Review and impacts of climate change uncertainties

Review and impacts of climate change uncertainties

850 REVIEW AND IMPACTS OF CLIMATE CHANGE UNCERTAINTIES Mark E. Fernau, William J. Makofske and David W. South This article examines the status of...

1MB Sizes 1 Downloads 26 Views

850

REVIEW AND IMPACTS OF CLIMATE CHANGE UNCERTAINTIES Mark E. Fernau,

William

J. Makofske

and David W. South

This article examines the status of the scientific uncertainties in predicting and verifying global climate change that hinder aggressive policy making. More and better measurements and statistical techniques are needed to detect and confirm the existence of greenhouse-gas-induced climate change, which currently cannot be distinguished from natural climate variability in the historical record. Uncertainties about the amount and rate of change of greenhouse gas emissions also make prediction of the magnitude and timing of climate change difficult. Because of inadequacies in the knowledge and depiction of physical processes and limited computer technology, predictions from existing computer models vary widely, particularly on a regional basis, and are not accurate enough yet for use in policy decisions. The extent of all these uncertainties is such that moving beyond no-regrets measures such as conservation will take political courage and may be delayed until scientific uncertainties are reduced.

The

atmospheric

capable

concentrations

of changing

Revolution,

with

a sharp

policy

makers

believe

global

climate

change

US Environmental is considerable of global

climate,

upswing

immediate caused

Protection

emy of Sciences’

of certain

the Earth’s

disagreement

climate

change

in the past several actions

by these Agency

and the United its

are required emissions,

(EPA),’

Nation?

among and

greenhouse

(CHCs), since

decades.

Some

to deal with

the US

Congress,2

implications.

makers

One

scientists

major

and

including

the National

the issue.

and policy

potentially

the Industrial

the possibility

and organizations

are addressing

scientists

policy

gases

have been increasing

Acad-

However, about

of the

there

the nature

reason

for

this

Mark E. Fernau and David W. South are with the Technology and Environmental Polrcy Section, Environmental Assessment and Information Sciences Division, Argonne National Laboratory, Argonne, IL 60439, USA. William Makofske is Professor of Physics, Ramapo College of New Jersey, 565 Ramaoo Vallev Road, Mahwah, NI 07430, USA, and a Visitrng Scientist in the Technology and Environmental’Policy Section at Argonne National Laboratory. The issues discussed in this paper stem from research funded by the US Department of Energy, Assistant Secretary for Fossil Energy, under contract W-31.109-Eng-38..The views expressed are those of the authors and should not be construed as those of the US Department of Energy or Argonne National Laboratory.

0016-3287/93/0885G14

@ 1993 Butterworth-Heinemann

Ltd

FUTURES October

1993

Review

and impacts

of climate

change

851

uncertainties

-0.4 -0.5 -0.6

L 1860

1870

1880

1890

1900

1910

1920

1930

1940

1950

1960

1970

1980

1990

Figure 1. Annual average global surface temperature expressed as a deviation from 19X-79 mean in “C. Data analysed by P. D. Jones, T. M. L. Wigley and P. B. Wright, as reported in T. A. Boden, P. Kancurnk and M. P. Farrell. Trends ‘90, A Compendium of Data on Global Change (Oak Ridge, TN, Carbon Dioxide Information Analysis Center, Environmental Sciences Division, Oak Ridge National Laboratory Report ORNL/CDIAC-36, 1990).

controversy is the substantial uncertainty surrounding the scientific aspects and modelling of potential climate change from increases in concentrations of CHCs. While the greenhouse (CH) effect is a verified phenomenon in science, the real question hinges on the magnitude of the ‘enhanced CH effect’, ie the change in the CH effect as CHC concentrations in the atmosphere increase. Computer models predict that the additional trapped energy from increased concentrations of CHCs will result in global temperature increases, changes in precipitation, changes in sea level and other climatic effects. This article outlines the current status of scientific uncertainties in predicting and verifying global climate change. The first section focuses on observation of the global warming signal; the second examines climate modelling uncertainties; the third section discusses potential surprises; and the conclusion provides some thoughts about potential policy implications.

Uncertainty

in the global warming

signal

Enough CHCs have been released in the past two centuries that, all things being equal, their effects by now should be evident in the global temperature record. indeed, data for the past 100 years show that the Earth has warmed by slightly less than l”F,j at the lower end of the range of what models predict should be the increase to date from emissions (see Figure I). However, there is uncertainty surrounding whether this trend is real6 and, if real, whether it can be attributed to CHCs.’ It has been suggested that the recent series of global annual temperature records broken in the 1980s and early 1990s may be an indication that the CH enhancement is now visible in the temperature record, but [email protected] has pointed out that, once a new temperature record has been set, the statistical likelihood of

FUTURES October

1993

852

Review

and impact.5 of climate

its being broken

multiple

trend

in the data.

is present The

global

warming location

leaving

especially Attributing

oceans

when

emissions

were

There

are other

factors

CHC

volcanic

change

the Earth,” particles

and cooling

and from

particles.‘”

This

trend

burning

seen

has

been

offset

most

tropospheric Sulphates Because

of the

years

of the irregularity

and small

number

effects

of Mt Pinatubo to

with

pollutants to

data

temperature to provide match

present

statistically precipitation,

for

significant

in

reduce

to space.

it is not thought However,

June 1991

stratosphere

the

eruption

in the stratosphere

and is

and

of aerosols variability,

often

satellite

and

effects

factors

past do

data, together

partitioning

an additional

altered

century, not

with

of observed

it will

since

exist.

more

and other

However, detailed

temperature

the

surface

changes

decade or two of data taking

be

reliable

as

is needed

evidence.”

of attempting

to observe

with

models

is tenuous

the climate

observed (CCMs)

data.

USA

to simulate

from

for temperatures

an increase

in the

1895

warming

Given

daily

signal

lack

1989

not

in winter

minimum

of

(see next

to link

have

is to try

of ability

climate

attempts

to

and changes

mean

the

regional

at best. Not surprisingly,

central

trends

although

can

of solar

of the

may allow

the

climate

record

predictions

this approach

The

the

temperature

circulation

predictions

the loss

sunlight

eruptions,

dioxide

warmed

of the

confounding

definitive

regional

general

section),

these

method

for

ozone

an increase

eruptions more

significant.

sources

unfortunately,

more

Another

although

volcanic

natural

of Nimbus-7

data sets,

origin;

and

the

on

collection

to their

understanding

human

interpret

factor

year timeframe.‘”

of sulphur

biomass

extentI

of volcanic

troposphere,

be why

II. Recently,

patterns.“’

better

from

historical

the

pollution

and could

climate

by reflecting

can be quite

megatons

cooled

circulation

Even

CCM

put 15-30

from

by pollution

that stratospheric

to some

on the l&l00

on temperature

have

atmospheric

recent

effect is important

sunlight

War

from

that reaches

due to the

significant

stratosphere

cooling

radiation

of incoming

halocarbons,

this

subsequent

short-term

difficult

in the

short-term

from

by El Nina

and lifetime

believed

from

surface

trend

exemplified

of solar

World

a potentially

forcing

a warming

variability,”

following

may counter

Global

to the 197Os-a

is felt to be important

it is now

com-

ever.12

reflection

several

that the volcanic

thought

CH

levels

formed for

as

difficulties,

however.

to or mask

reflectivity

decades

and

represen-

of 0.45” + / - 0.15~~~~

the 1940s

temperatures),

effect, especially,

li In addition,

ozone

temperatures

from

the gradual

underrepresented,

these

from

in the amount

in cloud

identified

than

is natural

sea-surface

resulting

in the

hemisphere.

faster

can contribute

changes

changes

pollution

was

southern may

in Pacific

change

from

if a

in the method

countries

is not an easy task, little

There

ash and sulphates,14

World

especially

geographical

an average warming

increasing

concentrations.

to biases

‘” Acknowledging

in CHCs

that

is high,

changes

the data have poor

and Third

showed

years

up, and from

hemisphere.

measurements

(a periodic

to

of the

southern

several

are subject

are built

In addition,

this to an increase

increased

no

measurements

land and ocean data sets suggest

temperature time

most

in the

uncertainties

in the following

areas as they

of measurement.”

tation, bined

times

temperature

of urban

change

regional

found

any

and summer

temperature

was

found.” So far, scientists global

temperature

has been observed.

have not been able to sort record

and to say with

It may be two

more

out

certainty

decades

all these that

before

uncertainties

the enhanced this

in the

CH effect

is possible.

FUTURES October

1993

Review

Climate modelling

and impact-s of climate

change

uncertainties

853

uncertainties

At present, the most common way to predict climate change is through the use of GCM computer models. CCMs are three-d~mensionai grid models that use the laws of physics governing heat, momentum and moisture, together with simplified descriptions of other physical processes, to predict climate variables such as temperature, precipitation and soil moisture. Such models are validated to some extent by simulating correctly the gross features of the present climate and the of Mars and Venus, and past climates.z-’ changing seasons, the atmospheres However, a recent study of model simulations from 14 CCMs applied to present climate found significant differences, pointing to ‘systematic’ deficiencies in the models.“” When models are compared to each other and to observations on a regional scale, disagreements are highlighted.25 For the Great Lakes region, for example, the models cannot even agree on whether summer rainfall---of importance to agriculture--will increase or decrease 26 (see Figure 2). Even superimposing finer grid structures over selected regions has not worked well over regions of complex terrain and surface characteristics, particularly for surface hydrology.27 When CCMs are used to simulate future climate resulting from an instantaneous doubling of carbon dioxide (CO& most mode! predictions yield a range in temperature increase of 2.7”F to 8.1”F (1.5”C to 4.5”CIz8 The models predict substantially greater warming in the polar regions and less in the tropics. The models also predict that global rainfall should increase. Differences occur not only among models, but among different versions of the same model depending on assumptions about physical processes and feedbacks.29 There are numerous reasons for the uncertainties in and the discrepancies among models; some major ones are described below.

Uncertainties

in G/-C emissions and concentrations

Results from climate models are dependent on inputs of present and future CHG emissions and concentrations. Examples of anthropogenic GHC sources are fossil fuel burning and deforestation for carbon dioxide, rice paddies and biomass burning for methane, and refrigerants and propellants for chlorofluorocarbons (CFCs). The contribution of each CHC to global warming depends on its radiative properties, atmospheric lifetime and concentration. In order to produce estimates about contributions to global warming, researchers must make assumptions about the rate of growth in the concentrations of GHCs. These estimates and the emission estimates on which they are based depend on many factors, such as future energy use, population growth, adoption of control strategies, and the like; these factors are difficult to estimate accurately. Even if one could model the behaviour of the climate system perfectly, the uncertainties in future CHC emissions and atmospheric concentrations would make it difficult to predict the timing and magnitude of climate change.

Uncertainties

in sinks

In addition, concentration estimates are affected by uncertainties in sinks, processes that remove CHGs from the atmosphere. Examples include uptake of CO, by vegetation, trees and ocean processes, and photochemical decomposition for nitrous oxide and CFCs. However, questions arise regarding the quantity of CO,

FUTURES

October 1993

854

Review and impact.5 of climate

Winter 0.6

b

change

Spring

uncertainties

Summer

Autumn

Annual

Tl-l

-0.6

Winter

Summer

Spring

Annual

Autumn

Frgure 2. Regional clrmate change predictions for (a) temperature and (b) precipitation for three of the most respected global climate models: Goddard Institute of Space Studies (GISS); Geophysical Fluid Dynamics Lab (GFDL); and Oregon State University (OSU). Taken from US Environmental Protectron Agency, op tit, reference I.

that

is used

CO,

emissions

changing the rate

by vegetation is captured

CO2 levels

atmospheric is decreased,

downwards.

The

long the various

in photosynthesis. by the oceans,

or temperature.

concentration perhaps

the

It is also

j0 If the oceans

could

increase

oceans

would

regarding

sinks

have

gases

in the atmosphere

make

what

might

become

relatively

uncertainties remain

uncertain

or how that fraction

saturated

quickly.

more

time

it difficult

and, thus,

fraction

change with

If the to

mix

relative

CO,,

emission the

to determine

their

of

under

CO, how

import-

ance to CH warming.”

FUTURES October

1993

Review

and impacts

of climate

change

uncertainties

855

Feedbacks Feedback loops in climate modelling are extremely important. While the direct effect of doubling CHCs on the radiative balance is significant (about l”C), the resulting response of the climate system could result in greater or lesser ultimate change. Positive feedback amplifies the effects of original change while negative feedbacks dampen change. Feedback loops are complicated and uncertain, yet accurate prediction of the ultimate impact of increased CHCs requires getting the feedbacks correct in the models. Major feedback uncertainties are described briefly below. Water vapour. The warmer that the Earth is, the more water will evaporate into the atmosphere. Warmer air is capable of retaining more water vapour and, since water vapour is a powerful CHC, this will lead to a still warmer atmosphere. In most CCMs, doubling CO, leads to water vapour feedback that essentially doubles the predicted temperature increase, from 1°C to about 2”C.32 However, this feedback depends on how water vapour is distributed in the atmosphere by cloud transport and processes, and some have argued that increased convection will lead to decreased specific humidity in the upper troposphere, potentially leading to negative rather than positive feedback. 33 Observational studies have so far supported positive feedback, although it has been difficult to validate model water vapour feedback.34 Randall et a/j5 examined differences in 19 CCMs and concluded that the major deficiency of GCMs is in how they model the hydrological cycle. Clouds. Perhaps the dominant factor undermining the accuracy of computer models is the rudimentary representation of clouds. Clouds both reflect sunlight back to space and act as a GH absorber. Generally, for high clouds infra-red absorption (heating) predominates, while for low clouds reflection (cooling) dominates. Currently, clouds are thought to have a net cooling effect,36 but it is not known whether the clouds will warm or cool the Earth in the doubled CO, scenario. The percentage of cloud cover, the vertical location of clouds, their droplet size and the amount of water in the clouds all affect their ability to reflect or trap energy. Because of insufficient knowledge and the coarseness of the model grids, currently none of these parameters is modelled adequate1y.j’ For example, the same model with three different treatments of cloud processes yielded three different temperature increases for doubled CO,, ranging from 3” to 9”F.38 Another study, in which 19 different versions of climate models were compared to one another, showed that for clear skies the models yielded similar climates in response to a sea-surface temperature increase. When clouds were included, the models yielded very different results from each other, with the models not even all agreeing on the direction of the feedback.39 The cloud issue is complicated even more by the recent discovery that clouds formed from man-made sulphate particles are acting to cool the Earth.30 The relative contributions from direct reflection from the sulphate particles themselves v indirect changes to cloud reflectivity, is not yet known. For doubled CO,, GH warming is expected to dominate over sulphate cooling4’ It is also thought that sulphates from dimethylsulphide (DMS) produced by phytoplankton may have a significant effect in producing cloud condensation nuclei, although the evidence is not yet conclusive. 42 While GCMs do not currently address the sulphate issue,

FUTURES October

1993

856

Review and impacts of climate change uncertainties

corrections for sulphates using a simplified climate model reduce radiative forcing and temperature predictions for a doubling of CO,, generally make model predictions more consistent with the temperature record, and increase the climate sensitivity.“’ Oceans. The important role played by oceans in climate change demands that the models include a more accurate ocean component. This currently is not the case. Oceans possess a huge capability to store added heat. The top three metres of the ocean can store as much heat as the entire atmosphere.“4 Because oceans potentially can mix heat downwards several miles, they could act as a major sink for both heat and CO,, thus delaying any potential CH warming. The oceans also transport heat away from the equator in amounts equal to that transported by the atmosphere.45 To date, most climate change analysis has been done with models that have a simple ocean component. A few of the model runs done with more accurate and detailed ocean components result in smaller predicted temperature increases. Despite inclusion of a more realistic ocean component, these models suffer from uncertainties due to the coarse ocean grid resolution and incomplete knowledge about ocean processes. Transport of heat, moisture and salt is not modelled well, and the models cannot reproduce today’s climate without manual adjustment of the transport. 46 Advanced coupled ocean-atmosphere models simulating a transient response to CO, doubling have been found to require significant flux adjustments to prevent drifting to an unrealistic state.“’ A realistic coupled oceanatmosphere model is needed to predict accurately the magnitude and timing of global warming and, particularly, regional climate.“8 Yet accurate coupled oceanatmosphere modelling is presently only in its infancy. Semtner and Chervin4” have been able to resolve ocean eddies on a near-global domain with a parallel computer, but effective use of their model for climate change applications will require computer resources that do not exist yet. Sea ice and snow. The representation of sea ice thickness, extent and behaviour is very crude in the CCMs. 50 The accurate representation of sea ice is very important because its reflectivity has a significant effect on the severity of potential climate change. A model that assumes more initial sea ice will produce a greater temperature response to enhanced CHCs because of the change in reflectivity as the ice melts. Melting ice will also affect the vertical ocean currents that are so important to heat and CO, mixing. Current models ignore transport of the ice pack and the fact that holes and gaps in the ice affect reflectivity of sunlight and exchange of heat. Snow melting is expected to provide a positive feedback to global warming simply because land has a lower albedo than snow and will absorb more solar radiation. However, an intercomparison of 17 CCMs for the effects of snow feedback found substantial differences due to different treatments of cloud interactions and long-wave radiation. 51 It appears that positive feedbacks from changing ice and snow cover are countered by other negative feedback terms.52

Other

climate

modelling

uncertainties

There are other simplifications and knowledge gaps that limit the accuracy of model predictions and estimates of the potential effects of climate change. Only

FUTURES October

1993

Review

and impacts

of climate

change

uncertainties

857

CO, is currently included in models, yet studies have found that the other CHCs may lead to different feedbacks and different climate effects.53 The global models do not contain atmospheric chemistry, which can affect the concentrations of CHCs. The models do not represent realistically the effects of soil and land vegetation on climate. Evaporation and run-off of water depend on how the plants and soil interact with rainfall, and are represented crudely in models.54 As vegetation changes in response to climate change, it could affect the amount of solar radiation that is reflected from the Earth’s surface.55 Because of limits in computer technology and cost, horizontal grid resolution is typically several hundred kilometres, which makes it difficult to include accurately appropriate topography and geography. Mountains and lakes, which have a large influence on regional climate, are omitted or represented in a crude and inaccurate manner. Certain types of storms, clouds and winds that occur on a small scale cannot be represented correctly on such a coarse grid. Such poor resolution enhances uncertainties, particularly for regional climate predictions. Also, the frequent practice of instantly doubling CO, levels to save computer time has been shown to yield regional climates different from those obtained when CO, is slowly increased in a more realistic manner.56 Based on predicted warming, oceanographers have predicted that sea level will rise as the oceans warm and as the glaciers and polar ice-caps begin to melt. Recently, the amount of predicted sea-level rise associated with increasing CHCs has decreased substantially. The current best estimate is an average rise of about 7” by 2030 and about 19” by the end of the 21st century, with a range of almost no change to about 4ft. 57 Clearly, there are substantial uncertainties present in estimating any sea-level rise that might occur.

Summary

of modeling

uncertainties

The reason why models do poorly is a combination of the fact that some important physical processes that influence climate are not understood well and that some processes that are understood well are characterized poorly because of limits in computer technology or modelling capability. The problems regarding water vapour, clouds, oceans, snow and ice, and the other sources of uncertainty (see Figure 3) cause difficulties in modelling the current climate, thus limiting the reliability of predictions under changing CHC conditions. The models perform relatively well in terms of the large-scale circulations, temperatures and seasonal behaviour. However, heat and moisture transport are not modelled well and, when regional areas are examined, many departures from reality can be found; agreement among regional predictions of the best available CCMs continues to be poor. 58 Im p rovements in computing power alone will not greatly improve climate simulations until the physics of climate processes is represented more correctly.5g According to Mahlman?” [To improve model predictions, substantial research efforts are required in] cloud-radiation feedback; upper-troposphere water vapor; convection-cloud coupling; atmosphericPocean fluxes; ocean circulation; sea ice; CO,, CH, and N,O biogeochemistry; radiative-chemical interactions, including ozone; land surface hydrological/biospheric processes; basic budgets and radiative effects of natural and anthropogenic aerosols; atmospheric and oceanic dynamics; turbulence; mathematical modeling techniques; There are now several ongoing programmes

FUTURES

October

1993

which are attempting

to reduce the

858

Revrew and impact.5 of climate

c

change

uncerta!nties

A

c,

2

C-----T--

-

c

Cloud

Effects

Effects of clouds that form on sulphur particles emitted by man-made sources and marine organisms

of

Effects

Hydrological

of snow

processes:

.-f-ype .Amount .Height *Liquid water content .Droplet distribution

cycle

Effects of changes in the sea’s surface temperature on the amount, type

Parameterizations soil moisture

of

Figure 3. Schematic of various imperfectly known factors that complicate global climate change understanding and predictions. Note: Simplification, parameterization and lack of knowledge of the processes, effects and feedbacks depicted here lead to uncertainty in the results of general circulation models.

uncertainties include

present

the

(ARM)

US

and

Computer

(CHAMMP) ment

in

general

Department

(MECCA);

the

the Global

System

Energy’s

Hardware,

programmes;

Observing

circulation

of

Advanced

Model

Energy

(EOS);

and

model

These

Radiation

Mathematics

Evaluation

and Water

the

results.

Atmospheric

and

Consortium

Cycle

Atmospheric

for

Experiment

Model

programmes Measurement Model

Physics

Climate

Assess-

(CEWEX);

the Earth

Intercomparison

Project

(AMIP).hl

Palaeoclimatology The

ability

to simulate

understanding tions. may

model

Indeed,

past

palaeoclimatic

is evidence

which

there

for

25 000

explanations the North collisions,hh

sharp

years

that derive

of uncertainty

changes

in climate

about

accepted 59 million

ago, and other

thermocline

and changes

is important

range

for past climatic Atlantic

change

and for developing

analyses

the

is no readily

deep sea temperature around

climate

feedbacks

be able to narrow

there for

and potential surprises

changes

in CCM over

at different

circulation,“4

These

times

massive

in CO,. h7 Mechanisms

sensitivity

relatively

climatic

short

include abrupt

shifts6’ include

volcano

beyond

validation,

the

for

in CCM CH predicparameter

predictions.62

ago, irregular

significant

model

the climate

explanation. years

for

confidence

However,

time

intervals

a rapid

rise

climate

Some abrupt

in

shifts

suggested changes

in

eruptions,“’

asteroid

Milankovitch

theory,

FUTURES October

1993

Review

and impacts

of climate

change

uncertainties

859

which explains previous climate change based on changes in solar insolation due to the Earth’s orbital variations, appear to be [email protected]’ There is also continued debate over the past and future stability of the West Antarctic ice sheet.6g The present models predict increasing but continuous global temperature rises in response to increases in CHCs. However, there may be missing processes and feedbacks that could lead to rapid discontinuous changes in climate. Some potential candidates for climatic surprises include the release of methane from clathrates in shallow sea sediments, the collapse of the West Antarctic ice sheet, rapid changes in the North Atlantic ocean circulation, collapse of local hydrological cycles in tropical rainforests, a rapid increase in tropospheric ozone build-up, and an increased potential of an Arctic ozone hole. ‘OUntil past climate changes are understood better, it is difficult to rule out the possibility of discontinuous climate change with potentially serious impacts.

Conclusions Currently, it is impossible to be confident of the prediction accuracy for the timing, amount and regional distribution of any potential future climate change resulting from increased emissions of GHCs. This conclusion stems from the existence of uncertainties in predicting future emissions and concentrations of CHCs, the lack of detection of the predicted warming that should already have occurred according to models, and the many uncertainties present in all aspects of climate modelling. Balanced against these daunting uncertainties is the certainty that concentrations of CHCs in the atmosphere are increasing and that emissions of CHCs continue to rise. That these additional CHGs cause radiative forcing is well established scientifically. Unfortunately, our present knowledge of the complex climate system is inadequate for accurate predictions of climate change. Conversely, while these uncertainties raise serious questions for policy makers and scientists, they do not rule out the possibility that the climate change that is predicted by current models could be realized. A large global warming could have serious impacts on human society and ecosystems. In addition, there is a potential for surprises,‘l both in the response of the climate system to change and in human and ecological responses. There has been much debate over whether societies should act to reduce or prepare to live with climate change. ‘* There has also been considerable activity to estimate economic costs and potential dislocations of mitigation, given the present infrastructural dependence on fossil fuels.73 Given the large range of uncertainties and the real possibilities of surprises, reducing the probability of extreme adverse impacts is desirable. It is also clear that many mitigation efforts have significant environmental and economic benefits, independent of GH warming reduction,74 and should be and are being undertaken. However, it is likely that additional reductions in emissions beyond these ‘no-regrets’ efforts will be necessary to reduce GHG concentrations, and climate change uncertainties have made policy makers reluctant to act thus far. It appears that it will take several decades before all the scientific questions are answered satisfactorily. In the meantime, researchers are working hard to give policy makers better information and an awareness of its associated uncertainty. Reduction in uncertainty will lend more comfort as the policy maker considers the value judgment: should we buy insurance against possible climate change and how much should we spend?

FUTURES October

1993

860

Review

and impacts

of climate

change

uncertainties

Notes and references 1.

2.

3.

4.

5. 6. 7. 8. 9. 10. 11. 12. 13.

14. 15.

16.

17.

18.

19. 20. 21.

US Environmental Protection Agency, J. 8. Smith and D. A. Tirpak (editors), The Potential Effects of Global Climate Change on the United States (Washington, DC, Report to Congress, US Environmental Protection Agency, EPA-230-05-89-050, 1989). US Congress, Office of Technology Assessment, Changing by Degrees: Steps to Reduce Creenhouse Gases (OTA-0-482, 1991, available from Superintendent of Documents, US Government Printing Office, Washington, DC, 20402). For a review, see A. L. Fish and T. J. Elliott, A Comparison of Global Climate Change Legislation in the 102nd Congress (Argonne, IL, Argonne Natronal Laboratory, Internal Report, 1992). National Academy of Sciences (NAS), Policy implications of Greenhouse Warming (Washington, DC, National Academy Press, 1990). There are three separate panel reports: Mitigation Panel; Adaptation Panel; and Synthesis Panel. IPCC, J. T. Houghton, C. J. Jenkins and J. J. Ephraums (editors), Climate Change: The //‘CC Scientific Assessment (Cambridge, Cambridge University Press, 1990); IPCC, J. T. Houghton, B. A. Callander and S. K. Varney (editors), Climate Change 1992. The Supplementary Report to the /KC Scientific Assessment (Cambridge, Cambridge University Press, 1992). IPCC, Climate Change 1992, op tit, reference 4. P. I. Michaels and D. E. Stooksburv. ‘Global warming: a reduced threat?‘, Bulletin of the American Meteorological Society, 73, 1992: pages 1563-l 577. A. R. Solow and J. M. Broadus, ‘On the detection of greenhouse warming’, Climatic Change, 75, 1989, pages 449-453. G. W. Bassett Jr, ‘Breaking recent global temperature records’, Climatic Change, 21, 1992, pages 3033315. T. R. Karl and P. D. Jones, ‘Urban bias in area-averaged surface air temperature trends’, Bulletin of the American Meteorological Society, 70, 1989, pages 2655270. R. S. Lindzen, ‘Some coolness concerning global warming’, Bulletin of the American Meteorological Society, 71, 1990, pages 2888299. IPCC, Climate Change 1992, op tit, reference 4. Michaels and Stooksbury, op tit, reference 6. T. M. L. Wigley and S. C. B. Raper, ‘Natural variability of the climate system and detection of the greenhouse effect’, Nature, 344, 1990, pages 324-327; D. D. Houghton, R. G. Callimore and L. M. Keller, ‘Stability and variability in a coupled ocean-atmosphere climate model: results of IOOyear simulations’, Journal of Climate, 4, 1991, pages 557-577. C. Brasseur and C. Cranier, ‘Mount Pinatubo, aerosols, chlorofluorocarbons and ozone depletion’, Science, 257, 1992, pages 123991241. J. E. Hansen and A. A. Lacis, ‘Sun and dust versus greenhouse gases: an assessment of therr relative roles in global climate change’, Nature, 346, 1990, pages 713-719; P. M. Kelly and T. M. L. Wigley, ‘The influence of solar forcrng trends on global mean temperature since 1861’, Nature, 347, 1990, pages 46&462; P. M. Kelly and T. M. L. Wigley, ‘Solar cycle length, greenhouse forcing and global climate’, Nature, 360, 1992, pages 328330; M. E. Schlesinger and N. Ramankutty, ‘Implications for global warming of intercycle solar irradiance variations’, Nature, 360, 1992, pages 33(f333; IPCC, C/imate Change 1992, op tit, reference 4. T. M. L. Wigley, ‘Possible climate change due to SO,-derived cloud condensation nuclei’, Nature, 339, 1989, pages 365-367; T. M. L. Wigley, ‘Could reducing fossil-fuel emissions cause global ‘Aerosols, clouds and radiation’, warming?‘, Nature, 349, 1991, pages 503-506; S. Twomey, Atmospheric Environment, 25A, 1991, pages 2435-2442; R. J. Charlson et al, ‘Perturbation of the northern hemisphere radiative balance by backscattenng from anthropogenic sulfate aerosols, Tel/us, 43AB, 1991, pages 152-163; R. J. Charlson et a/, ‘Climate forcing by anthropogenic aerosols’, Science, 255, 1992, pages 423-430; J. Lelieveld and J. Heintzenberg, ‘Sulfate cooling effect on climate through in-cloud oxidation of anthropogenic SO,‘, Science, 258, 1992, pages

117-120. Y. 1, Kaufman

and R. S. Fraser, ‘Fossil fuel and biomass burning effect on climate-heating or cooling?‘, /ourna/ of Climate, 4, 1991, pages 5788588; J. E. Penner, R. E. Dickinson and Cy A. O’Neill, ‘Effects of aerosol from biomass burning on the global radiation budget’, Science, 256, 1992, pages 1432-1433. V. Ramaswamv. M. D. Schwarzkoof and K. P. Shine, ‘Radiative forcing of climate from halocarbon-induced global stratospheric ozone loss’, Nature, 355, 1992, page‘; 81&812; M. D. Schwarzkopf and V. Ramaswamy, ‘Radiative forcing due to ozone in the 1980s: dependence on altitude of ozone change’, Geophysical Research Letters, 20, 1993, pages 205-208. Hansen and Lacis, op tit, reference 15. Brasseur and Cranier, op tit, reference 14. ‘Monitoring global monthly mean surface K. E. Trenberth, J. R. Christy and J. W. Hurrell, temperatures’, /ourna/ of Climate, 5, 1992, pages 1405F1423; P. E. Ardanuy, H. L. Kyle and D. Hoyt, ‘Global relationships among the Earth’s radiation budget, cloudiness, volcanic aerosols,

FUTURES

October

1993

Review

22. 23. 24.

25.

26. 27.

28. 29. 30.

31. 32.

33. 34.

35. 36.

37.

38.

39.

40.

41. 42.

43.

and impacts

of climate

change

uncertainties

861

and surface temperature’, /ourna/ of Climate, 5, 1992, pages 1120-1141. T. R. Karl, R. R. Heim and R. C. Quale, ‘The greenhouse effect in central North America: if not now, when?‘, Science, ,X7,1991, pages 1058-1061. S. H. Schneider, ‘The greenhouse effect: science and policy’, Science, 243, 1989, pages 771-781. C. J. Boer et a/, ‘Some results from an intercomparison of the climates simulated by 14 atmospheric general circulation models’, /ourna/ of Geophysical Research, 97, 1992, pages 12771-12786. ‘The use of general circulation models to predict S. L. Crotch and M. C. MacCracken, regional climate change’, /ourna/ of C//mate, 4, 1991, pages 286303; W. J. Cutowski, Jr, D. S. Gutzler and W-C. Wang, ‘Surface energy balances of three general circulation models: implications for simulating regional climate change’, /ourna/ of Climate, 4, 1991, pages 121-134. US Environmental Protection Agency, op tit, reference 1. F. Ciorgi, M. R. Marinucci and G. Visconti, ‘A 2 X CO, climate change scenario over Europe generated using a limited area model nested in a general circulation model 2-climate change scenario’, Journal of Geophysical Research, 97, 1992, pages 10011-10028. IPCC, Climate Change 1992, op tit, reference 4. J. F. B. Mitchell, C. A. Senior and W. J. Ingram, ‘CO, and climate: a missing feedback?‘, Nature, 347, 1989, pages 132-134; Boer et a/, op tit, reference 24. T. Takahashi, ‘Elements of oceanic CO, feedback in response to climatic changes’, in 8. Manowitz (editor), Global Climate Feedbacks: Proceedings of the Brookhaven National Laboratory Workshop/une 336, 7990cprepared for US Department of Energy, Washington, DC, CONF9006134 (BNL-45321), 1990, available from NTIS, Springfield, VA 22161, USA), pages 95-100; P. P. Tans, I, Y. Fung and T. Takahashi, ‘Observational constraints on the global atmospheric CO, budget’, Science, 247,1990, pages 1431-1438; W. S. Broecker and Y-H. Peng, ‘Interhemispheric Nature, 356, 1992, pages 587-589; 1. L. transport of carbon dioxide by ocean circulations’, Sarmiento and E. T. Sundquist, ‘Revised budget for the oceanic uptake of anthropogenic carbon dioxide’, Nature, 356, 1992, pages 589-593. D. A. Lashof and D. R. Ahuja, ‘Relative contributions of greenhouse gas emissions to global warming’, Nature, 344, 1990, pages 529-531. M. E. Fernau and D. W. South (editors), Effects of Scientific Uncertainties on the Accuracy of Global Climate Change Predictions: A Survey of Recent Literature (Argonne, IL, Argonne National Laboratory, Internal Report, 1991). Lindzen, op tit, reference 10. A. Raval and V. Ramanathan, ‘Observational determination of the greenhouse effect’, Nature, 342, 1989, pages 758-761; D. Rind et a/, ‘Positive water vapour feedback in climate models confirmed by satellite data’, Nature, 349, 1991, pages 5OC-503; M. T. Chahine, ‘The hydrological cycle and its influence on climate’, Nature, 359, 1992, pages 373-380. D. A. Randall et al, ‘Intercomparison and interpretation of surface energy fluxes in atmospheric general circulation models’, /ourna/ of Geophysical Research, 97, 1992, pages 3711-3724. V. Ramanathan et al, ‘Cloud-radiative forcing and climate: results from the Earth Radiation Budget Experiment’, Science, 243, 1989, pages 5763; E. F. Harrison et a/, ‘Seasonal variation of cloud radiative forcing derived from the Earth Radiation Budget Experiment’, /ourna/ of Geophysical Research, 95, 1990, pages 18687718703. K. E. Taylor and S. J. Ghan, ‘An analysis of cloud liquid water feedback and global climate sensitivity in a general circulation model’, /ourna/ of Climate, 5, 1992, pages 907-919; J. F. B. Mitchell and W. J. Ingram, ‘Choice of cloud parameterization scheme has a major effect on the climate sensitivity’, /ourna/ of Climate, 5, 1992, pages 5-21. C. A. Wilson and J. F. B. Mitchell, ‘A doubled CO, climate sensitivity experiment with a global climate model including a simple ocean’, /ourna/ of Geophysical Research, 92, 1987, pages 13315-13343; Mitchell et al, op tit, reference 29. R. D. Cess et a/, ‘Intercomparison and interpretation of climate feedback processes in 19 atmospheric general circulation models’, /our& of Geophysical Research, 95, 1990, pages 16001-16615. M. Engardt and H. Rodhe, ‘A comparison between patterns of temperature trends and sulfate aerosol pollution’, Geophysical Research Letters, 20, 1993, pages 117-120; IPCC, Climate Change 7992, op tit, reference 4; all works cited in reference 16. J. Langner et al,“Anthropogenic influence on the distribution of tropospheric sulphate aerosol’, Nature, 359, 7992, pages 712-715; Kaufman and Fraser, op tit, reference 17. R. J. Charlson et a/, ‘Oceanic phytoplankton, atmospheric sulfur, cloud albedo and climate’, Nature, 326, 1987, pages 655661; T. L. Anderson and R. J. Charlson, ‘Ceophysiology of natural marine sulfate aerosols’, Atmospheric Environment, ZSA, 1991, pages 244552447. T. M. L. Wigley and S. C. B. Raper, ‘Implications for climate and sea level of revised IPCC emission scenarios’, Nature, 357, 1992, pages 293-300.

FUTURES

October 1993

862

44.

45. 46.

47. 48.

49. 50. 51. 52. 53. 54.

55. 56. 57. 58. 59. 60.

61

62. 63.

64.

65. 66. 67. 68.

69.

Review

and impacts

of climate

change

uncertainties

E. S. Sarachik and R. H. Gammon, The Role of the Ocean in the NOAA Program ‘Climate and Global Change’, (NOAA Climate and Global Change Program Special Report No 1, produced by the University Corporation for Atmospheric Research, Boulder, CO, 1989). Chahine, op cif, reference 34; IPCC, Climate Change, op tit, reference 4. K. Bryan, ‘Ocean thermal transients: a program of data analysis, modeling, and monitoring of the Atlantic Ocean thermohaline circulation’, in Manowitz (editor), op tit, reference 30, pages 633 70. S. Manabe et a/, ‘Transient responses of a coupled ocean-atmosphere model to gradual changes of atmospheric CO,. Part 1: annual mean response’, /ourna/of ?/imate, 4, 1991,-pages 785-878. W. M. Washington and C. A. Meehl, ‘Climate sensitivity due to increased CO,: experiments with a coupled atmosphere and ocean general circulation model’, Climate Dynamics, 4, 1989, pages l-38. A. J. Semtner, Jr and R. M. Chervin, ‘Ocean general circulation from a global eddy-resolving model’, Journal of Geophysical Research, 97, 1992, pages 549335550. W. D. Hibler, Ill, ‘Sea ice response to global climatic change’, in Manowitz (editor), op tit, reference 30, pages 1255150. R. D. Cess et a/, ‘Interpretation of snow-climate feedback as produced by 17 general circulation models’, Science, 253, 1991, pages 888-892. J. Cohen and D. Rind, ‘The effect of snow cover on the climate’, /ourna/ of Climate, 4, 1991, pages 689-706. G. Molnar and W. C. Wang, ‘Effects of cloud optical property feedbacks on the greenhouse warming’, Journal of Climate, 5, 1992, pages 814-821. P. C. D. Milly, ‘Potential evaporation and soil moisture in general circulation models’, /ourna/ of Climate, 5, 1992, pages 209-226; C. Thomas and A. Henderson-Sellers, ‘An evaluation of proposed representations of subgrid hydrologic processes in climate models’, /ourna/ of C/imate, 4, 1991, pages 898-910. C. B. Bonan, D. Pollard and 5. L. Thompson, ‘Effects of boreal forest vegetation on global climate’, Nature, 359, 1992, pages 716718. Washington and Meehl, op tit, reference 48. M. F. Meier, ‘Reduced rise in sea level’, Nature, 343, 1990, pages 1155116; IPCC, Climate Change 1992, op tit, reference 4; Wigley and Raper, op tit, reference 43. Crotch and MacCracken, op tit, reference 25; Cutowski et al, op tit, reference 25. Randall et al, op tit, reference 35. J. D. Mahlman, ‘Session on understanding climate change’, /oint Climate Project to Address Decision Makers’ Uncertainties (Washington, DC, Science and Policy Associates, Inc, 19921, page 5. W. L. Gates, ‘AMIP: the Atmospheric Model Intercomparison Project’, Bulletin of the American MeterologicalSociety, 73, 1992, pages 1962-1970; Ardanuy et a/, op tit, reference 21; Chahine, op tit, reference 34. M. I. Hoffert and C. Covey, ‘Deriving global climate sensitivity from paleoclimate reconstructions’, Nature, 360, 1992, pages 573-576. E. J. Barron, ‘Eocene equator-to-pole surface temperatures: a significant climate problem?‘, Paleoceanography, 2,1987, pages 729-739; W. J. Dansgaard et a/, ‘The abrupt termination of the Younger Dryas climate event’, Nature, 339, 1989, page 532; S. J. Johnson et a/, ‘Irregular glacial rnterstadials recorded in a new greenhouse ice core’, Nature, 359, 1992, pages 311-313. W. S. Broecker, ‘Unpleasant surprises in the greenhouse?‘, Nature, 328, 1987, page 123; T. F. Stocker, D. G. Wright and L. A. Mysak, ‘A zonJly averaged, coupled ocean-atmosphere model for oaleoclimate studies’, lournal of Climate, 5, 1992, oases 773-797; S. T. Lehman and 1. D. Keigwin, ‘Sudden changes in North Atlantic circulation &r&g the last deglaciation’, Nature, 356, 1992, pages 757-762. M. R. Rampino and S. Self, ‘Volcanic winter and accelerated alaciatron following ,_ the Toba supereruption’,‘Nature, 359, 1992, pages 50-52. L. W. Alvarez, W. Alvarez, F. Asaro and H. V. Michel, ‘Extraterrestrial cause for the CretaceousTertiary extinction’, Science, 208, 1980, pages 1095-1108. C. Lonus et a/, ‘The ice-core record: climate sensitivity and future greenhouse warming’, Nature, 347, 1990, pages 139-145. I. J. Winograd et a/, ‘Continuous SOO,OOO-yearclimate record from vein calcite in Devils Hole, Nevada’, Science, 258, 1992, pages 255-260; H. Gallee et a/, ‘Simulation of the last glacial cycle by a coupled, sectorially averaged climate-ice sheet model. 2. Response to insolation and CO, variations’, Journal of Geophysical Research, 97, 1992, pages 15713-15740; Stocker et a/, op tit, reference 64. D. R. MacAyeal, ‘Irregular oscillations of the West Antarctic ice sheet’, Nature, 359, 1992, pages 29-32; P. J. Barrett et a/, Ceochronological evidence supporting Antarctic deglaciation three million years ago’, Nature, 359, 1992, pages 816818.

FUTURES October 1993

Review

70.

71. 72. 73.

74.

and impacts

of climate

change

uncertainties

863

W. C. Clark, ‘Visions of the 21st century: conventional wisdom and other surprises in the global interactions of population, technology, and environment’, in K. Newton, T. Schweitzer and I. P. Voyer (editors), Perspective 2000(Ottawa, Canadian Government Publishing Center, Catalogue no EC22-167-1990E, 19901, pages 7-32; H. Brooks, ‘Generic issues regarding climate change’, Joint Climate Project to Address Decision Makers’ Uncertainties, Report of Findings (Washington, DC, Science and Policy Associates, Inc, 1992); J. Austin, N. Butchart and K. P. Shine, ‘Possrbility of an Arctic ozone hole in a doubled-CO, climate’, Nature, 360, 1992, pages 221-225. Clark, op tit, reference 70. NAS, op tit, reference 3. Ibid; W. D. Nordhaus, ‘To slow or not to slow: the economics of the greenhouse effect’, The Economic/oumal, 701, 1991, pages 92&937; W. D. Nordhaus, ‘An optimal transition path for controlling greenhouse gases’, Science, 258, 1992, pages 1315-1319. E. S. Rubin et al, ‘Realistic mitigation options for global warming’, Science, 257, 1992, pages 14& 266.

FUTURES October

1993