1. Stephen H. Schneider*
Department of Biological Sciences
and
Woods Institute for the Environment
Stanford University, California, USA.
“Key Vulnerabilities” and the
Risks of Climate Change?
Michigan State University
Ides of March 2007
*[Website for more info: www.climatechange.net.]

2. The role of the scientific community
#1: Provide climate change scenarios
The IPCC’s Special Report on Emissions Scenarios (SRES) ­ 2000

3. What will be our future
emissions?
Higher
Lower
Source: Intergovernmental Panel on Climate Change

6. WG 1Approved AR 4 SPM:
In this Summary for Policymakers, the
following terms have been used to indicate
the assessed likelihood, using expert
judgment, of an outcome or a result:
Virtually certain > 99% probability of
occurrence, Extremely likely > 95%, Very
likely > 90%, Likely > 66%, More likely than
not > 50%, Unlikely < 33%, Very
unlikely < 10%, Extremely unlikely < 5%.

7. ???
Approved
uncertainties
language for
10­33%
likelihood:
“unlikely”

9. Pre­
Plenary
proposed

10. “Low”
“High”

11. From the Sports Pages: Type 1 Error Aversion
Denver is allowing 7.3 points per game.
Broncos defensive end Kenard Lang was
asked to predict a point total for the Colts
offense this Sunday:
"I‘m not predicting nothing. All I'm going to
predict is a good game and I'm hoping the
Broncos come out on top. If I make
predictions and it goes the opposite way,
then I'll look like a horse's fanny. And I ain't

12. From the Sports Pages: Type 1 Error Aversion
Denver is allowing 7.3 points per game.
Broncos defensive end Kenard Lang was
asked to predict a point total for the Colts
offense this Sunday:
"I'm not predicting nothing. All I'm going to
predict is a good game and I'm hoping the
Broncos come out on top. If I make
predictions and it goes the opposite way,
then I'll look like a horse's fanny. And I ain't

13. From the Sports Pages: Type 1 Error Aversion
Denver is allowing 7.3 points per game.
Broncos defensive end Kenard Lang was
asked to predict a point total for the Colts
offense this Sunday:
"I'm not predicting nothing. All I'm going to
predict is a good game and I'm hoping the
Broncos come out on top. If I make
predictions and it goes the opposite way,
then I'll look like a horse's fanny. And I ain't
trying to look like a horse's fanny right now."

14. “Type 1” versus “Type 2" errors and their
consequences
Decision Forecast Forecast
proves false proves true
Accept forecast— Type I Correct
policy response error decision
follows [Squandered
resources]
Reject or ignore Correct Type 2
forecast (e.g., “too Decision error
much” uncertainty)—
no policy response

15. “Type 1” versus “Type 2" errors and their
consequences
Decision Forecast Forecast
proves false proves true
Accept forecast— Type I Correct
policy response error decision
follows [Squandered
resources]
Reject or ignore Correct Type 2
forecast (e.g., “too Decision error
much” uncertainty)— [Unmitigated
damages]
no policy response

16. “Type 1” versus “Type 2" errors and their consequences
Decision Forecast proves Forecast proves
false true
Accept forecast—policy response Type I error Correct decision
follows
Reject or ignore forecast (e.g., “too Correct Decision Type 2 error
much” uncertainty)—no policy
response
*************************************************
Role of Scientists: Assess Risk (= Consequence X Probability of Occurrence)
as function of alternative policy choices ; confidence in the assessment of risks;
distribution of risks and benefits; traceable account of aggregations.
­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­
Role of Decision­makers: Negotiate acceptability of risks and policies that alter
risks; make policy choices; guide assessment process.

17. “Type 1” versus “Type 2" errors and their consequences
Decision Forecast proves Forecast proves
false true
Accept forecast—policy response Type I error Correct decision
follows
Reject or ignore forecast (e.g., “too Correct Decision Type 2 error
much” uncertainty)—no policy
response
*************************************************
Role of Scientists: Assess Risk (= Consequence X Probability of Occurrence)
as function of alternative policy choices ; confidence in the assessment of risks;
distribution of risks and benefits; traceable account of aggregations.
­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­­
Role of Decision­makers: Negotiate acceptability of risks and policies that alter
risks; make policy choices; guide assessment process.

18. Competing paradigms between science and policy
communities.
It is common in policy analysis to refer to an
incorrect forecast that was taken to be true as a
“type 1 error” and a decision to ignore an
uncertain forecast that turns out to be true as a
“type 2 error”. The prime paradigm within the
scientific community is to view the type 1 error
as the more egregious mistake, whereas within
the policy arena, the type 2 error is often more
concerning. Decision­makers often prefer to
hedge against a potentially damaging event
rather than wait for it to possibly happen.

25. Chapter 19 (draft, do not quote) identifies seven criteria for
assessing and defining key vulnerabilities:
• magnitude
• distribution
• timing
• persistence and reversibility
• likelihood and confidence
• potential for adaptation
• “importance” of the vulnerable system
No single metric can adequately aggregate the diversity of key
vulnerabilities, nor determine their ranking.

26. What will be our future
emissions?
Higher
Lower
Source: Intergovernmental Panel on Climate Change

30. Not nearly
stabilized at
2100
Nearly
stabilized at
2100

31. Emissions Scenarios
6
Overshoot to
5 stabilization
Radiative Forcing
4 600 ppm CO e
2
3 500 ppm CO e
2
2
Gradual increase to
stabilization
1
0
2000 2050 2100 2150 2200 2250
Year
(O’Neill and Oppenheimer, 2004)

32. DT
Source: Schneider and Mastrandrea, PNAS, Oct 2005

33. Exceedence of
DAI threshold:
dependence on
scenarios
Source: Schneider and Mastrandrea, PNAS, Oct 2005

34. The great “greenhouse
gamble”…(for 2100)
<1°C (4.1%; 1 in 24 odds)
1 to 1.5°C (11.4%; 1 in 9 odds)
1.5 to 2°C (20.6%; 1 in 5 odds)
2 to 2.5°C (22.5%; 1 in 4 odds)
2.5 to 3°C (16.8%; 1 in 6 odds)
3 to 4°C (16.2%; 1 in 6 odds)
4 to 5°C (4.6%; 1 in 22 odds)
>5°C (3.8%; 1 in 26 odds)
Source: MIT Joint Program on the Science and Policy of Climate Change

35. HOW CAN WE EXPRESS THE VALUE OF A
CLIMATE POLICY UNDER UNCERTAINTY?
Compared with What would we A NEW WHEEL
NO POLICY buy with STABILIZATION with lower odds
of CO2 at 550 ppm? of EXTREMES

36. HOW CAN WE EXPRESS THE VALUE OF A
CLIMATE POLICY UNDER UNCERTAINTY?
Compared with What would we A NEW WHEEL
NO POLICY buy with STABILIZATION with lower odds
of CO2 at 550 ppm? of EXTREMES

37. Risk = Probability x
Consequence
[What metrics of harm?]
­$/ton C avoided
­lives lost/ton C avoided
­species lost/ton C avoided
­increased inequity/ton C avoided*
­quality of life degraded/ton
*Perception that prime generators of the risks are not accepting
responsibility for their emissions or helping victims to adapt (e.g.,
OECD countries refusing to join in Kyoto Protocol) itself creates
risks.
[Source: “The Five Numeraires”, Schneider, Kuntz­Duriseti and Azar 2000]

40. Risk of catastrophic fires
(and other disturbances)
(and other disturbances)

42. Agriculture: The Wine Industry
• ‘Potentially devastating’ effect on
industry
•Water availability
•Temperature
•Storms

43. A u s t r a l i a n w i n e r e g i o n s
M e a n J a n u a r y T e m p e r a t u r e o f 2 3 C h i g h l i g h t e d
( m a x i m u m v a l u e ( $ / h e c t a r e ) )
Climate change scenario A1B
MJT23C2000 CSIRO Mk 3 model
Leanne Webb CSIRO and Melbourne University

44. A u s t r a l i a n w i n e r e g i o n s
M e a n J a n u a r y T e m p e r a t u r e o f 2 3 C h i g h l i g h t e d
( m a x i m u m v a l u e ( $ / h e c t a r e ) )
MJT23C2000 Climate change scenario A1B
MJT23C2030 CSIRO Mk 3 model
Leanne Webb CSIRO and Melbourne University

45. A u s t r a l i a n w i n e r e g i o n s
M e a n J a n u a r y T e m p e r a t u r e o f 2 3 C h i g h l i g h t e d
( m a x i m u m v a l u e ( $ / h e c t a r e ) )
MJT23C2000
MJT23C2030 Climate change scenario A1B
MJT23C2050 CSIRO Mk 3 model
Leanne Webb CSIRO and Melbourne University

48. The cost to stabilize the atmosphere
Global GDP
250
200
Trillion USD/yr
Bau
150
350 ppm
450 ppm
100
550 ppm
50
0
1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
Source: Azar & Schneider, 2002.

49. The cost to stabilise the atmosphere
Global GDP Delay time to 500% richer
per capita with tough
250
climate policy ~ 1­2 years
200
Trillion USD/yr
Bau
150
350 ppm
450 ppm
100
550 ppm
50
0
1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100
Year
Source: Azar & Schneider, 2002.

50. Questions?
Comments??

52. Risk = Probability x
Consequence
[What metrics of harm?]
­$/ton C avoided
­lives lost/ton C avoided
­species lost/ton C avoided
­increased inequity/ton C avoided*
­quality of life degraded/ton
*Perception that prime generators of the risks are not accepting
responsibility for their emissions or helping victims to adapt (e.g.,
two OECD countries refusing to join in Kyoto Protocol) itself
creates risks.
[Source: “The Five Numeraires”, Schneider, Kuntz­Duriseti and Azar 2000]

53. Munich Re:
“We need to stop this dangerous
experiment humankind is
conducting on the Earth’s
atmosphere.”

54. What does “dangerous” climate
change really mean?

55. Article 2 of the UN Framework Convention on Climate Change
(UNFCCC) states that: “The ultimate objective of this Convention
and any related legal instruments that the Conference of the
Parties may adopt is to achieve, in accordance with the relevant
provisions of the Convention, stabilization of greenhouse gas
concentrations in the atmosphere at a level that would prevent
dangerous anthropogenic interference with the climate system”.
The Framework Convention on Climate Change further suggests
that “Such a level should be achieved within a time frame
sufficient
· to allow ecosystems to adapt naturally to climate change,
· to ensure that food production is not threatened and
· to enable economic development to proceed in a sustainable
manner.”

56. Climate Uncertainty
• Inherent uncertainty in projections of future
climate
• Best guess à Range à PDFs
à à PDFs

57. Climate Uncertainty
Climate Uncertainty

58. Climate Uncertainty
0.04
Density 0.03
0.02
0.01
0
0 1 2 3 4 5
o
Temperature Change above 2000 ( C)

59. Climate Uncertainty
0.04
Temperature
probability density
0.03
Density
function for 2100
0.02
based on PDF for
climate sensitivity
0.01
0
0 1 2 3 4 5
o
Temperature Change above 2000 ( C)

61. The great “greenhouse
gamble”…
<1°C (4.1%; 1 in 24 odds)
1 to 1.5°C (11.4%; 1 in 9 odds)
1.5 to 2°C (20.6%; 1 in 5 odds)
2 to 2.5°C (22.5%; 1 in 4 odds)
2.5 to 3°C (16.8%; 1 in 6 odds)
3 to 4°C (16.2%; 1 in 6 odds)
4 to 5°C (4.6%; 1 in 22 odds)
>5°C (3.8%; 1 in 26 odds)
Source: MIT Joint Program on the Science and Policy of Climate Change

63. Governor of
California:
80% reduction in
emissions by 2050

64. Strategic Plan for SA

66. ‘Changes in the location of
Goyder’s line’ 2070
1
2
3
4
5
Quorn 6
Port Augusta
7
8
9
10
Adelaide
Goyder’s Line
Study Site
Mark Howden CSIRO sustainable ecosystems
Howden and Hayman – Greenhouse 2005

70. Governor of California:
80% reduction in emissions by 2050
Premier of South Australia:
60% reduction in emissions by 2050

71. Risk = Probability x
Consequence
[What metrics of harm?]
­$/ton C avoided
­lives lost/ton C avoided
­species lost/ton C avoided
­increased inequity/ton C avoided*
­quality of life degraded/ton
*Perception that prime generators of the risks are not accepting
responsibility for their emissions or helping victims to adapt (e.g.,
OECD countries refusing to join in Kyoto Protocol) itself creates
risks.
[Source: “The Five Numeraires”, Schneider, Kuntz­Duriseti and Azar 2000]