Transcript Climate change: How do we tell if it`s really happening?

Can we really predict future
climate change?
Ellie Highwood
Department of Meteorology, University of Reading
BA Thames Valley
February2004
Outline
• Before we start…
• Are we qualified to predict future
climate?
• The science of climate prediction
• Future predictions and uncertainty
• Ensemble forecasting (and how
you can help!)
• Global warming vs ice age theory
BA Thames Valley
February2004
Is climate change responsible?
• Record hot temperatures in Europe
during 2003?
• Flooding and climate change?
BA Thames Valley
February2004
Scorchio! Summer 2003
Record summer temperatures
happened in Kent on 10
August 2003…
And, summer 2003 was very
warm across Europe
compared to historical
temperatures
BA Thames Valley
February2004
Floods… e.g. Abingdon, January
2003
Photo by Guy Shepherd
BA Thames Valley
February2004
But is it global warming?
“Climate” is very variable, day to day, year to
year and place to place, and some of these
variations are due to internal chaotic
interactions between components of the
climate system -
“Climate is what we expect, weather is what we
get “ (Apologies to Ed Lorenz, 1976)
So, it’s not easy to say if any one event, such as
a hot spell or a flood is due to global warming,
But…
BA Thames Valley
February2004
“Record hot” events are more
likely in a generally warmer world
BA Thames Valley
February2004
Flooding and climate change?
• Little change in flood occurrence, although
high volume flows have lasted longer in the
past 30-50 years (Tyndall centre).
• Increased frequency in “intense” events in
mid and high latitude areas (IPCC) UK:
increase in frequency of intense rainfall
events in the winter and a decrease in the
summer (Tim Osborn, CRU)
• But, flood occurrence isn’t just about
rainfall – catchment changes and channel
changes are also important. Impact vs.
occurrence?
BA Thames Valley
February2004
What is climate anyway?
• Expected values
of temperature
and rainfall?
• Includes seasonal
cycles, year-toyear and decade
to decade
variations, and
place-to-place
variations
BA Thames Valley
February2004
What causes “variations” in
climate?
Internal chaotic interactions
e.g. El Nino
+
Natural but external forcing
of the climate system
(e.g. changes in the Sun, volcanic eruptions)
This “natural variability” is here to stay!
But then we add Human activities…
BA Thames Valley
February2004
Are we qualified to predict?
• Do we know enough about the
fundamental processes
determining the Earth’s climate?
• Do we understand (and can we
reproduce) any changes that have
occurred in the past?
• Are we capable of building models
to simulate climate, and of being
confident in their results?
BA Thames Valley
February2004
Fundamental processes
• Greenhouse
effect
• Clouds
• Oceanatmosphere
interactions
• Land surface
+ vegetation
+ hydrological
cycle
• Snow and ice
• Humans?
BA Thames Valley
February2004
THE GREENHOUSE EFFECT: identified in 1896!
..most escapes to outer space
and cools the earth...
SUN
…but some IR is trapped by
some gases in the air, thus
reducing the cooling….
Sunlight
passes
through the
atmosphere..
..and warms the earth.
BA Thames Valley
February2004
Arrhenius 1896
Infra-red radiation
is given off by the earth...
Major Players in the Greenhouse Effect
Water Vapour:
The natural greenhouse gas which makes our planet habitable
Human influences on the greenhouse effect
climate change
Carbon dioxide Fossil fuels, deforestation
100 years
Methane
10 years
Agriculture, natural gas
Other gases (nitrous oxide, CFCs, ground-level ozone..) 2 weeks
Aerosols
BA Thames Valley
February2004
Power generation, transport
2 weeks
CLIMATE CONFUSION
•The greenhouse effect is a natural phenomenon
which makes life on Earth possible
•The enhanced greenhouse effect is expected to lead
to global warming
•Detecting a change in climate is not the same as
attributing it to a particular cause (e.g. human activity)
BA Thames Valley
February2004
Do we understand past changes?
There is evidence that the Earth has been very
warm in the past 10 years...
BA Thames Valley
February2004
Although it’s really about global change… some
places have seen very large warming, other
regions have actually cooled.
BA Thames Valley
February2004
The evidence for global warming is
mounting!
INCREASE
• Length of freeze free
season
• Length of growing
season
• More frequent heat
waves
• Wetter winters/Drier
summers
• Sea level rise
BA Thames Valley
February2004
DECREASE
• Extent of mountain
glaciers
• Sea-ice amounts
and thickness
• Fewer Frosts
Climate change is not
unprecedented
The earth has been
capable of rapid
transitions in the
past.
But …..
BA Thames Valley
February2004
The northern hemisphere is probably the warmest it’s
been for at least 1000 years…
Mann et al., Science 1999 (Northern Hemisphere only)
BA Thames Valley
February2004
… the rise in temperature is strongly correlated
with a rise in the emission of certain gases into
the atmosphere by human activities….
Concentrations of these
gases are at record levels
in the atmosphere
BA Thames Valley
February2004
… And the present
trends in CO2 and
methane are
outside those
experienced on
earth for at least
the last 400,000
years…..
BA Thames Valley
February2004
More greenhouse gases leads to more
trapping of IR and a warmer surface
(As has been observed)
SUN
Roughly 2 Wm-2
more energy is
absorbed by the
Earth than it emits
Sunlight
passes
through the
atmosphere..
..and warms the earth.
BA Thames Valley
February2004
CO2
CO2
CO2
Infra-red radiation
is given off by the earth...
Arrhenius 1896
Image: BMRC
Circumstantial evidence?
Eventually we have to use
computer models, to understand,
and to predict
The Science of Climate Prediction
Computer models are our laboratory.
BA Thames Valley
February2004
What is a climate model anyway?
• Huge computer codes based on
fundamental mathematical equations of
motion, thermodynamics and radiative
transfer.
• These govern:
– Flow of air and water - winds in the
atmosphere, currents in the ocean.
– Exchange of heat between the atmosphere and
the earth’s surface
– Release of latent heat by condensation during
the formation of clouds and raindrops
– Absorption of sunshine and emission of
thermal radiation
BA Thames Valley
February2004
The Met Office
The earth is represented
by a grid of squares,
typically of length 250 km,
and by a stack of layers.
This gives us a 3-D picture
of the circulation of the
atmosphere and oceans
BA Thames Valley
February2004
The earth system is also very complex…..
Climate models must address the possible interactions and
feedbacks between the various components. And they must
consider human behaviour and our response to climate change.
BA Thames Valley
February2004
Are our models
qualified to predict?
First we have to make sure
our models are getting the
present climate about right
Figures from the
IPCC 2001 report
BA Thames Valley
February2004
We test different combinations of potential causes
against observations…
NATURAL ONLY
NATURAL + MAN-MADE
BA Thames Valley
February2004
MAN - MADE ONLY
“Most of the observed warming
over the last 50 years is likely to
have been due to the increase in
greenhouse gas concentration”
(IPCC 2001)
And look for “fingerprints” in
data and model results
BA Thames Valley
February2004
We finally make predictions …
How quickly the climate will change in the future
depends on:
• How much greenhouse gas emissions grow
–depends on population growth, energy use, new
technologies, etc (“scenario” uncertainty)
• How sensitive the climate system is to
emissions
–how clouds, ice, oceans etc respond to the extra
heating (“structural” uncertainty in climate
models and impact models)
Uncertainty in the predictions
BA Thames Valley
February2004
Climate changes over the next few decades are
predicted to be much larger than we have seen
so far, but there is quite a range of
“uncertainty” in the predictions…
By 2100:
•Increase in global
mean temperature
of 1.4°C - 5.8°C.
•Sea level rise by
a further 0.09m –
0.88m.
BA Thames Valley
February2004
IPCC (2001)
We can also produce very detailed predictions of
climate change, with little idea of how reliable they
might be.
2080
temperature
change (K)
2080
precipitation
change (%)
Source: Mat Collins, Hadley Centre
We need to be able to judge the probability of a
particular change, and to quantify the likely
uncertainty in our predictions.
BA Thames Valley
February2004
Sources of uncertainty in climate
prediction
• Uncertainties in models (“physics”
ensembles)
• uncertainties in emissions (“boundary
condition” ensembles)
• Uncertainties in initial conditions
(very important for Weather forecasts)
“Ensemble” prediction: run many
simulations with small changes to
the model setup each time, giving
a probablistic forecast
BA Thames Valley
February2004
Why is there model uncertainty?
• Many simplifications and assumptions
must be made to make useable models of
the climate system
• There are often different ways of doing
this and sometimes there isn’t a “best”
method. Some of these are based on
observations – which have an inherent
uncertainty
• These variations leads to differences in
the feedbacks in models and differences
in the future predictions
BA Thames Valley
February2004
You can help us produce the forecast!
Be part of a huge ensemble
experiment using spare processor
capacity of millions of
home/work/school PCs!
Run your own unique version of the
model, helping us to understand the
causes of uncertainty in our climate
predictions.
http://www.climateprediction.net
BA Thames Valley
February2004
Standard Visualisation Package
What will you see?
www.climateprediction.net
BA Thames Valley
February2004
What will we end
up with?
BA Thames Valley
February2004
What about the “UK ice age”
theory?
“Shut down” of thermohaline
circulation leads to cooling of UK
BA Thames Valley
February2004
Is it likely?
IPCC 2001
BA Thames Valley
February2004
So far, no
model has
shown total
shutdown
before 2100,
but climate
change may
make it more
likely in
successive
centuries
From Vellinga and Wood, 2002
Temperature response 20
years after THC collapse
If it did shut
down, it
would
produce a
cooling
tendency
But, if you include forcing from increases in
greenhouse gases in the model up until
2100, EU temperatures still warm overall
BA Thames Valley
February2004
Communicating Uncertainty
“Uncertainty” conveys a lack of
knowledge, lack of understanding or
disagreement between scientists.
But, as we’ve seen…
Our predictions can be uncertain despite
actually knowing a lot about the climate
system and climate change
BA Thames Valley
February2004
Communicating uncertainty?
• Often it is the least certain (extreme
weather) or least likely (thermo-haline
shutdown) phenomena that are used
as evidence of climate change or
predictions of the future
• Including probabalities will help balance
the discussion. As long as we can
communicate the meaning adequately
– To customers?
– To the public?
This is not just a climate change issue!
BA Thames Valley
February2004
“ If we begin with certainties, we
shall end in doubts, but if we begin
with doubts, and are patient in
them, we shall end in certainties”
Francis Bacon
“ Real knowledge is to know the
extent of one’s ignorance”
Confucius
BA Thames Valley
February2004
BA Thames Valley
February2004
Climateprediction.net design

Expt 1: UM with simple, thermodynamic ocean. (HadSM3)
 Aim: To identify parameter combinations which have little
effect on the mean climate but a large effect on climate
sensitivity.
Double CO2
15 yr, 2 x CO2
Calibration
15 yr spin-up

Derived fluxes
15 yr, base case CO2
Diagnostics from final
8 yrs.
Control
Expt 2: Fully coupled model. (HadCM3 - the “coupled” model)
 Distribute pre-packaged simulations of 1950 -2050.
 Downweight or eliminate runs which compare badly with
observations.
 Re-distribute the surviving versions to simulate 2000-2050.
BA Thames Valley
February2004