Transcript Inaugural Lecture

SAGES
Scottish Alliance for Geoscience, Environment & Society
CESD
Climate Change: Observing and
Simulating the Past; Predicting
the Future
Simon Tett, Chair of Earth System Dynamics & Modelling
With thanks to Gabi Hegerl, Ben Santer, Phil Jones, Keith Briffa, Peter Thorne,
Philip Brohan, Nick Rayner, John Kennedy, Peter Stott, Myles Allen, Gareth
Jones, John Mitchell, Geoff Jenkins, Chris Folland, David Parker, Jonathan
Gregory, Bob Harwood, Richard Kenway and Claire Jones
1
What are we trying to understand?
CESD
How might the
earth system
evolve in the
future?
How and why
did it evolve in
the past?
Image created by
Reto Stockli with the
help of Alan Nelson,
under the leadership 2
of Fritz Hasler
What are we modelling?
CESD
From Space Science and Engineering Center, University of Wisconsin-Madison
3
Overview
CESD
•
•
•
•
Basic physics
Modelling the climate system
Observations of climate change
Using climate models
–
–
–
–
Understanding 20th century climate change
Role of natural drivers in natural variability
Predictions of future change
Importance of external drivers
• Concluding thoughts
4
CESD
Radiation – the driver of the climate
system
• Key ideas
– Lots of incoming shortwave radiation (“Visible”) from
sun
– Same total energy going out from Earth but peaks in
Infra-red. (“Heat”)
– Surface is warmer than you’d expect from simple
radiation budget.
• The bit of the climate system that radiates energy to space is
high up (where it is cooler).
• Atmosphere cools with height
• So surface is warmer the “greenhouse” effect
• Changing the height of the atmosphere where energy gets to
space will then affect the surface temperature
5
Lapse Rate
CESD
Tropical Pacific lapse rate
Temperature falls
with height
From
http://tamino.wordpress.c
om/
6
Feedbacks
CESD
• Act to amplify (or decrease) warming from changes in CO2, other
greenhouse gases and other climate drivers.
– Blackbody – warmer planet emits more radiation and so cools.
(Negative feedback)
– Water vapour – warmer atmosphere can store more water vapour.
Water vapour absorbs “heat” radiation so is a Greenhouse gas.
• Most important in the upper troposphere
• Warmer world will have more moisture in the atmosphere and so will trap
more heat. Positive feedback.
– Clouds
• Positive feedback – “trap” “heat” radiation. Particularly true for high clouds
• Negative feedback – reflect back solar radiation. Particularly true for low
clouds
– Ice/Albedo feedback.
• Ice is white and reflects lots of solar energy back to space.
• Melt ice and more solar radiation absorbed which in turn warms the climate..
7
Snow/Ice Feedback
CESD
Summer
Winter
Image courtesy
NASA/GSFC/JPL, MISR Team.
8
See
http://visibleearth.nasa.gov/
Climate Modelling
CESD
• Atmospheric modelling has long history – first
attempts, using computers, made in 1950’s.
• General Circulation Models (GCM’s) developed
from numerical weather prediction models
– Take physical laws and apply them to atmospheric
and oceanic motions.
– Key is that GCM’s are built bottom up.
– Interested in “Emergent Phenomenon”, such as
statistics of data, rather than detailed evolution.
• Other approaches but not covered in this lecture.
9
Modelling the Climate System
CESD
Main
Message:
Lots of
things
going on!
10
Karl and Trenberth 2003
General Circulation Models
CESD
3-D model of the circulation of the atmosphere and
ocean
Fundamental equations:
• Conservation of momentum
• Conservation of mass
• Conservation of energy
• Equation of state
11
Parameterized Processes
CESD
•Unresolved motions and processes affect the large
scale flow so their effect needs to be parameterized.
12
Slingo
From Kevin E. Trenberth, NCAR
What are we trying to parameterize?
CESD
What is there…
How we
parameterise
13
Numerical Modelling
CESD
Cray Y-MP ~ 1990
L. F. Richardson circa
1920
Since the 1960’s super-computer
computational power increased by
factor of 16 every decade. Over
my career increased 200-300 fold
HECToR –
2008
14
Observing Climate Change
CESD
• Observing system not stable
• Climate changes slowly compared to observing
system.
15
Global Mean Temperature
CESD
16
From Brohan et al, 2006
The longer perspective
CESD
Recent warming unprecedented
17
CESD
Changes in Upper Ocean
temperatures
The upper ocean is
warming at, when looked
one way, at roughly the
same rate everywhere
From Palmer et al, 2007
18
CESD
Changes in the free atmosphere:
Large Observational Uncertainty
Left plot shows cooling in the
tropical atmosphere. Contradicts
climate models which predict
largest warming in the tropics.
From Thorne et al, 2005 & Titchner et al, 2008
Right hand plot shows range of possible
temperature changes in tropical free
atmosphere due to uncertainties in
observations. Sometimes models are
more reliable than observations!
19
Model Applications
CESD
• Understanding 20th century climate
change
• The role of natural and human drivers in
climate variability
• Future scenarios
• Summary: external drivers important in
explaining observed climate variability and
future climate change
20
What might cause observed change?
CESD
21
Internal variability – variability
CESD generated within the climate system
Recent tropical
Pacific ocean
temperatures
from IRI
The North Atlantic Oscillation
source
22
http://www.ldeo.columbia.edu/NAO by
Martin Visbeck
Natural Factors that might effect climate:
CESD
Volcanoes
Volcanic Aerosol depth
0.2
0
1850
2000
Large tropical volcanoes inject sulphur dioxide into the
Stratosphere where it stays for 2-3 years. Effect is to make
23
an aerosol that scatters light and so cools climate.
Natural Factors that might effect climate:
CESD
Solar Irradiance
200
0
Sunspot Number
1700
2000
Solar activity (sunspots etc) & irradiance changes with 11year solar cycle. There are long term changes in solar activity
– the Maunder Minimum being one example. Converting this
to changes in solar irradiance can be done though very
uncertain. “Sun-like” starts which show activity variations
have been used to estimate irradiance changes. Recent work
(astronomical) and modelling (Lean et al) suggests there may
24
be no significant long term variation in solar irradiance.
CESD
Human Factors that might affect
climate: Aerosols
Thanks to Met Office
25
CESD
380
Human Factors that might affect
climate: Greenhouse gases
CO2 MMR*106
Mauna Loa
Observatory
360
1800
1400
320
1200
300
1000
1900
Year
2000
Flasks
1600
340
280
260Ice cores
1700
1800
CH4 MMR*109
Ice cores
800
600
1700
1800
1900
Year
2000
Greenhouse gas
concentrations have
changed over the last
century. Their effect is
to decrease the
transmission of heat
radiation by the
atmosphere. So
should warm climate.
26
CESD
Understanding and Attributing Climate
Change in the 4th Assessment
Globe, Land, Ocean and individual continents all likely
show human induced warming. Warming effect of
greenhouses gases likely offset by other human and
27
natural drivers
Modelling the last 500 years
CESD
• How important are external drivers
compared to internal climate variability?
• Simulation with fixed drivers – “internal”
variability alone.
• Simulation with only natural drivers
– Sun & Volcanic eruptions
• Simulation with human and natural drivers
– Natural + changes in greenhouse gases,
aerosols, and land-surface properties
28
Natural Drivers
CESD
Annual: Slow
changes with large
negative forcings
(from volcanoes)
25-year Gaussian filter. Solar and
Volcanic forcing as important as one
another. “Maunder Minimum” includes
volcanic contribution. Tambora is
largest eruption of last 500 years. Late
20th century is also a volcanically active
period.
Solar
Volcanic
29
Effect of natural drivers
CESD
SH has less
variability
(as more
ocean)
than does
NH
Both hemispheres change together as does the land & ocean though there are some
differences. Natural variability is about ±0.3K compared to “internal” variability
30
of ±0.1K. So Natural forcings are an important driver of global-scale temperature
variability
Naturally driven variability
CESD
Effect of natural drivers is to increase
variability in the tropics
31
Adding human drivers
CESD
Aerosols
and
Greenhouse volcanoes
gases
offset some
GHG and
solar
warming
Total Human
Sun
Tot. Natural
Volcanoes
Aerosols
32
CESD
Temperature Changes with human
drivers included
Effect of human
drivers is to warm
climate so that it
warms outside
envelope of natural
variability by midlate 19th century in
southern
hemisphere land
and by mid 20th
century in northern
hemisphere
33
Effect of human drivers of climate
CESD
Shows impact of human drivers on zonal-average
temperature. Tropics warm first and warming is
significant by mid 19th century. Northern hemisphere
warming delayed by aerosol cooling in simulation 34
Predicting the Future
CESD
Material in this section from IPCC 4th
assessment report.
Results based on multi-model archive.
Typically show average across all model
simulations with uncertainties from range
Scenarios used to drive models. Selfconsistent atmospheric concentrations of
CO2 and other greenhouse gases. Based
on different human development paths
35
Projections of Future Changes in Climate
CESD
Best estimate for
low scenario (B1)
is 1.8°C (likely
range is 1.1°C to
2.9°C), and for
high scenario
(A1FI) is 4.0°C
(likely range is
2.4°C to 6.4°C).
36
Projections of Future Changes in Climate
CESD
Projected warming
in 21st century
expected to be
greatest over land
and at most high
northern latitudes
and least over the
Southern Ocean
and parts of the
North Atlantic
Ocean
37
Projections of Future Changes in Climate
CESD
Precipitation increases very likely in high latitudes
Decreases likely in most subtropical land regions
38
CESD
Is climate changing faster than we
thought it would?
• Lot of argument has been about reality of
climate change
– Are observations good?
– Is the sun responsible for warming?
– Feedbacks are weak so that future warming not likely
to be a great threat?
• General consensus (see 4th Assessment report)
is that climate is changing, likely due to human
influences and agreement between different
models as to likely warming.
• But could models be underestimating future
climate change?
39
What does the future hold?
CESD
Climate Sensitivity –
measure of feedbacks.
“Long tail” suggests there
may be strong feedbacks.
Ensemble of “perturbed physics” models
showing large uncertainty range of
future warming. Which are right?
40
Sea-ice (its ½ what is should be)
CESD
Is this unexpected? Are we missing
something fundamental in our understanding
of the Earth system?
41
Sea-Ice
CESD
NASA/GODDARD SPACE FLIGHT CENTER SCIENTIFIC
VISUALIZATION STUDIO; (DATA) ROB GERSTON, GSFC
42
Circulation change important for regional
changes
CESD
Human influence
detected on Sea Level
Pressure BUT magnitude
under-simulated in
Northern Hemisphere
(e.g. Gillett et al., 2005)
Observations
Model mean
NH
SH
These problems will affect
regional model
simulations and regional
predictions
From Gabi Hegerl
Multi-model archive
43
UK Extreme events
CESD
Tewkesbury 2007Photograph: Daniel
Berehulak/GettyImages
Met Office figures show
that May to July in the
England and Wales
Precipitation is the
wettest in a record that
began in 1766.
We must learn from the events of recent days. These rains
were unprecedented, but it would be wrong to suppose that
such an event could never happen again…. (Hazel Blears,
House of Commons, July 2007)
44
Is it human induced climate change or natural variability?
UK changes
CESD
Precipitation (blue) and
temperature (red) for
1931-80 and 1981-date
(dashed)
High summer drying
and warming. Rest
warming and
45
moistening
Change over last century
CESD
Natural
All
Obs
Does model underestimate high
summer changes?
Observations
distinct from
zero,
consistent
with all and
inconsistent
with natural.
Implies
human
influence on
46
UK climate.
Summary & Conclusions
CESD
• Basic understanding of the climate system explains
greenhouse effect and why would expect warming in
response to changing atmospheric composition
• Details of response come from feedbacks
• Climate models are built “bottom up” not top down.
Uncertainties arise from need to parameterize
unresolved phenomenon
• Interested in the emergent behaviour which is not easily
predictable from basic physics in model.
• Instrumental observations of surface temperature back
to mid 19th century
• This, and other observations, show clear evidence of
warming and climate change.
47
Summary & Conclusions
CESD
• Using models and observations to establish that:
– 20th century climate change is likely to be human driven with
greenhouse gas warming being offset by natural and other
human drivers
– That external drivers are an important driver of natural climate
variability
– That humans might have affected 19th century tropical climate.
– Climate change has already happened and will continue to
happen regardless of what we do.
– But will be large if emissions are not reduced.
• Models may be underestimating changes to come
particularly those related to changes in atmospheric
circulation.
– This has important consequences for regional (i.e. UK) climate
change.
48
CESD
The End!
Thanks for listening
Any Questions?
49
Extra Material
CESD
50
CESD
(Natural) Variability in Extreme events
can be large
20’s
60’s
90’s
From Allan et al, 2008
51
Trends since 1800
CESD
18002006
July-Aug
CET
18002006
Oct-May
CET
18002006
July-Aug
EWP
Obs
0.42
K/Cent
0.50
K/Cent
-8.2 %
/Cent
Model
0.35
K/Cent
+/- 0.2
0.41K/Ce 2.6 %
nt
/Cent +/9%
+/- 0.3
Biggest
difference
is +/- 0.3
18002006
Oct-May
EWP
6.3%
/Cent
1.7%
/Cent +/4%
Biggest
difference
is 10%
Model not
capturing
drying
trend
52
Digitisation as a source of new data
CESD
• Available observed weather
data are limited before 1950 and
almost non-existent before 1850.
• Many more observations exist,
in logbooks, reports and other
paper records (mostly in the UK).
If we digitised them we could
improve the climate record and
extend it back to 1800.
• Hadley Centre digitised
observations from Royal Navy
Ships logbooks for WW2. These
give a much-improved picture of
1940s climate.
53
1998 & 2007
CESD
54
Another feedback: The Carbon cycle
CESD
From Friedlingstein et al, 2006. Plots shows additional CO2 from
feedbacks between climate change and carbon cycle. Values vary
between 25 and 225 ppm at 2100 mostly due to land-carbon cycle
feedbacks.
55
Relative Contributions
CESD
Greenhouse
gases
Natural
Other
human
56
Assessing Recent Change
CESD
Observed trend (K/decade) marked with X where outside maximum absolute 50-year
trend from Natural. + where outside maximum trend. Recent changes are outside
simulated natural variability over large parts of the world. Suggests that natural
systems are already being affected by climate change
57
CESD
Zonal 20th century
precipitation change
Observed (black) and simulated 1901-1998 precipitation trends
Fig 9.18b
Observations
Multi-model mean
Model range
Thin solid line  model’s all
forcing response detected in obs
Changes in rainfall over NH underestimated by
58
models?
Figure from IPCC WGI Ch9 (Hegerl, Zwiers et al)
Ranking
CESD
From John
Kennedy & the
Met Office
59
Radiation
CESD
60
Timeseries of UK records
CESD
61
Actual heat waves
CESD
62
CESD
Future changes in the Hydrological
Cycle
63