Anthropogenic Climate Change

1
Anthropogenic Climate Change

• Myles Allen
• Department of Physics, University of Oxford
• myles.allen_at_physics.ox.ac.uk

2
The evidence for human influence on climate and
its implications for the future

• The warming trend of 0.13-0.23oC per decade
  experienced since 1980 is largely a predictable,
  and predicted, response to the observed
  anthropogenic increase in greenhouse gases.
• Alternative explanations have failed either to
  predict or to account for this recent observed
  warming.
• Natural factors may well have played a larger,
  and underestimated, role on longer timescales.
• Without very large reductions in GHG emissions,
  warming will continue at or above the current
  rate.
• The impacts of climate change are beginning to be
  felt through changing risks of extreme weather.

3
Warming of the climate system is unequivocal
4
Models driven with natural factors alone simulate
a modest cooling over the past 50 years
Colours Simulations with natural influences alone
5
Models driven with anthropogenic and natural
factors are consistent with observed changes
Colours Simulations with human and natural
influences
6
Model-data agreement extends to other variables,
including tropospheric trends
Simulated and observed MSU 2LT trends 1979-99,
courtesy of Ben Santer, 2007
7
and regional scale temperature changes, provided
anthropogenic factors are included
Anthropogenic and natural factors Natural factors
alone
8
Simulated and observed changes in Arctic
September sea ice extent
9
But is the fit too good? Could the models and/or
drivers have been tuned to fit the data?
10
A genuine out-of-sample prediction
FAR 1st Assessment Report, 1990
11
C.f. another prediction made at the time
Lindzen (1992,1994) Climate sensitivity
0.30.2oC
12
And a third that global temperatures follow the
solar cycle length (SCL)
Friis-Christensen and Lassen, 1990
13
And a third that global temperatures follow the
solar cycle length (SCL)
Predicted values for smoothed SCL
Friis-Christensen and Lassen, 1990
14
And a third that global temperatures follow the
solar cycle length (SCL)
Including post-1990 data on SCL from Laut, 2004
15
All indicators of solar forcing of climate
indicate no trend since 1980
(a) Sunspot number (b) Open solar flux (c) GCR
flux (d) TSI (e) global temperature Lockwood
and Fröhlich, 2007
16
But there is evidence that some models
underestimate the response to solar variability
Combined forcing
17
But there is evidence that some models
underestimate the response to solar variability
Combined forcing, doubling solar response
18
A possible explanation GCR-cloud interactions
amplifying the response to solar forcing
Figure kindly provided by Henrik Svensmark, DNSC
19
Conclusions from attribution studies

• Most of the warming over the past 50 years is
  very likely due to the observed increase in
  greenhouse gas concentrations.
• Any underestimation of the response to solar
  forcing is a (possibly large) error in a
  relatively small number.
• Solar forcing may have played a proportionally
  larger role in the (much more uncertain) warming
  that has occurred since the Little Ice Age
• but this has no direct bearing on our
  understanding of the origins of current trends or
  on near-term predictions of anthropogenic
  warming.

20
Assessing what this means for the future the
process of climate projection
21
High consensus on response over coming decades
Distributions of global temperature change
relative to 1980-2000 mean under two SRES
emission scenarios
22
But who cares about global temperature?

• Lorenz definition of climate
• Climate is what you expect, weather is what you
  get.
• Updated for the 21st century
• Climate is what you affect, weather is what gets
  you.
• Most impacts of climate change over this century
  will be due to changing risks of extreme weather.
• In any given instance, we will not be able to say
  but for human influence, this would not have
  occurred.

23
The European heat-wave of 2003
European temperatures in early August 2003,
relative to 2001-2004 average
From NASAs MODIS - Moderate Resolution Imaging
Spectrometer, courtesy of Reto Stöckli, ETHZ
24
Impact of 2003 heat-wave on daily mortality in
Baden-Württemberg, Germany
2003 Summer Heat-wave
Influenza epidemic
Schär Jendritzky, 2004
25
Changing risks of European heatwaves (Stott et
al, 2004)
Return periods for European heat-waves
9x increase in risk
26
Anthropogenic contribution to the risk of the
2003 heat-wave
Threshold for civil liability
Range of uncertainty
27
Stratospheric aerosols affect weather as well
Orange red Regions affected by drought in the
year following the eruption of Mount Pinatubo
(Trenberth and Dai, 2007)
28
Conclusions from analysis of changes in extreme
weather risk

• Assessments of climate change impacts and costs
  focus on deterministic changes in mean climate.
• For most people alive today, the more important
  impacts will be through stochastic weather.
• Doubling the risk of an adverse event is
  typically used as a threshold for civil
  liability, and events may already be happening
  that pass this test.
• Tobacco-style litigation against fossil fuel
  producers is thought unlikely, but not because of
  problems with establishing causation.
• Potential liability will be a serious, probably
  fatal, impediment to active geo-engineering.

29
Europeans can adapt to more frequent heat-waves,
but other impacts are more challenging
Tuyuksu Glacier, Kazakhstan a vital water source
30
High level of consensus on the reponse to a given
emissions scenario
Climate response to the IS92a scenario as
predicted by 2001 IPCC models and by Patrick
Michaels, University of Virginia