The Water-Energy Nexus in a Changing World

Introduction

Water and energy are the two most fundamental ingredients of modern civilization.  Without water, people die.  Without energy, we cannot grow food, run computers, or power homes, schools, or offices.  The interdependency between the world’s two most critical resources is receiving more and more attention from academia, economist, and engineers as well as the general public.  A comprehensive and in-depth understanding of the water-energy nexus is essential to achieve sustainable resource management.

The paradox of the water-energy nexus has gained additional traction with greater concerns regarding various climate change scenarios.  Drought in the Southwest United States could produce power systems compromising this region in the context of supply-demand balance, thermal and hydro-capacity loses, reserve margin reductions, and overall system reliability and vulnerability.

I live in Texas where Texas is on the front page of concerns about the water-energy nexus.  As Texas has seen the evolution of one energy industry (i.e., hydrofracking for expanded natural gas development – keep in mind that the “hydro-“ portion of fracking is essential) it has seen the decline of another portion of the energy matrix.  This was recently reported in the New York Times (Malewitz, 2014):

“Faced with dwindling water supplies, the Lower Colorado River Authority, which supplies water and energy to much of Central Texas, is limiting downstream water releases for activities like rice farming.  Aside from stirring controversy among water users, the changes have shrunk the amount of electricity the agency generates from its six Colorado River dams. 

“Your hydropower becomes an innocent bystander of the conditions around it,” said Robert Cullick, a former River Authority spokesman who is now a consultant.

Hydroelectricity makes up a sliver of the L.C.R.A.’s energy portfolio, a mix of coal, natural gas and wind energy, and its further decline would probably not affect the region’s energy reliability.  But its possible extinction would close the book on a fuel source that played a major role in the history of Central Texas and the creation of the River Authority, whose dams make up about 40 percent of the state’s hydropower capacity.”

Hydropower is not the only energy source at risk in Texas in an era of extreme weather events and concerns relating the dependencies embedded in the water-energy nexus.  The Texas Comptroller of Public Accounts issued the following warning regarding the Texas drought of 2011 (a year in which Amarillo, Texas received the same annual rainfall as Damascus, Syria – roughly 5.5 inches) (Texas Comptroller of Public Accounts, 2012):

“Extended drought may affect the price and availability of electrical power in Texas, due both to the demand for summer air conditioning and the fact that most power plants use large amounts of water for cooling.

On December 1, 2011, the Electric Reliability Council of Texas (ERCOT) warned that another hot, dry summer could push the state’s power reserves below the minimum target next year.

More than 11,000 megawatts of Texas power generation – about 16 percent of ERCOT’s total power resources – rely on cooling water from sources at historically low levels.  If Texas does not receive “significant” rainfall by May, more than 3,000 megawatts of this capacity could be unavailable due to a lack of water for cooling.”

Consider the following as water and energy interdependencies increasingly become problematic across the United States (Webber, 2008):

“The paradox is raising its ugly head in many of our own backyards.  In January, Lake Norman near Charlotte, N.C. dropped to 91.7 feet, less than a foot above the minimum allowed level for Duke Energy’s McGuire Nuclear Station.  Outside Las Vegas, Lake Mead, fed by the Colorado River is now routinely 100 feet lower than historic levels.  If it dropped another 50 feet, the city would have to ration water use, and the huge hydroelectric turbines inside Hoover Dam on the lake would provide little or no power, potentially putting the booming desert metropolis in the dark.

Research scientist Gregory J. McCabe of the U.S. Geological Survey reiterated the message to Congress in June 2013.  He noted that an increase in average temperature of even 1.5 degrees Fahrenheit across the Southwest would compromise the Colorado River’s ability to meet the water demands of Nevada and six other states, as well as that of the Hoover Dam.  Earlier this year scientists at the Scripps Institution of Oceanography in La Jolla, Calif., declared that Lake Mead could become dry by 2021 if climate changes as expected and future use is not curtailed.”

The national press and television outlets have widely reported on the continuing drought in the Southwest and West.  This particular drought has had negative impacts on reservoir storage.  Reservoir storage can be critical in the context of some forms of hydro and thermal power systems.  Consider the following (Fulp, 2005):

“The effect of the drought is immediately evident in reservoir storage.  In 1999, reservoirs on the Colorado River collectively were more than 90 percent full.  Today [2005] the system-wide storage is about 50 percent, a decrease in volume of some 25 million acre-feet of water.  Although the situation is very serious, the reservoir system is clearly doing its job, so about 30 million acre-feet of water remain in storage, or nearly two full years of average inflow into the system.”

Problems in the water-energy nexus are enormously complex.  The continuing drought in California highlights the challenges and interconnections between water and energy.  The Wall Street Journal highlighted this in a recent article (Heard on the Street Column, 2014):

“More than half of California is classified as being in a state of extreme drought, according to the U.S. Department of Agriculture.  Recent storms have brought some snow to parched slopes – including Tahoe’s – but they come very late: California’s snowpack is just 10% of normal levels, according to Citigroup.  The problem extends up the coast, with the Northwest River Forecast Center earlier this month reporting lower-than-normal precipitation in the region this season.

This matters because almost half of U.S. hydroelectric lies in California, Washington, Oregon, Idaho and Montana.  California, the country’s second-largest electricity market after Texas, got 17% of its power this way in the decade ending 2012.

So if rivers are low, the state has a problem – even more so when other sources of energy are stress as well.”

Extreme winter weather and constraints in the water-energy nexus forced prices for natural gas entering California up to $15 a million BTUs in late January from $4.19 at the end of 2013.  From the Wall Street Journal article:

“Adding to this is the fierce cold and snow battering New York a host of other places across the U.S.  Natural-gas-fired power plants make up more than 60% of California’s capacity, so these take the strain when the rivers run dry.  The problem is when the rest of the country needs gas for heat, supplies can be constrained.”

Thermal-based power facilities such as nuclear and coal-fired plants, are critically dependent on water for cooling purposes.  This enables them to maintain high production efficiencies, but also means that they use tremendous volumes of water every day.  Thermal power plants – those that consume coal, oil, natural gas or uranium – generate more than 90 percent of U.S. electricity, and they are water hogs (Webber, 2008).  The sheer amount required to cool the plants impacts the available supply to everyone else.  Although a considerable portion of the water is eventually returned to the source (some evaporates), when it is emitted it is at a different temperature and has a different biological content that the source, threatening the environment.  All of this takes place in a context in the U.S. where 520 billion kilo-watt (KWh) is required to move, treat, and heat its water, which accounts for up to 60% of the energy bill in some cities – equating to 13% of the entire electricity use in the United States (the carbon dioxide emission for the water portion of the water-energy nexus is equal to the annual emissions of 53 million cars) (Smedley, 2013).

Purpose of Paper

The goals of this paper focus on two areas.  The first is a general understanding of the operating environment that power generation and transmission systems face in drought and temperature stress environments.  The general impacts of drought and higher temperatures on power systems is as follows (Argonne National Laboratory, 2012):

Thermo-Electric Plants

·       Use surface water for cooling, fuel processing, and emissions control.

·       Low water level limits the amount of water that can be withdrawn (minimum water elevation limits).

·       Intake structures could be exposed (above water level).

·       High water temperatures at intake may lead to violation of water discharge regulations.

·       High temperatures lowers plant heat rate (efficiency).

Hydro-Electric Plants

·       Lower inflows means low power output.

·       Lower reservoir levels mean less water available for power generation and degraded water-to-energy conversion factors.

Gas-Fired Plants

·       High ambient temperatures limit cooling ability of air-cooled systems.

·       High temperatures decrease efficiency and capacity.

Photovoltaic Cells

·       High temperatures reduce efficiency and outputs of photovoltaic units.

Transmission Lines

·       High temperatures lower the thermal limits of transmission lines and circuit breakers.

·       High temperatures increase transmission loss and operation cost.

·       High ambient temperatures lower throughputs of transformers.

An important point in the water-energy nexus goes beyond just volumetric concerns.  Rising water temperatures play an important part in the water-energy nexus.  Accordingly, the second goal of the paper is the presentation of a model developed by researchers at the Norwegian School of Economics for the German electric markets in the context of electricity prices, river temperatures, and cooling water scarcity (McDermott & Nilsen, 2012).

Several examples outlined in the paper will be from the Southwest United States.  This is an area of the U.S. which is subject to drought and climate change concerns coupled with population increases.  As background, the installed capacity mix by fuel type in this area of the U.S., including ERCOT, is as follows (Argonne National Laboratory, 2012):

Fuel Type	Nameplate Capacity (MW)	Percent Share (%)
Coal	49,458	19%
Hydro	19,556	7%
Natural Gas	157.987	59%
Oil	1,523	1%
Others	23,107	9%
Uranium	13,925	5%
Total	265,555	100%

 

Given the capacity of the Southwest, it is important to examine the share of high-risk drought capacity also in the Southwest.  Some 61% of the capacity has been rated at high-risk in the context of droughts (Argonne National Laboratory, 2012):

Drought Risk Type	Capacity (MW)	Percent Share (%)
High-Risk Thermal	149,336	54%
High-Risk Hydro	19,552	7%
Low-Risk Thermal and Others	102,667	39%
Total	265,555	100%

 

 

Power Generation and Our Risky Climate

Over the past 15 years, increasing concerns about the risks to the electric grid from severe drought and hotter temperatures have grown for managers, engineers, and owners of electricity generating plants.  Recent drought events in the Pacific Northwest and California in 2001, in the Southwestern United States in 2007 and 2008 and in Texas in 2011, along with the uncertain impacts of climate change, have heightened these concerns.

Harto and Yan have written extensively about drought impacts on electricity production in the Western and Texas interconnections of the United States.  Several of their more important and interesting conclusions are outlined below (Harto & Yan, 2011):

·       The greatest precipitation shortage occurs during the winter months (which is exactly California is experiencing winter 2013-14).

·       The climate change observed in the 20^th century (an increase of 1-3 degrees in spring temperature, a decline in spring show pack, and snow water equivalent, and a shift to early peak runoff) has been projected to continue throughout the 21^st century in much of the western United States.

·       For a water resource region that has significant storage capacity, a one-year drought is expected to have a limited impact.  For an individual river system that has limited storage capacity, it is likely that droughts with durations of 1-5 years would have significant impacts on flow reduction.

·       Drought events in the Pacific Northwest and Great Basin regions show a longer duration and lower frequency, whereas droughts in the Texas Gulf have shorter duration and higher frequency.

·       During the 2011 drought, California and the Pacific Northwest saw significantly reduced hydroelectric power generation, resulting in tight electricity supplies and high prices.

·       The Union of Concerned Scientists reported that the Tennessee Valley Authority (TVA) was forced to temporarily shut down its Browns Ferry Facility, and a few others were forced to reduce generation during a particularly acute period of drought in August 2007.  During this period, TVA was forced to purchase electricity from the grid to meet demand.  This outrage appears to have been more a result of an increase in the temperature of the cooling water source than due to limitations in the availability of water.

·       In general, the literature indicates that hydro generation is far more significantly affected by drought than thermoelectric generation.

·       While hydro generation has been shown to vary by large margins depending upon hydrological conditions, there have been limited reports of the thermoelectric functions being forced to shut down involuntarily.

·       Regional areas that are more prone to drought are more resilient than areas with less experience and are likely to have put more effort into planning and developing mitigation strategies.

·       Of the 423 power plants reviewed, 43% were identified as having cooling-water intake heights of less that 10-feet below the typical water level of their water source.

·       Of 580 power plants reviewed, 60% (representing approximately 90% of the total generating capacity) were deemed to be vulnerable on the basis of either supply or demand – related criteria.

Concerns regarding water-energy issues in the Southwest U.S. are highlighted throughout this paper.  Population and the water-energy nexus are tightly linked.  Consider the following in the context of the Southwest (Fisher & Ackerman, 2011):

“Since 1950, there has been a strong increase in the proportional growth of suburban populations.  In 2000, suburbanites accounted for 50% of the population.  Southwestern suburban developments, in which 70% or more of the water is often used for landscaping, amplify the water demands exerted by the increasing population.  Sabo et al. estimate that per-capita virtual water footprints are seven times higher for cities in the arid West than in the East.  They suggest that with a doubling of population, the West would require the equivalent of more than 86% of its total stream-flow to meet human use at current per-capita levels.”

Climate change ramifications in the Southwest have broad concerns for the water-energy nexus.  The 21^st century will be marked by increasing risk as outlined below (MacDonald, 2010):

“. . . with greenhouse gas concentrations at their current levels, we likely will not escape significant warning and resulting increased aridity over the 21^st century.  Coupled with the demographic projections, the climatic estimates for the next decade compel us to develop water resources strategies that adapt to these changing conditions and promote sustainability in the face of increasing general aridity as well as more serve episodic droughts.  Finally, the proximal economic costs of reducing greenhouse gas emissions are often cited as a rationale for inaction on emissions reduction.  Because climate warming will exacerbate water sustainability problems, the Southwest is likely to experience some of the highest economic expenses and environmental losses related to climate change.  As the papers in this issue illustrate, the ultimate costs of inaction in curbing greenhouse gas emissions will be particularly high for the Southwest.”

I live in Southlake, Texas (between Dallas and Fort Worth) and have experienced the continuing drought in Texas.  The following comments are on the minds of many electricity providers in Texas (Cayan, 2010):

“One way that climatologists measure drought is by comparing annual rainfall to a long-term norm.  A serve drought in Texas early in the twentieth century was caused by a deficit of 10-12 inches of rain compared to the long-term norm, and one in the 1950s was brought on by a 6- to 8-inch deficit.  In 2011, the deficit was 13 inches, making the resulting drought the most severe recorded since record keeping had begun in 1896.  The 2011 drought also covered a huge area compared to previous droughts in Texas.  By August of 2012, a 62 percent of the contiguous US was declared to be under “moderate to exceptional drought.””

Many organizations are starting to make plans for a world of climate change risk.  The seven Colorado River basin states recently published Study of Long-Term Augmentation Options for the Water Supply of the Colorado River System (Colorado River Water Consultants, 2013).  Study outlined various augmentation options within the basin.  One has a direct impact on the water-energy nexus:

·       Add to Rainfall – weather modification, such as cloud seeding.

·       Reduce Evapotranspiration – vegetation management and reservoir evapotranspiration control.

·       New Water/Reuse from the System – Use desalted brackish water, reuse wastewater, use desalted or inland water.

·       Add to Groundwater – Conjunctive use.

·       Reduce Outflow from System – Reduction of power plant water consumptive use and stormwater storage.

·       Add to Inflow – Water from coalbed methane production and importation alternative.

Provided below is summary of the various alternatives evaluated in the study with the cost in dollars per acre-feet (Colorado River Water Consultants, 2013):

Alternative	Quantity Evaluated (Acre-feet per year)	Cost ($/Acre-feet)
Brackish Water Desalination	4,000 – 5,000	$700 - $2,000
Coalbed Methane Produced Water	3,000 – 20,000	$900 - $4,600
Conjunctive Use	8,000 – 40,000	$400 - $700
Ocean Water Desalination	20,000 – 100,000	$1,100 - $1,800
Power Plants – Reduce Consumptive Use	1,500 – 160,000	$1,000 - $4,000
Reservoir Evaporation	0 – 270,000	$500 - $2,000
River Basin Imports	30,000 – 700,000	Needs more refinement
Stormwater Storage	0 – 100,000	$600 +
Vegetation Management	20,000 – 150,000	$30 - $100
Water Imports Using Ocean Routes	10,000 – 300,000	$1,400 - $4,000
Water Reuse	20,000 – 800,000	$900 - $1,700
Weather Modification	150,000 – 1,400,000	$20 - $30

 

One can see numerous interface points with the water-energy nexus with the planning list detailed above.  Clearly the Power Plant – Reduction of Consumptive Use is a category critical to this research and paper.  The report further outlines the following in the context of power plants (Colorado River Water Consultants, 2013):

“Thermoelectric power generation requires a significant amount of water within the Basin to provide cooling to power plants and remove waste heat from the power generation cycle.  Evaporative cooling is the most common method used within the Basin.

The White Paper compared “wet cooled” systems, such as once-through cooling systems and recirculated cooling systems water systems, to air-cooled systems, which use an air-cooled condenser instead of the typical water-cooled condenser.  It was found that air-cooled systems eliminate the consumptive use of water for plant cooling, but at the cost of lower plant efficiencies and increased plant capital costs.  The Technical Committee determined that this option should be addressed by individual States.”

Any attempt to reduce the consumptive use of water for power plants in the region must first consider the scale of the endeavor.  This a sample of power plants in the Colorado River System (Colorado River Water Consultants, 2013):

Plant Name	Plant Capacity (MW)	Consumptive Use (Acre-feet per year)	Water Source
Navajo	2,409	27,366	Lake Powell
Jim Bridger	2,312	25,266	Green River
Four Corners	2,270	22,515	San Juan River
San Juan	1,848	19,981	San Juan River
Hunter	1,441	18,968	Cottonwood Creek
Huntington	996	12,307	Huntington Creek
Bonanza	500	7,964	Green River
Reid Gardner	612	7,500	Muddy River
Naughton	707	6,081	Hams Fork River
Hayden	465	2,896	Yampa River
Carbon	189	2,679	Price River
Craig	1,339	2,534	Yampa River
South Point Energy Center	708	1,955	Colorado River
Desert Basin Power	646	1,810	Central Arizona Project Canal Water
Nucla	114	1,520	San Miguel River

 

In some respects, the power generation component of the water-energy nexus is more embedded with our water delivery systems than people and policy makers realize.  For example, the energy needed to move agricultural water in California exceeds the electricity used by everyone in San Diego.  Consider the following (Los Angeles Times , 2014):

“The energy cost of water also varies by locale.  Water transported to Southern California is almost three times as energy intensive as water moved in the southern part of the state.  Likewise, it’s more expensive to bring water to houses atop the Hollywood Hills than by those in the flatlands.  Some uses consume more energy as well.  Watering outdoor plants, which results in no sewage treatment, demands less energy than, say, running the dishwasher.

The biggest spigot, though, is the agricultural sector, which consumes 60% of the state’s water.  Water used for farming requires less treatment and doesn’t wind up in the sewer, so it’s less energy intensive per gallon.  Still, the volume dwarfs any other water use.

The energy needed to move agricultural water exceeds the electricity used by everyone in San Diego.”

In closing, it should also be noted that The World Bank is also increasingly concerned about the water-energy nexus in both the developed and developing world (The World Bank, 2014).  Their new initiative aims to address the interconnection between energy and water head-on by providing countries with “assessment tools and management frameworks” to help governments “coordinate decision-making” when planning for future energy and water infrastructure.  The name alone of the World Bank website paints a picture for the energy and power industries – Thirsty Energy.  Consider the following from a report issued this year (The World Bank, 2014):

“Today, more than 780 million people lack access to potable water, over 1.3 billion people lack access to electricity.  At the same time, estimates show that by 2035, global energy consumption will increase by 35%, while water consumption by the energy sector will increase by 85%.  Climate change will further challenge water and energy management by causing more water variability and intensified weather events, such as severe floods and droughts.

While a global water crisis could take place in the future, the energy challenge is present.  Water constraints have already adversely impacted the energy sector in many parts of the world.  In the U.S., several power plants have been affected by low water flows or high water temperatures.  In India, a thermal power plant recently had to shut down due to a severe water shortage.  France has been forced to reduce or halt energy production in nuclear power plants due to high water temperatures threatening cooling processes during heatwaves.  Recurring and prolonged droughts are threatening hydropower capacity in many countries, such as Sri Lanka, China and Brazil.”

Thermal pollution has a range of impacts on water quality.  These are discussed in the following from Wikipedia (Wikipedia, 2014):

“Elevated temperature typically decreases the level of dissolved oxygen of water. This can harm aquatic animals such as fish, amphibians and other aquatic organisms. Thermal pollution may also increase the metabolic rate of aquatic animals, as enzyme activity, resulting in these organisms consuming more food in a shorter time than if their environment were not changed.   An increased metabolic rate may result in fewer resources; the more adapted organisms moving in may have an advantage over organisms that are not used to the warmer temperature. As a result, food chains of the old and new environments may be compromised. Some fish species will avoid stream segments or coastal areas adjacent to a thermal discharge. Biodiversity can be decreased as a result.

High temperature limits oxygen dispersion into deeper waters, contributing to anaerobic conditions. This can lead to increased bacteria levels when there is ample food supply. Many aquatic species will fail to reproduce at elevated temperatures.

Primary producers are affected by warm water because higher water temperature increases plant growth rates, resulting in a shorter lifespan and species overpopulation. This can cause an algae bloom which reduces oxygen levels.

Temperature changes of even one to two degrees Celsius can cause significant changes in organism metabolism and other adverse cellular biology effects. Principal adverse changes can include rendering cell walls less permeable to necessary osmosis, coagulation of cell proteins, and alteration of enzyme metabolism. These cellular level effects can adversely affect mortality and reproduction.

A large increase in temperature can lead to the denaturing of life-supporting enzymes by breaking down hydrogen- and disulphide bonds within the quaternary structure of the enzymes. Decreased enzyme activity in aquatic organisms can cause problems such as the inability to break down lipids, which leads to malnutrition.

In limited cases, warm water has little deleterious effect and may even lead to improved function of the receiving aquatic ecosystem. This phenomenon is seen especially in seasonal waters and is known as thermal enrichment. An extreme case is derived from the aggregational habits of the manatee, which often uses power plant discharge sites during winter. Projections suggest that manatee populations would decline upon the removal of these discharges.”

Water Requirement Examples for Electric Power Plants

The BP Energy Outlook 2035, January 2014 outlines several trends that will ultimately indirectly impact the water-energy in the coming decades (the report offers no direct water-nexus concerns which should be noted):

·       Primary energy demand increases by 41% between 2012 and 2035, with growth averaging 1.5% per annum.

·       We are leaving a phase of very high energy consumption growth.  The 2002-2012 decade recorded the largest ever growth of energy consumption in volume terms over a ten year period, and this is unlikely to be surpassed in our timeframe.

·       Coal’s contribution to growth diminishes rapidly.

·       Energy consumption grows less rapidly than the global economy.

·       One of the longest established trends in energy is the increasing coal of the power sector.

·       Coal’s share declines in all sectors.  In power generation, the largest coal consuming sector, the share of coal will decline from 43% in 2012 to 37% by 2035, as renewables gain share.

·       Looking beyond 2030 illuminates a potential turning point for nuclear energy.  Many reactors among the first adopters of nuclear technology, such as the U.S. and Europe, will approach technical retirement, while only a few countries plan to add new capacity.  Even allowing for additional lifetime extensions, we may well see a peak in nuclear energy.

·       Historically, as economies grew richer and more sophisticated, the fuel mix became more diversified.  The scope for changes in the fuel mix depends on technology, resource endowments and tradability, and the underlying economic structure.  As incomes rise we put a premium on cleaner and more convenient fuels.  The actual substitution between fuels is typically guided by relative prices.

The outlook of the U.S. Energy Information Administration paints a better and more clear picture of the challenges embedded in the water-energy nexus during this century (Hadian & Madani, 2013):

“The outcomes reveal the amount of water required for total energy production in the world will increase by 37% - 66% during the next two decades, requiring extensive improvements in water use efficiency of the existing energy production technologies, especially renewables.”

The cooling system is an essential component in most electric power plants.  Several different types of systems are available that have unique impacts to the water-energy nexus.  These are (Carney, 2010):

·       Once-through, fresh water cooling systems are more likely to be affected by lower water levels in lakes, rivers, and streams that occur in sustained drought periods.  The vast majority of once through cooled power plants in Texas withdraw water cooling reservoirs that were constructed by a utility to support the power plant.  Cooling water is pumped through a condenser to condense the steam which is then pumped back to the boiler to complete the cycle.  Virtually all the cooling water is returned to the cooling reservoir where it re-circulates, cools naturally, and can be pumped back to the condenser or used for other purposes.  Once-through cooling systems are the simplest, least expensive, and most effective technology for condensing steam, providing the best power plant efficiency (i.e., the most electricity is produced for the amount of fuel burned).

·       Once-through, salt water cooling systems withdraw and discharge from larger bodies of water (oceans, bays and sounds) which are slightly less likely to be affected by lower water levels in sustained drought periods, though they can be affected by temperature regulations.

·       Wet cooling tower systems pump water from a water source (which can be municipal wastewater plant effluent, captured rain and storm water runoff, groundwater, and/or surface water) through a condenser and then to a cooling tower.  Large fans (forced draft) or hyperbolic designs (natural draft) provide air flow to dissipate the transferred heat from the cooling water to the air, primarily by means of evaporation.

·       Closed-cycle, hybrid and other systems either reuse water after withdraw or use very little water for cooling.  Over half of the cooling systems at U.S. power plants re-use water through a cooling tower, though some of the larger plants in the nation have once-through systems from fresh water sources.  There are currently no power plants in Texas with hybrid cooling systems.  Hybrid cooling systems are dual cooling systems that have both a wet cooling component and a dry cooling component.  The two primary types of hybrid cooling systems and plume abatement systems and water conservation systems.

·       Dry-air cooled systems use essentially no water for cooling purposes but are not in wide use at this time.  There are only two power plants with dry cooling systems currently operating in Texas.  Both of these plants employ air-cooled condensers (ACCs) to condense steam, this is known as direct dry cooling.  Because more electricity must be used to operate the cooling equipment, less net electricity is produced form the fuel burned.  This translates to increased fuel consumption.

The Texas Water Resources Institute recently completed a study of 24 Texas power plants with one-through cooling systems (Water Conservation & Technology Center, 2012).  The water consumption information is provided below.  While this is not an exhaustive list, it is fairly representative of Texas plants using once-though cooling.

Facility	Water Consumed (ACFT/Plant Unit)	Water Consumed Per Electric Generation (ACFT/ 1,000 MWH)	Water Consumed Per Electric Generation (Gallons/KWH)
Plant 1	11,914.4	1.05	0.49
	4,718.0
Plant 2	250.0	1.24	0.40
	23,522.0
Plant 3	9,774.3	1.04	0.34
Plant 4	2,602.0	1.40	0.46
Plant 5	206.0	1.20	0.41
Plant 6	3,707.6	0.62	0.20
Plant 7	1,797.0	1.20	0.40
Plant 8	3,509.0	0.78	0.25
Plant 9	379.9	0.54	0.18
Plant 10	13,896.4	1.04	0.34
Plant 11	426.3	1.25	0.41
Plant 12 (2 Units Combined)	37,893.0	1.79	0.58
Plant 13	21,066.3	0.99	0.33
Plant 14	505.8	1.00	0.33
Plant 15	405.9	1.20	0.40
Plant 16	5,176.0	0.99	0.32
Plant 17	13,262.2	1.75	0.57
	9,688.6
Plant 18	9,366.1	1.18	0.39
Plant 19	219.9	1.00	0.33
Plant 20	680.2	1.30	0.42
Plant 21	2,779.6	1.71	0.56
Plant 22	35.1	1.00	0.33
Plant 23	636.1	0.92	0.30
Plant 24	285.7	1.06	0.35
Average		1.14	0.38

 

In summary, there are a range of cooling systems.  However, two types of systems account for the vast majority of power plant cooling.  The first system (open-loop wet cooling) withdraws a lot of water but consumes relatively little of what it withdraws; the second system (closed-loop wet cooling) withdraws less water but consumes a larger proportion of what it withdraws.  Unfortunately there is a tradeoff between water withdrawal and water consumption.  Either withdrawal is relatively high but consumption is relatively low or withdrawal is relatively low and consumption high (Argonne National Laboratory, 2012).

The 1970s saw a shift in how power plant cooling systems were designed (Carney, 2010).  Plants built before the 1970s tended to withdraw large amounts of water via open-loop-wet cooling systems.  In response to concerns about their impact on marine life, most plants built since the 1970s use closed-loop wet cooling systems that withdraw relatively less water, but consume large quantities of water. 

Engineers would agree the existing closed-loop wet cooling in gas-fired power plants consumers approximately 180 gallons to produce one MWh of electricity (one MWh is roughly the electricity required by an average plasma screen TV per year).  All thermoelectric power plants including natural gas, coal, oil, nuclear and solar thermal also have options for alternative cooling systems (dry or hybrid).  However these options generally reduce the efficiency (the heat rate increases) and more expensive).

Thermoelectric power plants use water to cool down (condense) steam after it has been used to turn a stream turbine to generate power.  For once-through cooling systems fed by fresh water sources, the need to withdraw significant amounts of water makes these plants more vulnerable to deratings or outages when water levels drop or water temperatures rise.  When water levels fall significantly, water intake structures may be exposed above the water surface, causing the plant to become nonoperational.  Additionally, at higher water temperatures, generators are less efficient, reducing the power capability of the plant.  Some areas also place regulatory limits on the temperature of the water a cooling system discharges.  At times of excessive heat, power plants are not allowed to raise water temperatures past levels safe for species of fish and other aquatic life (The Johnson Foundation at Wingspread, 2013).

The table below outlines water consumption for electric generation in the Southwest States (Argonne National Laboratory, 2012).  Keep in mind the water-energy nexus deals with both water withdrawal (electric power plants account for more than 40% of water withdrawal in the U.S. while electric generation in the Southwest States consumes less than 2% of the total amount of water withdrawn).

Water Consumption for Electric Generation in Southwest States
State	Withdrawal Rate (cfs)	Consumption Rate (cfs)	Percent Consumed (%)	Net Generation 2010 (MWh)	Net Generation per Water Consumed (MWh per cfs)
AZ	694.4	667.1	96%	18,762,284	28,125
CA	173,750.0	589.7	0.3%	16,244,290	27,547
CO	932.9	799.1	86%	19,145,034	23,958
NM	255.9	270.2	106%	7,938,534	29,380
NV	1,228.8	187.9	15%	9,349,924	49,760
TX	285,244.3	4,902.6	1.7%	102,596,558	20,927
UT	1040.8	1,040.8	100%	18,836,843	18,098
Total	463,147.1	8,457.4	1.8%	192,873,466	22,805

 

 

Modeling the Impact of River Temperatures and Electricity Prices

Thermal-based power facilities, such as nuclear and coal-fired, are critically dependent on water for cooling.  This enables them to maintain high production efficiencies (i.e., lower heat rates).  As previously mentioned, the thermal industry accounts for roughly 40% of all freshwater withdrawals in the United States.  The majority of these withdrawals are actually returned to their source.  The excess thermal energy absorbed by cooling water during the heat exchange will naturally cause it to warm up prior to being released back into the river of lake from which it was withdrawn.  This can ultimately raise ambient temperature of the water source itself and cause detrimental effects to the aquatic ecosystem.

The context of the water-energy nexus is typically stated in terms of quantity.  Will there be enough water for cooling?  But another issue is quality.  The Fourth Assessment Report of the intergovernmental Panel on Climate Change suggested that future energy generation will be vulnerable to higher temperatures and a reduced availability of cooling water for thermal power stations.  This is a key point regarding climate change and drought conditions – rising water temperatures reduce the cooling efficiency of thermal power plants.

McDermott and Nilsen of the Norwegian School of Economics have extensively studied electricity prices, river temperature, and cooling water scarcity in the German energy markets.  It is a well-known operational fact thermal energy can be converted into electrical energy more efficiently in the presence of an external coolant, such as water – in other words the production of electricity is contingent on the difference in temperature of the discharge water at the outlet point.  This is illustrated in the following equation:

Q = A(T_EW-T) x W, where (also see Exhibit 1)

Q = Production of electricity

TEW = Temperature of the discharge water at the outlet point

T = Temperature of the water at the intake point

The production of electricity by thermal-based power plants is subject to the following constraint:

W/S x TEW + (S-W/S) x T ≤ T*, where

T* = Cap on the temperature of the downstream river (typically set by environmental authorities)

S = River volume

S – W = River water not used for cooling

W/S = Share of total river water for cooling

The constraint equation implies that rather than completely shutting a power plant down, the operators of the plant have the option of reducing the flow of discharge relative to the volume of downstream mixing water when the temperature of each unit of discharge water, T_EW, is relatively hot.  The authors point out as the temperature of the river water itself approaches the regulatory limit (e.g., during the very hot summer months common in the Southwest) the plant management has little room for maneuvering and will likely have to decrease electricity output.

Remember that environmental authorities will also typically impose limits on the temperature of the discharge water itself (T_EW) and/or on the temperature differential between river water at the intake point and the discharge.

The strategic managerial decision variable to power plants in the model is quantity.  As pointed out in this class, electricity is a homogenous product that cannot be stored.  Demand must be perfectly balanced by supply at all times.

The authors further outline the model for profit, ∏, for thermal-based plants as follows:

∏= p(Q + F) Q – c(Q) – p_W(RL) x W, where

P(Q + F) = The inverse demand function and total electricity demand is the sum of power produced by the analyzed plants, Q, together with electricity imports and the other sources that aren’t dependent on cooling water (e.g., wind power), F

C(Q) = The marginal costs associated with the production of additional quantities of electricity

p_W(RL) x W = Reflects the fact that there are costs associated with drawing cooling water, W, from the external coolant (river).  These are said to be a function of the river level, RL, such that p_W<0

The research of McDermott and Nilsen rests on two demand and supply equations, where electricity prices and quantities are jointly determined in a “market-clearing equilibrium.”  The supply and demand equations that form the basis for their regression model is as follows:

Supply equation

lnP_t = β_o+β₁lnQ_t+β₂lnRiver Level+β₃lnRiver Temp.+β₄lnF_t+β_tT_t+V_t

Demand equation

lnQ_t=α₀+α₁lnP_t+α₂lnHDD_t+α₃lnCDD_t+α₄lnNWD_t+α_tT_t+W_t

Where,

P = Daily clearing price for electricity

Q = Daily electricity consumption

River Level = The aggregated river level

River Temp. = River temperature

F = Fuel (input) costs

HDD = Heating degree-day (degrees below 18⁰C outside air)

CDD = Cooling degree-day (degrees above 22⁰C outside air)

NWD = Non-work days (i.e., either a weekend or public holiday (0/1))

T = A set of seasonal and trend variables

The authors are primarily interested in the supply equation since this captures how electricity production is effected by access to cooling water.  The supply of electricity is defined by its price (P), which is then a function of quantity (Q) and several supply-related variables.  The supply side regressors of greatest interest for this particular study are river levels (River Level) and river temperatures (River Temp).  These two coefficients should reflect how electricity supply is constrained by diminishing cooling water availability, due to either relative scarcity (i.e., falling river levels) or regulatory concerns (i.e., river temperatures breaching environmentally sensitive thresholds).

The demand equation includes two terms that capture the nonlinear effect of changing temperatures on electricity demand – Heating degree day (HDD) and cooling degree day (CDD) capture the extent to which air temperatures fall outside a given comfort zone.  These two variables thus allow the demand function to respond to the discomfort presented by both cold and warm weather. 

The following is a summary regarding data collection for the model:

·       The data consist of daily values over the period 2002 to 2009.

·       Data on German spot electricity prices and values were obtained from the European Energy Exchange AG. 

·       The focus was exclusively on the base load – power plants most vulnerable to water-related factors – such as nuclear and coal-fired plants – are all base load electricity operators.

·       Air temperatures were obtained from the German Meteorological Service (Deutscher Wetterdienst).

·       Hydrological data, in the form of river levels and temperatures, was obtained from the Federal Institute of Hydrology (Bundesantalt Fur Gewasserkunde).

·       Data measurements were taken from gauging stations situated at various German rivers – the Elba, Main, Necka, and Rhine.

·       These rivers acted as the water source for a number of nuclear plants during the 2003-2009 period, in addition to several coal-fired plants that also suffered reduced capacity due to restrictions.  The dataset was able to capture the relevant effects of cooling water scarcity and environmental regulations.

·       Apart from being log-transformed, data from the River Level series were entered directly into the regression model.  The authors made two adjustments to the River Temperature series to better capture how regulation of thermal pollution impacts electricity prices.  The first was to generate a standard dummy variable that tests for a difference in the price intercept when river temperatures exceed a defined regulatory limit of 25⁰C.  The second is to specially measure the continued rise is temperature above 25⁰C.  The authors point out this formulation is aimed at ensuring some flexibility and allows for a non-linear temperature effect around the regulatory threshold.  See Exhibit 2.

·       While oil-fired plants do not play a substantial role in the German electricity market, oil is widely used as a proxy for natural gas and it is even used within the power industry to forecast the general price movements of coal.

Table 1 reflects the results of the model efforts for the four primary German River basins.  The following results illustrate the key results and implications (McDermott & Nilsen, 2012):

·       The “Base Volume” (i.e., Elbe 8.099) is the coefficient on the contemporaneous volume of electricity that denotes the short-run, instantaneous impact of a change in quantity on price.

·       The long-run multiplier is found by incorporating the lagged endogenous variables of the model and can be calculated for the Neckar River Basin as [(8.081-3.672-2,271)/(1-0.623-0.0168)] = 4.699.  Testing this figure reveals it to be statistically significant at the 1% level.  What this means is that a 1% increase in electricity volumes will lead to a 4.7% increase in price over the course of a full week.  This describes a very inelastic supply curve.

·       Looking at the effect of river temperatures for the Main River Basin reveals that there is a positive impact of electric prices once the 25⁰C threshold is breached (all four river basins have a positive impact).  A 1% increase in river temperatures above the 25⁰C mark will yield an increase in contemporaneous prices equal to 0.22%.  The equivalent long-run effect is 0.98%.  Thus a temperature rise from 25⁰C to 26⁰C would bring about an immediate price increase of approximately 9.14% over the next seven days.  These effects are all statistically significant at the 1% level.

·       As can be seem from Table 1, the four river basins have individual river level coefficients that are all negative and thus indicative of a higher electric price when river levels fall.  In the case of the Neckar River Basin, a 1% drop in river levels will lead to a 0.6% rise in contemporaneous prices, or a 1.8% rise in the long run.

In conclusion the authors point out the following that all U.S. water and energy managers should make note of (McDermott & Nilsen, 2012):

“We have argued that Germany serves as a good case study to investigate these issues, and have based our analysis of daily data taken over a period of seven years.  Having successfully controlled for various demand effects within a simultaneous equation framework, our results indicate that electricity prices are significantly affected by both falling river levels and higher river temperatures.  The magnitude of these relationships varies according to the exact specifications of the regression model at hand and we have explored several contemporaneous and dynamic settings.  Qualitatively, however, they all tell a very similar story:  electricity prices are driven higher by falling river levels and high river temperatures.  Under a fully contemporaneous setting, the electricity price is expected to rise by around one percent for every one percent that river levels fall.  The dynamic specification, on the other hand, suggests that the price will rise at about half the rate in the short-run, before increasing to approximately one and a half percent in the long-run.  With regards to river temperatures, the models imply that the price of electricity will increase by roughly one percent for every degree that temperatures rise above a 25⁰C threshold.  Incorporating the longer-run effects implied by a dynamic model shows that prices will rise by nearly four percent over the course of a week.  In addition to this slope effect, we test for a price discontinuity on either side of this 25⁰C threshold.  However, we do not find evidence of a marked price jump once the threshold is breached.  An explanation, which is consistent with our theoretical model and the surveyed literature, is that power plants reduce their output in stages rather than simply shutting down.  This allows them some additional scope for managing thermal pollution, although a decrease in output – and hence in price – cannot be fully avoided.”

Van Vliet, Vogele, and Rubbelke have also examined the impacts on electricity prices in the context of water constraints in Europe from climate change (T.H. van Vliet, Vugele, & Rubbelke, 2013).  As previously outlined, climate change is likely to impact electricity supply in terms of both water availability for hydropower generation and cooling water usage for thermoelectric power production.

The authors utilized simulations of daily river flow and water temperature projections using a physically based hydrological-water temperature modelling framework with climate model data for 2031-2060.  These projections for river flows and water temperatures were used in a thermoelectric power and hydropower production model to calculate impacts on power generating capacity. 

Provided below is a summary of the methodology and results of the research (T.H. van Vliet, Vugele, & Rubbelke, 2013):

·       The thermoelectric model calculates water demands of power plants based on their efficiency, installed capacity, cooling system type and the maximum allowed water temperature (increase).

·       The authors focused on 68 thermoelectric power plants in Europe.  Selection was based on the availability of information.

·       The authors quantified the impacts of replacing a particular cooling system type – (a.) replacement of all once-through recirculation (tower) cooling systems, and (b.) replacement of all once-through systems by recirculation cooling systems and replacement of all coal lignite and oil-fueled to gas-fired power plants.

·       The research focused on the changes associated with wholesale electricity prices, production, and electricity producer surplus.

·       A key assumption was that in any point in time, electricity supply must meet electricity demand.

·       Climate change scenarios illustrated a sharp difference in mean annual river flow between northern and southern Europe.

·       North of 52⁰N is projected to have river flow increases of 3-5% while south will have declines of 13-15%.

·       For example, Greece is projected to have declines of more than 20%.

·       Increase in mean water temperatures are largest (>1⁰C) in central Europe (e.g., Switzerland, Austria, Slovenia, Hungary, Slovakia) and south-eastern parts (e.g., Romania, Bulgaria, Croatia, Serbia).

·       A combination of strong increases in water temperatures and decline in low river flow is generally most critical for cooling water use.  These conditions are mainly projected for southern, central, and south-eastern European.

·       The largest declines in mean useable capacity under “baseline setting” are estimated for countries in southern and south-eastern Europe.

·       Replacement of cooling systems and changes in the sources of fuel lead to an overall reduction in the vulnerability of thermoelectric power plants to climate change.

·       The authors concluded that overall higher wholesale prices would be expected for most countries; because the limitations in water availability and exceeded water temperature limits mainly affect power plants with low production cost (e.g., hydroelectric and nuclear power plants).  Strongest increases in mean annual wholesale prices are projected for Slovenia (12-15%), Bulgaria (21-23%), and Romania (31-32%) for 2031-2060 relative to 1971-2000 for “baseline setting”.  Sweden and Norway are exceptions, because mean water availability is projected to increase in these countries, and consequently more electricity will be produced there by “low-cost” hydroelectric power plants, putting costlier power plants out of operations.

The authors offer the following conclusions (T.H. van Vliet, Vugele, & Rubbelke, 2013):

“Overall, more electricity will be traded with changes in power plant availabilities in Europe under future climate and changes in power plant stock.  Autonomous adaptation via the European electricity market provides opportunities to partly compensate for the loss of power generating capacity in one subsector or location.  However, considering the high shares of hydropower, coal-fuelled and nuclear-fuelled power plants in most European countries, the vulnerability to declines in summer river flow and increased water temperatures can be high.  Planned adaptation strategies are therefore highly recommended, especially in the southern, central and south-eastern parts of Europe, where overall largest impacts on thermoelectric and hydropower generating capacity are projected under climate change.  Considering the high investments costs, retrofitting or replacement of power plants night not be beneficial form the perspective of individual power plant operators, although the social benefits of adaptation could be substantial.”

An attempt was made to explore the findings outlined in the previous European studies in the context of the Texas electrical markets.  The Big Brown Power Plant was selected for review.  The Big Brown Power Plant is located in Fairfield, Texas (Luminant, 2014).  The fuel source is lignite from Texas coal fields and is supplemented by Powder River Basin coal.  The operating capacity is 1,150 MW (2005 net generation of 8,549,084 MWhr) – enough to power about 575,000 homes in normal conditions and 230,000 homes in periods of peak demand.  Unit #1 was constructed in 1971 and Unit #2 was constructed the following year.  The plant is owned by Luminant.

The Big Brown Power Plant utilizes once through cooling (Ross, 2012).  In 2005, the water use was 6,093 acre-feet.  Cooling water consumption was 2,703 acre-feet.  Cooling water comes from Fairfield Lake.  Water Rights Permit No. 2351 A was issued to Texas Power and Light Company (presently Luminant) on May 9, 1968 and authorized the construction of a dam to impound 50,600 acre-feet of water.  Of that total, 19,700 acre-feet of water could be diverted annually by pumping from the Trinity River.  The permit allowed annual usage not to exceed 14,150 acre-feet of water for cooling a steam-electric generation plant.

Fairfield Dam and appurtenant structures consist of a rolled-earthfilled embankment approximately 3,250 feet in length, with a maximum height of 77 feet and a crest elevation of 322.0 feet (Texas Water Development Board, 1999).  The service spillway is located at the north abutment and is a concrete chute with an ogee crest.  The crest is 60 feet in net length at 299.0 feet.  Two tainter gates, each 14 feet tall and 30 feet wide, control the service spillway.  The emergency spillway located to the south of the dam, is an earth trench through the natural ground.  The uncontrolled broad-crested weir is 500 feet in length at elevation 314.0 feet.  The minimal operating elevation for the intake to the power plant is 305.0 feet.

No information on lake levels or temperature could be obtained via online report searches or repeated e-mail requests to the Texas Water Development Board.   Research did come across a major fish fill that occurred August 25-26, 2010 – which corresponds to a period of extreme drought in Texas (Leschper, 2010).  Because the watershed for Lake Fairfield is small relative to the lake volume, make-up water is pumped from the Trinity River to maintain elevation.  Trinity River water is high in nutrients (i.e., all of the wastewater effluent generated in the Dallas and Fort Worth area is discharged into the Trinity River), which are further concentrated in Lake Fairfield due to evaporation and a lack of water discharged through the dam.  This high level of nutrients of contributes to high phytoplankton and fish production in Lake Fairfield but also contributes to dissolved oxygen depletion during cloudy weather.  An estimated 1,255,674 fish were killed, which was higher than previous years (914,189 in 2009 and 121,568 in 2008). 

The Lake Fairfield situation is a good example of the water-energy-environment nexus.  Consider the following from the article (Leschper, 2010):

“TPWD biologists began to unravel the ecological factors contributing to fish kills on Lake Fairfield in fall 2009.  By combining oxygen data from datasondes with solar radiation data from a local weather station, biologists were able to understand the mechanisms leading to repeated kills at Lake Fairfield.

In late August and September, water temperature and bacterial activity are still high but day length shortens incrementally, in power-plant reservoirs such as Fairfield, water temperature and day length can become out of phase and increase the probability of fish kills.

Similar fish kills have also been reported at other power-plant lakes such as Victor Braunig and Calaveras near San Antonio but are much lower magnitude than those at Lake Fairfield.”

Having reviewed water level and water quality issues for Lake Fairfield, a specific time period around August 20, 2010 can be reviewed for electricity pricing information.  Pricing information was obtained for the Electric Reliability Council of Texas (ERCOT) (Potomac Economics, 2011).  Consider the following information for this time period:

Price Hub	Delivery End Date	Average Price ($/MWh)	Daily Volume (MWh)
ERCOT Houston	August 19, 2010	64.31	3,200
ERCOT Houston	August 20, 2010	63.93	5,600
ERCOT Houston	August 23, 2010	77.11	14,400
ERCOT Houston	August 24, 2010	81.92	21,600
ERCOT Houston	August 25, 2010	45.44	6,400
ERCOT Houston	August 26, 2010	40.35	5,600
ERCOT Houston	August 27, 2010	39.65	4,000
ERCOT Houston	August 30, 2010	45.98	8,000
ERCOT Houston	August 31, 2010	47.42	4,800
ERCOT Houston	September 1, 2010	44.59	6,400

 

The data is inconclusive regarding the water-energy nexus and pricing, but one can see the extra ordinary increase in power demand at the same time the Lake Fairfield generation station was utilizing water from other sources (i.e., the Trinity River).  This period of high demand, water resource constraints, and declining water quality produced an environment suitable for a massive fish kill.

Conclusion

One of the most pressing problems associated with the water-energy nexus is the lack of cross-sector collaboration.  Neither group (i.e., water or the energy industrial sectors) fully understands or appreciates the others operational needs and constraints.  In addition, both groups have a lack of incentives to take risks, both have financing challenges, and both face huge regulatory and policy constraints.  Table 2 illustrates what productive discussion between the two sectors might look like.

Consider the following (The Johnson Foundation at Wingspread, 2013):

“At a fundamental level, there is a lack of understanding between sectors about their respective operational needs and constraints, as well as a lack of broad systems thinking about the interdependencies between them.  Furthermore, water and energy utilities often have different goals and reward structures that create conflicting interests.  Players in the water sector need to develop a better understanding of how the power sector is regulated and how the electrical grid is managed.  While the power sector is quite heterogeneous, there are national reliability standards developed and enforced by the North American Electric Reliability Corporation, according to which all electric utilities must design their systems.  The nonstandardized nature of small-scale energy generation projects at wastewater facilities, therefore, is one reason electric utilities find it challenging to incorporate distributed generation sources into their portfolios.  In addition, interconnection fees and approval processes, as well as net metering policies, present hurdles to connecting distributed generation from wastewater treatment plants to the electric grid.”

This environment of multiple hurdles to cross-sector collaboration and navigating toward the infrastructure of integrating water and electricity is also under pressure from climate change and extreme weather events.  The science of climate is rather clear, given that science is always a closed-looped system where new information is constantly feed into old assumptions.  The scientific community is projecting the following in the context of climate change (Friedrichs, 2014):

“At the end of the twenty-first century, average global temperatures are projected to be roughly 2-7⁰ C above preindustrial levels.  This can be decomposed into about 0.6⁰C of global warming from about 1750 to 1990, plus an additional 1.1-6.4⁰C in the period from 1990 to 2100.  Overall, it seems safe to say that the world must be prepared for global warming of at least 2⁰C, and probably more, above present temperatures by the end of the twenty-first century.”

The Water Research Foundation also paints a dim picture of the water-energy nexus in terms of water resources planning, climate change, and water supply reliability (Water Research Foundation, 2008):

·       For the North Hemisphere, climate models project a broad pattern of drying in the subtropics, including the Mediterranean Basin and the U.S. Southwest.  The models project wetter conditions north of about 50⁰ latitude, but for other temperate areas, projected changes in total precipitation and runoff remain inconsistent across models.

·       Higher temperatures will increase potential evaporation, which may diminish water availability.

·       Changes in water supply reliability will broadly mirror the changes in regional precipitation and runoff, although changes in seasonal runoff patterns and the intensity and frequency of precipitation events will also affect supply reliability.

·       Longer dry spells and heavier precipitation events appear likely in most temperate areas including all of Europe, and the contiguous states of the U.S.  The supply reliability effects of such changes will depend upon the capacity of surface and groundwater systems to capture and store water from the heavier precipitation events for subsequent supply augmentation.

·       The impacts of warmer temperatures on seasonal flow timing will be especially significant in areas that currently depend on melting mountain snowpacks for summer water supplies, such as Western North and South America.  Impacts on supply security will likely be greatest for water users depending on direct streamflows in small watersheds, who also have limited access to storage capacity, or alternative sources of supply.

In conclusion, this paper hopefully has pointed out society and organizations cannot separate water and energy issues and constraints.  The two issues have many different and complex levels of interdependencies.  Future problems, such as climate change, will have cascading impacts for both water and energy.  Looking into the future, it is important for the two industries to further develop a joint operational model that allows for increased understanding, visualization, and discussions by key stakeholders and policy makers.  Engineers, managers, and policy makers in both industries must become more comfortable with a new operational environment of imperfect choices.

Bibliography

Argonne National Laboratory. (2012). Impacts of Long-term Drought on Power Systems in the U.S. Southwest. Washington D.C.: U.S. Department of Energy.

Carney, B. (2010). Water Vulnerabilities for Existing Coal-Fired Power Plants. Washington, DC: National Energy Technology Laboratory.

Cayan, D. R. (2010). Future Dryness in the Southweat US and the Hydrology of the Early 21st Century Drought. PNAS, 21271-21276.

Colorado River Water Consultants. (2013). Study of Long-Term Augmentaion Options for the Water Supply of the Colorado River System. Las Vegas, NV: Southern Nevada Water Authority.

Fisher, J., & Ackerman, F. (2011, February). The Water-Energy Nexus in the Western States: Projections to 2100. Stockholm, Sweden: Stockholm Environment Institue.

Friedrichs, J. (2014). The Future is Not What It Used to Be: Climate Change and Energy Scarcity. New York: Random House.

Fulp, T. (2005, March/April). How Low Can It Go? Southwest Hydrology, pp. 16-18.

Hadian, S., & Madani, K. (2013). The Water Demand of Energy: Implications for Sustainability Energy Policy Development. Sustainability, 4675-4687.

Harto, C., & Yan, Y. (2011, December). Analysis of Drought Impacts on Electricity Production in the Western and Texas Interconnections of the United States. Oak Ridge, TN, USA: Argonne National Laboratory - Environmental Science Division.

Heard on the Street Column. (2014, February 18). California Power Market Gets Freeze-Dried. The Wall Street Journal. New York, New York, USA: The Wall Stree Journal.

Leschper, W. (2010, September 14). Lake Fairfield fish kill includes sustantial numbers of game fish. Retrieved from Examiner.com: www.examiner.com

Los Angeles Times . (2014, March 5). Water Conservation's Other Benefit: It's a Power Safer. Los Angeles Times Op-Ed, p. 33.

Luminant. (2014, March 19). Big Brown Power Plant and Mines. Luminant.

MacDonald, G. M. (2010). Water, Climate Change, and Sustainabilityin the Southwest. PNAS, 21256-21262.

Malewitz, J. (2014, March 9). Drought Hastens End of a Region's Hydropower Era. The New York Times, p. 29A.

McDermott, G. R., & Nilsen, O. A. (2012, September). Electricity Prices, River Temperatures and Cooling Water Scarcity. Bergen, Norway: IZA Discussion Paper No. 6842.

Potomac Economics, L. (2011). 2010 State of the Market Report for the ERCOT Wholesale Electricity Markets. Austin: ERCOT.

Ross, D. L. (2012). Water for Coal-Fired Power Generation in Texas: Current and Future Demands. Glean Rose: Glenrose Engineering, Inc.

Smedley, T. (2013, February 1). Energy for water may have greater impact on the nexus. The Guardian, p. 7.

T.H. van Vliet, M., Vugele, S., & Rubbelke, D. (2013). Water Constraints on European Power Supply Under Climate Change: Impacts on Electricity Prices. Environmental Research Letters IDP Publishing, 1-10.

Texas Comptroller of Public Accounts. (2012). The Impact of the 2011 Drought and Beyond. Austin: The State of Texas.

Texas Water Development Board. (1999). Volumetric Survey of Fairfield Lake. Austin: Texas Water Development Board.

The Johnson Foundation at Wingspread. (2013). Building Resilient Utilities: How Water and Electric Utilities Can Co-Create Their Futures. New York: The Johnson Foundation.

The World Bank. (2014, February 7). Securing Energy in a Water-Constrained World. Washington, DC: The World Bank.

Water Conservation & Technology Center. (2012). Viability and Impacts of Implementing Various Power Plant Cooling Technologies in Texas. College Station: Texas A&M University System.

Water Research Foundation. (2008, April 30). Technical Briefing Paper (7): Water Resources Planning: Climate Change and Water Supply Reliability. Washington, DC: Water Research Foundation.

Webber, M. E. (2008, August). Catch 22: Water vs. Energy. Scientific American, pp. 34-41.

Wikipedia. (2014, March 3). Thermal Pollution. Retrieved from Wikipedia: http://en.wikipedia.org/wiki/Thermal_pollution