A	J\ United States
Environmental Protection
^1 M m Agency
EPA 600/F-16/179 | July 2016
Innovative Research for a Sustainable Future
Research Summary
Water Energy Nexus
1.0 Introduction
In the water-energy nexus the electric power sector stands out as a crucial link between
water and energy systems. The U.S. energy system is comprised of primary energy
resources such as coal, natural gas, petroleum, uranium and renewables like wind, solar,
hydropower, geothermal and biomass. Refineries and electric power facilities convert
these primary resources to useable forms of energy such as electricity, gasoline, diesel
and other products. The end-use sectors include residential, commercial, industrial and light and heavy duty
transportation.
According to the Energy Information Administration more than 90% of the U.S. electricity supply is generated by
thermoelectric power plants that require an abundant supply of water from nearby water bodies such as rivers and
lakes for cooling. The U.S. Geologic Survey has shown the electric sector withdraws more water than any other
sector in the U.S., accounting for 45% of all total water withdrawals and 38% of total freshwater withdrawals for all
The Water-Energy Nexus
i

-------
For water consumption there is less recent data, but the electric sector has been estimated to account for between
3% and 6% of total water consumption in the U.S. This is expected to grow to 9% by 2030 however, making the
electric sector the fastest growing water consumer. A heavy dependence on reliable water supply means that the
U.S. electric sector is highly vulnerable to changes in water resource availability. The electric sector will be
negatively impacted when the water supply is compromised, affecting the ability to produce adequate cooling
causing reduced electricity generation or total shut down of a facility.
Thermoelectric cooling can be compromised in three ways. First, the water level of rivers and other cooling water
sources can fall below the cooling water intakes of nearby power plants. Second, high source water temperatures
can reduce the efficiency of power plant cooling systems and consequently limit plant generating capacity. Lastly,
state regulations may limit power plants from discharging heated cooling water into water bodies that already
exceed temperature thresholds designed to protect water quality and ecosystems.
NRMRL's research assessing the water demand of the energy system focuses primarily on changes in water
withdrawals and consumption. Withdrawals represent the amount of water diverted from a water body such as a
freshwater river, lake, reservoior, or saline or brackish water sources. Consumption represents the portion of water
lost to evaporation (e.g., steam from a cooling tower) or other losses that may occur where the water does not
return to the original water body. Our research has examined both the water consumption and withdrawal for a
range of scenarios generated by the MARKAL (MARKet ALIocation) energy system model using the EPA's U.S. nine-
region (EPAUS9r) database. A 2014 study, performed in collaboration with researchers from UNC Chapel Hill,
assessed the water demands for scenarios of system-wide C02 mitigation for the electric sector, looking only at
electric sector water use (Cameron et al. 2014). A more recent study, currently in journal review (Dodder et al.
2016), expanded the system boundares to asssess water use under a range of both C02 and water constrained
scenarios, estimating not only the water use by electric power generation facilities, but the life cycle water use for
the fuels they consumed (coal, natural gas, biomass, etc.) and in their construction. Finally, in 2014 NRMRL
produced a critical review of the literature examining a range of modeling tools and approaches used to assess the
future water demands of the energy system (Dodder, 2014).
2.0 The water demands of energy
w
Our research conducted at the National Risk Management Research Lab
(NRMRL) Air Pollution Prevention and Control Division (APPCD) addresses the
interaction between the energy and water sectors, by modeling the water
demands from the full energy system. This includes not only the water
demands for thermoelectric cooling and other electric power operational
water needs, but also water use for the full energy system. This means also
taking a full life-cycle perspective, including the water use for the primary
resource supplies as well as the water use in the manufacturing, construction
and decommissioning of power plants and other energy-production facilities.
2

-------
Electric sector water use under carbon constrained scenarios
Cameron et al. (2014) assessed the impact of long-range (2055) energy sytem
scenarios on electric sector water withdrawal and consumption to 2055. The
EPAUS9r MARKAL model was used to estimate changes in U.S. electric sector water
withdrawal and consumption through 2055 resulting from energy system-wide C02
emissions reduction targets of 10%, 25%, and 50%, as well as nine sensitivity
analysis scenarios. C02 reduction strategies accelerated the deployment of more
water-efficient thermoelectric generation technologies and increased the use of
renewables, nuclear, and carbon capture. The 50% C02 reduction scenario also prompted electrification in end-use
demand sectors, most notably transportation, increasing electricity demand 36% by 2055. In aggregate, C02
reduction strategies decrease national electric sector water withdrawal in all scenarios (-31% to -46% by 2055), but
have a varied impact on water consumption (-4% to +42% by 2055). These changes in electricity generation
technology and the resulting change in water use would likely reduce broad electric sector vulnerability to droughts
and heat waves, however the potential for more localized vulnerabilies exists. In comparison with previous work on
this topic, this study addressed a wider range of scenarios than was previously modeled and captured interactions
between multiple sectors (e.g., between electric power and transportation or other end-use demands) not
previously explored in the context of U.S. electric sector water use.
Broader energy system water use under carbon and water constrained scenarios
Currently, the operational (e.g., cooling systems) water use for thermoelectric generation dominates life cycle water
withdrawal and consumption for the electric sector. Water use for the associated fuel cycle (e.g. natural gas, coal)
and power plant manufacturing is substantially lower on a life-cycle basis. However, a remaining question is whether
shifting the full electricity mix toward low carbon and low water operations can lead to trade-offs across the life
cycle. Dodder et al. (2016) compared business-as-usual with scenarios of carbon reductions and water use
constraints using the MARKet ALIocation (MARKAL) energy system model. Our results showed that for water
withdrawals, the trade-offs appeared to be minimal - operational water accounts for over 95% of life cycle
withdrawals across all scenarios. For water consumption, however, this analysis identified some potential trade-
offs. Some scenarios showed significant increases in the water consumption in the upstream stages of the life cycle.
Nationally, water use for the fuel cycle and power plant manufacturing could reach up to 26% of the total life cycle
consumption. In the western and midwestern regions, non-operational water consumption in 2050 could even
exceed operational water use. In particular, water use for biomass feedstock irrigation and for the
manufacturing/construction of concentrating solar power facilities could increase with high deployment of these
low carbon options. As the U.S. moves toward lower carbon and lower water electric power operations, decision
makers may consider looking at shifting water demands in order to avoid unintended consequences.
3.0 Expected Outcomes I Results
The U.S. electric sector's reliance on water makes it vulnerable to climate variability and change. This research
informs the water use implications of approaches to decarbonization of the energy system, by exploring a range of
pathways and their changing water demands. The modeling also examines regional differences in the energy mixes
and associated water use. Using these studies, APPCD evaluates various carbon mitigation policies and how they
would improve or exacerbate electric sector water reliance. While these studies inform carbon and water trade-offs,
they do not address facility level water withdrawal with sufficient spatial and temporal detail to consider impacts on
3

-------
electric sector vulnerability to changes in water availability during periods of drought, for example. Future efforts
will aim to extend this work by linking the water demands with estimates of changes in water resources. In addition,
while there is a robust body of literature assessing the water demands of long range electricity and energy system
scenarios from a water quantity standpoint, there are limited studies that assess water quality implications of
broader energy systems scenarios, that capture the full life cycle of water use.
The expected outcome is a broader understanding of the potential water-related co-benefits or tradeoffs between
climate migitation strategies (carbon reductions) and climate adaptation and resiliency strategies (water use
reductions).
Keywords
Environmental Protection Agency (EPA), APPCD, climate mitigation, energy system modeling, MARKAL, carbon
dioxide (C02), water quantity, consumption, water-energy nexus, electric power, renewables
Contact
Rebecca Dodder
dodder. rebecca(a)epa.gov
919.541.5376
Lead
4

-------