University of Southern California Dissertations and Theses

Developing high-resolution spatiotemporal methods to model and quantify water use for energy

(USC Thesis Other)

Developing high-resolution spatiotemporal methods to model and quantify water use for energy

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Developing High-Resolution Spatiotemporal Methods to Model and Quantify
Water Use for Energy
by
Rebecca Peer
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Environmental Engineering)
May 423;
Dissertation Committee
Kelly T. Sanders, Ph.D. (Chair)
Lucio Soibelman, Ph.D.
Bistra Dilkina, Ph.D. (Outside Member)
Copyright 423; Rebecca Peer
Everything is energy and that’s all there is to it.
— ALBERT EINSTEIN
Acknowledgements
The author acknowledges the partial financial support of the University of Southern California’s
Provost Fellowship, the Women in Science and Engineering (WiSE) top-off Fellowship, the PEO
Scholar Award, as well as the National Science Foundation’s Early-concept Grant for Exploratory
Research program (CBET 3854;67).
I would like to thank the members of my qualifying and defense committees for their helpful
comments, support, and guidance throughout my degree.
I would like to thank all of my friends and family that endured hours of late-night revisions
and practice sessions, talks about life and science, my numerous existential crises, endless cups
of coffee, mindless television shows, and countless other things all in the name of support for me
during my doctoral studies.
Finally, I would like to thank my advisor, Kelly Sanders, for her unyielding support and eter-
nal confidence. You always believed in me, even when I didn’t.
iii
Table of Contents
Acknowledgements iii
List Of Tables vi
List Of Figures viii
Abstract x
Chapter 3: Introduction 3
3.3 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3.4 Summary of research gaps and contributions of this work . . . . . . . . . . . . . . 6
3.5 Structure of document and resulting publications to date . . . . . . . . . . . . . . 8
Chapter 4: Background :
4.3 Factors that influence water use for electricity
generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . :
4.3.3 Cooling systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ;
4.3.4 Fuel type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.3.5 Prime movers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.3.6 Ambient climate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.3.7 Life cycle water use considerations for the electricity sector . . . . . . . . . 34
4.4 Other environmental considerations for electricity
generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.4.3 Greenhouse gas emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.4.4 Air quality pollutants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.4.5 Spatial and temporal environmental trade-offs of across power generation
technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Chapter5: Building a database of cooling water usage rates for US thermoelectric
power plants 3;
5.3 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3;
5.4 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
5.5 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
5.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
Chapter 6: Quantifying the impacts of recent power sector transitions on cooling
water usage at US thermoelectric power plants 56
6.3 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
6.4 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
6.5 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
iv
6.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
Chapter 7: Developing methods to estimate the environmental externalities of
the power sector with high spatio-temporal resolution 7:
7.3 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7:
7.4 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
7.5 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
7.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Chapter 8: Quanitfying Regional Water Use Rates for the Electricity Sector 97
8.3 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
8.4 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
8.5 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . :3
8.5.3 eGRID regional variability in electricity-associated consumptive water in-
tensity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ::
8.5.4 Data uncertainties and limitations . . . . . . . . . . . . . . . . . . . . . . . ;2
8.5.5 Policy drivers and implications of variability in water consumption from
electricity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ;3
8.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ;4
Chapter 9: Conclusion ;6
Reference List ;9
Appendix A
Supplemental information for chapter 4: Using self-reported data to build a water use
rate database for US power plants . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
A.3 Nationwide Cooling Water System and Source
Characterization Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
A.4 Plant-by-Plant Water Withdrawal and Consumption Rate Analysis . . . . . . . . . 338
A.5 Comparison to literature and heat budget models . . . . . . . . . . . . . . . . . . . 33:
A.6 Suggestions for EIA data improvement . . . . . . . . . . . . . . . . . . . . . . . . . 345
Appendix B
Supplemental information for chapter 6: A case study of spatial and temporal environ-
mental impacts of electricity generation in Texas . . . . . . . . . . . . . . . . . . . 346
B.3 Natural gas combined cycle conversion results . . . . . . . . . . . . . . . . . . . . . 346
B.4 Peak-shifting analysis results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34;
Appendix C
Supplemental information for chapter 7: Quanitfying Regional Water Use Rates for the
Electricity Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
C.3 Primary energy consumptive water intensity . . . . . . . . . . . . . . . . . . . . . . 356
C.4 Point of generation consumptive water intensity . . . . . . . . . . . . . . . . . . . . 35:
C.5 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
v
List Of Tables
5.3 US thermoelectric generation by cooling system type . . . . . . . . . . . . . . . . . 46
5.4 US thermoelectric generation by cooling water source type and quality . . . . . . . 47
5.5 Water withdrawal and consumption factors for thermoelectric generation units in
the US . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4:
6.3 Net changes in generation, capacity, water withdrawal, and water consumption
across the US Power Sector between 422: and 4236 . . . . . . . . . . . . . . . . . 66
7.3 Relative distribution of generation, emissions, and water consumption by generat-
ing technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8:
7.4 Waterconsumptionconsequencesofconversiontonaturalgascombinedcyclepower
plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
8.3 Regional water consumption intensity for different generating technologies . . . . . :5
A.3 Level of confidence in characterizing cooling sources . . . . . . . . . . . . . . . . . 334
A.4 Level of confidence in characterizing cooling system technologies . . . . . . . . . . 336
A.5 US power generation by cooling source type . . . . . . . . . . . . . . . . . . . . . . 337
A.6 US power generation by cooling source quality . . . . . . . . . . . . . . . . . . . . 337
A.7 Dry cooled power generation in US by state . . . . . . . . . . . . . . . . . . . . . . 338
A.8 EIA reported water usage data as a percentage of total 4236 generation . . . . . . 339
A.9 Comparison of operational water use rate studies . . . . . . . . . . . . . . . . . . . 344
B.3 Change in emissions and water use in natural gas combined cycle conversion scenarios349
B.4 Socioeconomic impacts of air pollution from natural gas combined cycle conversion 34:
B.5 Shifts in water consumption with natural gas combined cycle conversions . . . . . 34;
vi
B.6 Shifts in environmental externalities with different demand response profiles . . . . 353
C.3 Description of eGRID regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
C.4 EIA fuel code proxies for fuel classifications . . . . . . . . . . . . . . . . . . . . . . 357
C.5 Upstream consumptive water intensity factor by fuel . . . . . . . . . . . . . . . . . 358
C.6 Definition of coal regions used in this work . . . . . . . . . . . . . . . . . . . . . . . 358
C.7 Separation of Eastern and Western Kentucky counties . . . . . . . . . . . . . . . . 359
C.33 Water consumption assumptions for each generating technology classification . . . 35:
C.8 EIA fuel code proxies for electricity fuel classifications . . . . . . . . . . . . . . . . 363
C.9 EIA prime mover code proxies for prime mover classifications . . . . . . . . . . . . 364
C.: EIA cooling system code proxies for cooling system classifications . . . . . . . . . . 364
C.; EIA water type code proxies for water type classifications . . . . . . . . . . . . . . 365
C.32 EIA water quality code proxies for water quality classifications . . . . . . . . . . . 365
C.34 Regional volumetric water consumption across generating technologies . . . . . . . 365
C.35 Upstream consumptive water intensity by fuel for each eGRID region, 4236 . . . . 366
C.36 Point of generation consumptive water intensity by fuel by eGRID region, 4236 . . 367
C.37 Total consumptive water intensity by fuel by eGRID region, 4236 . . . . . . . . . . 368
vii
List Of Figures
3.3 Pathway displaying the research contributions from this work . . . . . . . . . . . . 7
4.3 Cooling water withdrawals in the US by industrial sector . . . . . . . . . . . . . . ;
5.3 Geographic representation of 4236 thermoelectric generation separated by cooling
system type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
5.4 Geographic representation of US water withdrawals and consumption for thermo-
electric power plant cooling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5.5 Box and whiskers plot of water withdrawal and consumption rates across thermo-
electric generating technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5.6 Comparison of monthly average water usage rates versus monthly generation for
once-through cooled power plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
6.3 Changes in thermoelectric generation and capacity requiring cooling between 422:
and 4236 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
6.4 Changes in cooling water withdrawals and consumption from thermoelectric ge-
neation requiring cooling between 422: and 4236 . . . . . . . . . . . . . . . . . . . 6:
6.5 Changes in generation, water withdrawals, and water consumption for added and
retired power plants by HUC-: basin between 422: and 4236 . . . . . . . . . . . . 72
7.3 Electricity generation, emissions, and water consumption profiles for EGUs in the
ERCOT region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
7.4 Averaged intensities of environmental externalities across ERCOT . . . . . . . . . 89
7.5 A few power plants contribute a large fraction of environmental damage in the
ERCOT grid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8;
8.3 Regional water consumption allocation approach . . . . . . . . . . . . . . . . . . . :3
8.4 Total regional water consumption intensity and generation . . . . . . . . . . . . . . :4
8.5 Regional water consumption, separated by fuel type . . . . . . . . . . . . . . . . . :6
viii
8.6 Regional share of upstream and point of generation water consumption in electric-
ityâĂŹs consumptive water footprint . . . . . . . . . . . . . . . . . . . . . . . . . . :8
8.7 Regional water consumption, separated by water quality and type . . . . . . . . . :9
A.3 Distribution of water-cooled thermoelectric power plants . . . . . . . . . . . . . . . 33;
A.4 Regression analysis for impact of heat rate on water usage . . . . . . . . . . . . . . 342
A.5 Regression analysis for impact of heat rate on all water usage rates . . . . . . . . . 343
B.3 ERCOT fleet with : coal-fired plants converted to natural gas combined cycle . . . 347
B.4 ERCOT fleet with : coal-fired and 4 nuclear plants converted to natural gas com-
bined cycle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348
B.5 Peak shifted load for August 3, 4233. . . . . . . . . . . . . . . . . . . . . . . . . . . 352
C.3 Map of eGRID regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
C.4 Regional distribution of generation, separated by fuel type. . . . . . . . . . . . . . 362
C.5 Regional distribution of generation, separated by cooling system type. . . . . . . . 36:
C.6 Regionalwaterconsumptionatthepointofgeneration, separatedbycoolingsystem
type. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36:
ix
Abstract
Electricity systems are integral in the developed and developing world and access to electricity
improves overall quality of life and enables basic human needs, such as communication, education
and sanitation. The electricity system has complex relationships with the environment and is
especially linked to the water system. Electricity is required for water treatment, transport, and
delivery and water is required for primary fuel extraction, preparation, transport, and conversion
to electricity. Although this relationship is understood, there are few efforts prior to this body
of work that present usable data and methodologies to quantify the water used for the electricity
sector. Specifically, this body of work quantifies water use for different generating technologies,
examines the changes in water use (in magnitude and across space) in response to changes in
the electricity grid, develops a methodology to quantify water and emissions with high temporal
resolution, and calculates regionally-specific life cycle (i.e., from resource extraction to electricity
generation) water consumption volumes and intensities for the United States (US) electricity
sector. Collectively, this work represents a set of tools for researchers, water and energy planners,
and decision makers to better account for the water requirements of electricity systems. This
work is particularly relevant in a world with a changing climate, where water resource availability
and distribution are uncertain and electricity grids are shifting quickly in response to climatic,
political, and economic drivers.
x
Chapter 3
Introduction
3.3 Motivation
Energy and water systems are interconnected and share a complex, symbiotic relationship. We
rely on energy to heat, transport, and treat water to useable standards. Similarly, we rely on
water to extract, refine, and process fuels, as well as to generate electricity. This body of work
is focused on the latter half of this relationship, particularly the water that is used to generate
electricity. In their Annual Energy Outlook, the US Energy Information Administration (EIA)
predicts that global electricity demand will continue to increase until 4272, which in turn requires
a predicted increased installation of power plant capacity [383]. Although the relationship between
the electricity and water sectors is understood, there is very little understanding of the volume of
water required for electricity generation and how this magnitude varies as a function of technology
configuration, power plant location, shifts in the energy sector, across time, or across the life cycle
of fuels used to generate electricity. This work fills this large knowledge gap by providing novel
databases and modelling techniques that can be used to quantitatively assess the water impacts of
the electricity grid in current, past, and future grid configurations. Although the work presented
in this dissertation is central to the United States, the methods and databases can be applied for
similar research in other areas of the globe.
3
There are many technology configurations available for electricity generation in the United
States that vary widely across the nation. The majority (:7% in 4239 [37;]) of electricity gener-
ated in the US is provided from thermoelectric power plants where heat is used from the combus-
tion of a fuel (e.g., coal, natural gas, biomass, etc.) or from the fuel itself (e.g., geothermal, solar
concentratingpower, etc.) inaboilertoconvertwatertosuperheatedsteamthatmovesamechan-
ical turbine to generate electricity [8:]. Steam exiting the turbine is condensed into water using a
condenser and pumped back into the boiler to complete the steam (Rankine) cycle [8:,33;]. Some
thermoelectric generators couple a steam turbine with a combustion turbine in a combined cycle
to generate electricity more efficiently by utilizing waste heat from the combustion turbine in the
Rankine cycle. All thermoelectric generators using a steam cycle require a cooling fluid (typically
water) to condense steam back into the liquid phase to begin the cycle again. Gas combustion
turbines contribute a small fraction of thermoelectric generation, typically during times of peak
power demand, and require much less water than other thermoelectric power plants because they
do not require a cooling system for operation. The remainder of generation in the US is provided
by non-thermoelectric renewable generating technologies, such as solar photovoltaics, wind tur-
bines, and hydroelectric turbines, which require no water for cooling. Quantifying the water use
for thermoelectric generators and assessing the variability across technology types is the first step
in this research effort and constitutes the foundation for the body of this work. Assessing the
water use variability across technology types informs our assessment of the water implications of
transitions in the energy sector.
In the last decade, the US electricity sector has undergone large transitions away from the
traditional grid of the industrial era towards the grid of the future. These changes have been mo-
tivated by economics, policy, social opinion, and technology advancements [8;,33:]. For example,
the use of hydraulic fracturing and horizontal drilling for oil and gas recovery has significantly in-
creased the production of oil and natural gas resources, which has driven natural gas fuel prices to
cost-competitive levels with both coal and nuclear fuels. Furthermore, federal and state policies,
4
such as the proposed Clean Power Plan, the Clean Water Act, the Clean Air Act, the Mercury
and Air Toxics Rule, and State Renewable Portfolio Standards have driven shifts in generating
technologies away from conventional coal-fired power plants toward renewable generation. There-
fore, research in the energy field is ripe to evaluate these shifts in grid structure and to investigate
the associated impacts on the environment.
Further, it is important to evaluate water use in the electricity sector on a time scale that
is relevant for water planners and decision makers. Developing a methodology to assess the
temporal variability of water use in the electric sector with high spatial resolution is another
important output of this work. The electricity grid has a relationship with the water sector, in
terms of water required for power plant cooling or hydroelectric generation, with air quality, in
terms of the air quality pollutants that are emitted from the combustion processes, and with
climate change, in terms of the greenhouse gas (GHG) emissions from combustion of fossil fuels.
Theelectricitygeneratingunits(EGUs, orgenerators)thatcompriseUSpowerplantshaveuneven
environmental impacts that vary in time and space. For example, generating technologies that are
beneficial for air quality or GHG emissions could be detrimental for the water sector (e.g., nuclear,
concentrating solar power) or for the reliability of the power grid (e.g., intermittent renewables).
Prior to this work, there was a lack of methods available to quantify and compare the multi-
faceted environmental trade-offs of power generation on the spatio-temporal scales pertinent for
decision-making.
Although US electricity generating infrastructure is becoming “cleaner” over time and there is
an increased focus on the water impacts of electricity generation, processes occurring upstream of
the power plant to prepare primary fuels are often neglected in environmental assessments. The
final contribution of this work further widens the scope at which we evaluate water use for the
electricity sector in the research community by evaluating the water use embedded in processes
occurring upstream of the power plant for primary fuel extraction, refining, and processing. This
work presents a more holistic understanding of the resources that are embedded in the electricity
5
we consume across the grid, which can be incorporated when evaluating the environmental impact
of electricity transitions, electrification, and urbanization. Ultimately, this work contributes to a
greater understanding of the water impacts of the electricity sector from a systems-scale perspec-
tive, drawing the boundary of our system from resource extraction to conversion to electricity.
3.4 Summary of research gaps and contributions of this
work
The overall goal of this research plan is to develop holistic frameworks to quantify and mitigate
the water-related externalities of power generation. To do so, methods are devised to address
data gaps (particularly in regards to the cooling requirements of power plants), to calculate the
environmental impacts of power generators with high spatio-temporal resolution, and finally, to
calculate regionally specific life cycle water consumption intensities for the US power grid. The
research pathway for this dissertation is displayed in Figure 3.3, showing that the results from
each prior study informs the next as the scope of the body of work increases. In all, this work
outlines and addresses the following research gaps:
3. There are a lack of cooling water usage data available for power plants in the literature for
the research community and decision makers. My work addresses this large data need in
Chapter 5 by analyzing the primary data for water use reported to the EIA in 4236.
4. The water-related consequences of recent power sector transitions have not been evaluated
with meaningful spatial resolution. My research addresses this gap in the literature in
Chapter 6 by systematically evaluating the water use consequences of the recent changes
in fuel, cooling system, and cooling water sources across the US power sector using the
database and methods developed in the previous study.
6
5. Methods to quantify multiple environmental externalities with high spatio-temporal resolu-
tion are lacking. My research addresses this gap in Chapter 7 and provides new information
on the time-variability of water use & emissions as well as a greater understanding of the
spatial variability of water use for US power plants using information from the two previous
studies.
6. Spatially-relevant and updated estimates of life cycle water consumption for electricity (i.e.,
including water consumed for primary fuel preparation occurring upstream of the power
plant) are lacking in the research community. My work addresses this gap in Chapter 8
by combining knowledge from the three previous studies and using a bottom-up analysis
to assess life cycle water use for every power generating facility greater than 3 megawatt in
capacity (i.e., connected to the grid) in the United States, using a 4236 base year.
Figure 3.3: Pathway with research summaries displaying the flow and connectivity of research
contributions from this work.
Water use
across
generating
technologies
Variability in water
use across space
and time
High-resolution spatio-
temporal modelling of
environmental impacts of
generation
Accounting for life cycle water
consumption in water use
estimates
Increasing scope
The work in this dissertation addresses the research gaps noted above by providing databases
and modelling tools for the research community and decision makers to better quantify and un-
derstand the water requirements of the electric sector. This work contributes a set of tools that
can be used in environmental studies of grid capacity expansion, electrification, and urbanization
to better understand the water consequences of planned and possible future changes to the elec-
tricity grid. This has obvious implications for assessing policies that will affect the our energy,
and more specifically our electricity, future.
7
3.5 Structure of document and resulting publications to
date
This document is organized into five additional chapters, and a concluding chapter. Chapter 4
describes the background information and literature to inform this body fo work. The four fol-
lowing chapters each correspond to a research gap noted above. Chapter 5 details the collection
and analysis of raw data from the EIA to create a database of operational water withdrawal and
consumption rates for thermoelectric power plants requiring cooling in the US. Chapter 6 uses
this database to asses the impact of recent transitions in the power sector, notably toward nat-
ural gas-fuel and recirculating-cooled power plants, on freshwater withdrawals and consumption
across the US. Chapter 7 creates a methodology to quantify the water, air, and GHG impacts
of individual power generators across entire gird networks with high spatio-temporal resolution,
using the Electric Reliability Council of Texas (ERCOT) as a case study. Chapter 8 describes the
calculation of regional water consumption and water consumption intensities for the US electric
sector, including the water consumption embedded in primary fuel processing upstream of the
power plant. The work described in Chapters 5, 6 and 7 has been published and the work in
Chapter 8 has been submitted for publication in the following peer-reviewed journals:
• Chapter 5: Peer & Sanders (4238) “Characterizing cooling water source and usage patterns
across US thermoelectric power plants: a comprehensive assessment of self-reported cooling
water data.” Environmental Research Letters, 33(34): 3-32 (423: Impact Factor: 6.763)
• Chapter 6: Peer & Sanders (4239) “The water consequences of a transitioning US power
sector.” Applied Energy, in press (423: Impact Factor: 9.;)
• Chapter 7: Peer, Garrison, TImms, and Sanders (4238) “A Spatially and Temporally Re-
solved Analysis of Environmental Trade-Offs in Electricity Generation.” Environmental
Science & Technology, 72: 6759-6767 (423: Impact Factor: 8.875)
8
• Chapter8: Peer,Grubert&Sanders(423;)“ARegionalAssessmentoftheWaterEmbedded
in the US Electricity System.” in review
Finally, Chapter8 summarizes the above studies and their contributions to the body of knowledge
in water and electricity.
9
Chapter 4
Background
4.3 Factors that influence water use for electricity
generation
Water “use” is separated into water withdrawals and water consumption: Withdrawals represent
the total amount of water removed from a source (river, reservoir, ocean, etc.), while water
consumption represents the amount of water that is not returned to that source (i.e. evaporative
losses) [65,8:,:5,33;,34:,356]. According to 4232 USGS data, approximately 577,222 million
gallons of water is withdrawn for use each day in the United States, almost half (63%) of which
is used for thermoelectric power generation (Figure 4.3) [56,::]. Approximately 5% of this
withdrawn water is consumed [56]. The volume of cooling water required for thermoelectric
generation varies by fuel type, generating technology (prime mover), cooling technology, pollution
controls, and ambient climate [8:,33;,344]. This volume can be several orders of magnitude
greater than the volume of water circulating in the steam loop in a given generating unit [:5]. A
less thermally efficient unit usually requires more primary fuel, and consequently, more water for
coolingperunitofelectricitygenerated. Likewise, thecooling waterrequired perunit ofelectricity
risesasthetemperaturerises,allelsebeingequal. Theefficiencyofathermalpowerplantgenerally
rises as the ratio of the temperature of the cooling fluid to the operating temperature of the boiler
:
Figure 4.3: Cooling thermoelectric power generation requires more water than any other sector
in the US, withdrawing 93:7 million gallons per day in 4232. (Data for this figure from [::])
decreases. However, factors such as auxiliary systems and the natural environment also affect
efficiency. The main factors influencing water use at thermoelectric power plants include cooling
system technology, fuel type, prime mover (generating technology), and ambient climate. Water
is also used for the preparation of primary energies upstream of the power plant, the volume of
which is heavily influenced by geography.
4.3.3 Cooling systems
Most thermoelectric facilities in the U.S. today use water as a cooling fluid in an open-loop
or closed-loop wet cooling system. Open-loop (once-through) cooling systems withdraw large
volumes of water that pass through the condenser, which condenses steam exiting the turbine, and
are discharged at a higher temperature to a receiving water reservoir. Closed-loop (recirculating)
cooling systems withdraw smaller volumes per unit of generation by recirculating the cooling
water withdrawn continuously in cooling towers or cooling ponds [8:,33;]. Open-loop cooling
systems demand large water withdrawals (and therefore require larger stream flows) compared
to closed-loop cooling systems, but lose a very small fraction of this water as evaporation (i.e.
;
“water consumption”). Closed-loop cooling systems, on the other hand, typically require smaller
water withdrawals but lose the majority of this water via evaporation. Dry cooling systems that
use ambient air as a cooling fluid appear attractive as an alternative for cooling systems where
water availability might be an issue. However, when compared to wet cooling systems, dry cooling
systemsaretypicallymoreexpensive, lessefficient, andhave largerland requirements[87]. Hybrid
cooling systems combine the use of wet and dry cooling to increase cooling system efficiency.
However, these systems are typically cost-prohibitive and suffer the same shortcomings as dry
cooled systems in the absence of water resources [34;].
4.3.4 Fuel type
The differences in combustion (or conversion) characteristics of different primary energies (i.e.
coal, natural gas, petroleum, uranium, biomass, solar, or geothermal energy) affect the efficiency
of transforming primary energy into finished electricity, thereby altering the cooling water require-
ments of their respective generating unit. Converting thermal energy derived from natural gas to
finished electricity, for example, is generally more efficient than using coal or nuclear fuels. Some
renewable energy facilities using fuels such as geothermal, biomass, and concentrating solar power
require water for cooling, because they also operate using thermal power generation cycles. Wet
cooled concentrating solar power facilities use more water than conventional generation technolo-
gies, which could limit the expansion of concentrating solar power in water-scarce regions if dry
cooling is not used as an alternative [38,3:,4;].
4.3.5 Prime movers
Non-thermoelectric generating technologies (e.g., solar PV and wind turbines) do not require
water for operation, with the exception of water used to clean solar panels and supress dust
[32,38,:4]. Prime movers used in thermoelectric generation include steam cycles, combined cycles,
and combustion turbines. Steam cycle generators follow a simple Rankine cycle to generate
32
electricity from a turbine. Combined cycle generators couple steam turbines and combustion
turbines such that one third of the net electricity output is generated using a steam turbine and
the remaining two thirds is generated using combustion turbines. Because combustion turbines
use little to no water for operation, water use at combined cycle facilities is greatly reduced
[:5,33;,344]. Combustion turbines require little to no water, but are generally small and less
efficient than larger steam or combined cycle units. However, despite their ability to operate with
almost no water requirements, combustion turbine facilities tend to be much more expensive to
operate per unit of electricity generated than larger thermoelectric facilities.
4.3.6 Ambient climate
Climatic variables such as air temperature, water temperature, stream flow, as well as the oc-
currence of extreme events can impact generation technologies and their water requirements
[55,79,88,93,;6,;:,342,343,347,38;,399]. Ambient air temperature influences the condenser
and turbine back pressure, which both rise with increasing temperature thus decreasing overall
efficiency. A decrease in efficiency results in more cooling water required per unit of electricity
generated [99,;4,343]. Extended periods of increased or decreased temperature would likely also
cause increases in electricity demand due to space cooling and heating demand, requiring water
for cooling excess generation and placing strain on the system. The efficiency of a thermoelectric
generating facility is influenced by cooling reservoir temperature because the temperature differ-
ential between the boiler and the cooling water moderates the efficiency of the thermal power
cycle [;4]. Therefore, if cooling water reservoir temperature rises, the cooling system becomes
less efficient and more water is required to produce each unit of electric output. Cooling water
withdrawalscanbesubjecttorestrictionsifstream flowapproachesminimumlimitssettosupport
aquatic ecosystem health [6,36,78,395]. Changes in stream flow are also important for hydro-
electric generation because small changes in flow can result in amplified changes in generation
output [387]. Extreme events, such as storms, wildfires, and especially drought can have adverse
33
effects on generating technologies, transmission and distribution systems, and the reliability of the
grid [7,45,55,93,342,343]. These climatic variables can cause power plant curtailments and shut-
downs to maintain the reliable operation of the grid. In instances of prolonged drought, adequate
water supply becomes a real concern for traditional thermoelectric generators [45,322,32;,399]
and hydroelectric facilities [79,88,;6,;:,38;].
4.3.7 Life cycle water use considerations for the electricity sector
In addition to water used for thermoelectric power plant cooling, water is also used for the
preparation of primary fuels for electricity generation and in life cycle stages occurring upstream
of electricity generation. The volume of water used for fuel extraction, processing, transportation,
and conversion into electricity varies a great deal across fuel types, production techniques, and
electricity generation technologies [75].
4.4 Other environmental considerations for electricity
generation
4.4.3 Greenhouse gas emissions
Greenhouse gas emissions from electricity generation are a direct result of fossil fuel combustion
for the operation of thermoelectric power plants. In the US, coal and natural gas are the two
most common fossil fuels used in power generation and together account for almost two thirds of
generating capacity in the country. The combustion of these carbon-based fuels releases CO
2
into
the atmosphere, a globally well-mixed pollutant that has climate-forcing capabilities [83]. GHG
emissions from electricity generation are a global issue, accounting for approximately 62% of total
worldwide anthropogenic CO
2
equivalent emissions [83]. GHG emissions from power plants are a
function of carbon content of fuels and efficiency of power plants. For example, natural gas has a
34
lower carbon content than most coal combusted in the US for power generation, therefore emits
less CO
2
per unit of fuel combusted. Additionally, a combined cycle power plant would emit less
CO
2
per unit of electricity generated than a steam cycle power plant due to the plant’s increased
efficiency. CO
2
emissions can be reduced by up to 82% at a natural gas combined cycle power
plant when compared to a steam cycle coal power plant [:3].
Current strategies to mitigate GHG emissions from power generation are technology or policy-
driven. Many generating technologies exist that do not emit GHG gases to the atmosphere (e.g.
nuclear, wind, solar, etc.), but these technologies have other trade-offs with water use, public
perception, and reliability. Common policies used today for mitigating GHG emissions from
power generating facilities include carbon taxes and carbon cap and trade programs. A carbon
tax is a simple fee placed on emissions of CO
2
emissions. Cap and trade programs set a limit
on CO
2
emissions with penalties for exceedences, but also the opportunity for power plants to
buy unused credits from other plants, essentially raising their limit. However, the efficacy of thee
programs are moderated by the cost imposed on polluters. New policies placing restrictions or
pricepenaltiesonemissionscouldforcethedevelopmentofpollutioncontroltechnologiesthathave
yet to be widely implemented. For example, carbon capture and sequestration (CCS) systems
can reduce CO
2
emissions from power generating facilities via CO
2
scrubbers. In 4236, work
began on the first large-scale CCS system (Petra Nova) to be implemented in the US, for a 462
MW coal-fired facility in Texas with a projected 3.6 million tons of CO
2
capture per year [4].
However, the implementation of this technology not only has negative water use consequences
but also parasitic load requirements and efficiency reductions [42,46,58,84,;2,366,3:2]. A CCS
system typically consumes 42-52% of total electricity output and can more than double the water
requirements at a generation facility [77,339,39;].
35
4.4.4 Air quality pollutants
Air quality pollutants of concern from electricity generating facilities are primarily nitrous and
sulfur oxides (NO
x
and SO
x
, respectively). Similar to GHG emissions from power plants, these
emissions result from the combustion of fossil fuels. These air quality pollutants in their primary
and secondary (transformed) forms are harmful to both human and environmental health, con-
tributing to the formation of photochemical smog, tropospheric ozone formation, acid rain, as well
as increases in respiratory illnesses [83]. Air quality pollutants are spatially relevant on a local
or regional scale, as these pollutants are short-lived in the atmosphere (hours to days) and are
therefore not well-mixed in the atmosphere. Additionally, these pollutants are relevant on a daily
to hourly time scale, as they can undergo photochemical reactions to form secondary pollutants,
such as ozone.
Pollution controls for the reduction of air quality pollutants at generating facilities exist today,
but typically require auxiliary systems that introduce parasitic power losses and additional water
requirements [339]. Alternatively, there are generating technologies that do not use fossil fuels
and do not emit air quality pollutants to the atmosphere, but these technologies can introduce
additional water requirements for cooling compared to fossil fueled power plants (i.e. nuclear) or
introduce intermittancy and reliability concerns for the grid (i.e. renewables like solar and wind).
4.4.5 Spatial and temporal environmental trade-offs of across power
generation technologies
The environmental impacts of generating electricity are uneven across generating technologies,
time, and space. Some technologies, such as wind turbines, solar PV, and nuclear generating
facilities require no pollution controls as they do not emit pollutants at the point of generation,
but vary in water requirements [7,:4,:5,;2,338]. Other technologies can be used to reduce
water requirements for cooling near to zero (i.e. dry cooling), but introduce parasitic load losses
36
and decrease the overall efficiency of the power plant. Pollution controls such as SO
x
scrubbers
(commonly used at coal facilities to reduce SO
x
emissions) offer large reductions in emissions,
but significantly increase water consumption by up to 8: gallons per MWh (which represents a
32-49% increase in consumption for average steam-cycle coal facilities, depending on the cooling
system used) [77].
The water use and emissions implications of thermoelectric generators can cause tensions be-
tween environmental priorities and resilience to water shortages. For example, the majority of
the western US has experienced some level of drought during the past five years [325]. Prolonged
periods of heat waves, dryness and water scarcity can place stress on the operation of the grid.
Baseload generators, such as coal-fired and nuclear facilities, offer benefits to grid reliability be-
cause they are typically large capacity power plants that operate almost continuously, with the
exception of regular maintenance shut-downs (i.e., they have large capacity factors). However,
these facilities usually withdraw and consume large amounts of water. Although nuclear gener-
ation is emissions-free, most US nuclear facilities operate with a once-through cooling system,
where insufficient volumes of water or insufficiently cool water can cause curtailments in genera-
tion because the cooling system cannot reliably cool the reactor core. Additionnally, it is projected
that this emissions-free electricity source will continue to decline, largely due to the pressure from
low natural gas prices and lack of policy incentives [;8].
Concentrating solar power, which can provide stability to the grid due to storage capabilities,
is also an emissions-free generating technology, but has greater water requirements than conven-
tionalthermoelectricgenerationbecauseit’sRankinecycleoperatesatrelativelylowtemperatures
compared to fossil fuel generators and cannot discharge heat through flue gases [77]. Baseload
coal-fired generation operating with once-through cooling systems are also subject to curtailments
when water is scarce or water temperatures are high due to the thermal discharge limits imposed
by the EPA. Additionally, regardless of the type of wet cooling system used, elevated cooling
water temperatures decreases the efficiency of the Rankine cycle, requiring more fuel per unit of
37
energy generated, thus more GHG and air quality pollutants are released into the atmosphere at
fossil-fueled power plants. Dry cooling systems have been implemented for coal-fired and natural
gas-fired power plants in water-scarce regions in the US (e.g., California, Nevada) to reduce de-
pendency on freshwater sources, but these systems still require water for operation and introduce
a parasitic load to the facility, decreasing overall efficiency and increasing emissions of GHGs and
air quality pollutants. Water-lean technologies such as wind turbines and solar PV generation are
also seen as a technological solution to water availability issues, but without grid-wide available
storage facilities, the intermittency of these technologies (i.e., their reliance on meteorological con-
ditions) does not make them suitable for reliable dispatching in the absence of other generating
technologies to meet demand.
The United States represents a variety of climates and geographies across the country; thus,
the country’s electricity grid is diverse in generating technologies, fuels, and cooling systems
with diverse, and often uneven, impacts in space. For exaple, the Western half of the US repre-
sents approximately 3:% of total generation in the country, but accounts for only 6% of water
withdrawals for thermoelectric generation [374]. The majority of the country’s open-loop cooled
facilities are located in the Eastern US [7,::,355]. Consequently, thermal discharge exceedances
at once-through cooled power plants have been more of an issue for thermoelectric units across
the Eastern US region, than the Western US. Many older open-loop cooling systems have been
retrofitted to closed-loop or dry cooled systems in response to new and proposed policies in the
Southwestern US [397]. The continued development of water-intensive generating technologies in
drier regions is unlikely due to concern over water availability [364].
Different fuels and prime movers are used throughout the US as well. Often, generating
technologies are most concentrated in areas that are closest to their fuel source. For example,
most CSP facilities (;4% of U.S. facilities) are located in Arizona, California, and Nevada, where
the sunny and arid climate allows for maximum potential energy production, but water scarcity
can be an issue [38]. The differences in generating technologies can also influence water use
38
across the country, as some technologies require large amounts of water for operation (e.g. coal),
and some require almost none (e.g. solar PV and wind). On top of this, some environmental
regulations have influenced the placement of generating technologies in the US. For example,
California’s Assembly Bill 54 (AB54), which aims to reduce greenhouse gas (GHG) emissions by
52% by 4242, has already forced the extinction of coal-fired generating facilities in the state and
welcomed more natural gas combined cycle and renewable facilities [374].
Water is also used for the preparation of primary energies (fuel extraction, processing, trans-
portation) before these fuels arrive at a power plant and are used to generate electricity. At the
power plant, most water that is used is fresh surface water [363], however water used for primary
fuel preparation more commonly includes the consumption of significant volumes of groundwater
due to the geologically driven needs for water removal from sub-surface resource deposits like coal
and natural gas [75]. Water is also used for hydroelectric generation, although the quantification
of withdrawal and consumption values can be contentious [76]. Most recently, consumption from
reservoir-associated hydropower is defined as net evaporation (which is to say, the water consump-
tion that would not otherwise have occurred from the land associated with the reservoir). This
evaporation is driven by regionally variable factors like weather and land cover [76]. Therefore,
hydropower can have a much broader range of possible consumptive water intensities than other
generating technologies.
During periods of drought, areas that rely on hydropower to provide a significant portion of
generation, such as California and the Pacific Northwest, have seen increased overall generation
costs [72]. The hydroelectric system in the Pacific Northwest is particularly vulnerable to changes
in water availability since it has a much lower storage to flow ratio than other regions in the
US [79]. Due to the decrease in available hydropower, there is increasing penetration of natural
gas-fired, wind, and solar facilities to maintain the stability of the grid. In February and March
of 4236, wind generation surpassed hydroelectric generation for the first time in California [374].
During the ongoing drought in California, the decrease in hydroelectric generation has not only
39
raised electricity costs, but has also placed pressure on natural-gas fired generating facilities to
meet demand, causing an increase in statewide GHG emissions from the power sector [5;,72].
Although the tension between water availability and hydropower is a major concern for grid
operation moving forward, it is not a major focus of this work.
3:
Chapter 5
Building a database of cooling water usage rates for US
thermoelectric power plants
This chapter reflects work published in Environmental Research Letters in 4238 [5].
5.3 Motivation
There are a lack of cooling water usage data available for power plants in the literature for the
research community and decision makers, which reduces the ability to manage energy and water
priorities together. Several federal agencies publish data regarding the cooling water requirements
of power plants, but each data source has various shortcomings. Likewise, a few peer-reviewed
studies have been published but their analyses have been limited to a small sample of power
plants. Here we discuss the existing datasets available and shortcomings of each.
The US Energy Information Administration (EIA) collects self-reported data from power plant
operators across the nation and reports these data in several “Forms.” The most pertinent forms
for this study are Forms ;45 & :82, which combined contain detailed information on power
plant type, location, generation, fuel use, cooling water use, cooling water source, and cooling
water type. Power plant operators for plants with capacity 3 MW or greater are required to
report generation, fuel use, and boiler characteristics to the EIA. However, power plants are
3;
only required to report information on cooling system type, water use, water source, and water
type to the EIA if they have a capacity of 47 MW or greater. Furthermore, until 422:, the
EIA did not require nuclear power plants to report water use, despite these power plants being
collectivelyresponsibleforthelargestvolumeofwaterwithdrawalsacrossthethermoelectricpower
sector. Operators are required to report data annually to the EIA, although some operators report
monthly data that is made available on the EIA website throughout the year. In previous peer-
reviewed studies, these data have been criticized for poor quality and inconsistent reporting across
US generation technologies [;,:4]. However, since the lack of data availability surrounding water
useatthermoelectricpowerplantswasfirsthighlightedin422;bytheGovernmentAccountability
Office [3], the data available from the EIA has markedly improved. Nuclear power plants are
now required to report water use data to the EIA and the number of non-reporting facilities,
or facilities that reported a volume of zero for water use has significantly decreased. The EIA
data are difficult to use in practice as generation data are reported by unit specific prime mover,
while water use data are collected and reported according to a cooling water system identification
number. Since power plants often have multiple fuels, cooling systems, and/or prime movers,
these data, although large in number, are not straightforward to analyze, and therefore, have not
been used in many studies to date [;,:4,:6]. Despite recent efforts by the EIA to improve data
quality, no analysis has been completed to re-assess self-reported values since the 422: data were
first analyzed by Averyt et al. [;]. The researchers concluded that the 422: data were too poor
(incomplete and lacking for certain types of generators) to effectively calculate operational water
use at US power plants. However, given the improvements in the EIA’s data collection since422:,
this work reassess this rich source of data to calculate operational water use rates for hundreds of
real us power plants.
In the peer-reviewed literature, there exist only a few vetted datasets that detail the water
requirements of US power plants. Macknick et al. compiled one of the first reviews of cooling
water use rates (i.e. cooling water volume per electrical energy output) based on reported values
42
from primary literature sources [:4,:6]. This compilation of water use rates has been central to
most recent studies evaluating cooling water use at the operational phase of power production [;,
33,39,49,4:,53,56,64,94,;2,;9,364,366,367,397–39:]. It characterizes power facility cooling water
consumption and withdrawal rates based on fuel, cooling system, and prime mover configuration
for a small sample of generators (on average four facilities per technology classification) reflecting
the best available data at the time of publication [:4]. Another recent report published by the US
Geological Survey (USGS) estimated the water consumption and withdrawal rates for a large set
of power plants based on heat budget models. While the USGS dataset represents a statistically
significant sample size of power plants, water usage rates do not reflect the unique configurations
of each individual power plant and are not reported by fuel or prime mover [56].
Although there has been a growing body of analyses exploring the cooling water requirements
of the power sector in the peer-reviewed literature across various energy futures [39,49,4:,;2,329,
364,366,367,399], these studies lean almost exclusively on published water usage rates based on a
small subset of power plants. Additionally, little analysis has been done to characterize emerging
trends such as the expansion of dry-cooled and recirculating tower cooled power generation or the
use of alternative sources of cooling water, such as reclaimed water from municipal and industrial
wastewatertreatmentfacilities. Givengrowingconcernsregardingthewaterusageofpowerplants,
an updated and expanded investigation is needed.
The purpose of this study is to systematically analyze 4236 self-reported cooling water data
published by the EIA in terms of plant-by-plant water usage rates, cooling water source type
and quality, and geospatial trends in power plant cooling by watershed. The resulting vetted
database of the cooling water characteristics of hundreds of power plant facilities is available in
full upon request, offering the research community a statistically significant and geographically
diverse database of plant-specific cooling water data for US power plants.
43
5.4 Methodology
EIA Forms ;45 [378] and :82 [377] were used to characterize the cooling system and cooling
sourcesforeachpowergeneratorin4236. TheseformsaresenttooperatorsatUSpowergeneration
facilities of 3 MW or greater that are connected to a regional power grid [376]. EIA Form :82
Schedule 8D details each cooling system by type, ID number, operational characteristics and
annual cooling water usage data. Thermoelectric generators are prompted to characterize their
cooling water sources in terms of four type (i.e. surface water, groundwater, plant discharge
water or other) and five quality (i.e. brackish water, freshwater, reclaimed water, saline water
or other) classifications, respectively. Cooling data in EIA Form ;45 Schedule :D were used to
cross-check information and identify cooling system and water source when data were missing.
In cases when a generator reported no cooling source in Schedules :D or 8D, specific cooling
source names (e.g. ”Colorado River” or “wells”) reported in the EIA Form :82 Schedule 4 were
manually analyzed and recorded into the prescribed type and quality classifications. Generators
with missing cooling technology and/or cooling source data records were generally small facilities.
(Full details regarding data cleaning and assumption assignments for missing data are available in
theSI).Thedataavailableinthe4236self-reportedEIAsurveysoutnumberspreviousdatarecords
by an order of magnitude for common generating technologies, such as steam-fired, tower-cooled
coal and natural gas facilities.
Operational water use rates were also calculated using cooling water data collected through
EIA Forms ;45 [378] and :82 [377]. Each data record in Form ;45 Schedules 5A & 7A, which
provide information on annual primary energy consumption by fuel type and electricity output,
was compared to :82 Schedule :D based on power plant identification number. Each power plant
identification number associated with a facility using one type of fuel, one type of prime mover
(plus all combined-cycle facilities), and one type of cooling system was filtered into a sub-set of
generators, which were assigned a code designating fuel type, prime mover type, cooling system
44
type, and combined heat and power (CHP) status. To increase the generation available for anal-
ysis, power plants that generated over ;7% of their annual generation from coal or natural gas in
steam or combined cycle facilities were also added to this filtered sub-set. Although these facilities
generated up to 7% of their electrical output using other fuels, the impact on cooling water was
assumed to be minimal. Only coal, natural gas, and nuclear generation facilities were considered
in the water use rate analysis due to data availability constraints for other types of generation
facilities.
Generators reporting multiple fuels, prime movers or cooling systems made some data difficult
to synthesize. Although EIA Form Schedule 8A associates Boiler ID to Cooling ID for a selection
of generators, most generators were not included in this form. Thus, for most power plants with
multiple fuels, prime movers, and/or cooling systems, there was no way to disaggregate water
use by fuel and configuration (e.g. linking the specific cooling system to the specific prime mover
system). Therefore, these facilities are omitted from the final filtered water usage rate dataset. In
total, the dataset of generators represented roughly :8% of total 4236 US thermoelectric, water-
cooled generation.
The filtered data records were processed by dividing annual cooling water withdrawals and
consumption by total annual power generation to determine a final water withdrawal and con-
sumption rate, respectively. The Appendix Instructions detailed in Form ;45 Schedule :D were
followed to determine water consumption in recirculating cooled facilities, which was defined as
the volumetric difference between water withdrawals and water discharges [373]. This calcula-
tion was only performed on recirculating facilities that reported values for water withdrawal and
discharge, but not for consumption. Generally, these facilities were cooled using recirculating
pond(s)/canal(s). Outliers were detected using a modified Z-score, based on methods described
by Iglewicz and Hoaglin [82]. Although zero-values were not considered outliers if the absolute
value of the Z-score was less than 5.7, no zero values met this criteria. Thus, generators that
45
reported zero-values or no cooling water data were discarded from the final filtered dataset. Confi-
denceintervals wereused asan alternative to standard deviation to describe the probable bounded
region within which the true values of the cooling water estimates were located [324].
5.5 Results and Discussion
Table 5.3: Cooling system technologies were characterized for all 4236 US electricity gen-
erators reporting to the EIA (listed from most to least water withdrawal intensive, on
average). Eighty-six percent of this generation was produced in thermoelectric power facil-
ities requiring a cooling system. Wet recirculating tower cooling systems are now utilized
more than any other type of cooling system.
Cooling System Type 4236 Generation (billion
kWh)
No Cooling 786± 4:.4 35.:%
Dry (air) cooling System 338± 7.;4 4.:6%
Hybrid: recirculatingwithforceddraftcoolingtower(s)withdry
cooling
:.4;± 2.636 2.425%
Hybrid: recirculating with induced draft cooling tower(s) with
dry cooling
:.:2± 2.749 2.437%
Recirculating with Induced Draft Cooling Tower 3622± 94.; 56.4%
Recirculating with Natural Draft Cooling Tower 723± 47.7 34.4%
Once through with Cooling Ponds 327± 8.78 4.79%
Recirculating with Cooling Ponds 548± 38.: 9.;9%
Once through without cooling pond(s) or canal(s) 3286± 77.3 48.2%
Total 62;5 322%
Table 5.3 summarizes total 4236 US power generation by cooling system. Over half of all
thermoelectricpowergenerationwascooledwithwetrecirculatingcoolingtowers,whileabout64%
was produced in facilities utilizing once-through cooling or some type of cooling pond. Dry-cooled
facilities generated nearly 5% of total US thermoelectric generation, nearly three times previous
estimates in the literature [65]. Nevada, California and New York represented approximately38%,
37%, and 35% of total US dry-cooled generation, respectively (Figure 5.3). Much of California’s
coastal generation seen in Figure 5.3 recently switched from once-through cooled facilities using
saline water to dry cooled facilities using reclaimed water because of new regulations [7;]. The
average cooling system in-service year for power plants listing their primary cooling technology
46
as once-through without cooling ponds, induced draft cooling towers, or dry cooling was 3;85,
3;::, and 4226, respectively, confirming the general shift towards lower withdrawal systems over
time [377].
While fresh surface water represents the majority of US cooling water, reclaimed water is
used to cool nearly 8% of thermoelectric generation (See Table 5.4). These facilities are generally
locatednearbigcitieswhereeffluentfromwastewatertreatmentorindustrialfacilitiesisabundant.
Although reclaimed water is often utilized within dry cooled generation facilities, this trend is not
captured in Table5.4 since dry-cooled facilities were not consistent in reporting a cooling source of
water versus air. However, some of this geospatial coupling is observed in Figure 5.3 by comparing
theHUC-:subbasinsmappedonthelowertwomapsintheFigure, especiallyinCalifornia. Plants
using saline or brackish surface water are generally older once-through facilities, as water with
high total dissolved solids causes fouling in cooling towers.
Table 5.4: Cooling sources were characterized for all US 4236 electricity generators reporting to
the EIA. Nearly 97% of US generation produced in power facilities requiring cooling utilized fresh
surface water.
Water Source Source Type 4236 Generation (billion kWh)
Surface Water
Freshwater 4848± 392 86.4%
Brackish 363± :.59 5.66%
Saline 384± :.2; 5.;8%
Groundwater
Freshwater 452± 4;.4 7.85%
Brackish 5.58± 2.38: 2.2:4%
Saline 6.55± 2.439 2.328%
Recycled 5.;4± 2.3;8 2.2;8%
Other 8.39± 2.52; 2.373%
Plant Discharge Recycled 424± 33.9 6.;6%
Other
Freshwater 32.;± 2.766 2.488%
Saline 2.472± 2.234 2.228%
Other 44.9± 3.54 2.775%
Dry-cooled Dry-cooled
a
338± 7.63 4.:6%
No Cooling No Cooling 786± 4:.4 35.:%
Total 62;5 322%
a
Although dry-cooled facilities consume approximately 32% of a
wet-recirculating tower facility [33:], generators were not consistent
in reporting cooling water sources and are not classified in this
table.
47
Figure 5.3: Total 4236 generation (top) and total generation by each respective cooling tech-
nology/source are aggregated in each map across each HUC-: watershed. Once-through cooled
facilities are concentrated in the eastern US and coastal regions where water is generally abun-
dant. Water constrained locations typically use recirculating cooling towers, which avoid large
water withdrawals at the expense of higher water consumption. Dry cooled generation and gen-
eration cooled with reclaimed water have expanded in recent years, lowering both freshwater
consumption and withdrawals compared to other cooling systems.
48
Figure 5.4: Total cooling water withdrawal and consumption volumes utilized by thermoelectric
power generators in 4236, aggregated here by HUC-: watershed, reflect cooling system technology
trends.
The geospatial distributions of water withdrawals and water consumption in Figure 5.4 reflect
cooling system configurations. Once-through cooled facilities that require large flow rates are
typically located on larger rivers or coastal locations, where water availability is high. Thus, most
once-through cooled capacity is located in the water-rich eastern US (Figure 5.3).
Median, minimum, and maximum water use rates (Table 5.5), in addition to full statistical
distributions(Figure5.5), werecharacterizedforallavailabletechnologyconfigurationsconsidered
in this study. The number of power plants analyzed in each category, n, reflects the total number
of data points, n
T
, less outliers, n
o
, and zeroes, n
z
(i.e.n =n
T
-n
o
-n
z
) (Table5.5). In total, 866
water withdrawal rates and 6;; water consumption rates (including all non-zero and non-outlier
values) were calculated in this analysis, representing data from a total of 894 unique power plants.
The majority of non-reporting facilities were smaller than 3222 MW or had low net generation in
4236 (Figure 6). Larger plants that were excluded from this analysis were mostly power plants
using both coal and natural gas. Approximately ;4%, ;2%, and :7% of total nuclear, coal,
and natural gas-fired generation in 4236, respectively, was classified with a single prime mover,
fuel, and cooling system and reported non-zero and non-outlier values for either cooling water
withdrawals or consumption. Collectively, these plants represent :8% and :6% of once-through
and recirculating cooled generation in 4236, respectively.
49
Table 5.5: Water withdrawal and consumption factors for electric generation units in the US based on calculations using self-reported
cooling water data from 4236 EIA forms ;45 and :82 [377,378].
a
Cooling Fuel Prime CHP Water Withdrawal (Gal/MWh) Water Consumption (Gal/MWh)
System Mover Status Median Min. Max. n nz no n
T
Median Min. Max. n nz no n
T
Once through
with ponds
Nuclear Steam N 54,597 54,597 54,597 3 2 2 3 373 373 373 3 2 2 3
Coal Steam N 52,68; 45,587 66,848 32 3 2 33 562 3;2 ;84 7 2 8 33
Natural Gas Steam N 364,975 93,92: 4:2,94; 6 2 2 6 694 587 79; 4
b
2 4 6
Natural Gas Combined Cycle N 62,2;4 46,33; 92,542 8 2 2 8 3,42; 457 4,3:6 4
b
2 6 8
Once through
without ponds
Nuclear Steam N 59,;46 43,436 78,935 4: 4 2 52 585 4: 3,398 7 3 46 52
Coal Steam N 63,328 788 ;6,4;: 334 9 2 33; 426 2.3 3,238 64 4 97 33;
Coal Steam Y 82,;62 52,:62 342,;85 9 2 2 9 3,;;9 3,;;9 3,;;9 3 2 8 9
Natural Gas Steam N 33:,6;2 34,6:: 643,6:; 52 9 2 59 547 87 845 : 3 4: 59
Natural Gas Combined Cycle N 4:,229 37,2;; 89,563 42 3 3 44 3:: 4.6 595 7 2 39 44
Natural Gas Combined Cycle Y 62,;52 45,746 7:,557 4 2 2 4 - - - - - - -
Recirculating
ponds
Nuclear Steam N 53,7:; 43,864 67,649 9 2 2 9 72; 594 867 4
b
2 7 9
Coal Steam N 57,55: 46,365 74,586 3: 6 2 44 58: 2.5 3,625 3: 3 5 44
Coal Steam Y 49 49 49 3 2 2 3 - - - - - - -
Natural Gas Steam N 36;,66; 4.8 735,9:3 32 2 2 32 539 6.; ;46 : 2 4 32
Natural Gas Combined Cycle N 8,259 96 67,797 : 3 2 ; 37: 75 626 8 2 5 ;
Natural Gas Combined Cycle Y 35,36; 35,36; 35,36; 3 2 2 3 3: 3: 3: 3 2 2 3
Recirculating
Tower (induced
draft)
Nuclear Steam N 3,372 97: 4,697 8 3 2 9 97: 839 :;; 9 2 2 9
Coal Steam N 75; 427 3,328 93 36 2 :7 6:9 53 3,:42 :4 5 2 :7
Coal Steam Y 768 374 3,7;6 35 5 2 38 6:9 348 3,6:; 36 3 3 38
Natural Gas Combined Cycle N 487 37 72: 378 3: 7 39; 439 52 5;4 374 3; : 39;
Natural Gas Combined Cycle Y 429 36 69: 4: 5 4 55 3:5 3.2 654 4; 4 4 55
Natural Gas Combined Cycle Single Shaft N 449 3; 688 9 3 2 : 427 ;5 642 8 3 3 :
Natural Gas Combined Cycle Single Shaft Y 495 495 495 3 2 2 3 45: 45: 45: 3 2 2 3
Recirculating
Tower (natural
draft)
Nuclear Steam N 3,526 9;7 4,:43 35 3 2 36 894 747 :95 35 3 2 36
Coal Steam N 829 46; 3,673 43 5 2 46 626 3;: 3,:97 46 2 2 46
Natural Gas Steam N 3,457 346 5,::5 46 5 3 4: :55 86 4,44; 48 3 3 4:
Natural Gas Combined Cycle N 489 489 489 3 2 2 3 43: 43: 43: 3 2 2 3
Hybrid
Coal Steam N 692 692 692 3 2 2 3 5;4 5;4 5;4 3 2 2 3
Natural Gas Combined Cycle N ;5 :9 ;: 4
b
2 3 5 ;4 :8 ;: 4
b
2 3 5
Natural Gas Combined Cycle Y 547 547 547 3 2 2 3 422 422 422 3 2 2 3
a
n: number of data points, excluding outliers and zero-values; nz: number of data points reporting zero-values; no: number of data points classified as outliers according to calculated modified Z-scores
(see B for full table including outliers.); n
T
: total number of data points. Median, minimum, and maximum values reflect n filtered values. Facilities with multiple cooling systems, prime movers,
and/or fuels are not included in the final filtered dataset.
b
Median reflects average of Min and Max values.
4:
The influence of fuel, prime mover, and cooling system on cooling water usage rates are evident
in Figure 5.5. The difference between water withdrawal rates for different cooling systems differed
by up to three orders of magnitude, while the difference between consumption rates was typically
within one order of magnitude. Generally, once-through cooled facilities reported the highest
waterwithdrawalrates, whiletherecirculatingtower-cooledfacilitiesreportedhigherconsumption
rates, in comparison. A regression analysis was performed to investigate the role of generation
unit efficiency on water usage. Results did not show a strong correlation between average water
withdrawal and consumption rates and monthly average heat rate. However, results were very
dependent on accurate monthly primary fuel usage, generation and volumetric water usage data;
evensmallmarginsoferrorinreportingwouldbeexpectedtoweakentheregressionanalysis. (Full
details of this regression analysis are available in the SI document.) Additionally, the influence of
CHP and pollution controls on water usage rates were evaluated but data were not sufficient to
draw meaningful conclusions.
Results of the EIA data analysis were compared to published cooling water usage rates in
recent years. Calculated median water withdrawal and consumption rates for most recirculating
tower-cooled facilities were comparable to values presented by Macknick et al. [:4,:6] and heat
budget models from the USGS [56]. There were also a large amount of data available for these re-
circulating tower cooled facilities, increasing the value of the resulting distribution. (Quantitative
comparisons to previous studies are available in the SI document.)
Calculated water usage values for once-through facilities were also similar to previous analyses;
however, there were cases in which calculated water withdrawal and/or consumption rates were
one or more orders of magnitude higher than expected. Figure 5.6 illustrates the distribution of
calculatedmonthlywithdrawalandconsumptionratesversusmonthlygenerationforonce-through
cooled generation units considered in the analysis. Calculated water use rates for a subset of these
generators were markedly higher than values reported in prior review and heat budget studies.
In general, these generators with water withdrawal or consumption rates characterized as outliers
4;
Figure 5.5: Distributions of 866 water withdrawal rates and 6;; water consumption rates, char-
acterized by cooling system, fuel and prime mover configuration, represent data from a total of
894 unique power plants (excluding power plants reporting outliers and zero-values). Box-and-
whisker plots showcase the quartile distribution of data (points shown to the left of the plots)
for each generating technology analyzed. Minimum, maximum, and median values are identified
on box-and-whiskers as horizontal lines; mean values are identified as hollow squares. Fuel Coal
(CL), Natural Gas (NG), Nuclear (NUC); Prime Mover: Steam Turbine (ST), Combined Cycle
(CC), Combined Cycle Single-Shaft (CS); Combined heat and power (CHP): Yes (Y), No (N)
(shown with hollow shapes in Figure 5.6) had low capacity factors, were constructed pre-3;92 and
were located on large bodies of water (e.g. rivers with very fast flow such as the Mississippi River
or the Pacific Ocean). Thus, these high water use rates were most likely driven by large incoming
flows of water used to cool small amounts of generation in infrequent intervals. This insight points
to the importance of considering water usage on a plant by plant basis, as averages might not
correctly characterize water usage at power plants with unique locations or operating conditions.
Such underestimations could lead to an underestimation of ecosystem impacts associated with
52
high flow rates (i.e. through entrainment or entrapment) or thermal pollution. Most once-through
cooled natural gas steam cycle generators, in particular, had calculated values larger than322,222
gal/MWh, which were far higher than previous studies.
Figure 5.6: Monthly average water usage rates versus monthly generation for the subset of once-
through cooled power plants suggest that older units with low capacity factors located on large
bodies of water often report very high water withdrawal and/or consumption rates.
Water withdrawal rates for generating units using recirculating pond/canal or once-through
pond/canal cooling systems were generally much greater than values reported by Macknick et
al. [:4] and Diehl and Harris [56], whereas reported non-zero consumption rates were similar in
value. Diehl and Harris (4236) make a clear distinction between recirculating and once-through
cooling ponds or canals. They define a once-through pond system as a large reservoir, typically
located within a large watershed that receives enough natural runoff to maintain normal flow
rates, while a recirculating pond system is typically within a smaller watershed with little to no
runoff [56]. Once-through systems would be expected to have higher withdrawal rates and lower
consumption rates than recirculating systems based on this distinction. However, the calculated
median water use rates from EIA data for all generating systems using ponds/canals do not show
noticeable differences between open-loop and recirculating systems. Although the EIA provides
53
an “Appendix for Schedule :D” [373] to plant operators to supplement ;45 instructions, there are
evident inconsistencies in reported cooling system definitions.
5.6 Conclusion
This study characterized US cooling water usage trends in 4236 based on self-reported cooling
water data from thermoelectric power generation units, published in EIA’s 4236 ;45 and :82
forms. Results indicate that shifts in the power sector toward wet recirculating cooling towers, dry
cooling, and reclaimed water use in recent years. Water withdrawal and water consumption rates
were calculated according to fuel, prime mover, and cooling system classification for 894 unique
power plants, which is an order of magnitude larger than available water usage rates published
in the literature based on real power plant data. While some reported data are incomplete or
erroneous,resultssuggestthatwaterusageratesatpowerplantswithuniquelocationsoroperating
conditions might not be accurately characterized by “average” facilities. Although characteristics
such as power plant efficiency, and consequently, pollution controls and combined heat and power
configurations would be expected to affect the water usage of power plants, the 4236 EIA dataset
analyzed was not sufficient to draw meaningful conclusions about these characteristics, so future
research should be dedicated to analyzing the impacts of more complex power plant configuration
on cooling water usage.
Water withdrawal and consumption rates calculated for power plants cooled with wet recir-
culating towers in this analysis were particularly close to values presented in the literature and
calculated in heat budget models [56,:4,:6], suggesting that there is a good understanding of the
water requirements of recirculating cooling systems. Some calculated water withdrawal and/or
consumption rates were one or more orders of magnitude higher than estimates in the literature
for some technologies, notably natural gas steam cycle facilities with once-through cooling. These
high water use rates could be the result of large incoming flows of cooling water used for small
54
amounts of generation in infrequent intervals. This new insight highlights the need for building
a more robust understanding of the cooling water usage of power plants with unique locations
and/or operating conditions.
55
Chapter 6
Quantifying the impacts of recent power sector transitions
on cooling water usage at US thermoelectric power plants
This chapter reflects work published in Applied Energy in 4239 [335].
6.3 Motivation
Recent shifts in resource availability, economics, environmental policy and public opinion have
prompted large transitions in the US electricity generation fuel mix [8;,33:]. Between 4227
and 4237, domestic natural gas production increased by almost 62% largely due to advances in
horizontal drilling and hydraulic fracturing techniques used for US shale gas extraction, putting
downward pressure on natural gas fuel costs and prompting large investments in natural gas-fired
generation units [379]. This growth in natural gas-fired generation, as well as renewable electricity
in recent years, has reduced the competitiveness of coal-fired and nuclear power plants in many
US regions.
These technological and market transformations across the power sector have translated into
environmental consequences that have yet to be quantified. Although a growing body of literature
has addressed the emissions ramifications of increasing natural gas-fired and decreasing coal-fired
generation [8;,92,:3,3:3], much less analysis in the literature has been dedicated to assessing how
56
recent fuel transitions in the power sector have affected US water availability or water quality at
the national level. Recent studies have analyzed the cooling water trade-offs that follow more
general shifts in fuel use [77,323,334], pollution controls [3;,329,366,39;,3:2], cooling system
technologies [356–358,365], environmental fees [33;], and generator dispatch order [:,32;,333].
Grubert et al. (4234) completed a detailed comparison of the water intensity of natural gas and
coal extraction, cooling for electricity production, and emissions controls at fossil-fueled power
plants in Texas using the peer-reviewed literature and government data. The researchers note that
the efficiency benefits of switching coal-fired power plants to natural gas combined cycle offer the
potential for a 82% reduction in annual freshwater consumption, even given the water-intensity
of hydraulic fracturing for the natural gas fuel [77]. Stillwell et al. (4233) assessed the reduction
in water diversions for thermal power plant cooling in Texas from switching traditional once-
through cooling systems to alternatives such as recirculating towers or dry cooling using a water
availability model from the Texas Commission on Environmental Quality. The authors noted
potential reductions in annual diversions up to 922 million m
3
from switching from coal-fired to
natural gas-fired combined cycle power plants, which could contribute to increased stream flow
and reduced water stress along the Texas Gulf in particular [356]. Another study by Tidwell et
al. (4236) assessed the transition of the whole US thermoelectric fleet to alternative cooling water
sources (dry cooling or wet cooling using reclaimed water) to achieve zero freshwater withdrawals
using a custom algorithm incorporating cost models, geographic proximity to water resources,
and resource availability. The results suggest that retrofits could be beneficial in the East by
reducing plant vulnerabilities to thermal discharge limits and in the West by reducing freshwater
consumption during times of drought or reduced water availability [365]. Similarly, a case study
by Stillwell and Webber (4236) investigated the potential of utilizing reclaimed water as a cooling
source for thermoelectric power plant cooling in Texas using a geospatial multi-criteria analysis.
They found that over 82% of thermoelectric capacity in the state is located within 47 miles
57
of a reclaimed water source and could be feasibly retrofitted to help alleviate water availability
concerns [359].
A recent body of work has also evaluated the long-term water use impacts of various electricity
futures. In 422:, the Department of Energy (DOE) completed a report estimating the volumes
of freshwater required to meet future electricity demand based on five scenarios defined in the
US Energy Information Administration’s (EIA) 422: Annual Energy Outlook forecast. With the
exception of the business-as-usual scenario (i.e. no changes), all case studies showed decreases
in water withdrawals and increases in water consumption for thermoelectric cooling, largely due
to transitions away from once-through cooling and towards recirculating cooling [54]. Clemmer
et al. (4235) modeled low-carbon electricity futures through 4272 using the Regional Energy
Deployment System (ReEDS) model developed by the National Renewable Energy Laboratory
(NREL) to calculate changes in national and regional cooling water use, finding that investments
in energy efficiency and renewable energy technologies resulted in considerable water savings over
other technology-based investments, such as carbon capture and sequestration [49]. Another
study modeled changes in cooling water usage using a GAMS optimization model to estimate wa-
ter withdrawals and consumption at thermoelectric, non-thermoelectric, and dry-cooled facilities
based on energy portfolio scenarios developed by NREL for high renewables penetration and a
scenario retrofitting all existing wet cooling systems to recirculating cooled systems through4272.
The study found that significant water withdrawal and consumption reductions are achieved un-
der the high renewable energy scenario, while only water withdrawal reductions are achieved in
the second scenario but at the expense of increased water consumption [35:]. In another study
that evaluates changes in the electricity fleet through 42;7 using an integrated assessment model
(GCAM) to investigate the electric sector’s global water demand, water withdrawals remained
relatively constant over the five scenarios examined (i.e. three climate change futures and two
strategic technology improvement scenarios), mainly due to the retirement of once-through cool-
ing systems [94]. The water use implications of a global 4
o
C warming policy (by end of century)
58
were analyzed by Fricko et al. (4238) using a global integrated assessment model. The authors
found that noticeable reductions in water withdrawals are achieved if large transitions toward
recirculating cooling systems occur, but water consumption increased for all electricity futures
analyzed [68]. On a smaller spatial scale, the influence of 4
o
C of warming, prolonged drought,
and population growth on water use until mid-century in the southwestern US showed a continued
or increased reliance on fossil fuels in the business as usual and Annual Energy Outlook scenar-
ios, leading to greater water stress. Conversely, carbon policy, renewable energy integration, and
increased energy efficiency led to decreased water stress and carbon emissions [397].
Despite the large changes that have occurred to the US generation fleet recently, no study to
the authors’ knowledge has evaluated the cooling water trade-offs resulting from these transitions
at the national scale. This research fills this knowledge gap by evaluating how recent shifts in
thermoelectric power generation affected the spatial and volumetric distribution of US cooling
water withdrawals and consumption between 422: to 4236.
6.4 Methodology
Self-reported data by power plant operators from EIA forms ;45 [372,378] and :82 [36;,377]
were used to characterize US power plants and their respective generation units in the years 422:
and 4236. Power plant operators are required to complete these forms for all plants of 3 MW
capacity or greater that are connected to a regional power grid [59].
EIA Form :82 details power plant locations (i.e. latitude and longitude), as well as power
plant cooling system information including cooling system ID number and cooling water source
type (i.e. surface water, groundwater, plant discharge water, etc.). In some cases, cooling water
source type data were missing, but information was available on the physical source (i.e. wells,
rivers, ocean, etc.), which enabled an adequate estimation of cooling water source type for many of
these plants. Information on cooling water quality (i.e. freshwater, reclaimed water, saline water,
59
etc.) was only available for 4236 power plants. Although all power plant operators are required
to report generation, fuel use, and boiler information for generating units with capacity 3 MW or
greater, they are not explicitly required to report volumetric water usage via the EIA :82 form
unless they have a capacity of 322 MW or greater. While annual cooling water usage data in 4236
were relatively abundant, these data for 422: are considerably less complete [36;,372,377,378].
In addition, there is no streamlined methodology imposed upon power plant operators for data
collection to ensure consistent reporting of water use. Consequently, many facilities use different
methodologies for measuring water withdrawals, consumption, diversions, and discharge [32].
EIA Form ;45 details electricity generation unit technology, fuel type, combined heat and
power (CHP) status, and annual generation for operational units at each US power plant. When
applicable, this form was also used to cross-check and identify missing cooling system and wa-
ter source data from the EIA :82 form. Each unit operating at a thermoelectric power plant
requiring a cooling system was categorized by fuel type, generation technology, CHP status, cool-
ing technology, and cooling water source type. Full details of this categorization procedure are
documented in Peer and Sanders (4238) [5].
Plant-specific cooling water consumption and withdrawal factors (i.e. rates in gallons/MWh)
calculatedbyPeerandSanders[5]usingEIA’s4236waterusagedatawereappliedtopowerplants
based on generator technology (i.e. fuel type, prime mover type, and cooling system type) when
all units within the plant reported a single fuel, prime mover, and cooling system. These water
use factors were applied consistently to power plants that were operating in both 422: and 4236
and/or only in 4236 (i.e. new power plants). The Union of Concerned Scientists’ (UCS) vetted
database of 422: water use at thermal power plants were used to calculate total water withdrawal
and consumption volumes for power plants that were operating 422: but had no reported 4236
generation or water use (i.e. units that were retired before 4236) [36:].
There were cases when assumptions and approximations had to be made due to data issues in
the EIA databases. These typically resulted when power plants were missing cooling-related data
5:
or reported multiple fuels, prime movers or cooling systems. (Collectively, the excluded power
plants from this analysis represent approximately 32% of thermoelectric generation requiring
cooling for both 422: and 4236). The following rules were followed when possible to fill in data
gaps.
• New power plants less than 322 MW (less than 2.5% of annual 4236 generation) in capacity
were not assigned a water use rate.
• For power plants that reported a single fuel and prime mover, but did not specify the type
of recirculating tower (i.e. natural draft, forced draft, induced draft) or cooling pond (i.e.
once through or recirculating), an average of the water use rates calculated by Peer and
Sanders (4238) for all recirculating cooling tower or cooling pond types was applied.
• Power plants that did not report a cooling system in 4236, but were operating in 422: were
assumed to have maintained the same cooling system throughout the time period.
• Power plants that reported using a dry-cooling system were assigned consumptive water use
rates equal to 32% of a recirculating tower cooled plant’s water use rate for its respective
fuel and prime mover category based on the literature, as power plant operators were very
inconsistent in reporting water use for these facilities.
• Water use rates from the literature were applied to geothermal and concentrating solar
power (CSP) for both 422: and 4236 based on [:4], as these generators were not consistent
in reporting.
• For power plants that reported a single prime mover and single cooling system, but reported
multiple fuels, the median water use rate consistent with their main fuel source (i.e. coal or
natural gas) was applied.
• Custom water use rates were also calculated with reported EIA data for water-cooled power
plants with multiple cooling systems that had only one fuel and one prime mover type. The
5;
custom calculated water use rates for 4236 were applied to power plants reporting multiple
cooling systems for both 422: and 4236 and/or only in 4236. For power plants reporting
multiple cooling systems only in 422:, a custom calculated rate based on 422: data was
applied. Power plants with multiple cooling systems that omitted water use data or reported
a withdrawal or consumption volume of zero were unable to be classified and were excluded
from this analysis.
• Power plants that reported multiple cooling systems and fuels, multiple fuels and prime
movers, multiple prime movers and cooling systems, or multiple fuels, prime movers, and
cooling systems that could not be otherwise classified (via a custom rate if water use was
reported or a water use rate from the literature) were excluded.
The EIA did not require reporting water use for nuclear facilities in 422:. Given the large
water implications of these generators, they were handled carefully on a case by case basis. For
nuclear power plants reporting multiple cooling systems in 422:, the median water use rate
consistent with the reported 4236 cooling system from Peer and Sanders (4238) was applied.
Custom water use rates based on plant-specific EIA reported water use data were calculated for
nuclear power plants that reported multiple cooling systems in 4236, to reduce error, as nuclear
power plants were often the most significant cooling water users across an individual watershed.
A geographic inspection to confirm cooling system type with Google Earth was performed for new
power plants added to the grid with capacity greater than 322 MW that did not report a cooling
system.
The geographic analysis of water use for thermoelectric power plants was performed using :-
digitHydrologicUnitCode(HUC-:)watershedareas. Eachthermoelectricpowerplantwaslinked
to its respective HUC-: code based on its geographic location (i.e. latitude, longitude) provided
in EIA form :82. Changes to net generation, capacity, and water withdrawal and consumption
62
between422: and4236 were evaluated for each basin containing at least one thermoelectric power
plant requiring cooling.
The equations used to calculate changes in generation, installed capacity, water withdrawal,
and water consumption are defined in Equations 6.4 - 6.4. X represents annual generation,
nameplate capacity, water withdrawal, or water consumption, depending on the metric of interest.
y represents the 3: categories for fuels, cooling systems, and cooling water with which the data
were classified (7 fuel, 9 cooling system, and 8 cooling water categories). ΔX
net,y
is defined as
“net change between 422: and 4236 for category y”, meaning any changes to X regardless of
whether the change is a result of (3) a new power plant, (4) a power plant that was retired, or
(5) changes in total annual generation at an existing power plant in the year 422: as compared
to 4236. On the other hand, ΔX
pp,y
, only accounts for changes in X that result because of a
newly installed power plant, added at any time from the beginning of 422; to the end of 4236,
or the loss of generation from a power plant that retired any time during 422: up to the end
of 4235. (ΔX
pp,y
would not include individual units that are added or retired at power plants
with other units operational in both 422: and 4236.) ΔX
net,y
and ΔX
pp,y
only include changes
to all thermoelectric units requiring cooling. Equations 6.4& 6.4 were used to calculate values in
Table 6.3 and Equations 6.4 & 6.4 were used to calculate watershed specific results, where the
subscript ws refers to a HUC-: watershed.
ΔX
net,y
= Σ
n
i=1
X
2014,y
− Σ
n
i=1
X
2008,y
(6.3)
ΔX
pp,y
= Σ
n
i=1
X
add2009−2014,y
− Σ
n
i=1
X
ret2008−2013,y
(6.4)
ΔX
net,ws,y
= Σ
n
i=1
ΔX
net,y
(6.5)
63
ΔX
pp,ws,y
= Σ
n
i=1
ΔX
pp,y
(6.6)
A water stress metric developed by the Aqueduct Water Risk Atlas [394], describing overall
globalwaterriskwasusedtogivefurthercontexttotheimpactsofatransitioningelectricitysector
on water resources in the US. The metric reports an aggregated risk value (i.e. low risk, low to
medium risk, medium to high risk, high risk, and extremely high risk) based on twelve water-
stresscategoriesincludedinphysicalwaterquantity(baselinewaterstress, inter-annualvariability,
seasonalvariability, floodoccurrence, droughtseverity, upstreamstorage, andgroundwaterstress),
physical water quality risk (return flow ratio, and upstream protected land), and regulatory &
reputational risk (media coverage, access to water, and threatened species). This metric has also
been applied in recent energy-water studies [363].
6.5 Results and Discussion
A summary of net changes in the US power sector for generation, capacity, water withdrawal
volume,andwaterconsumptionvolumeacrossallUSthermoelectricpowerplantsrequiringcooling
systems (i.e. including steam cycle and gas cycle units, but excluding pure gas turbine units) is
shown in Table 6.3, separated by fuel type, cooling system technology, and cooling water type.
The“NetChangesfrom422:-4236”columns, calculatedwithEquation3, reflect(3)fullyretiredor
newly installed power plants, (4) added, retired, or retrofitted generation units at existing power
plants, and (5) changes incurred by differences in generation across the two years of study. The
“Changes from new or retired power plants” columns, by contrast, are calculated with Equation
4 and only reflect changes in (3) and are detailed in Figure 6.5 below. It should be noted that
generation and capacity data in Table 3 is representative of all US thermoelectric power plants
requiring cooling systems, regardless of whether or not they reported data to the EIA. However,
64
the cooling water withdrawal and consumption estimates in Table 3 are only representative of
cooling water used power plants included in the analysis.
65
Table 6.3: Net changes across the US Power Sector between 422: and 4236 were characterized for 3,:58 thermoelectric power plants requiring
cooling systems in the US, as well as the subset of changes resulting exclusively from the installation and retirement of 386 and 375 unique
thermoelectric power plants requiring cooling systems that were newly installed and retired, respectively. *Surface water does not include
ocean water. **Totals for thermoelectric power plants with cooling systems only
Net Changes from 422:-4236 Net changes across power plants in-
stalled and retired from 422:-4236
Generation
(TWh)
Capacity
(GW)
Withdrawn
water
(Billion
gallons)
Consumed
water
(Billion
gallons)
Generation
(TWh)
Capacity
(GW)
Withdrawn
water
(Billion
gallons)
Consumed
water
(Billion
gallons)
Fuel type
Coal -58: -8.3; -9332 -373 -2.:42 -6.42 -:8.7 -;.6:
Natural Gas 875 97.7 3972 32: ;3.: 37.8 -385 35.4
Nuclear 32.6 -2.3:4 3382 -7.43 -34.3 -4.:3 -758 -5.47
CSP 3.88 2.884 2.847 2.847 3.22 2.872 2.248 2.248
Other -332 -63.4 -43:4 -4.4; -9.69 -6.;4 -5:8 -3.3;
Cooling
system
type
Once-through 93.7 7:.7 858 54.8 -67.6 -39.; -3952 -:.;:
Recirculating tower 637 9;.4 45.7 85.3 89.2 36.2 -42.5 32.4
Cooling ponds 46.8 39.4 -78; -6.57 39.9 2.337 796 -4.72
Hybrid cooling 37.; 5.5; 7.45 6.46 4.95 2.876 2.::8 2.766
Dry cooling 82.: :.8; 2.3:5 2.367 52.6 9.5: 2.338 2.326
Complex -5;: -357 -8692 -368 - - - -
Not Reported -4.95 -5.26 - - - - - -
Cooling
water type
Surface water* 339 84.8 -348 -54.4 44.3 -6.74 -648 -8.42
Ocean water -3;2 -83.5 -92;2 -6:.: -35.3 -8.98 -8;7 -5.22
Groundwater ::.5 9.;3 -33.5 32.4 :.37 4.28 -47.; 2.943
Reclaimed water 7:.4 -38.5 -3:2 7.57 68.4 33.2 -45.8 9.96
Dry cooling 5;.: 33.5 2.273 2.275 :.76 4.27 2.253 2.253
Multiple types 95.8 46.7 3252 37.2 2.655 2.572 -2.243 -2.225
Totals** 3:9 4:.8 -85:2 -72.6 94.6 6.53 -3392 -2.8;4
66
Net electricity generation from thermoelectric power plants requiring cooling systems grew by
3:9 TWh between 422: and 4236 (7.;2% increase from 422: thermoelectric generation requir-
ing cooling). Generation from new net capacity installations (including units added at existing
facilities as well as new facilities) surpassed losses in generation and capacity from retired units
during this time, which is expected as newer power plants are generally more efficient than older
units, and therefore, are often running more frequently. Net generation from natural gas com-
bined cycle facilities increased approximately 942 TWh from 422: to 4236 across the US, while
capacity additions were 329 GW, prompting decreased use of coal-fired generators. (The increase
across the entire natural gas category shown in Table 3 is tempered by retirements of natural
gas steam cycle generators.) Nuclear retirements occurred, but net generation increased (due to
higher capacity factors for operating facilities) during this period. Net negative generation in the
“Other” category reflects the decreased use of oil-, bio-, and multi-fueled (i.e. combination of coal,
natural gas, various types of oil, jet fuel, etc.) units.
Net changes in US electricity generation and power generation capacity between 422: and
4236 were calculated with Equation 3 and are shown in Figure 6.3A & B, respectively. The subset
ofchangestoUSelectricitygenerationandpowergenerationcapacityresultingfromneworretired
power plants during this period (i.e. Equation 4) are illustrated directly below, in Figure 6.3C &
D, respectively.
The largest capacity retirements for thermoelectric generation requiring cooling were in Texas,
California, Pennsylvania, Florida and Arizona, which retired 8.9, 8.3, 7.2, 5.9 and 5.3 GW of ca-
pacity, respectively, between 422: and 4236 [36;,377]. Collectively, these retirements represented
6;% of US retirements during this period and were comprised of coal, natural gas, and nu-
clear steam-fired retirements, primarily. The largest capacity additions were in Texas, California,
Florida, North Carolina and Louisiana, which added 34.:, ;.:, 7.;, 6.2, and 5.; GW, respectively,
between422: and4236 [36;,377]. In total, these states installed about68% of new thermoelectric
67
Figure 6.3: Left: (A) Net changes in US thermoelectric generation requiring cooling and (C)
the subset of these changes to US electricity generation resulting exclusively from new or retired
power plants between 422: and 4236. Right: (B) Net changes in US power generation capacity
and (C) the subset of these changes to US power generation capacity resulting exclusively from
new or retired power plants between 422: and 4236. These maps only include units that are
greater than 3 MW.
0 370 740 1,110 1,480 185
Miles
$
< -3.5 -3.5 - -6.5 -0.65 - 0 0 - 15 > 15
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE
EEE EEEE
EEEE
EEEE
EEE
EEE
EE
EE
EE
EEE
EEE
EE
EE
EE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EEE
EEE
EE
EE
EE
EEE
EEE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEEE
EEEE
EEEE
EE
EE
EE
EEE
EEE
EEE
EE
EE
EE
EE
EEE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EE
EE
EE EEE
EEE
EEE
EEE
EEE
EEEE
EEEE
EE
EE
EE
EE
EE
EE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EEEE
EEEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EE
EE
EE
EE
EE
EEEE
EEEE
EEEE
EE
EE
EEE
EEE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EE
EE EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EE EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE EEE
EEE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
E
E
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EEEE
EEEE
EEEE
EEEE
EEEE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE EEE
EEE
EEE EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE EEE
EEE
EEE
EE
EE EEE
EEE
EEE
EEE EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EEE
EEE
EEE
EEE
EE
EE
EE
EEE
EEE
EE
EE
EE
EE
EEE
EEE
EEE EE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEEEEEEEE
EEEEEEEEE
EEEEEEEEE
EEEEEEEEE
EEEEEEEEE
EE
EE
EE EEE
EEE
EEE
EEE
EEE
EEE EEE
EEE
EEE
EEE
EEE
EEEE
EEEE
EE
EE
EE
EEE
EEE
EEE
EEEE
EEEE
EEEE
EEEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEEEE
EEEEE
EEEEE
EEEEE
EEE
EEE
EEE
EEEE
EEEE
EEEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEEE
EEEE
EEEE
EE
EE
EE
EE
EE
EE
EEE
EEE
EEE EE
EE
EE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EEE
EEEE
EEEE
EEEE
EEEE
EEEE
EEEE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EE
EE
EEEE
EEEE
EEEE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE EE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EE
EE
EEE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EEE
EEE
EEE
EEE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EEE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE
EEE
EE
EE
EEE
EEE
EEE
EEEE
EEEE
EEEE
EEEE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EEE
EEE
EEE
EEE
EEE
EE
EE
EE
EE
EEE
EEE
EEE
EEE
EEE
EEE
EE
EE
EEE
EEE
EE
EE
EE
EE
EE
EE
EE
EE
EEE
EEE
EEE
EEEE
EEEE
EEEE EEE
EEE
EEE
EE
EE
EE
< -1000 -1000 - -250 -250 - 0 0 - 250 250 - 1000 > 1000
EEEEEE
EEEEEE
EEEEEE
EEEEEE
No Change in Capacity
0 370 740 1,110 1,480 185
Miles
$
< -10 -10 - -5 -5 - -2.5 -2.5 - -1 -1 - 0 0 - 1 1 - 2.5 2.5 - 5 5 - 10 > 10
Net change in generation from 2008 to 2014 (TWh)
Net change in generation from power plants that were
added or retired between 2008 and 2014 only (TWh)
Net change in capacity from power plants that were
added or retired between 2008 and 2014 only (GW)
Net change in capacity from 2008 to 2014 (GW)
0 370 740 1,110 1,480 185
Miles
$
< - 1500 -1500 - -700 -700 - -300 -300 - -100 -100 - 0 0 - 100 100 - 300 300 - 700 700 - 1500 > 1500
A B
C D
> -10 -10 - -5 -5 - -2.5 -2.5 - -1 -1 - 0 0 - 1 1 - 2.5 2.5 - 5 5 - 10 > 10 > -1.5 -1.5 - 0.7 -0.7 - -0.3 -0.3 - -0.1 -0.1 - 0 0 - 0.1 0.1 - 0.3 0.3 - 0.7 0.7 - 1.5 > 1.5
< -3.5 -3.5 - -6.5 -0.65 - 0 0 - 15 > 15
generation capacity requiring cooling between 422: and 4236. Coal and natural gas steam-fired
as well as natural gas combined cycle units represented the majority of additions.
Our results indicate that the total water withdrawn and consumed for power generation de-
creased between 422: and 4236. The relative water withdrawal intensity of the grid in 422: was
approximately 42,222 gallons per MWh versus approximately 38,222 gallons per MWh in 4236.
The relative water consumption intensity of the grid in 422: was approximately 622 gallons per
MWh versus approximately 567 gallons per MWh in 4236. Water use intensities only reflect
power plants reporting both generation and water use (or power plants that could be assigned a
water use rate if no water use was reported) to reduce error in intensity values. In total, water
data from units representing 559 TWh of generation in 422: and 582 TWh of generation in 4236
68
were excluded from the analysis (typically from very small generators), which is assumed to have
a small impact on these results.
Retired power plants were generally fueled by coal, nuclear and natural gas-steam boiler units
cooled using once-through cooling systems, while added power plants were mostly natural gas
combined cycle cooled with recirculating towers. New natural gas combined cycle power plants
withdrew approximately 32 times less water per unit of electricity generated on average compared
to retired steam-fired power plants. However, most newly installed natural gas combined cycle
power plants use recirculating tower cooling systems and the water consumption per unit of
electricity generated is comparable to average steam-fired, once-through cooled power plants that
were retired during this period. Volumetric cooling water withdrawals and consumption decreased
in every major fuel category, when only net changes across new and retired power plants were
considered, except for the water consumed by natural gas generators. When all changes in net
generation were considered, total net water consumption also rose for natural gas generators
and despite retirements of large once-through cooled facilities, total net water withdrawals rose
for nuclear generators, further reflecting the transition toward recirculating cooled natural gas
generation as well as increases in demand for electricity.
Net changes in the volume of water withdrawn and consumed for thermoelectric power plant
cooling are illustrated in the left-hand (A, C) and right-hand (B, D) sides of Figure 6.4, respec-
tively. The net changes in water withdrawals are much larger than net changes in water con-
sumption, given that water consumption is a subset of water withdrawals, and water withdrawal
rates (i.e. cooling water withdrawn per MWh) can span several orders of magnitude depending
on cooling technology. It is important to note that the volumetric changes in water withdrawals
and water consumption occurring across HUC-: subbasins reflect a combination of factors such
as the spatio-temporal distribution, technological composition, and operational characteristics of
the generation fleet, as well as spatio-temporal changes in the magnitude of electricity demand,
69
Figure 6.4: Left: (A) Net changes in US cooling water withdrawals and (C) the subset of these
changes in withdrawals resulting exclusively from new or retired power plants between 422: and
4236 . Right: (B) Net changes in US cooling water consumption and (B) the subset of these
changes to US cooling water consumption resulting exclusively from new or retired power plants
between 422: and 4236. Includes thermoelectric power generation units requiring cooling greater
than 322 MW
> -500 -500 - -150 -150 - 0 0 - 150 150 - 500 > 500 < -5 -5 - -1.5 -1.5 - 0 0 - 1.5 1.5 - 5 > 5
> -100 -40 - -100 -10 - -40 0 - -10 0 - 10 10 - 40 40 - 100 > 100 > -10 -4 - -10 -1 - -4 -1 - 0 0 - 1 1 - 4 4 - 10 > 10
Net change in volume of cooling water withdrawn from power plants
that were added or retired between 2008 and 2014 only (Billion gallons)
Net change in volume of cooling water consumed from power plants
that were added or retired between 2008 and 2014 only (Billion gallons)
Net change in volume of cooling water withdrawn from
2008 to 2014 (Billion gallons)
Net change in volume of cooling water consumed from
2008 to 2014 (Billion gallons)
A B
C D
> -100 -40 - -100 -10 - -40 -10 - 0 0 - 10 10 - 40 40 - 100 > 100 > -10 -10 - -4 -4 - -1 -1 - 0 0 - 1 1 - 4 4 -10 > 10
> -500 -500 - -150 -150 - 0 0 - 150 150 - 500 > 500 > -5 -5 - -1.5 -1.5 - 0 0 - 1.5 1.5 - 5 > 5
in addition to any added and retired power plants. This is evidenced in the differences between
Figure 6.4A and B compared to Figure 6.4C and D.
Net generation for facilities using dry cooling increased approximately 82 TWh with capacity
additions of almost ; GW over this time period. It should be noted that Table 3 shows changes in
water use across each respective category. Thus, although the increased generation and capacity
from facilities reporting dry cooling resulted in increased water withdrawals and water consump-
tion across its own category, the impact of retrofitting cooling systems to dry cooling across the
power sector decreases the net water intensity of the fleet, as it requires approximately 32% of
the water required per unit of electricity of an average recirculating cooled power plant. Net
6:
generation cooled using reclaimed water increased 7: TWh from 422: to 4236, while capacity
was reduced by just over 38 GW.
Figure 6.5 illustrates changes in generation, calculated with Equation 4 from added (i.e. pos-
itive values of generation) or retired (i.e. negative values of generation) power plants between
422: and 4236 are quantified for each US HUC-: subbasin. Results are ordered from West to
East based on subbasin location. The shaded blue and white regions distinguish the larger HUC-4
regions (containing many HUC-: subbasins), which correspond to the numbered HUC-4 regions
labeled on the US map of Figure 6.5. The first three rows detail identical generation data, but
categorize these data in terms of the fuels (row 3), cooling systems (row 4), and cooling water
sources (row 5) utilized by each added or retired power plant(s) in the subbasin. Corresponding
cooling water consumed and withdrawn for these added/retired power plants are illustrated in
the fourth and fifth rows of Figure 6.5, respectively.
The spatial location and generation of newly installed and retired natural gas-fired and coal-
fired power plants during the period of study are illustrated in the bottom map of Figure 6.5.
New natural gas-fired combined-cycle power plants dominated installations in western and eastern
regions of the US, while a handful of new coal-fired steam cycle power plants were added in the
central US. Generally, new natural gas combined cycle power plants were larger in number but
smaller in capacity, compared to new coal-fired power plants.
Figure 6.5 does not include new generation units installed at existing power plants or changes
in generation from existing power plants from increased demand (these net shifts are summarized
in Table 6.3). A much clearer shift from coal-fired to natural-gas fired generation is seen when
changes in net generation are considered. During the study period, low natural gas prices in
combinationwithhigherefficiencycombined-cycleunitscausednaturalgascombined-cycledplants
to out-compete many existing coal-fired generators. This trend resulted in increased natural
gas generation and decreased coal-fired generation, even when capacity at existing power plants
remained the same. Installations of natural gas-fired units at existing power plants totaled almost
6;
Figure 6.5: Top: Generation (rows 3 - 5), water consumption (row 6), and water withdrawal
(row 7) changes from the retirement of 375 and addition of 386 thermoelectric power plants with
cooling systems between 422: and 4236 in each HUC-: watershed. Changes in generation are
colored according to fuel type (row 3), cooling system (row 4), cooling water type (row 5), while
changes in water usage are colored according to water risk (rows 5 and 6). HUC-: watersheds
are ordered from west to east based on HUC-: number. Shaded white and blue regions represent
HUC-4 watersheds, numbered on the bottom of the figure. Bottom: Geographic representation of
the 375 retired and 386 added thermoelectric power plants greater than 3 MW requiring cooling,
classified by fuel type and scaled to represent power plants added (circles) or retired (squares)
between 422: and 4236. Shaded regions correspond to shaded HUC-4 regions in the bar chart
(top). The “Other” fuel type classification includes geothermal, solar, biofueled, and multi-fueled
power plants.
" )
" ) " )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" ) " )
" )" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" ) " )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" ) " )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" ) " )
" )
" )
" )
" )
" )
" )
" )
" )
" ) " )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" )
" ) " )
" )
" )
" )
" )
" ) " ) " )
" )
" )
" ) " )
" )
" )
" )
" ) " ) " )
" )
" ) " )
" )
" )
" )
" )
" )
" )
" )
USGS The National Map: National Hydrography Dataset
0 370 740 1,110 1,480 185
Miles
$
" )
< - 5
" )
-5 - -3
" )
-3 - -1.5
" )
-1.5 - -0.5 " ) -0.5 - 0 0 - 0.5 0.5 - 1.5 1.5 - 3 3 - 5
> 5
Coal
Natural gas
Nuclear
Coal and Natural gas
Other
Net change in generation from power plants that were added or retired between 2008 and 2014 only (TWh)
18 16 14 12 10 8 6 4 2
18
16
14
12
10
8
6
4
2
For power plants that were added or retired between 2008 and 2014:
72
46 GW of new net capacity, whereas 33 GW of coal capacity was retired at existing power plants.
These trends in retrofitting or adding new generation units to existing power plants has resulted
in 37 power plants operating with both coal and gas units in 4236. These trends are discussed
more in the Discussion section.
Shifting Fuels
Recent shifts in the US electricity generating sector are driven primarily by declining natural gas
prices, which dropped from $;.24 to $7.22 per MMBTU for the electric power industry between
422: and 4236, respectively. (For reference, the average natural gas price in the power sector
between 4226 - 422: and 4232 -4236 was $9.67 and $6.73, respectively) [37:]. Natural gas
prices have fallen because of technical advances in horizontal drilling and hydraulic fracturing,
which have increased domestic production significantly during this period [92]. Low natural gas
prices prompted a market-based response in the electricity sector, as natural gas combined cycle
facilities gained competitive market advantage over coal-fired facilities due to lower fuel costs.
Additionally, newer natural gas-fired combined cycle units are often more efficient than older
units, which can increase their cost-competiveness with existing generators. Thus, net changes in
generation between 422: and 4236 are skewed towards newer, more efficient units.
Concurrently, more stringent environmental policies and regulations (e.g. Coal Combustion
Residuals rule, Mercury and Air Toxics Standards for Utilities, Clean Power Plan, Renewable En-
ergy Portfolio Standards, etc.) have also reduced the competitiveness of coal-fired power plants
compared to natural gas-combined cycle facilities due to higher operational costs [67]. The rel-
ative efficiency improvements of combined-cycle facilities versus other steam-cycle facilities also
improve the cost competitiveness of new natural gas units especially against coal as well as nuclear
generators.
Electricity markets close to the Marcellus shale have seen a particularly large switch from coal-
fired to natural-gas generation because of their proximity to areas of high gas production. As a
73
result, newly installed gas plants have replaced a significant amount of coal-fired power generation
in this region. In Pennsylvania alone, just over 5.6 GW of coal-fired capacity was retired from
422: to 4236. Many coal-fired power plants, especially older plants in the Ohio valley, have been
retired in the past six years (Figure 6.5, HUC-4 region 27). In this basin, 39 coal-fired power
plants were retired and 9 existing facilities retired coal-fired generating units from 422: to 4236,
totaling almost 7.9 GW in retired capacity. Some of coal’s national decline has been tempered by
a few new large coal-fired generating units, particularly in Texas, which added over 6 GW of new
coal capacity during the period of study.
Shifting Cooling Technologies
Shifts in cooling water technologies have an impact on operational cooling water usage at power
facilities. A fraction of retired power plants illustrated in Figure 6.5 represented relatively large
volumes of water withdrawals, corresponding to once-through cooled power plants. Most new
power plants withdraw relatively small volumes of water, reflecting a transition in the electricity
sector towards recirculating cooled facilities, which markedly reduce water withdrawals per unit of
electricity, typically at the expense of higher rates of consumption across similar fuel, cooling tech-
nology, prime mover classifications [:4]. However, these water benefits are unevenly distributed
across US watersheds (and more importantly, unevenly distributed across water-stressed water-
sheds) as illustrated in Figure 6.5. Net changes in water consumption are driven by increased
electricity demand, as the relative consumption rate (i.e. total reported consumption normalized
per unit of electricity generated) in 422: compared to 4236 remained relatively stable. Aside
from capacity additions, increases in net water consumption on an individual plant basis were
typically linked to increases in generation. Decreases in water withdrawals versus water consump-
tion offer different benefits. For example, a decrease in net water withdrawals across a watershed
may reduce the risk of a curtailment due to low water flows during droughts or heat waves and
also reduces the occurrence of the entrainment and impingement of living organisms across intake
74
screens. A reduction in water consumption can increase the water that can be allocated to other
water users in periods of severe drought, as net evaporative losses across a watershed are reduced.
The trend towards the retirement of once-through (i.e. open-loop) cooled power plants and the
addition of recirculating (i.e. closed-loop) tower cooled power generation units (Figure 6.4) reflect
environmental regulations, namely the 538 a&b amendments to the Clean Water Act (CWA).
These amendments address the ecosystem impacts associated with the thermal pollution and
entrapment/entrainment issues caused by once-through cooled facilities, respectively, are driving
transitions away from once-through cooled facilities [4:,33:,369]. For example, in California,
7.67 GW of reported once-through cooled thermoelectric capacity was retired between 422: and
4236. The state was the first to actively force the phase out of once-through cooled power plants
(Figure 6.5 HUC-4 region 3:) [7;]. Only 33 power plants remain in the state reporting once-
through cooling as their only cooling technology. Of these facilities, nine are natural gas-fired,
one is nuclear powered, and one reports the use of multiple fuels. With the exception of two
facilities, all of these plants operate a simple steam cycle. Five of the existing once-through
cooled facilities in 4236 are greater than 3 GW in capacity, the largest of which is Diablo Canyon
Nuclear Generating Facility. All of these large power plants reported using surface water from the
Pacific Ocean for cooling. The remaining once-through cooled generating facilities in California
(and elsewhere in the US) are currently under evaluation for compliance with the538 amendments
to the CWA. In fact, the planned decommissioning of Diablo Canyon Nuclear Generating Facility
in 4247 was announced in 4238, partly due to concerns over the environmental impacts of its
cooling system [32:].
There is a clear trend toward the use of dry cooling systems for new power plants, despite
higher capital and operational costs [8:]. Although dry cooling technologies impose efficiency
penalties, the technology can relieve freshwater dependency in drought-prone and water-stressed
regions,especiallywhencoupledwithreclaimedwaterusage. Approximately4.6GWofgenerating
capacity (:.86 TWh of generation) in 4236 came from power plants that retrofitted to dry cooling
75
systems. Furthermore, capacity additions from dry-cooled power plants added to the grid between
422: and 4236 totaled approximately 9.: GW (53 TWh of generation). In fact, dry cooling
systems were used to cool almost 5.4% of thermoelectric generation requiring cooling in 4236.
Shifting Cooling Water Sources
The transition to less water withdrawal intensive generation is also reflected in the source of water
used for cooling new power plants. Surface water is the most common water source for cooling
thermal power plants. However, this analysis reveals a number of transitions to other cooling
sources. Many new power plants are utilizing alternative water sources, such as reclaimed water
from municipal waste water treatment facilities, relieving some of the pressure on fresh surface
water [5]. However, at the same time, there is an appreciable growth in new power plant capacity
using groundwater for cooling, which could add pressure to groundwater depletion if not managed
sustainably. Net decreases in groundwater withdrawals coupled with net increases in consumption
highlight the transition towards using groundwater for cooling recirculating towers. A noticeable
decrease in the use of ocean water for cooling in California and on the East Coast of the country
highlights the influence of regulations and policies on cooling water sources for thermoelectric
generators, namely CWA a&b (Figure 6.5).
As ocean water cooling has decreased, reclaimed water use for cooling has increased, tempering
the trend’s impact on freshwater consumption. Between422: and4236, :.6 GW of capacity tran-
sitioned fully to reclaimed water for cooling, 8.6 GW of capacity reported using some reclaimed
water for cooling (in addition to groundwater or surface water), and 35.6 GW of capacity cooled
using reclaimed water was added to the grid from new power plants (representing approximately
52, 2.5, and 75.8 TWh of generation, respectively). The use of reclaimed water can also relieve
cooling water related disruptions in the power sector, especially in drought-prone areas, by reduc-
ing dependence on freshwater usage and alleviating down-stream water stress. California, Nevada,
and New York are among the top states for reclaimed water use at power plants (Figure 6.5). In
76
California, many of the large facilities that use reclaimed water are recirculating or dry cooled nat-
ural gas combined cycle plants; smaller reclaimed water users (in terms of power plant capacity)
are geothermal plants and plants using multiple biofuels, oils, and/or other gases.(Figure 6.5).
The CWA amendments are also driving a shift towards reclaimed water use, as many coastal
power plants must use an alternative cooling water source to the ocean. Due to policy, water
availability, and drought vulnerability, many new power plants are using reclaimed water for cool-
ing recirculating or coupled with dry cooled systems. Alternative cooling water sources represent
a much larger fraction of water-cooled thermoelectric generation than previously understood. In
fact, reclaimed water was used to cool almost :% of thermoelectric generation requiring cooling
in 4236. In addition, the feasibility of reclaimed water use for thermoelectric cooling has been
studied and research in Texas shows that reclaimed water sources are located close enough to
;4 power plants (72% of fresh water-cooled thermoelectric generation in Texas) to be a suitable
cooling water source [359]. Therefore, it is reasonable to assume that the transition toward al-
ternative cooling water sources will continue, as water resources become increasingly scarce and
policies push thermoelectric generators to look for alternatives to traditional power plant cooling.
6.6 Conclusion
This study investigates how recent shifts in thermoelectric power generation in the US has im-
pactedthespatialdistributionofcoolingwaterusageacrossHUC-:basins. Resultsfromthestudy
illustrate some key trends that occurred in the US power sector between 422: and4236, including
transitions (3) from coal-fired steam to natural-gas combined cycle units, (4) from once-through
cooling to wet-recirculating towers and dry cooling systems, and (5) from traditional fresh and
saline surface water to reclaimed water sources. Consequently, the electricity sector is moving
towards more water-withdrawal efficient technologies, which can result in a water consumption
penalty (in the case of wet-recirculating towers) or a water consumption benefit (in the case of
77
dry cooled systems). In addition to the expansion of dry cooling across the power sector, much
of the added capacity in recent years has been from natural gas combined-cycle facilities with
recirculating towers, which benefit from large efficiency increases compared to the once-through
steam-cycle facilities they are typically replacing; thus, this water consumption penalty due to
cooling technology transitions has not yet been significant, although results are highly variable by
watershed. Accordingly, the net impact to cooling water usage has been a reduction in the average
volume of water withdrawn and relatively flat average water consumption when normalized per
unit of electricity produced in4236 versus422:. Some of these trends have been tempered slightly
by increased electricity demand during this time. Because of the uneven spatial distribution of
new and retired power generation capacity, changes in the relative cost-competitiveness of one
generator to another (affecting frequency of dispatch and duration of operation), and the shifts
in demand, the changes to water usage and the ratio of water withdrawals to water consumption
vary significantly across watersheds.
Overall, the most important regulatory efforts affecting the water consumed and withdrawn
by power plants has been the CWA 538 a&b amendments that are driving the power fleet from
once-through towards recirculating technologies. These amendments have mixed-implications for
water use by the power sector. While most retirements that required high water withdrawals
have been replaced with relatively high water consumptive technologies, much of the existing
once-through cooled capacity in the interior US is where water is relatively abundant since these
systemsgenerallyrequirehigher-flowratesforoperation. Inwater-constrainedstatessuchasTexas
and California that utilize a lot of ocean-cooled capacity, this transition needs to be managed with
freshwater impacts in mind. Although these shifts are generally ubiquitously good for ecosystem
health, the transition away from once-through cooled facilities that use ocean water for cooling,
can result in an increase of freshwater usage, since recirculating tower power plants typically do
not use ocean water. Reclaimed water and dry cooling technologies are attractive options to avoid
these freshwater impacts, although they can induce other operational trade-offs (e.g. decreased
78
efficiency in the case of dry cooling or increased fouling concerns in the case of reclaimed water
use).
There are not clear trends in terms of net increases or decreases of cooling water usage as a
function of relative water scarcity. However, there are signs that freshwater availability is affecting
trends in decisions regarding new power plant installations. Dry and hybrid cooling systems, as
well as the use of alternative water sources, such as reclaimed water, are growing in terms of net
capacity. However, power plant capacity utilizing groundwater in water-constrained locations is
also growing, which is a trend that will be important to moderate moving forward, as aquifers
across much of the US are being exploited at rates much faster than they recharge.
This analysis points to the importance of considering the water use requirements of power
plants when new installations are being planned, as freshwater availability across the US varies
significantly. While the framework developed in this analysis provides an estimate of the cooling
water demands across HUC-: subbasins, resolving data limitations (i.e. omissions, zero-values,
erroneous values, variable measurement techniques) associated with federal cooling water usage
reporting by power plants would improve the utility of the results presented here.
79
Chapter 7
Developing methods to estimate the environmental
externalities of the power sector with high spatio-temporal
resolution
This chapter reflects work published in Environmental Science & Technology in 4238 [334].
7.3 Motivation
The power sector’s contribution to global greenhouse gas (GHG) emissions and air pollution emis-
sions as well as its wide-spread water use impacts are undeniable. Research by the International
Panel on Climate Change (IPCC) has shown that the power sector is one of the leading contribu-
tors to climate change via the emission of approximately 62% of GHG equivalents worldwide [83].
Additionally, it is well known in the field of air pollution research that sulfur oxides (SO
x
) and
nitrogen oxides (NO
x
), two primary air pollutants emitted from thermoelectric power plants, in
the atmosphere cause environmental impacts such as photochemical smog, tropospheric ozone for-
mation, and acid rain as well as adverse health impacts [83]. The US Geological Survey (USGS)
has highlighted the magnitude of water use (withdrawals and consumption) for the power sector
in their quinquennial report on estimated use of water in the US since the National Water Project
began in 3;9: [::]. Although the GHG emissions, air pollution emissions, and water use of power
7:
plants impact ecosystems and human health unevenly in space and time, there has been rela-
tively little research of the associated environmental trade-offs on both a spatially and temporally
relevant scale.
Manypreviousanalyseshaveevaluatedhowchangesingridconfigurationsanddemandprofiles
impact cooling water requirements, [7,4:,65,:5,:6,;2,32;,33:,33;,344,34:,352,359,35:,364,
388,397] GHG emissions, [7,44,52,336,346,348] as well as air pollutant emissions [:,9;,38:].
However, very few analyses have addressed air emissions and water use simultaneously [49,32;,
392]. Webster et al. [392] found that a restriction on CO
4
emissions (e.g. 97% reduction below
business as usual) at ERCOT power plants would also reduce water withdrawals. However, more
stringent restrictions on CO
4
emissions (e.g. greater than 97% below business as usual) would
likely increase water withdrawals due to the implementation of pollution controls (which often
require large amounts of water); restrictions on both CO
4
emissions and water withdrawals would
force the ERCOT region to shift towards different (and sometimes more expensive) technologies
(e.g. hybrid or dry cooling systems, carbon capture and sequestration). Clemmer et al. [49] used
the National Renewable Energy Laboratory’s Regional Energy Deployment System (ReEDS),
a long-term capacity expansion model, to assess the impact of various low-carbon electricity
futures on national water use. The researchers found that the 4272 reference case scenario
(projected business as usual) had the lowest electricity costs, but at the expense of high carbon
emissions, water consumption, and a poorly diversified electricity mix; conversely, they found that
significant technical changes would be required to meet stringent carbon reductions by 4272, with
varied water savings, carbon emissions reductions, and public health benefits depending on the
technologies used (i.e. a focus on nuclear increases water consumption, but reduces emissions; a
focus on renewablesreduces both water consumption and emissions, with the potential for reduced
electricity costs). Pacsi et al. [32;] studied the feasibility and air quality impacts of drought-based
dispatching in the ERCOT footprint using a power flow model, finding that dispatching power
7;
plants based on water availability, although expensive, could result in coupled water savings and
air pollution reductions in drought-stricken areas of the state.
Although previous studies have analyzed the systems-level water or air emissions impacts of
the electricity sector under a variety of scenarios, [49,4:,:5,33;,388,392] to date no analysis
has evaluated the trade-offs in terms of water consumption, GHG, and air quality pollutant
emissions with high spatio-temporal resolution [83,336,33;,359]. This analysis fills this existing
research gap by quantifying water, air, and climate trade-offs with high temporal (i.e. hourly)
and spatial resolution, which will be important to consider as the US makes new investments in
power infrastructure to meet different, and often competing environmental priorities.
7.4 Methodology
This work utilizes a Unit Commitment and Dispatch Model (UC&D) model of the Electric Re-
liability Council of Texas (ERCOT) grid developed and described by Townsend, [368] Garri-
son [69,6:] and Sanders [33;]. The model, implemented in Energy Exemplar‚s PLEXOS
®
soft-
ware, simulates the dispatching algorithm utilized by ERCOT to coordinate the reliable provision
of electric power services. The UC&D framework simulates EGU dispatch and reserve supply
according to a least cost optimization method, solving each day using load and available supply
conditions defined for the previous, current, and following day (using 6-hour look-ahead win-
dows averaging grid requirements for the following day) such that the Short Run Marginal Cost
(SRMC) is minimized. Thus, capital costs are excluded [349]. User inputs include generator type
definitions, associated generator information (e.g. EGU capacity, ramp rates, minimum operat-
ing level, maintenance and downtime, minimum runtime requirements, etc.), system-wide hourly
load profiles, and ancillary generation requirements. Model constraints limit overall generator
82
operations to reflect realistic EGU performance (see Garrison et al. (4237) for full model docu-
mentation) [69]. SRMC (Equation 4.3) is a function of variable operation and maintenance costs
(C
O&M
), fuel costs (C
fuel
), and power plant heat rate (HR)
SRMC = C
O&M
+C
fuel
×HR (7.3)
Thus, the model’s algorithm determines the optimal generation fleet dispatch required to meet a
prescribed load for every hour of the year such that the total dispatch cost of generating electricity
is minimized. Model results are quantified on hourly time steps for a 4233 ERCOT load curve
using 4233 fuel prices and the 4233 ERCOT generation fleet.
Equation 4.4 was used to calculate EGU specific environmental impact, X
k,t
, based on gen-
eration, G
k,t
, for each EGU, k, at time, t; EGU-specific emissions rate, E
k
(in gallons per MWh
generated or pounds of emissions per MMBTU of fuel consumption); and time-varying, EGU-
specificheatrate,HR
k,t
. X
k,t
wascalculatedforwaterconsumption, CO
2
, NO
x
, orSO
x
emissions,
respectively.
X
k,t
= G
k,t
×E
k
×HR
k,t
(7.4)
Next, the hourly environmental intensity of the entire ERCOT grid, X
t
, (i.e. the total grid-
wide impact per hourly electricity generation) was calculated for each environmental impact,
X
k,t
.
Intensity, X
t
=
P
252
k=1
X
k,t
P
252
k=1
G
k,t
(7.5)
Water consumption rates were assigned according to a report published by King et al. for the
Texas Water Development Board in 422: for each EGU in the ERCOT region [89]. Average
annual emissions rates were calculated from the Environmental Protection Agency‚s Air Markets
83
Program [62,63]. Heat Rates are modeled in UC&D simulations with piece-wise linear segmented
second-order polynomial function to better represent generators’ actual fuel consumption during
all stages of operation. Results from the UC&D model were post-processed to interface with
ArcGIS to perform spatial analysis.
The UC&D modeling approach facilitates high temporal and spatial resolution of environmen-
tal criteria results. Although emissions data are available on an hourly timescale from the EPA,
to date, cooling water use data have only been available with yearly resolution. The model also
allows for the evaluation of peak-shifting scenarios and their impact on EGU dispatch resulting
from any prescribed load scenario. Most importantly, this framework allows for the realistic sim-
ulation of power plant retrofitting and replacement, since dispatch order moderates overall grid
operation. This is to say that retrofitting and/or replacing power plants with identical generating
capacity produces non-linear environmental impacts. Additionally, this methodology provides a
framework for computing emissions and water use at hourly timescales for projected scenarios.
7.5 Results and Discussion
Temporally resolved environmental impacts
Hourly generation (row 3), CO
2
(row 4), NO
x
(row 5), and SO
x
(row 6) emissions, as well as
water consumption (row 7) data were aggregated by EGU fuel type, cooling technology, and
prime mover, as displayed in Figure 7.3. Each resulting data type is shown for the entire year
(column 3), as well as for a sample summer week (column 4) and a sample summer day (column
5) in August 4233 to represent seasonal, weekly, and daily variability, respectively.
Coal and nuclear EGUs provide relatively constant baseload generation throughout the year
to ensure grid reliability, minimize the SRMC, and avoid ramping (these EGUS typically have
slow ramp rates, large minimum down times, and high minimum operating levels). The majority
of this baseload generation in the ERCOT footprint is provided by EGUs that use once-through
84
(OT) cooling systems, typically older units that withdraw large volumes of water. While baseload
nuclear generation does not generate air or GHG emissions, baseload coal generation is responsible
for a large portion of air quality and GHG emissions in the ERCOT region.
Natural gas combined cycle plants are able to follow load more than baseload power plants, but
are unable to react to rapid changes in demand. They are also dispatched before other older, more
expensive or inefficient natural gas EGUs, such as steam boiler (NGBlr) plants, since EGUs are
generally dispatched in order of least to highest cost to meet demand. Natural gas combustion
turbines (i.e. OCGT and internal-combustion natural gas (NGIC)) are much smaller and can
react much quicker than baseload and combined cycle units. Combustion turbine EGUs are often
used for ancillary services to maintain the reliable operation of the grid by quickly ramping up
and ramping down to balance supply with demand. However, these EGUs are typically expensive
to operate, and emissions intensive, since they are not as efficient as larger combined cycle units.
Electricity demand in the ERCOT region is the highest during the hottest months of the year,
due largely to the use of space cooling systems (Figure 7.3) [;7,38;]. The weekly generation
profile shows a slight decrease in electricity demand during the weekend due to less business and
industrial activites [336]. Daily electricity demand peaks in the late afternoon and early evening
as people generally arrive home from work and perform energy intensive tasks. This observation
is particularly pronounced during the hottest months of the year, when space cooling usage is
highest.
The emissions profiles of CO
2
and NO
x
are similar in shape to the electricity generation profile
(except for the obvious omission of nuclear) because carbon based fuel (primarily coal-fired and
natural gas-fired) EGUs are responsible for all CO
2
and NO
x
emissions and also represent a large
fraction of generation. However, the shape of these emissions profiles is skewed towards older,
open-loop cooled coal-fired EGUs and away from newer combined-cycle natural gas-fired EGUs
when compared to the generation shape profile (Figure 3). The SO
x
emissions profile is flat as
85
Figure 7.3: The electricity generation profiles for EGUs in the ERCOT region are generally flat
for coal and nuclear EGUs (i.e. baseload EGUs) and variable for natural gas EGUs, which vary
according to fluctuations in demand. CO
2
and NO
x
emissions follow the shape profiles of coal and
natural gas generation, while SO
x
emissions are primarily associated with coal generation. Water
consumption is skewed towards recirculating cooled thermoelectric generators. Large seasonal
and daily fluctuations reflect energy use patterns and climatic characteristics. Data reflect 4233
load characteristics. Daily profiles are for August 3, 4233. Fuel and prime mover type: nuclear
(Nucl), coal (Coal), natural gas combined cycle (NGCC), natural gas boiler (NGBlr), open-cycle
natural gas (OCGT), internal-combustion natural gas (NGIC), hydroelectric (Hydro); Cooling
technology: once-through (OT), recirculating (RC), no cooling (NA)
86
SO
x
emissions are primarily a result of coal combustion. When summed over the entire year, coal-
fired EGUs contributed 98.3%, :7.4%, and ;;.;% of the total CO
2
, NO
x
, and SO
x
emissions,
respectively in the ERCOT region.
Baseload coal-fired and nuclear EGUs represent a significant fraction of water consumption.
However, the variability in seasonal, weekly, and daily profiles is driven primarily by closed-loop
cooled combined-cycle natural gas-fired EGUs, which also consume water at higher rates than
once-through cooled plants. The water consumption profile varies slightly from the profiles of CO
2
andNO
x
asitisskewedtowardsEGUswithclosed-loopsystems, whichareoftennewerandcleaner
than less water-consumptive (but emissions intensive) once-through cooled plants. However, these
older open-loop cooled baseload plants represent the majority of water withdrawals in ERCOT,
which can have negative ecosystem impacts within the water reservoirs that they pump water
to and from during the cooling cycle through impingement, entrapment, and thermal discharge
issues [6,34,33;,353].
Grid-wide environmental emissions and water consumption intensities
Total grid-wide CO
2
, NO
x
, SO
x
, and water consumption intensities (i.e. total impact per unit of
hourly electricity generated) were calculated using Equation 5 and are presented in Figure 7.4.
The variations in each profile are largely a function of the fraction of generation represented by
coal and/or natural gas-fired EGUs (Figure 7.4, top row). Because coal and natural gas-fired
EGUs both emit CO
2
and NO
x
, and represent a large fraction of total generation across all hours,
the profiles of these two emissions intensities remain relatively flat. For short periods during the
day and during the summer months, when increased demand is met with more emissions-lean
generating technologies, the emissions profiles decrease slightly. SO
x
emissions are almost entirely
a result of coal-fired EGUs, which maintain a fairly stable baseload generation (Figure 7.3).
Therefore, when the fraction of generation represented by coal-fired EGUs is high (i.e. during the
colder months and during the night), the SO
x
emissions intensity is also high. Conversely, when
87
the fraction of generation represented by coal-fired EGUs decreases as other generating units are
used to meet demand, the SO
x
emissions intensity falls. Water consumption is largely a function
of coal-fired and natural gas combined cycle EGUs with closed-loop cooling systems. During
the night, the water consumed per MWh generated is high; during the day, it decreases as more
water-lean generators come online. Thus, in considering Equation 5, the numerator changes much
less than the denominator for SO
x
emissions and water consumption intensities across the year,
translating to shape profiles that have much more defined peaks than the shape profiles of CO
4
and NO
x
emissions intensities in Figure 7.4.
As a result, electricity generation during the night and during cooler months of the year is
more emissions- and water-intensive per MWh generated as compared to electricity generated
during the peak of the day or the hottest months of the year. However, it is important to
note that total emissions and water consumption are highest when demand is greatest because
of high electricity load. Consequently, reducing a unit of electricity generated during the night
or when temperatures are cooler is actually more advantageous in terms of emissions and water
consumption reductions than to reduce a unit of electricity generated during the day or when
temperatures are hottest. (Although, typically peak demand reductions are advantageous for grid
reliability and cost standpoints.) To evaluate the grid-scale impacts of shifting generation from
the peak of the day to nighttime hours, two additional simulations (72% peak shift and flat load)
were executed to quantify shifts in environmental externalities. Both scenarios result in minor
grid-wide increases in all evaluated environmental externalities. (Full results are available in the
SI document).
Spatially resolved environmental impacts
ResultsfromtheUC&Dmodelwerealsoanalyzedonanindividualpowerplantlevel, byevaluating
the percentage of CO
2
, NO
x
, and SO
x
emissions as well as water consumed by each power plant
relative to total ERCOT-wide emissions and water use for the year (Figure 7.5). Fractional
88
Figure 7.4: Averaged CO
2
, NO
x
, SO
x
, and water consumption intensity generally trend with the
fraction of baseload coal-fired power on the grid at any given hour.
89
contributions of each power plant are organized from highest to lowest percentage value, where
each section of the figure corresponds to a power plant with one or many EGUs of the same
type. Figure 7.5 illustrates that large fractions of water use and emissions from the grid are
attributable to a handful of EGUs, mostly coal-fired or nuclear generation facilities with open-
loop cooling systems (as these tend to be the oldest). As seen in Table 7.3, emissions in the
ERCOT region are disproportionately skewed towards coal-fired EGUs and water consumption
is disproportionately skewed towards closed-loop cooled EGUs when compared to total relative
generation. Therefore, a simultaneous reduction in GHG, air pollution emissions, and water
consumption could be achieved if EGUs producing the largest environmental externalities were
targeted.
Table 7.3: Relative distribution of generation, emissions, and water consumption varies by gener-
ating technology.
EGU
type
% total
generation
% total CO
4
emissions
% total NO
x
emissions
% total SO
x
emissions
% total H
4
O
consump.
Biomass 0.06 − − − 0.02
Wind 9.3 − − − −
Hydro 0.44 − − − −
NGIC-NA 0.13 0.12 0.27 − 0.02
OCGT-NA 1.5 1.9 5.7 0.02 0.22
NGBlr-OT 1.4 1.2 0.99 − 1.1
NGBlr-RC 0.05 0.05 0.08 − 0.08
NGCC-OT 0.68 0.52 0.17 − 0.82
NGCC-RC 28 20 7.6 0.07 18
Coal-OT 33 56 57 79 20
Coal-RC 12 20 28 21 37
Nucl-OT 13 − − − 23
Figure7.5alsoshowsthelocationsofthe32powerplantsrepresentingthelargestcontributions
of air, water, and GHG emissions, as well as the total fractional contribution of each power
plant to emissions and water consumption across the ERCOT footprint. The geographic context
8:
GENCO
2
NO
x
SO
x
WC
0
5%
10%
15%
20%
GENCO
2
NO
x
SO
x
WC
0
5%
10%
15%
20%
GENCO
2
NO
x
SO
x
WC
0
5%
10%
15%
20%
GENCO
2
NO
x
SO
x
WC
0
5%
10%
15%
20%
GENCO
2
NO
x
SO
x
WC
0
5%
10%
15%
20%
GENCO
2
NO
x
SO
x
WC
0
5%
10%
15%
20%
GENCO
2
NO
x
SO
x
WC
0
5%
10%
15%
20%
GENCO
2
NO
x
SO
x
WC
0
5%
10%
15%
20%
GENCO
2
NO
x
SO
x
WC
0
5%
10%
15%
20%
GENCO
2
NO
x
SO
x
WC
0
5%
10%
15%
20%
0
20
40
60
Generation
(1000MW)
Biomass
Wind
Hydro
NGIC-NA
OCGT-NA
NGBlr-OT
NGBlr-RC
NGCC-OT
NGCC-RC
Coal-OT
Coal-RC
Nucl-OT
0
20000
40000
60000
80000
CO
2
Emissions
(1000 lb)
0
10
20
30
40
50
NO
x
Emissions
(1000 lb)
20
40
60
80
100
120
SO
x
Emissions
(1000 lb)
Total Gen.
Total CO
2
Total NO
x
Total SO
x
Total Water
Consumption
0% 20% 40% 60% 80% 100%
Figure 7.5: A geographic representation of power plants contributing the largest fraction of total
CO
2
, NO
x
, SO
x
emissions, and water consumption across the ERCOT region is shown overlain
with Texas state boundaries, river basins, and state population density. Colored dots represent
power plant type, physical location, as well as generation (by size). Percent of total CO
2
, NO
x
,
SO
x
emissions, water consumption, and generation for each power plant in the ERCOT region is
displayed in the bar graph.
8;
is particularly important for discussion surrounding NO
x
and SO
x
emissions, as well as water
consumption. NO
x
and SO
x
emissions are important to consider in terms of regional air quality,
as these emissions contribute to the formation of localized air pollution via particulate matter
(PM) [83]. Therefore, power plants located in proximity to areas with larger population density
cancontributetovisualpollutionandhealthimpactsviaPMformation. ThestateofTexassuffers
from air pollution, particularly in urban areas, and has struggled with non-attainment status for
national ambient air quality standards (NAAQS), particularly ozone, in the Houston-Gavelston-
Brazoria and Dallas-Fort Worth areas [386]. Although CO
4
is a globally well-mixed pollutant,
total magnitude concentrated across a small subset of generators is substantial. Collectively, the
32 highlighted power plants represent 68%, 72%, 85%, 89%, and 82% of the generation, CO
2
,
NO
x
, SO
x
, and water consumption in the ERCOT region, respectively.
Spatial trends for water use are important because shifts in water use can impact drought
resiliency for a power plant and for regions sharing a watershed with water consumptive genera-
tors. Retrofitting power plants from once-through cooling to recirculating cooling systems could
reduce water diversions, making more water available to junior water rights holders [356]. While
decreased reliance on water withdrawal is important for drought resiliency, retrofitting to recir-
culating cooling can increase water consumption, actually decreasing net water availability in the
basin [32;,356]. However, if newer RC plants displace generation from older, more expensive RC
plants in a UC&D regime, there still might be net water consumption benefits.
Environmental implications of natural gas combined cycle retrofitting
The impact of natural gas combined cycle retrofitting was evaluated by investigating two conver-
sion scenarios, considering the replacement of the EGUs with the greatest environmental impacts
on the grid mentioned in the previous section.
92
• Scenario 3: conversion of : coal-fired power plants (4 recirculating cooled and 8 open-loop
cooled facilities) identified in Figure 5, to natural gas combined cycle facilities with equal
capacities
• Secnario 4: conversion of : coal-fired power plants and 4 nuclear power plants (both open-
loop cooled facilities) identified in Figure 5, to natural gas combined cycle facilities with
equal capacities.
In the current environment of high production of natural gas and low natural gas prices in Texas,
natural gas-fired combined cycle EGUs are an attractive option for new or retrofitted genera-
tion due to economic, regulatory, and environmental benefits over other generation technologies,
especially coal [47,49,5:,74,77,9;,:2,:4,389,393].
For Scenario 3, if more modern natural gas combined cycle facilities replaced the selected
coal-fired generators, total ERCOT-wide reductions of approximately 4;%, 77%, 84%, and 35%
in CO
2
emissions, NO
x
emissions, SO
x
emissions, and water consumption, respectively, would be
achieved when considering optimum dispatch based on 4233 fuel costs. In Scenario 4, ERCOT
wide reductions in CO
2
emissions, NO
x
emissions, SO
x
emissions, and water consumption are
approximately 3;%, 73%, 82%, and 49%, respectively. Scenario 3 results show a greater impact
on grid-wide emissions reductions, while Scenario 4 results show a lesser impact on reducing emis-
sions and a greater impact on reducing water consumption compared to the baseline (Table 7.4).
These results are expected, as retrofitting old, inefficient coal plants will have a greater impact
on emissions than retrofitting emissions-free nuclear plants. Similarly, retrofitting once-through
cooled thermal power generators to recirculating cooled power generators actually increases the
basin-wide water consumption where new facilities are installed (although water withdrawals and
thermal pollution would be significantly reduced).
93
Table 7.4: Converting from once-through cooled to recirculating cooled NGCC EGUs increases
water consumption (WC) in Texas Water Basins where replacements occur, regardless of fuel
conversion.
Baseline Scenario 3 Scenario 4
Basin
Name
Basin
Electric
WC [362]
WC
Share of
electric
WC
WC
Share of
electric
WC
WC
Share of
electric
WC
Brazos 76.; 35.3 46% 42.: 5:% 45.6 65%
San
Antonio
39.2 8.35 33% 32.7 3;% 34.5 44%
Colorado 45.7 39.; 55% 46.8 67% 42.: 5:%
Cypress 32.: 5.6; 8.6% ;.9; 3:% 33.3 42%
Red 4.24 3.84 5.2% 9.59 35% 34.6 45%
Sabine 3:.6 9.3; 35% ;.93 3:% 33.: 44%
Trinity 33.2 4.;2 7.5% 32.5 3;% 33.: 44%
While some basins have increased water consumption where new recirculating NGCC plants
replace once-through cooled EGUS, grid-wide water consumption savings are achieved via off-
setting generation from older, less efficient recirculating cooled power plants in the least cost
dispatching regime. For example, conversions of once-through cooled coal-fired power plants re-
sulted in increased annual water consumption at the plant level on the order of 57-529% across
the two scenarios. The Sabine, Cypress, Brazos, Trinity, Colorado, San Antonio, and Red river
basins would all experience increases in total electric water consumption (Table 7.4) as a result of
conversion to natural gas combined cycle facilities with recirculating cooling systems. Although
the Brazos river basin incurs one of the highest water consumption penalties from retrofitting the
conversion scenarios, the water withdrawal savings that would be incurred would likely be most
meaningful because it contains the longest section of river in the state, and services many large
urban areas, agricultural water users, and industrial water users [356]. (Full results of conversion
scenarios are available in A.)
94
7.6 Conclusion
Developments in policy and technology over the past decade are pushing the U.S. electric sector
towards an important state of transition [6,7,38,46,93,:2,345,352,389]. Separate, and often
conflicting, policies have been issued to reduce pollution, greenhouse gas emission, water usage
or ecosystem impacts [33:]. However, this analysis demonstrated that environmental priorities
are not always aligned. Some power plants that are water lean have emission penalties (e.g.
dry-cooled fossil fueled EGUs), while some emissions lean technologies are water intensive (e.g.
nuclear or concentrating solar power plants). There are also “win-win” scenarios. Solar PV panels
and wind turbines require little to no water. Retrofitting the most environmentally taxing power
plants to more emissions-lean and water-lean natural gas combined cycle facilities offers significant
benefitsintermsofoverallreductionsinCO
2
, NO
x
, SO
x
emissions, andwaterconsumptionforthe
region. However, new and retrofitted recirculating cooled power plants may individually consume
more water than older, once-through cooled power plants. Grid-wide reductions of environmental
impact could result in significant improvements in regional air quality and water availability for
other users that share airsheds and watersheds with these generators, relatively cost-effectively.
Therefore, new policies must consider the multi-faceted spectrum of environmental impacts to
avoid creating unintended consequences to environmental or human health. Retrofitting coal-fired
EGUs results in significant changes in air pollution emissions and therefore, significantly impacts
the overall socioeconomic impacts of these EGUs. In conversion Scenario 3, approximately $822
Million (4233 USD) in social costs of air pollution impacts would be mitigated by retrofitting of
: coal-fired generators (See SI for full results of socioeconomic analysis).
This highly temporally-resolved evaluation of the environmental trade-offs between water con-
sumption, GHG emissions, and air pollutant emissions in ERCOT serves as a foundation for more
holistic analyses of the electricity grid and the associated environmental impacts moving forward.
Many of the trends identified in this analysis were driven by the fact that baseload generation
95
is provided almost exclusively by coal-fired and nuclear EGUs, while seasonal and daily peaks in
demand are generally accommodated by flexible natural gas-fired EGUs, which are able to ramp
up or down to follow load. We found that water consumption follows a similarly shaped profile
to generation, since flexible combined-cycle natural gas-fired EGUs require a significant fraction
of water in addition to baseload nuclear and coal-fired EGUs. When compared to generation
profile, water consumption is distorted towards EGUs with closed loop cooling systems, which
reflects the fact that these systems are very water consumptive per MWh of electricity generated
compared to other types of cooling. CO
2
and NO
x
emissions in the ERCOT region also follow
a similarly shaped profile as electricity generation, with coal and natural gas-fired EGUs being
the primary contributors to CO
2
and NO
x
emissions. Conversely, coal-fired EGUs are shown
in this analysis to be almost entirely responsible for SO
x
emissions, demonstrated by an almost
flat profile. When normalized over the grid, average water consumption, CO
2
, NO
x
, and SO
x
emissions intensities (i.e. output per MWh) are highest when electricity demand is the lowest,
being provided by primarily baseload EGUs. A large fraction of total CO
2
, NO
x
, SO
x
emissions,
and water consumption across the ERCOT region were found to be provided by only a handful of
power plants, mainly baseload coal-fired generators. The results of this analysis, which offer re-
sults with unprecedented spatio-temporal fidelity, can be used as proxies to evaluate the trade-offs
of power generation in other regions.
96
Chapter 8
Quanitfying Regional Water Use Rates for the Electricity
Sector
8.3 Motivation
Electricity generation is responsible for substantial water resource consumption in the United
States (US), especially when accounting for water used for processes upstream of the point of
electricity generation (PoG). Water consumption for fuel extraction, processing, transportation,
and conversion into electricity varies a great deal across fuel types, production techniques, and
electricity generation technologies [75]. Water consumption for electricity also varies in space and
time, evenforsimilarfuelsandgenerationtechnologies. Furthermore, electricityasacommodityis
a product of a diverse generation portfolio, not an individual power plant, which means that quan-
tifying water consumption associated with electricity is especially challenging [48]. Understand-
ing the water intensity of electricity is relevant for diverse stakeholders, including utilities [34:],
sustainability planners and environmental analysts [337], and electricity consumers trying to un-
derstand the water footprints of their consumption patterns [7:]. Specifically, understanding the
water intensity of electricity at a regional level is important because of the regional nature of
water availability and water management practices [49], and because thoughtful comanagement
of energy and water resources can lead to improved environmental outcomes [43,8:]
97
Previous work on water use for electricity has focused on quantifying water use or water use
ratesforelectricitygeneratingfacilities(i.e., powerplants)usingliteraturesourcesandgovernment
reports [:4], heat budget models [56], and self-reported data [5]. One recent study from Argonne
National Laboratory presents regional water consumption factors for North American Electric
Reliability Corporation (NERC) regions, accounting for 85% of US thermal generation in 4236
[97]. These studies represent valuable estimates for the electricity sector’s water use, but they
neglect water use occurring during production, processing, or transportation of fuels (referred to
as âĂIJupstreamâĂİ stages in this letter). These upstream stages can be important contributors
to the overall water intensity of electricity [:;]. Other studies have addressed upstream stages
[75,;;], but they often rely on coarse or generic data or do not specifically characterize the water
intensity of electricity.
Thisletterbuildsonpriorworktopresentregionalestimatesfortheconsumptiveupstreamand
PoGwaterintensityofelectricityintheUS,usinga4236baseyearandaregionalizationconsistent
with common emissions intensity estimates. Specifically, we analyze the US electricity system at
the generator level to estimate operational (i.e, water consumed as a result of ongoing electricity
production) life cycle water consumption from resource capture through electricity generation for
each generator reporting generation to the Energy Information Administration (EIA) [377,378],
then aggregate consumption across the regions used by the Environmental Protection Agency’s
(EPA’s) Emissions & Generation Resource Integrated Database (eGRID) [385]. This database,
which tracks greenhouse gas and other air pollutant emissions at the PoG for the US, is used
extensively for modeling and analysis. Example applications include estimating the life cycle
emissions impacts of fuels used to generate electricity [85], evaluating emissions tradeoffs of fuel
transitionsforelectricity[:3,327,332,38:,39:], electrification studies [73,:7,35;], and urbanization
[6;] studies.
Notably, our use of eGRID regionalization means that the results presented here can easily be
integrated to Life Cycle Assessment (LCA) studies in the US. LCA is a common environmental
98
assessment method used for decision support, and it is heavily reliant on data. Regionalization
[396] and data on water consumption [337] are both areas of need in LCA. Our work is compatible
with life cycle inventory (LCI) data from the US LCI database [95], which is commonly used both
directly and as a source for US data in other LCI databases. This analysis addresses essentially all
US electricity generation, enabling our contribution of regionalized consumptive water intensities
and total water consumption for both upstream and PoG uses. We also present details on the
types of water being consumed (e.g., by source and quality), which aids in evaluation of the
environmental impact of water use [98], and technological variability across regions that can
inform similar analyses elsewhere.
8.4 Methodology
We characterize total operational water consumption for electricity by US eGRID region, with
a base year of 4236 (consistent with national estimates of water use for energy from Grubert
and Sanders [75]). We note that temporal variability in water consumption is more associated
with technology changes than, e.g., precipitation. That is, our technology-specific water intensity
findings are unlikely to change substantially, but our regional water intensity findings will not
apply if fuel and technology mixes change dramatically. Here, âĂIJtotalâĂİ refers both to the
electricity’s full life cycle, from resource capture to power plant, and to water of all types. We
distinguish among twelve water classifications combining water type (surface, ground, and reuse)
and water salinity (fresh, brackish, saline, and not treatable by reverse osmosis) [75]. The focus
on operational water consumption means that water embedded in infrastructure like power plants,
pipelines, and other fixed assets is not included. We focus on water consumption (water that is
removed from its source and not directly returned) rather than water withdrawals (water that
is removed from its source, whether or not it is directly returned) largely because withdrawals,
which are driven by increasingly rare once-through cooling systems at power plants [75], are more
99
consistently reported and are relatively well understood (though note that as of the 4237 data
year, the US Geological Survey now reports consumptive use by thermoelectric power plants [57]).
Also, consumption is a metric more consistent with other water footprinting exercises [48], as
withdrawals are rarely used as a metric outside of the context of thermoelectric power plant
cooling.
Toassessregionalconsumptivewaterintensity,werelyonabottomupanalysisofUSelectricity
generation at the level of individual generators at power plants. Using federally reported data
for grid-connected US generators with capacity greater than one megawatt (MW) [377,378] and
recent studies of US consumptive water intensity by fuel [75,76] and generator type [5], we assign
water consumption associated with the fuel (upstream of the plant) and direct use (at the PoG) to
each generator. Then, we aggregate water consumption embedded in electricity over all reported
generators and generation in a given eGRID region, using the eGRID 4238 boundaries [384].
This allows us to report both total water consumption and consumptive intensity (i.e., water
consumption per unit of net electricity generation exported to the grid) by eGRID region.
We first calculate water consumption associated with producing, processing, and transporting
power plant fuels (upstream processes), which might or might not occur in the same eGRID region
as the power generation. In general, water consumption from upstream processes is calculated at
the generator level by multiplying the generator’s 4236 electric fuel consumption (from EIA Form
;45 [378]) by a fuel-specific upstream consumptive water intensity factor based on values from
Grubert and Sanders (423:) [75], which also uses a 4236 base year. Specifically, we multiply 4236
fuel consumption (FC) for generatori by the water intensity factor given by dividing total water
consumption (WC) across water typesS for upstream life cycle stagesj for electricity fuelF [75]
by total 4236 electric FC for fuel f in US power plant generators k [378], per Equation 8.3:
WC
i
=FC
i
×
X
F
X
S
Σ
j
WC
j,s,f
Σ
k
FC
k,s,f
(8.3)
9:
SM Table C.4 relates fuel codes from EIA Form ;45, Schedule 4-7 to codes used for this
research. SM Section C.3 describes how the consumptive water intensity factor (SM Table C.5)
is calculated for each fuel, including fuel classifications and allocation schemas (e.g., when a fuel
is used for purposes in addition to electricity).
After calculating upstream water consumption associated with each generator’s fuel, we cal-
culate water consumption at the PoG -âĂŤ that is, consumption by the generators at each power
plant. PoG water consumption occurs in the eGRID region where the power plant is located and
is primarily associated with cooling, but it can also include water use for e.g., scrubbers, human
use, and fire suppression [75]. PoG water consumption (WC
∗
) is calculated by multiplying fuel-
and technology-specific 4236 US water consumption rates (WCR) for generator classifications c
from Peer and Sanders (4238) [5] and Grubert and Sanders (423:) [75] by net 4236 generation
(GEN) per generator i, per Equation 8.4:
WC
∗
i
=
X
S
WCR
c,s
×GEN
i,c,s
(8.4)
Generator classifications c account for fuel, cooling system, and prime mover [5,377,378].
Data from [5] are used for natural gas, coal, and nuclear plants, while data from [75] are used for
other technologies. (Note that values from [5] are source data for natural gas, coal, and nuclear
conversion values in [75]: this work recalculates due to a focus on regionalization rather than fuel
averages as in [75].)
Individual generators are related to cooling systems via relations to the boiler(s) at each power
plant using EIA Form :82, Schedule 8-4 [377]. Generators linked to multiple cooling systems can
have multiple classifications c. For these generators, cooling water consumption is calculated by
separating the fraction of generation cooled by each cooling system and using corresponding water
consumption intensities from [5] for technology configurations matching the fuel, prime mover,
and cooling systems at each multi-cooled generator. SM Section C.4 provides additional details,
9;
and SM Table C.33 shows water intensity by generator classification. Our results do not account
for water consumption downstream of the power plant (e.g., for waste management past the power
plant gate) or water production from combustion of hydrocarbons, though see [35,75].
Aftercalculatingtotalgenerator-levelwaterconsumption(summingtheresultsofEquation8.3
and Equation 8.4), we spatially join latitude and longitude data from EIA Form :82, Schedule
4 [377] with EPA shapefiles [384] to aggregate over eGRID regions.
Hydroelectric facilities are assigned to eGRID regions slightly differently, described further in
SM Section C.3. Regional water consumption (WC
R
, in cubic meters, m
3
) and regional water
consumptionintensity(WCI
R
, inm
3
permegawatt-hour(MWh)deliveredtothegrid, (ΣGEN
R
))
for each eGRID region R are reported based on the location of electricity generators i in region
R:
WC
R
=
X
i
(WC
i
+WC
∗
i
) (8.5)
WCI
R
=
WC
R
Σ
i
GEN
i
(8.6)
Notably, these values do not correspond to totals and intensities associated with the location
of water consumption or the location of electricity consumption. The nature of electricity systems
is such that water is often virtually transferred as embodied water in a unit of input fuel or
electricity [48], which means there are multiple ways to express water intensity by region. In
this letter, we report consumptive water intensity as volume of water per unit of net electricity
generation, or electricity first exported to the grid in a given region–this can also be seen as the
virtual plus local water footprint of electricity at the power plant. That is, if water is consumed
during fuel extraction in eGRID region 3, but that fuel is then exported to eGRID region 4, the
water physically consumed in region 3 is assigned as embedded water for region 4 generation.
Similarly, if electricity produced in eGRID region 4 is exported to eGRID region 5, embedded
:2
Figure 8.3: Regional water consumption allocation approach, whereα andβ are fractions of water
consumed for primary energy preparation (WC
1
and WC
2
) and electricity generation (WC
3
)
and the red (upstream water consumption) and blue (PoG water consumption) arrows show
water consumption attributed to a power plant from inside (solid) or outside (dashed) a given
region. Each eGRID regional water consumption is calculated as all water embedded in electricity
generated within the eGRID region’s borders, which is not the same as water consumed within
the region for electricity or water embedded in electricity consumed within the region’s borders.
water in the electricity remains assigned to region 4 based on the power plant’s location (Figure
8.3).
8.5 Results and Discussion
Overall, we find that consumptive water intensity associated with US electricity varies by a factor
greater than 42 (i.e., from a low of 2.64 m
3
/MWh in AKMS to a high of ;.4 m
3
/MWh in AZNM)
across eGRID regions (Figure 8.4, Table 8.3), confirming that spatial specificity is relevant for
environmental assessments of electricity that include water consumption. SM Table C.3 defines
eGRID region acronyms, and SM Table C.34 presents total volumetric water consumption for
electricity by eGRID region.
Thermal power plants accounted for the majority (:;%) of US electricity generation in 4236
[378]. These generators are fundamentally more water intensive than most nonthermal generating
technologies, as they rely on cooling, which is typically provided with water. Coal, natural gas,
and nuclear-powered electricity (;9% of 4236 US thermal generation) collectively accounted for
93% of life cycle water consumption associated with US electricity production, about half of
:3
Figure 8.4: Regional water consumption intensity is highly variable across the country and is a
function of not only technology mix, but also geography and climate (i.e., consumption intensity
does not necessarily scale with regional generation). Each region is represented by a bar, colored
by water consumption intensity (location displayed on the map with matching color). Total water
consumption intensity and generation of each region are shown by the height and width of each
bar, respectively, where a 3×3 square on the chart represents a volume of 3×10
9
m
3
. The bars are
ordered from left to right in decreasing total volumetric consumption.
:4
Table 8.3: Regional water consumption intensity for different generating technologies
Water Consumption Intensity (m
3
/MWh)
a
eGRID
region
all thermal
b
non-
thermal
c
fossil-
fueled
d
non-hydro
renewables
e
all
renewables
f
no
combustion
g
AKGD 0.98 1.1 0.0020 1.2 0.0075 0.0018 0.0020
AKMS 0.42 2.9 0 2.9 0.011 0 0
AZNM 9.2 2.0 64 1.3 3.9 51 23
CAMX 1.6 1.6 1.3 0.82 3.7 3.4 3.1
ERCT 1.1 1.3 0.13 1.2 0.016 0.13 0.96
FRCC 1.4 1.4 0.0031 1.3 3.4 3.2 1.6
HIMS 2.2 3.0 0.0088 3.2 0.83 0.74 0.45
HIOA 3.0 3.1 0.011 3.1 2.2 2.2 0.011
MROE 2.2 1.3 17 1.2 1.3 10 4.2
MROW 2.2 1.4 4.5 1.3 0.19 4.4 3.2
NEWE 1.2 1.4 −0.72 0.94 2.4 1.2 1.1
NWPP 1.1 1.7 0.40 1.6 0.78 0.53 0.59
NYCW 1.1 1.1 0 0.81 3.7 3.7 1.4
NYLI 1.3 1.3 0.011 1.1 3.5 3.5 0.011
NYUP 0.63 1.4 −0.73 1.0 0.66 −0.56 0.42
RFCE 1.5 1.6 −0.97 1.3 1.5 0.52 1.8
RFCM 1.3 1.4 −0.82 1.2 0.76 0.27 1.8
RFCW 1.8 1.8 1.4 1.8 0.30 1.4 1.8
RMPA 1.1 1.4 −0.67 1.4 0.026 −0.65 −0.67
SPNO 1.2 1.4 0.011 1.3 0.018 0.018 0.96
SPSO 1.5 1.4 1.6 1.4 0.25 1.6 1.6
SRMV 1.8 1.5 37 1.3 2.5 15 3.9
SRMW 1.5 1.4 5.2 1.2 0.026 5.1 2.6
SRSO 1.4 1.6 −4.5 1.5 2.5 −0.55 1.0
SRTV 2.4 2.4 2.4 2.7 2.1 2.4 1.8
SRVC 1.8 1.5 18 1.4 2.4 7.4 2.2
National
h
1.8 1.6 3.8 1.5 1.3 3.9 2.6
a
Values are rounded to two significant digits.
b
Biogas, biomass, coal, geothermal, natural gas, nuclear, oil, solar
thermal.
c
Hydroelectricity, solar, wind. Includes solar thermal.
d
Coal, natural gas, oil.
e
Biogas, biomass,
geothermal, solar, wind.
f
Biogas, biomass, geothermal, hydroelectricity, solar, wind.
g
Geothermal,
hydroelectricity, nuclear, solar, wind.
h
National totals may not match generation-weighted averages presented in
table due to rounding.
:5
Figure 8.5: Water consumption (upstream, point of generation, and total), separated by fuel type
for each eGRID region (ordered from west to east). Fuel type (i.e., generation mix) is a driver
of regional water consumption variability at the PoG, whereas consumption upstream is driven
more by geology (e.g., coal, gas, and oil resource extraction) and climate (e.g., consumption from
hydroelectric reservoirs).
which is attributable to coal (5:% of water consumption and 5;% of total generation). Reservoir-
associated hydroelectricity also induced substantial water consumption though evaporation and
accounted for essentially all of the water consumption associated with nonthermal generators
(45% of electricity-associated water consumption and 8% of US electricity generation in 4236)
(Figure 8.5).
Although thermal plants drive total water consumption, fuel mix is the primary driver of
regional variability in electricity-associated consumptive water intensity (here defined as m
3
of
water consumed from fuel extraction through conversion per net MWh generated in the region)
(Table 8.3). A grid powered solely by wind and solar photovoltaic resources, for example, would
:6
not be as consumptively water intensive as a grid powered solely by water-cooled thermal gener-
ators. In general, thermal electricity is associated with higher consumptive water intensity than
renewable resources typically eligible under renewable portfolio standards (RPS), like wind and
solar, but not renewables in general or renewables relying on combustion.
This outcome is largely driven by high consumptive water intensity of reservoir-associated
hydropower (nonthermal, no combustion, and renewable, but not generally accepted as an RPS
resource) and nuclear power (thermal, no combustion, and nonrenewable) [75]. Fuel mix also
drives regional variability within fuel categories. For example, regions with higher penetration by
thermal renewables tend to have higher RPS renewable consumptive water intensity than those
primarily reliant on low-water resources like wind and solar photovoltaics. SM Tables C.35, C.36,
and C.37 show fuel mix and consumptive water intensities upstream, at PoG, and overall for each
eGRID region.
Another main finding of this work is that water consumption upstream of the point of gen-
eration is both nonnegligible and highly variable by region (Figure 8.5, Figure 8.6). Using a net
generation-weighted average, upstream water consumption accounts for about 5:% of US water
consumption for electricity generation, ranging from essentially none to essentially all of the con-
sumptive water footprint of electricity in a given eGRID region (i.e., from 5% in SPNO to 322%
in AKMS).
Part of the reason for variability shown in Figure 8.6 is that upstream consumptive water
intensity tends to be more site specific than PoG intensity. PoG intensities are usually thermally
driven (see SM Figures C.5 and C.6 for a view of generation and PoG water consumption
by cooling type), while upstream intensities are often geologically or geographically driven [75].
For example, water consumption for oil, natural gas, coal, and uranium mining depends on the
characteristicsoftheresourcedeposit, includingrockpermeabilityandwatersaturationlevels, and
evaporationfromreservoirsforhydroelectricityorirrigationneedsforbiofuelcropsdependonlocal
climate. One major implication is that when water consumption is an environmental indicator
:7
Figure 8.6: Share of upstream and point of generation water consumption in electricity’s con-
sumptive water footprint by US eGRID region, 4236. The fraction of upstream and PoG water
consumption is highly variable across regions, representing a unique combination of generating
mix, geology, and climate in each region. Note that some regions experience negative consumption
from hydropower: net negative values have been zeroed on this chart for clarity.
of interest, consumption upstream of the power plant cannot be considered negligible without
confirmation. This contrasts with typical practice regarding water withdrawals, as withdrawal
intensities are dominated by power plant cooling systems [75].
A final general finding is that most (:2-;7%) water consumption for electricity is fresh (Fig-
ure 8.7). Consistent with previous findings [363], this reliance on freshwater suggests that the
electricity sector competes with other freshwater users (e.g., agriculture, ecosystems, and mu-
nicipalities) for water access. Fresh surface water comprises the majority of water consumed
for power plant cooling and all the water evaporated from reservoirs for hydroelectricity. When
power plants use other water sources, it is typically because of fresh surface water scarcity (e.g.,
groundwater or recycled water in AZNM, CAMX, ERCT, FRCC, and SRMV) or proximity to
an ocean (e.g., FRCC, SRVC, RFCE, NEWE). Upstream, although surface water still comprises
the majority of water consumption, groundwater consumption is more common, largely because
of geologically driven needs for water removal from resource deposits like coal mines and natural
gas reservoirs [75]. Water consumption is defined in this work as the water removed from a source
:8
Figure 8.7: Water consumption (upstream, point of generation, and total), separated by water
quality (fresh, brackish, saline, not RO treatable) and type (RU = reuse, GW = groundwater,
SW = surface water) for each eGRID region (ordered from west to east). Water consumption
in every region (and nationally) is largely fresh, surface water, which is most heavily influenced
by hydroelectric (upstream) and thermoelectric (at the PoG) for power plant cooling. Ground-
water (mostly fresh) and reclaimed water (fresh and brackish) are consumed largely for resource
extraction (e.g., coal and gas) and as an alternative source for power plant cooling, respectively.
and not returned. Thus, groundwater removed during fuel extraction processes and not returned
(e.g., water removed from aquifers during natural gas or coal extraction that is released to surface
water or injected into different aquifers) is considered consumed, even if it is a nondiscretionary
byproduct rather than a required water input.
:9
8.5.3 eGRID regional variability in electricity-associated consumptive
water intensity
Average regional consumptive water intensity (m
3
of water consumed from fuel extraction through
conversion per net MWh generated in the region) is highest for AZNM and lowest for AKMS.
Thisrangedemonstratestheinfluenceofthedynamicsofevaporationfromreservoir-associatedhy-
dropowerplantsonoverallconsumptivewaterintensity(Figure8.4). Consumptionfromreservoir-
associated hydropower, defined as evaporation that would not otherwise have occurred from the
land associated with the reservoir (that is, net evaporation), is driven by regionally variable
factors like weather and land cover. This work uses net evaporation values calculated for the
United States, based on gross evaporation volumes calculated using a Penman-Monteith model
less estimated evapotranspiration (ET) volumes associated with the most common landcover at
proxy facility locations, based on National Land Cover Database mapping and landcover-specific
ET coefficients. More discussion, and the full models, can be found in [76]. Hydropower thus
has a broader range of possible consumptive water intensities than other resources in part be-
cause it can result in lower water consumption than would have otherwise occurred in a region,
as when reservoirs replace water intensive land cover like wetlands. This negative consumption
is observed in the northeast (NEWE, NYUP, RFCE), the forested southeast (SRSO), and the
northwest (NWPP, RMPA) (Table 8.3). By contrast, reservoirs in regions with limited regional
landcover evapotranspiration (ET) and high evaporation potential, notably the arid southwest
(AZNM, ERCT) and the midwest (MROW, MROE, RFCM, SRMV, SRMW), experience very
high evaporative losses from reservoirs (Table 8.3). Note that these results reflect allocation of
the evaporative burden of reservoirs across multiple purposes: see [76] for details.
Hydroelectricity is a major driver of variability on its own. A more typical profile for a given
fuel is to have either consistently high or consistently low consumptive intensity, and regional
variability is driven by variation in use of different types of fuel. For example, oil, nuclear, and
::
coalaretypicallylargewaterconsumers(SMTableC.37), whichisreflectedinhigher-than-average
consumptive water intensity in eGRID regions with high contributions by those fuels (e.g., HIOA,
HIMS, MROE, MROW, and SRTV, Table 8.3). Regions with high relative fractions of natural
gas generation (e.g., CAMX, NEWE, NYCW, RFCM) have noticeably lower water consumption
intensities for fossil fuel generation (Table 8.3), largely because many natural gas plants rely at
least partly on gas turbines (versus steam turbines) that do not need to be water cooled.
High consumptive water intensity for electricity does not always mean high local water con-
sumption or even high freshwater consumption, though. For example, Hawaii (HIOA, HIMS)
is unusually reliant on oil, which is not produced locally and is less reliant on freshwater than
most resources [75]. Similarly, high penetration by typically water-intensive fuels does not always
mean high overall consumptive water intensity for the region. SRMW and SPNO both have high
penetration by coal and nuclear facilities, but neither has higher than average overall consumptive
water intensity due to the relatively low upstream water consumption intensity of coal used in
these regions (See SM Table C.35).
For SPNO, high wind penetration (37% of 4236 generation) also lowers the overall average
water intensity of generation, given that wind consumes essentially no operational water [75]. In
general, just as some fuels tend to drive consistently high water consumption, wind and solar
photovoltaic use tends to drive low water consumption (SM Table C.37). With the exception of
HIMS (with its oil drivers) and MROW (with its hydroelectric drivers), all eGRID regions with
wind penetration above 32% in 4236 had lower-than-average consumptive water intensity (ERCT,
RMPA, SPNO, SPSO). For solar, the effect of higher penetration in lowering water consumption
is somewhat masked by the fact that the largest solar-using regions (AZNM and CAMX, at 7%
and 5% of 4236 generation, respectively) also use unusually water-intensive hydroelectric and
geothermal resources. Growth in use of wind and solar photovoltaic resources is expected to
continue to outpace growth in use of other resources [8,9], however, so these fuels might more
clearly lower regional consumptive water intensity of electricity in the future.
:;
Variability in regional cooling system profiles is a secondary driver of the consumptive water
intensity of electricity. Wet-cooled facilities display relatively low variation (i.e., same order of
magnitude) in consumptive water intensity across generating technologies, but hybrid and dry-
cooled facilities consume significantly less water per unit of output (SM Table C.33). Note that
some variability in cooling systems is fuel mix driven: some generation does not require cooling,
including wind, solar photovoltaics, and gas turbines using natural gas or biogas (SM Table C.33).
Additional context can be found in SM Section C.5.
8.5.4 Data uncertainties and limitations
This work is largely based on modern estimates of water intensity from [75] and [5], which
represent up-to-date, best-guess assessments of consumptive water use for the energy system.
Comparing data from these publications to other published estimates is challenged by the fact
that many other estimates are based on older information that does not reflect current technology.
Thus, a precise quantitative estimate of uncertainty in the values presented in this letter is not
possible. Major data limitations in this work are thus similar to limitations articulated by [75] and
[5]. Most significantly, information about water consumption in general is not measured, centrally
tracked, or published, so data are based on a combination of empirical, self-reported, and derived
values based on physical relationships. Uncertainty is high, but the nature of that uncertainty is
unknown given serious limits to data availability and quality. In general, estimates for processes
that are not thermodynamically driven (e.g., process water for coal cleaning) are more uncertain
than estimates for processes that are thermodynamically driven, as the physical relationship
between process and water consumption is less clear. Hydroelectricity-related consumption is
thermodynamically driven, but data are highly sensitive to assumptions about allocation across
multicriteria purposes and dam location [76]. PoG water consumption estimates are likely the
most accurate, given some reporting requirements and multiple methods for triangulating true
valuesbasedonphysicalandotherrelationships. Foragivenfuelandagiventechnology, estimates
;2
can be checked against values reported for other similar plants, thermodynamic laws, and EIA
reports.
In addition to data quality-related uncertainty, this work is affected by the choice in [75] to
conservatively overestimate freshwater consumption versus consumption of other water qualities,
by assuming water quality is fresh when it is not known. The impact of this choice is estimated
to be small, based on a general preference for freshwater in industrial and other contexts. Fur-
ther, uncertainty associated with carried-through ambiguity related to units and definitions in
the source data, discussed in more depth in [75] and [5], is also present in this work. Thus,
overall data uncertainty is high but difficult to productively quantify. Note that given reporting
requirements related to fuel use and outputs from electricity generators, uncertainty related to
regional electricity production is very low.
8.5.5 Policydriversandimplicationsofvariabilityinwaterconsumption
from electricity
The values presented in this work are based on technology and fuel mix conditions from 4236.
These conditions are likely to change, however, in part due to changing policies. Policy can
influence consumptive water intensity associated with electricity. As this work describes, fuel mix
and cooling systems at thermal power plants affect water consumption intensity. Both are policy
targets. For example, fuel mixes are increasingly affected by Renewable Portfolio Standards in
many US states, which aim to increase the amount and/or share of renewable generation. Federal
tax credits similarly foster investment in various fuels. Cooling systems are affected by policy like
Section 538(b) of the Clean Water Act, which aims to reduce the impact of water withdrawals
on aquatic organisms [33:]. State implementation varies, leading to further differentiation. For
example, California has implemented 538(b) by establishing cooling system technology-based
standards aimed at reducing withdrawals of ocean and estuarine surface water, which has tended
;3
to shift once-through ocean water cooling to non-water cooled or recirculating systems in the
state [37]. Through Section 538(a), the Clean Water Act also sets operationally relevant limits on
temperature increases due to water use for cooling [:8,;5], which influences choices about cooling
system technologies. Although variability in regional consumptive water intensity for electricity
is related to circumstance (e.g., fuel availability), geology (e.g., fossil fuel deposit characteristics),
and geography (e.g., air and water temperature), it is also affected by policy.
Although this letter specifically investigates the US, many of the same conditions affecting
water consumption from electricity in the US apply elsewhere [;;,33:,354]. Just as policy can
drive water consumption associated with electricity, water consumption can drive policy, par-
ticularly in response to pressures like drought [365], strained water supply [358], rising water
and air temperatures [:8,;3,33:], and others. In considering potential future policy changes,
one implication of this work is that water consumption from processes upstream of generators
is both large enough and regionally variable enough to motivate specific consideration of local-
ized data. PoG water consumption is relatively similar from place to place after accounting for
fuel and technology, but accurately assessing upstream consumption requires more specific data.
Expanding the existing literature on water for electricity, future policy-motivated analysis might
productively investigate regionally specific temporal patterns of water consumption from electric-
ity, higher spatial resolution, different intensity metrics (e.g., water consumption physically in a
region or water consumption per unit of electricity consumed in a region), and more regions (and
see, e.g., [66,86,96,9:,354]).
8.6 Conclusion
Electricitysystemsconsumelargevolumesofwater. Basedonthisanalysisand[75], upstreamand
PoG water consumption associated with electricity systems collectively accounts for about 7% of
total US water consumption. Given that agriculture accounts for about 97% of total (nonrainfall)
;4
US water consumption [:9], and given that much of the water associated with electricity systems
is fresh surface water in relatively populated areas, this is a substantial volume. The volume of
water consumed by electricity systems is driven by thermal power plants, and particularly by
PoG water consumption from cooling systems. Consumptive water intensity varies by region,
however, and this variability is largely driven by fuel mix. The consumptive water intensity of
hydroelectricityinparticularishighlyregionallyvariablebecauseoftheinfluenceofgeographyand
climate on evaporation from reservoirs. Beyond hydroelectricity, fuel mix differentiates regions
due to variability in water required for a given fuel (e.g., coal versus wind). Notably, water
consumption upstream of the power plant is highly regionalized, while water consumption at the
PoG tends to be similar for similar technologies. One implication is that studies of electricity’s
water intensity cannot easily adapt upstream water intensity values from one context to another
even for the same fuel, while they can adapt PoG water intensity values if the technologies are
similar. Policy affects consumptive water intensity of electricity, particularly via regulation about
fuel mix and access to cooling water. In turn, changing drought, scarcity, temperature, and other
conditions might tend to affect policy making regarding water consumption for electricity in the
future.
;5
Chapter 9
Conclusion
The research described throughout this dissertation focuses on developing methods for quantify-
ing the water resource impacts of electricity generation on spatial and temporal scales relevant
for holistic planning, decision-making and engineering modeling. Water is required across the
electricity sector: for thermoelectric power plant cooling, hydroelectric generation, and primary
fuel preparation. This water comes from a variety of sources and can be of different quality, both
of which are important metrics for water management. Prior to this work, the magnitude, as
well as the spatial, and temporal distributions of this water use was not well studied. This body
of work addresses several knowledge and data gaps in the water-electricity literature, notably it
answers the following research questions:
• How much water is used to cool thermoelectric power plants and how does this vary across
generating technologies?
• Howhaverecentchangesintheenergysector(e.g., frackingandhorizontaldrilling)impacted
the electricity sector and water use for electricity across the US?
• How can the environmental impacts of electricity generation be holistically modelled on
relevant time scales?
;6
• What is the life cycle (i.e., from fuel extraction to electricity generation) water consumption
embedded in electricity regionally across the US?
Chapter 5 statistically analyzed how the technological characteristics of a generating unit
affect the cooling water use for thermoelectric power plants in the US, based on self-reported
data. This work characterized 894 unique power plants, representing ;4%, ;2%, and :7% of
total nuclear, coal, and natural gas-fired generation in 4236, and increasing the representation of
power plants across the US by one to two orders of magnitude across generating technologies.
Chapter 6 extended this work by quantifying the water withdrawal and consumption conse-
quences of changes in the electric sector between 422: and 4236, capturing the years following
the US “fracking boom.” This study noted shifts from coal-fired steam to natural gas-fired com-
bined cycle generating units, from one-through cooling systems to recirculating and dry cooling
systems, and from fresh and saline to reclaimed water sources which resulted in a more water
withdrawal efficient grid per unit of electricity generated, on average. The shift toward reclaimed
water sources is a particularly interesting conclusion of this work, as previous literature under-
estimated the fraction of generation cooled from this water source by approximately a factor of
:.
Chapter 7 developed a methodology to quantify water and emissions with high temporal res-
olution using the Texas Interconnection as a case study. This work enabled holistic modelling of
multiple environmental externalities with novel insights in the time variability of water use and
emissions and environmental impacts of retrofitting generating technologies to natural gas com-
binedcycleunits. Thisworkservesasabenchmarkforfuturestudiesevaluatingtheenvironmental
trade-offs of electricity generation, particularly on relevant time scales.
Chapter 8 calculated regionally-specific life cycle (i.e., from resource extraction to electricity
generation) water consumption volumes and intensities for the US electricity sector. These values
canbeintegratedintolifecycleanalysis(LCA)oftheelectricitysectorandarecompatiblewithlife
cycle inventory (LCI) data, which already incorporate other environmental impacts of electricity
;7
generation, notably carbon emissions. The results of this work show that there are nonnegligable
differences in water consumption across regions. These differences are mostly driven by water
consumption for primary fuel preparation upstream of the power plant, which is influenced by
geography, and hydroelectric generation and vary as much as a factor of 42 between regions.
Collectively,thisbodyofworkofferswaterandenergyplanners,researchersandotherpertinent
stakeholders data and methods to better account for the water required by the electricity system.
Climate forcing causes uncertainty in the future spatio-temporal distribution and magnitude of
water availability. Without representative data and modelling tools, it is difficult to accurately
quantify and qualify the environmental impacts of the electricity system in a changing climate.
Furthermore, as the electricity grid continues to change in response to not only climatic, but
also policy and economic drivers, understanding the water requirements of a new electricity grid
becomes increasingly important for grid resiliency. Examining these impacts from a life cycle
perspective is also important for resource management and accountability, particularly as fuel
and water resources increase in scarcity. In all, this work presents a valuable collection of data
and methods that can be used to continue to assess the water requirements of the electricity sector
under uncertain future water availability, grid configuration, and climate with high spatial and
temporal specificity.
;8
Reference List
[3] ENERGY-WATER NEXUS: Improvements to Federal Water Use Data Would Increase
UnderstandingofTrendsinPowerPlantWaterUseENERGY-WATERNEXUSHighlights.
Technical Report October, United States Government Accountability Office, 422;.
[4] Construction starts on first post-combustion carbon capture project in US, 4236.
[5] Rebecca A M Peer and Kelly T Sanders. Characterizing cooling water source and usage
patternsacrossUSthermoelectricpowerplants: acomprehensiveassessmentofself-reported
cooling water data. Environmental Research Letters, 33(34):346252, 4238.
[6] Robert H Abrams and Noah D Hall. Framing Water Policy in a Carbon Affected and
Carbon Constrained Environment. Natural Resources Journal, 72(3):5–92, 4232.
[7] Frank Ackerman and Jeremy Fisher. Is there a waterâĂŞenergy nexus in electricity gener-
ation? Long-term scenarios for the western United States. Energy Policy, 7;:457–463, aug
4235.
[8] Energy Information Administration. Annual Energy Outlook 423:, 423:.
[9] Energy Information Administration. Electricity Data Browser, 423:.
[:] Nawaf S Alhajeri, Pearl Donohoo, Ashlynn S Stillwell, Carey W King, Mort D Webster,
Michael E Webber, and David T Allen. Using market-based dispatching with environ-
mental price signals to reduce emissions and water use at power plants in the Texas grid.
Environmental Research Letters, 8(6):26623:, oct 4233.
[;] K Averyt, J Macknick, J Rogers, N Madden, J Fisher, J Meldrum, and R Newmark. Water
use for electricity in the United States: an analysis of reported and calculated water use
information for 422:. Environmental Research Letters, :(3):237223, mar 4235.
[32] KAveryt, JMeldrum, PCaldwell, GSun, SMcNulty, aHuber-Lee, andNMadden. Sectoral
contributions to surface water stress in the coterminous United States. Environmental
Research Letters, :(5):257268, sep 4235.
[33] Jonathan Baker, Kenneth Strzepek, William Farmer, and C. Adam Schlosser. Quantifying
the Impact of Renewable Energy Futures on Cooling Water Use. JAWRA Journal of the
American Water Resources Association, 24377:n/a–n/a, apr 4236.
[34] Lawrence W. Barnthouse. Impacts of entrainment and impingement on fish populations: A
review of the scientific evidence. Environmental Science & Policy, 53:36;–378, aug 4235.
[35] E. L. Belmont, F. T. Davidson, Y. R. Glazer, E. A. Beagle, and M. E. Webber. Accounting
for water formation from hydrocarbon fuel combustion in life cycle analyses. Environmental
Research Letters, 34(;):2;623;, 4239.
;9
[36] Reed D Benson. So Much Conflict, Yet So Much in Common: Considering the Similarities
between Western Water Law and the Endangered Species Act. Natural Resources Journal,
66:4;–98, 4226.
[37] California State Water Quality Control Board. CWA 538(b) Thermal Discharges-Cooling
Water Intake Structures, 423:.
[38] Nathan Bracken. Water and energy in the west: the legal and institutional issues that affect
water availability for energy-related activities. Technical Report April, 4237.
[39] Nathan Bracken, Jordan Macknick, Angelica Tovar-hastings, Paul Komor, Margot Ger-
ritsen, and Shweta Mehta. Concentrating Solar Power and Water Issues in the U . S .
Southwest. Technical Report March, Western States Water Council, 4237.
[3:] John Burkhardt. Life Cycle Assessment of a Parabolic Trough Concentrating Solar Power
Plant. 4232.
[3;] E A Byers, J W Hall, J M Amezaga, G M O’Donnell, and A Leathard. Water and climate
risksto powergenerationwith carbon capture and storage. Environmental Research Letters,
33(4):246233, feb 4238.
[42] Edward a. Byers, Jim W. Hall, and Jaime M. Amezaga. Electricity generation and cooling
water use: UK pathways to 4272. Global Environmental Change, 47:38–52, mar 4236.
[43] Ximing Cai, Kevin Wallington, Majid Shafiee-Jood, and Landon Marston. Understanding
and managing the food-energy-water nexus âĂŽÃĎÃň opportunities for water resources
research. Advances in Water Resources, 333:47;–495, January 423:.
[44] Ken Caldeira, Atul K Jain, and Martin I Hoffert. Climate sensitivity uncertainty and the
need for energy without CO4 emission. Science (New York, N.Y.), 4;;(7837):4274–6, mar
4225.
[45] California Independent System Operator Corporation. 4236 Summer Load & Resources
Assessment. Technical report, 4236.
[46] Munish K. Chandel, Lincoln F. Pratson, and Robert B. Jackson. The potential impacts of
climate-change policy on freshwater use in thermoelectric power generation. Energy Policy,
5;(32):8456–8464, oct 4233.
[47] Jiangang Chen, Mohammed H Al-Wadei, Rebekah C M Kennedy, and Paul D Terry. Hy-
draulic Fracturing: Paving the Way for a Sustainable Future? Journal of environmental
and public health, 4236:878:46, 4236.
[48] Christopher M. Chini, Lucas A. Djehdian, William N. Lubega, and Ashlynn S. Stillwell.
Virtual water transfers of the US electric grid. Nature Energy, 5(34):3337, December 423:.
[49] SClemmer, JRogers, SSattler, JMacknick, andTMai. Modelinglow-carbonUSelectricity
futures to explore impacts on national and regional water use. Environmental Research
Letters, :(3):237226, mar 4235.
[4:] Heather Cooley, Julian Fulton, and Peter Gleick. Water for Energy : Future Water Needs
for Electricity in the Intermountain West. Technical report, 4233.
[4;] Kerstin Damerau, Keith Williges, Anthony G. Patt, and Paul Gauché. Costs of reducing
water use of concentrating solar power to sustainable levels: Scenarios for North Africa.
Energy Policy, 5;(9):65;3–65;:, jul 4233.
;:
[52] Steven J Davis, Ken Caldeira, and H Damon Matthews. Future CO4 emissions and climate
change from existing energy infrastructure. Science, 465338(September):422:–4233, 4232.
[53] Tyler A. DeNooyer, Joshua M. Peschel, Zhenxing Zhang, and Ashlynn S. Stillwell. Integrat-
ing water resources and power generation: The energyâĂŞwater nexus in Illinois. Applied
Energy, 384:585–593, 4238.
[54] Department of Energy. Estimating Freshwater Needs to Meet Future Thermoelectric Gen-
eration Requirements. Technical report, 422:.
[55] Department of Energy. U.S. energy sector vulnerabilities to climate change and extreme
weather. Technical report, Department of Energy, 4235.
[56] T.H. Diehl and M.A. Harris. Withdrawal and Consumption of Water by Thermoelectric
Power Plants in the United States , 4232 Scientific Investigations Report 4236 âĂŞ 73:6.
Technical report, 4236.
[57] Cheryl A. Dieter, Molly A. Maupin, Rodney R. Caldwell, Melissa A. Harris, Tamara I.
Ivahnenko, John K. Lovelace, Nancy L. Barber, and Kristin S. Linsey. Estimated use of
water in the United States in 4237. USGS Numbered Series 3663, U.S. Geological Survey,
Reston, VA, 423:. IP-2;265;.
[58] Rosa Domenichini, Silvio Arienti, Paolo Cotone, and Stanley Santos. Evaluation and anal-
ysis of water usage and loss of power in plants with CO4 capture. Energy Procedia, 6:3;47–
3;54, jan 4233.
[59] Eia. FORM EIA-;45 POWER PLANT OPERATIONS REPORT INSTRUCTIONS.
[5:] Joshua D. Eichman, Fabian Mueller, Brian Tarroja, Lori Smith Schell, and Scott Samuelsen.
Exploration of the integration of renewable resources into California’s electric system using
the Holistic Grid Resource Integration and Deployment (HiGRID) tool. Energy, 72(3):575–
585, 4235.
[5;] Energy Information Administration. California drought leads to less hydropower, increased
natural gas generation, 4236.
[62] Environmental Protection Agency. Air Markets Program Data | US EPA, 4236.
[63] Environmental Protection Agency. Index of Air Markets Program Data, 4236.
[64] Paul Faeth. The Impacts of EPA ’ s Clean Power Plan on Electricity Generation and Water
Use in Texas. Technical Report November, CNA, 4236.
[65] Thomas J. Feeley, Timothy J. Skone, Gary J. Stiegel, Andrea McNemar, Michael Nemeth,
Brian Schimmoller, James T. Murphy, and Lynn Manfredo. Water: A critical resource in
the thermoelectric power industry. Energy, 55(3):3–33, jan 422:.
[66] Kuishuang Feng, Klaus Hubacek, Yim Ling Siu, and Xin Li. The energy and water nexus
in chinese electricity production: A hybrid life cycle analysis. Renewable and Sustainable
Energy Reviews, 5;:564 – 577, 4236.
[67] Lesley Fleischman, Rachel Cleetus, Jeff Deyette, Steve Clemmer, and Steve Frenkel. Ripe
for retirement: An economic analysis of the U.S. coal fleet. Electricity Journal, 48(32):73–
85, 4235.
;;
[68] Oliver Fricko, Simon C Parkinson, Nils Johnson, Manfred Strubegger, Michelle TH van
Vliet, and Keywan Riahi. Energy sector water use implications of a 4 ÂřC climate policy.
Environmental Research Letters, 33(5):256233, mar 4238.
[69] Jared Brett Garrison. A Grid-Level Unit Commitment Assessment of High Wind Penetra-
tion and Utilization of Compressed Air Energy Storage in ERCOT. PhD thesis, University
of Texas at Austin, 4236.
[6:] Jared Brett Garrison, Stuart M. Cohen, Aaron Keith Townsend, and Michael E. Webber.
Developmentandbenchmarkingofaunitcommitmentmodelforamodernelectricitymarket
(under review). International Journal of Electric Power and Energy Systems, 4237.
[6;] EdwardL.GlaeserandMatthewE.Kahn. Thegreennessofcities: Carbondioxideemissions
and urban development. Journal of Urban Economics, 89:626–63:, 4232.
[72] Peter Gleick. Impacts of California ’ s Ongoing Drought: Hydroelectricity Generation.
Technical Report March, 4237.
[73] Joshua S. Graff Zivin, Matthew J. Kotchen, and Erin T. Mansur. Spatial and temporal
heterogeneity of marginal emissions: Implications for electric cars and other electricity-
shifting policies. Journal of Economic Behavior and Organization, 329(PA):46:–48:, 4236.
[74] K. B. Gregory, R. D. Vidic, and D. a. Dzombak. Water Management Challenges Associated
with the Production of Shale Gas by Hydraulic Fracturing. Elements, 9(5):3:3–3:8, jun
4233.
[75] Emily Grubert and Kelly T. Sanders. Water Use in the United States Energy System: A
National Assessment and Unit Process Inventory of Water Consumption and Withdrawals.
Environmental Science and Technology, 74(33):88;7–8925, 423:.
[76] EmilyA.Grubert. WaterconsumptionfromhydroelectricityintheUnitedStates. Advances
in Water Resources, ;8(July):::–;6, 4238.
[77] Emily a Grubert, Fred C Beach, and Michael E Webber. Can switching fuels save water?
A life cycle quantification of freshwater consumption for Texas coal- and natural gas-fired
electricity. Environmental Research Letters, 9(6):267:23, dec 4234.
[78] S. Hardy. The Endangered Species Act: on a collision course with human needs. The Public
Land Resources Law Review, 35::9–;:, 3;;4.
[79] C.B. Harto, Y. E. Yan, Y.K. Demissie, Deborah Elcock, Vincent C. Tidwell, K.C. Hallett,
Jordan Macknick, M.S. Wigmosta, and T.K. Tesfa. Analysis of drought impacts on electric-
ity production in the Western and Texas Interconnections of the United States. Technical
report, 4233.
[7:] Arjen Y. Hoekstra and Mesfin M. Mekonnen. The water footprint of humanity. Proceedings
of the National Academy of Sciences, 32;(;):5454–5459, February 4234.
[7;] ICF Jones & Stokes, Global Energy Decisions, and Matt Trask. Electric Grid Reliability
Impacts from Regulation of Once-Through Cooling in California. Technical Report April,
422:.
[82] Boris Iglewicz and David Hoaglin. How to Detect and Handle Outliers. In Edward F.
Mykytka, editor, Volume 38 of ASQC Basic References in Quality Control: Statistical Tech-
niques, page :9. ASQC Quality Press, 3;;5.
322
[83] International Panel on Climate Change. IPCC AR7 Report on Climate Change. Technical
report, 4236.
[84] Wendy S. Jaglom, James R. McFarland, Michelle F. Colley, Charlotte B. Mack, Boddu
Venkatesh, Rawlings L. Miller, Juanita Haydel, Peter a. Schultz, Bill Perkins, Joseph H.
Casola, Jeremy a. Martinich, Paul Cross, Michael J. Kolian, and Serpil Kayin. Assessment
of projected temperature impacts from climate change on the U.S. electric power sector
using the Integrated Planning Model®. Energy Policy, pages 3–38, jul 4236.
[85] Paulina Jaramillo, W. Michael Griffin, and H. Scott Matthews. Comparative life-cycle air
emissions of coal, domestic natural gas, LNG, and SNG for electricity generation. Environ-
mental Science and Technology, 63(39):84;2–84;8, 4229.
[86] Daqian Jiang, Anuradha Ramaswami, Hall J W Byers E A, Amezaga J M, CCTV News,
Robins N Chan W-S, Knight Z, Xu Y Chen H, He Y, Cohen E A, Ramaswami, Crompton P
Y,Wu, LeeCLeeJDennenB,LarsonD,TellinghuisenS,StiegelGJJrMcNemarANemeth
M Schimmoller B Murphy J T Feeley T J I I I, Skone T J, Manfredo L, Fthenakis V C,
Kim H, Förster H J, Lilliestam, Gnansounou E J, Dong, Kahrl F D, Roland-Holst, Hutson
S S Linsey K S Lovelace J K Kenny J F, Barber N L, Maupin M A, Kimmell T A A, Veil J,
Koch H S, Vögele, Koulouri A J, Moccia, Li H N, Mo, Liu J J, Diamond, Heath G Macknick
J, Newmark R, Hallett K, Marsh D M, Hutson S S Lovelace J K Barber N L Maupin M
A, Kenny J F, Linsey K S, National Bureau of Statistics of China, National Development
(NDRC), Reform Commission, Rio Carrillo A M C, Frei, Clemmer S Davis M Flores-Lopez
F Frumhoff P Kenney D Macknick J Madden N Rogers J, Averyt K, Meldrum J, Shiu A
P-L, Lam, Pierce R R Solley W B, Perlman H A, Sovacool B K, Sovacool B K E, Sovacool
K, Malczynski L A Klise G Tidwell V C, Kobos P H, Castillo C R, Wang B, Conway D
Zhang L Xiong W Holman I P Wang J, Rothausen S G A, Li Y, Woldeyesus T A, Yang X
B, Dziegielewski, Zoebisch M A Chen G Xie G Zhen L, Routray J K, and Cheng S. The
âĂŸthirsty’ water-electricity nexus: field data on the scale and seasonality of thermoelectric
powergeneration’swaterintensityinChina. Environmental Research Letters,32(4):246237,
feb 4237.
[87] Arturo a. Keller, Stacy Tellinghuisen, Cheryl Lee, Dana Larson, Bliss Dennen, and James
Lee. Projection of California’s Future Freshwater Requirements for Power Generation. En-
ergy & Environment, 43(4):3–42, 4232.
[88] Iftekhar Khan, Firoz Alam, and Quamrul Alam. The global climate change and its effect
on power generation in Bangladesh. Energy Policy, 83:3682–3692, oct 4235.
[89] Carey W. King, Ian J Duncan, and Michael E. Webber. Water Demand Projections for
Power Generation in Texas, Report prepared for the Texas Water Development Board.
Technical report, 422:.
[8:] Carey W King, Ashlynn S Stillwell, Kelly M Twomey, and Michael E Webber. Coherence
Between Water and Energy Policies. Natural Resources Journal, 75:339–437, 4235.
[8;] Christopher Knittel, Konstantinos Metaxoglou, and Andre Trindade. Are we fracked? The
impact of falling gas prices and the implications for coal-to-gas switching and carbon emis-
sions. Oxford Review of Economic Policy, 54(4):463–47;, apr 4238.
[92] Christopher R. Knittel, Konstantinos Metaxoglou, and Andre Trindade. Natural Gas Prices
and Coal Displacement: Evidence from Electricity Markets. Technical report, National
Bureau of Economic Research, Massachusetts, MA, 4237.
323
[93] Natalie Kopytko and John Perkins. Climate change, nuclear power, and the adapta-
tionâĂŞmitigation dilemma. Energy Policy, 5;(3):53:–555, jan 4233.
[94] Page Kyle, Evan G.R. Davies, James J. Dooley, Steven J. Smith, Leon E. Clarke, James A.
Edmonds, and Mohamad Hejazi. Influence of climate change mitigation technology on
global demands of water for electricity generation. International Journal of Greenhouse
Gas Control, 35:334–345, mar 4235.
[95] National Renewable Energy Laboratory. U.S. Life Cycle Inventory Database, 4234.
[96] Morten Andreas Dahl Larsen and Martin Drews. Water use in electricity generation for
water-energy nexus analyses: The European case. Science of The Total Environment,
873:4266–427:, February 423;.
[97] Uisung Lee, Jeongwoo Han, Amgad Elgowainy, and Michael Wang. Regional water con-
sumption for hydro and thermal electricity generation in the United States. Applied Energy,
432:883–894, 423:.
[98] Susana LeâĹŽÂčo, Philippe Roux, Montserrat NâĹŽâĹńâĹŽÂśez, ElâĹŽÂľonore Loiseau,
Guillaume Junqua, Agata Sferratore, Ywann Penru, and Ralph K. Rosenbaum. A
worldwide-regionalised water supply mix (WSmix) for life cycle inventory of water use.
Journal of Cleaner Production, 394:524–535, January 423:.
[99] Shuya Li, David W Coit, Saltuk Selcuklu, and Frank A Felder. Electric Power Generation
Expansion Planning : Robust Optimization Considering Climate Change. In Proceedings
of the 4236 Industrial and Systems Engineering Research Conference, 4236.
[9:] Xiawei Liao, Li Chai, Junping Ji, Zhifu Mi, Dabo Guan, and Xu Zhao. Life-cycle water uses
forenergyconsumptionofChinesehouseholdsfrom4224to4237. JournalofEnvironmental
Management, 453:;:;–;;7, February 423;.
[9;] Xi Lu, Michael B McElroy, Gang Wu, and Chris P Nielsen. Accelerated reduction in
SO4 emissions from the U.S. power sector triggered by changing prices of natural gas.
Environmental science & technology, 68(36):9::4–;, jul 4234.
[:2] Xi Lu, Jackson Salovaara, and Michael B McElroy. Implications of the Recent Reductions
in Natural Gas Prices for Emissions of CO4 from the US Power Sector. Environmental
Science & Technology, 68(7):5236–5243, 4234.
[:3] RogerLueken, KellyKlima, W.MichaelGriffin, andJayApt. Theclimateandhealtheffects
of a USA switch from coal to gas electricity generation. Energy, 32;:3382–3388, 4238.
[:4] J Macknick, R Newmark, G Heath, and K C Hallett. Operational water consumption and
withdrawal factors for electricity generating technologies : a review of existing literature.
Environmental Research Letters, 9(267:24):3–32, 4234.
[:5] J Macknick, S Sattler, K Averyt, S Clemmer, and J Rogers. The water implications of
generating electricity: water use across the United States based on different electricity
pathways through 4272. Environmental Research Letters, 9(6):267:25, dec 4234.
[:6] Jordan Macknick, Robin Newmark, Garvin Heath, and K C Hallett. A Review of Opera-
tional Water Consumption and Withdrawal Factors for Electricity Generating Technologies
A Review of Operational Water Consumption and Withdrawal Factors for Electricity Gen-
erating Technologies. Technical Report March, 4233.
324
[:7] NathanD.Macpherson,GregoryA.Keoleian, andJarodC.Kelly. FuelEconomyandGreen-
house Gas Emissions Labeling for Plug-In Hybrid Vehicles from a Life Cycle Perspective.
Journal of Industrial Ecology, 38(7):983–995, 4234.
[:8] N Madden, a Lewis, and M Davis. Thermal effluent from the power sector: an analy-
sis of once-through cooling system impacts on surface water temperature. Environmental
Research Letters, :(5):257228, sep 4235.
[:9] Landon Marston, Yufei Ao, Megan Konar, Mesfin M. Mekonnen, and Arjen Y. Hoekstra.
High-Resolution Water Footprints of Production of the United States. Water Resources
Research, February 423:.
[::] M. a. Maupin, J. F. Kenny, S.S. Hutson, John K Lovelace, N. L. Barber, and K. S. Linsey.
Estimated use of Water in the United States in 4232 Circular 3627. Technical report, 4236.
[:;] J Meldrum, G Heath, and J Macknick. Life cycle water use for electricity generation : a
review and harmonization of literature estimates. 237253.
[;2] J Meldrum, S Nettles-Anderson, G Heath, and J Macknick. Life cycle water use for electric-
ity generation: a review and harmonization of literature estimates. Environmental Research
Letters, :(3):237253, mar 4235.
[;3] Ariel Miara, Jordan E. Macknick, Charles J. VâĹŽâĹĆrâĹŽâĹĆsmarty, Vincent C. Tid-
well, Robin Newmark, and Balazs Fekete. Climate and water resource change impacts and
adaptation potential for US power supply. Nature Climate Change, 9(33):9;5–9;:, October
4239.
[;4] Ariel Miara, Charles J Vörösmarty, Robert J Stewart, Wilfred M Wollheim, and Bernice
Rosenzweig. Riverine ecosystem services and the thermoelectric sector: strategic issues
facing the Northeastern United States. Environmental Research Letters, :(4):247239, jun
4235.
[;5] Ariel Miara, Charles J VâĹŽâĹĆrâĹŽâĹĆsmarty, Robert J Stewart, Wilfred M Woll-
heim, and Bernice Rosenzweig. Riverine ecosystem services and the thermoelectric sec-
tor: strategic issues facing the northeastern united states. Environmental Research Letters,
:(4):247239, 4235.
[;6] Torben K. Mideksa and Steffen Kallbekken. The impact of climate change on the electricity
market: A review. Energy Policy, 5:(9):579;–57:7, jul 4232.
[;7] N.L. Miller, K. Hayhoe, J. Jin, and M. Auffhammer. Climate, extreme heat, and electricity
demand in california. Journal of applied meteorology and climatology, 69:3:56–3:66, 4229.
[;8] M Granger Morgan, Ahmed Abdulla, Michael J Ford, and Michael Rath. US nuclear power
: The vanishing low-carbon wedge. Proceedings of the National Academy of Sciences, pages
3–8, 423:.
[;9] Ioanna Mouratiadou, Anne Biewald, Michaja Pehl, Markus Bonsch, Lavinia Baumstark,
David Klein, Alexander Popp, Gunnar Luderer, and Elmar Kriegler. The impact of climate
change mitigation on water demand for energy and food: An integrated analysis based on
the Shared Socioeconomic Pathways. Environmental Science & Policy, 86:6:–7:, 4238.
[;:] Pierre Mukheibir. Potential consequences of projected climate change impacts on hydro-
electricity generation. Climatic Change, 343(3):89–9:, sep 4235.
325
[;;] Mesfin M.ÂňâĂăMekonnen, P. W.ÂňâĂăGerbens-Leenes, and Arjen Y.ÂňâĂăHoekstra.
The consumptive water footprint of electricity and heat: a global assessment. Environ-
mental Science: Water Research & Technology, 3(5):4:7–4;9, 4237.
[322] V. Nanduri and W. Otieno. A New Water and Carbon Conscious Electricity Market Model
for the Electricity-Water-Climate Change Nexus. Electricity Journal, 46(;):3–9, 4233.
[323] National Energy Regulatory Comission. Potential Impacts of Future Environmental Reg-
ulations. Technical Report November, North American Electric Reliability Corporation,
4233.
[324] National Institute of Standards and Technology and SEMATECH. Standard and Expanded
Uncertainties. In Engineering Statistics Handbook, chapter 4. 4234.
[325] National Integrated Drought Information System. U.S. Drought Portal.
[326] National Research Council. Hidden Costs of Energy: Unpriced consequences of Energy
Production and Use. Technical report, 4232.
[327] Adam Newcomer, Seth A. Blumsack, Jay Apt, Lester B. Lave, and M. Granger Morgan.
Short Run Effects of a Price on Carbon Dioxide Emissions from U.S. Electric Generators.
Environmental Science & Technology, 64(;):535;–5366, may 422:.
[328] United States Army Corps of Engineers. CorpsMap: The National Inventory of Dams
(NID), April 4237.
[329] Yang Ou, Haibo Zhai, and Edward S. Rubin. Life cycle water use of coal- and natural-gas-
fired power plants with and without carbon capture and storage. International Journal of
Greenhouse Gas Control, 66:46;–483, 4238.
[32:] Pacific Gas and Electric. RETIREMENT OF DIABLO CANYON POWER PLANT ,
IMPLEMENTATION OF THE JOINT PROPOSAL , AND RECOVERY OF ASSOCI-
ATED COSTS THROUGH PROPOSED RATEMAKING MECHANISMS. Technical re-
port, 4238.
[32;] Adam P Pacsi, Nawaf S Alhajeri, Mort D Webster, Michael E Webber, and David T Allen.
Changing the spatial location of electricity generation to increase water availability in ar-
eas with drought: a feasibility study and quantification of air quality impacts in Texas.
Environmental Research Letters, :(5):25724;, sep 4235.
[332] Adam P. Pacsi, Nawaf S. Alhajeri, Daniel Zavala-Araiza, Mort D. Webster, and David T.
Allen. Regional Air Quality Impacts of Increased Natural Gas Production and Use in Texas.
Environmental Science & Technology, 69(4):5743–5749, mar 4235.
[333] Adam P Pacsi, Kelly T Sanders, Michael E Webber, and David T Allen. Spatial and
Temporal Impacts on Water Consumption in Texas from Shale Gas Development and Use.
4236.
[334] Rebecca; Peer, Jared; Garrison, and Kelly Sanders. A Spatially and Temporally Resolved
Analysis of Environmental Trade-Offs in Electricity Generation. Environ. Sci. Technol., in
press, 4238.
[335] Rebecca A M Peer and Kelly T Sanders. The water consequences of a transitioning US
power sector. Applied Energy, In Press, 4239.
326
[336] Garielle Pétron, Pieter Tans, Gregory Frost, Danlei Chao, and Michael Trainer. High-
resolution emissions of CO 4 from power generation in the USA. Journal of Geophysical
Research, 335(G6):G2622:, oct 422:.
[337] Stephan Pfister, Samuel Vionnet, Tereza Levova, and Sebastien Humbert. Ecoinvent 5:
assessing water use in LCA and facilitating water footprinting. The International Journal
of Life Cycle Assessment, 43(;):356;–3582, September 4238.
[338] Anna Mercè Rio Carrillo and Christoph Frei. Water: A key resource in energy production.
Energy Policy, 59(33):6525–6534, 422;.
[339] Gary T Rochelle. Amine scrubbing for CO4 capture. Science (New York, N.Y.),
547(7;6:):3874–6, sep 422;.
[33:] Kelly T Sanders. Critical Review: Uncharted Waters? The Future of the Electricity- Water
Nexus. Environmental science & technology, 4237.
[33;] Kelly T Sanders, Michael F Blackhurst, Carey W King, and Michael E Webber. The impact
of water use fees on dispatching and water requirements for water-cooled power plants in
Texas. Environmental science & technology, 6:(34):934:–56, jun 4236.
[342] Jayant a. Sathaye, Larry L. Dale, Gary a. Fitts, Kevin Koy, Sarah M. Lewis, and André
Frossard Pereira de Lucena. Estimating Risk to California Energy Infrastructure from
Projected Climate Change. Technical report, California Energy Comission, 4233.
[343] Jayanta.Sathaye, LarryL.Dale, PeterH.Larsen, Garya.Fitts, KevinKoy, SarahM.Lewis,
and André Frossard Pereira de Lucena. Estimating impacts of warming temperatures on
California’s electricity system. Global Environmental Change, 45(4):6;;–733, apr 4235.
[344] Bridget R Scanlon, Robert C Reedy, Ian Duncan, William F Mullican, and Michael Young.
Controls on water use for thermoelectric generation: case study Texas, US. Environmental
science & technology, 69(3;):33548–56, oct 4235.
[345] Christopher a. Scott, Suzanne a. Pierce, Martin J. Pasqualetti, Alice L. Jones, Burrell E.
Montz, and Joseph H. Hoover. Policy and institutional dimensions of the waterâĂŞenergy
nexus. Energy Policy, 5;(32):8844–8852, oct 4233.
[346] Christine Shearer, John Bistline, Mason Inman, and Steven J. Davis. The effect of nat-
ural gas supply on US renewable energy and CO4 emissions. Environmental Research,
;(2;622:):3–:, 4236.
[347] Jeannette Sieber. Impacts of, and adaptation options to, extreme weather events and
climate change concerning thermal power plants. Climatic Change, 343(3):77–88, sep 4235.
[348] Kyle Siler-Evans, Inês Lima Azevedo, and M Granger Morgan. Marginal emissions factors
for the U.S. electricity system. Environmental science & technology, 68(;):6964–:, may
4234.
[349] S. Soft. Power System Economics: Designing Market for Power. IEEE Press, Piscataway,
NJ, 4224.
[34:] Benjamin K. Sovacool. Running on empty: the electricity-water nexus and the U.S. electric
utility sector. Energy Law Journal, 52(33):33–74, 422;.
[34;] Benjamin K Sovacool and Alex Gilbert. Developing adaptive and itegrated strategies for
managing the electricity-water nexus. University of Richmond Law Review, 6:(;;9):;;9–
3254, 4236.
327
[352] Benjamin K. Sovacool and Kelly E. Sovacool. Identifying future electricityâĂŞwater trade-
offs in the United States. Energy Policy, 59(9):4985–4995, jul 422;.
[353] Benjamin K Sovacool and Kelly E Sovacool. Preventing National Electricity-Water Crisis
Areas in the United States. Columbia Journal of Environmental Law, 56:555–5;5, 422;.
[354] E S Spang, W R Moomaw, K S Gallagher, P H Kirshen, D H Marks, Barker B, Whiteman A
Cushion E, Dieterle G, DOE, EIA, EIA, EIA, Elcock D, Fthenakis V C, Kim H, Gleick P H,
Hoekstra A Y K, Chapagain A, IAEA, Heath G Macknick J, Newmark R, Hallett K C, Av-
eryt K Clemmer S Macknick J, Sattler S, Rogers J, Mekonnen M M Y, Hoekstra A, Heath G
Meldrum J, Nettles-Anderson S, Macknick J, Anadon L Mielke E, Narayanamurti V, Mit-
tal A, Hagens N Mulder K, Fisher B, NETL, NREL, Orcutt M, Platts, USGS, Vassolo S
P, Döll, WCD, WEC, WEC, WISE, Wang M Wu M, Mintz M, Arora S, Yang X B, and
Dziegielewski. The water consumption of energy production: an international comparison.
Environmental Research Letters, ;(32):327224, oct 4236.
[355] Robert J Stewart, Wilfred M Wollheim, Ariel Miara, Charles J Vörösmarty, Balazs Fekete,
RichardBLammers, andBerniceRosenzweig. Horizontalcoolingtowers: riverineecosystem
services and the fate of thermoelectric heat in the contemporary Northeast US. Environ-
mental Research Letters, :(4):247239, jun 4235.
[356] Ashlynn S Stillwell, Mary E Clayton, and Michael E Webber. Technical analysis of a
river basin-based model of advanced power plant cooling technologies for mitigating water
management challenges. Environmental Research Letters, 8(5):256237, jul 4233.
[357] Ashlynn S. Stillwell and Michael E. Webber. A novel methodology for evaluating economic
feasibility of low-water cooling technology retrofits at power plants. Water Policy, 37(4):3–
54, 4234.
[358] Ashlynn S Stillwell and Michael E Webber. Evaluation of power generation operations in
response to changes in surface water reservoir storage. Environmental Research Letters,
:(4):247236, jun 4235.
[359] Ashlynn S Stillwell and Michael E Webber. Geographic, technologic, and economic analysis
of using reclaimed water for thermoelectric power plant cooling. Environmental science &
technology, 6:(:):67::–;7, apr 4236.
[35:] Kenneth Strzepek, Jonathan Baker, William Farmer, C Adam Schlosser, and Report No.
Modeling Water Withdrawal and Consumption for Electricity Generation in the United
States. Technical Report 444, 4234.
[35;] Mili Ann M. Tamayao, Jeremy J. Michalek, Chris Hendrickson, and Ineîs M.L. Azevedo.
Regional variability and uncertainty of electric vehicle life cycle CO4 emissions
across the United States. Environmental Science and Technology, 6;(36):::66–::77, 4237.
[362] Texas Water Development Board. Historical Water Use Estimates, 4233.
[363] Vincent Tidwell and Barbie Moreland. Mapping water consumption for energy production
around the Pacific Rim. Environmental Research Letters, 33(;):35, sep 4238.
[364] Vincent C Tidwell, Peter H Kobos, Len A Malczynski, Geoff Klise, and Cesar R Castillo.
Exploring the Water-Thermoelectric Power Nexus. Journal of Water Resources Planning
and Management, 35::6;3–723, 4234.
328
[365] Vincent C. Tidwell, Jordan Macknick, Katie Zemlick, Jasmine Sanchez, and Tibebe Wold-
eyesus. Transitioning to zero freshwater withdrawal in the U.S. for thermoelectric genera-
tion. Applied Energy, 353:72:–738, oct 4236.
[366] Vincent C Tidwell, Leonard a Malczynski, Peter H Kobos, Geoffrey Taylor Klise, and Erik
Shuster. Carbon Capture and Sequestration: Potential Impacts on US Water Resources.
Environmental Science & Technology, 69::;62–:;69, 4235.
[367] Vincent C Tidwell, Barbara D Moreland, Katie M Zemlick, Barry L Roberts, and Howard
Passell. Water as a Constraint for New Thermoelectric Generation in the WECC. Technical
report.
[368] Aaron Keith Townsend. A Grid-Level Assessment of Compressed Air Energy Storage in
ERCOT. PhD thesis, University of Texas at Austin, 4235.
[369] Union of Concerned Scientists. The Energy-Water Collision: 32 Things You Should Know.
Technical report, Union of Concerned Scientists, 4232.
[36:] Union of Concerned Scientists. UCS EW5 Energy-Water Database V.3.5. Technical report,
4234.
[36;] U.S. Energy Information Administration. 422: Form EIA-:82 Data. Technical report,
422;.
[372] U.S. Energy Information Administration. 422: From EIA-;45 Data. Technical report,
422;.
[373] U.S. Energy Information Administration. 4235 Form EIA-;45 Instructions Appendix for
Schedule :D. Technical report, 4235.
[374] U.S. Energy Information Administration. 4235 Form EIA-;45 Data. Technical report,
4236.
[375] U.S. Energy Information Administration. FORM EIA-:82 INSTRUCTIONS ANNUAL
ELECTRIC GENERATOR REPORT. Technical report, 4236.
[376] U.S. Energy Information Administration. FORM EIA-;45 POWER PLANT OPERA-
TIONS REPORT INSTRUCTIONS, 4236.
[377] U.S. Energy Information Administration. 4236 Form EIA-:82 Data. Technical report,
4237.
[378] U.S. Energy Information Administration. 4236 Form EIA-;45 Data. Technical report,
4237.
[379] U.S. Energy Information Administration. Monthly Energy Review - October 4238. Tech-
nical report, 4238.
[37:] U.S. Energy Information Administration. U.S. Energy Information Administration (EIA) -
Electricity Data, 4238.
[37;] U.S. Energy Information Administration. 4239 Form EIA-;45 Data. Technical report,
423:.
[382] U.S. Energy Information Administration (EIA). 4237 Uranium Marketing Annual Report.
(May):3–75, 4237.
329
[383] U.S. Energy Information Administration (EIA). Annual Energy Outlook 423; with projec-
tions to 4272. Technical report, 423;.
[384] U.S. Environmental Protection Agency. eGRID4238 Shapefiles, 4238.
[385] U.S. Environmental Protection Agency. Emissions & Generation Resource Integrated
Database (eGRID), 4238.
[386] U.S. Environmental Protection Agency Office of Air. Current Nonattainment Counties for
All Criteria Pollutants, 4237.
[387] U.S. Global Change Research Program. Global climate change impacts in the United States.
Cambridge University Press, New York, NY, 422;.
[388] Michelle T H van Vliet, Stefan Vögele, and Dirk Rübbelke. Water constraints on European
power supply under climate change: impacts on electricity prices. Environmental Research
Letters, :(5):257232, sep 4235.
[389] Avner Vengosh, Robert B Jackson, Nathaniel Warner, Thomas H Darrah, and Andrew
Kondash. A Critical Review of the Risks to Water Resources from Shale Gas Develop-
ment and Hydraulic Fracturing in the United States. Environmental science & technology,
38(6):8:5:, 4236.
[38:] Aranya Venkatesh, Paulina Jaramillo, W Michael Griffin, and H Scott Matthews. Impli-
cations of near-term coal power plant retirement for SO4 and NOX and life cycle GHG
emissions. Environmental science & technology, 68(3:):;:5:–67, sep 4234.
[38;] Edward Vine. Adaptation of California’s electricity sector to climate change. Climatic
Change, 333(3):97–;;, oct 4234.
[392] Mort Webster, Pearl Donohoo, and Bryan Palmintier. Water-CO4 trade-offs in electricity
generation planning. Nature Climate Change, 5(34):324;–3254, oct 4235.
[393] James H Williams, Andrew Debenedictis, Rebecca Ghanadan, Amber Mahone, Jack Moore,
William R Morrow Iii, Snuller Price, and Margaret S Torn. 4272 : The Pivotal Role of
Electricity. Science (New York, N.Y.), 557(April):75–82, 4234.
[394] World Resource Institute. Aqueduct | Water Risk, 4236.
[395] A.W. Wortzel. Energy versus water: the gorwing role of water in controlling energy deci-
sions. University of Richmond Law Review, 6::3273–3284, 4236.
[396] Yi Yang. Toward a more accurate regionalized life cycle inventory. Journal of Cleaner
Production, 334:52:–537, January 4238.
[397] D Yates, J Meldrum, and K Averyt. The influence of future electricity mix alternatives on
southwestern US water resources. Environmental Research Letters, :(6):267227, dec 4235.
[398] D Yates, J Meldrum, F Flores-Lopez, and Michelle Davis. Integrated impacts of future
electricity mix scenarios on select southeastern US water resources. Environmental Research
Letters, :(5):257264, sep 4235.
[399] David Yates and Kathleen A Miller. Climate Change : Impacts and Responses the South-
west. 6(3).
[39:] Katie Zemlick, Vincent C Tidwell, Barry L Roberts, and Cesar R Castillo. the Electricity
Generation Demands in 4252. Journal of Contemporary Water Research & Education,
(373):;7–327, 4235.
32:
[39;] Haibo Zhai and Edward S. Rubin. Performance and cost of wet and dry cooling systems for
pulverized coal power plants with and without carbon capture and storage. Energy Policy,
5:(32):7875–7882, oct 4232.
[3:2] Haibo Zhai and Edward S. Rubina. Carbon capture effects on water use at pulverized coal
power plants. Energy Procedia, 6:445:–4466, jan 4233.
[3:3] Xiaochun Zhang, Nathan P Myhrvold, and Ken Caldeira. Key factors for assessing climate
benefits of natural gas versus coal electricity generation. Environmental Research Letters,
;(33):336244, nov 4236.
32;
Appendix A
Supplemental information for chapter 4: Using
self-reported data to build a water use rate database for
US power plants
Data used in this analysis are publicly available on the EIA website for both forms :82 and
;45 [377,378]. Instructions for EIA Form :82, Form ;45, and an illustrated appendix for cooling
systems for from ;45 schedule :D are available via the EIA website [373,375,376].
A.3 Nationwide Cooling Water System and Source
Characterization Analysis
Cooling water system, water source, and water quality data were compiled based on EIA forms
:82/;45 Schedules :D, 8D, and 4, as described in the main body of this manuscript. However,
cooling system and cooling water data, especially for smaller generators were missing for a fraction
of generators. After the manual analysis of EIA data was complete, approximately 9% and 7% of
all generation produced in thermoelectric power plants requiring cooling listed multiple cooling
systems or no cooling system technology, respectively, making it difficult to allocate generation
across specific cooling water types. Generation at facilities with multiple wet cooling systems was
split equally across all fuel-prime mover categories, introducing some estimation error. Generators
with both dry-cooling and wet-cooling systems were small in percentage, and thus, evaluated on
a case by case basis to make allocations of generation for each system. Thermoelectric generators
missing cooling system data were assigned cooling systems based on the fractional breakdown of
cooling systems from all reporting facilities in the US.
About34%ofall4236thermoelectricgenerationwasmissingcoolingsourcetypeand/orquality
data, but more than ;2% of this generation had at least one of the two identifiers (i.e. type or
quality), narrowing down error in the estimation of missing data significantly. For example, many
generators listed “wells” in the Schedule 4 form, but did not indicate quality. Thus, the source
could be identified as groundwater, but quality had to be estimated. Several assumptions were
employed in assigning estimates to missing data:
• Generators that identified their cooling systems as once-through or cooling ponds were never
assigned reclaimed water as a cooling medium.
• Generators reporting cooling towers as a cooling source were not assigned brackish or saline
water quality sources, since these sources typically cause substantial fouling without expen-
sive pre-treatment.
• When multiple cooling sources were listed, generation was split evenly across each source.
332
• Whenno“educated”guesscouldbemaderegardingcoolingsourcetypeorquality, fractional
breakdowns were assumed based on nationally well-characterized generation with complete
cooling information.
The EIA chooses a selection of monthly respondents each year to report detailed monthly data
to reduce the data reporting burden for the whole set of power plants required to report to the
EIA. Thus, while some monthly data are measured and reported by power plants, most monthly
data are estimates calculated by the EIA based on this smaller subset of reporting facilities [375].
Annual data, by contrast, are actual reported data, so we use these values for the majority of the
analysis.
333
Table A.3: Level of Confidence in Characterizing Cooling Sources by Total Annual Generation (TWh)
Cooling Water Classification Confidence
Water Source Source Type High (;7% Confidence) Medium (92% Confidence) Low (52% Confidence) Grand Total
Dry-cooled Dry-cooled 324 ::.3% 7.:9 7.26% 9.;7 8.:5% 338± 9.99
Groundwater
Recycled 5.;4 322% 5.;4± 2.3;8
Brackish 5.58 322% 5.58± 2.38:
Freshwater 3:6 9;.:% ;.23 5.;3% 59.8 38.5% 452± 52.2
Other 8.39 322% 8.39± 2.52:
Saline 6.55 322% 6.55± 2.439
No Cooling No Cooling 786 322% 786± 4:.4
Other
Freshwater 32.; 322% 32.;± 2.766
Other 43.4 ;5.8% 3.72% 3.34 6.;5% 44.9± 3.54
Saline 2.472 322% 2.472± 2.2347
Plant Discharge Recycled 396 :8.2% 46.6 34.3% 5.;7 3.;7% 424± 33.9
Surface Water
Brackish 343 :8.4% 3;.4 35.8% 2.354% 363± :.59
Freshwater 4493 :8.7% 424 9.8:% 375 7.:5% 4848± 389
Saline 382 ;:.6% 2.:94 2.759% 3.93 3.28% 384± :.2:
Grand Total 5849 :6.9% 483 8.5:% 428 7.24% 62;6
334
Filling data gaps for generators that did not report cooling system technology type and/or
cooling water source data introduced some estimate error into the analysis. To track estimation
errorthroughouttheanalysis, threeconfidenceestimateswereassignedtoeachcategoryincluding,
“High”, “Medium” and “Low”, corresponding to ;7%, 92% and 52% confidences, respectively.
The ‘High” confidence category described generation that was produced in facilities that reported
full cooling system technology and sources. A percentage of these generators had one missing
cooling system or source identifier that could be estimated with high confidence from other infor-
mation (e.g. a generator listing “Colorado River” as its cooling source was assumed to use fresh
surface water). The “Medium” confidence plants generally had multiple wet cooling systems, such
that generation was split evenly across cooling system types. Although dry cooled plants were
analyzed in more detail, power plants with dry and wet systems were also captured in this cat-
egory to give conservative estimates of uncertainty. Generation captured in the “Low” category
represented generators that had missing data records that were assigned based on the fractional
breakdowns of cooling systems or cooling water sources according to prime mover classifications.
Thus, if 42% of generation from steam generation facilities were cooled with Recirculating with
Induced Draft Cooling Towers, 42% of the categorized generation in this category was assigned
to that cooling system. Overall, about :;%, 8% and 7% of 4236 generation fell into the “High”,
“Medium” and “Low” confidence categories, respectively. Categorical breakdowns of generation
included in this national analysis are included in Tables A.3 and A.4.
335
Table A.4: Level of Confidence in Characterizing Cooling System Technologies by Total Annual Generation (TWh)
Cooling Water Classification Confidence
Cooling System High (;7% Confidence) Medium (92% Confidence) Low (52% Confidence) Grand Total
No Cooling 786 322% 786± 4:.4
Dry (air) cooling System 324 ::.3% 7.:9 7.26% 9.;7 8.:5% 338± 7.;4
Hybrid: recirculating with forced
draft cooling tower(s) with dry
cooling
:.4; 322% :.4;± 2.636
Hybrid: recirculating with induced
draft cooling tower(s) with dry
cooling
9.7: :8.4% 3.44 35.:% :.:2± 2.749
Recirculating with Induced Draft
Cooling Tower
342: :8.5% ;;.7 9.33% ;4.: 8.85% 3622± 94.;
Recirculating with Natural Draft
Cooling Tower
665 ::.5% 55.9 8.94% 47.3 7.22% 723± 47.7
Recirculating with Cooling Ponds 4:6 :9.4% 45.5 9.37% 3:.5 7.84% 548± 38.:
Once through with Cooling Ponds :5.; 9;.:% 38.2 37.4% 7.3: 6.;5% 327± 8.78
Once through without cooling
pond(s) or canal(s)
;48 :9.2% :5.2 9.:2% 77.3 7.3:% 3286± 77.3
Grand Total 5849 ::.8% 483 8.5:% 428 7.24% 62;5
336
A total uncertainty value was assigned to each cooling system technology category, as well as
cooling source type and quality category. Uncertainty estimates considered anticipated error in
estimatingthevaluesofmissingdatarecords; thus, uncertaintywasnotassignedtovaluesreported
by power plants to the EIA. Uncertainty was calculated with expression: U
tot
=(
P
g
2
×x
2
)
1/2
,
where g is the generation within each category and x refers to the estimated classification of
confidence in each power plant data record, with the expressionx = 3-confidence [324]. “High”,
“Medium” and “Low” confidence categories were assigned x values of 7%, 52% and 92% as
very conservative estimates, respectively. Total uncertainty estimates for each category, U
tot
are
summarized in Tables A.3 and A.4.
Table A.5: 4236 US Power Generation by Cooling Source Type
Cooling Source Type 4236 Generation (billion kWh)
Surface Water 4;4;.4 93.78%
Groundwater 46:.4 8.28%
Plant Discharge (e.g. Reclaimed) 424.4 6.;6%
Dry-cooled 338.5 4.:6%
Other Source 55.: 2.:47%
No Cooling Required (Non-thermal generators) 786.2 35.:%
Total 62;5 322%
Table A.6: 4236 US Power Generation by Cooling Source Quality
Cooling Source Quality 4236 Generation (billion kWh)
Freshwater 4:89.6 92.2%
Reclaimed 428.3 7.25%
Saline 388.: 6.2:%
Brackish 366.4 5.74%
Dry-cooled 338.5 4.:6%
Other Quality 4:.: 2.92%
No Cooling Required (Non-thermal generators) 786.2 35.:%
Total 62;5 322%
TablesA.5and A.6presentsasummaryoftotal4236generationbasedonthecoolingfluidtype
and quality utilized by US generation units producing electricity. Most thermoelectric generators
in the US are cooled using fresh surface water. Approximately 7% of generation is cooled with
reclaimed water and 5% of generation is cooled using dry cooling (Table A.5, A.6). A breakdown
of dry cooling at thermoelectric power plants in the US is shown in Table A.7. Nevada (38%),
California (36%), and New York (35%) represent the highest fraction of dry-cooled generation
in the country. The use of dry cooling in both the eastern and western regions is likely due
to policy (CWA 538(A)&(B)), to address issues of thermal discharge from existing wet-cooled
infrastructure as well as the entrapment, entrainment, and impingement of aquatic species in
water intake structures.
A geospatial analysis was performed in this study on a HUC-: unit level using the data
provided in the ;45 and :82 forms. For this analysis, all reporting plants with capacity greater
than 3 MW were included in the 4236 generation map (see Figure 5.3 in main text). Power
plants reporting exclusively (i.e. only one cooling system type or cooling water quality type) the
specified cooling system and water quality were used for the four remaining maps in Figure 3
of the manuscript. Figure 5.4, detailing water withdrawals and consumption by volume in each
337
Table A.7: Dry Cooled Power Generation in US by State
State 4236 Dry-cooled generation (MWh)
AZ 38;,8;2 2.38%
CA 37,8:8,784 36%
CO 3,:24,9;8 3.9%
CT 6,743,268 6.4%
FL 768 2.22%
IL 5:,769 2.26%
LA 523,2;2 2.4:%
MA 32,375,677 ;.6%
ME 2 2.22%
MN 63,924 2.26%
MO 59,93; 2.25%
MS 4,:8:,;:: 4.8%
NJ 8,27:,:49 7.8%
NM 4,927,346 4.7%
NV 39,689,3:7 38%
NY 36,32;,299 35%
OH 5:4,:4: 2.57%
PA 5,:56,74: 5.7%
RI 3,675,97; 3.5%
SC ;4,733 2.2;%
SD 57:,9;; 2.55%
TN 2 2.22%
TX 8,939,:;3 8.4%
UT 4,6;9,65; 4.5%
VA 9,998,938 9.4%
WA 4,765,9:7 4.5%
WY 8,95:,;;3 8.4%
Total 32:,57;,822 322%
HUC-: unit, includes all plants reporting water use values to the EIA, regardless of cooling system
configuration, fuel use, or prime move classification (i.e. includes some plants reporting multiple
cooling systems, multiple fuels, and/or multiple prime movers). If a facility reported zero volume
forwithdrawalorconsumption, medianwateruseratescalculatedfromthe4236datawereapplied
where possible (i.e. if the power plant configuration was detailed in the main study) to calculate
water withdrawal and consumption volumes.
A.4 Plant-by-Plant Water Withdrawal and Consumption
Rate Analysis
The total amount of generation characterized for the calculation of operational water use rates in
this analysis, both including and excluding zero-values and outliers, is summarized in Table A.8.
In addition to zero and outlier data points, power plants with multiple fuels (i.e., defined as the
primary fuel representing less than ;7% of total generation), prime movers (excluding combined-
cycle facilities), and/or cooling systems were not characterized in this study regardless if cooling
338
water data was reported. In total, 866 water withdrawal rates and 6;; water consumption rates
(including all non-zero and non-outlier values) were calculated, representing data from a total
of 894 unique power plants, surpassing data availability in previous government and academic
studies as well as previous studies of earlier EIA data (i.e. the 422: EIA data studied by Averyt
et al. [;] contained water use data reported for at most 58: unique power plants, excluding
zero-values). (Additional facilities reported zero-value or outlier water use rates, but would have
skewed water withdrawal and consumption rates if not excluded.)
Table A.8: US 4236 water usage data that was characterized in water withdrawal and water
consumption rate analysis as a percentage of total generation.
Actual 4236
Generation
(TWh)
4236 Generation
Analyzed (TWh)
Non-outlier, non-
zero generation
(TWh)
Fuel
Coal 3722 36:3 ;;% 3574 ;2%
Natural gas 3267 ;98 ;5% :;2 :7%
Nuclear 9:7 9:7 322% 947 ;4%
Other 3;8 2 2% 2 2%
Prime
Mover
Steam 46;; 4528 ;4% 4372 :8%
Combined cycle 3248 ;58 ;3% :79 :6%
Cooling
System
Once-through only 3269 3237 ;9% ;9: ;5%
Recirculating only 3;75 3;2; ;:% 3946 ::%
Hybrid 38 38 322% 38 322%
Dry Cooling 322 :8 :8% :8 :8%
Multiple 632 2 2% 2 2%
CHP
Status
Yes 529 396 79% 388 76%
No 543: 5277 ;7% 4984 :8%
Figure A.3 (top) illustrates all of the thermoelectric generation units required to report their
annual generation in the EIA;45 data form. While all generators greater than3 MW are required
to report electric generation, those less than 322 MW nameplate capacity are not required to
report water usage volumes, and thus, generally do not. Data points in black represent facilities
that reported no water usage. These units were typically smaller than 472 MW or had low net
generation in 4236. (Figure A.3 (bottom) shows the subset of these thermoelectric generation
units less than 472 MW.) Larger plants that were excluded from this analysis were mostly power
plants using both coal and natural gas.
Additional analysis was performed to asses the influence of pollution controls on water use
requirements for coal and natural gas generators; however, the results provided no discernible
trends in water use rates. Although parasitic loads can increase the water requirements of power
generators substantially [77,366], the differences in power plant characteristics (e.g. age, technol-
ogy and pollution control configurations, cooling source characteristics, etc.), as well as the wide
range of couplings between various pollution control technologies, made it difficult to parse out
meaningful trends across technologies in this meta-analysis.
Aregressionanalysiswasperformedtoinvestigatetheroleofgenerationunitefficiencyonwater
usage. Average monthly heat rate was calculated based on monthly primary fuel consumption for
electricity and monthly electricity generation. Monthly water usage rates were calculated based
339
on monthly water consumption, monthly water withdrawals, and monthly electricity generation.
Consequently, each power generation unit included in Figure A.4 and Figure A.5 has34 associated
markers. AnoutlieranalysisusingmodifiedZ-scoreswasperformedonreportedwateruserates; all
reported zero-values and outliers (absolute value of Z-score greater than5) were excluded from the
regression in Figure A.4. Heat rates outside of 7-42 MMBTU/MWh were also considered outliers
for this regression analysis. No outlier analysis was performed for the regression in Figure A.5,
but reported zero-values were omitted. The resulting plots and R-squared values were determined
based on various power plant technology configurations and are detailed in Figures A.4 and A.5.
Results in Figure A.4 did not show a strong correlation between average water withdrawal and
consumption rates and monthly average heat rate. This result is not very surprising as the
regression was very dependent on accurate reporting of monthly primary fuel usage, generation,
and volumetric water usage data. However, results in Figure A.5 showed stronger correlation
for consumption rates and heat rates, particularly for coal and natural gas steam generating
technologies.
The efficiency of a power plant (i.e. heat rate) is moderated by how effectively a power plant
is able to remove heat from the steam exiting a steam turbine. In the absence of any temperature
differential between the inlet of the turbine (where hot steam enters) and the outlet of the turbine
(where steam exits), the turbine would not move and electricity would not be generated, since the
pressure differential between a turbineâĂŹs inlet and outlet is driven by the temperature differ-
ential. Thus, colder cooling water is more effective at removing heat than warmer cooling water.
(The Carnot Efficiency is the simplest way to express cooling water temperatureâĂŹs impact
on a generation unitâĂŹs efficiency for an ideal power plant.) Consequently, water consumption
rates would be expected to increase as a plant becomes less efficient because more water would be
required to remove a unit of thermal energy. Put another way, a power plant on a hot day would
require more water to produce a unit of electricity than the same power plant on a cool day. This
relationship might not be as obvious for high water withdrawal plants, but thermodynamically
it would hold for recirculating (closed-loop) cooled plants where water consumption approaches
water withdrawals
A.5 Comparison to literature and heat budget models
Generally, calculated water withdrawal and consumption rates for this study were comparable to
previous heat budget analysis and literature review. Table A.9 offers comparison to these studies
discussedinthemainmanuscript[56,:4]. Discrepanciesbetweenvaluesreportedinthisstudyand
others in the literature were noticeable for once-through and pond-cooled (both recirculating and
open-loop) generating technologies, especially for water withdrawal rates. In general, the facilities
reporting large withdrawal rates are co-located with large, fast-moving water sources, such as the
Mississippi River, and have very low capacity factors; thus, these facilities withdraw large volumes
of water to cool relatively small amounts of generation (i.e. a high water withdrawal rate). These
discrepancies suggest that there is possibly a lack of understanding of actual operational water
usage for plants operating with once-through (or once-through like) cooling technologies.
33:
Figure A.3: (Top) The distribution of all water-cooled thermoelectric generation plants with
capacity of 3 MW or greater in the US shows that small capacity plants or low generating plants
represent the majority of non-reported values for this study (represented in black). (*Larger
capacity plants that were not characterized in the final filtered dataset generally had multiple
cooling system types and/or multiple fuels, typically dual natural gas and coal units.) (Bottom)
The subset of total thermoelectric generation units above, filtered to those units with capacity of
3-472 MW in the US. The smallest capacity plants (less than 322 MW nameplate capacity) are
almost entirely non-reporting facilities, since they are not required to report.
33;
Figure A.4: A regression analysis, exploring the impact of generation unit heat rate and water
usage, did not indicate strong correlation when zero-values and outliers were omitted.
342
10 20 30 40
0e+00 2e+06 4e+06
10 15 20 25
0 5000 10000 15000
5 10 15 20 25 30
0 50000 100000
5 10 15 20 25 30
0 40000 80000 120000
6 8 10 12 14
0e+00 2e+05 4e+05
6 8 10 12 14
0 500 1500 2500
10 30 50 70
0.0e+00 1.0e+07 2.0e+07
15 20 25
0 5000 15000
10 30 50
0 50000 100000
10 30 50
0 2000 6000 10000
15 20 25 30
0e+00 2e+06 4e+06
15 20 25 30
0 500 1000 1500 2000
10 15 20 25 30 35 40
5000 10000 20000
10.5 10.7 10.9
0.5 0.6 0.7 0.8
0 10 20 30 40 50 60
0 500 1500 2500
0 10 20 30 40 50 60
0 500 1000 2000
11.0 12.0 13.0
0 2000 6000 10000
9 11 13 15
0 100 300 500
10.46 10.48 10.50
50000 100000 150000
10.46 10.48 10.50
0 2000 4000 6000
10.45911 10.45911
2000 6000 10000
10.45911 10.45911
500 1000 1500 2000
10.45911 10.45911
0e+00 2e+05 4e+05 6e+05
10.45911 10.45911
200 400 600 800
Heat Rate (MMBTU/MWh)
Withdrawal (gal/MWh) Withdrawal (gal/MWh) Withdrawal (gal/MWh) Consumption (gal/MWh) Consumption (gal/MWh) Consumption (gal/MWh)
Once-Through Cooled Recirculating Tower Cooled Pond Cooled
Coal
Natural Gas Steam
Natural Gas
Combined Cycle
Nuclear
r
2
= 0.377
n = 71
r
2
= 0.042
n = 32
r
2
= 0.042
n = 184
r
2
= 0.097
n = 183
r
2
= 0.063
n = 27
r
2
= 0.017
n = 22
r
2
= 0.041
n = 21
r
2
= 0.306
n = 4
r
2
= 0.007
n = 11
r
2
= 0.482
n = 11
r
2
= 0.043
n = 10
r
2
= 0.152
n = 10
r
2
= 0.018
n = 7
r
2
= 0.028
n = 2
r
2
= 0.0004
n = 173
r
2
= 0.0002
n = 175
r
2
= 0.029
n = 8
r
2
= 0.095
n = 8
r
2
= 0.00
n = 28
r
2
= 0.004
n = 8
r
2
= 0
n = 22
r
2
= 0
n = 22
r
2
= 0
n = 9
r
2
= 0
n = 4
Figure A.5: A regression analysis, exploring the impact of generation unit heat rate and water
usage, showed a stronger correlation for consumption rate of once-through cooled technologies
when outliers were included, compared to excluding outliers.
343
Table A.9: Comparison of operational water use rates across the current study, Macknick et al. (4234) [:4], and USGS (4236) [56] studies.
For this comparison, all recirculating tower cooling system classifications were considered “recirculating tower(s).”
Cooling Prime CHP Median Water Withdrawal (Gal/MWh) Median Water Consumption (Gal/MWh)
System Fuel Mover Status This Study n Macknick n USGS n This Study n Macknick n USGS n
Once-Through
no pond
Nuclear Steam No 59,;46 3: 66,572 6 5,922 - 585 7 48; 6 622 -
Coal Steam No 63,328 325 49,2:: 5 54,222 - 426 64 335 5 552 -
Coal Steam Yes 82,;62 9 - 2 - 2 3,;;9 3 - 2 - 2
Natural Gas Steam No 33:,6;2 43 57,222 3 5;,222 - 547 : 462 4 662 -
Combined Cycle No 4:,229 5 33,5:2 4 33,222 - 3:: 7 322 5 342 -
Combined Cycle Yes 62,;52 3 - 2 - 2 - 2 - 2 - 2
Once-through
pond/canal
Nuclear Steam No 54,597 3 - 2 - 2 373 3 - 2 - 2
Coal Steam No 52,68; 32 39,;36 5 - 2 562 7 99; 5 - 2
Natural Gas Steam No 364,975 6 - 2 - 2 694 4 - 2 - 2
Combined Cycle No 62,2;4 8 7,;72 3 - 2 3,42; 4 462 3 - 2
Recirculating
pond/canal
Nuclear Steam No 53,866 6 9,272 4 9:2 - 9:7 4 832 4 9:2 -
Coal Steam No 52,:6; 39 39,;35 5 872 - 54: 35 99; 5 872 -
Steam Yes 49 3 - 2 - 2 - - - 2 - 2
Natural Gas Steam No 356,365 7 - 2 4,3;2 - 686 6 - 2 43;2 -
Combined Cycle No 55,:;4 4 7,;72 3 572 - 543 4 462 3 572 -
Combined Cycle Yes 35,36; 3 - 2 - 2 3: 3 - 2 - 2
Recirculating
tower
Nuclear Steam No 3,5:2 42 3,323 5 :32 - 898 44 894 8 7:2 -
Coal Steam No 769 ;3 7:9 : 852 - 6:4 ;: 69; 9 652 -
Steam Yes 7:8 35 - 2 - 2 6:9 36 - 2 - 2
Natural Gas Steam No 3,457 46 3,425 4 :42 - :55 45 :48 6 7:2 -
Combined Cycle No 488 379 477 9 4;2 - 43: 374 427 8 422 -
Combined Cycle Yes 429 4: - 2 - 2 3:5 4; - 2 - 2
Combined Cycle Single Shaft No 449 3; - 2 - 2 427 8 - 2 - 2
Combined Cycle Single Shaft Yes 495 3 - 2 - 2 45: 3 - 2 - 2
Hybrid
Coal Steam No 692 3 - 2 - 2 5;4 3 - 2 - 2
Natural Gas Combined Cycle No ;5 4 - 2 - 2 ;4 4 - 2 - 2
Combined Cycle Yes 547 3 - 2 - 2 422 3 - 2 - 2
344
A.6 Suggestions for EIA data improvement
Although the EIA dataset includes a fair amount of erroneous, infeasible, zero value or missing
coolingwaterdata,thisstudypointstothesubstantialamountofdatarecordsthatarecomparable
to other studies and offer a statistically significant sample of cooling water values that can help
improve understanding of the water requirements of the power sector. Improvements in reporting
instructions, namelyschematicsofdifferentcoolingsystemtechnologiesandwithdrawal, diversion,
consumption, and discharge definitions, in recent years seem to have vastly improved the quality
of reported data since previous criticisms from Averyt et al. [;]. The following recommendations
would further improve self-reported data collection:
• Increase self-reporting of cooling water data by smaller facilities (i.e. between 3 to 3222 MW
nameplate capacity).
• Increase the reporting in EIA Form :82 8A so that generator IDs can be linked to cooling
system when possible, enabling the assessment of facilities with multiple fuels, prime movers,
and/or cooling systems.
• Standardize the formatting of EIA data reports, such that synthesis of multiple forms is
facilitated.
• Add operator reporting requirements to record average monthly water consumption and
water withdrawal rates (i.e. volume per unit of generation) so that climate-related variability
and reliability can be assessed.
345
Appendix B
Supplemental information for chapter 6: A case study of
spatial and temporal environmental impacts of electricity
generation in Texas
B.3 Natural gas combined cycle conversion results
The impact of natural gas combined cycle retrofitting was evaluated by investigating two con-
version scenarios, considering the replacement of the 32 EGUs with the greatest environmental
impacts on the grid mentioned in the previous section.
• Scenario 3: conversion of : coal-fired power plants (4 recirculating cooled and 8 open-loop
cooled facilities) to natural gas combined cycle facilities with equal capacities
• Secnario 4: conversion of : coal-fired power plants and 4 nuclear power plants (both open-
loop cooled facilities) to natural gas combined cycle facilities with equal capacities.
The conversion to natural gas-fired facilities is a reasonable option for reducing the environmental
impacts of electricity generation, particularly for coal-fired EGUs. In the current environment of
highproductionofnaturalgasandlownaturalgaspricesinTexas,naturalgas-firedcombinedcycle
EGUs are an attractive option for new or retrofitted generation due to economic, regulatory, and
environmental benefits over other generation technologies, especially coal [47,49,5:,74,77,9;,:2,
:4,389,393]. Overall results are presented in Table B.3 and Figures B.3 and B.4. Table B.4 details
the socioeconomic cost benefits of natural gas combine cycle retrofitting for the two conversion
scenarios, relying on socioeconomic costs for air pollutants ($/ton) published by the National
Research Council [326] and adjusted to 4233 US dollars. Table B.5 details the shifts in water
consumption for the affected water basins compared to baseline electric water consumption values.
346
Figure B.3: The electricity generation, CO
2
, NO
x
, SO
x
emissions, and water consumption profiles
for the ERCOT fleet with : coal-fired plants converted to NGCC shows overall decreases in
emissions and water consumption.
347
Figure B.4: The electricity generation, CO
2
, NO
x
, SO
x
emissions, and water consumption profiles
for the ERCOT fleet with : coal-fired and 4 nuclear plants converted to NGCC shows lesser
decreases in emissions than pure coal conversion.
348
Table B.3: Change in emissions and water use in natural gas combined cycle conversion scenarios
Environmental
externality
Baseline Scenario 3 Scenario 4
CO
4
Emissions
(3222 tons)
445222 37;222 -4;% 3:3222 -3;%
NO
x
Emissions
(3222 tons)
343 77 -77% 82 -73%
SO
x
Emissions
(3222 tons)
62; 378 -84% 384 -82%
Water
Withdrawal
(Trillion gallons)
;.5 8.; -48% 3.9 -:4%
Water
Consumption
(Trillion gallons)
2.34 2.32 -35% 2.2; -49%
349
Table B.4: Socioeconomic impacts of air pollution from natural gas combined cycle conversion (social costs of air pollutants calculated using
values from NRC 4232 report [326].)
Baseline Scenario 3 Scenario 4
Power Plant
Name
NOx
(3222
tons)
Social Cost
(Million
USD)
SOx
(3222
tons)
Social Cost
(Million
USD)
NOx
(3222
tons)
Change in
Social Cost
(Million
USD)
SOx
(3222
tons)
Change in
Social Cost
(Million
USD)
NOx
(3222
tons)
Change in
Social Cost
(Million
USD)
SOx
(3222
tons)
Change in
Social Cost
(Million
USD)
Coal
Big Brown 9.56 34.9 99.: 357 2.655 -34.2 2.234 -357.3 2.6;: -33.8 2.236 -357
Fayette
Power
8.85 33.7 7.35 :.;3 2.683 -32.9 2.235 -:.:5 2.75: -32.4 2.237 -:.92
JK Spruce 6.24 8.;: 5.48 7.88 2.667 -8.4 2.234 -7.7: 2.739 -7.97 2.236 -7.67
Limestone 37.7 49.2 49.6 69.7 2.666 -48.4 2.234 -69.7 2.73: -47.9 2.237 -69.5
Martin Lake 3:.2 53.5 9;.: 35; 2.664 -52.7 2.234 -35; 2.733 -52.3 2.236 -35:
Monticello 33.9 42.6 92.2 343 2.657 -3;.8 2.234 -343 2.729 -3;.4 2.236 -343
Oak Grove 6.8 9.;7 7.6: ;.73 2.683 -9.37 2.235 -;.65 2.75: -8.89 2.237 -;.52
Oklaunion :.5 36.6 6.82 9.;; 2.4;; -35.; 2.22: -9.;6 2.549 -35.8 2.22; -9.:8
Nuclear
STP 2 2 2 2 2 2 2 2 2.7;9 3.64 2.239 2.458
Commanche
Peak
2 2 2 2 2 2 2 2 2.793 3.58 2.238 2.448
Total 98.4 354 496 697 5.64 -348 2.2;8 -696 7.34 -342 2.366 -695
34:
Table B.5: Shifts in water consumption (WC, billion gallons) with natural gas combined cycle
conversions
Baseline Scenario 3 Scenario 4
Plant Name WC WC
Difference
from
Baseline
WC
Difference
from
Baseline
Big Brown 2.90 10.3
477%
11.8
529%
Feyette Power 3.83 10.5
397%
12.8
455%
JK Spruce 6.13 10.5
93%
12.3
322%
Limestone 8.46 10.1
42%
12.1
65%
Martin Lake 7.19 9.71
57%
11.8
86%
Monticello 3.49 9.79
3:2%
11.1
43;%
Oak Grove 4.67 10.6
34:%
11.3
365%
Oklaunion 1.62 7.37
576%
12.4
883%
STP 14.0 14.1
2%
8.09
-64%
Commanche Peak 12.2 12.2
2%
13.3
;%
Total 64.5 105
85%
117
:3%
B.4 Peak-shifting analysis results
To evaluate the grid-scale impacts of peak-shifting, two additional simulations were executed to
quantify shifts in environmental externalities:
• Scenario A: 72% load adjustment
• Scenario B: Flat load adjustment
Smoothing the actual historic load profile to a 72% adjusted scenario was accomplished by reduc-
ing the difference between the actual load and the daily average by 72%. This smoothing task
was performed for every hour of the year and is illustrated in Figure S5 for August 3, 4233. The
flat load profile was created by using the average load as the constant load for each day of the
year. This method ensures a flat load profile, but more importantly maintains the net daily load
and net annual load for 4233 (Figure B.5). These adjusted profiles were inserted into the existing
model to evaluate the environmental impacts of peak shifting across the grid (Table B.6).
34;
Figure B.5: Peak shifted load for August 3, 4233.
352
Table B.6: Shifts in environmental externalities with different demand response profiles
Environmental
externality
Baseline Scenario
A
Difference
from
baseline
Scenario
B
Difference
from
baseline
CO
4
Emissions
(3222 tons)
44522 44622 2.79% 44622 2.84%
NO
x
Emissions
(3222 tons)
342 343 2.88% 343 2.7;%
SO
x
Emissions
(3222 tons)
62; 642 4.88% 645 5.69%
Water
Withdrawal
(Trillion gallons)
8.5: 8.62 2.46% 8.5; 2.3:%
Water
Consumption
(Trillion gallons)
2.324 2.325 2.73% 2.325 2.:2%
353
Appendix C
Supplemental information for chapter 7: Quanitfying
Regional Water Use Rates for the Electricity Sector
Introduction
This appendix includes additional detail on methods and results for Chapter 8, which estimates
US consumptive water intensity for electricity by eGRID region, defined in Table C.3 and shown
in Figure C.3.
Figure C.3: Map of eGRID regions as defined by the US Environmental Protection Agency [384]
354
Table C.3: Description of eGRID regions [384]
Code Description
AKGD Alaska Systems Coordinating Council Alaska Grid
AKMS Alaska Systems Coordinating Council Miscellaneous
AZNM Western Electricity Coordinating Council Southwest
CAMX Western Electricity Coordinating Council California
ERCT Electric Reliability Council of Texas All
FRCC Florida Reliability Coordinating Council All
HIMS Hawaiian Islands Coordinating Council Miscellaneous
HIOA Hawaiian Islands Coordinating Council Oahu
MROE Midwest Reliability Organization East
MROW Midwest Reliability Organization West
NEWE Northeast Power Coordinating Council New England
NWPP Western Electricity Coordinating Council Northwest
NYCW Northeast Power Coordinating Council NYC/Westchester
NYLI Northeast Power Coordinating Council Long Island
NYUP Northeast Power Coordinating Council Upstate NY
RFCE Reliability First Corporation East
RFCM Reliability First Corporation Michigan
RFCW Reliability First Corporation West
RMPA Western Electricity Coordinating Council Rockies
SPNO Southwest Power Pool North
SPSO Southwest Power Pool South
SRMV SERC Reliability Corporation Mississippi Valley
SRMW SERC Reliability Corporation Midwest
SRSO SERC Reliability Corporation South
SRTV SERC Reliability Corporation Tennessee Valley
SRVC SERC Reliability Corporation Virginia/Carolina
355
C.3 Primary energy consumptive water intensity
Water consumption embedded in fuel upstream of the point of generation is calculated using a
bottom-up approach, aggregating individual generators at power plants up to eGRID region level.
Tofacilitateapplicationofconsumptivewaterintensityvaluesbyfuel,usingdatafrom[75],Energy
Information Administration (EIA) fuel codes for power plants are simplified and categorized.
Table C.4 provides the translation from reported EIA fuel types to fuel classifications used in this
work, adapted from Data File 3, Sheet âĂIJEIA DefinitionsâĂİ in [75]. Water types are defined
as in [75].
Fuel inputs to US generators for 4236 are taken from EIA Form ;45 Data “Elec Fuel Con-
sumption MMBtu” [378]. This metric excludes fuel inputs for heat at combined heat and power
(CHP) plants, unlike the similar metric “Total fuel consumption MMBtu.” Note that by using
total electrical fuel consumption, this work only allocates water consumption associated with in-
put fuel at electricity generators to electricity production, rather than across multiple products.
Across US electricity generation, the choice is fairly inconsequential: electricity fuel consumption
is about ;6% of total fuel consumption for generators reporting on EIA Form ;45 [378]. The
difference is relevant in a few cases, however, mainly for various biomass and oil products (which
are often burned in CHP contexts at industrial facilities like pulp and paper plants or refineries)
and for midwestern coal. Energy allocated to electricity from bituminous coal is, on average, ;9%
of total energy burned at power plants, but in the midwest, it is only 4;% [378]. Electricity fuel
for natural gas is, on average, about ;2% of total natural gas consumed at power plants [378],
with limited regional differentiation.
Upstream water consumption for production, processing, and transportation life cycle stages
for each fuel is drawn from Grubert and Sanders (423:) [75]. Factors used in this work are
presented in Table C.5. For power plants consuming coal, regional water consumption values
are applied based on the state where the supplier coal mine is located, based on EIA reporting.
Table C.6 details the separation of coal-producing states into coal provinces, as defined in Grubert
and Sanders [75].
For Kentucky, counties were individually assigned East or West tags based on their geographic
location (Table C.7) because of varying coal characteristics in the state. Upstream water con-
sumption from imported coal from countries outside the US was excluded, as this analysis only
examines the water consumption within the boundaries of the US.
Some US coal is used for steel making, as metallurgical (also called met or coking) coal. This
work assumes that all met coal is produced in the Appalachian region and that consumptive
water intensity is identical for met and thermal coal. Met coal accounts for about 8% of US coal
energy [75], which is about 36% of Appalachian coal energy, so upstream Appalachian coal water
consumption in this work is assumed to be about :8% of the dewatering and processing volumes
presented in [75].
Coal is not the only fuel used for multiple purposes. Both oil and natural gas are primarily
used for purposes other than electricity in the US. This work thus allocates upstream water
consumption from [75] to oil and natural gas proportionally based on the amount of energy used
for electricity for each fuel, which is about 8% of total oil energy and about 55% of total natural
gas energy.
The water intensity of geothermal plants is particularly site specific. This work specifically
allocates plant-specific water consumption for reservoir augmentation to the appropriate eGRID
regions, and water removal in the form of steam at The Geysers steam field (CAMX) is also
allocated specifically to the appropriate region. Water for well drilling, the remaining upstream
water consumer, is allocated across US geothermal sites based on total generation. See [75] for
details on how these values were calculated and accounted.
356
Table C.4: EIA fuel code proxies for fuel classifications
EIA Code EIA Definition This Work Classifies
AB Agricultural By-Products Solid Biomass and RDF
ANT Anthracite Coal n/a
BFG Blast Furnace Gas Bituminous Coal
BIT Bituminous Coal Bituminous Coal
BLQ Black Liquor Solid Biomass and RDF
DFO Distillate Fuel Oil. Including diesel, No. 3, No. 4, and No. 6 fuel
oils.
Oil
GEO Geothermal Geothermal
JF Jet Fuel Oil
KER Kerosene Oil
LFG Landfill Gas Biogas
LIG Lignite Coal Lignite Coal
MSB Biogenic Municipal Solid Waste Solid Biomass and RDF
MSN Non-biogenic Municipal Solid Waste Solid Biomass and RDF
MWH Electricity used for energy storage n/a
NG Natural Gas Natural Gas
NUC Nuclear. Including Uranium, Plutonium, and Thorium. Uranium
OBG Other Biomass Gas. Including digester gas, methane, and other
biomass gases.
Biogas
OBL Other Biomass Liquids Solid Biomass and RDF
OBS Other Biomass Solids Solid Biomass and RDF
OG Other Gas Natural Gas
OTH Other Fuel Case-by-case allocation
PC Petroleum Coke Oil
PG Gaseous Propane Oil
PUR Purchased Steam Case-by-case allocation
RC Refined Coal Bituminous Coal
RFO Residual Fuel Oil. Including No. 7 & 8 fuel oils and bunker C fuel
oil.
Oil
SC Coal-based Synfuel. Including briquettes, pellets, or extrusions,
which are formed by binding materials or processes that recycle ma-
terials.
Bituminous Coal
SGC Coal-Derived Synthesis Gas Bituminous Coal
SGP Synthesis Gas from Petroleum Coke Oil
SLW Sludge Waste Solid Biomass and RDF
SUB Subbituminous Coal Subbituminous Coal
SUN Solar Solar
TDF Tire-derived Fuels Solid Biomass and RDF
WAT Water at a Conventional Hydroelectric Turbine and water used in
Wave Buoy Hydrokinetic Technology, current Hydrokinetic Tech-
nology, Tidal Hydrokinetic Technology, and Pumping Energy for
Reversible (Pumped Storage) Hydroelectric Turbines.
Hydropower
WC Waste/Other Coal. Including anthracite culm, bituminous gob, fine
coal, lignite waste, waste coal.
Bituminous Coal
WDL Wood Waste Liquids, excluding Black Liquor. Including red liquor,
sludge wood, spent sulfite liquor, and other wood-based liquids.
Solid Biomass and RDF
WDS Wood/Wood Waste Solids. Including paper pellets, railroad ties,
utility poles, wood chips, bark, and other wood waste solids.
Solid Biomass and RDF
WH Waste Heat not directly attributed to a fuel source Case-by-case allocation
WND Wind Wind
WO Waste/Other Oil. Including crude oil, liquid butane, liquid propane,
naphtha, oil waste, re-refined motor oil, sludge oil, tar oil, or other
petroleum-based liquid wastes.
Oil
357
Table C.5: Upstream consumptive water intensity factor by fuel
Fuel Classification Water consumption Unit
Oil 3.4;×10
−1
m
3
/GJ
Subbituminous coal (northern great plains) :.67×10
6
m
3
Bituminous coal (appalachia) 3.42×10
8
m
3
Bituminous coal (interior) 6.67×10
8
m
3
Bituminous coal (rocky mountain region) 5.5:×10
7
m
3
Lignite coal (gulf coast) 5.4:×10
8
m
3
Lignite coal (northern great plains) 6.82×10
5
m
3
Natural gas 4.79×10
−2
m
3
/GJ
Uranium 3.29×10
7
m
3
Hydropower 2 m
3
Wind 3.;8×10
6
m
3
Solid biomass & RDF 9.58×10
7
m
3
Biogas 2 m
3
Geothermal (excluding California & Nevada) 5.78×10
4
m
3
Geothermal (California) 3.6;×10
8
m
3
Geothermal (Nevada) 3.26×10
6
m
3
Solar PV 3.8;×10
5
m
3
Solar thermal 3.98×10
5
m
3
Table C.6: Definition of coal regions used in this work
Province Abbreviation States
Northern Great Plains NGP MT, ND, WY
Appalachia/Eastern APP AL, eastern KY, MD, OH, PA, TN, VA, WV
Interior INT AR, IL, IN, KS, western KY, MO, OK
Gulf Coast GFC LA, MS, TX
Rocky Mountain Region RMR AZ, CO, NM, UT
358
Table C.7: Separation of Eastern and Western Kentucky counties
FIPS County ID County Name KY E-W Classification
35 Bell County East
3; Boyd County East
47 Breathitt County East
73 Clay County East
7; Daviess County West
87 Estill County East
93 Floyd County East
;7 Harlan County East
329 Hopkins County West
337 Johnson County East
33; Knott County East
343 Knox County East
349 Lawrence County East
353 Leslie County East
355 Letcher County East
35; Livingston County West
36; McLean County West
375 Magoffin County East
37; Martin County East
399 Muhlenberg County West
3:5 Ohio County West
3;5 Perry County East
3;7 Pike County East
425 Rockcastle County East
439 Taylor County West
447 Union County West
455 Webster County West
NA Daviess County (based on coal mine name) West
359
Water consumed for hydroelectricity
Regional eGRID water consumption rates for hydroelectricity are calculated using the method in
Grubert (4238) [76]. Each reservoir associated with a dam defined as having primary purpose =
hydroelectricityintheNationalInventoryofDams(NID)[328]isassignedbothtoitscentroidfrom
[76]andtoaneGRIDregion. Hydroelectricfacilitiesthathaveprimarypurpose=hydroelectricity
in the NID that also appear in eGRID (about 9;% of relevant reservoir surface area) are assigned
to their reported eGRID region, and others (about 43% of relevant reservoir surface area) are
assigned based on a spatial join of NID reported coordinates and the 4238 eGRID boundaries
[384] using QGIS. Note that using coordinates to define location for hydroelectric facilities is less
straightforward than for single-site power plants, as reservoirs are areally extensive, and dams,
powerhouses, and the single-point location of a reservoir might be quite distant from each other.
Thewaterintensityofhydroelectricityiscalculatedastotalwaterconsumptionassociatedwith
dams with NID primary purpose = hydroelectricity divided by total hydroelectricity generation in
the region, regardless of whether the plants are at dams with primary purpose = hydroelectricity
(see [76] for discussion and justification), so water consumption for relevant reservoirs in each
eGRID region is calculated based on reservoir surface area, not powerhouse generation. Water
consumption from reservoir evaporation upstream of the point of generation (PoG) is calculated
using a Penman-Monteith model, with dams aggregated by statistically determined centroids.
The values presented in this work are evaporation net of pre-reservoir landcover. See [76] for
details and a copy of the model.
C.4 Point of generation consumptive water intensity
Water consumption at the point of generation (PoG) is the water consumed at power plants,
mainly for cooling at thermoelectric generating facilities. This work classifies generators by fuel
type, prime mover, and cooling system. To simplify calculations and reduce the number of genera-
tor codes (fuel-prime mover-cooling system), reported fuel codes, prime mover codes, and cooling
system codes from the EIA were reclassified as described in Table C.8, Table C.9, and Table C.:,
respectively. Water types are defined in Table C.; and Table C.32, based on EIA Form :82.
All gas turbine, wind turbine, and photovoltaic generators are assigned a PoG water consump-
tion rate of zero, as there is no reported operational water consumption for these technologies.
Energy storage technologies are also assumed to have zero water consumption. Median water
consumption rates from 4236 EIA data [5] were applied for all natural gas, coal, and nuclear (ura-
nium) generating technologies. Generating technologies reporting no cooling system (i.e., “NA”)
were assigned a generation-weighted average consumption intensity based on the consumption
rates for other technology classifications with the same fuel and prime mover.
Estimates of total water consumed for power plant cooling from Grubert and Sanders [75] are
used to assign water consumption rates for biomass, biogas, oil, solar thermal, and geothermal
generating technologies. The total estimates reflect data and calculations from the US Geological
Survey and the Union of Concerned Scientists [56,36:]. These estimates were divided by total
water cooled generation for each fuel type for an average water consumption rate to be applied
for each generator classification. The source and assumed consumptive water intensity for each
PoG classification is presented in Table C.33.
Table C.33: Water consumption assumptions for each generating technology
classification
Code (F-PM-CS) Cooling Rate Assumption Consumption Rate
(m
3
/MWh)
BG-CC-NA Water use from [75] for biogas 1.13
BG-ES-NA No cooling required 0
35:
BG-GT-NA No cooling required 0
BG-ST-ON Water use from [75] for biogas 1.13
BG-ST-RC Water use from [75] for biogas 1.13
BG-ST-NA Water use from [75] for biogas 1.13
BM-GT-NA No cooling required 0
BM-ST-DC Water use from [75] for solid biomass & RDF 1.82
BM-ST-ON Water use from [75] for solid biomass & RDF 1.82
BM-ST-PN Water use from [75] for solid biomass & RDF 1.82
BM-ST-RC Water use from [75] for solid biomass & RDF 1.82
BM-ST-NA Water use from [75] for solid biomass & RDF 1.82
CL-CC-PN Consumption rate for natural gas from [5] 0.598
CL-CC-NA Generation-weighted average of consumption
rates for natural gas combined cycle from [5]
0.713
CL-OT-NA Generation-weighted average of consumption
rates for coal steam from [5]
1.42
CL-ST-DC Consumption rate for coal from [5] 0.184
CL-ST-ON Consumption rate for coal from [5] 0.772
CL-ST-PN Consumption rate for coal from [5] 1.39
CL-ST-RC Consumption rate for coal from [5] 1.84
CL-ST-NA Generation-weighted average of consumption
rates for coal steam from [5]
1.42
ES-ES-NA No cooling required 0
GEO-BT-NA Water use from [75] for geothermal binary tur-
bine, wet cooled
5.09
GEO-ST-NA Water use from [75] for geothermal, wet cooled 0.193
NG-CC-DC 32% of [5] rates for natural gas CC, tower cooled 0.082
NG-CC-HB Consumption rate for natural gas from [5] 0.348
NG-CC-ON Consumption rate for natural gas from [5] 0.712
NG-CC-PN Consumption rate for natural gas from [5] 0.598
NG-CC-RC Consumption rate for natural gas from [5] 0.821
NG-CC-NA Generation-weighted average of consumption
rates for natural gas combined cycle from [5]
0.713
NG-CS-DC 32% of [5] rates for natural gas CS, tower cooled 0.078
NG-CS-RC Consumption rate for natural gas from [5] 0.776
NG-CS-NA Generation-weighted average of consumption
rates for natural gas combined cycle single shaft
from [5]
0.428
NG-ES-NA No cooling required 0
NG-GT-ON No cooling required 0
NG-GT-NA No cooling required 0
NG-OT-NA No cooling required 0
NG-ST-DC Consumption rate for natural gas from [5] 0.315
NG-ST-ON Consumption rate for natural gas from [5] 1.23
NG-ST-PN Consumption rate for natural gas from [5] 1.20
NG-ST-RC Consumption rate for natural gas from [5] 3.15
NG-ST-NA Generation-weighted average of consumption
rates for natural gas steam from [5]
1.69
NUC-ST-ON Consumption rate for nuclear from [5] 1.37
NUC-ST-PN Consumption rate for nuclear from [5] 1.93
NUC-ST-RC Consumption rate for nuclear from [5] 2.54
NUC-ST-RT Consumption rate for nuclear from [5] 2.54
NUC-ST-NA Generation-weighted average of consumption
rates for nuclear from [5]
1.81
OIL-CC-RC Water use from [75] for oil 2.36
OIL-CC-NA Water use from [75] for oil 2.36
OIL-GT-NA No cooling required 0
OIL-ST-DC Water use from [75] for oil 2.36
OIL-ST-ON Water use from [75] for oil 2.36
OIL-ST-PN Water use from [75] for oil 2.36
OIL-ST-RC Water use from [75] for oil 2.36
OIL-ST-NA Water use from [75] for oil 2.36
OTH-GT-NA No cooling required 0
OTH-OT-NA No cooling required 0
SUN-CC-NA Water use from [75] for solar thermal 4.39
35;
SUN-ES-RC No cooling required 0
SUN-ES-NA No cooling required 0
SUN-PV-NA No cooling required 0
SUN-ST-DC Water use from [75] for oil 4.39
SUN-ST-NA Water use from [75] for oil 4.39
WAT-ES-NA No cooling required 0
WAT-HY-NA No cooling required 0
WND-WT-NA No cooling required 0
NA-NA-NA No cooling required 0
C.5 Results
Figure C.4 shows generation by fuel in each eGRID region, and Table C.34 shows total water
consumption by technology for each eGRID region.
Figure C.4: Regional distribution of generation, separated by fuel type.
TablesC.35,C.36, and C.37presentfuel-specificconsumptivewaterintensitybyeGRIDregion,
for upstream, PoG, and overall intensities.
362
Table C.8: EIA fuel code proxies for electricity fuel classifications [378]
EIA Code EIA Definition Proxy for this work
LFG Landfill Gas Biogas
OBG Other Biomass Gas. Including digester gas, methane, and other
biomass gases.
Biogas
AB Agricultural By-Products Biomass
BLQ Black Liquor Biomass
MSB Biogenic Municipal Solid Waste Biomass
MSN Non-biogenic Municipal Solid Waste Biomass
OBL Other Biomass Liquids Biomass
OBS Other Biomass Solids Biomass
SLW Sludge Waste Biomass
TDF Tire-derived Fuels Biomass
WDL WoodWasteLiquids, excludingBlackLiquor. Includingredliquor,
sludge wood, spent sulfite liquor, and other wood-based liquids.
Biomass
WDS Wood/Wood Waste Solids. Including paper pellets, railroad ties,
utility polies, wood chips, bark, and other wood waste solids.
Biomass
BIT Bituminous Coal Coal
DFO Distillate Fuel Oil. Including diesel, No. 3, No. 4, and No. 6 fuel
oils.
Coal
JF Jet Fuel Oil
KER Kerosene Oil
LIG Lignite Coal Coal
PC Petroleum Coke Oil
RC Refined Coal Coal
RFO Residual Fuel Oil. Including No. 7&8 fuel oils and bunker C fuel
oil.
Oil
SC Coal-based Synfuel. Including briquettes, pellets, or extrusions,
which are formed by binding materials or processes that recycle
materials.
Coal
SUB Subbituminous Coal Coal
WC Waste/Other Coal. Including anthracite culm, bituminous gob,
fine coal, lignite waste, waste coal.
Coal
WO Waste/Other Oil. Including crude oil, liquid butane, liquid
propane, naphtha, oil waste, re-refined motor oil, sludge oil, tar
oil, or other petroleum-based liquid wastes.
Oil
GEO Geothermal Geothermal
BFG Blast Furnace Gas Natural Gas
NG Natural Gas Natural Gas
PG Gaseous Propane Natural Gas
SGC Coal-Derived Synthesis Gas Coal
SGP Synthesis Gas from Petroleum Coke Natural Gas
OG Other Gas Natural Gas
NUC Nuclear. Including Uranium, Plutonium, and Thorium. Nuclear
SUN Solar Solar
WAT Water at a Conventional Hydroelectric Turbine and water used in
Wave Buoy Hydrokinetic Technology, current Hydrokinetic Tech-
nology, Tidal Hydrokinetic Technology, and Pumping Energy for
Reversible (Pumped Storage) Hydroelectric Turbines.
Hydro
WND Wind Wind
ANT Anthracite Coal n/a
MWH Electricity used for energy storage n/a
OTH Other Fuel Case-by-caseallocation
PUR Purchased Steam Case-by-caseallocation
WH Waste Heat not directly attributed to a fuel source Case-by-caseallocation
363
Table C.9: EIA prime mover code proxies for prime mover classifications used in this work [378]
EIA Code EIA Definition Proxy for this work
CA Combined-Cycle – Steam Part Combined Cycle
CS Combined-Cycle Single-Shaft Combustion Turbine and Steam Turbine
share of single generator
Combined Cycle
CT Combined-Cycle Combustion Turbine Part Combined Cycle
BA Energy Storage, Battery Energy Storage
CE Energy Storage, Compressed Air Energy Storage
CP Energy Storage, Concentrated Solar Power Energy Storage
ES Energy Storage, Other (Specify on Schedule ;, Comments) Energy Storage ES
FW Energy Storage, Flywheel Energy Storage
PS Energy Storage, Reversible Hydraulic Turbine (Pumped Storage) Energy Storage
GT Combustion (Gas) Turbine. Including Jet Engine design Combustion Turbine
IC Internal Combustion (diesel, piston, reciprocating) Engine Combustion Turbine
HA Hydrokinetic, Axial Flow Turbine Hydrokinetic
HB Hydrokinetic, Wave Buoy Hydrokinetic
HK Hydrokinetic, Other Hydrokinetic
HY Hydraulic Turbine. Including turbines associated with delivery of water
by pipeline.
Hydraulic
PV Photovoltaic Photovoltaic
BT Turbines Used in a Binary Cycle. Including those used for geothermal
applications.
Binary Turbine
ST Steam Turbine. Including Nuclear, Geothermal, and Solar Steam (does
not include Combined Cycle).
Steam Turbine
WT Wind Turbine, Onshore Wind Turbine
WS Wind Turbine, Offshore Wind Turbine
FC Fuel Cell Energy Storage
OT Other Other
Table C.:: EIA cooling system code proxies for cooling system classifications used in this work
[377]
EIA Code EIA Definition Proxy for this work
DC Dry (air) cooling System Dry cooling (DC)
H Hybrid (non-specified) Hybrid (HB)
HRC Hybrid: recirculating cooling pond(s) or canal(s) with dry cooling Hybrid (HB)
HRF Hybrid: recirculating with forced draft cooling tower(s) with dry cooling Hybrid (HB)
HRI Hybrid: recirculating with induced draft cooling tower(s) with dry cool-
ing
Hybrid (HB)
HT Helper Tower Recirculating (RT)
O Once through (non-specified) Once through (ON)
O + R Once through (non-specified) and Recirculating (non-specified) n/a
OC Once through with Cooling Ponds Pond (PN)
ON Once through without cooling pond(s) or canal(s) Once through (ON)
R Recirculating (non-specified) Recirculating (RT)
RC Recirculating with Cooling Ponds Pond (PN)
RF Recirculating with Forced Draft Cooling Tower Recirculating (RT)
RI Recirculating with Induced Draft Cooling Tower Recirculating (RT)
RN Recirculating with Natural Draft Cooling Tower Recirculating (RT)
364
Table C.;: EIA water type code proxies for water type classifications used in this work [377]
EIA Code EIA Definition Proxy for this work
SW Surface Water (ex: river, canal, bay) Surface Water
GW Ground Water (ex: aquifer, well) Groundwater
PD Plant Discharge Water (ex: wastewater treatment plant discharge) Recycled Water
Table C.32: EIA water quality code proxies for water quality classifications used in this work [377]
EIA Code EIA Definition Proxy for this work
BR Brackish Water Brackish Water
FR Fresh Water Freshwater
BE Reclaimed Water (ex: treated wastewater effluent) Freshwater
SA Saline Water Saline Water
Table C.34: Regional volumetric water consumption across generating technologies
Volumetric Water Consumption (million m
3
)
a
eGRID
region
all thermal
b
non-
thermal
c
fossil-
fueled
d
non-hydro
renewables
e
all
renewables
f
no
combustion
g
AKGD 4.7 4.7 0.0027 4.7 0.0027 0.0027 0.0027
AKMS 0.46 0.46 0.00060 0.46 0.00060 0.00060 0.00060
AZNM 1,400 260 1,100 120 55 1,200 1,300
CAMX 300 250 44 99 120 170 180
ERCT 360 360 4.7 290 0.93 4.9 71
FRCC 300 300 − 240 17 17 46
HIMS 6.4 6.4 0.0085 5.8 0.64 0.64 0.34
HIOA 22 22 0.0040 21 1.4 1.4 0.0043
MROE 78 43 35 26 3.5 39 48
MROW 450 230 220 150 8.0 230 290
NEWE 130 140 −6.5 52 29 22 51
NWPP 360 300 64 250 24 88 100
NYCW 45 45 0 19 1.4 1.4 24
NYLI 15 15 0 11 3.5 3.5 0.00074
NYUP 53 75 −22 26 4.0 −18 23
RFCE 410 420 −6.9 190 13 6.4 200
RFCM 110 110 −2.9 75 4.7 1.6 32
RFCW 950 920 22 730 4.7 27 220
RMPA 93 100 −8.0 100 0.37 −7.8 −8.0
SPNO 72 72 0.19 55 0.25 0.25 17
SPSO 250 210 40 200 6.4 45 40
SRMV 380 300 84 200 10 94 170
SRMW 300 250 44 180 0.27 44 120
SRSO 370 400 −31 290 23 −8.7 54
SRTV 580 530 45 440 4.7 50 130
SRVC 520 440 79 220 24 100 280
National
h
7,500 5,800 1,700 4,000 360 2,100 3,300
a
Values are rounded to two significant digits.
b
Biogas, biomass, coal, geothermal, natural gas, nuclear, oil, solar
thermal.
c
Hydroelectricity, solar, wind. Includes solar thermal.
d
Coal, natural gas, oil.
e
Biogas, biomass,
geothermal, solar, wind.
f
Biogas, biomass, geothermal, hydroelectricity, solar, wind.
g
Geothermal,
hydroelectricity, nuclear, solar, wind.
h
National totals may not match generation-weighted values presented in
table due to rounding.
365
Table C.35: Upstream consumptive water intensity by fuel for each eGRID region, 4236
Water Consumption Intensity (m
3
/MWh)
a
eGRID
region
Biogas Biomass Coal Geothermal Hydro Oil Natural Gas Solar PV Solar Thermal Uranium Wind
AKGD 0 0 0.037 0 0 1.2 0.27 0 0 0 0.011
AKMS 0 0 0 0 0 13 0.41 0 0 0 0.011
AZNM 0 1.6 0.21 12 126 1.5 0.21 0.011 0.078 0.013 0.011
CAMX 0 1.5 0.87 12 2.8 1.4 0.20 0.011 0.078 0.013 0.011
ERCT 0 1.5 0.20 0 13 2.1 0.20 0.011 0 0.013 0.011
FRCC 0 1.7 1.0 0 0 1.4 0.21 0.011 0.078 0.013 0
HIMS 0 1.0 0 0 0 1.4 0 0.011 0 0 0.011
HIOA 0 1.7 0.19 0 0 1.2 0.15 0.011 0 0 0.011
MROE 0 1.2 0.031 0 33 1.2 0.22 0 0 0.013 0.011
MROW 0 1.4 0.013 0 19 1.7 0.24 0.011 0 0.013 0.011
NEWE 0 1.4 0.25 0 −0.97 1.4 0.21 0.011 0.078 0.013 0.011
NWPP 0 1.2 0.16 0.30 0.47 1.5 0.21 0.011 0 0.013 0.011
NYCW 0 1.9 0 0 0 1.2 0.24 0 0 0.013 0
NYLI 0 1.9 0 0 0 1.5 0.25 0.011 0 0 0
NYUP 0 1.3 0.30 0 −0.85 1.3 0.21 0.011 0 0.013 0.011
RFCE 0 1.8 0.32 0 −2.2 1.5 0.22 0.011 0 0.013 0.011
RFCM 0 1.3 0.06 0 27 1.3 0.25 0 0 0.013 0.011
RFCW 0 1.0 0.51 0 6.0 1.4 0.22 0.011 0 0.013 0.011
RMPA 0 1.6 0.064 0 −3.8 1.5 0.23 0.011 0 0 0.011
SPNO 0 1.4 0.012 0 0 2.1 0.30 0 0 0.013 0.011
SPSO 0 0.8 0.14 0 12 1.4 0.22 0.011 0 0 0.011
SRMV 0 0.72 0.076 0 37 1.4 0.23 0.011 0 0.013 0
SRMW 0 1.0 0.046 0 37 1.5 0.24 0.011 0 0.013 0.011
SRSO 0 1.0 0.47 0 −4.6 1.5 0.20 0.011 0 0.013 0
SRTV 0 0.52 1.8 0 2.4 1.5 0.22 0.011 0 0.013 0.011
SRVC 0 1.1 0.38 0 21 1.5 0.21 0.011 0 0.013 0
National
h
0 1.2 0.40 9.5 6.8 1.4 0.21 0.011 0.078 0.013 0.011
a
Values are rounded to two significant digits.
b
National totals may not match generation-weighted averages presented in table due to rounding.
366
Table C.36: Point of generation consumptive water intensity by fuel by eGRID region, 4236
Water Consumption Intensity (m
3
/MWh)
a
eGRID
region
Biogas Biomass Coal Geothermal Hydro Oil Natural Gas Solar PV Solar Thermal Uranium Wind
AKGD 0 0 1.4 0 0 1.9 0.62 0 0 0 0
AKMS 0 0 0 0 0 0 0 0 0 0 0
AZNM 0 1.8 1.7 0.71 0 2.0 0.67 0 2.3 2.5 0
CAMX 0.53 1.8 1.4 0.32 0 0 0.60 0 5.1 1.8 0
ERCT 0 1.8 1.4 0 0 0.22 0.74 0 0 1.7 0
FRCC 0.52 1.8 1.3 0 0 0.46 0.73 0 4.4 1.6 0
HIMS 0 1.8 1.4 1.3 0 1.9 0 0 0 0 0
HIOA 0 1.7 1.8 0 0 2.3 0 0 0 0 0
MROE 0.017 1.2 1.2 0 0 0.28 0.83 0 0 1.4 0
MROW 0.10 1.8 1.3 0 0 1.0 0.68 0 0 1.6 0
NEWE 0.45 1.8 1.5 0 0 2.2 0.53 0 0 1.6 0
NWPP 0.23 1.8 1.7 3.6 0 0.70 0.62 0 0 2.5 0
NYCW 0 1.8 0 0 0 2.1 0.48 0 0 1.4 0
NYLI 0 1.8 0 0 0 1.9 0.67 0 0 0 0
NYUP 0 1.8 0.82 0 0 2.4 0.69 0 0 1.7 0
RFCE 0.31 1.8 1.4 0 0 1.3 0.71 0 0 1.9 0
RFCM 0.17 1.8 1.1 0 0 0.27 0.74 0 0 2.5 0
RFCW 0.04 1.8 1.5 0 0 0.44 0.87 0 0 1.9 0
RMPA 0 1.8 1.4 0 0 0.89 0.60 0 0 0 0
SPNO 0 1.8 1.3 0 0 1.7 0.96 0 0 1.9 0
SPSO 0 1.8 1.5 0 0 0.050 0.92 0 0 0 0
SRMV 0 1.8 1.7 0 0 0.064 0.75 0 0 2.1 0
SRMW 0.022 1.8 1.2 0 0 2.2 0.65 0 0 2.0 0
SRSO 0.40 1.7 1.5 0 0 0.94 0.78 0 0 1.8 0
SRTV 0 1.8 1.3 0 0 0.51 0.65 0 0 1.6 0
SRVC 0 1.8 1.3 0 0 1.3 0.67 0 0 1.6 0
National
h
0.22 1.8 1.4 1.1 0 1.1 0.71 0 3.5 1.8 0
a
Values are rounded to two significant digits.
b
National totals may not match generation-weighted averages presented in table due to rounding.
367
Table C.37: Total consumptive water intensity by fuel by eGRID region, 4236
Water Consumption Intensity (m
3
/MWh)
a
eGRID
region
Biogas Biomass Coal Geothermal Hydro Oil Natural Gas Solar PV Solar Thermal Uranium Wind
AKGD 0 0 1.5 0 0 3.2 0.89 0 0 0 0.011
AKMS 0 0 0 0 0 13 0.41 0 0 0 0.011
AZNM 0 3.5 1.9 13 126 3.5 0.88 0.011 2.4 2.6 0.011
CAMX 0.53 3.3 2.3 13 2.8 1.4 0.81 0.011 5.2 1.8 0.011
ERCT 0 3.3 1.6 0 13 2.4 0.94 0.011 0 1.7 0.011
FRCC 0.52 3.5 2.3 0 0 1.9 0.94 0.011 4.5 1.6 0
HIMS 0 2.8 1.4 1.3 0 3.3 0 0.011 0 0 0.011
HIOA 0 3.3 2.0 0 0 3.5 0.15 0.011 0 0 0.011
MROE 0.017 2.4 1.2 0 33 1.5 1.0 0 0 1.4 0.011
MROW 0.10 3.2 1.4 0 19 2.7 0.92 0 0 1.7 0.011
NEWE 0.45 3.2 1.8 0 −0.97 3.6 0.74 0.011 0.078 1.6 0.011
NWPP 0.23 2.9 1.9 3.8 0.47 2.2 0.82 0.011 0 2.6 0.011
NYCW 0 3.7 0 0 0 3.3 0.72 0 0 1.4 0
NYLI 0 3.8 0 0 0 3.4 0.92 0.01 0 0 0
NYUP 0 3.2 1.1 0 −0.85 3.7 0.90 0 0 1.7 0.011
RFCE 0.31 3.6 1.8 0 −2.2 2.9 0.93 0.011 0 1.9 0.011
RFCM 0.17 3.1 1.2 0 27 1.6 0.99 0 0 2.6 0.011
RFCW 0.04 2.8 2.0 0 6.0 1.9 1.1 0.011 0 1.9 0.011
RMPA 0 3.4 1.5 0 −3.8 2.4 0.83 0.011 0 0 0.011
SPNO 0 3.2 1.3 0 0 3.7 1.3 0 0 1.9 0.011
SPSO 0 2.6 1.6 0 12 1.4 1.1 0.011 0 0 0.011
SRMV 0 2.5 1.8 0 37 1.5 0.97 0.011 0 2.1 0
SRMW 0.02 2.8 1.2 0 37 3.7 0.89 0.011 0 2.0 0.011
SRSO 0.40 2.7 2.0 0 −4.6 2.4 0.98 0.011 0 1.8 0
SRTV 0 2.3 3.2 0 2.4 2.0 0.86 0.011 0 1.6 0.011
SRVC 0 2.9 1.7 0 21 2.8 0.87 0.011 0 1.6 0
National
b
0.22 3.0 1.8 11 6.8 2.5 0.92 0.011 3.6 1.8 0.011
a
Values are rounded to two significant digits.
b
National totals may not match generation-weighted averages presented in table due to rounding.
368
Technological and fuel variability in water consumption from electricity
Total induced upstream water consumption for coal produced for electricity generation is only
44% of the total life cycle water use illustrated in Figure 8.5; thus, the coal category is dominated
by cooling water use at the point of coal-fired power plant generation. In total, coal represented
approximately 44% of total US water consumption associated with upstream production and
processing of fuels consumed in power plants, but 6:% of total US water consumption at the
point of electricity generation.
Because of regional differences in production methods (e.g. underground vs surface extraction)
and geological characteristics, such as the accumulated water present in a coal seam, the water
consumed for coal mining is non-linear and can vary significantly across regions. Thus, regional
water consumption rates for coal production and processing upstream of the point of power gener-
ation were utilized. Although the water consumed for coal mining appears to be relatively large in
some regions, much of this water consumption is associated with dewatering and depressurization
of mines, that is, water trapped within the coal mine is released to facilitate production. We
consider this water removal consumptive since water is displaced from its original formation and
is not returned. However, unlike many other consumptive uses of water, this water might not
directly compete with regional water users as it might not have been produced in the absence of
coal mining.
Water consumption for natural gas-fired generators is also dominated by cooling water require-
ments at the point of generation, which represents 99% of the total life cycle water consumption
for the fuel. The introduction of hydraulic fracturing (HF) and horizontal drilling has increased
the fraction of generation supplied by natural gas, but the induced water consumption upstream
of generation only represents :% of the national upstream water consumption. Similarly, the wa-
ter consumption at the point of generation represents 39% of national consumption, highlighting
the increased average efficiency of the natural gas generation fleet.
Results indicate that the large volumes of water consumption associated with nuclear power
generation are driven by cooling water requirements. The upstream induced water consumption
for uranium is relatively small (only 3% of total US upstream water consumption) because the
uranium extraction, processing, and refining industry is highly international [75]. In 4236, only
8% of the uranium purchased for nuclear power production originated from the US, with the
remaining ;6% originating from foreign countries, namely Kazakhstan, Australia, Canada, and
Russia [382]. This work only reflects water that is consumed in the United States, therefore
the water embedded in upstream process of imported uranium used for electricity generation
is excluded. Conversely, the water consumption at the point of generation for nuclear-fueled
generation represents approximately one third (53%) of total US point of generation consumption.
This consumption is driven heavily by slightly increased consumption rates and the exclusive use
of wet cooling systems at nuclear generating facilities.
Hydropower represented approximately 8% of total US electricity generation and 45% of
total water consumption in 4236. Water consumption associated with hydroelectricity occurs as
evaporative losses upstream of the PoG, i.e., from the reservoir.
The water consumption associated with hydropower generation has noticeably large regional
variability, including some regions with negative water consumption. Negative water consumption
can occur due to our use of a net water consumption approach for calculating water consumption
relative to evapotranspiration from preexisting landcover. This condition applies to about 42% of
4236 generation, mostly in New England and the forested southeast. Hydroelectric generation in
theNortheastPowerCoordinatingCouncilNewEngland(NEWE),NortheastPowerCoordinating
Council Upstate NY (NYUP), SERC Reliability Corporation South (SRSO), Western Electricity
Coordinating Council Rockies (RMPA), and Reliability First Corporation East (RFCE) eGRID
subregions has negative water consumption, indicating that the presence of impoundments in
369
Figure C.5: Regional distribution of generation, separated by cooling system type.
Figure C.6: Regional water consumption at the point of generation, separated by cooling system
type.
these regions actually result in the net accumulation of water into the environment as a result of
human activity.
Although the Western Electricity Coordinating Council Northwest (NWPP) generated more
electricity from hydropower than any other subregion in 4236 (76% of total US hydro gener-
ation), its net water consumption from this generation is relatively small (5.9% of total US
hydropower water consumption). By contrast, the Western Electricity Coordinating Council
Southwest (AZNM), represents a relatively small amount (5.7%) of total US hydroelectricity
generation, ranking 36th of the 48 eGRID subregions, but its upstream water consumption far
exceeds any other subregion (87% of total US upstream consumption for hyroelectricity). As
a result, consumptive losses associated with hydropower generation represent 82% of total up-
stream water consumption despite only accounting for 7.;% of generation within the region. The
disproportionate water consumption intensity of hydropower generation in the AZNM region is
driven primarily by large dams along the Colorado River (e.g. The Hoover Dam, etc.) that drive
net increases in evaporation (as compared to the baseline evapotranspiration in the region).
The total water consumption associated with electricity generation (i.e. upstream and at the
point of generation) is driven by cooling thermal power plants in most regions (Figure 8.5). The
cooling water consumption intensity of these thermal generators will vary slightly from facility
to facility due to factors such as cooling system, prime mover configuration, power generation
efficiency, local climate conditions, and to some extent, regulatory policy [33:]. The variability
of consumptive rates (i.e. m
3
/MWh) for wet-cooled facilities is relatively low across generating
technologies (e.g. on the same order of magnitude), but is significantly reduced for hybrid and
dry-cooled facilities. As such, the calculated water consumption at the point of generation scales
largely with the fraction of generation from thermal generation (namely coal, natural gas, and
nuclear) in each region. Figures C.5 and C.6 show electricity generation and water consumption
by cooling system type, for context.
36:
In sharp contrast to coal mining, the water consumed at the point of coal-fired electricity
generation scales roughly with the amount of electricity generated at coal-fired facilities in each
subregion. The same trends are seen for nuclear generators, although their average water con-
sumption rates are slightly higher than coal-fired facilities. Unlike coal-fired generators, nuclear
facilities cannot discharge heat through flue gases. Natural gas-fired generators in the US have
higher average efficiency than coal-fired and nuclear generation units, in part due to use of com-
bined cycle configurations. Both the higher efficiency and use of gas turbines in a combined cycle
configuration contribute to lower cooling water requirements per unit of generation [77]. These
trends are evident in Figures 8.5 and C.4, where the water consumed at the point of genera-
tion for each type of thermal generator generally scales with its respective fraction of electricity
generation in each subregion.
36;

Abstract (if available)

Abstract Electricity systems are integral in the developed and developing world and access to electricity improves overall quality of life and enables basic human needs, such as communication, education and sanitation. The electricity system has complex relationships with the environment and is especially linked to the water system. Electricity is required for water treatment, transport, and delivery and water is required for primary fuel extraction, preparation, transport, and conversion to electricity. Although this relationship is understood, there are few efforts prior to this body of work that present usable data and methodologies to quantify the water used for the electricity sector. Specifically, this body of work quantifies water use for different generating technologies, examines the changes in water use (in magnitude and across space) in response to changes in the electricity grid, develops a methodology to quantify water and emissions with high temporal resolution, and calculates regionally-specific life cycle (i.e., from resource extraction to electricity generation) water consumption volumes and intensities for the United States (US) electricity sector. Collectively, this work represents a set of tools for researchers, water and energy planners, and decision makers to better account for the water requirements of electricity systems. This work is particularly relevant in a world with a changing climate, where water resource availability and distribution are uncertain and electricity grids are shifting quickly in response to climatic, political, and economic drivers.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Developing frameworks to quantify the operational and environmental performance of energy systems within the context of climate change

PDF

Investigating the role of climate in affecting residential electricity consumption through high spatiotemporal resolution observations

PDF

Residential electricity demand in the context of urban warming: leveraging high resolution smart meter data to quantify spatial and temporal patterns…

PDF

Integrating systems of desalination and potable reuse: reduced energy consumption for increased water supply

PDF

Using demand-side management for decarbonization: developing methods to quantify the impact of altering electricity consumption patterns

PDF

Thermally driven water treatment with membrane distillation: membrane performance, waste heat integration, and cooling analysis

PDF

Evaluating the role of energy system decarbonization and land cover properties on urban air quality in southern California

PDF

Evaluating energy consuming behaviors and the sufficiency of urban systems in the context of extreme heat hazards

PDF

Building occupancy modeling and occupancy-loads relationships for building heating/cooling energy efficiency

PDF

System stability effect of large scale of EV and renewable energy deployment

PDF

Defending industrial control systems: an end-to-end approach for managing cyber-physical risk

PDF

Integration of energy-efficient infrastructures and policies in smart grid

PDF

Groundwater contaminant transport predictions in a sustainable water management scenario

PDF

Synthesis and characterization of novel donor-acceptor systems for solar water splitting

PDF

Behavioral form finding using multi-agent systems: a computational methodology for combining generative design with environmental and structural analysis in architectural design

PDF

Electric vehicle integration into the distribution grid: impact, control and forecast

PDF

Smart buildings: employing modern technology to create an integrated, data-driven, intelligent, self-optimizing, human-centered, building automation system

PDF

Energy-efficient computing: Datacenters, mobile devices, and mobile clouds

PDF

Distribution system reliability analysis for smart grid applications

PDF

Impacts of heat mitigation strategies and pollutant transport on climate and air quality from urban to global scales

Asset Metadata

Creator Peer, Rebecca Allyson Marie (author)

Core Title Developing high-resolution spatiotemporal methods to model and quantify water use for energy

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Engineering (Environmental Engineering)

Publication Date 04/29/2019

Defense Date 03/07/2019

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag electricity systems,energy and environment,environmental impacts,OAI-PMH Harvest,power plants,water use

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Sanders, Kelly (committee chair), Dilkina, Bistra (committee member), Soibelman, Lucio (committee member)

Creator Email peerr@usc.edu,rebecca.peer@canterbury.ac.nz

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-159574

Unique identifier UC11660357

Identifier etd-PeerRebecc-7352.pdf (filename),usctheses-c89-159574 (legacy record id)

Legacy Identifier etd-PeerRebecc-7352.pdf

Dmrecord 159574

Document Type Dissertation

Format application/pdf (imt)

Rights Peer, Rebecca Allyson Marie

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA