Energy And

THE Environment

Wind and Water

 Energy And
THE Environment

Wind and Water

JOHN TABAK, Ph.D.

 For Rick Cardenas, an old friend, a good man.

WIND AND WATER

Copyright © 2009 by John Tabak, Ph.D.

All rights reserved. No part of this book may be reproduced or utilized in any form
or by any means, electronic or mechanical, including photocopying, recording, or by
any information storage or retrieval systems, without permission in writing from the
publisher. For information contact:

Facts On File, Inc.
An imprint of Infobase Publishing
132 West 31st Street
New York NY 10001

Library of Congress Cataloging-in-Publication Data

Tabak, John.
  Wind and water / John Tabak.
   p. cm. — (Energy and the environment)
  Includes bibliographical references and index.
  ISBN-13: 978-0-8160-7087-9
  ISBN-10: 0-8160-7087-3
  1. Renewable energy sources. I. Title.
  TJ808.T33 2009
  333.79′4—dc22 2008028247

Facts On File books are available at special discounts when purchased in bulk
quantities for businesses, associations, institutions, or sales promotions. Please call
our Special Sales Department in New York at (212) 967-8800 or (800) 322-8755.

You can find Facts On File on the World Wide Web at http://www.factsonfile.com

Text design by Erik Lindstrom

Illustrations by Accurate Art
Photo research by Elizabeth H. Oakes

Printed in the United States of America

Bang Hermitage 10 9 8 7 6 5 4 3 2 1

This book is printed on acid-free paper.

 Contents

Preface ix
Acknowledgments xiii
Introduction xiv

Part I  Hydroelectric Power 1

1 Waterpower: A Brief History 3
Designs from Antiquity 3
The Industrial Revolution and Michael Faraday 7
Creating Electricity Demand and Supply 12
Niagara Falls 16

2 Theory and Practice 19

Turbine Design 23
The Water Supply 28
Pumped Storage 30
Base Load versus Peak Load 33

3 Costs and Policies 37

The Costs of Hydropower 38
Global Warming 44
More about Environmental Costs 46
Methane Emissions and Hydropower 52
The Future of Hydropower 54
Part II  Electricity from the Oceans 57
4 Wave Power 59
Sea Snakes 62
Blow Holes 67
The Archimedes Wave Swing 72

5 Tidal Power 75
The French and Canadian Projects 77
Turbines without Dams 82

6 Heat Engines 88
The Theory of Heat Engines 88
Practical Applications 92
Ocean Thermal Energy Conversion Tests 94

7 The Role of Government in Promoting

New Technologies 99
The United States: Creating Supply and Demand 102
An Interview with Dr. Stan Bull on Research
  at the NREL 106
Denmark: Creating Supply, Demand,
  and Goodwill 111
Germany: Rapid Growth, Ambitious Goals 114

Part III Wind Power 117

8 Wind Power: A Brief History 119
Windmills 121
Wind Turbines 131
The Altamont Pass Wind Resource Area 136

9 The Nature of Wind Power 139

How Much Energy Is in the Wind? 142
Estimating Capacity 148
 Storing the Wind 154
Wind Power, Topography, and the Environment 154

10 Wind Energy: Economic and Public Policy

  Considerations 160
The Costs of Wind Power 162
The Role of Economic Class 165
The Future of Wind Power 170

Conclusion 173

Chronology 176
List of Acronyms 180
Glossary 181
Further Resources 184
Index 189
 Preface
N ations around the world already require staggering amounts
of energy for use in the transportation, manufacturing, heat-
ing and cooling, and electricity sectors, and energy requirements
continue to increase as more people adopt more energy-intensive
lifestyles. Meeting this ever-growing demand in a way that mini-
mizes environmental disruption is one of the central problems of
the 21st century. Proposed solutions are complex and fraught with
unintended consequences.
The six-volume Energy and the Environment set is intended to
provide an accessible and comprehensive examination of the his-
tory, technology, economics, science, and environmental and social
implications, including issues of environmental justice, associated
with the acquisition of energy and the production of power. Each
volume describes one or more sources of energy and the technology
needed to convert it to useful working energy. Considerable empha-

ix
  Wind and Water

sis is placed on the science on which the technology is based, the

limitations of each technology, the environmental implications of
its use, questions of availability and cost, and the way that govern-
ment policies and energy markets interact. All of these issues are
essential to understanding energy. Each volume also includes an
interview with a prominent person in the field addressed. Interview
topics range from the scientific to the highly personal, and reveal
additional and sometimes surprising facts and perspectives.
Nuclear Energy discusses the physics and technology of energy
production, reactor design, nuclear safety, the relationship between
commercial nuclear power and nuclear proliferation, and attempts
by the United States to resolve the problem of nuclear waste disposal.
It concludes by contrasting the nuclear policies of Germany, the
United States, and France. Harold Denton, former director of the
Office of Nuclear Reactor Regulation at the U.S. Nuclear Regulatory
Commission, is interviewed about the commercial nuclear industry
in the United States.
Biofuels describes the main fuels and the methods by which
they are produced as well as their uses in the transportation and
electricity-production sectors. It also describes the implications of
large-scale biofuel use on the environment and on the economy with
special consideration given to its effects on the price of food. The
small-scale use of biofuels—for example, biofuel use as a form of
recycling—are described in some detail, and the volume concludes
with a discussion of some of the effects that government policies have
had on the development of biofuel markets. This volume contains
an interview with economist Dr. Amani Elobeid, a widely respected
expert on ethanol, food security, trade policy, and the international
sugar markets. She shares her thoughts on ethanol markets and their
effects on the price of food.
Coal and Oil describes the history of these sources of energy.
The technology of coal and oil—that is, the mining of coal and the
drilling for oil as well as the processing of coal and the refining
of oil—are discussed in detail, as are the methods by which these
 Preface  xi

primary energy sources are converted into useful working energy.

Special attention is given to the environmental effects, both local
and global, associated with their use and the relationships that have
developed between governments and industries in the coal and oil
sectors. The volume contains an interview with Charlene Marshall,
member of the West Virginia House of Delegates and vice chair of
the Select Committee on Mine Safety, about some of the personal
costs of the nation’s dependence on coal.
Natural Gas and Hydrogen describes the technology and scale of
the infrastructure that have evolved to produce, transport, and con-
sume natural gas. It emphasizes the business of natural gas production
and the energy futures markets that have evolved as vehicles for both
speculation and risk management. Hydrogen, a fuel that continues to
attract a great deal of attention and research, is also described. The
book focuses on possible advantages to the adoption of hydrogen as
well as the barriers that have so far prevented large-scale fuel-switch-
ing. This volume contains an interview with Dr. Ray Boswell of the U.S.
Department of Energy’s National Energy Technology Laboratory about
his work in identifying and characterizing methane hydrate reserves,
certainly one of the most promising fields of energy research today.
Wind and Water describes conventional hydropower, now-conven-
tional wind power, and newer technologies (with less certain futures)
that are being introduced to harness the power of ocean currents, ocean
waves, and the temperature difference between the upper and lower
layers of the ocean. The strengths and limitations of each technology
are discussed at some length, as are mathematical models that describe
the maximum amount of energy that can be harnessed by such devices.
This volume contains an interview with Dr. Stan Bull, former associate
director for science and technology at the National Renewable Energy
Laboratory, in which he shares his views about how scientific research
is (or should be) managed, nurtured, and evaluated.
Solar and Geothermal Energy describes two of the least objec-
tionable means by which electricity is generated today. In addition
to describing the nature of solar and geothermal energy and the
xii  Wind and Water

processes by which these sources of energy can be harnessed, it de-

tails how they are used in practice to supply electricity to the power
markets. In particular, the reader is introduced to the difference
between base load and peak power and some of the practical differ-
ences between harnessing an intermittent energy source (solar) and
a source that can work virtually continuously (geothermal). Each
section also contains a discussion of some of the ways that govern-
mental policies have been used to encourage the growth of these
sectors of the energy markets. The interview in this volume is with
John Farison, director of Process Engineering for Calpine Corpora-
tion at the Geysers Geothermal Field, one of the world’s largest and
most productive geothermal facilities, about some of the challenges
of running and maintaining output at the facility.
Energy and the Environment is an accessible and comprehensive
introduction to the science, economics, technology, and environ-
mental and societal consequences of large-scale energy production
and consumption. Photographs, graphs, and line art accompany the
text. While each volume stands alone, the set can also be used as a
reference work in a multidisciplinary science curriculum.
 Acknowledgments
T he author extends special thanks to Stan Bull, Associate Di-
rector for Science and Technology for the National Renewable
Energy Laboratory, for sharing his time and considerable insights.
Also important to the preparation of this work were Ken McDon-
nell, Senior Media Relations Specialist with ISO New England;
Elizabeth Oakes, for her creativity in assembling the photography;
and Frank Darmstadt, executive editor.

xiii
 Introduction
E nergy is one of the fundamental problems of the 21st century,
although there is no lack of it. There are enormous reserves of
energy in the winds, the tides, and in the temperature difference
between the upper and lower regions of the oceans. It is an oft-
­repeated claim that each of these sources of energy is so abundant
that if the energy of any one of them were converted into electrical
energy, it would satisfy the electricity demands of the entire world
many times over. This claim is true, but the implication that one
could under any circumstance convert more than a tiny fraction of
any one of these energy sources into electricity is false. There is no
way to effect such a conversion—not now, not ever. Understanding
both the promise and the limitations of these energy sources is one
of the principal aims of this book.
All of the technologies in this book have benefited from sig-
nificant government subsidies. Conventional hydroelectric power

xiv
 Introduction  xv

projects with their enormous dams and their enormous reservoirs

cannot be built without significant government funding and other
forms of assistance. Once built, they are often operated in ways very
different from those of for-profit generating stations and often for
good reasons. As for the other technologies discussed in this book,
they also require significant subsidies to construct—and so far they
have also required significant subsidies to operate. Understanding
how these power producers benefit from government support, the
economics of operating these types of generating stations, and the
ways that they contribute to the power pool is an important part of
understanding their potential, and that is the other main objective
of Wind and Water.
Chapters 1 through 3 describe conventional hydroelectric power.
Measured by the size of its contribution, conventional hydroelectric
power is the most important technology discussed in this volume,
but it is a mature technology, highly refined and not subject to much
additional improvement. In most nations, most of the commercially
valuable sites have already been developed. Significant changes in
this sector are unlikely.
The second part of the book consists of chapters 4 through 7,
and describes wave energy converters, tidal mills, ocean thermal
energy converters, and tidal barrages. The first three of these tech-
nologies are rapidly evolving. They have, therefore, attracted a great
deal of media attention, but so far they have generated relatively
little power. Tidal barrages, an old idea, is being reexamined as
those nations with suitable sites reconsider the technology in light
of higher fossil fuel prices, increased demand for electricity, and the
desire to generate more electricity with fewer emissions. All these
technologies are heavily dependent on government support for
their construction and operation. They are, therefore, economically
uncompetitive. But given help, they may become competitive. Some
believe that if governments intervene aggressively enough on be-
half of these technologies, they can rapidly create markets for them.
xvi  Wind and Water

The role of government in developing markets for power-produc-

tion technologies is examined in chapter 7, with special reference
to wind power, the most mature of these power technologies. Pro-
wind policies in Denmark, Germany, and the United States are de-
scribed and compared. This chapter also contains an interview with
Dr. Stan Bull, associate director for science and technology for the
National Renewable Energy Laboratory, the nation’s premier site
for research into renewable energy, about the challenges involved
in establishing and maintaining high-quality government-funded
research programs.
The last part of the book, consisting of chapters 8 through 10,
describes the technology of wind turbines, the ways that wind tur-
bines can be harnessed to contribute to the power supply, and the
politics of wind power, especially the role of economic class in the
siting of wind farms. All of these issues have so far played an im-
portant role in the development of this rapidly growing segment of
the power market.
From an environmental viewpoint, all of the technologies de-
scribed in this volume are relatively benign. The environmental im-
pacts of even the most disruptive of them—namely, the very large
hydroelectric projects—are generally local in nature. In particular,
none of these technologies affect the global climate. But for a power-
production technology to be adopted, it is not enough that it not
pollute. It must also produce power at the right time and in suf-
ficient quantities to meet demand.
Wind and Water seeks to bring the engineering, economic,
environmental, and public policy aspects of these technologies to-
gether to provide an overview of their place in the power sector
and thereby contribute to the ongoing debate about the ways that
electricity should (or even can) be generated in the future.
 Part I

Hydroelectric Power
 1

Waterpower:
A Brief History
C onventional hydroelectric power, the technology that converts
the energy of moving water into electrical energy, depends upon
a complex union of ideas and technologies. Some of these ideas and
technologies were pioneered thousands of years ago; some are of
much more recent origin. What is certain is that no one person in-
vented the technology of hydroelectric power. Instead, hydropower
evolved in response to numerous and sometimes conflicting engi-
neering, economic, and societal requirements. To appreciate how
modern hydroelectric plants work, it helps to know something of
their history. That is the purpose of this chapter.

Designs from Antiquity

The earliest devices to convert the energy of moving water into work
were waterwheels. The ancient Greeks used a type of waterwheel to
grind grain. The earliest extant reference to these waterwheels is in


  Wind and Water

Waterwheel on the Orontes River, photographed between 1898 and 1914. The
size of the project indicates the importance that the people of the time at-
tached to this device. (Library of Congress)

a poem by the Greek poet, Antipater of Thesselonica (ca. 85 b.c.e.),

who wrote the following:

Cease grinding, ye women who toil at the mill; sleep late,

even if the crowing cocks announce the dawn. For Demeter
has ordered the Nymphs to perform the work of your hands,
and they, leaping down on the top of the wheel, turn its axle,
which with its revolving spokes, turns the heavy Nysarian
millstones. We taste again the joys of the primitive life,
learning to feast on the products of Demeter without labor.
 Waterpower: A Brief History  

It is believed that the type of waterwheel Antipater praises was

placed horizontally in the water with only part of the wheel in
the flowing stream. The vertical axle extended upward, where the
grinding device was attached to the axle. The water, pushing against
part of the wheel, provided the force necessary to do the grinding.
The ancient Romans, Chinese, and Japanese all used waterwheels
of various types. Some wheels were oriented vertically, some hori-
zontally; sometimes gears were used. The exact details of construc-
tion were determined by the speed with which the water flowed, the
volume of water pushing against the wheel per unit time, and the
technical sophistication of the designers. What all waterwheels have
in common is that they convert linear motion into rotary motion.
In particular, they use moving water to turn a shaft. This is critical
because a rotating shaft can drive any piece of machinery.
Waterwheel design evolved over time, but most of the early
records regarding their construction and use have been lost. It is
known, for example, that the Romans constructed an enormous
complex at Barbegal, located near Arles, France, of 16 seven-foot
(2 m) waterwheels. These waterwheels were built in pairs, and the
pairs were constructed at descending levels along a hillside. An aq-
ueduct brought water to the top of the complex. The water flowed
downward from the aqueduct turning each pair of waterwheels in
sequence. The turning wheels provided the power necessary to grind
grain. Estimates of the capacity of this enormous facility vary wide-
ly. Various authorities estimate that the Romans ground sufficient
grain at this one facility to provide food for a population consisting
of from 12,000 to 80,000 people. Whatever the number, it is certain
that its construction required a large commitment of resources, that
its maintenance involved numerous skilled individuals, and that it
served a large community. The facility was a valuable resource.
Waterwheels remained important throughout the Middle Ages.
The Domesday Book, the main source of information on 11th-centu-
ry England, reveals that there were more than 6,000 water-powered
  Wind and Water

mills in England when the book was compiled. All of these mills
were, apparently, used to grind grain.
But a waterwheel turns only when there is sufficient water flow-
ing past it. Seasonal variations in rainfall as well as local weather
conditions affect the volume of water flowing past any mill. When-
ever the water dries up, so does the machine’s output. One way to
introduce more predictability into watermill performance involves
coupling the watermill to a dam. The dam enables the mill owner to
store water when it is plentiful and the flow is abundant, and release
water whenever it is needed. While this does not guarantee that
there will always be sufficient water—a long drought can exhaust
the storage capacity of the reservoir—it does enable the waterwheel
owners to maintain production during periods of short-term water
scarcity.
In addition to providing water storage, dams often serve to
increase waterwheel efficiency. By raising the water level, one can
increase the force with which the water impinges on the blades of
the waterwheel, with the result that more work can be extracted
per unit of water. Ancient waterwheels (and modern water turbines)
work between water levels. If the water level upstream of the wheel
and the water downstream of the wheel are equal, no water will
flow; the wheel will not turn. The larger the difference in height
between the upstream and downstream levels, the faster the water
flows and the more work can be accomplished per unit volume of
water. Increasing this difference in height is one way that ancient
and modern designers extract more work per unit of water.
As with waterwheels, early records of dams are also scarce. It is,
however, known that dams have been constructed since the begin-
ning of recorded history and perhaps earlier. In particular, it seems
likely that dam technology is at least as old as waterwheel technol-
ogy. Dams were, for example, constructed in ancient Egypt. Egyp-
tian records describe a dam constructed across the Nile in 2900
b.c.e. That dam is gone, but another, built in Syria on the Orontes
 Waterpower: A Brief History  

River in 1300 b.c.e., remains in use. Early dams were constructed

for the purpose of storing water, which could later be used to fur-
nish drinking water or water for irrigation. The decision by design-
ers to couple one or more waterwheels to a dam was an important
step forward in the development of waterpower, but when this first
occurred is unknown.

The Industrial Revolution and

Michael Faraday
The Industrial Revolution is the term used to describe the Western
transformation from an agrarian economy to an industrial one.
Eighteenth-century Britain was the first to experience the Indus-
trial Revolution, but it spread rapidly—first to Belgium and France
and soon thereafter to the United States. In Britain coal provided
the energy needed to power the transformation. While coal was
important everywhere, in the United States waterpower was also an
important power source.
Much of the eastern United States enjoys annual rainfall levels
of about 40 inches (1 m) per year. The topography of the eastern
United States is such that water flows rapidly to the sea. This is an
area that is ideally suited to the practice of harnessing water to do
mechanical work. Lowell, Massachusetts, was the site of the first
successful large-scale development of waterpower for manufactur-
ing purposes. It began during the 1820s when Boston merchants be-
gan to build the nation’s first planned industrial city along the Paw-
tucket Falls section of the Merrimack River. Along the Falls and in
less than a mile (1.6 km), the Merrimack River descends 32 feet (9.6
m). To harness this drop in water level, city planners constructed
nearly six miles (10 km) of canals, and they constructed a dam at
the head of the falls to divert water through the canals. The flowing
water turned waterwheels and their more modern analogues, water
turbines, and the spinning shafts of these devices were connected
to the machinery in the textile mills. There was no electricity in
  Wind and Water

these early mills; this was a purely mechanical use of hydropower.

Mill machinery was driven by often-complicated systems of belts
and pulleys that were harnessed to the spinning shafts that were
attached to the turbines that were driven by the flowing waters of
the Merrimack River. This was not hydroelectric but rather hydro-
mechanical power. The mills’ machines could not be located farther
from the water than one could stretch a series of drive belts. De-
spite this severe limitation, the mills at Lowell produced enormous
wealth and were considered by the people of the era a tremendous
success.
Within 20 years from the time that work at Lowell had begun,
multistory mills lined the canals with a combined capacity of 10,000
horsepower. Lowell had gone from a tiny village to the second larg-
est city in the state. The Lowell experiment served as a model for
other industrial cities located elsewhere in the nation. Holyoke,
Massachusetts, Nashua, New Hampshire, Saco, Maine, Augusta,
Georgia, and Cohoes, New York, were some of the cities whose de-
velopment was patterned on that of Lowell. The power of water to
drive machinery had been demonstrated in Lowell as never before.
The next step in the development of hydroelectric power grew
out of the work of the British scientist Michael Faraday (1791–1867).
Faraday, who was originally trained as a bookbinder and who began
his scientific work in the field of chemistry, is, today, best remem-
bered for his discoveries about the interplay between electricity and
magnetism. Endowed with an almost numbing patience, a lively
imagination, and a profound intellect, Faraday was also an accom-
plished inventor, and during his decades-long series of experiments
with electrical and magnetic phenomena he created a remarkable
collection of devices to illustrate his scientific ideas and sometimes
to discover new ones.
Here is one of Faraday’s most important discoveries: By spin-
ning a copper disk between the poles of a magnet and placing one
 Waterpower: A Brief History  

A cutaway diagram of an early type of water turbine called a Fourneyron tur-

bine. Water flows downward along the shaft and is then directed outward by
stationary vanes at g. As the water flows away from the shaft, it pushes against
so-called buckets at g’. The buckets are attached to the rim of a wheel, caus-
ing the wheel and shaft assembly to spin. The shaft transmits power upward
where it is harnessed to do work.
10  Wind and Water

Michael Faraday. His generator

converted mechanical energy into
electrical energy. (University of Ab-
erdeen, Department of Physics)

end of a circuit in contact with the rim of the disk and the other end
in contact with its center, he discovered that he could create a steady
electrical current. This was the first electric generator. By turning
the shaft on which the disk was mounted, he produced an electrical
current for as long as he kept the shaft spinning. This was extremely
important because in contrast with the already-invented battery,
Faraday’s power source functioned indefinitely; it never ran out
of electricity; it never went dead. He had turned the generation of
electricity into a mechanical—as opposed to a chemical—problem.
He demonstrated that to produce a continuous supply of electricity,
one need only harness a continuous power source to the shaft of the
generator.
But Faraday was not satisfied. He tried replacing the disk with
wires and again with a metal sphere. He investigated different geo-
metrical configurations for his generator and attempted to compare
each to the other in order to determine its efficiency in producing
current. He knew that given sufficient time and funding he could
have built larger, more powerful generators, but he did not do so. His
 Waterpower: A Brief History  11

interests lay elsewhere. Faraday sought to discover broad physical

principles; he sought to reveal new laws of nature. In effect, Faraday
left the following two problems for future generations of engineers
to solve: (1) Increase the size and efficiency of the generator in order
to produce greater electrical power; and (2) Harness power sources
capable of steadily spinning the shafts of these ever more massive
generators. The engineers succeeded. They continue to succeed, and
the result is an increasingly abundant supply of electricity.
Although the technology has changed a great deal since Faraday
constructed the first generator, the concept has not. Broadly speak-
ing, a modern power plant generator, the piece of technology that
actually produces electricity, has two pieces, a rotor and a stator. The
cylindrical rotor fits inside the hollow (and stationary) stator the way

Schematic of Faraday’s generator

12  Wind and Water

an ink cartridge fits inside a pen, and then, as its name implies, the
rotor rotates. For as long as it rotates within the stator an electric
current emanates from the generator: The more powerful the source
used to spin the rotor, the larger and more powerful the generator
that can be used.
To be clear, generators convert mechanical energy into electrical
energy. They do not create energy. One cannot extract more electri-
cal energy from the generator than the amount of mechanical energy
that one uses to spin the rotor. In particular, the greater the “load,”
or electrical demand, placed on the generator, the more force is re-
quired to turn the rotor. These simple-sounding facts explain why
generators come in different sizes and produce differing amounts of
electricity. In the case of water turbines, for example, moving water
pushes directly against the blades of the turbine, which converts the
linear motion of the water into the rotary motion necessary to spin
the rotor. The harder the water pushes, the more electrical energy
can be generated.
Faraday’s invention made it possible, in theory, to harness the
spinning shafts of turbines and waterwheels to generators and pro-
duce electricity, but his work did not immediately lead to hydroelec-
tric power, because initially there was no demand for it. There were
no electrical appliances.

Creating Electricity Demand

and Supply
Sir Humphry Davy (1778–1829), British chemist and teacher of
Michael Faraday, was a prominent scientist in his own right and
received numerous honors during his life for his discoveries. One
discovery made by Davy was the arc lamp. He found that by placing
two pieces of carbon close together—he used charcoal—and passing
a current between them an arc of brilliant light formed. The light
emanates from the electric arc that forms across the gap between
the two pieces of carbon and from the ends of the pieces of car-
 Waterpower: A Brief History  13

bon. In 1807, Davy used charcoal sticks and an enormous battery

to produce the light. The invention had no practical value because
no practical source of inexpensive electricity yet existed, but Davy’s
discovery did inspire a number of engineers to begin the process
of improving the technology. Several important modifications were
made by a number of engineers over the next several decades with
the result that arc lighting was sometimes used to illuminate public
places on special occasions. (Arc lighting is too bright and too harsh
to use as indoor illumination.) The technology was not, however,
widely adopted, because batteries were simply too expensive to use
as a regular source of power, and no other practical source of power
was yet available.
Another source of light, the incandescent lightbulb, was in-
vented by the American inventor Thomas Edison (1847–1931) in
1879. To make money from his invention Edison had to overcome
several obstacles. In particular, he had to construct an electric
power infrastructure to generate electricity and transmit it to the
consumers who wanted to use his bulbs to illuminate their homes
and businesses. Edison’s incandescent lamp, which was well-suited
for indoor illumination, and the arc lamp, which was well-suited
for outdoor illumination, created a demand for electricity. Initially,
however, hydropower was poorly suited to supply this market.
To appreciate how the demand for electricity was met, keep in
mind that electricity can be delivered as direct current (DC) or al-
ternating current (AC). As their names imply, direct current flows
along a wire in one direction only, much as water flows within a
pipe. By contrast, alternating current regularly reverses direction.
A cycle of AC current consists of flow in one direction and then in
the other. Today, in North America it is standard practice to deliver
AC electricity to homes at 60 cycles per second. This is called the
frequency of the current. Edison, however, provided DC current,
the same type of current generated by batteries, to his customers,
and at first Edison’s choice seemed reasonable. There were only a
14  Wind and Water

few customers, and the generating stations were built near where
the demand was. The problem with DC current—a problem that has
since been overcome—was that it was poorly suited for transmission
over longer distances. The losses were too great—that is, as the elec-
tricity traveled along the power line some of the electrical energy
was converted into thermal energy due to the resistance of the wire
to the flow of electricity. The farther the customer was located from
the generator, the less electrical power he or she received. Transmit-
ting DC current using the technology of Edison’s time was akin
to sending water through leaky pipes. The problem was a serious
one, because most major sources of hydroelectric power—Niagara
Falls is the most famous example—were located too far from most
potential consumers to supply them with DC electricity using the
technology available at the time. At this point in the history of hy-
droelectric power, there were electrical appliances, which created a
demand for electricity, and simple generators and turbines, which
provided a supply. But there was not yet the technology available to
connect electricity suppliers with electricity consumers if they were
located more than a few miles apart.
While Edison was busy promoting the use of DC current, an-
other approach to power generation and transmission was being
proposed. This second approach used AC current. The backers of
AC proposed a somewhat more complicated method of distributing
power that depended on devices called transformers. Electrical trans-
formers change the voltage of AC current. One feeds electricity into
one side of the transformer at a given voltage and, depending on how
the transformer is built, the AC current emerges from the other side
with the same frequency but with a voltage that has a different but
very specific relationship to the original voltage. One can construct
transformers that increase the voltage by a factor of 2, 10, 100, or any
other ratio. These are called step-up transformers. Similarly, one can
use transformers to lower the voltage by any predetermined factor.
These voltage-lowering transformers are called step-down transform-
 Waterpower: A Brief History  15

ers. This is important because when AC electricity is transmitted

along electrical lines at high voltages the losses that plagued Edison’s
low voltage DC system are very much reduced—the higher the volt-
age at which the electricity is transmitted, the lower the losses—and
when the electricity nears its destination, the voltage can be lowered
by passing the electricity through a step-down transformer. This was
the technology that would make it possible to connect hydroelectric
power plants with consumers located in distant cities.
More than anyone else, the American engineer and businessman
George Westinghouse (1846–1914) was responsible for the technol-
ogy used to bridge U.S. consumers with hydroelectric suppliers. He
was not the first person with the idea of using AC power. The idea
was pioneered in Europe. Westinghouse had, in fact, bought the
American rights to patents that had already been obtained in Eu-
rope by the French and English team of Lucien Gaulard and John
D. Gibbs. Westinghouse improved on the technology, but he was far
from the originator of it.
What made the technology championed by Westinghouse so
important from the outset is that most significant potential sources
of hydroelectric power were not located near major markets. This
was especially true in the western regions of the United States.
There, rapidly flowing streams are often located in the mountains,
but the population centers of the time were largely confined to the
coast. Portland, Oregon, for example, was one of the earliest cit-
ies to receive electric service. In 1889, the Willamette Falls Electric
Company began operation of a new facility along the Willamette
River in Oregon City. The AC electricity produced in Oregon City
was sent to Portland, located 13 miles (21 km) away, along a 4,000-
volt transmission line. In Portland it was stepped down to 50 volts.
Again, it was Westinghouse’s AC technology that provided the
bridge between U.S. consumers and producers of electricity.

(continued on page 18)

 16  Wind and Water

Niagara Falls

T oday most people know Niagara Falls as a tourist attraction, but

during the 19th century, when an energy-poor United States was
seeking the sources of energy it would need to develop, Niagara Falls
was perceived primarily as a potential energy source. Attempts to begin
utilizing the power of falling water at Niagara began in the 1820s when
two canals were built along the rapids that exist just above the falls.
The energy of the water that flowed through these canals was used to
power mills, but utilizing the main source of energy, the 180-foot (55-m)
drop that occurs at the falls themselves, was beyond the technology of
the times.
Nineteenth-century development at Niagara took two paths. Canal
builders continued to expand the canal system, first along the rapids and
then below the falls along Niagara Gorge, which was soon lined with an
unbroken row of mills. But the canal systems did not tap the real power of
the falls themselves. Instead, they made use of the relatively small drop in
height along the river immediately before and immediately after the falls.
By the 1880s, the turbines and generators needed to produce the power
and many of the electrical appliances that would consume it all existed.
In 1886, the Niagara River Hydraulic Tunnel Power and Sewer Company
was organized to dig deep tunnels that would divert water to the nearby
city of Niagara Falls. The tunnels would produce hydraulic heads of 79 to
125 feet (24.1 to 38.1 m). (The hydraulic head is defined as the difference
in height between the surface of the water upstream and the surface of
the water downstream.)
Financial and legal problems prevented the success of the plan, and
the Cataract Construction Company purchased Niagara River Hydraulic.
Cataract Construction would be responsible for the final tunneling. But
this was a project designed to produce profit as well as power. The cen-
tral financial barrier to the project was that the city of Niagara Falls, with
a population of only 20,000, was simply too small a market for such a
large and expensive-to-develop power source. The nearest big market
was Buffalo, New York. But Buffalo was located 26 miles (42 km) away.
Bridging such a long distance was a new type of technical challenge.
 Waterpower: A Brief History  17

Cataract solicited ideas for power transmission schemes from many

sources. Thomas Edison, America’s premier inventor, was one of the first
to be consulted by Cataract, and, not surprisingly, he lobbied for install-
ing a DC transmission line, which was the technology in which he had in-
vested. The difficulties of transmitting DC power 26 miles were, however,
apparent to everyone concerned. Others advocated using compressed
air to transmit power. Compressed-air technology was receiving some
attention in Europe at the time.
The technology that eventually found favor was the one championed
by George Westinghouse: Electricity would be generated at one volt-
age, the voltage would be increased (to 11,000 volts) for transmission to
Buffalo and then stepped-down to 110 volts for small motors, arc lamps,
and incandescent lighting, 100 or 250 volts for streetcar lines, 240 to 2,000
volts for various industrial applications, and so on. In 1895, the facility at
Niagara began serving local customers. In 1896, the system operators
began transmitting power to Buffalo. The generating station installed at
(continues)

Niagara Falls, mills on the American shore, ca. 1900. A century ago one of
America’s most famous tourist attractions was an industrial site. (Library
of Congress)
 18  Wind and Water

(continued)
Niagara Falls was a financial success. It not only met demand, it created
it. Soon energy-hungry industries—two famous examples of which are
ALCOA, then known as the Pittsburgh Reduction Company, and Union
Carbide—established large facilities in the vicinity of Niagara Falls to take
advantage of the power produced there. The Niagara Falls project proved
the viability of large-scale hydroelectric generation and large-scale elec-
tricity transmission, and set the pattern for the many large hydroelectric
projects that followed.

(continued from page 15)

By the beginning of the 20th century, all the necessary subsys-
tems for the production, transmission, and consumption of hydro-
electric power were in place. The hydroelectric power production
and distribution systems of the time were, by today’s standards,
extremely primitive, but they had, nevertheless, been developed to
a high enough degree to make large-scale projects practicable. Nu-
merous innovations were on the way, but the hydropower resources
of the United States, Canada, and other countries with a sufficient
technological base were now open for development. Electricity de-
mand surged in the United States during the early decades of the
20th century as did production capacity, and for a while, hydro-
electric power kept pace. During the 1920s, about 40 percent of the
nation’s electrical needs were met with hydropower.
 2

Theory and Practice

H ydroelectric power occupies a special niche in the electricity
markets. That niche has evolved as the electricity markets have
evolved. The purpose of this chapter is to understand some of the
fundamental principles that govern the operation of hydroelectric
generating stations and how these principles have affected the role
of hydroelectric plants in the electric power markets.
The purpose of a hydroelectric facility is to convert the power
of moving water into electrical power. Power is defined as energy
per unit time, so statements about power are statements about the
rate at which energy is supplied. Plant efficiency is the relationship
between the power provided to the plant (in the form of moving
water) versus the power manufactured by the plant (in the form of
electricity). When thinking about “power in” (in the form of water)
versus “power out” (in the form of electricity), the following very im-
portant question arises: Given the height of the water column above

19
20  Wind and Water

Cutaway view of a hydroelectric plant

the turbine—that is, given the hydraulic head—and the amount of

water flowing past the turbine per unit time, what is the most elec-
trical power that a hydroelectric facility can possibly generate? It
is an important question, because the answer places limitations on
what can be obtained, even in theory, from a hydroelectric station.
When discussing the maximum amount of electrical power that
can be obtained from a stream of flowing water, it is helpful to think
of the power of the flowing water as the input to the hydroelectric
station and the electrical power as the output. The plant converts
some of the input into output, but some input is lost to friction and
other inefficiencies. Or to put it still another way: The power of
flowing water is the raw material on which the turbine-generator
system acts; the electricity is the finished product. And because the
hydroelectric station only converts one form of energy into another,
the amount of power produced by a generating station cannot ex-
 Theory and Practice  21

ceed the amount of power supplied. (This is a consequence of the

statement that energy cannot be created.)
For a hydroelectric station, the power of flowing water is most
easily expressed in terms of an algebraic equation:

P = qha (2.1)

where the letter P represents power, q is the volume of water flow-

ing through the facility per unit time, h is the hydraulic head, and
a is a constant, the value of which depends on the units in which
P, q, and h are expressed. If P is expressed in watts, q is expressed
in cubic meters per second, and h is expressed in meters, then the
scaling factor a is approximately 9,800 newtons per meter cubed.
(Electrical output is usually not expressed in terms of horsepower,
but to find the output of a turbine in horsepower, multiply its output
expressed in kilowatts by 1.34.)
As one might expect, the actual amount of power produced by
a hydroelectric station is always less than the maximum possible
amount, but it is proportional to P. If Pact is the actual power output
of a hydroelectric power station, then

Pact = eqha

where e is the efficiency of the conversion process. (The exact value

of e depends on the design of the station, the way that the equip-
ment is maintained, and so forth. In all cases, e is a number between
0 and 1, and the closer e is to 1 the more efficient the station is.)
Because P and Pact are proportional, and because the formula for P
is a little simpler than the formula for Pact, the following discussion
will use equation (2.1), the equation for P, but the discussion can
also be repeated word-for-word in discussing the equation for Pact.
Equation (2.1) reveals some very important facts about hydro-
electric power. To generate a given amount of electrical power one
can use a small amount of water provided the hydraulic head is high
enough. Conversely, one can generate the same amount of power
22  Wind and Water

with a low hydraulic head provided a large enough volume of water

flows through the turbine. Equation (2.1) even shows how the two
quantities must vary in order that the power remains constant: h
must vary inversely with q. It is not, therefore, the hydraulic head or
the volume of water that matters; it is their product.
Equation (2.1) explains why a tall dam with, say, twice the
hydraulic head of a shorter dam, will generate twice the power of
the shorter dam when operating at the same volumetric flow rate.
Taller dams make better use of the available water. Of course, one
can also release more water through a shorter dam to achieve the
same power output, but there are physical limits on the amount of
water that can be released. One cannot, for example, release water
through a dam at a rate that is higher than the rate at which the
water is replenished from upstream—at least not for very long. On
the other hand, taller dams tend to cost much more to build than
shorter ones. There is, therefore, a balance to be struck between the
initial construction costs and a station’s expected power output.
(Environmental concerns, which are described in chapter 3, further
limit the rates at which water can be released.)
The accompanying diagram (on page 23) is a graph of a curve
called an isoquant, a line of constant power. The horizontal axis
represents the height of the hydraulic head, and the vertical axis
represents the volumetric flow rate so the first coordinate of a point
on the isoquant represents the value of h, and the second coordi-
nate represents q. (The value a appearing in equation (2.1) is just a
scaling factor. It can be ignored—or set equal to 1—for purposes
of this discussion. The important characteristics of the curve do
not depend on the value of a.) The product of the coordinates of
any point on a given isoquant equals the product of the coordi-
nates of any other point on the same isoquant. Each isoquant is,
therefore, the set of all values of h and q such that their product
equals a given power output. The isoquant illustrates the fact that
to produce a given amount of power with a low hydraulic head, one
 Theory and Practice  23

Isoquant curve, a curve representing the set of conditions—hydraulic head

and flow rate—that together yield a given power output

must release a very large volume of water. To put it another way, the
isoquant is a graphical demonstration that, all other things being
equal, taller dams are simply more efficient than shorter ones be-
cause they generate more power per unit of water. These are facts of
nature. No technology can overcome them. Equation (2.1) provides
a framework for evaluating the potential of a particular hydroelec-
tric power project.

Turbine Design
There was a time during the second half of the 19th century and
the early years of the 20th century when many would-be inventors
devoted considerable effort to designing water turbines. This was
an era when the principles of water turbine design were not entirely
understood; there was plenty of room for additional innovation,
and large sums of money were being spent developing new water-
power sites. Many designs were proposed, and a few found favor
with those responsible for designing power stations. In retrospect,
most turbine designs fall into one of two categories: Some work best
24  Wind and Water

The remains of a Fourneyron turbine. These turbines played an important part

in the industrialization of the United States. (French River Land Company)

when operating under high hydraulic heads, and some work best
when operating under low hydraulic heads. The higher head tur-
bines are called impulse turbines, and the lower head turbines are
called reaction turbines.
Impulse turbines depend largely on the percussive force of wa-
ter flowing under a high hydraulic head. In hydropower jargon, a
high hydraulic head is usually taken to be in excess of 1,000 feet
(300 m). Impulse turbines capture the water with what designers
call buckets. A specially designed nozzle directs the water, which
shoots out under very high pressure, at the turbine’s buckets. The jet
of water strikes the buckets and spins the turbine. The basic design
for the first successful impulse turbine was patented in 1889 by the
American engineer and inventor Lester Allen Pelton (1829–1908). It
is called the Pelton turbine.
 Theory and Practice  25

The story is that sometime during the 1870s Pelton was watch-
ing a water turbine spin as water struck the buckets that were placed
along its rim. These “buckets” would have been rounded and shal-
low and looked more like cereal bowls. They were aligned so that
the stream of high-velocity water driving the turbine struck each
bucket in the middle. The force of impact produced a lot of splash-
ing. As Pelton watched, the turbine became misaligned, and the wa-
ter began striking the edge of each bucket. The splashing stopped,
and instead the water was directed around the curve of each bucket
to shoot off the other side of the bucket in a coherent stream. The
buckets had reversed the direction of flow of the water. The effect
was an increase in turbine speed. Pelton turbines make use of this
observation.
The Pelton turbine has been somewhat modified over the years
and today has an efficiency of slightly more than 90 percent—that
is, it converts a little more than 90 percent of the power in a stream
of water. These turbines look somewhat like steel waterwheels.
The buckets have a cross section that looks like a rounded “W,”
and the water is directed toward the wedge in the middle. The
wedge splits the stream of water, and the curved surfaces direct
the stream backward. As the water emerges from the sides of the
bucket it propels the bucket and spins the turbine. This kind of
design depends on a high velocity stream of water, a fact that ex-
plains why Pelton turbines are usually reserved for facilities with
high hydraulic heads.
Sometimes engineers cannot generate a high hydraulic head
with which to work. Designers of hydroelectric facilities are con-
strained by the topography of the land. There may, for example, be a
large river flowing through a flat landscape. In this case, it is point-
less to construct a tall dam, because it will not constrain the water
upstream of the dam. Blocking the flow of the river would only
cause the water to flow around the dam. If the river flows through a
shallow valley, then a tall dam would fill the valley, possibly putting
26  Wind and Water

a large area of land underwater, but it could not create a deep reser-
voir behind the dam since the valley will constrain the water only
until the level of the water reaches the lowest point along the valley
wall. Consequently, it is sometimes necessary to also use turbines
that are designed to operate under lower hydraulic heads than those
characteristic of Pelton turbines. These low- and medium-head tur-
bines are called reaction turbines.
There are several very different designs for reaction turbines.
All of them are placed in chambers that are completely filled with
water. As the water flows through the chamber it creates a pressure
difference; the pressure is much higher on the upstream side of
each blade than on the downstream side, causing them to spin.
The earliest reaction turbine was created by the French engineer
Benoît Fourneyron (1802–67) and is called a Fourneyron turbine.
The Fourneyron turbine is mounted on a vertical axis. As the wa-
ter flows down through the chamber it encounters a set of curved
blades that do not spin. These fixed blades direct the water outward
against a runner—a round wheel—upon which a second set of
blades are mounted. The runner blades are curved in the direction
opposite that of the fixed blades. The force exerted by the water
on the second set of blades causes the runner to spin. Fourneyron
turbines were an important step forward. Fourneyron himself cre-
ated turbines that operated at 80 percent efficiency, a tremendous
technical breakthrough for the time. His most famous turbine
weighed only 40 pounds (13 kg), had a runner that was only one
foot (30 cm) across, and produced 60 horsepower (45 kW). These
turbines were used to power a great deal of industry in Europe,
and they also found favor in the textile mills of the eastern United
States. In 1895, Fourneyron turbines were installed at the facility
at Niagara Falls.
Not surprisingly, engineers have developed a number of other
reaction turbine designs since the first half of the 19th century
when Fourneyron created the first such turbine. They differ in the
 Theory and Practice  27

Glen Canyon power

plant generators (U.S.
Department of the Interior,
­Bureau of Reclamation)

way in which the water flows through the turbine. Fourneyron’s

design directs the water away from the axis of symmetry of the
turbine, while other designs, most notably the Francis turbine, di-
rect the water toward the axis of symmetry, and still other designs
are a combination of the two. Designs changed as engineers kept
searching for better solutions to perceived shortcomings of earlier
designs. (There are even special applications such as the pumped
storage facilities, which are described in the sidebar accompanying
this chapter, where a Francis turbine rather than an impulse turbine
is operated under a high hydraulic head because the Francis turbine
can be operated in reverse as a pump, an application for which the
impulse turbine is ill-suited.)
28  Wind and Water

In North America by the 1930s, most sites for large-scale hy-

droelectric development were identified and many were under de-
velopment. In addition, numerous small-scale sites had also been
developed. At the time, these smaller sites offered an economical
method of generating mechanical or electrical power.
From 1945 until the early 1970s, many of the smaller sites were
abandoned because fossil fuels were so cheap that small-scale hy-
droelectric power was no longer competitive. In New England, for
example, the ruins of small-scale hydroelectric and hydromechani-
cal facilities can still be found along many streams and small riv-
ers. After the energy crisis of the early 1970s, interest in small-scale
hydropower facilities began to increase, and now a modest number
of sites are developed each year. But because these sites have neither
a large hydraulic head nor a large volumetric flow rate, their total
contribution would be relatively small even if they were all devel-
oped. In some areas, however, they make a welcome contribution to
the local economy.

The Water Supply

Although most large hydroelectric facilities depend upon dams,
most dams are not constructed for the purpose of generating elec-
tricity. Often dams are constructed to create a reservoir of drinking
water, or as a source of water for irrigation, or for flood control, or
for recreation, or to maintain the navigability of waterways, or a
combination of these reasons. Dams used at hydroelectric power
stations constitute only a fraction of the total number of dams. And
not every hydroelectric facility depends on a dam. Niagara Falls is an
example of a large-scale hydroelectric station that operates without
a dam. Nevertheless, dams are extremely important components of
most large-scale hydroelectric facilities. And bigger dams are better
than smaller ones, because with respect to power generation the
more water that can be stored behind the dam, the more reliably
and the more efficiently the station will be able to generate power.
 Theory and Practice  29

Storing a large supply of water increases the reliability of a hydro-

electric facility because a large store of water acts as a buffer in the
face of fluctuations in supply. Depending on the location of a particu-
lar dam, the rate at which water flows into the reservoir created by the
dam may undergo regular seasonal fluctuations, or it may vary spo-
radically and unpredictably due, for example, to drought. The larger
the quantity of water behind the dam, the longer the hydroelectric
facility can operate in the face of a shortage in the rate of supply.
Unusually high flow rates can also interfere with power gen-
eration at hydroelectric facilities. Floodwaters often transport large
amounts of debris, which may require the station’s operators to close
the penstocks, those tubes that connect the turbines with the water
upstream of the dam, in order to protect the turbines from dam-
age due to debris. A large volume of water in the reservoir serves
to dampen the effects of short-term surges just as it dampens the
effects of short-term deficits.
Efficiency in the use of the water resource is enhanced provided
the dam creates a reservoir with a high hydraulic head. As previ-
ously mentioned in the discussion of equation (2.1), the higher the
hydraulic head, the more power can be extracted from each unit
of water flowing through the penstocks. But even when it is not
possible to construct a facility with a high hydraulic head, it may
still be possible, depending on the topography of the site where the
dam is located, to impound a large quantity of water. A large water
supply will still work to the operators’ advantage even without a
high hydraulic head, because a large supply of water enables the
operators to release large volumes of water without immediately
exhausting the supply. Again, as equation (2.1) shows, by doubling
the volume of water flowing through the turbine per unit time, the
power output can double as well. To be sure, releasing large volumes
of water will cause the water level behind the dam to fall—that is,
it will cause the hydraulic head to decrease—and a reduction in the
(continued on page 32)
 30  Wind and Water

Pumped Storage

P umped storage technology illustrates all of the ideas described so far

in this chapter. As will be seen, it is a technology that consumes more
energy than it generates, but it has proven very useful in enabling power
producers to meet demand.
To build a pumped storage facility, a power producer generally re-
quires something along the lines of a small mountain. A reservoir is built
at the top of the mountain, and a second reservoir is built at the base of
the mountain. They are connected by a penstock, a conduit that allows
the water in the upper reservoir to empty into the lower reservoir. Near
the base of the penstock, turbines are installed to harness the power of
the falling water. These are specially designed turbines that can oper-
ate in reverse, and when they are operated in reverse they function as
pumps. As the water flows down from the upper reservoir to the lower
one, some of the energy of falling water is converted into electricity.
Later, the turbines are used to pump the water from the lower reservoir
back up the mountain into the upper reservoir. More energy is always
needed to pump the water back to the upper reservoir than is recov-
ered by allowing the water to flow down to the lower reservoir—that
is, pumped storage facilities always consume more energy than they
produce—so, one might ask, what is the point of building this kind of
power consuming machine?
Because electricity cannot be stored in large quantities for later use, it
must be produced upon demand. Consequently, timing matters. Pumped
storage facilities are valuable because they enable the producer to shift
power production, to times when demand is high from times when
demand is low. Here is how it works: At night, utilities generally depend
upon coal plants and nuclear plants to meet demand because coal and
nuclear plants produce very large amounts of power and work most ef-
ficiently when operated at a steady output for prolonged periods of time.
In fact, at night these plants can easily produce surplus power—that is,
they can produce more electricity than there is demand for it. This surplus
 Theory and Practice  31

power can then be used to “prime” the pumped storage facility. Surplus
electricity powers the turbines, which are used to raise the water from the
lower reservoir to the upper one. The water remains in the upper reservoir
until it is needed. During the day, when demand is high and the coal and
nuclear plants cannot meet the entire demand for electricity, the water
is released, and the turbines convert some of the energy of the flowing
water back into electricity.
Pumped storage facilities are profitable because they enable the
power producer to manufacture power when demand and price are high-
est. In a market with widely fluctuating levels of demand and price, it is
often more important to produce electricity at the right time than it is to
be a net producer of electrical power.

Tennessee Valley Authority pump storage facility, Raccoon Mountain (TVA)

32  Wind and Water

(continued from page 29)

hydraulic head reduces the amount of power obtained from each
unit of water. But this reduction will happen slowly, provided the
amount of water impounded behind the dam is large enough. As
demand lessens, the volume of water flowing through the turbines
can be reduced. The water supply will replenish itself, and the pro-
cess can be repeated during the next demand cycle.
Finally, storing water is, in a sense, equivalent to storing elec-
tricity, because water can be converted into electricity by opening
the penstocks. Unlike water, electricity cannot be stored in large
quantities. The inability to store electricity places a fundamental
constraint on the way that power must be produced: Electrical
power must be produced simultaneously with demand. The ability
to adjust the rate at which power is produced (by opening and clos-
ing the penstock gates) is one of the most important advantages that
hydroelectric facilities have over other forms of power production.
Simple adjustments to the flow rate enable hydroelectric facility op-
erators to respond to changes in demand almost immediately. Of all
the methods for generating electrical power, hydroelectric facilities
are the most flexible in this regard.
Although engineers became interested in building huge dams
for purposes of power production only a century or so ago, attempts
to build large dams began much further back in history. The essen-
tial challenge of designing a large dam is that large volumes of water
exert enormous forces. For a long time, there was no engineering
theory that could guide those charged with building these huge
structures. Despite this, some early dams display some of the same
ideas that one finds in modern dams. The Ma´rib Dam, located in
present day Yemen, was built more than 2,000 years ago, and at
both ends of the dam the designers included spillways, architec-
tural features that enabled excess water to flow away from the dam.
(Water that flows over the top tends to erode—and so weaken—the
base of the dam.) Spillways are features of all large modern dams.
 Theory and Practice  33

And early 14th-century Persians built the Keber Dam, an early arch
dam. Arch dams are curved away from the upstream water so that
the pressure of the water against the dam is transferred along the
dam to its ends. By using an arch the need for enormous quanti-
ties of material to hold back the upstream water supply is greatly
reduced without sacrificing strength.
Civil engineering, that branch of engineering of most relevance
to the construction of dams, grew in large measure out of the work
of British engineer and scientist William John Macquorn Rankine
(1820–72), whose pioneering studies into the mechanics of soils and
other materials formed the basis of a deeper understanding of how
dams are able to withstand the tremendous forces exerted on them.
Today, engineers have a great deal of insight into the construction
and operation of several different types of dams, and as a conse-
quence it has been possible to build ever larger dams with ever more
exacting levels of performance.
But as dam-building technology has evolved, the number of
undeveloped sites suitable for large-scale hydroelectric power sta-
tions has continued to decrease. While hydroelectric power plants
will retain an important role in the energy sector of many countries
for the foreseeable future, the amount of electricity generated by
hydroelectric plants is not expected to increase very much because
most of the sites with the potential for generating large amounts of
hydroelectric power have already been developed.

Base Load versus Peak Load

Electricity demand fluctuates, and the demand for electricity at any
given hour of the day is only partly predictable. Power producers
know, for example, that on hot days people will turn on their air
conditioners—air conditioning is a very energy-intensive technol-
ogy—and on those days demand will surge. What they do not know
far in advance is which days will be hot. In addition to unpredict-
able fluctuations in electricity demand, there are many fluctuations
34  Wind and Water

Hoover Dam control room. A small number of people controls one of the
­nation’s great hydroelectric power plants. (U.S. Department of the Interior,
Bureau of Reclamation)

in demand that are highly predictable. Demand goes up during the

day and down at night; during the day demand is higher at 3:00
p.m. than it is at 9:00 a.m.; demand is higher during the day from
Monday to Friday than it is during the day on Saturday and Sunday;
demand is higher during the summer than it is during the spring,
and averaged over several years the total demand for electricity rises
steadily. Enormous data sets have been collected for every major
U.S. electricity market that make these statements precise.
Underlying these fluctuations in demand is a minimum and
highly predictable power requirement that must always be met.
Hospitals, for example, consume electricity continuously as do
some manufacturing concerns. There are numerous other exam-
ples. The electricity required to meet the sum of these minimum
power demands is called the base load. Base load power plants are
 Theory and Practice  35

those plants that furnish this electricity. Power plants are not dis-
tinguished by the type of electricity they produce—the characteris-
tics of the electricity supplied to the grid are the same for all types
of power producers—rather, base load power plants are those that
furnish an essentially steady flow of electrical power for prolonged
periods of time. As a general rule, base load power plants shut down
only for maintenance and repairs.
In the United States hydroelectric power plants once provided
a large fraction of the nation’s base load power. Some hydroelectric
power plants are still used to provide base load, but today most base
load power is provided by coal and nuclear plants. Coal and nuclear
plants are large power producers and work best when operated at
relatively steady power levels for long periods of time. By contrast,
hydroelectric stations are more flexible and can be operated so as
to respond to fluctuating levels of demand. Electricity produced to
meet demand fluctuation above the base load requirement is called
peak load power. (Production is sometimes further divided into in-
termediate load and peak load, but in this presentation as in many
others, it is sufficient to call all power in excess of the base load
demand peak power.)
Because electricity cannot be stored, electricity production
must vary simultaneously with demand. Matching supply to de-
mand is a difficult technical problem. In a modern market, de-
mand is measured every few seconds, and as it fluctuates, supply
must be adjusted accordingly. Matching supply with demand is
further complicated by the fact that particular energy technolo-
gies may not be available when they are needed. Solar and wind
operate independently of demand. The Sun shines and the wind
blows (or not) without regard to the requirements of electricity
markets. Consequently, they cannot be counted upon to meet peak
demand—at least not in the same way that natural gas plants can
be relied upon to produce power on demand. If solar and wind
producers are producing power when it is needed, then that power
36  Wind and Water

can be utilized, but they are unreliable suppliers. Gas-fired and

oil-fired power plants are more reliable peak load power produc-
ers, but their fuels are very expensive, and they take time to start.
Starting a fossil fuel plant is not like flicking a switch. Before they
can be brought online they must reach operating temperature.
This takes hours. But running them simply because they might be
needed is costly both to the ratepayer and to the environment. By
contrast hydroelectric power is peculiarly well-suited for meeting
fluctuations in demand. As long as water is behind the dam, they
are ready to go. They can ramp up or ramp down in a few seconds,
because plant operators need only open or close a gate to increase or
decrease power production. So even though the technology in use
at most hydroelectric plants has not changed in any essential way
in many years, the way that many of these plants are operated has
changed. (Federal environmental legislation has further affected
how hydroelectric plants are operated. See the section “More about
Environmental Costs” in chapter 3.)
 3

Costs and Policies

A ll electric generating stations, whether powered by fossil fuels,
nuclear reactions, water, wind, the Sun, geothermal energy, fuel
cells, or any other technology, are energy conversion devices. Each
technology begins with one type of energy—it might, for example,
be kinetic or thermal—and converts it into electricity. Each type
of conversion technology has its own environmental and economic
costs. When it comes to generating electricity, societies have always
been willing to pay all of the costs involved because access to large
amounts of reasonably priced electricity make modern life possible.
Today the environmental costs of meeting the worldwide demand
for electricity are becoming clearer, and the search for alternative
technologies is becoming increasingly urgent as more nations adopt
westernized energy-intensive lifestyles and begin emitting the enor-
mous amounts of pollution that historically have made Western

37
38  Wind and Water

Hoover Dam with flag (U.S.

Department of the Interior,
Bureau of Reclamation)

lifestyles possible. There is no easy way to meet the world’s appetite

for electricity.
To compare different energy conversion technologies in a mean-
ingful way, it is not enough to understand how each technology is
used to generate electricity. It is just as important to understand the
costs associated with each technology. The purpose of this chapter
is to convey something of the costs associated with hydroelectric
power.

The Costs of Hydropower

Economists sometimes make a distinction between the words price
and cost. The price of a commodity is what the consumer pays. The
cost is interpreted more broadly. By way of example, for a consumer
 Costs and Policies  39

the price of electricity, no matter how it is generated, is stated clearly

on the monthly electric bill that each consumer receives. The cost
of electricity includes the cost of building the generating station,
which may or may not have been heavily subsidized; it includes
damage to the environment caused by generating the electricity; it
includes research and development costs; and it includes numerous
other items that for one reason or another may or may not affect
the price of the electricity. The distinction between price and cost is
particularly helpful in assessing the value of hydropower.
Large hydroelectric projects cost a lot. The exact nature of the
costs associated with any particular hydroelectric power project
depends upon the location of the project, its size, and how it is
operated, but the economic, social, and environmental costs of the
big projects are often so large that only governments are capable of
undertaking them. And because many large hydroelectric projects
are paid for and operated by governments, they are often operated
in ways that are very different from the ways that commercial gen-
erating stations are operated.
Hydroelectric generating stations, their dams, and associated
reservoirs are some of the largest engineering projects in the world.
Consider the Grand Coulee Dam, which is located in the state of
Washington. It is 550 feet (168 m) tall and approximately one mile
(1.6 km) long. Although it began operation more than 70 years ago,
it remains the largest concrete structure in the United States. To
construct the dam required thousands of workers. In fact, when
work began in 1934, the consortium of companies hired to build
the dam found it necessary to build an entire town, called Mason
City, to house some of its employees. While employment levels
fluctuated during the construction period, 6,000 workers were
employed during 1937. The project did not begin producing power
until 1941. The hydroelectric station, which has been upgraded
more than once, currently has a capacity of 6,480 MW (megawatts)
(6.48 gigawatts).
40  Wind and Water

This pattern of long construction times coupled with enormous

outlays of money prior to generating any income is still typical of
these types of projects. Consider the Nathpa Jhakri hydroelectric
power project, one of the largest hydroelectric projects in recent
times. It is located on the Sutlej River in India’s northern state of
Himachal Pradesh. Construction began in 1989, and it began gen-
erating power in 2003. Approximately 7,000 workers were engaged
in this project at its peak. The Nathpa Jhakri plant has a capacity of
1,500 MW.
Most large hydroelectric plants also require large amounts of
land. The Depression-era Hoover Dam, located about 39 miles (62
km) south of Las Vegas, Nevada, created Lake Meade, which covers
approximately 247 square miles. The Egyptian Aswan High Dam,
which took approximately a decade to build—it was completed in
1970—created a roughly 300-mile (480-km) long reservoir behind
it. And the Itaipú hydroelectric power plant, located on the border
of Brazil and Paraguay, required the builders to change the course
of the Paraná River, the seventh largest river in the world. Only
governments can build on such an enormous scale.
To construct the large upstream reservoirs characteristic of these
projects requires the power of the government for another reason:
Many large hydroelectric projects involve the forced displacement
of large populations. In the United States, for the most part, this
has not been a characteristic of large-scale hydroelectric develop-
ment because the largest projects were built in the western part of
the nation at a time when the area was still sparsely populated. To
be sure, there has been some resettlement, but it is estimated that
throughout the history of the United States only about 30,000 people
have been forced to relocate as a consequence of hydroelectric plant
construction, and the majority of these (approximately 18,000) were
displaced as a result of a single project, the Norris Dam, located on
the Clinch River near present-day Norris, Tennessee. (The Norris
Dam, built during the years 1933 to 1936, forms Lake Norris, with a
 Costs and Policies  41

surface area of about 53 square miles [137 km2]. The project has a ca-
pacity of about 131 MW.) By contrast, the Three Gorges Dam, which
is built across the Yangtze River (Chang Jiang) in Hubei Province
in China, involved the displacement of more than 1 million people.
(The construction of the Three Gorges Dam began in 1993, and it
is expected to be fully operational in 2009. It will have a capacity of
18,000 MW.) Nor is the Three Gorges Dam the only such project to
cause large-scale dislocations. The Sanmenxia Dam, also in China,
involved the relocation of 410,000 people; China’s Xinanjian Dam
required the resettlement of 306,000 people, and India’s Bargi Dam
required the resettlement of 113,000 people. To effect such large-
scale displacements requires the power of the state.
Large-scale hydroelectric power projects also involve changes
to both the upstream and downstream environments. Upstream of
the dam a large lake is often formed that changes the ecology all
along that section of the river. This degrades the environment for
some species and improves it for others. Downstream of the dam,
the effects of the hydroelectric plant are more complex and depend
on how the dam is operated. Operation of the dam is determined
by the interaction of government regulations and market pressures.
If, however, the goal is to maintain the environment along the river
so that it remains as unaltered as possible, the construction of a
hydroelectric power plant presents a serious challenge. The situa-
tion is complex and is discussed in more detail in the next section
of this chapter.
The effects of the hydroelectric plant on the broader envi-
ronment are also complex and are due, in part, to the way that
hydroelectric power is priced as well as the way that the water
behind the dam is utilized. Governments often choose to subsi-
dize the cost of hydropower—that is, the electricity is sold for less
than it costs to produce. In one sense this makes the hydroelectric
plant a money-loser. But electricity is more than a simple com-
modity. Inexpensive electricity is a necessary factor for all types
42  Wind and Water

of economic growth. In other words, even if a power project never

makes money through the sale of electricity, it can still serve an
important and positive economic function. The history of the
Grand Coulee project clearly illustrates the widespread economic
effect of hydroelectric power.
Prior to the time that the Grand Coulee hydroelectric plant be-
gan limited operations in 1940, there was no aluminum industry in
the Pacific Northwest. (The aluminum industry depends upon large
amounts of inexpensive electricity, and aluminum plants will often
locate near a source of cheap electricity.) In 1946, only four years
after Grand Coulee began full-power operation, 36 percent of the
aluminum producing capacity in the United States was located in
the Northwest. Not all of the electricity upon which the aluminum
plants depended came from the Grand Coulee Dam, but much of it
did, and the rest came from other hydroelectric stations that were
completed in the same general area at roughly the same time. The
Grand Coulee Dam increased the size of the regional economy; it
increased the tax base; it drew workers to the area; but it generated
no immediate profit from the sale of electricity. In the United States
and in some other countries as well, hydroelectric power plants
were operated so as to cause economic growth.
The effects of the Grand Coulee Dam on the development of
the Northwest were even more widespread than indicated in the
preceding paragraph, because Grand Coulee was also built to pro-
vide sufficient water to irrigate more than 1 million acres (405,000
hectares) of land. Prior to Grand Coulee’s construction there were
few farmers in the area. The soil was rich but too dry to farm profit-
ably. After completion of the dam, a large and vibrant agricultural
sector was established in the region. Again, the result of the Grand
Coulee Dam project was a further influx of workers—this time
agricultural workers—and the environment was further changed
as large tracts of previously undeveloped land were converted into
farmland.
 Costs and Policies  43

Although each large-scale hydroelectric facility is unique, it is

fair to say that the effects of many other large-scale hydropower
projects have been somewhat similar to those at Grand Coulee.
One should not conclude from the previous paragraphs that
economic development is wrong or unhealthy. Many would con-
sider all of these economic effects to be positive—that is, the eco-
nomic effects (and the accompanying environmental changes) that
resulted from the construction and operation of the Grand Coulee
project should be counted as benefits rather than costs. Certainly
there are many people whose lives have been enhanced by what the
Grand Coulee and other regional hydropower projects have made
possible. Whatever one’s opinions, the Grand Coulee dam project
illustrates that the effects of large-scale hydroelectric projects on
the environment extend well beyond the immediate vicinity of the
dam and the river on which it depends.
With such high economic, social, and environmental costs,
one might question the value of hydroelectric projects, but it is just
as important to remember the alternatives. Whatever the costs of
hydroelectric power, they are, at least, local—that is, the costs of
hydroelectric power development are restricted to the area in which
the plant is located. This area might be fairly large—as noted pre-
viously, the construction of the Grand Coulee Dam affected the
development of the entire Pacific Northwest—but the costs of dam
construction and operation are, at least, not global. Compare the
costs of hydroelectric plants with those of fossil fuel plants, espe-
cially the costs of emissions from fossil fuel generating stations.
Today, most electricity is generated by burning coal, and coal tech-
nology also has important local costs. The mining of coal involves
significant environmental disruption, and mining is a dangerous
business; miners are injured and killed on a regular basis in the per-
formance of their jobs. The routine operation of coal plants releases
a number of dangerous materials into the environment. Mercury
(continued on page 46)
 44  Wind and Water

Global Warming

G lobal warming, the gradual increase in the average temperature of

Earth’s oceans and atmosphere, continues to attract a great deal of
attention in the popular press and in scientific journals. Global warming is
due to changes in the chemical composition of Earth’s atmosphere. To see
the profound impact of Earth’s atmosphere on surface temperature com-
pare Earth with its neighbor, the Moon. When the Sun is shining directly on
a region located at the Moon’s equator, the surface temperature is about
230°F (111°C). But in the dark, the Moon’s surface temperature plunges
to -450°F (-233°C). There is no place on Earth’s surface with temperatures
that approach either of these extremes even though the Earth and Moon
are located at approximately the same distance from the Sun. Earth’s at-
mosphere is what accounts for the difference. The density and chemical
composition of Earth’s atmosphere affects both the amount of the Sun’s
energy that reaches Earth’s surface and the amount of that energy that
Earth retains. Change the chemical composition of the atmosphere and
the temperature of Earth’s surface changes as well. Gases in Earth’s at-
mosphere that effectively retain energy from the Sun and contribute to a
warmer climate are called greenhouse gases.
Human beings are altering the chemical composition of Earth’s
atmosphere—mostly by burning fossil fuels, a process that releases
carbon dioxide into the atmosphere. Carbon dioxide is a potent green-
house gas. Whatever the beneficial effects of heavy fossil fuel consump-
tion—and there have been many, at least in developed countries—it
has become clear that the consumption of fossil fuels at current levels
releases so much carbon dioxide into the atmosphere that the thermal
properties of Earth’s atmosphere have changed as a result. Earth now
retains more of the Sun’s energy than it did in the period immediately
preceding the time in which humans began to consume large amounts
of fossil fuels. Many of the long-term effects of this change in atmo-
spheric chemistry are not yet clear. What, for example, will be the effects
of increasing oceanic and atmospheric temperatures on the powerful
ocean currents, such as the Gulf Stream, that distribute thermal energy
 Costs and Policies  45

about the planet? The answer is currently unknown. But the question is
vitally important because changes in the distribution of heat may cause
some regions to cool even as the average global temperature increases.
The effects of an increase in average temperature can be complex and
difficult to predict.
What is clear—because it has been measured—is that the chemi-
cal composition of Earth’s atmosphere is changing. Because hydroelec­
tric plants convert kinetic energy rather than chemical energy into
(continues)

An iceberg in Drake’s passage, off the coast of Antarctica. Although the

atmospheric concentrations of greenhouse gases are easily measured—
and they are increasing—the effects of these gases are harder to predict.
(Cathy Webster)
 46  Wind and Water

(continued)
­ lectricity, they produce electricity with no greenhouse gas emissions.
e
(Actually, this statement is only true in temperate climates; in tropical
locations the situation is more complicated. See “Methane Emissions
and Hydropower,” later in this chapter.) While large-scale hydropower
projects have their own environmental problems, their value must be
assessed relative to the presently available alternatives. Many perceive
hydroelectric plants as an environmentally advantageous alternative to
fossil fuel consumption.

(continued from page 43)

emitted from coal plants in the United States Midwest, for example,
has found its way into the bodies of fish throughout the Northeast,
and greenhouse gases emitted from these same plants contribute
to global climate change. The costs associated with fossil fuel con-
sumption are distributed over a much broader area and have more
far-reaching consequences than those associated with hydroelectric
power. In contrast to the costs of hydroelectric generating stations,
the costs of fossil fuel consumption are shared by everyone.

More about Environmental Costs

The operator of every hydroelectric power plant must balance
profits, production, and environmental concerns. Sometimes these
goals coincide, and sometimes achieving one goal means sacrificing
another. Achieving the right balance for a particular power plant
is a complex problem. With respect to an investor-owned facility,
for example, it may not always be possible to maximize profits and
minimize environmental damage. There is room for disagreement
 Costs and Policies  47

Three Gorges Dam on the Yangtze River (Chang Jiang) in China (Wikimedia)

among people of goodwill about whether any particular solution is

the “right” solution.
Finding a reasonable balance between competing economic and
environmental goals is further complicated by the fact that there is
no generally agreed upon way to assign a value to environmental
damage. There is not even a generally agreed-upon standard as to
precisely what constitutes damage to the environment. How, for ex-
ample, should one determine the value of emissions-free electricity?
What value should be assigned to the preservation of a particular
ecosystem? Which changes constitute damage and which do not?
Even many ecologists tend to answer these questions using quasi-re-
ligious language or quasi-scientific ideas, but these kinds of answers
do not lead to rigorous solutions; they do not solve the problem of
finding the best way to balance the two competing values of meet-
ing the demand for reasonably priced electricity and minimizing
the environmental changes that result from the production of that
electricity.
Today, a great deal of attention is given to the difficult problem
of optimizing the operation of hydroelectric power plants. Finding
an optimal solution is further complicated by the fact that every
48  Wind and Water

solution, optimal or not, must conform to the regulatory environ-

ment determined by the nation in which the power station is lo-
cated. In the United States, over the last century, federal, state, and
sometimes even local governments have, for a variety of reasons,
passed laws and adopted regulations to constrain the way that
power producers operate. To appreciate some of the ways that these
factors affect one another, it is helpful to know a little about the
history of the regulation of power producers in the United States.
Dams are heavily regulated. They use a shared resource, in this case
a river, to produce a vital commodity, electricity. Ensuring that this is
done in a responsible way has long been an interest of government.
Historically, early electricity markets were local monopolies.
One company, the local utility, owned the electricity generating sta-
tions in its service area; it owned the high-voltage transmission lines
that brought the electricity from the location where it was produced
to the local market, and it owned the network of low-voltage power
lines that connected the high-voltage network to the businesses and
homes that depended upon the electricity. (As described in chap-
ter 1, transformers are used to increase and decrease the voltage,
a characteristic of electrical current analogous to the pressure of
water flowing through a pipe, so that the electricity can be safely
transmitted with minimum losses.)
Some of these utilities were privately owned monopolies; other
utilities were owned by a local government. The federal govern-
ment also owned some large generating stations, especially large
hydroelectric stations. The reasons for the monopolistic nature of
early electricity markets are primarily historical. Early private and
municipal companies had to build their own electricity supply and
distribution networks if they were going to sell their electricity
because originally there was no distribution infrastructure. Sup-
pliers became distributors in order to bring their product, elec-
tricity, to market. Moreover, because there was not enough room
on a single street for multiple sets of utility poles to accommodate
 Costs and Policies  49

multiple electricity suppliers, the first supplier to erect a distribu-

tion network for a particular region remained the only supplier
for that market.
This monopolistic system, which emerged out of the first efforts
of Thomas Edison and other technological pioneers, remained in
place until the latter decades of the 20th century. One reason for
its longevity is that for most of that time it worked well, providing
reasonably priced electricity to almost everyone while ensuring rea-
sonable profits for the power producers. But another reason that no
alternatives were proposed during these decades is that no one had
imagined an alternative. Utility monopolies were widely perceived
as “natural monopolies,” in the sense that there seemed to be no
practical alternative to this way of doing business.
But monopolies were also viewed with suspicion because mo-
nopolies of all sorts have sometimes abused their market position
and offered consumers limited services at inflated prices. There was
fear that a similar situation would arise in the electricity markets.
Electricity is, however, too valuable to the economic well-being of
the nation to trust to the good intentions of local monopolies. As a
consequence, various regulatory bodies were established to ensure
that electric rates were “fair” to both consumers and utilities. Dur-
ing this time, hydroelectric plants were usually operated with little
regard for the environment.
Beginning in the 1970s the federal government passed a series
of laws to constrain the operation of hydroelectric plants in order
to minimize the environmental effects these plants have. Of special
interest are the 1973 Endangered Species Act, which introduced
constraints on the operation of some hydroelectric plants, and the
Electric Consumers Protection Act of 1986, which required federal
regulators to balance hydroelectric power production with environ-
mental concerns. The effect of these and other later pieces of legisla-
tion was to require operators to operate their generating stations in
ways that lessened their environmental impacts.
50  Wind and Water

Another important set of changes began when the electricity

markets were restructured. The new markets, which are still evolv-
ing, are sometimes described as deregulated, but that is not accurate.
Instead, beginning in the 1990s, the government instituted a new
set of regulations that attempted to introduce competition between
power producers—that is, they sought to break up the old natural
monopolies. (Certain parts of the electricity infrastructure, most
notably high-voltage transmission lines, are, however, still viewed
as natural monopolies, and these remain tightly regulated by the
Federal Energy Regulatory Commission [FERC].)
The key to the government’s attempt at introducing competition
lay in a 1996 order by FERC that required owners of high-voltage
transmission lines to grant all power producers “nondiscriminatory”
access to the high-voltage lines that serviced an electricity market.
The idea was that all power producers would use the system of high-
voltage transmission lines as a sort of electric highway. Each power
producer would have equal access to the highway and each would at-
tempt to produce electricity at a price that would find a buyer. Open
access to the high-voltage network would enable multiple power
producers to compete for a limited customer base. If one power pro-
ducer could supply power at a more competitive price than another,
then the lower-price supplier would earn profits at the expense of the
higher-price supplier. Innovation, it was hoped, would flourish.
At least that was the theory. In the jargon of the industry, a
competitive market provides “price signals” that investors can use
to guide their investment decisions. The price of the electricity—not
its cost—would guide the way that the market evolved. This new
system has had a number of intended and unintended effects on the
electricity markets.
One consequence of the restructured markets was that some
hydroelectric plants were increasingly used for peak power produc-
tion. Unlike most other power generating technologies, hydroelec-
tric power works equally well providing peak power or base load
 Costs and Policies  51

power. Base load power is highly predictable and so utilities gener-

ally enter into long-term contracts with power producers, mostly
coal and nuclear, to provide base load power. Some hydroelectric
plants are used to provide base load power, but peak power, which
is sold on an hour-by-hour basis, is generally sold at a premium.
Provided that one can sell enough peak power, there is more profit
selling peak than base load power, a clear price signal that the peak
power market is the preferred market.
Four economists, Matthew Kotchen, Michael R. Moore, Frank
Lupi, and Edward Rutherford, have published “Environmental Con-
straints on Hydropower,” an interesting study of the effects of regu-
lation on the price and the costs of hydroelectric power production.
They sought to quantify what happened when, as a condition of a
relicensing agreement between FERC and the Consumers Energy
Company, Consumers Energy changed the way that it operated two
medium-sized hydroelectric plants on the Manistee River in Michi-
gan. The plants in question are the 20.1-MW generating station at
Tippy Dam and the 17-MW generating station at the Hodenpyl
Dam. Prior to the relicensing agreement both facilities were being
used to produce peak power.
If no thought is given to the environment, the most profitable
way to produce hydroelectric power is to leave all the water im-
pounded behind a dam and release it only intermittently. In particu-
lar, this means releasing little or no water during the evenings when
demand and prices are low, and releasing large amounts during the
day, when demand and profits are high, and for a while, Consumers
Energy operated its dams in just that way. But that operating regime
caused large fluctuations in water levels and flow rates on an almost
daily basis. As part of its relicensing agreement, Consumers Energy
agreed to operate its dams in so-called run of river mode—which
is another way of saying that it would monitor the amount of wa-
ter flowing into the reservoirs and release it through the dams at
roughly the same rate. The reason that FERC wanted run of river
 52  Wind and Water

operating mode is that releasing large amounts of water during pe-

riods of peak demand and withholding it otherwise causes a fair
amount of environmental damage downstream. In a single day the
water levels downstream of these dams varied from drought level to
flood stage and back again. Run of river, on the other hand, creates
downstream conditions that are more similar to what would exist if
the dams were not on the river.

Methane Emissions and

Hydropower

H ydropower is often described as an emissions-free source of electric-

ity and in temperate areas of the world this is certainly true, but the
situation is more complicated in tropical regions. Deep in the reservoirs
that form behind large tropical dams, bacteria feed on organic matter
in low-oxygen environments. One of the by-products of this process is
methane, which is released by the bacteria. At the high pressures that
exist deep beneath the surface, the methane simply accumulates in the
water. But when this water passes through turbines and encounters the
low (atmospheric) pressure on the downstream side of the dam, the
methane bubbles out of the water in much the same way that carbon
dioxide bubbles out of soda when the bottle is first opened. This has im-
portant environmental consequences because, once in the atmosphere,
methane retains heat far more efficiently than does carbon dioxide—that
is, methane is a far more potent greenhouse gas than is carbon dioxide
(about 20 times as efficient)—and for some dams, the amount of methane
released into the atmosphere is substantial. Scientists at Brazil’s National
Space Research Institute estimate that for some Brazilian hydroelectric
facilities, the methane released in this way contributes more to climate
change than would the carbon dioxide released by a fossil fuel plant with
a similar power output. In fact, they estimate that each year worldwide,
the total effect of the methane released during dam operations in tropical
 Costs and Policies  53

Under the new agreement, the two facilities produced less elec-
tricity during peak periods but began producing energy during off-
peak times. The consequences of the new operating procedures were
complicated: First, the belief that switching to run of river operating
mode would be better for the environment was justified. Chinook
salmon runs on the Manistee River more than tripled, from 100,000
per year to approximately 370,000 after the operators switched to run

environments has the same effect as releasing 800 million tons of carbon
dioxide, an enormous environmental load.
The solution, which is also being developed at the National Space
Research Institute, is to capture the methane before it passes through
the dam. An intake pipe will pull the methane-rich waters that exist deep
within the reservoir up to a natural gas plant that would be built on-site.
(Natural gas is almost pure methane.) As the methane separates from the
water, it is captured and burned to produce electricity. The water, now
nearly methane-free, is returned to the reservoir. Of course, the carbon
dioxide produced at the plant by burning the methane is also a green-
house gas, but its effect on the climate is far less than that of the meth-
ane from which it was derived. Moreover, the mass of the carbon dioxide
produced by burning the methane is roughly equal in mass to the carbon
dioxide removed from the atmosphere by the plants that produced the
organic matter on which the deepwater bacteria fed. Returning the car-
bon dioxide to the atmosphere simply “completes” the cycle, so in this
sense the Brazilian scheme is approximately carbon neutral. In areas of
the Amazon where the methane content is exceptionally high, the fossil
fuel plants that would be built to capture the methane could increase
the power outputs of their respective hydroelectric facilities by as much
as 50 percent.
54  Wind and Water

of river. Second, Consumers Energy’s costs for operating the dams

rose by approximately $300,000 per year. FERC had predicted that the
change in operating mode would enable Consumer’s Energy to main-
tain total power output. This, too, proved to be correct, but some of
that energy was now produced when there was less demand—and so
less profit for the producer—than under the peak-production mode.
The difference in timing accounts for what is essentially the $300,000
loss. Most interesting was the impact of changing the operating mode
of the dam on Consumers Energy’s greenhouse gas emissions.
Consumers Energy’s first responsibility is to meet the total
­minute-by-minute demand for electricity on the part of its customers.
Their choices for meeting this demand were hydroelectric power or
fossil fuel plants. Having cut back hydroelectric power when demand
was high, Consumers had to generate more peak power with fossil
fuels. But because Consumers increased hydroelectric power when
demand was low, they cut back on that portion of base load power
produced by fossil fuel generating stations. The difference, however,
was about more than timing. Base load generating capacity is, in this
case, largely done by coal-burning plants. Peak load generating ca-
pacity is generally met with natural gas plants. Burning coal releases
more carbon dioxide per unit of heat produced than does burning
natural gas. By changing power-production schedules, Consumers
Energy burned less coal and more natural gas. As a consequence, the
run of river operating regime also cut, if only a little, air pollution
levels and the emission of greenhouse gases.
Was it worth it? The answer depends on the relative values that
one places on the environment, company profits, and the price of
electricity. With respect to hydroelectric power, each plant is unique
and requires its own analysis. Even after 100 years there is still more
to learn about the best ways to employ these valuable resources.

The Future of Hydropower

Worldwide, the number of remaining sites suitable for traditional
large-scale hydropower development is not large and cannot be
 Costs and Policies  55

increased. There are more medium-sized sites with smaller hydrau-

lic heads and lower volumetric flow rates such as the Tippy Dam,
described in the previous section, and there are many sites suitable
for microhydropower development. Microhydropower systems are
defined to be those that generate no more than 100 kW, but they
often generate much less. The difficulty with medium and smaller-
sized facilities is that it takes so many of them to produce the same
amount of power as a single large unit. Looking back at equation
(2.1), one can see that to produce a large amount of electricity re-
quires a high hydraulic head and a large flow rate. To see why large
facilities are better in this respect, compare a larger and a smaller
generating station. Suppose a smaller station has half the hydraulic
head and half the volumetric flow rate of a larger facility. The result
is that the smaller facility can at best produce only one-fourth the
amount of power of the larger facility. Small facilities are not just
small in size; they are inefficient in the sense that they require more
water to produce the same number of kilowatts. Because of these
inefficiencies there are hard limits to what can be done with smaller
facilities.
In the early days of the United States’ Industrial Revolution,
engineers quickly identified locations along rivers in the eastern
part of the country that combined large vertical drops over short
distances with good flow rates. Factories sprouted in these places
as fast as the engineers could build the facilities needed to harness
the rivers. There was a limited number of such locations, of course,
and they were soon developed. These technological pioneers recog-
nized these locations as valuable natural resources to be exploited
for the public and the private good, though not necessarily in that
order. Today, as the consequences of the fossil fuel economy become
increasingly clear, there is renewed interest in developing hydro-
power as well as heightened interest in the more effective use of
hydropower. Both areas are of some importance for economic and
environmental reasons, but the value of conventional hydropower
is limited. Most countries simply do not have a sufficient number
56  Wind and Water

of sites to produce a large percentage of their power from flowing

water, and in developed nations most of the best sites have already
been developed. In most cases, countries interested in meeting ad-
ditional demand with reasonably priced emissions-free power will
have to look to other power-production technologies.
 Part II

Electricity from
the Oceans
 4

Wave Power
A nyone who has seen pictures of ocean waves crashing against
the shore knows that there is a great deal of power in waves.
They can reduce cliffs to rubble and destroy large ships. These
examples illustrate both the advantage and the disadvantage of
wave power: Waves are both powerful and (usually) destructive. A
machine designed to convert the energy of waves into electricity
could, at least in principle, produce a great deal of electrical energy
in such an energetic environment. The main barriers to exploiting
this resource are that (1) wave power occurs in surges, and (2) the
wave environment is a destructive one. It has so far proven difficult
to design a device that can withstand the routine violence of the
sea and still produce power reliably. But technology continues to
improve. This chapter considers three methods for converting wave
energy into electrical energy.

59
60  Wind and Water

Winter storm swells, North Pacific. The ocean can be a very energetic
­environment. (NOAA)

Before describing these conversion schemes, it is worthwhile to

consider the waves themselves. A wave is a disturbance in the surface
of the ocean. The types of waves of importance here are generally
caused by wind blowing across the ocean’s surface. Far from shore, if
the wind blows hard enough and over a long enough distance it can
create substantial disturbances (waves) that move across the surface
in a motion that some describe as rolling. It is the disturbance that
moves forward; the water of which it is momentarily composed sim-
ply oscillates in a (more or less) up and down motion—that is to say,
it is the wave rather than the water that moves across the ocean.
Waves are classified according to three characteristics: (1) the
amplitude of the wave, which is defined as one half of the vertical
distance from the wave’s peak to its trough, (2) the wavelength, L,
of a wave, which is defined to be the distance from one peak to
 Wave Power  61

the next, and (3) the wave period, T, which is the time that elapses
between the passing of wave crests. The speed with which a wave
travels depends in part on how deep the water is, but in deep water
wave speed is a simple function of the wavelength and the period:

v = L/T

where v represents the speed of the wave.

Wind generates waves of different wavelengths and periods and
so, as the preceding equation indicates, ocean waves travel at dif-
ferent speeds. A faster wave may overtake a slower wave, and as it
passes through the slower wave they will briefly combine to form a
new wave with an amplitude that is greater than the amplitude of
either of the individual waves. The faster wave continues to move
forward and eventually the two waves separate; their characteris-
tics—amplitude, wavelength, and period—will be the same after the
encounter as they were before. This phenomenon is called interfer-
ence, and it helps to explain why ocean waves may vary in height
and distance from one another.

Sine wave showing amplitude and wavelength

62  Wind and Water

Waves, having formed in a windy region of the ocean, last a

surprisingly long time. It is not uncommon for waves to roll across
the sea for many hundreds of miles even when the wind that created
them ceases to blow. As waves travel across the ocean they slowly
dissipate. Waves with shorter wavelengths fade first so that a group
of waves that have traveled a long way across the ocean will be com-
posed exclusively of waves of longer wavelengths. Near where the
waves are formed, however, there will be waves of many different
wavelengths, and this makes for a more turbulent sea.
Finally, the energy of the wave—and it is the energy that mat-
ters to those attempting to convert wave energy into electricity—is
proportional to the square of the amplitude of the wave. The equa-
tion that describes this situation is

E = pA2

where E represents the energy of the wave, A represents its ampli-

tude, and p is a constant of proportionality whose value depends
on the units used to express energy and amplitude. This equation
is important because it shows that if one doubles the amplitude of
the wave, the amount of energy in the wave increases by a factor of
four. Therefore, small changes in the height of waves represent large
changes in the energy of the waves.

Sea Snakes
One method of wave energy conversion has been developed by the
company Pelamis Wave Power. Their device is called the Pelamis
Wave Energy Converter, and it has already been deployed in small
numbers. (Pelamis is the name of a snake that swims across the
surface of the sea.)
The Pelamis Wave Energy Converter, or “the Pelamis” for short,
which is sometimes called the “snake” because of its shape and
motion, consists of four long cylindrical sections. The sections are
aligned one after the other and connected by three flexible joints to
 Wave Power  63

Pelamis wave energy farm (Pelamis Wave Power)

form a segmented string-like assemblage almost 12 feet (3.5 m) in

diameter with a total length of approximately 500 feet (150 m). The
Pelamis is designed to float at the ocean’s surface while held in place
by a tether.
Pelamis wave farms will usually be located a few miles from
shore, where the waves are more predictable, and in waters deep
enough so that the waves are unaffected by interactions with the
seafloor. These near-shore locations are important because the cost
of equipment needed to carry the power from a Pelamis wave farm
back to shore is high, and each unit must be towed to shore for
service and repair. Siting the “wave farm” as close to shore as pos-
sible—but not so close that the waves are weakened by interactions
with the seafloor—reduces these costs. The first Pelamis wave farm,
a small operation located three miles (5 km) off the Portuguese
64  Wind and Water

coast, is, as of the spring of 2008, under construction. Another

wave farm is planned for Scotland, where Pelamis converters will
be used to provide power for the Orkney Islands. Pelamis facilities
are modular—that is, to obtain more power one simply adds more
converters.
The devices work in the following way: They are tethered to the
ocean floor so that they can swing perpendicularly to incoming
waves. As waves pass along the length of the device, the buoyancy
force, the force exerted by the water against the body of the snake,
varies from point to point. The Pelamis, because it is jointed, tends
to sag in the troughs of the waves and arch upwards where the crest
is passing. The Pelamis’s motion is not really snakelike—its body
is far too rigid—but the cylinders do bend at the joints. As a joint
begins to flex, a great deal of force is exerted at the joint by the mas-
sive steel cylinders, and this force is used to power a type of pump
called a hydraulic ram. The hydraulic ram drives oil through a mo-
tor, which performs a function analogous to that of a turbine in a
hydroelectric facility; the hydraulic motor drives a generator, which
produces the electricity. The power is sent to a cable below the unit.
The cable carries the electricity to shore.
In concept the Pelamis wave energy converter is almost, but not
quite, that simple. Because the joints are powered by passing waves,
they flex only intermittently and irregularly. Without additional
modification, the output from the Pelamis would be too unsteady
to be useful. The engineers’ solution is to employ something called
an accumulator. The accumulator, which is placed between the mo-
tor and the hydraulic ram, captures the oil and releases it to the
motor at a steadier rate. This extra step smoothes the power output
of the device.
So far Pelamis has found buyers because the technology shows
promise. These devices produce power with zero emissions. They
have a low visual profile, which is an important advantage since
one of the main objections to wind energy is that the equipment is
 Wave Power  65

As waves move past the Pelamis, they cause it to flex, and as it flexes it
generates electricity.

visually intrusive. In contrast to the tall towers of wind turbines,

Pelamis converters float low in the water and are located miles from
land. And, as previously mentioned, these devices use a resource
that is more predictable than wind. In theory, Pelamis wave farms
could produce large amounts of power while occupying a smaller
area of Earth’s surface than wind farms with similar outputs. Even
now, for certain remote applications, the Pelamis system can make
an important contribution. Those are some of the main advantages
to the Pelamis system.
There are disadvantages to the system as well. At present these
wave-energy converters are not economically competitive with
more conventional power sources. As of this writing, power from a
Pelamis unit costs about twice that of power generated by wind tur-
bines, but Pelamis representatives claim that as production volumes
increase costs will quickly decrease. This is surely true, although
it is not as clear that costs will decrease enough to make Pelamis
power competitive with wind—never mind conventional sources of
power, which are often significantly cheaper than wind.
66  Wind and Water

The second disadvantage involves changes in energy availability.

Each snake has a rated power output of 750 kW, but unlike more
conventional generating stations—that is, coal, natural gas, oil, or
nuclear—the amount of power each unit actually produces is vari-
able and in most ways is beyond the control of the operator. Although
waves are more predictable than wind, wave energy also varies from
day to day and according to the season. Some of this variation is pre-
dictable, and some is not. Pelamis technology decouples supply and
demand in the sense that a wave farm may produce large amounts
of power when demand is low and small amounts of power when
demand is high. Once the site is chosen and the units deployed, the
level of power production is largely out of the operator’s control.
To their credit, Pelamis designers have found ways to increase
the efficiency of their devices during periods of low wave activity
and so lessen the effect of the inevitable variations in wave energy.
But even for the ocean test site used by Pelamis for data acquisi-
tion, the projections that they published indicate that a wave farm
at their site would generate more than twice as much power during
winter months as during summer months. In other words, produc-
ing the same amount of power during summer as during winter
would require deploying many more units than would be needed
to meet wintertime demand. Alternatively, one could deploy other
technologies in order to assure adequate power supplies when the
wave energy converters fall short. Both strategies are expensive. The
Pelamis is also designed to shut down during periods of intense wave
activity to prevent damage to the unit. This saves the wave farm, but
increases its unreliability as a source of energy.
These advantages and disadvantages need to be evaluated rela-
tive to the other options that are available. True, wave energy fluctu-
ates, but it does not exhibit as much day-to-day variation as does
wind. Buyers must choose the best technology from the available
alternatives. They do not have the luxury of choosing the best tech-
nology imaginable.
 Wave Power  67

The Pelamis converter may prove useful to buyers willing to

pay the premium necessary to diversify their energy base—espe-
cially if their goal is to produce some power from waves, some
from wind, and some from solar, for example. By distributing
one’s risk among several different intermittent energy sources,
one lessens the probability that there will be power production
shortfalls since it is less likely that the wind, waves, and Sun, for
example, will all fail simultaneously. But this approach also in-
creases costs since it requires the power producer to build parallel
systems, each using a different energy source and each system built
with enough capacity to meet the level of demand that will occur
when one or more of the other technologies fails to produce. By
distributing risk in this way, the system operator can decrease the
probability that it will be unable to meet the demand for power.
Alternatively, fossil fuel plants can be used to provide the neces-
sary margin of safety. What is certain is that the system requires
backup, and backup systems cost money. It is a truism that money
spent on electricity is money diverted from other uses, such as
health care, education, transportation, and even energy research.
Whether consumers and taxpayers are willing to pay the extra
costs involved in deploying the Pelamis and other similar systems
remains to be seen.

Blow Holes
The oscillating water column (OWC) energy converter is the name
of the technology described in this section. The first of these devices
to provide electricity to the grid is called the Limpet and is located
on the isle of Islay off the coast of Scotland. The Limpet plant be-
gan generating commercial amounts of electricity in the year 2000.
(Scotland, which invests heavily in renewable energy, is also where
Pelamis Wave Power is located.) The Limpet’s OWC technology is
produced by a company called Wavegen and in concept the idea is
elegant and simple.
68  Wind and Water

Cutaway views of an OWC. As the wave approaches air is forced out of the
opening on top, and airflow reverses as the wave recedes.

To see how an OWC energy converter works, imagine a bicycle

pump. Pushing on the handle of the pump drives the piston down-
ward through the cylinder. The piston pushes the air forward and
causes it to rush out the other end of the hose, which is attached
to the bottom of the cylinder. If the cross-sectional area of the
hose were the same as the cross-sectional area of the piston, the air
 Wave Power  69

would flow out of the pump only as fast as the pump handle moves
forward, but pumps are not designed that way. Instead, the cross-
sectional area of the hose is many times smaller than the surface
area of the piston. As a consequence the speed of air flowing out of
the hose is many times greater than the speed of the piston. There is
an easy algebraic equation to describe this situation:

SpAp = ShAh (4.1)

where Sp and Ap represent the speed of the piston and the area of the
piston, respectively, and Sh and Ah represent the speed of the air as
it passes through the hose and the cross-sectional area of the hose,
respectively. This equation was known to Leonardo da Vinci from
his studies of water moving through canals of variable cross sec-
tions. (From a mathematical point of view equation [4.1] is identical
to equation [2.1].)
Equation (4.1) shows that if the cross-sectional area of the hose,
Ah, is small then the speed, Sh, must be large in order that the prod-
uct of the two terms equals SpAp. In fact, even if Sp, the speed of the
piston, is not very fast, the product SpAp will be large as long as Ap is
large enough. Consequently, given SpAp, in order to force the air to
rush very quickly out of the hose, one need only choose a hose with
a small enough cross section. The entire process works the same in
reverse. If one pulls the handle of the pump back so that the piston
moves up the tube and away from the hose, air will rush up the
hose and into the piston at the speed predicted by the preceding
equation.
OWC technology works on the same principle as the pump.
First, a large structure is built in an area of strong wave action. The
structure is boxlike in shape and has two openings. The first is a
comparatively small hole at the top of the box. The second “hole” is
the bottom of the box, which is left open so that water can flow in
and out. The walls of the structure extend down below the water-
line so that the ocean forms an airtight seal at the base of the box.
70  Wind and Water

Each incoming wave causes water to flow under the walls and up
into the box. The rising water level inside the box displaces the air
in the box, which rushes out of the hole (or holes) at the top. (The
water acts like the piston in a pump.) As the wave recedes, the water
inside the box flows out of the bottom and creates a region of low
pressure inside the box. Air flows back down into the box through
the smaller opening (or openings) at the top. (This is equivalent to
pulling up on the handle of the pump and withdrawing the piston
from the tube.) By adjusting the relative sizes of the box and the
opening (or openings) at the top, it is possible to create a power-
ful wind flowing through the top first in one direction and then in
the other. The situation is completely analogous to adjusting the air
speed through the hose of a bicycle pump by adjusting the relative
cross-sectional areas of the piston and the hose.
To convert the up-and-down oscillations of the waves into elec-
trical power, the next step is to place a turbine in each opening at the
top of the structure. As air rushes in and out through the blades of
the turbine, the turbine converts the linear motion of the air into ro-
tary motion, and the rotary motion of each turbine is used to drive a
generator. Technically, a significant problem with this design is that
the airflow is continually reversing direction—out of the box when
the wave is rushing forward and into the box when the wave is reced-
ing. The solution is to use a turbine that turns in the same direction
regardless of the direction of airflow. These turbines now exist.
It is important to note that Wavegen is not the only company
engaged in developing wave power using an oscillating water col-
umn. The Japanese government has long supported research into a
similar idea called the Mighty Whale, and an Australian company,
Energetech, has further modified the ideas of OWC technology used
at the Limpet plant by employing a parabolically shaped wall to fo-
cus the energy of a large segment of an oncoming wave into a small
area with the goal of increasing the efficiency of the device—that is,
increasing the amount of electrical energy generated per wave.
 Wave Power  71

As with other wave energy converters, OWC technology pro-

duces no greenhouse gas emissions. The turbine, the main moving
part, can be easily removed for repair or maintenance because it
is on land. The converters can be placed along the shore, as with
Wavegen’s Limpet, or near the shore, as with the Mighty Whale and
the Energetech device—almost anyplace where there is significant
wave action. Because waves are more regular than wind, the output
of an OWC design is, in theory, somewhat more predictable than
that of a wind turbine.
Most of the drawbacks of OWC converters are similar to those
of the Pelamis: Power from the OWC converters is more expensive
than that produced by other more conventional power plants, but
at least part of this disadvantage would disappear if the units were
mass-produced. A second disadvantage is that, as with the Pela-
mis, the OWC’s output is dependent on the level of wave energy,
and wave energy varies day by day and according to the season.
The amount of energy available for conversion does not, however,
depend on the level of demand for electricity. Therefore, the power
that these devices produce may not be available when the power is
needed. Therefore, compared to more conventional power sources
such as fossil fuel plants, OWCs are not especially reliable. There
is nothing to be done about this. It is a characteristic of OWC and
other similar technologies. Finally, output per unit, even under the
best of conditions, is relatively small because by the time a wave is
near the shore much of its energy has dissipated. Interactions be-
tween the wave and seafloor weaken the wave, and OWCs are built
along the shore or in shallow water. As a consequence, it would
be necessary to construct many OWC plants in order to make a
substantive contribution to the power supply of any large energy
market. But if its value to the larger economy is limited, an OWC
plant can make a real contribution to more isolated island markets,
which is, after all, exactly where the first unit, the Wavegen unit,
was constructed.
72  Wind and Water

The Archimedes Wave Swing

The last wave energy converter to be considered in this section is
called the Archimedes Wave Swing (AWS), and it, too, is being de-
veloped in Scotland. The company building the AWS is AWS Ocean
Energy Ltd. The AWS operates under somewhat different principles
than the Pelamis or the OWC. The AWS is completely submerged
so that waves pass over it—recall that waves pass under the Pelamis
and crash into the Limpet.
The AWS operates on a pressure difference caused by each passing
wave. Pressure beneath the ocean is a simple function of depth: the
deeper one goes, the higher the pressure one experiences. The relation
between depth and pressure is summarized in the following algebraic
equation, where P represents pressure, w represents the weight of the
water per unit volume, and h represents the distance to the surface:

P = wh

The Archimedes Wave Swing is driven by the pressure changes of passing

waves.
 Wave Power  73

Water does not compress very easily, so at the depths at which the
AWS operates, w can be taken to be a constant. Changes in pressure
are, therefore, proportional to changes in h, the height of the col-
umn of water. Pressure changes caused by passing waves are what
activate the AWS. The larger the wave, the larger the change in pres-
sure, and the more electrical energy the AWS generates.
Here is how the AWS works: The designers of the AWS created a
large cylinder filled with air. (The support structure for the cylinder
is firmly attached to the seafloor.) The cylinder consists of a move-
able upper section and a fixed lower section. The air acts as a sort of
spring that serves to restore the cylinder after it has been compressed
by pressure increases caused by passing waves. When the peak of a
passing wave is positioned above the cylinder it creates a momentary
increase in h, and the resulting increase in h forces the upper part
of the cylinder to descend. The air inside the cylinder compresses
and the pressure increases until the pressure inside the cylinder is
sufficient to balance the pressure outside the cylinder. The crest of
the wave continues moving forward and away from the AWS. The
peak, of course, is followed by the trough of the wave. As the trough
passes over the AWS, the height of the water column above the AWS
takes on its minimum value. Consequently, the pressure on the AWS
is also at its minimum value. The air inside the cylinder, which was
compressed by the peak of the wave, now expands outward, pushing
the upper part of the cylinder upward until the pressure inside the
cylinder balances the pressure outside. The cycle is repeated with
each passing wave. The AWS is a large piston activated by pressure
differences caused by incoming waves. The up-and-down motion of
the piston drives a linear generator, a device that converts the piston’s
motion into electricity. As long as the waves continue to roll in, the
AWS will continue to generate electricity.
Interest in wave energy is driven by several factors: emis-
sions-free electricity, the large amount of energy in ocean waves,
the fact that waves are somewhat more predictable than wind,
74  Wind and Water

and what companies call the lower “visual profile” of wave farms
relative to wind farms. Large wind turbines are visible from far
away because they are so tall. In some places, people object to the
visual “intrusion” of wind towers and work hard to prevent the
wind turbines from being erected. By contrast, all wave energy
converters are less visible than wind turbines, and in the case of
the AWS, once deployed, they are impossible to see without diving
gear. The drawbacks associated with OWC and Pelamis technol-
ogy, described previously, also apply to AWS technology: AWS
technology is dependent on the supply of waves, a factor beyond
the control of the operators, and wave energy may be low when
electricity demand is high. And the cost of AWS technology re-
mains an issue just as it does for the Pelamis and OWC technology.
Electricity produced by these technologies costs more than wind,
and wind costs more than conventional sources. All of which illus-
trates just how difficult it is to replace—rather than simply supple-
ment—more conventional power sources, especially coal, which in
the United States, Germany, and China, three of the world’s largest
economies, remains the most important source of energy for gen-
erating electricity. It is by no means certain that these wave-energy
conversion technologies will ever provide more than a tiny fraction
of the electricity required by most large industrialized countries.
Whether these energy conversion technologies eventually flourish
will depend on how competitive they are with respect to all other
power technologies.
 5

Tidal Power
H arnessing the tides could, in theory, produce enormous quan-
tities of electricity. So far, however, the electrical energy ob-
tained from the tides has been miniscule. The difference between
theory and practice could not be more extreme. This chapter begins
with a short description of tides and then examines two schemes for
converting tidal energy into electrical energy.
Tides result from the deformation of the Earth’s oceans due to
the gravitational tug of the Moon and the Sun. The Moon, because
it is so much closer to Earth, exerts a gravitational pull on Earth’s
oceans that is about 2.2 times as strong as that exerted by the more
massive and more distant Sun. When the Earth, Sun, and Moon all
lie along a straight line, the effects of the Sun and Moon are addi-
tive—that is, their gravitational forces combine to amplify the tides.
These are called spring tides. When the Earth, Sun, and Moon lie
at the corners of a right triangle, the gravitational forces exerted by

75
76  Wind and Water

the Sun and Moon partially cancel one another, and the tides are
reduced in magnitude. These are called neap tides.
If Earth were a smooth sphere covered to a uniform depth by one
vast ocean, tides would be a simple matter to describe: Bulges would
form in the ocean due to the interaction of the ocean with the Sun,
the Moon, and the rotation of the Earth. These bulges would remain
more or less aligned with the Moon, so as Earth rotated about its
axis—causing the Moon to appear to rise and set—the bulges would
move across the planet’s surface. The tides would be higher when
the Earth, Moon, and Sun were aligned and lower when they were
not aligned. It would be that simple.
In practice, tides are far more complicated. The main reason is
the interaction of the oceans with Earth’s topography. The depth of
the ocean varies from several miles to a few inches. The seafloor is
often rough on a small-scale and on a large scale, and is comprised,
in part, of mountains, cliffs, and valleys, all of which serve to chan-
nel the oceans’ water as it flows steadily in response to the forces
acting upon it. In addition to the uneven seafloor, the continents
that bound the oceans have complex shapes. Harbors, reefs, rivers,
and barrier islands, for example, all have localized effects on the
height of the tides. These geographical features can redirect the tidal
motions in their vicinity and cause parts of the ocean to flow as if
they were large rivers. At the Bay of Fundy in Canada, the rising
tide pours 70 billion cubic feet (2 billion m3) into the bay twice each
day before rushing out again. Less dramatic but just as significant,
there are other locations in the ocean where tides are virtually ab-
sent. Tides are, therefore, a local phenomenon. Each potential site is
unique; each site requires its own analysis.
Throughout the years, tidal energy enthusiasts have described
the tremendous amounts of energy associated with these large and
regular water movements and made extravagant claims regarding
the economic potential of harnessing tidal energy. But there is a dif-
ference between theory and practice. The obstacles involved in har-
 Tidal Power  77

nessing tidal energy are formidable. Consequently, tidal energy has

so far made only a tiny contribution to world energy output. That
may soon change, however, as new approaches toward converting
the ebb and flow of the tides into electricity are developed.

The French and Canadian Projects

The technology discussed in this section, a method that employs
dam-like structures called tidal barrages, is hardly new. The Ro-
mans built conceptually similar structures that harnessed tidal
energy to mill grain. The Roman scheme worked as follows: A dam-
like structure was built in a small inlet. The dam was equipped with
a gate called a sluice. As the tide rose, the sluice was opened, allow-
ing water to flow through it and accumulate behind the dam. At
high tide the sluice was closed, trapping the water behind the dam.
As the tide ebbed it created a difference between the height of the
water behind the dam and the height of the sea in front of the dam.
When the tide was low enough—and the hydraulic head was high
enough—the Romans released the water behind the dam in a con-
trolled way. As the water flowed back to the sea it was used to drive
a waterwheel. The waterwheel powered a mill that ground grain.
Unlike a waterwheel driven by a river or stream, the tidal mill only
worked twice each day, because only two tidal cycles occur each
day. (A tidal cycle—high tide to low tide and back to high—occurs
approximately once every 12 hours and 25 minutes.)
The mills were advantageous when they could be constructed
in areas where the local streams moved slowly or where the flow of
the streams was irregular and unpredictable. Tides are highly pre-
dictable and unlike streams are not dependent on rainfall. But the
tides powered the mills only intermittently and at times that were
not always convenient for the mills’ operators. Because the length
of the tidal cycle does not evenly divide the length of the day, the
operating hours of the mill were constantly shifting, sometimes
occurring during the day and sometimes late at night. On balance
78  Wind and Water

Part of the Rance tidal power plant in Brittany, France. Completed in 1967, it
remains by far the largest facility of its kind. (La Rance Tidal Power Plant)

the mills operators must have felt that the benefits outweighed
the costs, because tidal mills were operated throughout Roman
times and some were built and operated throughout the European
Middle Ages. The Domesday Book, for example, the 11th-century
inventory of William the Conqueror’s English holdings, lists at
least one tidal mill.
Scaling up the simple-sounding tidal mill to create a commer-
cially viable electricity generating station is no small challenge.
Numerous proposals for harnessing tidal power were proffered
throughout the first half of the 20th century. But constructing this
 Tidal Power  79

kind of generating station is difficult because it involves undertak-

ing a massive civil-engineering project in the middle of a large tidal
basin. The problems associated with this type of project were first
solved by French engineers with the successful construction of the
La Rance Tidal Barrage, which is located in northern France. Con-
struction began in 1961. It has operated successfully since 1967, and
it remains the world’s only commercial-scale tidal barrage.
At La Rance the difference in height between high and low tides
averages about 28 feet (8.5 m). The tidal variation—the difference
between high and low tide—changes in a predictable way from neap
tide to spring tide, but the difference is always sufficient to generate
power. A large tidal variation is, however, by itself insufficient to
warrant the construction of a tidal power facility. To create a com-
mercial-scale facility, it is also necessary to impound enormous
amounts of water behind the barrage. In the case of La Rance, the
basin behind the barrage has a capacity of about 6.4 billion cubic
feet (180 million m3) of water in an area of 8.5 square miles (22
km2). The basin fills up and empties out twice each day. The La
Rance Tidal Barrage works as follows:
A 2,460-foot (750 m) long barrage was built across the mouth
of the basin effectively separating the basin from the ocean. (The
section of the barrage containing the turbines is 1,000 feet [330 m]
long.) As the tide rises, seawater flows from the ocean through gates
in the barrage and into the basin behind it. At high tide the gates
are closed, impounding the water behind the barrage. The operators
wait for the tide to ebb in order to create a hydraulic head sufficient
to drive the turbines. When there is enough difference between the
water levels in front of and behind the barrage, the water is allowed
to flow through the facility’s 24 turbines, each with a rated capac-
ity of 10 MW for a total capacity of 240 MW. As the water behind
the barrage flows back to the sea, it turns the turbines, which drive
the generators, which produce the electricity. Electricity produc-
tion continues until the water level behind the barrage is reduced
80  Wind and Water

Cutaway view showing how the La Rance facility produces power.

to a level where power production is no longer practicable, at which

point the sequence repeats.
This sequence of events can be modified in two different ways.
First, La Rance can be operated so as to produce electricity as water
flows into the basin as well as out. The incoming water is directed
through the turbines, which are allowed to generate electricity on
the incoming tide. Second, the La Rance facility can be operated in
a way somewhat analogous to a pump storage facility. The station
can pump water behind the dam in order to increase the height
of the hydrodynamic head. As with all pump storage facilities this
requires more electricity than is generated by the turbines as the
impounded water returns to the sea, but as with all pump storage
facilities, if the electricity required to pump the water is obtained
during a period of low demand, and electricity production can be
deferred to a period of high demand, the procedure can be justified
 Tidal Power  81

on environmental and economic grounds. (See the sidebar “Pumped

Storage” on page 30 for a description of this type of technology.)
The site that is always mentioned as a possible location for a La
Rance–type facility is the Bay of Fundy in eastern Canada, the site
of the world’s highest tides. The maximal tidal variation in the Bay
of Fundy is approximately 70 feet (21 m); the bay covers 3,600 square
miles (9,300 km2), and twice each day 70 billion cubic feet (2 billion
m3) of water flows into and out of the bay. The tidal variation varies
from point to point within the bay, but the flow is high essentially
everywhere. Engineers and planners have discussed building a tidal
energy facility here since the 1920s. In 1984, Canada opened a small
tidal barrage facility in the Bay of Fundy. It operates along the same
principles as the La Rance facility but is in every way much more
modest in scope. The Bay of Fundy facility has a capacity of 18 MW.
An even more modest facility was constructed in Russia on the Kola
Peninsula on the Barents Sea in 1968. It has a capacity of 0.4 MW.
As is apparent, tidal barrages share many of the characteristics
of low-head hydropower facilities. In particular, a large tidal barrage
facility has the potential to produce large amounts of electricity in
a predictable way with zero emissions. Even better: Building a tidal
barrage, though very expensive, carries far fewer social costs than
building many large hydropower projects because the construction
of the barrage cannot result in the displacement of a large number
of people. A barrage must be built between the ocean and a large
piece of land that is inundated by the ocean twice each day. No one
lives in tidal basins, and so no one will be displaced by the barrage.
There are also characteristics to tidal barrages that are disad-
vantageous. Although a large tidal barrage can produce a significant
amount of power each day, its output is determined more by the flow
of the tides than by consumer demand. The La Rance facility, for
example, produces enough power for 200,000 homes, but for part of
each day it produces no power whatsoever. The development of this
type of technology is further restricted by the scarcity of suitable
82  Wind and Water

sites, which require a large tidal variation and a large basin in which
to impound the water for later release. Approximately 20 sites have
been identified throughout the world for possible development, but
some are in remote locations devoid of infrastructure or consum-
ers. Tidal barrages can make a difference in the right area, but even
if all identified sites were developed their total contribution to the
world’s electric power production would still be modest.

Turbines without Dams

There is a second technology available for converting tidal energy
into electricity. This technology may be more widely applicable
than tidal barrage technology because it does not rely on barrages
at all. Instead, it extracts energy directly from currents generated
by the tides. Technically, this approach is new. Engineers have
only recently begun to work on the problems associated with
converting the energy of motion of ocean currents into electric-
ity, but conceptually the idea is similar to the much older idea
of the wind turbine. By placing a turbine in an ocean current,
the linear motion of the water can be converted into rotary mo-
tion by a device similar to a wind turbine. Some of these undersea
turbines even look like wind turbines, and as with wind turbines,
the rotary motion of the undersea turbine can be harnessed to
produce electricity. The ocean currents play the same role as air
currents. Before engineers decided to begin searching for techni-
cal solutions to accomplish this goal they had to decide whether
it was worth the effort—that is, they had to evaluate the potential
of ocean currents as an energy source. Deploying these turbines is
difficult and expensive, so is it worth it? How much energy can be
extracted from an ocean current?
The waters of the ocean are in continual motion. Some currents
occur far from shore; some are within a few miles of shore; some
run continuously and some run intermittently. And with today’s
technology some can be exploited to produce power and some can-
 Tidal Power  83

SeaGen marine current turbine awaiting installation in Strangford Lough,

Belfast

not. The costs associated with building and maintaining a deep-sea

turbine farm, for example, are prohibitive. Consequently, most oce-
anic currents cannot be exploited for electrical power production at
present. But currents that run near the shore are a different matter
because they are close enough to land to offer at least the possibility
of economical exploitation. These currents are heavily influenced
by the topography of the coast and the topography of the ocean bot-
tom, and many of these currents are driven, at least in part, by the
tides. (The machines designed to convert the energy of tidal flows
into electricity are called tidal mills.) The drawback of attempting
to harness tidal currents is that they stop and reverse direction four
times every 25 hours. Of course, when the currents stop, they fail to
drive the tidal mills and electricity output drops to zero, but tidal
currents are, at least, predictable, and some are reasonably swift.
For example, in Uldolmok Strait in South Korea, a proposed loca-
tion for a set of tidal mills, the tidal current has a top speed of about
20 feet per second (6 m/s). Tidal currents are not usually this fast,
84  Wind and Water

but in areas where there is a constriction in the channel along which

these currents flow, the velocity increases as the cross-sectional area
of the channel decreases. (The equation that relates the speed of the
current to the cross-sectional area of the channel through which it
passes is equation [4.1].) The best place to locate a mill is, therefore,
in a narrow channel.
The maximum power that can be converted into electricity from
an ocean current is best expressed in a simple algebraic equation.
That equation is as follows:

CdAv3
Pmax = — (5.1)
2
where Pmax represents the maximum power that can be converted
from the current; d represents the density of the water, and v repre-
sents the speed of the flow before it encounters the tidal mill. (The
mill slows the current as it converts some of the current’s energy of
motion into electricity.) The letter A represents the cross-sectional
area swept out by the mill’s rotors, and the letter C represents the
maximum efficiency attainable by the mill—or to put it another
way: C represents the maximum percentage of the current’s power
that can be converted into electrical power. (The quantity dAv3/2
is the power of the current flowing through the circular area swept
out by the blades of the tidal mill.)
Equation (5.1) explains why designers emphasize the impor-
tance of locations with faster currents. The speed in this equation is
cubed. This means that power depends very strongly on speed—so,
for example, if the speed of the current is doubled from v to 2v, the
power of the current jumps by a factor of 8 (2v × 2v × 2v = 8v3).
In particular, a five-mile per hour (8 kph) current possesses double
the power of a four-mile per hour (6 kph) current. Therefore, even
small increases in velocity can lead to substantial changes in the
amount of available power. By contrast, A, the cross-sectional area
of the flow passing through the mill’s blades, has less of an effect
 Tidal Power  85

on the mill’s output in the sense that doubling the cross-sectional

area only doubles the mill’s output. The density of the water, d, is,
of course, completely outside the control of the engineer, so with
respect to producing the maximum possible power, it is important
to build a tidal mill with the largest possible rotor and position it in
the area with the fastest possible current.
During the 1920s at Göttingen University, the German scientist
and engineer Albert Betz developed a theoretical description of how
windmills worked and how much energy they could produce. Betz’s
reasoning, however, also applies to tidal mills. Equation (5.1) was an
important part of his model. Betz’s mathematical model predicted
that C is no larger than 0.59—that is, no more than 59 percent of
the power of a moving current—no matter whether it is an air cur-
rent or a water current—can be converted into electricity. Because
water is very dense, which makes d in equation (5.1) very large, the
amount of power that can be harnessed from tidal mills is, in prin-
ciple, also very large, even in a slowly moving current. (More recent
research indicates that C is less 0.5, although its exact value remains
an open question.)
Tidal mills come in a variety of shapes. The easiest to recognize
are those that look like the more familiar wind turbines. Marine
Current Turbines, a company based in the United Kingdom, oper-
ates a single 300 kW prototype with a rotor that is 36 feet (11 m)
across leading to a cross-sectional area—represented by the letter
A in equation (5.1)—of about 1,000 square feet (95 m2). Because
tidal currents drive this prototype, it is not capable of producing
a steady current of 300 kW. As the tidal currents change direction
they first slow to a stop. This happens four times every 25 hours.
Electrical output drops accordingly, but the output does not remain
at zero for long. As the current reverses direction, the turbine be-
gins to produce electricity again—that is, power is produced as the
tide rises and as it falls. The goal of Marine Current Turbines is to
install larger dual turbine units that will produce 1 MW apiece. As
86  Wind and Water

of 2008, contracts have been signed to deploy units off the coast of
western Canada and Northern Ireland. Other designs very different
from the Marine Current Turbine design have also been developed.
One idea of particular interest incorporates a helical turbine, which
in 2001 was patented by the Russian engineer Alexander M. Gorlov.
(A helical turbine is mounted on a vertical axis and resembles an
eggbeater.) Gorlov’s turbine is said to be capable of converting 35
percent of the power of the current, which is quite good compared
to other, more standard designs.
The different designs currently in place or under development
have a number of commonalities so it is important not to overem-
phasize differences. All are located (or will be located) below the
surface of the ocean in locations where the current runs relatively
quickly. All designs convert a fraction of the energy of motion of
the tidal currents into electrical energy and send it back to shore
through a submarine cable. They produce no emissions. They do
not require a barrage to function, and their environmental impact
is less than that of a tidal barrage.
But it cannot be asserted that these devices have no environ-
mental impact. It is not clear, for example, what effect these huge
spinning blades will have on marine life, especially whales and
other large marine organisms. Furthermore, a line of large tidal
mills across the mouth of a bay would certainly change the flow
rate of the tides. The environmental effects of diminishing the tidal
currents in a particular location depend on considerations unique
to each location. These facts simply serve to emphasize the fact that
large-scale power production cannot be done without affecting the
environment.
A final commonality worth discussing here is that all of these
schemes are substantially more expensive than wind, which, as pre-
viously mentioned, remains more expensive than, for example, coal.
Tidal mills will, presumably, eventually be subject to the same laws
of the marketplace as other types of power generation technologies.
 Tidal Power  87

Spokespersons for the companies that manufacture these devices

always stress that it is early yet, and that as production volumes
and manufacturing experience increase, prices will come down.
This is certainly true, but whether they come down enough to be
competitive with other technologies without generous government
subsidies is not clear at the present time.
 6

Heat Engines
T his chapter describes how heat from the ocean has been har-
nessed to do work. But before one can appreciate the advantages
and disadvantages of this technology, it helps to know a little about
heat engines in general.

The Theory of Heat Engines

Heat engines are machines that convert heat into work. The first
heat engines, which were built more than 300 years ago, were steam
engines used to drive water pumps. Powered by coal, they were
used to pump water out of coal mines. In the 19th century, heat en-
gines—also powered by coal—provided the motive force for trains
and ships. Although the transportation sector still depends almost
exclusively on heat engines—jet engines and internal combustion
engines are both examples of this type of energy conversion tech-
nology—petroleum is now the fuel of choice. But in many countries,

88
 Heat Engines  89

Nicolas-Léonard-Sadi Carnot. He
discovered the principles governing
the operations of heat engines. (AIP
Emilio Segrè Visual Archives)

coal-fired steam engines remain the preferred method for generat-

ing electricity. The United States, China, Germany, and India, for
example, are all heavily reliant on coal technology to power their
national grids. Some countries also use natural gas–fired heat en-
gines to produce electricity, and some nations still find it necessary
to burn oil to meet some of their electricity demands.
Fossil fuels are not, however, the only primary energy source
used to supply heat engines with the thermal energy they need to
operate. Nuclear power plants, which harness the energy within the
nuclei of atoms, are also heat engines. In the United States nuclear
power is responsible for about 20 percent of all electricity produc-
tion. In France the figure is closer to 75 percent. Worldwide, most
electricity is generated by heat engines of one type or another.
One characteristic shared by all of the different types of heat
engines mentioned in the preceding paragraphs is that they have
fairly high operating temperatures. The exact operating tempera-
ture for a particular engine depends upon the type of engine under
consideration, the type of fuel it uses, and the way that the engine is
90  Wind and Water

operated. Nevertheless, all of these engines operate at temperatures

that are high enough to be hazardous to humans.
But a high operating temperature is not a requirement for a
heat engine because, strictly speaking, heat engines only require
a temperature difference to operate. Just as waterwheels operate
between two different water levels, heat engines operate between
two regions of different temperatures. With respect to waterwheels,
as water flows from the higher level to the lower level, the water-
wheel converts some of the water’s energy of motion into work.
With respect to heat engines, heat will flow from a region of higher
temperature to a region of a lower temperature, and as it flows, the
engine converts some of that thermal energy into work. This anal-
ogy between waterwheels and heat engines was proposed by one
of the 19th-century’s most insightful thinkers, the French scientist,
Nicolas-Léonard-Sadi Carnot (1796–1832) as he attempted to ex-
plain how heat engines generate power.
Carnot recognized that the efficiency of an engine is determined
solely by the percentage of “flowing” thermal energy that is convert-
ed into work. An engine that converted all of the thermal energy
that it produced into work, for example, would be 100 percent ef-
ficient. If it converted half of its thermal energy into work, it would
be only 50 percent efficient. Carnot discovered that each engine has
a maximal efficiency that it can attain, and that maximum value is
always less than 100 percent. This upper limit on its efficiency does
not depend on the details of the engine’s construction or the fuel it
uses. Nor does it depend on how the engine is operated. Instead, the
maximal efficiency is determined solely by the temperatures of the
two regions between which the engine operates. The lower tempera-
ture region is usually, but not always, taken to be the temperature
of the outside environment; the higher temperature is the operating
temperature of the engine. The bigger the temperature difference
between these two regions, the more efficiently the engine can, in
principle, operate. Again: Efficiency is the percentage of thermal
 Heat Engines  91

energy moving from the higher temperature region to the lower

temperature region that is converted into work.
This upper limit to an engine’s efficiency is easily summarized
by an algebraic equation:
Th – Tl
E =— (6.1)
Th
where Th represents the (higher) operating temperature of the
engine, and Tl represents the (lower) temperature of the environ-
ment, and both temperatures are measured in degrees kelvin. (A
temperature difference of one degree kelvin equals a temperature
difference of one degree Celsius, but 0°C equals 273.15°K.) The let-
ter E represents the maximal efficiency of the engine. For any given
upper temperature, Th, and lower temperature, Tl , it is easy to make
an engine that is less than maximally efficient, but it is impossible to
make an engine operating between these two temperatures whose
efficiency exceeds this upper limit.
To see how this works in practice, consider an engine whose
operating temperature, measured in degrees kelvin, is 1,200°K
(1,700°F, or about 927°C), and suppose that the temperature of
the environment is 300°K (80°F, or about 27°C). This engine can-
not convert more than 75 percent of the thermal energy produced
within the engine into work (0.75 = (1200 - 300)/1200). By contrast,
if the engine’s operating temperature is 315°K (108°F or 42°C), and
the temperature of the environment is 300°K (80°F or 27°C), the
engine can, at best, operate at about 4.8 percent efficiency, which is
another way of saying that it can convert no more than 4.8 percent
of the available thermal energy into work; the rest, which in this
case is 95.2 percent of the total, is “waste” in the sense that it cannot
be converted into work.
Equation (6.1) indicates why engineers design heat engines that
operate at temperatures that are as high as possible. An engine with
a higher operating temperature can, in principle, convert more of
92  Wind and Water

its thermal energy into work. Therefore higher temperature engines

will, in theory, waste less because they will require less thermal en-
ergy to produce the same amount of work. These ideas are extremely
important when evaluating proposals to tap the enormous supplies
of thermal energy contained in Earth’s oceans.

Practical Applications
A small fraction of the thermal energy of the world’s oceans can,
in principle, be converted into electricity by heat engines that oper-
ate between the warmer surface waters and the cooler depths. The
machines designed for this purpose are collectively called ocean
thermal energy converters (OTECs). The first proposal to build an
OTEC dates back to 1881 when the French scientist Jacques-­Arsène
d’Arsonval (1851–1940) suggested harnessing the temperature
difference between the warmer upper layers of the ocean and the
cooler lower regions to do work. D’Arsonval was a creative scientist
who was interested in both physics and biology and the interplay
between the two fields. He was also an inspirational teacher. His
student Georges Claude (1870–1960) was the first to attempt to de-
ploy an OTEC device. He tried twice—the first time at Matanzas
Bay, Cuba, and the second time in a cargo vessel moored near the
Brazilian coast. These machines were purely experimental in char-
acter—they required more power to operate than they produced.
Experience has shown that building a productive and economical
OTEC plant is a difficult technical challenge even today.
At its most basic level, OTEC plants run on solar energy. The
world’s oceans act as enormous, if not especially efficient, solar col-
lectors. As the Sun shines on the oceans, some of the light from
the Sun is reflected at the surface, but enough of the Sun’s energy
is absorbed by the water to raise the temperature of the water in a
predictable and potentially useful way. The effect, not surprisingly,
is especially strong in tropical regions. Absorbed light raises the
temperature of the water that absorbed it, and this increase in tem-
perature causes a small increase in volume. As a consequence, water
 Heat Engines  93

These are parts of a

­machine for generating
electricity based on the
differences in temperature
between the sea surface
and great depth. Devised
by Georges Claude in 1926,
this device was a fore-
runner of modern ocean
thermal energy conversion
(OTEC) projects. (Oceano-
graphic ­Museum of Monaco
and National Oceanic and
­Atmospheric Administration/
Department of Commerce)

near the surface of the sea is slightly less dense than water deeper in
the ocean; the warmer, less dense water floats on top of the colder,
denser water, and there is very little mixing between the lower and
upper regions. Because sunlight does not penetrate very deep into
the ocean, the cooler water is “only” about 3,000 feet (1,000 m) from
the warmer surface. In the tropics the temperature of the water near
the surface of the ocean is about 77°F (25°C or 298°K). The lower
layer has a temperature of about 40°F (5°C or 278°K).
As indicated in equation (6.1), heat engines operating between
these two temperatures cannot hope to be more than about 7 per-
cent efficient, and practical engines would operate at still lower
(continued on page 96)
 94  Wind and Water

Ocean Thermal Energy

Conversion Tests

O TEC technology is slowly evolving, but even after decades of work

only a few nations have deployed demonstration plants. Japan and
the United States were the first to establish significant OTEC research pro-
grams. For about one year, beginning in 1982, a group of Japanese com-
panies operated a demonstration plant on the island nation of Nauru. The
plant produced about 100 kW of electricity, about two-thirds of which was
used to run the plant’s pumps and other equipment. The United States
performed its OTEC experiments in Hawaii, the best known of which was
a demonstration plant that operated from 1993 until 1998. It produced up
to 225 kW, about 60 percent of which was used to run the plant.
These plants tested OTEC designs, and the results were not entirely
satisfactory and did not lead to commercial-scale plants. The United States
Department of Energy (DOE) even ceased funding research into OTEC.
During the early years of the 21st century, India built a one-MW OTEC
demonstration plant. But that project suffered from cost overruns, and
the plant, which was designed to operate on the open ocean, suffered
damage in rough seas. Funding was canceled. Yet interest in developing
OTEC technology also continues in India.
The attraction of OTEC rests on two observations. The first is easy
to appreciate: The fuel, which consists of warm water, is free. The
second reason OTEC continues to attract attention depends on the
observation that there is an enormous amount of energy available in
the temperature difference between the upper and lower regions of
tropical oceans—enough energy, in theory, to power a nation. And yet
not one of the three countries that invested in demonstration plants
has attempted to build a commercial-scale OTEC power plant. By con-
trast, for many years, all three nations have successfully operated com-
mercial nuclear power plants, which to many observers may appear to
be a technologically more difficult undertaking. Why not at least one
OTEC power plant?
 Heat Engines  95

OTEC power plants are, for several reasons, much harder to build
than, for example, a nuclear power plant. First, because an OTEC plant
operates between thermal reservoirs that are at almost the same tem-
perature, they are, under the best of circumstances, able to convert
into work no more than 6 or 7 percent of the thermal energy that they
produce. (Recall that the larger the temperature difference, the larger
the percentage of thermal energy that can, in theory, be converted into
work.) As a consequence, OTEC plants must operate very near their op-
timum level of efficiency in order to accomplish anything at all. By con-
trast, heat engines that operate between larger temperature drops have
a larger percentage of energy that they can convert into work. For high
temperature engines, inefficiencies are less of a problem. (Early steam
engines, for example, wasted almost all of the thermal energy they
produced yet usually managed to perform the functions for which they
were designed.) Second, as an OTEC plant grows in size, its power output
increases faster than its own internal power requirements. A small OTEC
plant will consume more power than it produces, while a large plant will
consume considerably less. Consequently, the only profitable OTEC plant
would be a very large OTEC plant, and the larger the better. This increas-
es the risk to investors, who must decide between building a large and
expensive plant or none at all. Investment risk would be especially high
during the first few generations of OTEC plant design, when engineers
and operators would still be learning how to make such a plant profit-
able. Finally, every OTEC plant operates with some—or in the case of the
Indian plant, all—of its equipment in or on the sea, an environment that
is corrosive and sometimes violent. It is a difficult place in which to locate
a large power plant of any sort.
With respect to OTEC technology there is a huge difference between
theory and practice. It is not known where or when the first commercial-
scale OTEC facility will be built.
96  Wind and Water

(continued from page 93)

­efficiencies. On the one hand, one could say that such low efficien-
cies are terrible, and an engine that is able to utilize no more than
7 percent of the heat flowing through it is hardly worth develop-
ing. For many applications, this is certainly true, but the oceans
are so huge and hold so much thermal energy that even if one were
only able to economically harness a tiny fraction of their thermal
reserves the implications would be enormous. And, of course, the
fuel, solar energy, is free and abundant.
Several different designs for OTECs have been proposed, and
various small-scale prototypes have already been built but with
limited success. One representative design will be considered here.
It works as follows: A very long, large pipe extends deep into the
ocean in order to draw cool water from the depths into the power
plant. Another pipe is used to draw warm surface water into the
plant. Inside the plant, situated between the supply of cool water
and the supply of warm water, is a third material called the working
fluid. The working fluid is chosen so that it will boil when warmed
by the warm water and condense when cooled by the cool water.
The working fluid does not come into physical contact with either
water source. Instead, only heat is exchanged between water and
working fluid via radiator-like devices called heat exchangers.
First the working fluid acquires heat from the warmer water via
a heat exchanger. As its temperature increases, the working fluid
turns to vapor. The expanding vapor is directed against the blades
of a specially designed turbine, causing it to spin. The turbine is
connected to a generator, which produces electricity. Once past the
turbine, the working fluid exchanges heat with the cooler water—
again through a heat exchanger—this time giving up some of the
heat it had earlier acquired from the warmer water. As the working
fluid cools, it condenses. The working fluid is then pumped back to
the first heat exchanger in order to acquire heat from fresh warm
water. The cycle repeats.
 Heat Engines  97

© Infobase Publishing

OTEC process diagram. The heat exchanger used to warm the working fluid
is called the evaporator. The heat exchanger used to cool the working fluid is
called the condenser.

There are a number of technical problems associated with build-

ing a large-scale OTEC generating station, but the fundamental
problem is the low efficiency of the unit. Because a practical OTEC
plant will only be able to convert into work a very small percentage
of the heat carried by the warm water, it requires enormous volumes
of warm water to produce useful amounts of electricity. And the
same statements hold true of the cool water: Very large volumes
of cool water must be brought up from the deep to cool the work-
ing fluid. Large-scale OTEC plants will require huge pipes, huge
pumps, and huge heat exchangers to facilitate the rapid transfer of
significant amounts of heat. Everything will have to operate very
efficiently in order to produce commercially meaningful amounts
98  Wind and Water

of electricity, and this raises costs and additional technical issues.

Furthermore, the energy needed to operate on this scale must be
drawn from the power output of the plant. Only what is left after the
plant’s energy needs have been met is available for sale.
The energy crises that occurred in the United States and other
Western nations during the 1970s as the result of rapid oil price in-
creases caused researchers in the United States and elsewhere to focus
on the possibility of developing OTEC plants. (At the time, it was
common practice to burn oil to produce electricity.) Research into
OTEC continued throughout the late 1970s and early 1980s. OTEC
technologies were attractive to the researchers of the time because the
“fuel”—solar energy—was perceived as free and a viable alternative to
high-priced oil. OTEC technology offered the promise of enormous
quantities of electricity for as long as the Sun shone. But the technical
problems encountered by engineers attempting to produce a work-
ing, economically competitive OTEC plant were insurmountable us-
ing 1970s technology. As the price of petroleum stabilized and began
to diminish, government support for OTEC research evaporated. In
1984 the U.S. Department of Energy spent just $8.2 million on OTEC
research, less than one-tenth of what it had spent on OTEC in 1978.
Today, the prices of natural gas and oil have once again become
volatile. Concerns about the effects of greenhouse gas emissions
have further focused the attention of government research facilities
on finding less expensive and less polluting technologies with more
predictable costs. Some engineers are renewing efforts to develop
OTEC technologies. The initial market would be tropical islands,
locations where fossil fuels must be imported at a premium and
where the temperature difference between the ocean’s upper and
lower layers is stable and comparatively large. The long-term goal,
however, is the harnessing of the oceans’ temperature difference to
produce electricity on a scale sufficient to supply larger markets.
The day that OTEC becomes commercially viable—if that day ever
arrives—is still many years in the future.
 7
The Role of Government
in Promoting New
Technologies
E very power generation technology described in this book is
(or has been) heavily dependent on government subsidies.
Large-scale conventional hydropower has often involved the dam-
ming of rivers, the flooding of valleys, and sometimes the forced
displacement of many people, all activities that are difficult to ac-
complish without enthusiastic government support. Once a project
is completed, however, conventional hydropower is dependable
and often competitive with all other technologies. (Although many
hydropower facilities could be operated profitably, sometimes, as
discussed in chapter 3, governments decide to continue to subsidize
the electricity produced at these facilities in order to spur economic
development.) There are other technologies—wind is the most
prominent example—that require subsidies if the facilities are to be
built, and they continue to require subsidies even after a project is
completed because the price of electricity produced by these power

99
100  Wind and Water

U.S. Department of Energy headquarters (DOE)

sources is not yet competitive with that of other more conventional

power sources.
To be fair, there are many other countries where conventional
power sources—a term that is shorthand for coal, natural gas, con-
ventional hydropower, and nuclear, the technologies that produce
the vast bulk of the world’s electricity—also receive government
subsidies of various sorts. In the United States, for example, where
roughly half of all the nation’s electricity is produced by burning
coal, many experts argue that the full costs of burning coal are not
reflected in the price U.S. consumers are charged for the electricity
these plants produce. These costs include but are not limited to the
environmental and social costs of extracting coal as well as the sig-
nificant environmental costs associated with burning it. The failure
to create a system that takes into account the environmental and
The Role of Government in Promoting New Technologies  101

social costs associated with coal use, they argue, is the most signifi-
cant governmental subsidy. They argue that the government should
create a price structure for electricity produced by coal-burning
plants that would better reflect the actual costs of that technology.
At present, the owners of coal plants retain the profits for them-
selves, but they pass along most of the environmental costs for all to
share. That, at least, is the perception of many. Nations have found
it difficult to act on this perception, however, because there is no
generally agreed upon method of assessing environmental costs,
and because the inexpensive and reliable electricity coal-burning
plants now produce is so beneficial to so many.
And what is true of coal is, in fact, true of every energy pro-
duction technology. How should the value of inexpensive, reliable
electricity production be assessed relative to the values of reduced
greenhouse gas emissions, energy security, and so on? What is the
value of any power-production technology relative to any other
power-production technology? And how can energy markets be de-
signed that incorporate these values? There are no generally agreed
upon answers to any of these questions.
Despite a lack of consensus on the details, the case for power-
production technologies that enhance energy security and reduce
greenhouse gas emissions is strong enough that many governments
have already undertaken programs to encourage the immediate
deployment of alternative technologies. But in the current energy
markets, investors have usually shown little interest in implement-
ing these alternatives without substantial government subsidies.
Some governments have, therefore, created policies to stimulate
interest by essentially guaranteeing the profits of those who in-
vest in government-sanctioned methods of electricity production.
The results of these governmental programs have varied widely.
It is, therefore, worthwhile to see what types of programs have
worked and what types have not. But rather than try to account
for all of the programs for each of the technologies described in
102  Wind and Water

this volume, the emphasis in this chapter is on the strategies used

by nations to encourage the deployment of wind power. Because
with the exception of conventional hydropower, wind is the most
widely deployed of the technologies described in this volume, and
it has long benefited from the largesse of several countries. More
recently, some governments have begun to apply wind power–like
subsidies to encourage the development of other power produc-
tion technologies, including those technologies described in the
preceding chapters. The policies of the United States, Denmark,
and Germany, three of the world’s biggest wind-power producers,
are of most interest. (The history, technology, and environmen-
tal implications of wind are discussed in chapters 8, 9, and 10,
respectively.)

The United States: Creating Supply

and Demand
In the United States, the federal government supports wind en-
ergy in three ways: direct subsidies, regulations to encourage wind
energy, and taxpayer-subsidized research and development into
wind-energy technologies. U.S. attempts to develop and deploy
wind-energy technology began in earnest with the energy crises of
the 1970s, which were marked by rapid increases in the worldwide
cost of oil and, in 1973, a brief embargo on oil shipments to the
United States by some oil exporters as retaliation against U.S. sup-
port for Israel during the 1973 war between Israel and several of its
neighbors. (Within the United States, some states also developed
their own programs to encourage the development of wind energy.
Most notable among these are the efforts of the state of California.
In fact, so successful was California in developing wind capacity
that during the 1980s most of the electricity produced via wind en-
ergy throughout the world was produced within California, despite
the fact that most of the best U.S. sites for wind farms are located in
states other than California.)
The Role of Government in Promoting New Technologies  103

Important early federal support for wind energy was contained

in the Public Utility Regulatory Policies Act (PURPA) of 1978.
PURPA, which was part of the National Energy Act of 1978, re-
quired utilities to buy electricity from nonutility power producers
that produced electricity using renewable energy technologies, pro-
vided that the capacity of these facilities did not exceed 80 MW. In
a general way the law even provided guidelines as to the price that
the utilities had to pay: Utilities were required to buy the power at
a rate equal to the amount of money that they would have spent
producing the power themselves. This was called the “avoided cost”
of the power. Avoided cost was, however, a flexible standard and dif-
ferent states interpreted the standard in different ways. California
and New York interpreted the standard in ways that favored the
independent power producers, while other states used interpreta-
tions that favored the utilities. (The differences among states in the
definition of avoided costs were large, and in 1995 the Federal En-
ergy Regulatory Commission established national standards. These
standards required that some states lower the rates at which utili-
ties were required to purchase renewable energy, and some states
were required to raise the rates that their utilities paid.) But when
the United States passed PURPA, wind energy technology was still
too primitive—in the sense that electricity from the wind was too
expensive to produce—to flourish with this level of support. Wind
technology of the 1980s was too inefficient to attract investors even
with the federal subsidy. Large-scale wind farms were only found in
California, which adopted additional subsidies generous enough to
attract investors for wind farms that produced only modest amounts
of very expensive electricity.
In 1992, Congress passed the Energy Policy Act (EPAct),
which provided a subsidy in the form of a production credit of 1.5
cents per kilowatt-hour of power produced during the first 10 years
of operation of each wind turbine. Private companies received a
production tax credit—that is, money off their tax bills—and ­public
104  Wind and Water

power producers received a similar sort of subsidy called a renew-

able energy production incentive. (The credit has been revised
upward and in 2008 is worth two cents per kilowatt-hour.) After a
slow start, the credit has had an important effect on the installation
of wind power capacity. The effect is most fully illustrated by what
has happened each time this subsidy has been allowed to expire.
This has occurred in the years 1999, 2001, and 2003. In each of
the years following the expiration of the credit—that is, in 2000,
2002, and 2004—installation of new wind facilities plummeted.
When the subsidy was reestablished, the installation of new facili-
ties surged. To get an idea of the scale of this one program, con-
sider that in 2005 the production credit for wind-power producers
amounted to $330 million, and the Congressional Joint Committee
on Taxation estimates that maintaining the credit will cost $2.8
billion from 2005 until 2015. EPAct has had the most effect of all
direct federal subsidies aimed at U.S. wind power producers. As of
this writing it is set to expire at the end of 2008, but if past practices
are any guide, it will probably be renewed shortly before it expires
or shortly thereafter.
The final type of subsidy considered here is government-spon-
sored research. The federal government has conducted research into
wind-power technology for several decades. Much of this research
has been conducted within the Department of Energy (DOE). Re-
search into wind energy began in the 1970s, a time when the U.S.
intensively studied many so-called nonconventional energy sources.
In 1979, the DOE spent almost $60 million on wind-power research.
But a few years later the price of petroleum declined and so did fed-
eral support for wind power research. By 1982, only $16.6 million
was allocated within the DOE for wind-power research. Despite the
sporadic nature of the funding, engineers have been very successful
in lowering the cost of converting wind energy to electrical ener-
gy—from roughly 50 cents per kilowatt-hour in the 1970s to about
five cents per kilowatt-hour today, a tremendous accomplishment.
The Role of Government in Promoting New Technologies  105

Wind-power research continues to attract the attention of the

federal government. Research occurs in three ways. First, the DOE
conducts its own program of research; second, the DOE sponsors
university research; and third, DOE researchers form partnerships
with industry and academia. More generally, the goal of DOE pro-
grams is to pursue research that would probably not occur—or at
least it would not be conducted with the same intensity—if the re-
search were conducted solely by private industry. It is worth noting
that although the DOE negotiates royalty-sharing agreements with
industry partners in the event that a particular research program
leads to a commercial success, the DOE has not required renewable
energy companies to pay any royalties when joint research has led
to profitable inventions. DOE research subsidies have created a fast-
growing industry but one that remains highly dependent on various
forms of government support.
The principal barriers to the development of wind farms within
the United States are not directly addressed by federal programs.
The first barrier to the development of some sites is local opposition.
The reasons expressed for opposition to the development of particu-
lar sites are varied, but many of the objections are surrogates for the
perceived ugliness of wind turbines. (See chapter 10 for a descrip-
tion of the opposition to the development of the United States’ first
offshore wind-power site.) The federal government has no policy to
deal with this type of opposition, but other nations have developed
strategies to deal with local objections. See, for example, the Danish
model described in the next section.
Another important difficulty in developing more remote wind-
power sites is associated with the difficulty of constructing high-
voltage transmission lines to connect producers with consumers.
Wind farms require large tracts of unpopulated land on which to
build the turbines. These sites must be free from trees and buildings
and are often found far from the cities that are their main markets.
(continued on page 110)
 106  Wind and Water

An Interview with
Dr. Stan Bull on
Research at the NREL

D r. Stanley R. (Stan) Bull is the Associate Director for Science and

Technology for the National Renewable Energy Laboratory, and in
this capacity he directs the research and development programs at the
Laboratory. He is also Vice President of the Midwest Research Institute. An
accomplished scientist—he has authored approximately 100 papers—of
more than 35 years, he has research experience in energy and related
applications including renewable energy, energy efficiency, bioenergy,
medical systems, nondestructive testing, and transportation systems, all
in addition to his work of planning and evaluating technical programs.
Here, he shares his ideas about the processes by which research projects
are chosen and evaluated, processes that are critical for scientific prog-
ress. (This interview took place July 9, 2007.)

Q: The National Renewable

Energy Laboratory [NREL] is in-
volved in research into a wide
variety of energy conversion
technologies—
A: Correct.
Q: For example, wind, hydro-
power, geothermal, biomass,
photovoltaic. But NREL doesn’t
concern itself with OTEC [ocean
thermal energy conversion],
and I don’t think that NREL has
ever investigated wave-energy
conversion technologies.
A: Let me clarify that. First, we
don’t research hydroelectric.
Stan Bull (Stan Bull)
We have not done research
The Role of Government in Promoting New Technologies  107

in that area. We are funded by the U.S. Department of Energy [DOE].

So what we do is dictated by the programs that are funded at the
Department of Energy (DOE). In the ’80s we were involved in ocean
research—primarily ocean thermal [energy conversion]—although we
were involved in wave energy and other forms of ocean energy. But in
terms of what we did in the laboratory, it was primarily ocean thermal.
That research ended because eventually DOE and Congress chose not
to fund that program any longer as funding for renewable energy was
reduced. For a good long while—approximately 15 years—we have not
done research into ocean technologies. We are now starting an ocean
program again. There is a modest amount of money for that this year,
and that would include both wave and underwater-current research,
but not ocean thermal.
Q: How are these technologies chosen? What is the mechanism?
A: This a hard question to answer.
We do what the Department of Energy funds us to do. The
Department of Energy does the research that Congress appropriates
funds for them to do. What does the federal process look like, and how
do we play in that? We support the Department of Energy in developing
multiyear program plans. There are a variety of plans. There are things
called roadmaps, multiyear plans, and annual plans. These are reviewed
by outside experts. Once a program gets going, there are generally in-
cremental changes to it. Things may happen that cause the increments
to be bigger or sometimes smaller.
There is the annual budget process that Congress goes through and
that federal agencies and the administration go through with Congress.
They submit a prepared budget and that gets submitted to Congress in
the State of the Union every year—you know all this, I’m sure.
Q: Sure.

(continues)
 108  Wind and Water

(continued)
A: That’s when the budget process is rolled out. Congress has hearings
that involve the Department of Energy folks and lab people like us—some
of our people and myself are invited to testify—and eventually Congress
appropriates, and generally it’s with a fair amount of definition about
where the dollars are appropriated. Those dollars go to the Department of
Energy, and the Department of Energy makes decisions on what fraction
goes to our laboratory and what fraction goes out by way of solicitations,
where they have private industries compete for awards or universities
compete. One of our goals is to support the Department of Energy in
an integrating role so that we can be an objective adviser to them and
give them good, sound technical input relative to the direction that the
program should go and what the research priorities should be. It’s a very
complex enterprise that is in play.
Q: The Department of Energy began research into wind power in the
1970s—
A: That’s correct. A lot of research priorities are driven by world events—
the world oil crisis in 1973 drove the U.S. government to seek alternatives.
You’ve also seen that phenomenon in more recent times when the price
of oil went up into the $70 per barrel range. One of the major points of
emphasis of the Congress and the administration today is, “What are
some ways to displace our need for foreign oil?” Or oil in general—but
our imports are larger than our domestic production today. And so the
major emphasis is to determine some of the alternatives, some of the
major alternatives. How do we get there? How do we accelerate that?
So things like the price of oil drives additional investment in renewable
energy—in alternative energy.
A vitally important question is energy efficiency. How can you use
our energy sources efficiently so that you don’t need to have more and
more of it?
Q: The researchers in the Department of Energy—to go back to wind
power as an example—reduced the price of a kilowatt-hour of electricity
The Role of Government in Promoting New Technologies  109

from the time that they started in the 1970s until today by about an order
of magnitude—
A: That’s a good summary. The price has been reduced by R & D, and with
the help of industry development, by approximately a factor of 10.
Q: How difficult is it to maintain a research effort over such a long time
frame?
A: It’s a challenge. One of the most important things with a research
effort is that it be steady, not cyclic. It takes a while to build a capability.
The human resource, the people, are the most important part of that. It
takes awhile to find the best people in this business, to attract them, and
get them into a situation with equipment and facilities where they can
be effective. So if the budget goes up and down, the best talent are the
first to find alternatives and leave. So you lose the strength of some of
your best capability over time. So the most important thing is a steady
budget.
But as you start to have success, you begin to hear the mantra that
“this is a mature technology, so we can reduce our R & D investment.” If
you think about it, the U.S. government still invests in coal technology, and
how old is coal technology? It’s very old. Nuclear energy is more than 50
years old, and we continue to invest in that. So these technologies at NREL,
while they may be maturing, they still can mature substantially more—
Q: When NREL undertakes a research project with such a long time frame,
how do you judge success in what is, from a practical point of view, almost
open-ended research?
A: First of all, virtually all of our research is applied research. Actually, I
would call it technology development. Sometimes we say research and
development, sometimes we use the phrase research, development, and
demonstration, and sometimes we say research, development, demon-
stration, and deployment, because we partner and work very closely with

(continues)
 110  Wind and Water

(continued )
industry. Everything we do is intended for industry. We don’t do anything
that we believe will have an outcome not intended for industry. So our
ultimate measure of success is, “Has it been useful to industry?” And, of
course, not everything we do ends up being adopted by industry, because
you do go down some blind alleys. You do go places that don’t work out,
but you learn from those experiences. That puts you on a different path,
and you go down that path and are successful. But we judge our success
by, “Was it useful for industry?” and “Is it going to be useful for industry?”
and “Did it eventually become so?”
Sometimes it takes longer than you think it should. We’ll develop a
technology, and it doesn’t seem to go anywhere, but then 10 years later
there is another improvement. Something else facilitates or enables that.
Or the price of oil doubles. There are a lot of factors that can ultimately
help make it successful.
Q: On balance, how satisfied are you with the mechanisms by which re-
search projects are chosen and pursued at NREL?
A: That’s a hard question for me to answer. I would say that we are not
always entirely happy with all of the directions that we need to follow, but
over time we are able to adjust our research portfolio such that we really
do believe that frankly we are on a pretty good track most of the time.
Q: I very much appreciate your time and your expertise. Thank you.

(continued from page 105)

It is, therefore, necessary to build corridors for high-voltage trans-
mission lines to connect power producers with the markets that
they serve. These high-voltage lines are often controversial because
their construction requires that many individuals surrender some
of their property to make construction possible. If the high-voltage
lines cannot be constructed there is little reason to develop more
The Role of Government in Promoting New Technologies  111

wind farms, as the producers will have no way to bring their prod-
uct to market.

Denmark: Creating Supply, Demand,

and Goodwill
Denmark is a country of about 16,639 square miles (43,094 km2)
and 5.3 million people, and it has long been a pioneer in wind pow-
er. Some of the earliest research into wind turbines was conducted
in Denmark by Poul la Cour (1846–1908), whose contributions are
summarized in chapter 8. Today, Danish companies are some of the
most successful in the wind-power market, and Danish-built tur-
bines can be found in countries around the world. Popular support
for renewable energy, especially wind, is very strong in Denmark.
This enthusiasm is reflected in government policies, and it is even
fair to say that some of the enthusiasm is a result of those policies.
Given the right weather, Denmark can generate as much as 25
percent of its electric power needs with nonconventional power
sources, the highest proportion of any developed nation. Most of
this power comes from the wind. It is worth noting, however, that
fossil fuel plants continue to comprise about three-fourths of Den-
mark’s generating capacity.
Denmark began passing legislation to encourage the develop-
ment of renewable power during the 1970s in response to the oil
crises of that time. In 1979, the government established a program
to encourage investment into a variety of renewable technologies,
including wind. The incentive was a 30 percent investment subsidy,
a program designed to reduce the cost of installing wind turbines.
The investment credit diminished steadily over the following 10
years, and in 1989, when it was at 10 percent, it was eliminated.
In 1981, Denmark began offering a production subsidy to reward
the production of electrical power rather than the construction of
turbines. By 1992, the motivation for renewable energy subsidies
had changed from energy security to the desire to lessen the en-
112  Wind and Water

vironmental impact of electricity production. This change in mo-

tivation was reflected in two pieces of legislation that were passed
in that year. First, Denmark instituted a carbon tax—that is, fossil
fuel technologies, which produce greenhouse gases as a by-prod-
uct, were taxed according to the amount of carbon they vented to
the atmosphere—and so the Danish government made the cost of
wind energy more competitive relative to fossil fuel technology by
making fossil fuel technology more expensive relative to wind. Sec-
ond, Denmark required utilities to purchase renewable energy at a
premium price, a price that was higher than what they would have
paid for fossil fuel–based electricity. Finally, Denmark established a
production credit, a direct payment from government to producer
of about 2.8 cents per kilowatt-hour. This payment was gradually
reduced, but when these measures were first instituted, independent
producers of wind energy benefited from a package of subsidies that
totaled approximately 4.4 cents per kilowatt-hour. Compared to the
United States, this was an extremely generous set of subsidies.
The Danish government also promoted the creation of a wind
turbine manufacturing industry. In 1990, the government guaran-
teed long-term financing of large-scale wind projects provided that
the projects used Danish manufactured turbines.
Unlike the United States, which has no real mechanism for en-
franchising potential local opposition to the siting of wind farms,
Denmark has worked hard to make wind power attractive to its citi-
zenry. This was accomplished in two ways. First, beginning in 1994,
all municipalities were required to create plans that would include
possible sites for the installation of wind turbines. The law did not
require each municipality to site a given number of turbines, but
the legislation did help to distribute among the broader population
the burden of accommodating wind turbines while simultaneously
allowing local governments a voice in how wind energy would be
implemented within their jurisdictions. In addition to this direc-
tive, Denmark also passed legislation that facilitated the formation
The Role of Government in Promoting New Technologies  113

of wind-turbine cooperatives. Early legislation encouraged many

small investors to buy shares in local wind projects so that the ben-
efits of wind-power production could be shared among those who
also bore the costs.
In the intervening years, all of the legislation for subsidies, co-
operatives, financing guarantees, and so forth has been continually
reviewed and regularly revised in a way that is gradual and predict-
able. This makes for a stable business climate, and stability enables
investors to successfully plan for the long term. The cumulative ef-
fect of this effort has been to foster investment in wind power. Wind
power in Denmark has grown rapidly and steadily for the last two
decades.
But success also involves overcoming technical challenges. Spe-
cifically, there are difficulties in integrating an intermittent power
source such as wind at the scale that Denmark has accomplished.
Because wind power is currently used to meet about 20 percent of
Danish electrical power needs—and that percentage will continue to
rise—the intermittent nature of wind can cause problems in terms of
balancing supply and demand. When the wind blows over large areas
of Denmark, a great deal of power is generated. But if the wind fails
over a large area, the Danes must find a way to cover a very significant
shortfall in national electricity production. This is the challenge.
Denmark has responded by improving grid connections with its
neighbors, Sweden and Norway, neither of which is dependent on
wind power. (Norway’s electricity is produced primarily through
hydroelectric facilities, and Sweden depends in roughly equal mea-
sure on nuclear power and hydroelectric facilities.) These grid con-
nections enable Denmark to balance supply and demand. Because
Denmark is small, has substantial wind resources, and is located
close to energy-rich neighbors, Denmark has found an answer to
its energy needs that works well for Denmark. The Danish model
may not be a suitable approach for a larger or more geographically
isolated economy.
114  Wind and Water

Germany: Rapid Growth,

Ambitious Goals
Germany has significant wind resources. Currently, it is heavily
dependent on coal, a fuel that it has in abundance, and nuclear
power. But the German government has pledged to gradually shut
down all nuclear facilities and to build no more. It has also pledged
to minimize coal consumption in order to reduce greenhouse gas
emissions. Simultaneously accomplishing both goals would be an
extraordinary technical accomplishment. Germany’s challenges are
further compounded by the fact that much of its natural gas supply
originates in Russia, and there are concerns that Russia will use its
supplies of natural gas to gain political advantage over adversar-
ies, as some claim it has already done with its neighbors Ukraine
and Belarus. Reluctant to use natural gas, determined to minimize
the use of coal, and having decided to eliminate nuclear power,
and with little in the way of additional undeveloped conventional
hydropower resources, Germany turned to wind as its best bet for
future energy production.
It is important to bear in mind that in comparison to Denmark,
Germany’s economy is enormous. Creating a wind-energy sector
capable of displacing much of the commercial nuclear output of one
of the world’s largest economies is a daunting challenge. Germany
already produces more of its energy from wind than any other na-
tion. In 2007 there were approximately 20,000 operating turbines
producing about 6.4 percent of its energy needs. Planners hope
wind’s contribution to the nation’s electrical energy output will
be about 20 percent by 2020. Because Germany pursues a strategy
of aggressive electrical energy conservation, the growth in energy
demand is small—less than 1 percent per year. Renewables are ex-
pected to grow at a rate of about 2.6 percent per year, so wind is on
track to gradually displace other more conventional sources of en-
ergy. But as with Denmark, the growth of the wind sector depends
heavily on government involvement in the energy markets.
The Role of Government in Promoting New Technologies  115

As with Denmark and the United States, German interest in

renewable energy dates to the oil shocks of the 1970s. At first, there
was increased investment in coal, nuclear, and the so-called renew-
able forms of energy. (Motivations have changed over the interven-
ing decades, and now the German government subsidizes renewable
energy principally for the perceived environmental advantages.)
Beginning in 1975 and continuing into the 1980s, Germany initially
concentrated on wind turbine research.
In 1986, the German federal government began subsidizing
the construction and operation of wind turbines. As designs were
evaluated and experience gained, subsidies began to increase. By
1989, participants accepted into a special program had the option
of choosing between a production subsidy—the government paid
4.3 cents per kilowatt hour—or a 60 percent grant to pay for the
construction of wind facilities. In 1991, the German government
passed its so-called Feed-in Law, which required utilities to pur-
chase renewable power at 90 percent of the retail cost of that power.
This requirement is similar in concept to the United States’ PURPA
legislation described earlier. All of the German subsidies were, how-
ever, more generous than those provided by the federal government
of the United States. And in 2000, Germany instituted the Renew-
able Energy Law, which stated that a new wind-turbine project
would be subsidized at a rate of 11 cents per kilowatt-hour for the
first five years of production, at which point the subsidy would be-
gin to decrease. This was a tremendous boon to independent power
producers, and construction surged ahead.
Germany maintains a very robust wind-energy sector, but as
is the case in the United States, the German wind-turbine sector
remains dependent upon the generosity of the government. This
means that as the industry grows so will the subsidies it requires. It
is even possible that the subsidies will grow faster than the capacity
of the industry to generate power, because as less profitable sites
are developed, larger subsidies may be necessary to make them at-
116  Wind and Water

tractive to investors. Technically, Germany can continue to expand

its wind capacity, but whether its populace will be willing to pay
the ever-increasing subsidies necessary to make this possible is not
clear. There are already some indications that support for wind
power has begun to weaken.
 Part III

Wind Power
 8

Wind Power:
A Brief History
S uccessful attempts to harness the power of moving air began
long ago. Wind power enabled early pharaohs to sail the Nile.
Windmills have been used for well over 1,000 years to pump water
and grind grain. For centuries European explorers depended upon
the wind to traverse the world’s oceans. During the latter years of
the 19th century, however, interest in wind power waned. Wind
technologies, some of which were already fairly advanced, were dis-
placed by fossil fuel technologies. Fossil fuels were easy to obtain,
relatively easy to transport, inexpensive, and most importantly,
fossil fuel–powered engines did not depend on the vagaries of the
weather.
Even today all of these statements are still true of coal, but
the situation has changed for oil and natural gas. Oil, once widely
used as fuel for power plants, is rarely used to generate electricity
anymore; its price is too volatile. Coal-fired and natural gas–fired

119
120  Wind and Water

At the USDA-ARS Con­

servation and Production
Research Laboratory, in
Bushland, Texas, wind
­turbines generate power
for submersible electric
water pumps that are far
more efficient than tradi-
tional windmills. (Scott
Bauer, USDA)

power plants are as important as ever—in the United States, they

provide about 70 percent of the nation’s electricity—but they are of-
ten used in different ways. Coal-fired plants are usually operated for
prolonged periods at fairly constant outputs. They do not compete
with wind because they do not serve the same function. Natural
gas–fired plants, by contrast, are often used intermittently, being
turned on and off as the demand for power rises and falls. The situa-
tion with respect to natural gas is especially relevant to this chapter,
because both gas–fired power plants and wind turbines are used to
produce peak power. Natural gas–fired plants, though functionally
very reliable, are often used intermittently because their fuel has be-
come too expensive to burn continuously. Wind turbines are used
intermittently because the wind only blows some of the time. But if
the two types of power generating technologies are used in roughly
 Wind Power: A Brief History  121

similar ways, their cost structures are very different. Because new
deposits of natural gas tend to be expensive to produce, and because
natural gas supplies are increasingly prone to disruption, natural
gas prices have become volatile, making long-term planning diffi-
cult for residential and industrial customers alike. By contrast, wind
energy has a well-understood cost structure and although intermit-
tent in supply, its physical characteristics, especially its average rate
of supply, are predictable. Wind, once forgotten, has again become
one of the fastest growing sectors of the power-generation market.
In addition to changes in the economics of fossil fuels, the envi-
ronmental costs associated with a heavy reliance on fossil fuels are
becoming increasingly apparent. The consumption of fossil fuels
results in the release of various pollutants. Some of the pollutants
are local in nature and contribute to smog or acid rain, for example.
Carbon dioxide emissions, however, are much longer lasting, and
because they affect the global climate, burning fossil fuels could not
have more widespread environmental implications. Wind turbines,
as will be seen, also have environmental effects. These effects are not
especially obvious only because wind farms generate relatively little
power at present. If wind becomes a large-scale source of electric-
ity, its effects will generate more controversy. Nevertheless, when
compared to fossil fuel–fired power plants, a heavy reliance on wind
has far fewer environmental implications.
Today, engineers and scientists are attempting to make the
maximum possible use of this resource, which is inexhaustible and
emissions-free. This chapter provides a brief history of the evolution
of wind-power technology.

Windmills
Not much is known about the very early history of windmills. They
seem to have been invented in Persia, which was located in present-
day Iran, or Mesopotamia, which was located in present-day Iraq.
Some claim windmills first appeared in China. The earliest written
122  Wind and Water

references are, however, Persian, and date to the seventh century c.e.
These early windmills were used to mill grain and possibly to pump
water. But if there is uncertainty about their origin and initial ap-
plication, there is, at least, general agreement on the basic elements
of their design: The arms of ancient windmills were mounted on a
vertical shaft so that they turned in a plane that was parallel to the
ground, and they were somewhat similar in appearance to revolv-
ing doors. The sails of these windmills were often made of reeds or
wood, and as with modern revolving doors, these early windmills
were also partially enclosed by walls.
If building a windmill indoors seems counterintuitive, imagine
what would happen if the windmill were not partially enclosed by

Top-view diagram of a vertical axis windmill—the wind blew through the

opening on the right and blew out through the opening on the left
 Wind Power: A Brief History  123

walls: As the wind blew on the sails on one side of the windshaft
(the axle to which the sails are affixed), it would also blow on the
sails on the other side of the windshaft. The result would be that
the force on one set of sails would be balanced by the force on the
other side and little or nothing would happen. The solution adopted
by the Persians was to enclose the windmill within a walled struc-
ture that had one opening through which the wind could enter and
one opening, located on the opposite side of the structure, through
which the wind could exit. The entrance was sometimes conical in
shape so that as the wind blew into the wide end of the hole, it accel-
erated, and emerging with increased velocity from the narrow end,
it pushed against the sails. The opening was placed so that the wind
was directed toward one side of the windshaft only. The sails on the
other side of the shaft were protected from the wind by a curved or
L-shaped wall. This allowed the wind to push preferentially on one
side of the windshaft and so cause it to turn in one direction only.
The main advantage of this design is its simplicity. There were
no gears. By mounting the sails on a vertical shaft, the first wind-
mill builders could attach millstones directly to the spinning shaft.
The wall might seem to be a significant disadvantage because if the
wind changes direction it will not pass through the openings, and
consequently, no power will be generated. These early mills were,
however, associated with an area of Persia that today is called Sistān
and is located near the Iran-Afghanistan border. In Sistān, the wind
blows steadily from the north from mid-June until mid-October.
Provided the mills were built facing north, they did not have to
be sensitive to changes in the wind because there were none. One
famous site in the ancient Persian city of Neh had 75 of these verti-
cal-axis windmills, all built in a single long row.
The main disadvantage to the vertical-axis design lies in the
fact that it makes inefficient use of the sails. Because the wind only
blew on half the sails at one time, only half the sails were transmit-
ting force to the spinning shaft. The other half was, at best, dead-
124  Wind and Water

weight. Moreover, even on the side of the shaft on which the wind
blew, some sails partially obstructed others from the full force of
the wind. Vertical-shaft windmills did not exert much force; their
output, when measured by the amount of processed grain they pro-
duced per hour, was not large.
Despite its limited efficiency, the vertical-shaft windmill has
never fallen entirely out of favor. Traditional designs were used in
parts of Iran and Afghanistan well into the 20th century, and Euro-
peans and Americans have experimented with the design for cen-
turies. The British physician Erasmus Darwin (1731–1802), grand-
father to the naturalist Charles Darwin, built a design of which he
was very proud: It consisted of a cylindrical tower equipped with
slats that could open and close. The (vertical) windshaft was placed
at the center of the tower and spanned its entire length. The sails,
which looked more like ceiling fans than revolving doors, were at-
tached to the top of the shaft. The slats on the windward side of
the tower would be opened at an upward angle. The wind would,
therefore, pass through the slats, and as it passed through the slats,
it was deflected upward through the tower. As the wind exited the
top of the tower, it acted against the sails and set the shaft turn-
ing. Passing the wind through narrow openings and changing its
direction from horizontal to vertical, not surprisingly, diminished
the force of the wind. This design was, therefore, capable of only
modest power even in a strong wind. (A more recent vertical-axis
design, called a Darrius wind turbine, looks like an eggbeater and
may prove better suited for use in tidal mills.)
The history of traditional horizontal-axis European windmills
begins during the 12th century. The origin of the European design
is unknown. Because the first European references to windmills oc-
cur after the First Crusade, which began in 1095 c.e., some think
that the concept, at least, is of Middle Eastern origin, but even the
earliest European designs employed an almost horizontal wind-
shaft, a design quite different from the vertical windshaft designs
 Wind Power: A Brief History  125

© Infobase Publishing

Post mill. Because the entire structure was mounted on a single pole, it could
be turned to face the wind no matter from what direction the wind blew.

that were still in use in the Middle East. European mills attempted
to harness the wind regardless of the direction in which it blew, a
very different goal from the early Persian design.
The earliest type of European windmill is called the post mill.
It was an extremely durable technology. A few commercial post
mills—they were used to grind grain—were operated in Great Brit-
ain well into the 20th century. Post mills were, therefore, economi-
cally viable for about 800 years. These mills were, for the most part,
126  Wind and Water

used to grind grain or pump water. The main structure of a post

mill, the enclosure in which the miller worked and which housed
the millstones, looks like a shed or small house supported above
the ground by a single large round post. The windmill’s sails were
mounted on a shaft that was inclined slightly upward—often at
about 15°. The windshaft was coupled by two or more gears to a
vertical shaft to which the millstones were attached.
Compared to Persian windmills, the post mill had the potential
to be more efficient because it used all of its sails all of the time—or
at least all of the time that the wind was blowing. Early sails were
made of large wooden lattices, each of which was attached to the
windshaft by a single spar and covered with canvas to provide a sur-
face against which the wind could push. (Etchings show that very
early windmill operators actually wove pieces of canvas through the
lattices.) The sails were slightly tilted relative to the plane in which
they rotated so that when the wind blew on the sails, it imparted
a turning force to the windshaft. (To see why this works, imagine
that the sails were oriented so that they lay in their plane of rotation,
and suppose that the windshaft pointed directly into the wind. In
this configuration, the entire force of the wind would be caught by
the sails and transmitted directly back along the windshaft without
exerting a turning force.)
The output of the windmill was sensitive to the direction of the
wind. A shift in wind direction could leave the canvas flapping inef-
fectually on a stationary shaft. This is the reason that the post mill
was mounted on a post. The entire structure could pivot on its post
as if it were a multiton weathervane. The turning procedure was
originally accomplished manually: A long pole extended out from
behind the mill. The pole was simply a very large lever. When the
direction of the wind changed, the mill operator went behind the
mill and pushed on the far end of the pole until the windmill faced
into the wind again. Later a secondary sail was sometimes added
to the back of the mill. It was oriented so that it was parallel to
 Wind Power: A Brief History  127

the windshaft. Its purpose was to provide a surface against which

the wind could push so that the mill would pivot without operator
intervention.
As one might expect, the larger the area of the sails, the more
of the wind’s power could be transmitted to the windshaft, and
the more work the mill machinery could perform per unit time.
All other things being equal, therefore, larger sails were better. But
larger sails required a larger support structure, and, in particular,
the structure had to be taller because the axis on which these lon-
ger sails turned had to be lifted higher off the ground so that the
spinning sails could clear the ground as they rotated. But this was
only one factor that caused designers to build taller mills. At least
as important is the fact that winds along the ground are weaker
than those at higher elevations due to interaction of the wind with
trees and buildings, a fact long known to windmill designers. This
resulted in two trends in windmill design: Mills became more mas-
sive so that they could resist the force of the wind against the larger
sails, and they became taller in order to raise the sails as high into
the oncoming wind as possible.
Building a very large post mill is problematic. As the structure
becomes more massive, it becomes more difficult to support on a
single post as well as more difficult to turn. The alternative to the
post mill was the tower mill. (The so-called smock mill is usually
listed as a different type of windmill because from the outside it
looks different from the tower mill—its shape is supposed to be
reminiscent of a smock—but mechanically, tower and smock mills
operate according to the same general design. This book will dis-
cuss only the older tower mills, but most of the statements made
about tower mills apply word for word to most smock mills.) The
tower mill was developed during the late Middle Ages. They were
anchored securely to the ground and were often several stories tall.
Only the top section of the mill, called the cap, turned so that the
sails could face the wind. The cap housed the windshaft. The much
128  Wind and Water

larger bottom section, which was often large enough for storage and
living quarters, was immobile.
Tower mills were often taller and more massive than post mills.
The biggest were approximately 100 feet (30 m) tall. At this scale
even the cap was massive, and the track on which the cap turned
had to support the massive sails and the windshaft, in addition to
the cap itself. To reduce stress on the track and to make the problem
of turning the cap manageable, tower mills were usually built in
the form of truncated cones—that is, the round base is massive and
wide, and the sides slope upward to a smaller and relatively lighter
cap. As with their smaller counterparts, tower mills continued to
make an important contribution to the economies of several Euro-
pean nations for several centuries and only began to fade late in the
19th century.
Throughout the history of European windmill design, the main
technical issue, for economic and safety reasons, concerned the de-
sign of the sails. The millwright had to be able to control the amount
of the wind’s force that was transmitted to the sails. An unexpect-
edly powerful wind could cause the sails to turn too fast. The flour,
the processing of which was the main reason that many of these
windmills were constructed, could be burned if it was milled when
the machinery was spinning too rapidly. Moreover, a rapidly turn-
ing windshaft could damage or destroy equipment. And working in
a mill experiencing unexpectedly strong winds was dangerous too.
Sometimes, the millwright was injured or even killed as unexpect-
edly high winds spun the mill’s machinery out of control.
There were also the problems associated with working in light
winds. For economic reasons the mill should operate whenever the
wind was blowing. But in order to overcome static friction, stron-
ger forces were necessary to start the windshaft turning than were
needed to continue operation once the sails had begun to rotate.
Poorly designed sails could prevent the windmill from starting in
light winds.
 Wind Power: A Brief History  129

Dutch tower windmill (no longer functional) (Harskampermolen Mariana)

To control the amount of the wind’s force transmitted to the

mill, millwrights developed methods for adding or reducing the
amount of canvas exposed to the wind. Recall that the canvas ­covered
130  Wind and Water

a ­lattice structure, so that using canvas to cover only part of the lat-
tice enabled the millwright to continue to operate the mill at a safe
speed under heavy winds. When the winds were lighter, more canvas
was deployed. Several ingenious methods were developed that en-
abled the millwright to adjust the amount of canvas without having
to climb all over the sails.
Changing the amount of sail exposed to the wind by altering
the amount of canvas was still time-consuming and occasionally
dangerous. Gusts of wind could still accelerate the sails in unex-
pected ways and occur too rapidly for the operator to respond.
A later invention involved replacing the canvas with devices that
looked like venetian blinds mounted on springs. The tension of the
springs could be adjusted from inside the mill via levers. When the
wind blew harder than some predetermined level of force, it blew
the blinds open, and the wind “spilled” through the lattice without
generating any turning force. When the force of the wind was below
that predetermined level, the springs closed the blinds, providing a
surface against which the wind could push.
With respect to the problem of starting the mill in light breez-
es, the best way was to design sails that were strongly tilted away
from the plane of rotation. This enabled the wind to exert a strong
sideways force—strong enough for a light breeze to begin turning
the sails. The problem is that as the speed of rotation increased the
tilted sails actually slowed the spinning shaft, decreasing its effi-
ciency. The problem was most pronounced at the ends of the sails,
which, after all, always turn faster than those parts closer to the
windshaft. The solution was to twist the surface of each sail so that
near the windshaft the sail’s surface was turned strongly away from
the plane in which the sails rotated. Nearer the tip, however, the
surface of the sails twisted so that they were more nearly parallel
to the plane of rotation. Designers believed that this design enabled
the mill to start turning in a light breeze, and still operate efficiently
as the speed of rotation increased. Various forms of several of these
innovations can still be found on contemporary wind turbines.
 Wind Power: A Brief History  131

Until late in the 19th century, tens of thousands of windmills

dotted the landscape of many parts of northern Europe. In 1900,
Finland, for example, had 20,000 operating windmills. Large num-
bers of relatively low-power windmills distributed over a large area
was an almost ideal solution to the demands of the time. Through-
out most of the history of the European windmill, the transpor-
tation infrastructure of most European nations was not especially
good. Grain was, therefore, grown and milled locally, and, provided
a windy site was located nearby, windmills were a good match for
this type of small-scale production.
The last large-scale effort to convert wind power into mechani-
cal energy—as opposed to electrical energy—occurred in the United
States. Farmers, faced with the vast treeless expanses of the Great
Plains, a region that often lacked sufficient surface water for their
agricultural needs, installed windmills to pump water from the
ground. These windmills used so-called annular sails, numerous
curved wood or steel slats that radiated out from the center. The sail
was held perpendicular to the wind by a vane in back. The tower
often consisted of an unadorned steel lattice that supported the sail
at a high enough height for it to operate efficiently. During the lat-
ter days of the 19th century, hundreds of American firms produced
hundreds of thousands of these wind-driven pumps. These devices
are still manufactured. Currently there are an estimated 1 million
to 2 million wind-driven water pumps in service worldwide, but
their value to agriculture has been surpassed by pumps powered by
electricity, gasoline, or diesel fuel, all of which produce more power
and operate without regard to the weather.

Wind Turbines
The term wind turbine is usually reserved for windmill-like ma-
chines that convert the kinetic energy of wind into electricity. Per-
haps the first such energy-conversion device, designed and built by
the American engineer and inventor Charles F. Brush (1849–1929),
is known as the Brush turbine. It was erected in Cleveland, Ohio,
132  Wind and Water

Lely wind farm, the Netherlands, one of the first offshore wind farms (United
Nations Atlas of the Oceans)

and began operation in 1888. The sail was 50 feet (19 m) across and
provided electricity to Mr. Brush’s very large house. Brush solved
the problem of windless days by using the modest output of his
enormous machine to charge batteries. His invention was highly
acclaimed—there is a detailed description of the Brush turbine in
the December 20, 1890, issue of Scientific American—but the idea
was not widely copied. Even Brush turned his attention to other
projects.
The earliest sustained effort to harness the wind was undertak-
en in Denmark beginning in 1891 by Poul la Cour (1846–1908), a
prominent Danish inventor and teacher. La Cour’s efforts were car-
ried out over several decades; his work was supported by the Dan-
ish government, and his discoveries had a more lasting effect than
 Wind Power: A Brief History  133

those of Brush. It is interesting to note that la Cour’s first solution

to the problem of power production on windless days was to use
the electricity produced by his wind turbine for electrolysis—the
separation of water molecules into oxygen gas and hydrogen gas.
The hydrogen could be stored and burned at the convenience of the
user, and the resulting heat could be harnessed as desired.
Modern grid-connected wind turbines have a great deal in com-
mon with the windmills that preceded them. The rotor, the analogue
of the sail, is usually mounted on a horizontal shaft. The shaft is con-
nected to a transmission, also called a gearbox. The gearbox is placed

Cutaway view of a modern wind turbine

134  Wind and Water

between the windshaft and the generator and is used to ensure that
the generator turns at an optimal speed. From a distance, the blades
of a wind turbine may appear to turn fairly quickly, but in order to
produce electricity suitable for the electric grid, the generator must
turn much faster—typically in the 1,200–1,800 rpm range. This
change in speeds is accomplished by the gearbox. (Early windmills
sometimes used sets of gears to change the rate at which the mill-
stones turned, but the difference in rotation rates was typically much
less than that obtained with modern wind turbine gearboxes.)
Just as larger sails were preferred to smaller ones by designers of
traditional windmills, contemporary engineers prefer longer rotors
to shorter ones and for the same reason: The amount of power con-
verted is proportional to the area swept out by the rotor. In order to
capture as much of the wind’s energy as possible, the wind turbine
should use the longest rotor practicable.
Height is another goal of wind turbine engineers just as it was
of their predecessors. Recall that early windmill designers built tall
windmills in order to raise the sails as high into the wind stream
as possible. Contemporary wind turbine designers also mount their
turbines on towers that are as high as practicable. Today, engineers
often mount turbines on towers that exceed 275 feet (84 m) in height.
Is the extra height worth the expense? A rule of thumb is that on an
open unobstructed landscape, wind speed will increase by about 25
percent as elevation above the ground increases from 50 feet (15 m)
to 200 feet (61 m).
Another area of commonality involves the problem of regu-
lating rotor speed. Wind currents are highly variable, and just as
windmills were designed to operate within certain limits, so, too,
are wind turbines. A very strong wind has the potential to spin the
rotor too quickly, which could result in expensive repairs. The ro-
tors on wind turbines are aerodynamically designed to produce a
pressure difference between one side of the rotor and the other as
wind flows past the rotor. The idea is similar to that used to create
 Wind Power: A Brief History  135

lift with an airplane wing. It is this pressure difference between the

two sides that causes the rotor to rotate. When the wind becomes
too strong, a control device changes the angle of the rotor relative
to the wind causing the pressure difference to diminish. In this way
the forces acting on the turbine are controlled, and the wind turbine
is protected from wind damage.
An early version of this protective technology was a spring-ac-
tivated linkage that was similar in concept to the spring-mounted
venetian blind–like devices on traditional windmills described ear-
lier. When the rotor began spinning too quickly the spring linkage
was released, and the angle at which the rotor met the oncoming
wind changed. The lift forces were eliminated, and the wind turbine
was made safe. This task is now usually handled by computer, which
monitors the rotor and continuously adjusts the angle at which the
rotor meets the oncoming wind in order to keep the forces acting on
the rotor within prespecified limits.
One of the main differences between traditional windmills and
contemporary wind turbines involves the speed with which they
rotate. The simple construction materials used in early windmills
meant that the windshaft could not rotate quickly. But a slow rate
of rotation—slow, that is, relative to the velocity of the oncoming
wind—also meant that the windshaft exerted a powerful turning
force. This is exactly the characteristic that is desirable for applica-
tions such as pumping water and processing grain. High rates of
rotation are better suited for electricity generation. The speed of the
tip of a modern three-rotor wind turbine will typically turn between
four times and six times the speed of the wind that drives it. A very
long rotor need not turn at a high rpm to achieve its optimal ratio
between tip speed and wind velocity. A shorter rotor, by contrast,
must turn much faster in order to maintain that ratio. This explains
why smaller wind turbines typically rotate faster than larger ones.
Achieving the goal of a high rate of rotation in moderate wind
speeds also helps to explain why wind turbine blades look so much
 136  Wind and Water

different from the sails one finds on traditional European windmills

or the sails one finds on the windmills that once were so common
on the Great Plains of the United States. Obtaining a high rate of

The Altamont Pass Wind

Resource Area

A ltamont Pass, located in California, is one of the earliest and at one

time was one of the largest of all wind farms located in the United
States. Altamont Pass was chosen because it is located near a high-volt-
age transmission corridor—building a high voltage line to connect a
power production facility to the grid is expensive—and because high
temperatures inland often result in the formation of a wind that blows-
from the ocean through Altamont Pass. Just as important—perhaps
more important at the time that Altamont was constructed—the state of
California offered substantial tax incentives to companies that built wind-
power facilities in the state. Altamont Pass was once widely hailed as an
environmentally friendly producer of electricity.
Since 1981, when construction began, thousands of wind turbines
have been built here. They are built along ridges and other locations where
the probability of strong winds is optimized. The sight of these wind tur-
bines was so impressive that Altamont Pass became an important tourist
destination. Row after row of moderately sized turbines spinning furiously
in the wind was, at one time, a sight that many found fascinating.
By today’s standards, many of the turbines at Altamont Pass are
small and inefficient, and they are being replaced with a smaller number
of larger, more efficient turbines. But even with old technology, in 2005
Altamont Pass had a total rated capacity of 548.3 MW—that is, if all the
turbines were operating at full power, Altamont Pass would generate
548.3 MW of power. Often, however, not all of the wind turbines operate
at full power, and there are some days when none of the turbines operate
at full power because there is not enough wind.
How valuable a project is Altamont Pass? For planning purposes, the
Monterey Bay Regional Energy Plan estimates that the percentage of time
that wind power resources produce their rated power is 20 percent. This
 Wind Power: A Brief History  137

rotation requires a small number of long thin blades. Obtaining a

slow motion with a high turning force is best accomplished with
wider, more numerous sails. While modern wind turbines typically

includes the Altamont Pass project. There is a big difference between the
so-called nameplate capacity, which is what a project generates when
it operates at full power, and the actual power output, which is always
much smaller.
The biggest criticism of Altamont Pass concerns its environmental
impact. To be sure, when the wind is blowing, the turbines produce emis-
sions-free power, but they also regularly kill golden eagles, burrowing
owls, and red-tailed hawks. These birds, and others, fly into the rotors.
The issue of the destruction of bird species has become one of the central
issues associated with the operation of Altamont Pass.
Studies have shown that bird mortality can be reduced by chang-
ing the design of the towers. Some tower designs attract more birds
than other designs, causing the birds to perch near the potentially lethal
rotors. And studies have shown that sometimes the location of a wind
turbine makes it particularly deadly: Turbines located in canyons tended
to destroy more golden eagles than other similar turbines placed else-
where. Presumably, mortality can be reduced by building replacement
units with less attractive towers, and excluding some locations from
development. But this may not be enough to satisfy many opponents
because the wind farm is also located along a bird migration route, and
some deaths of the golden eagle, a federally protected species, seem
unavoidable.
Some groups now advocate shutting down Altamont Pass during the
winter months, when bird mortality is highest. This proposal, if enacted,
would further diminish the economic value of the wind farm. Altamont
Pass illustrates the difficulties of identifying what constitutes “green en-
ergy.” Perceptions of what it means to produce power in environmentally
sound ways continue to evolve.
138  Wind and Water

use two or three thin blades, traditional windmills use four or more
wide sails.
Large modern wind turbines can, under the right conditions,
generate significant amounts of power. Details vary by manufacturer,
but General Electric, for example, currently manufactures wind tur-
bines in the 1.5–3.6 MW range. When all its occupants are home and
all of its appliances are drawing power, the average home will draw
anywhere from two to four kilowatts. This is the maximum rate at
which it draws power. An electrical grid must always be prepared
to meet maximum demand. (A system that only meets the average
demand will fail to meet the demand for power roughly half of the
time. Most people would agree that a 50 percent failure rate for an es-
sential service is unacceptable.) A three-MW turbine will, therefore,
be able to provide enough power for between 750 and 1,500 homes,
provided the wind is blowing. Clearly, powering a modern national
economy with wind turbines would require an enormous number
of wind turbines even when the wind is blowing hard, and when the
wind is not blowing, no collection of wind turbines is enough. How,
then, can one assess the value of wind turbines? How much reliance
can be placed on a power source that is inherently unreliable? These
questions are addressed in the next chapter.
 9

The Nature of Wind Power

W ind turbines are energy conversion devices. They convert
the kinetic energy of the wind into electrical energy. This
simple-sounding statement indicates that there is an upper limit on
the amount of energy that can be derived from the movement of a
particular mass of air. No wind turbine can produce more electri-
cal power than the amount of power in the wind itself. In fact, the
amount of electricity produced by any turbine must be substantially
less than the kinetic energy of the wind, since any device that con-
verted all of the kinetic energy of the wind into electrical energy
would have to stop the wind from blowing. Determining the maxi-
mum amount of energy that can be produced per unit time from
moving air—that is, determining an upper limit on the power gen-
erated by a wind turbine—is an important first step in understand-
ing the potential contribution that wind turbines can make toward
the electricity supply. This is one goal of this chapter.

139
140  Wind and Water

Because winds at higher altitudes are stronger and steadier, the towers are as
tall as is economically practical. (Cam Gordon)

But there is more to understanding wind energy than comput-

ing the amount of energy available on a windy day. Some days the
wind does not blow. Wind turbines are, therefore, fundamentally
different from coal and nuclear plants in that wind energy is inter-
mittent by nature. While it is true that coal and nuclear plants may
break down without warning, experience has shown that both types
of power plants produce power reliably and continuously for long
 The Nature of Wind Power  141

periods of time without interruption. Consequently, they are usu-

ally used to meet the minimum demands of an electricity market.
This is called base load power. Wind energy is poorly suited for this
type of function. Wind does not blow steadily enough—wind is not
reliable enough—to supply base load power. Consequently, wind
does not compete with coal and nuclear.
There is, however, another aspect of electricity production
called peak power. (Power is sometimes further subdivided, but for
purposes of this discussion, as for many others, it is sufficient to
divide power production into peak and base load.) Peak power is
the electricity that must be produced to supplement the base load
requirements on an hour-by-hour basis. Peak demand occurs daily
as electricity use rises during business hours or on hot days. (Air
conditioning, for example, is an energy-intensive technology.) Peak
power requirements usually occur for just part of each day. Natural
gas plants are often used to meet peak demand, and, in fact, some
natural gas–fired power plants are used only during peak periods.
The reason, as previously mentioned, is that natural gas has become
so expensive that electricity produced by natural gas–fired plants
is, when compared to that produced by coal and nuclear plants, un-
economical to produce. Each day, as demand rises, the order will go
out to turn on natural gas–fired plants so that they will be ready to
produce power at the required time. If the demand is high enough,
natural gas plants will be brought online to produce electricity for
the grid, and the owners will be paid accordingly. Later in the day,
as demand falls, the plants are turned off. In contrast to natural gas
plants, wind turbines produce power (or not) independently of the
demand for electricity because the wind blows without regard for
the electricity markets. In this sense, wind turbines are not well-
suited to peak power production either, because (again) they are
unreliable. And yet wind turbines do produce power, and some-
times many wind turbines produce a great deal of power. This raises
the question of how one should value the contribution that wind
142  Wind and Water

power can make. Answering this question is the second goal of this
chapter.
Finally, this chapter discusses the environmental effects of wind
power. Every technology that produces commercially significant
amounts of electricity has significant environmental effects. Wind
power is no exception.

How Much Energy Is in the Wind?

Ultimately, the winds are driven by the unequal heating of Earth’s
atmosphere. Wind is, therefore, a form of solar energy. Earth spins
rapidly—land at the equator is moving about Earth’s axis of rota-
tion at approximately 1,000 miles per hour (1,600 kph)—and be-
cause Earth is a sphere, different latitudes are exposed to differing
amounts of sunlight for different periods of time. And different
parts of Earth’s surface absorb the Sun’s heat in different ways.
The oceans, for example, absorb solar energy differently than land
does. Finally, air is transparent to the Sun’s light in the sense that as
the rays from the Sun pass through Earth’s atmosphere, they have
little effect on its temperature. Instead, the Sun’s rays heat Earth’s
surface, which absorbs some of the Sun’s light and radiates it back
into the atmosphere as heat. (Some light is also reflected.) Earth’s
atmosphere is, therefore, heated from below, not from above. The
uneven heating causes changes in the density and pressure of the air
near Earth’s surface, causing the air to move. The path that large-
scale motions of air take across Earth’s surface are influenced by
the planet’s rotation. This very dynamic process produces moving
regions of low- and high-pressure air that form and dissipate and in
the process produce winds. Some winds are local, and some sweep
over large areas of Earth’s surface. The processes involved are as old
as the planet on which they occur.
As with hydropower, wave power, and tidal power sites, windy
sites are natural resources, the locations of which are beyond hu-
man control. Simply put: Windy sites can be developed or not, but
 The Nature of Wind Power  143

they cannot be created. The suitability of each site should, of course,

be evaluated on its own merits, but the number of commercially
viable sites is not as large as one might think. The decision to forgo
development of one windy site for environmental, economic, or
aesthetic reasons is also a decision to either attempt to extract more
power from another site, assuming one is available, or to switch to
another technology to produce the electricity that no one is willing
to live without. (The problems associated with site development are
considered in more detail in the next chapter.)
Winds are a three-dimensional phenomenon, and the character-
istics of wind depend very much on the elevation above the ground
at which it is observed. Winds that are close to the ground are very

Diagram showing how wind speed increases with height. Turbulence

decreases with height.
144  Wind and Water

turbulent. They are accelerated as they pass between tall buildings

or through mountain passes. They are deflected by buildings and
trees, and as the wind passes over and around these obstacles eddies
form and roil along at ground level. This is most easily observed
during blizzards when snowflakes are swept along the complex, un-
certain paths that the wind takes as it moves across the landscape.
Turbulent air movements contain a great deal of kinetic energy,
but because much of this energy is contained in eddies and other
small-scale unsteady motions, it is difficult to convert the kinetic
energy of turbulent airflows into electrical energy. Moreover, as the
wind blows through trees and across “rough” terrain, some of the
energy of the wind is dissipated. It is a much easier process to convert
into electricity the smooth steady flows of air that occur hundreds of
feet above the ground. At higher altitudes the wind is unobstructed
and less turbulent, and so the process of converting its kinetic energy
into electricity is more efficient. This is one reason that commercial
wind turbines are so tall. They tower over the landscape in order
to lift the blades of the turbine into the smoother, steadier airflows
that exist hundreds of feet above the ground. Turbines mounted on
shorter towers are less productive, less efficient, and less profitable,
and consequently when mounted on shorter towers more turbines
are required to produce the same energy as fewer turbines mounted
on taller towers. There is, as a general rule, little economic sense in
mounting an expensive turbine on a short tower.
Companies seeking to produce electricity from the wind must
identify sites where the wind blows as steadily as possible. If they
receive approval to erect their turbines at such a site, they construct
towers that extend high enough to capture wind unperturbed by
interaction with the ground. Under these idealized conditions, how
much power can a wind turbine of a given size generate when wind
is blowing at a given speed?
The answer to the preceding question is contained in equation
(5.1), the same equation that was used to determine the maximum
 The Nature of Wind Power  145

amount of power that can be harnessed from ocean currents. That

equation is repeated here for ease of reference:

CdAv3
Pmax = — (9.1)
2
In this equation, Pmax still represents the maximum amount of
power that can be obtained from a turbine, but now d is the density
of air, A is the area swept out by the spinning (wind) turbine blades,
v is the speed of the wind, and C is a number that represents the
maximum percentage of the wind’s energy that can be harnessed.
(The expression dAv3/2 is the total amount of power contained in
air of density d blowing through a tube of cross-sectional area A
with velocity v.) Equation (9.1) reveals a great deal about the physics
of wind turbines.
First, d, the density of air, is very small compared to, for example,
the density of seawater. To see the difference between air and wa-
ter—and so the difference between wind turbines and tidal mills—it
is enough to know that the density of seawater is roughly 900 times
greater than the density of air at sea level. This is one reason why
tidal mills can, in principle, produce so much electricity from slowly
moving ocean currents: Although the velocity of the ocean current is
usually small compared to the velocity of wind currents, the amount
of power that can be generated by an ocean current can still be large
because of the relatively high density of seawater.
Second, the density of air decreases rapidly with height so, for
example, although the wind may blow steadily at the top of a tall
mountain, there will not be as much energy to convert at that al-
titude as there is at lower altitudes—all other things being equal—
because of the thinner mountain air. To take an extreme case, the
density of air at the top of Mount Everest is only about one-third the
density of air at sea level.
Third, equation (9.1) shows that the amount of power that can
be harnessed from the wind is highly sensitive to the velocity of the
146  Wind and Water

wind. As noted in chapter 5, if the speed, v, of the wind is doubled,

the power of the wind increases by a factor of 8 (2v × 2v × 2v =
8v3). To take a more realistic example, a wind turbine operating in
a wind that blows at 12.5 miles per hour (20 k/hr) can, in principle,
generate twice the power of a turbine operating in a 10-mph (16
k/hr) wind. (As will be shown later, wind turbines are designed to
produce a rated power output, so the situation is more complex than
that shown here. Equation [9.1] simply indicates that no matter how
one designs a wind turbine, the amount of power the turbine pro-
duces cannot exceed the amount predicted by this equation.)
Fourth, assuming that a power company can find a suitably
windy site that is reasonably close to its target market, it will want
to make sure that the turbine blades sweep out the largest area pos-
sible. This is done by equipping the turbines with the longest pos-
sible blades. The area, A, that is swept out by the rotor is a circle
that depends on r, the distance from the tip of the blade to the hub,
according to the following formula—a formula that every student
encounters in geometry:

A = πr 2

where π is a number that is a little larger than 3. By doubling the

length of a blade—that is, by doubling r—this formula shows that
the area swept out by the rotor will quadruple (2r × 2r = 4r 2). Con-
sequently, doubling the length of a blade quadruples the available
power. This explains why it is to the designer’s advantage to make
the blades as long as possible. The need for longer blades further
drives up the height of the tower. In order for the turbine to achieve
maximum efficiency, the lowest point reached by the blade should
be as high above the ground as practicable.
This leaves only the question of the value of C, and the answer
is that C is the same regardless of whether one considers wind cur-
rents or water currents. As mentioned in chapter 5, during theoreti-
cal investigations conducted at Göttingen University in Germany
 The Nature of Wind Power  147

during the 1920s, the German engineer and scientist Albert Betz
estimated that the symbol C in equation (9.1) was no larger than
0.59, which means that no more than 59 percent of the kinetic en-
ergy of the wind can be converted into electricity, and contempo-
rary research indicates that C is probably quite a bit less than 0.59.
Estimates vary.
Power producers that depend upon wind are especially concerned
with finding places where the wind blows reliably and rapidly. Their
choices are further constrained by the necessity of building on sites
where they will have access to high-voltage power lines. (Building
high-voltage lines is very expensive and always controversial. Suc-
cess in obtaining the necessary permits is not assured.)
Assuming, then, that a power producer receives approval to
build a wind farm and (if necessary) the high-voltage power lines
necessary to connect its site to its target market, the next question
is how should the utilities that might buy power from the power
producer compute the increase to system reliability that the wind
farm offers? A utility must produce power simultaneously with
demand. The reliability of the grid is important from a planning,
a safety, and a legal standpoint. In many areas there are legal re-
quirements that utilities maintain a reserve capacity at all times
in order to meet unanticipated events—either a surge in demand
or the failure of a particular power source. System reliability is
extremely important. How do wind farms contribute to system
reliability?
Although it is true that no power source operates indefinitely
without interruption, conventional power sources—especially coal,
nuclear, and natural gas plants, the main sources of electricity in
the United States and most other nations—are much more reliable
than is the wind. When a conventional plant is turned off, it is usu-
ally because of regularly scheduled maintenance. Predictability is
one of the key advantages of conventional power plants. To make
use of the wind, it is necessary to find a way to integrate an inher-
148  Wind and Water

ently unreliable power source into a system that must supply power
with a high degree of reliability.

Estimating Capacity
There is not yet a definitive answer to the problem of determining
what sort of contribution to the reliability of an electric grid a wind
farm makes. Different utilities and system operators place different
values on wind power. Differences in opinions arise because wind
power is often unavailable exactly when it is most needed—and here
the word “often” is relative to more conventional power producers
such as natural gas plants. But while it is evident that wind is not
as reliable as more conventional power sources, it has some value,
because it sometimes produces electricity when electricity is most
needed. This assertion does not, of course, identify the value that
one can assign to wind power; it only claims that wind power has
more than no value. But those who operate power distribution sys-
tems require a more precise statement in order to ensure the reli-
ability of the network. Reliability is one of an electrical network’s
most important properties.
An important characteristic with respect to wind turbine reli-
ability is its capacity factor. There are multiple definitions of capacity
factor, but a common one is defined to be the actual power output of
the turbine divided by the maximum possible turbine output over
the course of a year. The capacity factor is, then, the percentage of
the maximum output actually produced by the turbine in question
during one year of operation. The advantage of such a measure is
its simplicity. Its disadvantage is that it fails to entirely capture the
practical contribution made by a turbine.
In the business of power production, timing is everything.
Because electrical power cannot be stored for later use, it must be
produced simultaneously with demand. A power source that can-
not produce power when there is a demand for it is, it is generally
agreed, worthless. Very sophisticated systems are now in use to
 The Nature of Wind Power  149

Large wind farms require large amounts of land. (Indiana Office of Energy and
Defense Development)

track changes in demand and to use the least expensive supplies

to satisfy those demands. Provided that a wind farm is producing
power at a time when there is demand for it, that power can be used
in one of two ways: Either it can be used to meet demand, or, alter-
natively, the wind farm can—for as long as the wind is blowing—be
held in reserve to meet unexpected demand. Reserve power is also
important. All electrical power networks maintain a reserve capac-
ity, a safety margin, to ensure that demand will be satisfied in the
event that it were to suddenly increase or in the event that online
supplies were to suddenly decrease, a situation that would arise if,
for example, there was an unscheduled shutdown of a power pro-
150  Wind and Water

ducer. In many states, the obligation to maintain sufficient reserve

capacity, which may be as high as 15 to 20 percent of peak demand,
is a legal requirement. Electrical networks that operate without a
reasonable reserve capacity do so at their peril. California, for ex-
ample, had 2 percent reserve capacity during the year 2000. During
this time its network was highly unstable. The unpredictability of
the network was reflected in power blackouts and sharp increases in
electricity prices. Since that time, a great deal of attention has been
paid to creating and maintaining adequate reserve capacity for the
California system.
Some analysts say that a power source should be valued in ac-
cordance with its contribution to system reliability. A power source,
no matter whether it be wind, water, natural gas, coal, nuclear, or
“other,” only has value if it contributes to system reliability. The
more reliable the power source is, the more valuable it is. Unlike
the more simplistic definition of capacity described earlier, the ap-
proach about to be described takes into account both the amount of
power produced as well as the time at which it is produced.
As described in chapter 2, electrical power is usually divided
into base load power and peak power. Every electrical market has a
minimum power requirement that is characteristic of that market.
This minimum requirement is called base load power. Late at night,
for example, or early in the morning, base load power is all the
power that is needed to meet the demands of the marketplace. But
additional power—that is, power that exceeds base load demand—is
usually required during the day as schools, banks, stores, and other
predominantly daytime consumers of electrical power begin to
power up. Electricity used during periods of higher demand is
called peak power.
Utilities generally sign long-term contracts for base load power
because the demand for base load power is highly predictable. Peak
load power is less predictable, and it is often purchased day-to-day
or even hour-to-hour on what is called the spot market, and it is
 The Nature of Wind Power  151

generally supplied by other types of power-producing technologies

than coal or nuclear. Wind power, because of its intermittent na-
ture, is poorly suited for base load power, but it may be useful for
peak power production.
It is important to emphasize that no power source is perfectly
reliable. Each power source is occasionally unavailable because of
scheduled maintenance or because of mechanical failure. Mechani-
cally speaking, wind turbines are very reliable. They are, on average,
mechanically available more than 95 percent of the time, which is
another way of saying that they are broken or shut down for main-
tenance less than 5 percent of the time. But wind farms have an
additional source of uncertainty. Wind, which is their “fuel,” is only
intermittently available. But by itself this observation tends to over-
state their unpredictability. Although it is not possible to predict
wind speed one or two months into the future, weather reports can
be useful in predicting wind speeds one or two days into the future,
and such reports are especially useful for predicting wind speeds
in one or two hours. Accurate weather forecasting can enable wind
producers to participate in the spot market with some confidence.
(The spot market is the day-to-day and hour-to-hour market that
exists to meet peak power demands.) With accurate forecasting and
good timing, wind producers can bid for contracts and earn profits
in the spot market. The amount of profit depends on the timing of
the wind. Given two wind farms, if the wind usually blows during
peak hours at one site and during off-peak hours at another, the site
that produces power during peak hours is more valuable than the
off-peak site even if the off-peak site produces somewhat more total
power.
Wind data is the first requirement for estimating the potential
value of a wind turbine site. Measurements are taken hourly for at
least a year. This enables engineers to estimate both the amount of
power available at the site and the times that it will probably be pro-
duced. At most sites, some times of the year, and even some times of
152  Wind and Water

the day, are reliably windier than others. Using this data, engineers
will estimate the amount of power that the wind turbine will prob-
ably produce and when it will probably produce it. They compare
that estimate with a more predictable source of power called the
benchmark. The benchmark, which is a mathematical idealization
of an actual power plant, is often taken to be a natural gas plant
because natural gas plants are also often used for peak power pro-
duction and their performance characteristics are well understood.
Based on wind measurements and the known mechanical re-
liability of wind turbines, engineers compute the contribution to
system reliability that the wind turbine will probably make. Their
answer depends on how much power will probably be obtained from
the wind at the site as well as the times that the wind will prob-
ably be blowing. (These calculations require probability theory, that
branch of mathematics that seeks to quantify uncertainty.) Next
the power output from the benchmark producer is adjusted until
its contribution to system reliability equals that of the wind. The
contribution is calculated based on the idea that a more powerful
but intermittent source of electricity (wind energy) makes the same
contribution to system reliability that is made by a lower power,
more reliable source (the idealized natural gas plant). The calculated
output of the benchmark is noted, and becomes the effective load-
carrying capability of the wind farm. This is also sometimes called
the capacity of the wind farm. (The word capacity is used in several
different ways.) This capacity rating is the computed value of the
wind farm to grid reliability.
Depending on the site, wind turbines are generally valued be-
tween 20 and 40 percent of their full power output. Turbine manu-
facturers and turbine enthusiasts often quote a higher number, the
maximum power output of a turbine, also called the nameplate
capacity, as evidence of a turbine’s value as an emissions-free gen-
erator of electricity. But system planners and operators are guided
more by the effective load-carrying capability (or some other con-
 The Nature of Wind Power  153

ceptually similar measure) because it is a more accurate measure of

the contribution that can be expected from the turbines in actual
practice.
To see the difference between the two estimates, it is enough to
note that some wind enthusiasts will assert that wind turbines can
be substituted for conventional sources on a megawatt-for-megawatt
basis. By way of example, one might support replacing a 100-MW
natural gas-fired plant with 100 individual one-MW wind turbines,
but the preceding discussion shows that this is almost certainly not
a reasonable suggestion. A more realistic estimate is that it would
take between 250 and 500 one-MW wind turbines to provide the
same effective load-carrying capability as a single 100-MW natural
gas-fired plant. The lower effective load-carrying capability of wind
turbines is one reason that a large-scale switch to wind power would
be very expensive relative to producing an equal amount of power
with more conventional power sources.
But the fact that individual wind turbines are, due to the inter-
mittency of the wind, some of the least reliable of all power pro-
ducers does not mean wind turbines cannot be used to increase
system reliability. Probability theory shows that it is possible to
construct a system that is more reliable than each of the individual
power producers of which it is composed. By separating wind
farms geographically, it becomes less likely that the wind will fail
at all sites simultaneously. By building enough geographically
separated wind farms, the probability that all will fail to produce
power at the same time can be made as small as desired—pro-
vided, of course, that there are enough sites to accommodate the
building boom. Constructing such a system would be extremely
expensive; it would require a great deal of land, and a major and
very expensive expansion of the required high-voltage transmis-
sion network. But provided one is willing to pay the price, wind
power can make a significant contribution to system output and
system reliability.
 154  Wind and Water

Storing the Wind

T here is another way to overcome the problem of the intermittency of

wind turbines. This method makes use of the process of electrolysis.
Water molecules are comprised of one oxygen atom bound to two hy-
drogen atoms. In a process demonstrated in most high-school chemistry
classes, one can split the molecular bonds that bind hydrogen atoms
to oxygen atoms by passing an electric current through the water in a
process called electrolysis. By collecting the resulting hydrogen gas, it
is possible to store some of the kinetic energy of the wind as chemical
energy. The hydrogen gas that results from the electrolysis process is
highly flammable. When burned in a pure oxygen environment, the only
product of combustion is water—or in a somewhat different process, the
hydrogen can be used to power fuel cells, devices that convert chemical
energy directly into electrical energy without employing the combustion
process. The hydrogen produced by the turbine can be used on-site or
collected and used elsewhere.
In theory, this process has a great deal of promise. Burning hydrogen
produces no greenhouse gases and it is renewable in the sense that the
hydrogen is obtained from water and produces water when it is burned.
But electrolysis, as described here, has so far proved to be expensive and
inefficient. There are three main reasons. First, hydrogen is very difficult
to compress for storage. At low pressures, small amounts of hydrogen—
where “small” is measured in terms of its energy content—require large
storage containers and large amounts of hydrogen require huge storage
containers. Alternatively, compressing hydrogen or liquefying it so that

Wind Power, Topography,

and the Environment
As described previously, the maximum amount of wind energy that
can be converted into electricity is very sensitive to the speed of
the wind—more sensitive to wind speed than any other factor. This
 The Nature of Wind Power  155

it fits into smaller containers requires significant amounts of energy.

Both the higher pressure and lower pressure alternatives are, therefore,
costly. Second, electrolysis is not a very efficient process in the sense that
a great deal more work is required to produce the hydrogen gas than is
recovered by burning it. (Recall the discussion of the efficiency of heat
engines in chapter 6.) And although fuel cells are, in theory, more efficient
than heat engines, manufacturers have not yet found a way to produce
fuel cells that are robust enough and economical enough for daily use
in many common applications. Finally, because electricity is already a
valuable product, a producer would have to be paid a premium for the
hydrogen in order for it to be worthwhile to convert electricity that might
otherwise be sold immediately on the grid into hydrogen gas that would
later be used as fuel—especially given the inefficiencies of the additional
conversion processes.
Engineers and scientists continue to investigate all the relevant tech-
nologies associated with wind power, hydrogen production and storage,
and fuel cell manufacturing. All of these technologies are in a state of
flux. It is possible that the difficulties presented by these technologies will
be partially overcome, but it is also important to keep in mind that the
promise of hydrogen as the “fuel of the future” dates back to the 1970s.
Assertions that a new “hydrogen economy” is just over the horizon have
been made every decade since the 1970s, and hydrogen supporters con-
tinue to insist that hydrogen will soon make significant contributions to
the energy mix. Perhaps they are right.

raises two questions: Where are the best sites for development? And
what are the environmental consequences associated with develop-
ing these sites?
From a technical standpoint, the “best” sites should have strong
steady winds during periods of peak demand, and they should not
156  Wind and Water

A wind resource map that shows the annual average wind power estimated
at ca. 160 feet (ca. 50 m) above the surface of the United States (Source: U.S.
Department of Energy, NREL)

be located too far from the markets that they are intended to serve.
In the United States these sites have, for the most part, already been
identified. The federal government has expended a good deal of
 The Nature of Wind Power  157

money and effort creating what are called “wind resource maps,”
which, as the name implies, identify sites that could, in theory, be
developed to provide significant amounts of electrical power. These
maps show, for example, that the best sites occupy less than one
percent of the land area in the lower 48 states—a useful fact, but
that is only a beginning. Economically viable sites may be excluded
from development because they are in national parks, scenic areas,
wetlands, urban areas, or because the sites are home to endangered
bird or bat species. Some sites are too isolated to make development
attractive, and development of other sites is problematic because
of fierce local opposition that is often based on reasons that are
largely aesthetic. There is also the problem of scale: Commercial
wind farms require many turbines distributed across substantial
amounts of land.
Consider, first, the problem of building a wind farm on the crest
of a mountain. The area along a mountain crest, also called a ridge-
line, is often the best location for a wind turbine. In a mountainous
area, ridgelines are sometimes the only viable locations because
the valleys are sheltered from the wind by the mountains. Because
one wind turbine is not economically viable, developers seek to
place as many wind turbines as possible along a ridgeline. There
are not many ways to build an economically viable wind farm on
a ridgeline. The principal reason has to do with the wakes created
by the rotors. As wind passes through the blades of a wind turbine
some of the kinetic energy of the wind is converted into electricity.
Downwind of the rotor, the wind is weakened. It is also turbulent,
and ill-suited for driving another turbine. Consequently, those who
design wind farms never position one turbine directly behind an-
other. Along ridgelines, turbines must be positioned like beads on a
string: single file and with space between them. Proposals to build
such wind farms have sometimes run into vociferous opposition by
those who, on aesthetic grounds, object to lining ridgelines with
enormous wind turbines.
158  Wind and Water

Alternatively, consider the development of wind farms in regions

where the land is flat and the wind blows in one direction only. In
this environment, the turbines could be organized into rows and the
rows placed perpendicularly to the prevailing winds. Turbine spac-
ing is expressed in terms of rotor length, and for an array of rotors
on flat land with a steady wind, rows would ordinarily be separated
by a distance of approximately 10 rotor diameters to allow the turbu-
lence created by the upstream row the time and distance necessary
to dissipate before encountering the next row of turbines. Within
the same row, three to four rotor diameters between neighboring
turbines is considered a reasonable distance. Because turbines have
large rotors—230 feet (70 m) from tip to tip is not unusual—this
means that rows would be separated by at least 2,300 feet (700 m,
or about 0.5 mile), and within the same row, neighboring turbines
would typically be from 690–920 feet (210–280 m) apart. Conse-
quently, a wind farm consisting of three rows of 20 turbines would
require at least two square miles (6.7 km2) of land.
If the turbines in this model wind farm are rated at three MW
each, the farm would have a maximum output of 180 MW and an
effective load-carrying capability of between 36 and 72 MW—not a
very large output, but spread out over a fairly large amount of land.
Another way of describing the situation is that the wind farm could
be expected to produce (on average) 28–56 kW per acre (69–138
kW/hectare). Evidently, wind energy on an industrial scale is land-
intensive, and there is little to be done about it. Placing the rows
much closer together risks making the turbines downwind of the
leading row economically unviable—that is, the output from down-
wind turbines may not be enough to justify their construction if
they are too close to the row upstream. Placing neighboring turbines
much closer than about three rotor diameters has the same effect.
By way of contrast, the Edwin I. Hatch Nuclear Plant site, which is
located near Baxley, Georgia, and is home to two nuclear reactors,
occupies 2,240 acres, most of which is used for timber production
 The Nature of Wind Power  159

and wildlife habitat. The actual reactor site, which houses the reac-
tor containment buildings, the cooling towers, and all structures
associated with both reactors, occupies only about 300 acres. The
two reactors on this site produce more than 1,700 MW of electricity
or about 5,700 kW per acre (14,000 kW/hectare).
The preceding discussion may make it seem that one of the main
disadvantages to large-scale wind development is that producing
commercially valuable amounts of wind power requires large tracts
of land. By itself, however, the requirement for large amounts of
land does not necessarily lead to land-use problems, because wind
turbines can sometimes share the landscape. Farmers, for example,
have sometimes found it profitable to lease some of their land to
power producers. The rotors spin high above the crops, and the
amount of land required for the tower is relatively small. So the pay-
ments from the power producer to the farmer can more than offset
the loss of income suffered by the farmer due to the relatively small
amount of arable land taken out of production in order to make
room for the turbines and associated infrastructure.
It is, of course, not just the size of the wind farm but its loca-
tion that matters. As noted previously, the best locations occupy
less than 1 percent of the landmass of the contiguous United States.
(One can, of course, use sites where the wind is less powerful or
less reliable, but building on less-productive sites also means build-
ing significantly more turbines to achieve the same power output.
Producing power from wind energy at those less-desirable sites is
less profitable and so less attractive as an investment opportunity.)
Good sites for wind power are natural resources, as unique and ir-
replaceable as oil fields and coal mines.
 10
Wind Energy: Economic
and Public Policy
Considerations
E lectricity is a commodity, an article of commerce, and it can be
produced by a wide variety of methods. Because wind energy
is only one method among many for producing electricity, it must
compete with other technologies in what many like to call “the” free
market. There are, however, many free markets. Each nation creates
its own legislation to establish the rules by which its free markets
operate, and it establishes its own regulatory authorities to monitor
compliance. There are some who insist on philosophical grounds
that when it comes to free markets fewer rules are better, but this
belief is not always supported by the available data. The actual situ-
ation is far more complex.
Today, the principal methods for generating electricity in the
United States are coal, nuclear, and natural gas. If wind energy
is to become a significant contributor of electrical energy, it will
have to grow at the expense of one or more of these more conven-

160
Wind Energy: Economic and Public Policy Considerations  161

U.S. Capitol. The federal government has been trying to create a wind-energy
sector since the 1970s. (Architect of the Capitol)

tional sources of electrical power—that is, it is not enough for wind

power to grow at the same rate at which the demand for electricity
is growing. Rather, in order to displace other, more conventional
power sources, wind power must grow faster than the demand for
electricity. This is no easy task. A substantial infrastructure has
evolved over the years to support coal, nuclear, and natural gas
power plants. With respect to energy policies, each type of power-
production technology has its own political constituency, and each
fills a particular niche in the market. Natural gas plants, for ex-
ample, make use of a highly developed production and distribution
system to produce peak power (and sometimes base load power),
and the base load producers, nuclear and coal, currently benefit
from a highly developed infrastructure and substantial govern-
ment subsidies of their own.
As described in chapter 7, the wind-power industry has also
benefited from heavy government price subsidies and the benefits
162  Wind and Water

of government-funded research. But these subsidies are insufficient

to guarantee that wind power will eventually displace significant
amounts of more conventionally produced power. All power-gen-
eration technologies continue to develop and become more efficient
as power producers struggle to either reduce their costs or to shift
their costs onto others with the goal of becoming more competi-
tive. If wind is to displace other power production technologies it
must become more efficient than those other technologies or it must
receive more in the way of subsidies. From a producer’s point of
view, of course, increased subsidies and increased efficiencies are
not mutually exclusive goals.

The Costs of Wind Power

As has been demonstrated repeatedly in this volume, each power-
generation technology has its own unique costs. There are envi-
ronmental costs associated with every form of large-scale power
generation, and there are the costs of financial subsidies used to
encourage the further development and deployment of each tech-
nology. These costs are not always reflected in the prices charged
to the consumer. In fact, they almost never explicitly appear in the
monthly utility bill. They may, instead, appear in the form of higher
taxes, increased environmental problems, or changes in govern-
ment spending priorities—money may be shifted from health care
to energy subsidies, for example.
There is, as has been previously mentioned, no generally agreed
upon method of assigning value to particular environmental out-
comes. It may be known, for example, that a particular power tech-
nology disrupts the environment in a particular way, but it is often
less clear how to assign a cost to that disruption relative to the costs
associated with another type of power technology. No modern nation
can survive without access to large amounts of reasonably priced
electrical power. Therefore, the value of any power technology can
only be weighed relative to the available alternatives.
Wind Energy: Economic and Public Policy Considerations  163

Financial subsidies are not always easy to recognize, either. In

a budget as large and complex as that of the United States, for ex-
ample, subsidies are often distributed among numerous agencies,
and often they are not labeled as subsidies at all. In addition, higher
taxes and spending shifts within the budget are practically always
the result of numerous pressures, most of which have little to do
with the energy industry. The costs of power production are real,
but a full accounting of the costs associated with any method of
power production has never been carried out—partly because of the
difficulties just described and partly because the concept of cost is
continually evolving.
To be sure, some of the subsidies that some power producers
have received are related to campaign contributions, political pa-
tronage, and a host of other factors that sometimes gives politics
a bad name, but which subsidies fall in the “undesirable” category
is open to debate. As a general rule, energy subsidies simply reflect
the fact that the provision of affordable and reliable electricity is an
essential service upon which modern life depends. Electricity is a
public good as well as a source of private wealth.
Some of the subsidies directed toward wind-power produc-
ers—the federal production tax credit and research conducted or
at least sponsored by the Department of Energy, for example—
were described in chapter 7. And some states—New York and
California, for example—have historically also offered generous
subsidies for wind-power producers. Today, U.S. wind-power pro-
ducers benefit from substantial subsidies no matter where in the
country they are located, although some localities still offer more
financial support than others. Each year hundreds of millions of
dollars in public monies are spent making wind power attractive
to investors, who, without those subsidies, would likely spend
their money elsewhere. Many of the taxpayers on whose behalf
the subsidies are distributed do not use electricity produced by
wind turbines.
164  Wind and Water

And some wind-power costs will probably rise faster than the
rate of installation of new wind facilities because the number of
ideal sites is limited. As discussed in chapter 9, building a large
number of turbines requires a great deal of land because turbines
need to be placed far apart to prevent each turbine from interfering
with the operation of its neighbors. Trees cannot be permitted to
grow near turbines because they interfere with the flow of the wind.
Turbines cannot be built near residential structures for two reasons:
First, the buildings interfere with the wind just as the trees do, and
second, turbines are noisy when in operation. Wind farms, there-
fore, require large amounts of inexpensive, sparsely populated, un-
forested land situated far from the nesting sites or migratory flight
paths of rare or endangered bird and bat species. As the best sites
are developed, secondary sites—that is, less-profitable sites—will
be developed, and investors will require even higher subsidies to
build in these locations. This is why it is reasonable to expect that
subsidies will probably increase at a rate exceeding the rate at which
new facilities are installed.
Wind farms can also be situated at sea, and a few offshore wind
farms have already been built or are under construction in vari-
ous nations around the world, but at present offshore projects are
even more expensive than those on land and their successful opera-
tion requires even heavier subsidies. At present, it is not clear how
much electricity from offshore wind farms consumers will be able
to afford.
In addition to all of these restrictions, there is also the ques-
tion of market access. Often wind farms are located far from the
consumers who require electricity. By way of example, in New York
(as of this writing) there is a good deal of controversy surround-
ing attempts to build a 200-mile (320-km) corridor for high-voltage
power lines to connect upstate wind producers with downstate con-
sumers. The proposal is controversial among those who will have
to surrender some of their land for the corridor. From the point of
Wind Energy: Economic and Public Policy Considerations  165

view of the energy producers, however, the need for the project is
obvious. Unless a way is found to connect their wind farms with the
downstate energy markets, there is little point in developing further
upstate capacity.
When all of these factors are taken into account, it is apparent
that large-scale wind power costs a lot. But so do the alternatives. The
energy requirements of modern societies are enormous, and meet-
ing those requirements requires large-scale projects. If wind power
becomes an important part of the energy mix—in the United States
it currently supplies less than 2 percent of the total—it is reasonable
to expect that the environmental and economic costs of wind will
rise at least as fast as the output from the wind farms. There is no
easy way to satisfy the energy requirements of modern societies.

The Role of Economic Class

The United States was once the leader in installed wind capacity,
but despite a continuing commitment to wind power, it has not
completely kept pace with developments elsewhere. Building a
wind farm in the United States has become increasingly difficult.
To see the problems faced by independent power producers as they
attempt to increase wind capacity, consider the difficulties thus far
encountered by Cape Wind, the company that hopes to develop the
nation’s first offshore wind farm. This wind farm will, if approved,
be built just off the coast of Massachusetts on approximately 25
square miles (65 km2) of Nantucket Sound.
Nantucket Sound is bounded by Cape Cod and the islands of
Martha’s Vineyard and Nantucket. Cape Cod and “The Islands”
have some of the highest electricity prices in the nation, and in
2005 state regulators approved a 50 percent increase in electric
rates for these consumers. About 60 percent of the electricity used
by consumers in this area is produced by fossil fuel plants, and
most of this is produced by burning expensive natural gas and
oil. In addition to high-priced electricity and the accompanying
166  Wind and Water

Brant Point Lighthouse on Nantucket Sound. If the Cape Wind project is built,
it will be located miles from any land. (Nelson Fontaine)

greenhouse gas emissions, the power production capacity of the

network has not kept pace with growing demand. ISO–New Eng-
land, the organization responsible for ensuring sufficient generat-
ing capacity throughout New England, has consistently warned of
impending shortfalls in electricity production. One would think,
therefore, that construction of a 130-turbine emissions-free power
project would be welcome in an area with an impending energy
shortage, high energy costs, and a frequently expressed commit-
ment to support clean energy, but this has not been the case.
Opponents to the project include author and radio personality
Robert F. Kennedy, Jr., and his uncle, U.S. Senator Ted Kennedy,
Wind Energy: Economic and Public Policy Considerations  167

as well as the Alliance to Protect Nantucket Sound, which enjoys

a multimillion dollar budget, former Massachusetts governor and
past 2008 presidential candidate Mitt Romney, former Massachu-
setts attorney general Tom Reilly, U.S. Congressman William Dela-
hunt, whose district includes Cape Cod, and other organizations
and political figures. What are the objections?
During the course of the debate about the Cape Wind project,
various environmental objections have been raised. By way of ex-
ample, one early source of expressed concern was the effect of the
turbines on birdlife. Wind turbines kill birds—this is well estab-
lished—and when turbines are placed in areas with many birds, they
will kill many birds. The key to minimizing the impact of a wind
farm on avian life is to locate the turbines far from large populations
of birds and even to locate them away from small populations of
endangered birds. A thorough study of Nantucket Sound by inde-
pendent scientists was undertaken in order to predict the probable
impact of the project on birdlife. Scientists determined that most
birds avoid Nantucket Sound, and radar studies have shown that
although some migrating birds fly over the area, they fly at altitudes
much greater than the proposed maximum height of the turbine
blades. The impact on birdlife would, therefore, be small. Having
established that this concern was without merit, objections simply
shifted elsewhere.
Project opposition was, if not widespread, at least well financed.
In 2004, the Alliance to Protect Nantucket Sound received $4.67 mil-
lion in contributions, $2.9 million (or 62 percent) of which came from
just 15 individuals. As each objection was raised, it was answered.
Opponents, however, remained steadfastly opposed, although as their
objections were answered they began to offer less and less in the way
of explanation. In June 2006, for example, Senator Kennedy simply
noted that, “There are still some outstanding questions with respect
to the project,” and Representative Delahunt noted that “. . . this proj-
ect is destined for litigation that will be interminable.” ­Neither was
168  Wind and Water

more specific. Opponents remained committed to preventing the

project even after the 2007 announcement by Massachusetts Envi-
ronmental Secretary Ian Bowles that the proposed Cape Wind project
“adequately and properly complies” with state environmental
requirements.
Of course, not everyone is opposed to the project. Voters, for
instance, have for the most part not expressed their opposition
at the ballot box. In the September 2006 Democratic primary for
governor of Massachusetts, 53 percent of the voters on Cape Cod
voted for Deval Patrick. Mr. Patrick was a supporter of the Cape
Wind project. Christopher Gabrielli, another supporter of the Cape
Wind project, was also a candidate in that primary. Together he and
Mr. Patrick garnered 74 percent of the vote on Cape Cod, Martha’s
Vineyard, and Nantucket. Attorney General Tom Reilly, who was
also running for the Democratic nomination for governor, and who
vociferously opposed the Cape Wind project, ran a distant second
to Mr. Patrick among voters on the Cape. In November of that year
Mr. Patrick ran for governor in the general election against Repub-
lican Lt. Governor Kerry Healy. She was also firmly opposed to the
Cape Wind project. Mr. Patrick won the general election.
What opponents to the project do not dwell upon is that the
Kennedy family compound, which is located in Hyannis Port on
Cape Cod, overlooks Nantucket Sound, as do many other expen-
sive homes. The residents of these homes will see the wind turbines
on the distant horizon when they look toward Nantucket Sound.
Moreover, placing large wind turbines in the Sound may interfere
with recreational boating. Perhaps for the first time, affluent con-
sumers of electricity on Cape Cod are being asked to shoulder some
of the burdens associated with electricity production—even if these
burdens only affected their leisure activities and their ocean views.
While every large-scale power technology impacts the lives of
many people, the burdens associated with production are never
distributed equally. The residents of Appalachia have, for example,
Wind Energy: Economic and Public Policy Considerations  169

long borne much of the brunt of the United States’ pro-coal policies.
Mining activities have cost the lives of many mine workers and the
health of many more; burning coal to produce power in the Mid-
west has had a profound effect on the forests and freshwater wildlife
of the Northeast as pollution from Midwestern power plants has
settled on the forests of New York and New England; transporting
oil around the globe is a business that has long been punctuated
by one environmental disaster after another; and oil refineries have
adversely affected the health of some refinery workers. As a general
rule those most able to afford the power produced by these tech-
nologies have been the least affected by the negative aspects associ-
ated with their use.
Historically, some groups, usually low-income groups or those
identified as minority groups, have borne most of the environmen-
tal and health impacts of energy production technologies. This con-
tinues to be true. In response, those concerned with social justice
have developed the concept of “environmental justice,” the idea that
policymakers should avoid crafting policies that disproportionately
and adversely affect the health and environment of those commu-
nities least able to protect themselves. While everyone agrees that
environmental injustices are unfortunate, not everyone agrees on a
solution. Those advocating environmental justice assert that when
the benefits of particular technologies are enjoyed by many, public
policy should be crafted so that the consequences of implementing
those technologies are more evenly distributed among the benefi-
ciaries. As the experiences on Cape Cod demonstrate, the equitable
sharing of the burdens associated with power generation and distri-
bution is an ideal that is difficult to put into practice.
The lack of commitment to environmentally just policies is par-
ticularly evident in the development of wind power, because wind
farms cannot always be sited among the poor and less powerful.
Wind farms must be sited where the wind blows, and some of the
sites best suited for wind power development are precisely those
170  Wind and Water

that have already attracted high-priced homes with views of moun-

tain ridgelines or the open and relatively placid ocean vistas to be
found in sounds and bays along the nation’s coasts. Because wind
farms require many large wind turbines, the visual impact of these
projects cannot be entirely hidden from the view of those who live
in these areas. Consequently, despite the fact that wind power is
one of the most environmentally benign of all energy technologies,
the construction of wind farms has sometimes proved to be highly
controversial for reasons that are neither scientific nor economic.
Simply put: Often the major objection to wind farms—and this is
certainly the case for the Cape Wind project—is that those who will
benefit from the power do not like the view.

The Future of Wind Power

How large a contribution can wind power make toward meeting
the ever-growing demand for electricity in the United States and
elsewhere? Although the question sounds simple, the answer is
not. First, it is important to emphasize that wind power is only one
technology among several. The goal of investors is not to produce
power from wind but rather to produce power profitably. Profits can
be the result of increased efficiencies or government subsidies, but
without profits nothing will be built. Subsidies can flow directly to
the wind-power industry as is the case with the production credits
currently enjoyed by U.S. wind producers, or government support
for wind can be indirect as would occur if the government were to
tax power producers for emitting greenhouse gases, thereby creat-
ing a disincentive to future investment in fossil fuel plants, a policy
adopted in Denmark. But while a carbon tax could benefit wind
power, there is no guarantee that it will. Investments might be di-
rected toward other technologies—tidal power, for example, or wave
power—provided that tidal power or wave power are more attractive
investments than wind power at the time that investors make their
decisions. The effects of government policies on energy markets are
complex, and historically, they are full of unintended consequences.
Wind Energy: Economic and Public Policy Considerations  171

But the government, which establishes the legal regime in which

energy markets operate, cannot retreat from its responsibilities in
this area. One can only hope that legislators will approach the task
with caution and humility, in part because there are many previous
policies about which they have reason to be humble.
From a technical point of view, engineers have made numerous
important improvements in wind technology. Since government re-
searchers first became involved in the 1970s, the price of producing
one kilowatt-hour of electricity from the wind has decreased by an
order of magnitude—that is, it costs only about one-tenth as much
to produce a kilowatt-hour of electricity from wind today as it did
in the 1970s. The improvements have been incremental and largely
the result of government-sponsored wind research. Small changes
in blade designs, increasing the scale at which the machines are
built, and other innovations have all had a cumulative effect, and
the price of a kilowatt-hour of electricity produced by an “ordinary”
wind turbine is still coming down.
Early wind projects were concerned primarily with developing
land-based wind farms situated in areas with strong reliable winds,
and while these remain important, emphasis has begun to shift to
developing two other types of sites. First, engineers are increasingly
looking toward the oceans. Offshore winds, free of the obstructions
of mountains, trees, and buildings, tend to blow more predictably
than over land. There are also fewer interests competing for the use
of offshore sites than for land-based sites, and finally, as with the
Cape Cod project, many offshore sites are located near the consum-
ers who need the power. By contrast, many sites with strong winds
that are located on land tend to be situated far from consumers.
Second, researchers are interested in developing regions where
the wind speed is lower. The reason is that regions with lower but
still commercially viable wind speeds are 20 times more common
than regions with higher wind speeds. What one loses in output
per turbine, one can gain by deploying more turbines. Of course,
deploying multiple turbines simply to produce the same amount of
172  Wind and Water

power that a single turbine could produce in a region with higher

and more reliable wind speeds is a recipe for higher-priced electric-
ity. The goal is, then, to make these lower–wind speed turbines as
cheaply and as efficiently as possible.
While wind enthusiasts claim that there is enough wind energy
available within the United States and just off its coasts to meet a
large fraction of the total electric needs of the United States, there
are real questions about whether this energy can ever be exploited
at a cost that consumers are willing or even able to pay. Further-
more, although the average amount of power that can be generated
by the wind is large, consumers care little about averages. Instead,
they want—they need—power at the instant that they turn on lights,
computers, televisions, stoves, or heaters. In this sense, intermit-
tence remains the central problem with wind power, but this is a
problem that may have a solution as engineers find better ways to
integrate wind power into the grid. This research is ongoing, and
there are promising solutions under consideration. Time will tell if
these solutions are practical.
Even if all technical problems are solved, wind power develop-
ment may be stymied unless creative government policies can be
developed that enfranchise those most likely to object to a particu-
lar project, the individuals living within sight of the wind farms.
Such policies have already been crafted in Denmark. Perhaps simi-
lar policies will be adopted in other countries as well.
Currently, energy markets are in a state of flux as power produc-
ers seek to respond to increasing concerns about the cost of electric-
ity and the environmental impacts associated with its production.
New technologies, some of which have been described in previous
chapters, may soon be competitive with wind. Some are, in fact,
already competitive. While the long-term outlook for wind is diffi-
cult to predict, for the next several years it is likely that wind power
will continue to grow at a rate that exceeds the rate of increase of
electricity demand.
 Conclusion
T he technologies described in this volume can be divided into
two classes: (1) conventional hydropower, and (2) everything
else. Conventional hydropower is a mature technology. To be sure,
engineers still make occasional improvements in turbine design,
for example, and new ideas, such as pumped storage, are still oc-
casionally introduced, but conventional hydropower is a mature
technology in the sense that its limitations are well understood, and
its potential has, for the most part, already been developed. All the
other technologies in this book are still under development, and
the potential of each of these methods for generating electricity is
still a matter of some debate. Innovations in product design—as
well as innovations in legislation that might lead to greater market
penetration—are almost assuredly waiting to be discovered and
implemented.

173
174  Wind and Water

There are real advantages to developing these technologies.

Emissions-free electricity is, of course, the principal advantage,
but another important advantage is energy security. A disadvan-
tage shared by all the technologies described in this volume except
conventional hydropower is that the energy sources are all diffuse.
There are, to be sure, enormous amounts of energy in the oceans’
currents, Earth’s winds, and the differences in temperature between
the oceans’ upper and lower layers, but there is not much energy at
any given location. Harnessing diffuse sources of energy will al-
ways be expensive. By way of example, one can power a city with
the output of a single nuclear reactor occupying a total of only a
few hundred acres of land, but one would need to place thousands
of wind turbines along many miles of coastline to accomplish the
same goal. Even under these conditions the nuclear plant’s output
would be more reliable and more under the control of its operators
than would the output of the many turbines that would be required
to produce the same average power output as the nuclear plant.
As a practical matter, how much power is available from the
technologies described in this book? The answer to this question
is not yet clear—even in the case of wind, which is, with the ex-
ception of conventional hydropower, the most developed of all the
technologies. Problems associated with intermittence of supply, lo-
cal opposition, and the price of wind-generated electricity all con-
tinue to pose barriers to large-scale development. Subsidies help,
of course, but while it is easy for any nation to afford the subsidies
necessary to keep 100 turbines in operation, ensuring the profit-
ability of 100,000 turbines would tax the budgets of every nation.
As a general rule, as the number of turbines (and the subsidies they
require) increase, public support can be expected to decrease. The
future of any power-production technology is, in part, a matter of
national priorities. Which, for example, is more deserving of sup-
port: wind energy, wave energy, space exploration, higher education,
national defense, or health care? Each program has its proponents,
 Conclusion  175

but because even national budgets are finite, compromises must be

made. It is unclear whether the power technologies described in this
volume—with the exception of conventional hydropower—will ever
produce more than a tiny fraction of the United States’ power out-
put. This is not to rule out the possibility of large-scale low-subsidy
or subsidy-free electric power from these technologies, only to point
out that this has yet to be accomplished.
Energy is one of the most fundamental problems of the 21st cen-
tury. The ability to produce large amounts of electricity with mini-
mal environmental disruption and at a reasonable cost is necessary
if future generations are to build upon today’s accomplishments.
Each generation wants to see the next generation do better—a better
standard of living, a better health-care system, better science, more
peaceful relations between nations, an improved environment—but
a necessary precondition for all of this is that solutions are found to
the problems currently associated with energy production. Whether
this will be accomplished remains to be seen.
 Chronology

2900 b.c.e. Egyptians build first known dam across the Nile
85 b.c.e. Earliest reference to a watermill (in a poem by the Greek
poet Antipater)
ca. 650 Vertical axis windmills in use in Persia
ca. 1100 Horizontal-axis windmills appear for the first time in
Europe
1826 Lowell, Massachusetts, incorporated as a town. By 1850, it
will be world famous as an industrial center dependent on
entirely hydromechanical power
1827 First water turbine invented by the French inventor and
engineer Benoit Fourneyron
1881 French scientist Jacques-Arsène d’Arsonval suggests the
possibility of building what is now known as an ocean
thermal energy converter (OTEC)
1889 Design for an impulse turbine is patented by the American
engineer and inventor Lester Allen Pelton
1891 Danish inventor and teacher Poul la Cour begins to work
on wind energy
1896 The hydroelectric facility at Niagara Falls begins
transmitting power to Buffalo, New York, a major
engineering achievement

176
 Chronology  177

1930 French scientist and engineer Georges Claude builds the

world’s first OTEC plant. It is deployed off the coast of
Cuba
1941 Grand Coulee hydroelectric project completed
1970 Aswan High Dam completed in Egypt
1973 OPEC oil embargo
1978 The United States passes PURPA (Public Utility Regulatory
Policies Act), the first concerted federal effort to encourage
the construction of “alternative” energy sources
1979 U.S. funding for wind-energy research exceeds $50 million.
The Raccoon Mountain Pumped Storage plant, a project of
the Tennessee Valley Authority, is brought online. At the
time, it is the largest such facility in the world.
The first OTEC plant to produce usable amounts of energy
(15 kW) goes into operation off the coast of Hawaii
1981 Consortium of Japanese companies put an OTEC plant into
operation on the island nation of Nauru.
Denmark begins offering a production subsidy for wind
power producers
1982 U.S. government funding for wind research is cut to $16.6
million
1983 The Itaipu hydroelectric power plant begins operation. (The
last unit is brought online in 1991.) It supplies 78 percent
of Paraguay’s electricity and 25 percent of Brazil’s electricity
needs
1985 California achieves 1,000 MW of installed wind-generated
power capacity
1989 U.S. Department of Energy funding for wind-power
research reaches its lowest point since 1978
 178  Wind and Water

1990 The United States has 2,267 MW of installed wind-

generated capacity.
Denmark has 300 MW of installed wind-generated capacity.
Germany has somewhat less than 100 MW of installed
wind-generated capacity
1991 Germany passes the Electricity Feed Law guaranteeing
wind producers a market for their higher-priced power
1992 The U.S. Energy Policy Act (EPAct) is passed, providing
production credits to wind energy producers
1997 Germany surpasses the United States as the country with
the largest wind energy capacity
2000 The German government passes the Renewable Energy
Act with the goal of doubling the amount of electricity
produced by renewable sources by 2010
2001 First Limpet (Land Installed Marine Power Energy
Transmitter) begins operation on the isle of Islay off the
coast of Scotland
2003 The Nathpa Jhakri hydroelectric power project, located in
India and one of the more demanding civil-engineering
projects of its era, begins operation
2004 First Archimedes Wave Swing pilot plant is connected to
the grid.
Brazil produces 83 percent of its electricity from
hydropower
2005 The U.S. Energy Policy Act of 2005 increases subsidies
in wind power in order to make the technology more
attractive to investors
2006 Europe installs 7.6 GW of wind power nameplate capacity
in a single year for a cumulative wind power nameplate
capacity of 48 GW.
 Chronology  179

The United States installs 2.45 GW of wind power

nameplate capacity in 2006 for a cumulative wind power
nameplate capacity of 11.6 GW
2007 The first commercial wave farm using Pelamis technology
is installed off the coast of Portugal.
Scottish Power announces plans to install a Pelamis-type
wave farm
 List of Acronyms

AC alternating current
AWS Archimedes Wave Swing
DC direct current
DOE Department of Energy
EPACT Energy Policy Act
FERC Federal Energy Regulatory Commission
GW gigawatt
kW kilowatt
MW megawatt
OPEC Organization of Petroleum Exporting Countries
OTEC ocean thermal energy converter
OWC oscillating water column
PURPA Public Utility Regulatory Policies Act of 1978
R & D research and development
rpm revolutions per minute

180
 Glossary

alternating current  an electrical current that reverses direction at

regular intervals
amplitude  in a wave, one half the distance from the peak to the trough
base load  the minimum amount of electrical power delivered over a
given period; the total amount of power minus transient increases
in demand
capacity  (1) actual power production divided by the theoretical maxi-
mum amount possible, also called the capacity factor; (2) the theo-
retical maximum rated output of power production, also called the
nameplate capacity; (3) effective load-carrying capability
conventional hydroelectric power  a power plant in which the electri-
cal power is produced by a flowing stream that is usually regulated
by a dam
cost  (of power) the total value that must be surrendered in order to
obtain a given amount of electrical power
direct current  electrical current that flows in one direction only
effective load-carrying capability  a measure of the contribution that
a generating unit makes to grid reliability
efficiency  the ratio of the energy supplied to a machine to the energy
supplied by it
generator  a device for converting mechanical energy into electrical
energy
global warming  the phenomenon marked by a gradual increase in
average global temperatures caused by changes in the composition
of Earth’s atmosphere
greenhouse gases  those gases that when added to the atmosphere in
sufficient quantities cause global warming

181
182  Wind and Water

hydraulic head  the vertical distance between the surface water

upstream and the surface water downstream of a hydroelectric power
plant
hydraulic ram  a type of pump that requires no external source of
power other than the kinetic energy of the fluid flowing through it
impulse turbine  a turbine commonly found in hydroelectric facilities
with hydraulic heads in excess of 1,000 feet (1,600 m)
isoquant  in a coordinate system where each point h on the positive
part of the horizontal axis represents a value for the hydraulic head
and each point q on the positive part of the vertical axis represents
a volumetric flow rate (so that p = hq is proportional to the power
of a stream of water with hydraulic head h and volumetric flow rate
q), an isoquant is the curve consisting of all points (h,q) such that hq
equals a given value of p
kilowatt  1,000 watts
linear generator  a device for converting linear motion directly into
electricity
megawatt  one million watts
nameplate capacity  the maximum power output of a wind turbine
under ideal conditions
peak power  the amount of energy needed to meet electrical demand
that is over and above the base load demand
penstock  a conduit in a hydroelectric facility connecting the water
supply to the turbine(s)
post mill  a type of traditional European windmill in which the build-
ing that houses the windshaft and mill machinery rotates on a single
post in order to face the wind
price  the amount of money charged to a consumer for electricity, a
quantity that often has little relationship to the electricity’s actual
cost
production credit  a subsidy paid to a power producer, the amount of
which is proportional to the amount of power produced
pumped storage facility  a hydroelectric facility that generates power
during periods of peak demand by releasing water that was pumped
into an elevated reservoir during off-peak hours
reaction turbine  a turbine designed for a hydroelectric facility with a
medium or low (less than 600 feet [180 m]) hydraulic head
 Glossary  183

rotor  (1) that part of a wind turbine consisting of the blades and the
hub on which they are mounted (2) that part of a generator that
rotates within the stationary cylindrical device called the stator
sail  in a windmill, the large surface consisting of a (usually) wooden
lattice and a canvas or wooden covering. It transmits the force of the
wind to the windshaft
step-down transformer  a device used to reduce the voltage of an
alternating current
step-up transformer  a device used to increase the voltage of an alter-
nating current
subsidy  a grant by a government designed to facilitate research into,
development of, deployment of, or operation of a particular power-
generation technology
tidal barrage  a hydroelectric facility that depends on the rise and fall
of the tides in order to generate the necessary hydraulic head
tidal mill  a device conceptually similar to a wind turbine that converts
the energy of flowing waters generated by the tides into electricity
tower mill  a type of traditional European windmill. It is built in the
shape of a truncated cone, and the windshaft is housed in a move-
able cap
transformer  a device used to change the voltage of alternating current
turbine  a device used to convert the linear motion of a moving liquid
or gas into rotary motion
voltage  a measure of the difference in electrical potential, it is some-
times called the electrical pressure in analogy with water pressure
windshaft  in a windmill, the shaft to which the sails are attached; in a
wind turbine, the shaft to which the rotor is attached
 Further Resources

The means by which nations produce their electrical power has be-
come a very controversial topic. As this book indicates, there are
real physical constraints on every type of power-production tech-
nology, and the value of any particular technology is determined, in
part, by the nature of these constraints. The better these limitations
are understood, the more accurately the value of each technology
can be assessed. The following books and Web sites emphasize the
science of power production.

Berinstein, Paula. Alternative Energy: Facts, Statistics, and Issues.

Westport, Conn.: Oryx Press, 2001. The author has a tendency
to try to sell each type of energy described, but the chapters
contain many interesting facts not easily found elsewhere.
Boyle, Godfrey. Renewable Energy: Power for a Sustainable Future.
2nd ed. Oxford: Oxford University Press, 2004. Highly recom-
mended. This textbook is almost 500 pages long. Readers of the
present volume will be well-prepared to tackle those sections
of Boyle’s work that concern wind and waterpower. Renewable
Energy is as advanced and complete a treatment of this impor-
tant topic as one can find that does not use calculus.
Hay, Duncan. Hydroelectric Development in the United States,
1880–1940. Washington, D.C.: Edison Electric Institute, 1991.
Most of the big projects in the United States were built or at
least under construction by 1940. This book conveys something

184
 Further Resources  185

of the impact that these projects had on the development of the

United States as well as some of the design considerations that
went into the projects themselves.
Hostetter, Martha, ed. Energy Policy. Reference Shelf, vol. 74, no. 2.
New York: H.W. Wilson, 2002. A collection of articles, reprinted
from reputable magazines and newspapers, discussing various
aspects of energy and energy policy. Very interesting.
Houghton, John. Global Warming: The Complete Briefing. 3rd ed.
Cambridge: Cambridge University Press, 2004. Global warming
is the topic that drives much of the interest in so-called alterna-
tive forms of energy. Global Warming offers a thoughtful and
reasonably thorough description of what is presently known
about the phenomenon of human-induced climate change.
Hunt, V. Daniel. Windpower: A Handbook on Wind Energy
Conversion Systems. New York: Van Nostrand Reinhold, 1981.
During the late 1970s and early 1980s, a number of high-qual-
ity books about wind energy aimed at a general readership
were published. This is one of them. While the technology has
changed a lot in the intervening years, wind has not. This book
is still a first-rate source of information on the physics of wind
and the principles involved in the conversion of wind energy to
electrical energy.
National Research Council, Committee on Nuclear and Alterna-
tive Energy Systems. Energy in Transition, 1985–2010: Final
Report of the Committee on Nuclear and Alternative Systems,
National Research Council, National Academy of Sciences. San
Francisco: W.H. Freeman, 1980. Highly recommended to all
those who enjoy reading books that make predictions about
the future of energy research and development. This almost-
700-page book was written by some of the best engineers and
scientists in the United States at the time, and they got almost
everything wrong. For example, only four pages of this tome
are devoted to wind energy, but whole chapters are devoted to
186  Wind and Water

breeder reactors, controlled nuclear fusion, and solar energy. It

turns out that in the United States wind energy now produces
more power than all three of those other sources combined.
Paton, W.R., trans. The Greek Anthology. Vol. 3. New York: G.P.
Putnam, 1918. This book is the source of the description of the
ancient watermill found in chapter one.
Pool, Robert. Beyond Engineering: How Society Shapes Technology.
New York: Oxford University Press. 1997. While not about wind
and waterpower specifically, this book addresses the interplay
between society and technology and how the evolution of each
affects the other. This subject is of vital interest to those who
want to change the way power is generated.
Shaw, Jane S., and Manuel Nikel-Zueger. Energy. Critical Thinking
about Environmental Issues. Farmington Hills, Mich.: Green-
haven Press, 2004. A general discussion about energy and the
environment. Elementary but well-written.
Williams, Wendy, and Robert Whitcomb. Cape Wind: Money Ce-
lebrity, Class, Politics, and the Battle for Our Energy Future. New
York: Public Affairs, 2007. A revealing report on the difficulties
involved in obtaining approval to build the United States’s first
offshore wind farm.

Internet Resources
Energy production has become a very controversial subject. Each
technology has its proponents, and it is not always easy to find
an evenhanded description of how a particular method of power
generation works. As you read, be skeptical!

Danish Wind Industry Association. “Wind Turbines Deflect

the Wind.” Available online. URL: http://www.windpower.
org/en/tour/wres/tube.htm. Accessed August 13, 2008. As a
result of the pioneering work of Danish scientist and inventor
Poul la Cour, Denmark has been involved in the development of
 Further Resources  187

wind-generated electrical power right from the beginning. This

informative site gives an excellent overview of the basics sur-
rounding wind power.
Elliot, D.L., C.G. Holladay, W.R. Barchet, H.P. Foote, and W.F.
Sandusky. “Wind Energy Resource Atlas of the United States.”
Available online. URL: http://rredc.nrel.gov/wind/pubs/atlas/.
Accessed August 13, 2008. Prepared for the U.S. Department of
Energy, this document discusses the basics of wind, the raw ma-
terial on which wind turbines depend. There are wind resource
maps as well as explanations that convey the technical details of
how wind energy is quantified.
Energy Information Administration. Official Energy Statistics
from the U.S. Government. Available online. URL: http://www.
eia.doe.gov/. Accessed on August 13, 2008. This is certainly the
best source of statistics with respect to how energy is used. It
has separate sections on all major technologies and most minor
technologies. The Web site is not especially easy to search, but it
is well worth the effort.
Gompertz, Simon, Business Correspondent, BBC. “On the Brink
of a Wave Revolution.” Available online. URL: http://news.bbc.
co.uk/2/hi/programmes/working_lunch/4849540.stm. Accessed
August 13, 2008. At the bottom of the page are links to BBC
videos that show the Pelamis wave energy converter and a tidal
mill in action. The videos convey something of the size of these
projects as well as the difficult environments in which these
machines function.
Kotchen, Matthew J., Michael R. Moore, Frank Lupi, and Edward
S. Rutherford. “Environmental Constraints on Hydropower: An
Ex Post Benefit-Cost Analysis of Dam Relicensing in Michigan.”
Available online. URL: http://www.msu.edu/user/lupi/Kotch-
en_etal_InPress_LandEcon2006.pdf. Accessed August 13,
2008. This is the paper described in chapter 3. It is not particu-
larly easy reading, but it contains information on how federal
188  Wind and Water

l­egislation, economics, and physics interact to affect the way

that hydroelectric facilities are used.
National Renewable Energy Laboratory. “What Is Ocean Thermal
Energy Conversion?”. Available online. URL: http://www.nrel.
gov/otec/what.html. Accessed August 13, 2008. This site pro-
vides an overview of OTEC technology as well as a map that
shows where the oceanic temperature differences are best suited
for deploying OTEC technology.
U.S. Bureau of Reclamation. Columbia Basin Project. Available on-
line. URL: http//www.usbr.gov/dataweb/projects/Washington/
columbiabasin/history.html#Construction. Accessed August
13, 2008. This is an excellent and very thorough history of the
Grand Coulee Dam, one of the largest civil-engineering projects
in the history of the United States.
U.S. Department of Energy. “Wind Power Today.” Available
online. URL: http://www.nrel.gov/docs/fy06osti/39479.pdf.
Accessed August 13, 2008. Although this presentation is a little
one-sided, it describes a number of important wind-power
research projects and the potential contribution of wind energy
to United States energy markets.
U.S. National Park Service. “Building America’s Industrial Revolu-
tion: The Boott Cotton Mills of Lowell, Massachusetts.” Avail-
able online. URL: http://www.nps.gov/nr/twhp/wwwlps/
lessons/21boott/21boott.htm. Accessed August 13, 2008. De-
signed for teachers, this excellent site has a number of photo-
graphs and drawings in addition to the well-written text. It tells
an important part of the story of the industrialization of the
United States.
 Index

Note: Italic page numbers indicate alternative energy sources. See

illustrations and diagrams; page renewable energy
numbers followed by c indicate altitude
chronology entries. air density and 145
wind speed and 127, 134, 143,
A 143–144
AC. See alternating current aluminum industry 42
accumulator 64 Amazon region 53
Afghanistan 124 amplitude 60, 62
agriculture annular sail 131
and Grand Coulee Dam 42 Antipater of Thesselonica 4–5, 176c
windmills for 131 Appalachia 168–169
and wind power land use 158 applied research 109–110
air, density of 145 aqueduct 5
air pressure 69, 73 Arab-Israeli War (1973) 102
air speed 69, 70 arch dam 33
ALCOA 18 Archimedes Wave Swing (AWS)
Alliance to Protect Nantucket 72, 72–74, 178c
Sound 167 arc lamp 12–13
Altamont Pass Wind Resource Area Arsonval, Jacques-Arsène d’ 92,
136–137 176c
alternating current (AC) Aswan High Dam (Egypt) 40, 177c
DC v. 13–14 atmosphere. See global warming
for electric power distribution atmospheric pressure 52
13–15 Australia 70
and Niagara Falls power avoided cost 103
distribution 17–18 AWS. See Archimedes Wave Swing
Westinghouse and 15 axis of symmetry 27

189
190  Wind and Water

B Brush, Charles F. 131, 132

backup systems 67 Brush turbine 131, 132
bacteria 52 Buffalo, New York 16, 17, 176c
Barbegal waterwheel complex 5 Bull, Stan 106, 106–110
Barents Sea 81
Bargi Dam (India) 41 C
barrage. See tidal barrage California
base load (base load power) Altamont Pass Wind Resource
coal and nuclear for 141 Area 136–137
and Consumers Energy reserve capacity issues 150
Company study 54 wind power development 102,
defined 34 103, 177c
hydropower for 35 wind power subsidies 163
peak load power v. 33–36, Canada 76, 81, 86
150–151 canals 7, 16
power plants for 34–35 canvas 126, 129–130
in restructured electrical market cap (tower mill) 127, 128
51 capacity
unsuitability of wind for 141 of Altamont Pass Wind Resource
basin 79 Area 136
battery 10, 13 of early 20th-century
Bay of Fundy 76, 81 hydropower 18
Belgium 7 of Grand Coulee Dam 39
benchmark 152 of Nathpa Jhakri 40
Betz, Albert 85, 147 of oscillating water column
birds 137, 167 converter 71
blackouts 150 of tidal mill 84–85
blade, turbine 26, 146 of wind farm 152–153
blinds 130 of wind power 148–153
boating 168 of wind turbines 138
Bowles, Ian 168 capacity factor 148
Brazil capacity rating 152
hydropower capacity (2004) Cape Cod, Massachusetts 165, 168
178c Cape Wind project 165–168
Itaipú hydropower plant 40, carbon dioxide emissions
177c from fossil fuels 121
methane and hydropower 52–53 and global warming 44
Britain. See Great Britain methane emissions v. 52, 53
 Index  191

carbon tax 112 computer, wind turbine speed

Carnot, Nicolas-Léonard-Sadi 89, control by 135
90 condensation 96
Cataract Construction Company condenser 97
16, 17 Congress, U.S. 107, 108
China Congressional Joint Committee on
coal-fired power plants 89 Taxation 104
dam projects 41, 47 conservation 114
early waterwheels 5 construction costs 22, 39
early windmills 121 Consumers Energy Company
Chinook salmon 53–54 51–54
civil engineering 33, 39–41, 79 continents, tides and 76
class, economic 165–170, 166 conventional hydroelectric
Claude, Georges 92, 93, 177c power. See also hydroelectric
Cleveland, Ohio 131 power
climate change. See global as mature technology 173
warming origins of ideas behind 3
Clinch River (Tennessee) 40 subsidies for xiv–xv
coal-fired power plants conventional power sources
for base load power 35, 120 hidden costs 100–101
and Consumers Energy infrastructure for 161
Company study 54 predictability of 147
continuing importance of 74, wind power growth vs. 160–
89, 119–120 161
cost v. price 100–101 cooperatives, wind-turbine 113
developed infrastructure/ cost (of power). See also
subsidies for 161 environmental costs; hidden
environmental costs 43, 46, costs
100–101 and dam construction 22
and German energy policy 114, from hydropower plants 37–43,
115 46–54
reliability of power from natural gas v. wind power 121
140–141 from Pelamis unit 65
in restructured electrical market price v. 38–39
51 from tidal mills 86–87
coal mining 88, 168–169 from wind 160–165
commodity, electricity as 160 currents, ocean 82–85
competition 50, 160–162 cylinder 73
192  Wind and Water

D topography and siting 25–26, 29

dams water volume and siting 29, 32
arch dam 33 Xinanjian Dam (China) 41
Aswan High Dam (Egypt) 40, Darrius wind turbine 124
177c Darwin, Erasmus 124
Bargi Dam (India) 41 Davy, Sir Humphry 12–13
Chinese projects 41, 47 DC. See direct current
construction as part of power debris, at hydropower plants 29
costs 22 Delahunt, William 167
early history 6–7 demand. See also supply and
and efficiency 28 demand
environmental costs of and Archimedes Wave Swing 74
construction 41 base load v. peak load 33–36
and government regulation 48 creating, for wind power
Grand Coulee Dam 102–111
(Washington state) 39, in early days of commercial
42–43, 177c electricity production
height and efficiency 23, 28, 29 12–18
height of 22 and flexibility of hydropower
history of 32–33 32, 35
Hodenpyl Dam (Michigan) fluctuations in 33–34
51–54 and German energy policy 114
Hoover Dam (Nevada) 34, 38, pumped storage facility to
40 accommodate 30–31
to increase reliability of water and wind turbines 138, 147
flow 6 Denmark
and isoquant curve 23 government energy policies
Keber Dam (Persia) 33 111–113
and low hydraulic head 25–26 Poul la Cour’s wind turbine
Ma’rib Dam (Yemen) 32 132–133, 176c
as massive engineering projects wind power capacity (1990)
39–41 178c
for non-hydropower uses 28 wind power policy 112–113,
Norris Dam (Tennessee) 40–41 172, 177c
Sanmenxia Dam (China) 41 density, of air 145
Three Gorges Dam (China) 41, Department of Energy, U.S. (DOE)
47 National Renewable Energy
Tippy Dam (Michigan) 51–54 Laboratory funding 107, 108
 Index  193

ocean thermal energy converter and dam height 23, 28, 29

94, 98 dams to increase 6
wind energy budget cuts 177c of electrolysis 155
wind energy research 104–105, of Fourneyron turbine 26
108–109 of heat engines 90–92
depth, of ocean 72, 76 of hydropower plant 19, 21
deregulation. See restructured of ocean thermal energy
electrical markets converters 93, 95–98
diffuse energy sources, harnessing of Pelamis 66
of 174 of Pelton turbine 25
direct current (DC) 13–14 of small hydropower facilities
dissipation of waves 62 55
distributed risk 67 of tidal mill 84–85
distribution networks 48–49. See of wind turbines 144, 146–147
also high-voltage transmission Egypt. See also Nile River
lines; infrastructure Aswan High Dam 40, 177c
DOE. See Department of Energy, early dam construction 6, 176c
U.S. Electric Consumers Protection Act
Domesday Book 5–6, 78 (1986) 49
electricity, magnetism and 8, 10
E electricity demand. See demand
eagle 137 Electricity Feed Law (Germany)
Earth, rotation of 142 115, 178c
economic class 165–170, 166 electricity prices. See price
economic growth 42, 43 electrolysis, energy storage by 133,
Edison, Thomas 154–155
and DC current 14, 17 elevation. See altitude
and early distribution networks embargo (1973) 102
49 emissions. See also greenhouse gases
light bulb invention and from fossil fuel plants 43
electricity infrastructure 13 and OTEC research 98
Edwin I. Hatch Nuclear Plant from Pelamis Wave Energy
(Georgia) 157–158 Converter 64
effective load-carrying capability of tidal barrages 81
152–153 endangered species 137
efficiency. See also energy efficiency Endangered Species Act (1973) 49
and competitiveness of wind Energetech 70, 71
power 162 energy conservation 114
194  Wind and Water

energy content, of wind 142–148, and Cape Wind project

143 opposition 167
energy conversion and Danish energy policies
electric generating stations for 111–112
37 of fossil fuels 121
generators for 12 global warming 44–47, 45
by hydropower plant 19–23 of hydropower xvi
Pelamis Wave Energy Converter methane emissions and
62–67, 63, 65 hydropower 52–53
by tidal mills 82 of solar energy xvi
by wind turbine 131, 139 of tidal barrages 81
energy crises (1970s). See oil crises of tidal mills 86
energy efficiency 108–109 of wind power technology 121,
energy-intensive lifestyles 37–38 154–159
energy policy, national. See environmental justice 169
government energy policy environmental regulation 49
Energy Policy Act of 1992 (EPAct) EPAct. See Energy Policy Act of
103–104, 178c 1992
Energy Policy Act of 1995 178c Europe. See also specific countries,
energy security 101 e.g.: Germany
enfranchisement 112, 172 Fourneyron turbine 26
England 6. See also Great Britain windmills 124–131, 176c
“Environmental Constraints on wind power capacity (2006)
Hydropower” (Kotchen, 178c
Moore, Lupi, and Rutherford)
51 F
environmental costs factories, early hydropower for 55
assigning value to 47 Faraday, Michael 8, 10, 10–12
of coal 100–101 Federal Energy Regulatory
of conventional energy 168–169 Commission (FERC) 50–51,
of dam construction 41 54, 103
of electricity generation 37 Feed-in Law (Germany) 115,
hidden, for conventional power 178c
sources 100–101, 162 Finland 131
of hydropower 40–47, 52 fish 53–54, 86
environmental impact fluctuations
of Altamont Pass Wind Resource demand 33–34, 36
Area 137 wave energy 66
 Index  195

fossil fuel power plants. See also generator

specific fossil fuel plants, e.g.: Faraday design 10–12, 11
coal-fired power plants at Glen Canyon plant 27
as backup for Pelamis 67 in OTEC 96
and Consumers Energy in Pelamis converter 64
Company study 54 for wind turbine 134
and decline in small-scale Germany
hydropower facilities 28 coal-fired power plants 89
environmental costs 43, 44, 46 government energy policy
environmental impact of 121 114–116
and global warming 44 wind power capacity (1990)
hydropower v. 52–53 178c
for peak power supply 36 wind power capacity (1997)
and wind power in late 19th 178c
century 119 Gibbs, John D. 15
Fourneyron, Benoît 26, 176c Glen Canyon power plant
Fourneyron turbine 9, 24, 26–27 generators 27
France global warming xvi, 44–47, 45
Barbegal waterwheel complex from fossil fuels 121
5 methane and hydropower
Industrial Revolution in 7 52–53
nuclear power in 89 and OTEC research 98
La Rance Tidal Barrage 78, golden eagle 137
79–81, 80 Gorlov, Alexander M. 86
Francis turbine 27 government energy policy
free market 160 complexity of 170–171
fuel cells 154, 155 Denmark 111–113
future issues and energy investment 174–
hydropower 54–56 175
wind power 170–172 and free market 160
Germany 114–116
G and hidden costs of energy
Gabrielli, Christopher 168 162
gas-fired power plants. See natural United States 102–111
gas power plants and wind power 99–116,
Gaulard, Lucien 15 169–172
gearbox 133–134 government ownership 48
General Electric 138 government regulation 41, 48–54
196  Wind and Water

grain milling/grinding ocean thermal energy converter

tidal power for 77–78 as 82–98, 93, 97
waterwheels for 3–6 theory 88–92
windmills for 122, 125–126, heat exchanger 96, 97
128, 131 height. See also altitude
Grand Coulee Dam (Washington of water column 6, 19–20, 73
state) 39, 42–43, 177c of waves 62
grants 115 of windmills 127, 128
gravity 75 of wind turbines 134, 144
Great Britain 7, 125–126 helical turbine 86
Greece, ancient 3–5 hidden costs, of conventional
greenhouse gases power sources 100–101, 162
and alternative technologies 101 high-voltage transmission lines
and coal-fired plants 46 and Altamont Pass Wind
and Consumers Energy Resource Area 136
Company study 54 and early electricity markets 48
and German energy policy 114 and restructured electrical
and global warming 44 markets 50
and oscillating water column and wind farms 164–165
converter 71 and wind power 105, 110–111
and OTEC research 98 and wind turbine siting 147
and tropical hydropower 52, 53 history
grid, electrical of dams 32–33
integrating wind power into of tidal power 77–79
147–148 of waterpower 3–18
and maximum demand 138 of wind power 119–138
wind power and system Hodenpyl Dam (Michigan) 51–54
reliability 152, 153 Holland 129, 132
grinding. See grain milling/ Hoover Dam (Nevada) 34, 38, 40
grinding horizontal-axis windmill 124–126,
176c
H hydraulic head
Hawaii 94, 177c and dams 29
Healy, Kerry 168 high v. low 24
heat engine and hydropower equation 21
efficiency of 90–92 and hydropower plant output
for harnessing ocean power 20
82–98 and isoquant curve 22–23
 Index  197

and large hydropower facilities hydrogen 133, 154–155

55 hydromechanical power 8, 28,
low 26–27, 81 176c
at Niagara Falls 16
for Pelton turbine 25 I
in tidal barrage 79, 80 iceberg 45
and tidal-powered waterwheel illumination, electricity for 12–13
77 impulse turbine 24, 176c
and turbine design 24 incandescent light bulb 13
water volume v. 21–22 inconsistency. See intermittency
hydraulic ram 64 India
hydroelectric power 3–58, 20 Bargi Dam 41
base load v. peak load 33–36 coal-fired power plants 89
costs 37–43, 46–54 Nathpa Jhakri hydropower
early development of 15–18 project 40, 178c
energy conversion by power ocean thermal energy converter
plant 19–23 research 94
environmental costs 40–47 indirect government support 170
Faraday’s inventions 8, 10–12, 11 Industrial Revolution 7–8, 55
future issues 54–56 industry 110, 112
German energy policy 114 infrastructure
and global warming 45–46 for conventional power sources
Grand Coulee Dam 161
(Washington state) 177c in early days of electricity
Itaipú hydropower plant 177c production 13–15, 48–49
methane emissions 52–53 Edison’s construction of 13
Nathpa Jhakri hydropower for electric power distribution
project 178c 13–15
National Renewable Energy for Niagara Falls power
Laboratory and 106–107 distribution 17–18
Niagara Falls 16–18, 17, 176c and restructured electrical
pumped storage for 30–31, 31 markets 50
regulatory environments for interference (wave) 61
48–54 intermittency
theory and practice 19–36 and AWS 74
and transformers 15 dams to mitigate 29
turbine design for 23–28 and Danish wind energy 113
water supply for 28–33, 31 and limitation wind energy 172
198  Wind and Water

and OWC 71 Korea, South 83

and Pelamis 66, 67 Kotchen, Matthew 51
storage to mitigate 154
of water flow 6 L
of wind 140 la Cour, Poul 111, 132–133, 176c
of wind power 120, 121, lakes, artificial 41
147–148, 151 land use
investment credits 111 for hydropower 40
investment/investors and wind farm siting 157–158,
and alternative technologies 101 164
and Danish wind energy policy and wind power 105, 110–111,
113 149, 174
profit as goal of 170 Lely wind farm (Netherlands) 132
in wind power v. other alternative Leonardo da Vinci 69
energy sources 170 lifestyles, energy-intensive 37–38
Iran 124 light bulb 13
irrigation 42, 131 lighting, electricity for 12–13
islands, OWC power for 71 Limpet OWC converter 67, 71, 178c
ISO-New England 166 linear generator 73
isoquant curve 22–23, 23 linear motion
Itaipú hydropower plant 40, 177c and OWC converters 70
and tidal mills 82
J and waterwheels 5
Japan load, generators and 12
early waterwheel use 5 local opposition
Mighty Whale OWC 70 for aesthetic reasons 157
OTEC research 94, 177c Danish enfranchisement of 112
and economic class 165–170
K need for government policy to
Keber Dam (Persia) 33 enfranchise 172
Kennedy, Edward F. 166, 167 to U.S. wind farms 105
Kennedy, Robert F., Jr. 166 local utility 48
Kennedy family compound long-term contracts 51, 150
(Hyannis Port, Massachusetts) Lowell, Massachusetts 7–8, 176c
168 low-head hydropower 25–27, 81
kinetic energy 139, 144 low-income groups 169
Kola Peninsula tidal barrage project low-voltage power lines 48
(Russia) 81 Lupi, Frank 51
 Index  199

M Moore, Michael R. 51
magnetism, electricity and 8, 10 mountains, wind farm siting on
Manistee River, Michigan 51–54 157
Ma’rib Dam (Yemen) 32
Marine Current Turbines 85–86 N
marine life, tidal mills and 86 nameplate capacity
markets, for electricity 156–157, and Altamont Pass Wind
161, 164–165 Resource Area 137
Mason City, Washington 39 effective load-carrying capability
Massachusetts 165–168 vs. 152
Matanzas Bay, Cuba 92 European wind power 178c
U.S. wind power 179c
maturing technologies 109
Nantucket Island 165, 166
maximum demand 138
Nantucket Sound 165–168
Meade, Lake 40
Nathpa Jhakri hydropower project
mechanical energy, conversion into
(India) 40, 178c
electrical energy 12
National Energy Act (1978) 103
mercury 43, 46
national policy. See government
Merrimack River 7–8
energy policy
Mesopotamia 121
National Renewable Energy
methane emissions, hydropower
Laboratory (NREL) 106–110
and 52–53 National Space Research Institute
microhydropower 55 (Brazil) 52–53
Middle Ages 5–6, 78, 127 natural gas power plants
Mighty Whale 70, 71 as benchmark for wind power
milling. See grain milling/grinding production 152
millwright 128–130 and Consumers Energy
mining Company study 54
effect on Appalachia 168–169 continued importance of
environmental costs 43 119–120
human cost 43, 169 developed infrastructure for 161
steam engines for 88 for exploiting methane from
minorities 169 tropical hydropower plants
molecular bonds 154 53
monopolies 48–50 and German energy policy 114
Monterey Bay Regional Energy for peak power 36, 120, 141
Plan 136 wind farm equivalent for 153
Moon 44, 75–76 wind power v. 120–121
200  Wind and Water

natural monopoly 49, 50 developed infrastructure/

Nauru 94, 177c subsidies for 161
neap tides 76 and German energy policy 114,
Neh, Persia 123 115
the Netherlands 129, 132 as heat engine 89
New England ocean thermal energy converter
decline in small-scale v. 94–95
hydropower facilities 28 reliability of power from
effect of Midwestern coal- 140–141, 174
burning plants on 169 in restructured electrical market
Lowell, Massachusetts mills 51
7–8, 176c wind power land use v. 157–
New York State 158, 174
avoided costs 103
Niagara Falls 14, 16–18, 17, 26, O
28, 176c ocean, as wind farm site. See
subsidies for wind power 163 offshore wind farms
transmission lines controversy ocean currents 82–85
164–165 ocean power 59–98. See also ocean
Niagara Falls 17 thermal energy converter
DC and power distribution 14 and heat engines 92–98
development of 16–18 tidal power 75–87, 78, 80, 83
Fourneyron turbines at 26 wave power 59–74, 60, 61, 63,
hydroelectric plant opens 176c 65, 68, 72
hydropower without dams 28 ocean thermal energy converter
Niagara River Hydraulic Tunnel (OTEC) 82–98, 93, 97
Power and Sewer Company Georges Claude’s model 177c
16 earliest concept of 176c
Nile River 6, 119, 176c Hawaiian project 177c
nondiscriminatory access 50 NREL research 106, 107
Norris Dam (Tennessee) 40–41 process diagram 97
Northern Ireland 86 tests of 94–95
Norway 113 offshore wind farms 132, 164–168,
nozzle, for impulse turbine 24 171
NREL (National Renewable Energy oil 88, 119
Laboratory) 106–110 oil crises (1970s) 177c
nuclear power plants and Danish energy policies 111
for base load power 35 and German energy policy 115
 Index  201

and NREL funding 108 Pelamis Wave Energy Converter

and OTEC research 98 62–67, 63, 65, 179c
and U.S. wind power policy 102 Pelamis Wave Power 62
oil-fired power plants 36, 89 Pelton, Lester Allen 24–25, 176c
oil spills 169 Pelton turbine 24, 25
operating temperature, heat engine penstock 29, 30, 32
efficiency and 89–92 Persia 121–123, 176c
Organization of Petroleum petroleum. See oil
Exporting Countries (OPEC) physics of waves/wave power
177c 60–62
Orkney Islands 64 piston 69, 73
Orontes River waterwheel 4, 6–7 Pittsburgh Reduction Company 18
oscillating water column (OWC) policy. See government energy
67–71, 68, 178c policy
oscillation of waves 60 politics
OTEC. See ocean thermal energy and conventional power sources
converter 161
OWC. See oscillating water column and energy investment 174–175
oxygen atoms 154 and hidden costs of energy 163
pollution 37–38, 121, 169. See also
P emissions
Paraguay 40, 177c population displacement 40–41
parallel systems 67 Portland, Oregon 15
Patrick, Deval 168 Portugal 63–64, 179c
Pawtucket Falls 7 post mill 125, 125–126, 128
peak load (peak load power) power
base load v. 33–36, 150–151 defined 19
and benchmark 152 of flowing water 20–21
and Consumers Energy and hydropower equation 21
Company study 53, 54 from ocean current 84
defined 35 power demand, base load and
hydropower for 36, 50–51 34–35
matching supply and demand power grid. See grid
35–36 power output. See also capacity
natural gas and wind power construction costs v. 22
120 of tidal mill 84–85
wind power for 141 and water volume 29
wind speeds during 155 of wind turbines 138, 145–147
202  Wind and Water

predictability, of conventional pumped storage facility

power plants 147 Francis turbines for 27
predictability, of waves 65, 66 principles of 30–31
pressure difference 26, 72–73 Raccoon Mountain plant 31,
price 177c
and Cape Wind project 165–166 and La Rance tidal barrage 80–81
cost v. 38–39 PURPA. See Public Utility
of Danish renewable energy 112 Regulatory Policies Act of 1978
for renewable energy under
PURPA 103 R
and reserve capacity 150 Raccoon Mountain Pumped Storage
and restructured electrical Plant (Tennessee) 31, 177c
markets 50 rainfall 6
of wind power 171, 172 La Rance Tidal Barrage 78, 79–81,
price signals 50, 51 80
priorities, government 174–175 Rankine, William John Macquorn
probability theory 153 33
production credit reaction turbine 24, 26–27
and Danish energy policies 112 regulation. See Federal Energy
and EPAct 103–104, 178c Regulatory Commission;
profit government regulation; Public
and Consumers Energy Utility Regulatory Policies Act
Company study 54 of 1978
environmental damage v. 46–47 Reilly, Tom 167, 168
as goal of energy investors 170 reliability
and Grand Coulee Dam 42 and estimating wind power
and hidden costs of coal 101 capacity 148
maximizing, from hydropower of nuclear power 174
plants 51 of oscillating water column
and OTEC research 95 converter 71
and pumped storage facilities 31 of Pelamis 66
in restructured electrical market of wind power, as part of grid
50, 51 system 150–152
wind energy and 174 of wind turbines 147
public good, energy as 163 renewable energy. See also specific
Public Utility Regulatory Policies sources, e.g.: wind power
Act of 1978 (PURPA) 103, energy security and 101
177c NREL research 106–110
 Index  203

PURPA and 103, 177c S

wind power v. other sources of sail
170 for Brush turbine 132
Renewable Energy Law (Germany, for early windmills 122, 123
2000) 115 for post mill 126–127
research and development (R&D) for tower mill 128–130
and hydropower costs 39 for vertical axis windmill
National Renewable Energy 123–124
Laboratory 106–110 salmon 53–54
ocean thermal energy converter Sanmenxia Dam (China) 41
106, 107, 177c Scotland
for ocean thermal energy Archimedes Wave Swing 72
converter 94, 95, 98 Limpet OWC converter 67, 178c
subsidies for 104 Pelamis wave farm 64, 179c
success standards for 109–110 Scottish Power 179c
by U.S. Department of Energy seafloor 76
104–105, 108–109 SeaGen marine current turbine 83
wave power 106 “sea snakes.” See Pelamis Wave
for wind power 162, 176c Energy Converter
wind research budget cuts 177c secondary sail 126
reserve capacity 149–150 secondary sites 164
reservoir, pumped storage facility 30 shortages, water 29
resettlement 40–41 sine wave 61
restructured electrical markets Sistān, Persia 123
50–51 siting
ridgeline 157 of dams 25–26, 29, 32
risk, distributed 67 of Danish wind farms 112–113
Rome, ancient 5, 77–78 economic class issues 165–170,
Romney, Mitt 167 166
rotary motion 5, 70, 82 and future hydropower
rotation, of Earth 142 development 54–56
rotation speed, wind turbine of high-voltage transmission
135–138 lines 110–111, 164–165
rotor 11–12, 133–138 local opposition to wind farms
royalties 105 105
run of river mode 50–51, 53–54 in low windspeed areas 171–172
Russia 81, 114 on mountains 157
Rutherford, Edward 51 of new hydropower projects 33
204  Wind and Water

and power line access 147 state wind power programs 102
of reliable wind farm system stator 11, 12
153 steam engines 88, 89
and suitability of wind farm sites step-down transformer 14–15
142–143 step-up transformer 14
technical/environmental storage
problems with wind farms of water for hydropower 30–31,
154–159 31. See also dams
of tidal barrages 81–82 of wind power 154–155
wind data for wind farms storm swells 60
151–152 subsidies xiv–xv
and wind farm land use 157– and Altamont Pass Wind
158, 164 Resource Area 136
wind reliability and 147 and alternative technology
sluice 77 investment 101
smock mill 127 for conventional power sources
“snakes.” See Pelamis Wave Energy 100–101
Converter and Danish energy policies
social costs 81, 100–101 111–113
solar energy direct v. indirect 170
for ocean thermal energy EPAct provisions 104
converter 92–93, 96, 98 and German wind energy policy
unreliability for peak power 115–116
supply 35–36 and hidden costs of coal 101
wind power as 142 and hidden costs of energy 163
South America 40, 177c for hydropower 41
South Korea 83 for new technologies 99–100
spacing, of turbines on wind farms for offshore wind farms 164
157, 158 PURPA provisions 103
speed for secondary wind farm sites
of ocean currents 84 164
of waves 61 for U.S. wind power 102–104,
speed control 161–163, 170
for windmills 128–130 Sun
for wind turbines 134–135 and global warming 44
spillways 32 for OTEC energy 92–93
spot market 150–151 and tides 75, 76
spring tides 75 sunlight, wind production and 142
 Index  205

supply and demand tidal power 75–87

and Archimedes Wave Swing 74 early history 77–79
balancing of 32, 35–36, 80–81, physics of 75–77
148–150 tidal barrages 77, 78, 79–81, 80
Danish energy policy 111–113 tidal mills 77–78, 82–87, 83
and early electricity production tidal variation 79, 81
12–18 Tippy Dam (Michigan) 51–54
Pelamis and 66 topography
tidal barrage to balance 80–81 and dam siting 25–26, 29
U. S. energy policy 102–111 and ocean currents 83
surplus power storage 30–31 and tides 76
Sweden 113 tower mill 127–130, 129
Syria, ancient 6–7 towers, for wind turbines 134
system reliability (grid) 147–148, transformer 14–15
152, 153 transmission, of electrical power
alternating current for 13–15,
T 17–18
taxation, hidden costs of energy in early electricity markets
and 163 48–49
tax credit 103–104, 136 and restructured electrical
temperature, operating 89–92 markets 50
temperature difference 90–96 transmission lines. See high-voltage
Tennessee 31, 40–41, 177c transmission lines
Tennessee Valley Authority transportation, heat engines for
pumped storage facility 31, 88
177c tropical regions
textile mills 7–8, 26 methane and hydropower
thermal energy, heat engine 52–53
efficiency and 90–92 OTEC energy in 92, 98
Three Gorges Dam (China) 41, 47 turbines. See water turbines; wind
tidal barrage turbines
ancient Rome 77 turbulence 144, 157
Bay of Fundy 76, 81
Kola Peninsula project 81 U
La Rance 78, 79–81, 80 Uldolmok Strait (South Korea) 83
revival of concept xv unexpected demand, wind power
tidal currents 83–84 to meet 149
tidal mill 77–78, 82–87, 83 Union Carbide 18
206  Wind and Water

United Kingdom 85. See also USDA-ARS Conservation

specific countries, e.g.: England, and Production Research
Great Britain Laboratory (Bushland, Texas)
United States. See also specific 120
locations, e.g.: Lowell, utilities, local 48
Massachusetts
coal and natural gas power in V
120 value, reliability and 150
coal-fired power plants 89 velocity. See speed
creating supply and demand for vertical axis windmill 122, 122–
wind energy 102–111 124, 176c
electricity demand in early 20th voltage, power transmission and
century 18 14, 15
Fourneyron turbine in 26 volume, of water
government energy policy and dam siting 29, 32
102–111 hydraulic head v. 21–22
hydropower for base load power and hydropower equation 21
35 for ocean thermal energy
Industrial Revolution 7–8 converters 97
nuclear power 89 volumetric flow rate
OTEC research 94 and isoquant curve 22, 23
price of wind energy 172 and large hydropower facilities
resettlement from hydropower 55
projects 40–41
topography favoring water W
power 7 wake, of wind rotors 157
wind farms siting 154–159 walls, for early windmills 122–123
windmills for water pumps 131 water column 6, 19–20, 73
wind power capacity (1990) 178c water level, waterwheel efficiency
wind power capacity (2006) and 6
179c watermill 176c. See also
wind research budget cuts 177c waterwheel
wind resource map 156m, water molecules 154
156–157 waterpower
unpredictability brief history of 3–18
of demand fluctuations 33 designs from antiquity 3–7
of wind power 151 hydropower. See hydroelectric
unreliability. See intermittency power
 Index  207

during Industrial Revolution wave period 61

7–8 wave power 59–74
Niagara Falls 16–18, 17, 176c Archimedes Wave Swing 72,
ocean power. See ocean power 72–74, 178c
water pressure 72–73 National Renewable Energy
water pumps 88, 131 Laboratory research 106,
water shortages 29 107
water supply, for hydropower oscillating water column
28–33, 31 converter 67–71, 68, 178c
water turbines Pelamis converter 62–67, 63, 65,
designs 23–28 179c
Fourneyron turbine 9, 24, physics of 60, 60–62, 61
26–27, 176c weather 6, 119
Francis turbine 27 weather reports 151
generators and 12 Westinghouse, George 15, 17
impulse turbine 24, 176c Willamette Falls Electric Company
Marine Current Turbines 15
85–86 wind
in ocean thermal energy energy content of 142–148, 143
converter 96 and waves 60, 61
in oscillating water column wind data, for wind farm siting
converter 70, 71 151–152
Pelton turbine 24, 25 wind farms
in pumped storage facility 30 Altamont Pass Wind Resource
reaction turbine 24, 26–27 Area 136–137
SeaGen marine current turbine California in 1980s 103
83 Danish siting policy 112–113
for textile mills 7 economic class and siting of
in tidal barrage 79, 80 165–170, 166
in tidal mills 82–87, 145 land use for 149, 164
waterwheel 3–8, 4 Lely wind farm (Netherlands)
wave, physics of 60–62 132
wave energy, variations in 71 and market access 164–165
wave farms suitability of sites 142–143, 147
Pelamis 63–64, 66, 179c technical/environmental
wind farms v. 74 problems with siting of
Wavegen 67, 71 154–159
wavelength 60–62 wind data for siting of 151–152
208  Wind and Water

windmills 121–131, 122, 125, 129 wind farms. See wind farms
early use of 119 windmills 121–131, 122, 125,
for grain milling 122, 125–126, 129
128, 131 wind resource map 156m, 156–
post mill 125, 125–126, 128 157
tower mill 127–130, 129 windshaft 123, 126, 127
wind power 99–116, 119–172 wind speed
in California 103, 136–137, 177c and altitude 127, 134, 143,
capacity of 148–153 143–144
costs 160–165 and tower mills 130
creating supply and demand for and turbine output 145–146
102–111 and wind farm siting 154–156,
Danish energy policies 111–113 171–172
energy content of wind 142– wind tower 140
148, 143 wind turbines 120, 131–138, 132,
environmental impact 154–159 133
future issues 170–172 at Altamont Pass Wind Resource
German policies 114–116 Area 136–137
government policy 99–116, 160 Brush turbine 131, 132
history 119–138 Darrius wind turbine 124
natural gas v. 120–121 in Denmark 112, 113
nature of 139–159 in Germany 115
NREL research 108–109 Poul la Cour’s design 132–133
for peak load power 151 OWC v. 71
Pelamis v. 65 spacing of, on wind farms 157,
resource map 156m 158
storage of 154–155 standard definition 131
U.S. energy policies 102–111 working fluid 96, 97
U.S. research budget cuts 177c
unreliability for peak power X
supply 35–36 Xinanjian Dam (China) 41