, link, or contact County/local level data Crop Acreage Used to calculate total CWU Colorado Division of Water Resources http://water.state.co. us/Home/Pages/def ault.aspx http://cdss.state.co.us/onlineTools/ Pages/BulkHydroBaseDataExporte r.aspx Contact: James Heath using data2.xls Crop Yields Used to calculate virtual water content National Agricultural Statistics https://www.nass.us da.gov https://quickstats.nass.usda.gov Head of livestock Used to calculate livestock CWU National Agricultural Statistics https://www.nass.us da.gov https://quickstats.nass.usda.gov National level data Weight of livestock Used to calculate virtual water content Institute for Water Education https://www.unesco ihe.org http://waterfootprint.org/media/dow nloads/Report13.pdf Appendix IXa 1 Daily Feed Requiremen ts of Livestock Used to calculate virtual water cont ent Institute for Water Education https://www.unesco ihe.org http://waterfootprint.org/media/dow nloads/Report13.pdf Appendix IXa 1 Modeled data Effective precipitation by crop type Used to calculate total CWU Colorado Division of Water Resources http://water.state.co. us/Home/Pages/def ault.aspx http://cdss.state.co.us/onlineTools/ Pages/BulkHydroBaseDataExporte r.aspx Contact: James Heath using data2.xls Surface Water use by crop type Used to calculate total CWU Colorado Division of Water Resources http://water.state.co. us/Home/Pages/def ault.aspx http://cdss.state.co.us/onlineTools/ Pages/BulkHydroBaseDataExporte r.aspx Contact: James Heath using data2.xls Ground water use by crop type Used to calculate total CWU Colorado Division of Water Resources http://water.state.co. us/Home/Pages/def ault.aspx http://cdss.state.co.us/onlineTools/ Pages/BulkHydroBaseDataExporte r.aspx Contact: James Heath using data2.xls Other sub irrigation use by crop Used to calculate total CWU Colorado Division of Water http://water.state.co. us/Home/Pages/def ault.aspx http://cdss.state.co.us/onlineTools/ Pages/BulkHydroBaseDataExporte r.aspx

PAGE 213

191 type Resources Contact: James Heath using data2.xls Municipal and Industrial water use Used to calculate total CWU Colorado Division of Water Resources http://water.state.co. us/Home/Pages/def ault.aspx http://cdss.state.co.us/onlineTools/ Pages/BulkHydroBaseDataExporte r.aspx Contact: James Heath using data2.xls

PAGE 214

192 Table 30 : Model inputs for the CWU indicator (1980 1989) . Data sources for all data are presented in Table 29 . Pay close attention as the units change throughout the table.

PAGE 215

193

PAGE 216

194 Table 31 : Model inputs for the CWU indicator (1990 1999). Data sources for all data are presented in Table 29 . Pay close attention as the units change throughout the table.

PAGE 217

195

PAGE 218

196 Table 32 : Model inputs for the CWU indicator (2000 2009). Data sources for all data are presented in Table 29 . Pay close attention as the units change throughout the table.

PAGE 219

197

PAGE 220

198 Table 33 : Model inputs for the CWU indicator (2010 2014). Data sources for all data are presented in Table 29 . Pay close attention as the units change throughout the table.

PAGE 221

199

PAGE 222

200 Table 34 : Model outputs for the CWU indicator (1980 1990) .

PAGE 223

201

PAGE 224

202

PAGE 225

203

PAGE 226

204 Table 35 : Model outputs for the CWU indicator (1991 2001).

PAGE 227

205

PAGE 228

206

PAGE 229

207

PAGE 230

208 Table 36 : Model outputs for the CWU indicator (2002 2010).

PAGE 231

209

PAGE 232

210

PAGE 233

211

PAGE 234

212 C Future Scenario Modeling Figure 24 : Surface soil texture distribution among all irrigated lands in SLV .

PAGE 235

213 Figure 25 : Surface soil texture distribution among all alfalfa lands in SLV .

PAGE 236

214 Figure 26 : Surface soil texture distribution among the alfalfa lands sample in SLV .

PAGE 237

215 Figure 27 : Surface soil texture distributi on among all potato lands in SLV .

PAGE 238

216 Figure 28 : Surface soil texture distribution among the potato lands sample in SLV .

PAGE 239

217 Figure 29 : Surface soil texture distribution among a ll the small grain lands in SLV .

PAGE 240

218 Figure 30 : Surface soil texture distribution among the small grain lands sample in SLV .

PAGE 241

219 Figure 31 : S urface soil texture distribution among all meadow/ pasture lands in SLV .

PAGE 242

220 Figure 32 : Surface soil texture d istribution among the meadow/ pasture lands sample in SLV .