Climate Change: Lines of Evidence

7/31/2019 Climate Change: Lines of Evidence

1/40

answers to common questions about the science

of climate change

Climate

Change


2/40

Contents

Part I. Evidence for Human-Caused Climate Change 2

How do we know that Earth has warmed? 3

How do we know that greenhouse gases lead to warming? 4

How do we know that humans are causing greenhouse gases to increase? 6

How much are human activities heating Earth?9

How do we know the current warming trend isnt caused by the Sun? 11

How do we know the current warming trend isnt caused by natural cycles? 12

What other climate changes and impacts have been observed? 15

The Ice Ages 18

Part II. Warming, Climate Changes, and Impactsin the 21st Century and Beyond 20

How do scientists project future climate change? 21

How will temperatures be affected? 22

How is precipitation expected to change? 23

How will sea ice and snow be affected? 26

How will coastlines be affected? 26

How will ecosystems be affected? 28

How will agriculture and food production be affected? 29

Part III. Making Climate Choices 30

How does science inform emissions choices? 31

What are the choices for reducing greenhouse gas emissions? 32

What are the choices for preparing for the impacts of climate change? 34

Why take action if there are still uncertainties about the risks of climate change? 35

Conclusion 36


3/40

Just what is climate? Climate is commonly thought o as the expected weather conditions

at a given location over time. People know when they go to New York City in winter, they

should take a coat. When they visit the Pacic Northwest, they take an umbrella. Climate

can be measured at many geographic scalesor example, cities, countries, or the

entire globeby such statistics as average temperatures, average number o rainy days,

and the requency o droughts. Climate changereers to changes in these statistics

over years, decades, or even centuries.

Enormous progress has been made in increasing our understanding o climate change

and its causes, and a clearer picture o current and uture impacts is emerging. Research

is also shedding light on actions that might be taken to limit the magnitude o climate

change and adapt to its impacts.

This booklet is intended to help people understand what is known about climate change.First, it lays out the evidence that human activities, especially the burning o ossil uels,

are responsible or much o the warming and related changes being observed around

the world. Second, it summarizes projections o uture climate changes and impacts

expected in this century and beyond. Finally, the booklet examines how science can help

inorm choices about managing and reducing the risks posed by climate change. The

inormation is based on a number o National Research Council reports (see inside back

cover), each o which represents the consensus o experts who have reviewed hundreds o

studies describing many years o accumulating evidence.

Climate Change


4/40

2

But how has this conclusion been reached? Climate science,

like all science, is a process o collective learning that

relies on the careul gathering and analyses o data,

the ormulation o hypotheses, the development o

models to study key processes and make testable

predictions, and the combined use o observations

and models to test scientifc understanding.

Scientifc knowledge builds over time as new

observations and data become available.

Confdence in our understanding grows i multiple

lines o evidence lead to the same conclusions, or

i other explanations can be ruled out. In the case o

climate change, scientists have understood or more

than a century that emissions rom the burning o ossil

uels could lead to increases in the Earths average surace

temperature. Decades o research have confrmed and extended this

understanding.

The overwhelming

majority o climate

scientists agree that

human activities,

especially the burning

o ossil uels (coal, oil,

and gas), are responsible

or most o the climatechange currently being

observed.

EvidEncEfor Human-Caused

Climate ChangePart I


5/40

3

Scientists have been taking widespread measure-

ments o Earths surace temperature since

around 1880. These data have steadily improved

and, today, temperatures are recorded by ther-

mometers at many thousands o locations, both on

the land and over the oceans. Dierent research

groups, including the NASA Goddard Institute or

Space Studies, Britains Hadley Centre or Climate

Change, the Japan Meteorological Agency, and

NOAAs National Climatic Data Center have used

these raw measurements to produce records o

long-term global surace temperature change

(Figure 1). These groups work careully to makesure the data arent skewed by such things as

changes in the instruments taking the measure-

ments or by other actors that aect local tempera-

ture, such as additional heat that has come rom

the gradual growth o cities.

These analyses all show that Earths average

surace temperature has increased by more than

1.4F (0.8C) over the past 100 years, with much

o this increase taking place over the past 35 years.

A temperature change o 1.4F may not seem like

much i youre thinking about a daily or seasonal

fuctuation, but it is a signicant change when

you think about a permanent increase averaged

across the entire planet. Consider, or example,

that 1.4F is greater than the average annual

FIGURE 2FIGURE 1

NASAs Global Surace Temperature Record Esti-mates o global surace temperature change, relative

to the average global surace temperature or theperiod rom 1951 to 1980, which is about 14C (57F)rom NASA Goddard Institute or Space Studies showa warming trend over the 20th century. The esti-mates are based on surace air temperature measure-ments at meteorological stations and on sea suracetemperature measurements rom ships and satellites.The black curve shows average annual temperatures,and the red curve is a 5-year running average. Thegreen bars indicate the margin o error, which hasbeen reduced over time. Source: National ResearchCouncil 2010a

(bottom let) Climate monitoring stations on land and sea,such as the moored buoys o NOAAs Tropical Atmosphere

Ocean (TAO) project, provide real-time data on tempera-ture, humidity, winds, and other atmospheric properties.Image courtesy o TAO Project Oce, NOAA Pacic MarineEnvironmental Laboratory. (right) Weather balloons, whichcarry instruments known as radiosondes, provide verti-cal proles o some o these same properties throughoutthe lower atmosphere. Image University Corporation orAtmospheric Research. (top let) The NOAA-N spacecrat,launched in 2005, is the teenth in a series o polar-orbiting satellites dating back to 1978. The satellites carryinstruments that measure global surace temperature andother climate variables. Image courtesy NASA

How do we know that Earth has warmed?


6/40

4

temperature dierence between Washington, D.C.,

and Charleston, South Carolina, which is more

than 450 miles arther south. Consider, too, that

a decrease o only 9F (5C) in global average

temperatures is the estimated dierence between

todays climate and an ice age.In addition to surace temperature, other parts o

the climate system are also being monitored careully

(Figure 2). For example, a variety o instruments are

used to measure temperature, salinity, and currents

beneath the ocean surace. Weather balloons are

used to probe the temperature, humidity, and winds

in the atmosphere. A key breakthrough in the ability

to track global environmental changes began in the

1970s with the dawn o the era o satellite remote

sensing. Many dierent types o sensors, carried

on dozens o satellites, have allowed us to build atruly global picture o changes in the temperature

o the atmosphere and o the ocean and land

suraces. Satellite data are also used to study shits in

precipitation and changes in land cover.

Even though satellites measure temperature very

dierently than instruments on Earths surace, and

any errors would be o a completely dierentnature, the two records agree. A number o other

indicators o global warming have also been

observed (see pp.15-17). For example, heat waves

are becoming more requent, cold snaps are now

shorter and milder, snow and ice cover are

decreasing in the Northern Hemisphere, glaciers

and ice caps around the world are melting, and

many plant and animal species are moving to cooler

latitudes or higher altitudes because it is too warm

to stay where they are. The picture that emerges

rom all o these data sets is clear and consistent:Earth is warming.

How do we know that greenhouse gaseslead to warming?

As early as the 1820s, scientists began to ap-preciate the importance o certain gases inregulating the temperature o the Earth (see Box 1).Greenhouse gaseswhich include carbon dioxide

(CO2), methane, nitrous oxide, and water vapor

act like a blanket in the atmosphere, keep-

ing heat in the lower atmosphere.

Although greenhouse gases

comprise only a tiny raction

o Earths atmosphere, they

are critical or keeping the

planet warm enough to

support lie as we know it

(Figure 3).Heres how the

greenhouse eect

works: as the Suns energy

hits Earth, some o it is

refected back to space, but

most o it is absorbed by the

land and oceans. This absorbed

energy is then radiated upward rom Earths surace

in the orm o heat. In the absence o greenhouse

gases, this heat would simply escape to space,and the planets average surace temperature

would be well below reezing. But greenhouse

gases absorb and redirect some o

this energy downward, keeping

heat near Earths surace. As

concentrations o heat-

trapping greenhouse

gases increase in the

atmosphere, Earths

natural greenhouse

eect is enhanced(like a thicker blanket),

causing surace

temperatures to rise

(Figure 3). Reducing the

levels o greenhouse gases

in the atmosphere would

cause a decrease in surace

temperatures.


7/40

5

Amplication o the Greenhouse Eect The greenhouse eect is a natural phenomenon that is essential tokeeping the Earths surace warm. Like a greenhouse window, greenhouse gases allow sunlight to enter and thenprevent heat rom leaving the atmosphere. These gases include carbon dioxide (CO2), methane (CH4), nitrousoxide (N2O), and water vapor. Human activitiesespecially burning ossil uelsare increasing the concentrationso many o these gases, ampliying the natural greenhouse eect. Image courtesy o the Marian Koshland Science

Museum o the National Academy o Sciences

FIGURE 3

Box 1

Early Understanding o Greenhouse Gases In 1824, Frenchphysicist Joseph Fourier (top) was the rst to suggest that the Earthsatmosphere might act as an insulator o some kindthe rst proposalo what was later called the greenhouse eect. In the 1850s, Irish-born physicist John Tyndall (middle) was the rst to demonstrate

the greenhouse eect by showing that water vapor and otheratmospheric gases absorbed Earths radiant heat. In 1896, Swedishscientist Svante Arrhenius (bottom) was the rst to calculate thewarming power o excess carbon dioxide (CO2). From his calculations,Arrhenius predicted that i human activities increased CO2 levels in theatmosphere, a warming trend would result.


8/40

6

Discerning the human infuence on greenhousegas concentrations is challenging because manygreenhouse gases occur naturally in Earths atmo-

sphere. Carbon dioxide (CO2)is produced and con-

sumed in many natural processes that are part o the

carbon cycle (see Figure 4). However, once humans

began digging up long-buried orms o carbon such

as coal and oil and burning them or energy, addi-

tional CO2 began to be released into the atmosphere

much more rapidly than in the natural carbon cycle.

Other human activities, such as cement production

and cutting down and burning o orests (deoresta-

tion), also add CO2 to the atmosphere.

Until the 1950s, many scientists thought the

oceans would absorb most o the excess CO2

released by human activities. Then a series o

FIGURE 4

The Carbon Cycle Carbon is continually exchanged between the atmosphere, ocean, biosphere, and land on avariety o timescales. In the short term, CO2 is exchanged continuously among plants, trees, animals, and the airthrough respiration and photosynthesis, and between the ocean and the atmosphere through gas exchange. Otherparts o the carbon cycle, such as the weathering o rocks and the ormation o ossil uels, are much slower pro-cesses occurring over many centuries. For example, most o the worlds oil reserves were ormed when the remainso plants and animals were buried in sediment at the bottom o shallow seas hundreds o millions o years ago, and

then exposed to heat and pressure over many millions o years. A small amount o this carbon is released naturallyback into the atmosphere each year by volcanoes, completing the long-term carbon cycle. Human activities, espe-cially the digging up and burning o coal, oil, and natural gas or energy, are disrupting the natural carbon cycle byreleasing large amounts o ossil carbon over a relatively short time period. Source: National Research Council

How do we know that humans are causinggreenhouse gas concentrations to increase?


9/40

7

scientic papers were published that examined the

dynamics o carbon dioxide exchange between

the ocean and atmosphere, including a paper by

oceanographer Roger Revelle and Hans Seuss in

1957 and another by Bert Bolin and Erik Eriksson

in 1959. This work led scientists to the hypothesis

that the oceans could not absorb all o the CO2

being emitted. To test this hypothesis, Revelles

colleague Charles David Keeling began collecting

air samples at the Mauna Loa Observatory in Hawaii

to track changes in CO2 concentrations. Today,such measurements are made at many sites around

the world. The data reveal a steady increase in

atmospheric CO2 (Figure 5).

To determine how CO2 concentrations varied

prior to such modern measurements, scientists have

studied the composition o air bubbles trapped in

ice cores extracted rom Greenland and Antarctica.

These data show that, or at least 2,000 years beore

the Industrial Revolution, atmospheric CO2

concentrations were steady and then began to rise

sharply beginning in the late 1800s (Figure 6).

Today, atmospheric CO2 concentrations exceed 390

parts per millionnearly 40% higher than

preindustrial levels, and, according to ice core data,

higher than at any point in the past 800,000 years

(see Figure 14, p.18).

Human activities have increased the atmospheric

concentrations o other important greenhouse

gases as well. Methane, which is produced bythe burning o ossil uels, the raising o livestock,

the decay o landll wastes, the production and

transport o natural gas, and other activities,

increased sharply through the 1980s beore

starting to level o at about two-and-a-hal

times its preindustrial level (Figure 6). Nitrous

oxide has increased by roughly 15% since 1750

(Figure 6), mainly as a result o agricultural

FIGURE 5 FIGURE 6

Measurements o Atmospheric Carbon DioxideThe Keeling Curve is a set o careul measurementso atmospheric CO2 that Charles David Keelingbegan collecting in 1958. The data show a steadyannual increase in CO2 plus a small up-and-downsawtooth pattern each year that refects seasonalchanges in plant activity (plants take up CO2 duringspring and summer in the Northern Hemisphere,where most o the planets land mass and landecosystems reside, and release it in all and winter).Source: National Research Council, 2010a

Greenhouse Gas Concentrations or 2,000 YearsAnalysis o air bubbles trapped in Antarctic ice coresshow that, along with carbon dioxide, atmosphericconcentrations o methane (CH4) and nitrous oxide(N2O) were relatively constant until they started to risein the Industrial era. Atmospheric concentration unitsindicate the number o molecules o the greenhousegas per million molecules o air or carbon dioxideand nitrous oxide, and per billion molecules o air ormethane. Image courtesy: U.S. Global Climate ResearchProgram


10/40

8

ertilizer use, but also rom ossil uel burning and

certain industrial processes. Certain industrial

chemicals, such as chlorofuorocarbons (CFCs),

act as potent greenhouse gases and are long-lived

in the atmosphere. Because CFCs do not have

natural sources, their increases can be attributed

unambiguously to human activities.

In addition to direct measurements o CO2

concentrations in the atmosphere, scientists haveamassed detailed records o how much coal, oil,

and natural gas is burned each year. They also

estimate how much CO2 is being absorbed, on

average, by the oceans and the land surace. These

analyses show that about 45% o the CO2 emitted

by human activities remains in the atmosphere.

Just as a sink will ll up i water is entering it aster

than it can drain, human production o CO2 is

outstripping Earths natural ability to remove it

rom the air. As a result, atmospheric CO2 levels are

increasing (see Figure 7) and will remain elevated

or many centuries. Furthermore, a orensic-style

analysis o the CO2

in the atmosphere reveals the

chemical ngerprint o carbon rom ossil uels

(see Box 2). Together, these lines o evidence prove

conclusively that the elevated CO2 concentration in

the atmosphere is the result o human activities.

+ =

Emissions AtmosphericConcentration

Growth

Land Sink

Ocean Sink

FIGURE 7

Emissions Exceed Natures CO2 Drain Emissions o CO2 due to ossil uel burning and cement manuacture areincreasing, while the capacity o sinks that take up CO2or example, plants on land and in the oceanare

decreasing. Atmospheric CO2 is increasing as a result. Source: National Research Council, 2011a

Clues rom the fngerprint o carbon dioxide.In a process that takes place overmillions o years, carbon rom the decay o plants and animals is stored deep in the Earths crust inthe orm o coal, oil, and natural gas (see Figure 4). Because this ossil carbon is so old, it containsvery little o the radioisotope carbon-14a orm o the carbon that decays naturally over long timeperiods. When scientists measure carbon-14 levels in the atmosphere, they nd that it is much lowerthan the levels in living ecosystems, indicating that there is an abundance o old carbon. While asmall raction o this old carbon can be attributed to volcanic eruptions, the overwhelming majoritycomes rom the burning o ossil uels. Average CO2 emissions rom volcanoes are about 200 milliontons per year, while humans are emitting an estimated 36 billion tons o CO2 each year, 80-85% owhich are rom ossil uels.

Box 2


11/40

9

Greenhouse gases are reerred toas orcing agents because otheir ability to change the planets

energy balance. A orcing agent

can push Earths temperature

up or down. Greenhouse gases

dier in their orcing power.

For example, a single methane

molecule has about 25 times

the warming power o a single

CO2 molecule. However, CO2 has amuch larger overall warming eect than

methane because it is much more abundant and

stays in the atmosphere or much longer periods

o time. Scientists can calculate the orcing power

o greenhouse gases based on the changes in their

concentrations over time and on physically based

calculations o how they transer energy through the

atmosphere.

Some orcing agents push Earths energy balance

toward cooling, osetting some o the heating

associated with greenhouse gases. For example,

some aerosolswhich are tiny liquid or solid

particles suspended in the atmosphere, such as those

that make up most o the visible air pollutionhave

a cooling eect because they scatter a

portion o incoming sunlight back into

space (see Box 3). Human activities,

especially the burning o ossil uels,

have increased the number o

aerosol particles in the atmosphere,

especially over and around major

urban and industrial areas.

Changes in land use and land

cover are another way that human

activities are infuencing Earths climate.

Deorestation is responsible or 10% to 20% o

the excess CO2 emitted to the atmosphere each

year, and, as has already been discussed, agriculture

contributes nitrous oxide and methane. Changes in

land use and land cover also modiy the refectivity

o Earths surace; the more refective a surace, the

more sunlight is sent back into space. Cropland is

generally more refective than an undisturbed orest,

while urban areas oten refect less energy than

undisturbed land. Globally, human land use changes

are estimated to have a slight cooling eect.

When all human and natural orcing agents are

considered together, scientists have calculated that

the net climate orcing between 1750 and 2005 is

How much are human activitiesheating Earth?

Warming and Cooling Eects o Aerosols Aerosols are tiny liquid or solid particlessuspended in the atmosphere that come rom a number o human activities, such as ossil uelcombustion, as well as natural processes, such as dust storms, volcanic eruptions, and sea sprayemissions rom the ocean. Most o our visible air pollution is made up o aerosols. Most aerosolshave a cooling eect, because they scatter a portion o incoming sunlight back into space, althoughsome particles, such as dust and soot, actually absorb some solar energy and thus act as warmingagents. Many aerosols also enhance the refection o sunlight back to space by making cloudsbrighter, which results in additional cooling. Many nations, states, and communities have takenaction to reduce the concentrations o certain air pollutants such as the sulate aerosols responsibleor acid rain. Unlike most o the greenhouse gases released by human activities, aerosols onlyremain in the atmosphere or a short timetypically a ew weeks.

Box 3


12/40

10

pushing Earth toward warming (Figure 8). The extra

energy is about 1.6 Watts per square meter o Earths

surace. When multiplied by the surace area o

Earth, this energy represents more than 800 trillion

Watts (Terawatts)on a per year basis, thats about

50 times the amount o energy people consume

rom all energy sources combined! This extra energy

is being added to Earths climate system every

second o every day.

The total amount o warming that will occur in

response to a climate orcing is determined by a

variety o eedbacks, which either ampliy or dampen

the initial warming. For example, as Earth warms,

polar snow and ice melt, allowing the darker colored

land and oceans to absorb more heatcausing Earth

to become even warmer, which leads to more snow

and ice melt, and so on (see Figure 9). Another impor-

tant eedback involves water vapor. The amount o

water vapor in the atmosphere increases as the ocean

surace and the lower atmosphere warm up; warm-

ing o 1C (1.8F) increases water vapor by about

7%. Because water vapor is also a greenhouse gas,

this increase causes additional warming. Feedbacks

that reinorce the initial climate orcing are reerred

to in the scientic community as positive, or ampli-

ying, eedbacks.

There is an inherent time lag in the warming that

is caused by a given climate orcing. This lag occurs

because it takes time or parts o Earths climate

systemsespecially the massive oceansto warm

or cool. Even i we could hold all human-produced

orcing agents at present-day values, Earth would

continue to warm well beyond the 1.4F already ob-

served because o human emissions to date.

Warming and Cooling Infuences on EarthSince 1750 The warming and cooling infuences(measured in Watts per square meter) o various cli-mate orcing agents during the Industrial Age (romabout 1750) rom human and natural sources has beencalculated. Human orcing agents include increases ingreenhouse gases and aerosols, and changes in landuse. Major volcanic eruptions produce a temporarycooling eect, but the Sun is the only major naturalactor with a long-term eect on climate. The net e-ect o human activities is a strong warming infuence

o more than 1.6 Watts per square meter. Source: Na-tional Research Council, 2010a (Depiction courtesy U.S.Global Climate Research Program)

Energy (Watts/m2)

FIGURE 8

TEMPERATURES RISE

AS REFLECTIVE

ICE DISAPPEARS,

DARKER OCEAN

WATER ABSORBS

MORE HEAT

ARCTIC SEA

ICE MELTS

Climate Feedback Loops The amount o warmingthat occurs because o increased greenhouse gasemissions depends in part on eedback loops.Positive (ampliying) eedback loops increase thenet temperature change rom a given orcing, whilenegative (damping) eedbacks oset some o thetemperature change associated with a climate orcing.The melting o Arctic sea ice is an example o a positiveeedback loop. As the ice melts, less sunlight is refectedback to space and more is absorbed into the darkocean, causing urther warming and urther melting oice. Source: National Research Council, 2011dFIGURE 9


13/40

11

Another way to test a scientic theory is to in-vestigate alternative explanations. Becausethe Suns output has a strong infuence on Earths

temperature, scientists have examined records o

solar activity to determine i changes in solar output

might be responsible or the observed global warm-

ing trend. The most direct measurements o solar

output are satellite readings, which have been avail-

able since 1979. These satellite records show that

the Suns output has not shown a net increase dur-

ing the past 30 years (Figure 10) and thus cannot be

responsible or the warming during that period.

Prior to the satellite era, solar energy output had

to be estimated by more indirect methods, such as

records o the number o sunspots observed each

year, which is an indicator o solar activity. These

indirect methods suggest there was a slight increase

in solar energy reaching Earth during the rst ew

Measures o the Suns EnergySatellite measurements o the Sunsenergy incident on Earth, available since1979, show no net increase in solarorcing during the past 30 years. Theyshow only small periodic variationsassociated with the 11-year solar cycle.Source: National Research Council, 2010a

Warming Patterns in the Layers othe Atmosphere Data rom weatherballoons and satellites show a warmingtrend in the troposphere, the lowerlayer o the atmosphere, which extendsup about 10 miles (lower graph), anda cooling trend in the stratosphere,which is the layer immediately abovethe troposphere (upper graph). Thisis exactly the pattern expected rom

increased greenhouse gases, which trapenergy closer to the Earths surace.Source: National Research Council, 2010a

FIGURE 11

FIGURE 10

How do we know the current warming trend

isnt caused by the Sun?


14/40

12

decades o the 20th century. This increase may have

contributed to global temperature increases during

that period, but does not explain warming in the

latter part o the century.

Further evidence that current warming is nota result o solar changes can be ound in the

temperature trends in the dierent layers o the

atmosphere. These data come rom two sources:

weather balloons, which have been launched

twice daily rom hundreds o sites worldwide

since the late 1950s, and satellites, which have

monitored the temperature o dierent layers o the

atmosphere since the late 1970s. Both o these data

sets have been heavily scrutinized, and both show a

warming trend in the lower layer o the atmosphere

(the troposphere) and a cooling trend in the

upper layer (the stratosphere) (Figure 11). This isexactly the vertical pattern o temperature changes

expected rom increased greenhouse gases,

which trap energy closer to the Earths surace. I

an increase in solar output were responsible or

the recent warming trend, the vertical pattern o

warming would be more uniorm through the

layers o the atmosphere.

How do we know that the current warmingtrend is not caused by natural cycles?

Detecting climate trends iscomplicated by the actthat there are many natural

variations in temperature,

precipitation, and otherclimate variables. These

natural variations are

caused by many dierent

processes that can occur

across a wide range o

timescalesrom a particularly

warm summer or snowy winter

to changes over many millions o

years.

Among the most well-known short-term cli-

matic fuctuations are El Nio and La Nia, which

are periods o natural warming and cooling in the

tropical Pacic Ocean. Strong El Nio and La Nia

events are associated with signicant year-to-year

changes in temperature and rainall patterns across

many parts o the planet, including the United

States. These events have been linked to a number

o extreme weather events, such as the 1992 food-

ing in midwestern states and the severe droughts

in southeastern states in 2006 and 2007. Globally,

temperatures tend to be higher

during El Nio periods, such as

1998, and lower during La

Nia years, such as 2008.

However, these up-and-down fuctuations are

smaller than the 20th cen-

tury warming trend; 2008

was still quite a warm year

in the long-term record.

Natural climate variations can

also be orced by slow changes in

the Earths orbit around the Sun that

aect the solar energy received by Earth, as

is the case with the Ice Age cycle (see pp. 18-19)

or by short-term changes in the amount o volca-

nic aerosols in the atmosphere. Major eruptions,

like that o Mount Pinatubo in 1991, spew huge

amounts o particles into the stratosphere that

cool Earth. However, surace temperatures typically

rebound in 2-5 years as the particles settle out o

the atmosphere. The short-term cooling eects o

several large volcanic eruptions can be seen in the

20th century temperature record, as can the global

temperature variations associated with several


15/40

13

strong El Nio and La Nia events, but an overall

warming trend is still evident (Figure 12).

In order to put El Nio and La Nia events and

other short-term natural fuctuations into perspec-

tive, climate scientists examine trends over severaldecades or longer when assessing the human infu-

ence on the climate system. Based on a rigorous as-

sessment o available temperature records, climate

orcing estimates, and sources o natural climate

variability, scientists have concluded that there is a

more than 90% chance that most o the observed

global warming trend over the past 50 to 60 years

can be attributed to emissions rom the burning o

ossil uels and other human activities.

Such statements that attribute climate change

to human activities also rely on inormation rom

FIGURE 12

Short-term Temperature Eectso Natural Climate VariationsNatural actors, such as volcaniceruptions and El Nio and La Niaevents, can cause average globaltemperatures to vary rom one yearto the next, but cannot explain thelong-term warming trend over thepast 60 years. Image courtesy o theMarian Koshland Science Museum

FIGURE 13

Model Runs With and WithoutHuman Infuences Modelsimulations o 20th-century suracetemperatures more closely matchobserved temperature when bothnatural and human infuences are

included in the simulations. Theblack line shows an estimate oobserved surace temperatureschanges. The blue line shows resultsrom models that only includenatural orcings (solar activity andvolcanoes). The red-shaded regionsshow results rom models thatinclude both natural and humanorcings. Source: Meehl et al, 2011


16/40

climate models (see Box 4). Scientists have usedthese models to simulate what would have happened

i humans had not modied Earths climate during

the 20th centurythat is, how global tempera-

tures would have evolved i only natural actors

(volcanoes, the Sun, and internal climate variability)

were infuencing the climate system. These undis-

turbed Earth simulations predict that, in the ab-sence o human activities, there would have been

negligible warming, or even a slight cooling, over

the 20th century. When greenhouse gas emissions

and other activities are included in the models, how-

ever, the resulting surace temperature changes more

closely resemble the observed changes (Figure 13).

14

What are climate models? For several decades, scientists have used the worlds most ad-vanced computers to simulate the Earths climate. These models are based on a series o mathemati-cal equations representing the basic laws o physicslaws that govern the behavior o the atmo-sphere, the oceans, the land surace, and other parts o the climate system, as well as the interactionsamong dierent parts o the system. Climate models are important tools or understanding past,present, and uture climate change. Climate models are tested against observations so that scientistscan see i the models correctly simulate what actually happened in the recent or distant past. Imagecourtesy Marian Koshland Science Museum

Box 4

EVAPORATION

CONDENSATION &

CONVECTION

EXCHANGE OF HEAT &

GASES BETWEEN ATMOSPHERE,

SEA ICE & OCEAN

GLACIER MELT

RADIATIVE

EXCHANGE

TERRESTRIAL CARBON

CYCLE

OCEAN CARBON CYCLEOCEAN

CIRCULATION

WATER STORAGE

IN ICE & SNOW

SURFACE

RUN-OFF

AIR POLLUTION

WIND

CLOUDS & WATER VAPOR

EVAPOTRANSPIRATION


17/40

R ising temperatures due to increasing greenhousegas concentrations have produced distinct pat-terns o warming on Earths surace, with stronger

warming over most land areas and in the Arctic.

There have also been signicant seasonal dierences

in observed warming. For example, the second hal

o the 20th century saw intense winter warming

across parts o Canada, Alaska, and northern Europe

and Asia, while summer warming was particularly

strong across the Mediterranean and Middle East

and some other places, including parts o the U.S.

west (Figure 15). Heat waves and record high tem-

peratures have increased across most regions o the

world, while cold snaps and record cold tempera-

tures have decreased.

Global warming is also having a signicant im-

pact on snow and ice, especially in response to the

strong warming across the Arctic. For example,

the average annual extent o Arctic sea ice has

dropped by roughly 10% per decade since satel-

lite monitoring began in 1978 (Figure 16). This

melting has been especially strong in late summer,

leaving large parts o the Arctic Ocean ice-ree or

weeks at a time and raising questions about eects

on ecosystems, commercial shipping routes, oil

and gas exploration, and national deense. Many

o the worlds glaciers and ice sheets are melting in

response to the warming trend, and long-term av-

erage winter snowall and snowpack have declined

in many regions as well, such as the Sierra Nevada

mountain range in the western United States.

Much o the excess heat caused by human-emit-

ted greenhouse gases has warmed the worlds

oceans during the past several decades. Water ex-

pands when it warms, which leads to sea-level rise.

Water rom melting glaciers, ice sheets, and ice caps

also contributes to rising sea levels. Measurements

made with tide gauges and augmented by satellites

show that, since 1870, global average sea level has

risen by about 8 inches (0.2 meters). It is estimated

15

FIGURE 15

Patterns o Warming in Winter and SummerTwenty-year average temperatures or 1986-2005 comparedto 1955-1974 show a distinct pattern o winter and summer warming. Winter warming has been intense acrossparts o Canada, Alaska, northern Europe, and Asia, and summers have warmed across the Mediterranean andMiddle East and some other places, including parts o the U.S. west. Projections or the 21st century show asimilar pattern. Source: National Research Council, 2011a

What other climate changes and impactshave already been observed?


18/40

16

that roughly one-third o the total sea-level rise over

the past our decades can be attributed to ocean

expansion, with most o the remainder due to ice

melt (Figure 17).

Because CO2 reacts in seawater to orm carbonic

acid, the acidication o the worlds oceans is an-

other certain outcome o elevated CO2 concentra-

tions in the atmosphere (Figure 18). It is estimated

that the oceans have absorbed between one-quarterand one-third o the excess CO2 rom human activi-

ties, becoming nearly 30% more acidic than during

preindustrial times. Geologically speaking, this large

change has happened over a very short timerame,

and mounting evidence indicates it has the poten-

tial to radically alter marine ecosystems, as well as

the health o coral rees, shellsh, and sheries.

Another example o a climate change observed

during the past several decades has been changes in

the requency and distribution o precipitation. Total

precipitation in the United States has increased by

about 5% over the past 50 years, but this has not

been geographically uniormconditions are gener-

ally wetter in the Northeast, drier in the Southeast,

and much drier in the Southwest.

Warmer air holds more water vapor, which has led

to a measurable increase in the intensity o precipita-

FIGURE 16

Loss o Arctic Sea Ice Satellite-based measurementsshow a steady decline in the amount o September(end o summer) Arctic sea ice extent rom 1979 to2009 (expressed as a percentage dierence rom 1979-2000 average sea ice extent, which was 7.0 millionsquare miles). The data show substantial year-to-year

variability, but a long-term decline in sea ice o morethan 10% per decade is clearly evident, highlighted bythe dashed line. Source: National Research Council, 2010a

FIGURE 17

Contributors to Sea-Level RiseSea level has risen steadily over the past ew decadesdue to various contributors: thermal expansion in theupper 700 meters o ocean (red) and deeper oceanlayers (orange), meltwater rom Antarctic and Green-land ice sheets (blue), meltwater rom glaciers andice caps (purple), and water storage on land (green).Source: National Research Council, 2011a


19/40

17

tion events. In the United States, or example, the

raction o total precipitation alling in the heaviest

1% o rainstorm increased by about 20% over the

past century, with the northeastern states experienc-

ing an increase o 54%. This change has increased

the risk o fooding and puts additional stress on

sewer and stormwater management systems.

As the climate has changed, many species have

shited their range toward the poles and to higher

altitudes as they try to stay in areas with the sameambient temperatures. The timing o dierent

seasonal activities is also changing. Several plant

species are blooming earlier in Spring, and some

birds, mammals, sh, and insects are migrating

earlier, while other species are altering their seasonal

breeding patterns. Global analyses show these

behaviors occurred an average o 5 days earlier

per decade rom 1970 to 2000. Such changes can

disrupt eeding patterns, pollination, and other vital

interactions between species, and they also aect

the timing and severity o insects, disease outbreaks,

and other disturbances. In the western United

States, climate change has increased the population

o orest pests such as the pine beetle.The next section describes how observed climate

trends and impacts are predicted to continue i

emissions o human-produced greenhouse gases are

maintained during the next century and beyond.

FIGURE 18

Evidence o Ocean Acidication Withexcess CO2 building up in the atmo-sphere, scientists wanted to know i itwas also accumulating in the ocean.

Studies that began in the mid-1980sshow that the concentration o CO2 inocean water (in blue, calculated romthe partial pressure o CO2 in seawater)has risen in parallel with the increasein atmospheric CO2 (in red, part o theKeeling curve). At the same time, theocean has become more acidic, becausethe CO2 reacts with seawater to ormcarbonic acid. The orange dots are directmeasurements o pH in surace seawater(lower pH being more acidic), and thegreen dots are calculated based on thechemical properties o seawater. Source:National Research Council, 2010d

Climate change has increased the population o orest pests in the western United States. The red trees in thisphoto o Dillon Reservoir in Colorado have died rom an inestation o mountain pine beetle.


20/40

18

Perhaps the most dramatic example o natural climate variability over long timeperiods is the Ice Age cycle. Detailed analyses o ocean sediments, ice cores, and

other data show that or at least 800,000 years, and probably or the past 4 to 5

million years, the Earth has gone through extended periods when temperatures

were much lower than today and thick blankets o ice covered large areas o the

Northern Hemisphere. These long cold spells, which typically lasted or around

100,000 years, were interrupted by shorter warm interglacial periods, including the

past 10,000 years (Figure 14).

Through a convergence o theory, observations, and modeling, scientists have

deduced that the ice ages are caused by slight recurring variations in Earths

orbit that alter the amount and seasonal distribution o solar energy reaching theNorthern Hemisphere. These relatively small changes in solar energy are reinorced

over thousands o years by gradual changes in Earths ice cover (the cryosphere)

and ecosystems (the biosphere), eventually leading to large changes in global

the iCe ages

800,000 Years o Temperatureand Carbon Dioxide RecordsAs ice core records rom Vostok,Antarctica, show, the tempera-ture near the South Pole hasvaried by as much as 20F (11C)

during the past 800,000 years.The cyclical pattern o tempera-ture variations constitutes the iceage/interglacial cycles. Duringthese cycles, changes in carbondioxide concentrations (in red)track closely with changes intemperature (in blue), with CO2lagging behind temperaturechanges. Because it takes a whileor snow to compress into ice, icecore data are not yet availablemuch beyond the 18th century atmost locations. However, atmo-spheric carbon dioxide levels, as

measured in air, are higher todaythan at any time during the past800,000 years. Source: NationalResearch Council, 2010a

FIGURE 14


21/40

19

temperature. The average globaltemperature change during an ice age

cycle, which occur over about 100,000

years, is on the order o 9F 2F (5C

1C).

The data show that in past ice age

cycles, changes in temperature have

ledthat is, started prior tochanges

in CO2. This is because the changes

in temperature induced by changes

in Earths orbit around the Sun leadto gradual changes in the biosphere

and the carbon cycle, and thus CO2,

reinorcing the initial temperature trend.

In contrast, the relatively rapid release o

CO2 and other greenhouse gases since

the start o the Industrial Revolution

rom the burning o ossil uel has,

in essence, reversed the pattern: the

additional CO2 is acting as a climate

orcing, with temperatures increasing

aterward.

The ice age cycles nicely illustrate

how climate orcing and eedback

eects can alter Earths temperature, but

there is also direct evidence rom past

climates that large releases o carbon

dioxide have caused global warming.

One o the largest known events o this type is called the Paleocene-Eocene

Thermal Maximum, or PETM, which occurred about 55 million years ago,

when Earths climate was much warmer than today. Chemical indicators

point to a huge release o carbon dioxide that warmed Earth by another 9F

and caused widespread ocean acidifcation. These climatic changes were

accompanied by massive ecosystem changes, such as the emergence o

many new types o mammals on land and the extinction o many bottom-

dwelling species in the oceans.

The U.S. Geological Survey National Ice Core Labstores ice cores samples taken rom polar ice capsand mountain glaciers. Ice cores provide cluesabout changes in Earths climate and atmospheregoing back hundreds o thousands o years.


22/40

20

Fortunately, scientists have made great strides in

predicting the amount o temperature change that

can be expected or dierent amounts o uture

greenhouse gas emissions and in understanding

how increments o globally averaged

temperaturesincreases o 1C, 2C, 3C and

so orthrelate to a wide range o impacts.

Many o these projected impacts pose serious

risks to human societies and things people care

about, including water resources, coastlines,

inrastructure, human health, ood security, and

land and ocean ecosystems.

Warming, Climate Changesand impacts

in the 21st Century and BeyondPart II

In order to respondeectively to the risksposed by uture climate

change, decision makers

need inormation on the

types and severity o

impacts that might

be expected.


23/40

21

The biggest actor in determining uture globalwarming is projecting uture emissions o CO2and other greenhouse gaseswhich in turn depend

on how people will produce and use energy, what

national and international policies might be imple-

mented to control emissions, and what new tech-

nologies might become available. Scientists try to

account or these uncertainties by developing dier-

ent scenarios o how uture emissionsand hence

climate orcingwill evolve. Each o these scenarios

is based on estimates o how dierent socioeco-

nomic, technological, and policy actors will change

over time, including population growth, economic

activity, energy-conservation practices, energy tech-

nologies, and land use.

Scientists use climate models (see Box 4, p.14)

to project how the climate system will respond to

dierent scenarios o uture greenhouse gas concen-trations. Typically, many dierent models are used,

each developed by a dierent modeling team.

Each model uses a slightly dierent set o mathe-

matical equations to represent how the atmosphere,

oceans, and other parts o the climate system inter-

act with each other and evolve over time. Models

are routinely compared with one another and tested

against observations to evaluate the accuracy and

robustness o model predictions.

The most comprehensive suite o modeling ex-

periments to project global climate changes was

completed in 2005.1 It included 23 dierent models

rom groups around the world, each o which used

the same set o greenhouse gas emissions scenarios.

Figure 19 shows projected global temperature

changes associated with high, medium-high, and

low uture emissions (and also the committed

FIGURE 19

Projected temperature change or threeemissions scenarios Models project globalmean temperature change during the 21stcentury or dierent scenarios o utureemissionshigh (red), medium-high (green)and low (blue) each o which is based ondierent assumptions o uture populationgrowth, economic development, lie-stylechoices, technological change, and avail-

ability o energy alternatives. Also shown arethe results rom constant concentrationscommitment runs, which assume that atmo-spheric concentrations o greenhouse gasesremain constant ater the year 2000. Eachsolid line represents the average o modelruns rom dierent modeling using the samescenario, and the shaded areas provide ameasure o the spread (one standard devia-tion) between the temperature changesprojected by the dierent models. Source:National Research Council, 2010a

1The modeling experiments were part o the World Cli-mate Research Programmes Coupled Model Intercom-

parison Project phase 3 (CMIP3) in support o the Inter-governmental Panel on Climate Change (IPPC) FourthAssessment Report.

How do scientists projectfuture climate change?


24/40

22

warmingwarming that will occur as a result o

greenhouse gases that have already been emitted).

Continued warming is projected or all three uture

emission scenarios, but sharp dierences in global

average temperature are clearly evident by the endo the century, with a total temperature increase in

2100, relative to the late 20th century, ranging rom

less than 2F (1.1C) or the low emissions scenario

to more than 11F (6.1C) or the high emissions

scenario. These results show that human decisions

can have a very large infuence on the magnitude outure climate change.

How will temperatures be affected?

L ocal temperatures vary widely rom day to day,week to week, and season to season, but howwill they be aected on average? Climate modelers

have begun to assess how much o a rise in averagetemperature might be expected in dierent regions

(Figure 20). The local warmings at each point

on the map are rst divided by the correspond-

ing amount o global average warming and then

scaled to show what pattern o warming would be

expected. Warming is greatest in the high latitudes

o the Northern Hemisphere and is signicantlylarger over land than over ocean.

As average temperatures continue to rise, the

number o days with a heat index above 100F

FIGURE 20

Projected warming or three emissions scenarios Models project the geographical pattern o annual averagesurace air temperature changes at three uture time periods (relative to the average temperatures or the period19611990) or three dierent scenarios o emissions. The projected warming by the end o the 21st century is lessextreme in the B1 scenario, which assumes smaller greenhouse gas emissions, than in either the A1B scenario orthe A2 business as usual scenario. Source: National Research Council 2010a


25/40

23

(the heat index combines temperature and humid-

ity to determine how hot it eels) is projected to

increase throughout this century (Figure 21). By

the end o the century, the center o the United

States is expected to experience 60 to 90 ad-

ditional days per year in which the heat index is

more than 100F. Heat waves also are expected

to last longer as the average global temperature

increases. It ollows that as global temperatures

rise, the risk o heat-related illness and deaths also

should rise. Similarly, there is considerable con-

dence that cold extremes will decrease, as will

cold-related deaths. The ratio o record high tem-

peratures to record low temperatures, currently

2 to 1, is projected to increase to 20 to 1 by mid-

century and 50 to 1 by the end o the century or

a mid-range emissions scenario. FIGURE 21

Projections o Hotter Days Model projections sug-gest that, relative to the 1960s and 1970s, the numbero days with a heat index above 100F will increasemarkedly across the United States. Image courtesy U.S.Global Climate Research Program

How is precipitation expected to change?Global warming is ex-pected to intensiyregional contrasts in precipi-

tation that already exist: dry

areas are expected to get

even drier, and wet areas

even wetter. This is because

warmer temperatures tend

to increase evaporation romoceans, lakes, plants, and

soil, which, according to both theory and observa-

tions, will boost the amount o water vapor in the

atmosphere by about 7% per 1C (1.8F) o warm-

ing. Although enhanced evaporation provides more

atmospheric moisture or rain and snow in some

downwind areas, it also dries out the land surace,

which exacerbates the impacts o drought in some

regions.

Using the same general

approach as or tempera-

tures, scientists can project

regional and seasonal per-

centage change in precipita-

tion expected or each 1C

(1.8F) o global warming

(Figure 22). The results show

that the subtropics, where

most o the worlds deserts

are concentrated, are likely to see 5-10% reductions

in precipitation or each degree o global warming.

In contrast, subpolar and polar regions are expected

to see increased precipitation, especially during win-

ter. The overall pattern o change in the continental

United States is somewhat complicated, as it lies

between the drying subtropics o Mexico and the

Caribbean and the moistening subpolar regions o


26/40


27/40

25

expected to increase annually by two to our times

per degree o warming (Figure 24). At the same time,

areas dominated by shrubs and grasses, such as parts

o the Southwest, may experience a reduction in re

over time as warmer temperatures cause shrubs and

grasses to die out. In this case, the potential societal

benets o ewer res would be countered by the losso existing ecosystems.

FIGURE 24

Increased Risk o Fire Rising temperatures and in-creased evaporation are expected to increase the risk ore in many regions o the West. This gure shows thepercent increase in burned areas in the West or a 1Cincrease in global average temperatures relative to themedian area burned during 1950-2003. For example,re damage in the northern Rocky Mountain orests,marked by region B, is expected to more than doubleannually or each 1C (1.8F) increase in global averagetemperatures. Source: National Research Council, 2011a

% change in runoff per degree warming(relative to 1971-2001)

FIGURE 23

Changes in Runo per Degree Warming Enhanced evaporation caused by warming is projected to decrease the

amount o runothe water fowing into rivers and creeks in many parts o the United States. Runo is a keyindex o the availability o resh water. The gure shows the percent median change in runo per degree o globalwarming relative to the period rom 1971 to 2000. Red areas show where runo is expected to decrease, greenwhere it will increase. Source: National Research Council, 2011a


28/40

26

A

s global warming continues, the

planets many orms o ice aredecreasing in extent, thickness,

and duration. Models indicate

that seasonally ice-ree condi-

tions in the Arctic Ocean are

likely to occur beore the end

o this century and suggest

about a 25% loss in Septem-

ber sea-ice extent or each

1C (1.8F) in global warming.

In contrast to the Arctic, sea

ice surrounding Antarctica has, onaverage, expanded during the past

several decades. This increase may be linked

to the stratospheric ozone hole over the Antarc-

tic, which developed because o the use o ozone-

depleting chemicals in rerigerants and spray cans.

The ozone hole allows more damaging UV light to

get to the lower atmosphere and, in the Antarctic,

may have also resulted in lower temperatures as

more heat escapes to space. However, this eect is

expected to wane as ozone returns to normal levels

by later this century, due in part to the success othe Montreal Protocol, an international treaty that

banned the use o ozone-depleting

chemicals. Still, Antarctic sea icemay decrease less rapidly than

Arctic ice, in part because the

Southern Ocean stores heat

at greater depths than the

Arctic Ocean, where the

heat cant melt ice as easily.

In many areas o the

globe, snow cover is expect-

ed to diminish, with snowpack

building later in the cold season

and melting earlier in the spring.According to one sensitivity analysis,

each 1C (1.8F) o local warming may lead

to an average 20% reduction in local snowpack in

the western United States. Snowpack has impor-

tant implications or drinking water supply and

hydropower production. In places such as Siberia,

parts o Greenland, and Antarctica, where tem-

peratures are low enough to support snow over

long periods, the amount o snowall may increase

even as the season shortens, because the increased

amount o water vapor associated with warmertemperatures may enhance snowall.

How will sea ice and snow be affected?

How will coastlines be affected?

Some o Earths most densely populated regionslie at low elevation, making rising sea levela cause or concern. Sea-level rise is projected

to continue or centuries in response to human-

caused increases in greenhouse gases, with an

estimated 0.5-1.0 meter (20-39 inches) o mean

sea-level rise by 2100. However, there is evidence

that sea-level rise could be greater than expected

due to melting o sea ice. Recent studies have

shown more rapid than expected melting rom

glaciers and ice sheets. Observed sea-level rise has

been near the top o the range o projections that

were made in 1990 (Figure 25).

Quantiying the uture threat posed to particular

coastlines by rising seas and foods is challeng-

ing. Many nonclimatic actors are involved, such

as where people choose to build homes, and the

risks will vary greatly rom one location to the next.

Moreover, inrastructure damage is oten triggered

by extreme events, or example hurricanes and

earthquakes, rather than gradual change. However,

there are some clear hot spots, particularly in

large urban areas on coastal deltas, including those

o the Mississippi, Nile, Ganges, and Mekong rivers.

I average sea level rises by 0.5 meters (20 inches)

relative to a 1990 baseline, coastal fooding could


29/40

27

aect 5 million to 200 million people worldwide. Up

to 4 million people could be permanently displaced,

and erosion could claim more than 250,000 square

kilometers o wetland and dryland (98,000 square

miles, an area the size o Oregon). Relocations are

already occurring in towns along the coast o Alaska,

where reductions in sea ice and melting permarost

allow waves to batter and erode the shoreline.

Coastal erosion eects at 1.0 meter o sea-level

rise would be much greater, threatening many

parts o the U.S. coastline (Figure 26).

FIGURE 25

Comparison o Projected and Observed Sea-LevelRise Observed sea-level change since 1990 has beennear the top o the range projected by the Intergov-ernmental Panel on Climate Change Third Assessment

Report, published in 1990 (gray-shaded area). The redline shows data derived rom tide gauges rom 1970 to2003. The blue line shows satellite observations o sea-level change. Source: National Research Council, 2011a

FIGURE 26

Projected Eects o Sea-LevelRise on the U.S. East and GulCoasts I sea level were to rise asmuch as 1 meter (3.3-eet), theareas in pink would be susceptibleto coastal fooding. With a 6-me-ter (19.8-oot) rise in sea level,areas shown in red would also besusceptible. The pie charts showthe percentage area o some citiesthat are potentially susceptible at1-meter and 6-meter sea-level rise.Source: National Research Council,2010a


30/40

28

Whether marine or terrestrial, all organisms

attempt to acclimate to a changing environ-ment or else move to a more avorable location

but climate change threatens to push some species

beyond their ability to adapt or move. Special

stress is being placed on cold-adapted species on

mountain tops and at high latitudes. Shits in the

timing o the seasons and lie-cycle events such as

blooming, breeding, and hatching are causing mis-

matches between species that disrupt patterns o

eeding, pollination, and other key aspects o ood

webs. The ability o species to move and adapt also

are hampered by human inrastructural barriers(e.g., roads), land use, and competition or interac-

tion with other species.

In the ocean, circulation changes will be a key

driver o ecosystem impacts. Satellite data show that

warm surace waters are mixing less with cooler,

deeper waters, separating near-surace marine lie

rom the nutrients below and ultimately reducing

the amount o phytoplankton, which orms the

base o the ocean ood web (Figure 27). Climate

change will exacerbate this problem in the tropics

and subtropics. However, in temperate and polar

waters, vertical mixing o waters could increase,

especially with expected losses in sea ice. At the

same time, ocean warming will continue to push the

ranges o many marine species toward the poles.

Changing ocean chemistry can result in other

impactswarmer waters could lead to a decline

in subsurace oxygen, boosting the risk o dead

zones, where species high on the ood chain are

largely absent because o a lack o oxygen. Ocean

acidication, brought on as the oceans take in more

o the excess CO2 will threaten many species over

time, especially mollusks and coral rees. But not all

lie orms will suer: some types o phytoplankton

and other photosynthetic organisms may benet

rom increases in CO2. Ocean acidication will

continue to worsen i CO2 emissions continue un-

abated in the decades ahead.

How will ecosystems be affected?

FIGURE 27

Eects on the Ocean Food Web The growth rateo marine phytoplankton, which orm the base othe ocean ood web, is likely to be reduced over timebecause o higher ocean surace temperatures. This

creates a greater distance between warmer suracewaters and cooler deep waters, separating upper ma-rine lie rom nutrients ound in deep water. The gureshows changes in phytoplankton growth (verticallyintegrated annual mean primary production, or PP),expressed as the percentage dierence between 2090-2099 and 1860-1869) per 1C (1.8F) o global warm-ing. Source: National Research Council, 2011a

The American pika

is a cold-adaptedspecies that isbeing isolatedon mountaintopislands by risingtemperatures.Image courtesyo J. R. Douglass,YellowstoneNational Park.


31/40

29

The stress o climate changeon arming may threatenglobal ood security. Although

an increase in the amount o

CO2 in the atmosphere avors

the growth o many plants, it

does not necessarily translate into

more ood. Crops tend to grow

more quickly in higher temperatures,

leading to shorter growing periods

and less time to produce grains. In addition,a changing climate will bring other hazards,

including greater water stress and the risk o higher

temperature peaks that can quickly damage crops.

Agricultural impacts will vary across regions and by

crop. Moderate warming and associated increases

in CO2 and changes in precipitation are expected

to benet crop and pasture lands in middle to high

latitudes but decrease yield in seasonally dry and

low-latitude areas. In Caliornia, where hal the

nations ruit and vegetable crops are grown, climate

change is projected to decrease yields o almonds,

walnuts, avocados, and table grapes by up to 40

percent by 2050. Regional assessments or other

parts o the world consistently conclude that climate

change presents serious risk to critical staple crops

in sub-Saharan Arican and in places that rely on

water resources rom glacial melt and

snowpack.

Modeling indicates that the

CO2-related benets or some

crops will largely be outweighed

by negative actors i global tem-

perature rises more than 1.0C

(1.8F) rom late 20th-century values

(Figure 28), with the ollowing project-

ed impacts:

For each degree of warming, yields of corn in

the United States and Africa, and wheat in

India, drop by 5-15%

Crop pests, weeds, and disease shift in

geographic range and frequency

If 5C (9F ) of global warming were to be

reached, most regions of the world would

experience yield losses, and global grain prices

would potentially double

Growers in prosperous areas may be able to adapt

to these threats, or example by varying the crops

which they grow and the times at which they are

grown. However, adaptation may be less eective

where local warming exceeds 2C (3.6F) and will be

limited in the tropics, where the growing season is

restricted by moisture rather than temperature.

Loss o Crop Yields per DegreeWarming Yields o corn in the UnitedStates and Arica, and wheat in India,are projected to drop by 5-15% per de-gree o global warming. This gure alsoshows projected changes in yield perdegree o warming or U.S. soybeansand Asian rice. The expected impacts oncrop yield are rom both warming andCO2 increases, assuming no crop adap-tation. Shaded regions show the likelyranges (67%) o projections. Values oglobal temperature change are relativeto the preindustrial value; current globaltemperatures are roughly 0.7C (1.3F)above that value. Source: National Re-search Council, 2011aFIGURE 28

How will agriculture and

food production be affected?


32/40

30

As a result, decision makers o all typesincluding

individuals, businesses, and governments at all levels

are taking or planning actions to respond to climate

change. Depending on how much emissions are

curtailed, the uture could bring a relatively mild

change in climate or it could deliver extreme

changes that could last thousands o years. Thenations scientifc enterprise can contribute both by

continuing to improve understanding o the causes

and consequences o climate change and by improving

and expanding the options available to limit the magnitude o

climate change and to adapt to its impacts.

MakingclecHOicEsPart III

Astrong body o

evidence shows that

climate change is occurring,

is caused largely by human

activities, and poses signif-

cant risks or a broad range

o human and naturalsystems.


33/40

31

A s discussed in Part II o this booklet, improve-ments in the ability to predict climate changeimpacts per degree o warming has made it easier

to evaluate the risks o climate change. Policymak-

ers are let to address two undamental questions:

(1) at what level o warming are risks acceptable

given the cost o limiting them; and (2) what level

o emissions will keep Earth within that level o

warming? Science cannot answer the rst question,

because it involves many value judgments outside

the realm o science. However, much progress has

been made in answering the second.

Even with expected improvements in energy e-

ciency, i the world continues with business as

usual in the way it uses and produce energy, CO2

emissions will continue to accumulate in the atmo-

sphere and warm Earth.2 As illustrated in Figure 29,

to keep atmospheric concentrations o CO2 roughly

steady or a ew decades at any given level to avoid

increasing climate change impacts, global emissions

would have to be reduced by 80%.

Another helpul concept is that the amount o

warming expected to occur rom CO2 emissionsdepends on the cumulative amount o carbon emis-

sions, not on how quickly or slowly the carbon is

added to the atmosphere (Figure 30). Humans have

emitted about 500 billion tons (gigatonnes) o car-

bon to date. Best estimates indicate that adding

about 1,150 billion tons o carbon to the air would

lead to a global mean warming o 2C (3.6F).

How does science informemissions choices?

FIGURE 29

FIGURE 30

Illustrative Example: How Emissions Relate to CO2

Concentrations. Sharp reductions in emissions areneeded to stop the rise in atmospheric concentrationso CO2 and meet any chosen stabilization target. Thegraphs show how changes in carbon emissions (toppanel) are related to changes in atmospheric concen-trations (bottom panel). It would take an 80% reduc-tion in emissions (green line, top panel) to stabilizeatmospheric concentrations (green line, bottom panel)or any chosen stabilization target. Stabilizing emis-sions (blue line, top panel) would result in a continuedrise in atmospheric concentrations (blue line, bottompanel), but not as steep as a rise i emissions continueto increase (red lines). Source: National Research Council,2011a

2Other greenhouse gases are a actor, but CO2 is by arthe most important greenhouse gas in terms o long-termclimate change eects.

Cumulative Emissions and Increases in Global MeanTemperature Recent studies show that or a particularchoice o climate stabilization temperature, there wouldbe only a certain range o allowable cumulative carbonemissions. Humans have emitted a total o about 500 bil-lion tons (gigatonnes) o carbon emissions to date. Theerror bars account or estimated uncertainties in boththe carbon cycle (how ast CO2 will be taken up by theoceans) and in the climate responses to CO2 emissions.Source: National Research Council, 2011a


34/40

32

Adding CO2 more quickly would bring tempera-

tures to that value more quickly, but the value itsel

would change very little.

Because cumulative emissions are what matters,

policies oriented toward the very long term (several

decades into the uture) might be able to ocus less

on speciying exactly when reductions must take

place and more on how much total emissions are al-

lowed over a long periodin eect, a carbon budget.

Such a budget would speciy the amount o total

greenhouse gas that can be emitted during a speci-

ed period o time (say, between now and 2050).

Meeting any specic emissions budget is more

likely the earlier and more aggressively work is done

to reduce emissions (Figure 31). Its like going on a

diet. I a person wants to lose 40 pounds by a cer-

tain event in the uture, it would be much easier to

reach that goal i he or she begins eating less and

exercising more as soon as possible, rather than

waiting to start until the month beore the event.

Meeting an Emissions Budget Meeting any emissionsbudget will be easier the sooner and more aggressivelyactions are taken to reduce emissions. Source: NationalResearch Council

FIGURE 31

As discussed earlier, to limit climatechange in the long term, the most

important greenhouse gas to control

is carbon dioxide, which in the

United States is emitted primarily

as a result o burning ossil uels.

Figure 32 shows the relative

amount o emissions rom resi-

dential, commercial, industrial,

and transportation sources. Its

not really a matter o doing with-

out, but being smarter about how weproduce and use energy.

The United States is responsible or about

hal o the human-produced CO2 emissions already

in the atmosphere and currently accounts or

roughly 20% o global CO2 emissions, despite hav-

ing only 5% o the worlds population. The U.S.

percentage o total global emissions is projected to

decline over the coming decades as emissions

rom rapidly developing nations such as China and

India will continue to grow. Thus,reductions in U.S. emissions

alone will not be adequate to

avert climate change risks.

However, strong U.S. lead-

ershipdemonstrated

through strong domestic

actions, may help infu-

ence other countries to

pursue serious emission

reduction eorts as well.

Several key opportunities toreduce how much carbon dioxide

accumulates in the atmosphere are

available (Figure 33), including:

Reduce underlying demand or goods and ser-

vices that require energy, or example, expand

education and incentive programs to infuence

consumer behavior and preerences; curtail sprawl-

ing development patterns that urther our depen-

dence on petroleum.

What are the choices for reducinggreenhouse gas emissions?


35/40

33

Improve the eciency with which energy is

used, or example, use more ecient methods or

insulating, heating, cooling, and lighting buildings;

upgrade industrial equipment and processes to be

more energy ecient; and encourage the purchase

o ecient home appliances and vehicles.

Expand the use o low- and zero-carbon energy

sources, or example, switch rom coal and oil to

natural gas, expand the use o nuclear power and

renewable energy sources such as solar, wind, geo-

thermal, hydropower, and biomass; capture and

sequester CO2 rom power plants and actories.Capture and sequester CO2 directly rom the

atmosphere, or example, manage orests and soils

to enhance carbon uptake; develop mechanical

methods to scrub CO2 directly rom ambient air.

Advancing these opportunities to reduce emis-

sions will depend to a large degree on private sector

investments and on the behavioral and consumer

choices o individual households. Governments at

ederal, state, and local levels have a large role to

play in infuencing these key stakeholders through

eective policies and incentives. In general, there are

our major tool chests rom which to select policiesor driving emission reductions:

Pricing o emissions such as by means o a car-

bon tax or cap-and-trade system;

mandates or regulations that could include

direct controls on emitters (or example,

through the Clean Air Act) or mandates such as

automobile uel economy standards, appliance

eciency standards, labeling requirements,

building codes, and renewable or low-carbon

portolio standards or electricity generation;

public subsidies or emission-reducing choices

through the tax code, appropriations, or loan

guarantees; and

providing inormation and education and pro-

moting voluntary measures to reduce emissions.

A comprehensive national program would likely

use tools rom all o these areas. Most economists

and policy analysts have concluded, however, that

putting a price on CO2 emissions that is suciently

high and rises over time is the least costly path to

signicantly reduce emissions; and it is the most e-cient incentive or innovation and the long-term

investments necessary to develop and deploy energy

ecient and low-carbon technologies and inrastruc-

ture. Complementary policies may also be needed,

however, to ensure rapid progress in key areas.

U.S. greenhouse gas emissions in 2009 show the rela-tive contribution rom our end-uses: residential, com-mercial (e.g., retail stores, oce buildings), industrial,and transportation. Electricity consumption accountsor the majority o energy use in the residential andcommercial sectors. Image courtesy: U.S. EnvironmentalProtection Agency

FIGURE 32

FIGURE 33

Key Opportunities or ReducingEmissions A chain o actors deter-

mine how much CO2 accumulates inthe atmosphere. Better outcomes(gold ellipses) could result i thenation ocuses on several opportunitieswithin each o the blue boxes. Source:National Research Council, 2010b


36/40

34

Opportunities to Reduce Other Human-

Produced Warming Agents

There are opportunities to reduce emissions o non-

CO2 gases, such as methane, nitrous oxide, and

some industrial gases (e.g., hydrofuorocarbons),

which comprise at least 15% o U.S. greenhouse gas

emissions. Molecule or molecule, these gases are

generally much stronger climate orcing agents than

CO2, although carbon dioxide is the most important

contributor to climate change over the long-term

because o its abundance and long lietime.

Some non-CO2 greenhouse gases can be re-

duced at negative or modest incremental costs. For

example, reducing methane leaks rom oil and gas

systems, coal mining, and landlls is cost-eective

because there is a market or the recovered gas.

Reducing methane also improves air quality.

The largest overall source o non-CO2 green-

house emissions is rom agriculture, in particular,

methane produced when livestock digest their

ood, and also nitrous oxide and methane rom

manure and nitrogen ertilizer. These emissions can

be reduced in many ways, including by employing

precision agriculture techniques that help armers

minimize the over-ertilization practices that lead to

emissions, and by improving livestock waste man-agement systems.

Some shortlived pollutants that are not green-

house gases also cause warming. One example is

black carbon, or soot, emitted rom the burning

o ossil uels, biouels, and biomass (or example,

the dung used in cookstoves in many developing

countries). Black carbon can cause strong local or

regional-scale atmospheric warming where it is

emitted. It can also ampliy warming in some re-

gions by leaving a heat-absorbing black coating on

otherwise refective suraces such as arctic ice andsnow. Reducing emissions o these short-lived warm-

ing agents could help ease climate change in the

near term.

What are the choices for preparing for theimpacts of climate change?

A lthough adaptation planning andresponse eorts are under wayin a number o states, counties, and

communities, much o the nations

experience is in protecting its peo-

ple, resources, and inrastructure

are based on the historic record o

climate variability during a time o

relatively stable climate. Adaptation to

climate change calls or a dierent para-digmone that considers a range o possible

uture climate conditions and associated impacts,

some well outside the realm o past experience.

Adaptation eorts are hampered by a lack o solid

inormation about benets, costs, and the potential

and limits o dierent responses. This is due in part

to the diversity o impacts and vulnerabilities across

the United States and the relatively small body o re-

search that ocuses on climate change

adaptation actions. In the short term,

adaptation actions most easily de-

ployed include low-cost strategies

that oer near-term co-benets, or

actions that reverse maladaptive

policies and practices. In the longer

term, more dramatic, higher cost

responses may be required. Table 1

provides a ew examples o short-term ac-tions that might be considered to address some

o the expected impacts o sea-level rise.

Even though there are still uncertainties regarding

the exact nature and magnitude o climate change

impacts, mobilizing now to increase the nations

adaptive capacity can be viewed as an insurance pol-

icy against climate change risks. The ederal govern-

ment could play a signicant role as a catalyst and


37/40

35

coordinator o local and regional eorts by providing

technical and scientic resources, incentives to begin

adaptation planning, guidance across jurisdictions,

a platorm to share lessons learned, and support o

scientic research to expand knowledge o impacts

and adaptation. In addition to the direct impacts o

climate change, the United States can be indirectly

aected by the impacts o climate change occurring

elsewhere in the world. Thus, it is in the countrys

interest to help enhance the adaptive capacity o

other nations, particularly developing countries that

lack resources and expertise.

TABLE 1. Examples o some adaptation options or one expected outcome o sea-level rise.

Why take action if there are still uncertaintiesabout the risks of climate change?

Further research will never completely eliminateuncertainties about climate change and its risks,given the inherent complexities o the climate sys-

tem and the many behavioral, economic, and tech-

nological actors that are dicult to predict into the

uture. However, uncertainty is not a reason or inac-

tion, and there are many things we already know

about climate change that we can act on. Reasons

or taking action include the ollowing:

The sooner that serious eorts to reduce green-

house gas emissions proceed, the lower the risks

posed by climate change and the less pressure

there will be to make larger, more rapid, and

potentially more expensive reductions later.

Some climate change impacts, once maniest-

ed, will persist or hundreds or even thousands

o years and will be dicult or impossible to

undo. In contrast, many actions taken to re-

spond to climate change could be reversed or

scaled back i they somehow prove to be more

stringent than actually needed.

Each day around the world, major investments

are being made in equipment and inrastruc-

ture that can lock in commitments to more

greenhouse gas emissions or decades to

come. Getting the relevant incentives and poli-

cies in place now will provide crucial guidance

or these investment decisions.


38/40

36

Many actions that could be taken to reduce

vulnerability to climate change impacts are

common sense investments that also will oer

protection against natural climate variations

and extreme events.

The challenge or society is to weigh the risks andbenets and make wise choices even knowing there

are uncertainties, as is done in so many other realms,

or example, when people buy home insurance. A

valuable ramework or supporting climate choices is

an iterative risk management approach. This

reers to a process o systematically identiying risks

and possible response options; advancing a portolio

o actions that are likely to reduce risks across a

range o possible utures; and adjusting responses

over time to take advantage o new knowledge, in-

ormation, and technological capabilities.

ConclusionResponding to climate change is about makingchoices in the ace o risk. Any course o actioncarries potential risks and costs; but doing nothing

may pose the greatest risk rom climate change and

its impacts. Americas climate choices will be made

by elected ocials, business leaders, individuals,

and other decision makers across the nation; and

those choices will involve numerous value judg-

ments beyond the reach o science. However,

robust scientic knowledge and analyses are a

crucial oundation or inorming choices.


39/40

RefeRences

National Research Council, 2010a, Advancing the Science o Climate Change

National Research Council, 2010b, Limiting the Magnitude o Climate Change

National Research Council, 2010c, Adapting to the Impacts o Climate Change

National Research Council, 2011d, Inorming an Eective Response to Climate Change

National Research Council, 2010e, Ocean Acidifcation: A National Strategy to Meet theChallenges o a Changing Ocean

National Research Council 2011a, Climate Stabilization Targets: Emissions, Concentrations,and Impacts or Decades to Millennia

National Research Council, 2011c, Americas Climate Choices

For more inormation, contact the Board on Atmospheric Sciences and Climate at202-334-3512 or visit http://dels.nas.edu/basc. A video based on Part I o this booklet isavailable at http://americasclimatechoices.org.

This booklet was prepared by the National Research Council with support rom the National Oceanic andAtmospheric Administration (NOAA).

Documents

Climate Change: Lines of Evidence