Transcript Model

Present day climate modelling –
its status and challenges
Ulrich Cubasch
Institut für Meterologie
Freie Universität Berlin
sponsored by BMBF and EU
Lectures
• Present day climate modelling – its status
and challenges
• Development of a climate m
• odel
• Projection of future climate change
• Modelling past climate
Gliederung
•
•
•
•
•
•
•
Introduction
The forcings
Simulation of the Eemian
Simulating the climate of the last 1000 years
Scenario calculations
Summary
Outlook
The „Mann et al“- curve (hockey-stick)
Temperature-Reconstruction (treerings, corals, ice and sediment cores,
historical evidence) of the temperature of the northern hemisphere from the
year 1000 bis 1999 and instrumental temperature from 1902 to 1999
today
CO2
Temperature
CH4
The Vostock ice core
Scientific questions
• To what extend does a change in radiative forcing (sun, volcanoes,
greenhouse gases, aerosols) influence the climate?
– How does solar induced climate variability compare with
anthropogenic influences?
– How sensitive is the climate system?
– How will the climate of the future look like?
– Can a climate model simulate the historic climate variability?
• Does it confirm the reconstructions?
• Can it be used to substitute and/or assimilate proxy-data?
– Can the climate model simulate paleo-climatic conditions like ice
ages and warm periods as well as the transition between warm and
cold stadials?
• Does it confirm the climate archives?
• Can it be used to substitute and/or assimilate proxy-data?
The forcings
External Forcing
The climate system
Orbital parameter
excentricity
precession
excentricity
precession
obliquity
tilt of the earth axis
obliquity
~41 ky
excentricity
~100 ky
~ 23 und 19
ky
precession
Change of solar input
by orbital parameters
Solar variability
1978
1999
Composite solar flux measured by satellites
Yearly averaged solar sunspot number
The solar forcing anomaly
reconstructed by 3 different methods
Constituents of the atmosphere
Modelling
…with Earth?
climate change
experiments
or
…in the computer?
The physical laws
It is assumed that the atmosphere follows
physical laws:
• the Newtonian equations of motion (for the
wind fields)
• the laws of thermodynamics
• ideal gas equation
• the continuity equation for mass
The grid representation of a 3-D model
Model Resolutions
Model validation
Validation
• By comparison of the mean state
• By comparison of the
energy and momentum balance
• By comparison of the variability
• By comparison of the hydrological cycle
• By comparison of the processes
(cyclons etc.)
• By reactions to observed sea-surface
temperature changes
model error surface temperature
DJF
flux corrected
zonally averaged
not flux corrected
CMIP
Simulations
from past to future
Simulations of 125 ky BP
(Eemian) and 115 ky BP
today
CO2
Temperature
CH4
115 ka bp
125 ka bp
Parameters of the simulations
CO2 [ppm]
Eemian
125k
270
115k
265
presentday
353
CH4 [ppb]
630
520
1720
N2O [ppb]
260
270
310
Eccentricity
Obliquity
0.0400
23.79
0.0414
22.41
0.0167
23.44
Precession
127.3
290.9
282.7
125 ky BP
115 ky BP
The radiation anomaly compared to
present day
Reconstruction
Velichko et al., 1992
Model
July temperature change
125 ka bp
F. Kaspar
125 ka bp
115 ka bp
The near-surface temperature change (annual mean)
F. Kaspar, K. Prömmel
125 ky BP (Eemian)
Thickness of snow in summer
[m]
115 ky BP
Simulations of the last 1000
years
Volcanism
Solar Radiation
+
=
Effective Forcing
Experiments
1. Erik
starting at the year 1000
 ECHO-G I
2. Columbus starting at the year 1500
 ECHO-G II
Zorita et al, 2004
Forcing
TemperatureResponse
Trend
The solar and volcanic forcing and the model response
Comparison of modelled and reconstructed temperatures
Erik
HADCM nat. forc.
Columbus
A comparison with the Hadley-centre simulation
Scenario experiments
The information chain leading to a climate projection
A2
The globally averaged change of
the near surface temperature
relative to the years 1961-1990,
Simulated with coupled
ocean atmosphere models
B2
The annually averaged change of the near surface temperature for the years
2071-2100 relative to the years 1961-1990, simulated by globally coupled
ocean-atmosphere models for the A2-scenario
The annual mean change
of temperature (map) and
the regional seasonal change
(upper box: DJF; lower
box: JJA) for the scenarios
A2 and B2
The temperature change for all SRES marker scenarios
(simulated by a simplified model)
The temperature evolution of the last 1000 years and the projections
for the next 100 years
Probability density function for different scenarios and timeintervals, as calculated by HADCM
Stott et al, Nature, 2002
AII
IS92
Sea level rise (m)
1.0
0.8
0.6
Scenarios
A1B
A1T
A1FI
A2
B1
B2
All SRES envelope
including land-ice
uncertainty
0.4
0.2
0.0
2000
2020
2040
2060
Year
2080
2100
The projected sea level change
Bars show the
range in 2100
produced by
several models
THC
The ocean conveyor belt circulation
The change of the thermohaline circulation in the
North Atlantic for the IS92a scenario
Summary
• The paleo climate can be simulated with the
coupled ocean-atmosphere models previously
employed for climate change predictions
• The model simulates the Eemian and the
transition to an ice age
• Simulations of the climate of the last 1000 years
show a larger amplitude in the temperature
variability than the proxy reconstructions
• The models predict a climate change between 1.4
and 5.8 K. If the uncertainty is taken into account,
it might well extend beyond 8 K
Outlook
Probabilistic approach
Probability density functions of temperature change
simulated with the Hadley Centre model
Stott and Kettleborough, 2002
Probability density distribution of climate projection
Allen & Ingram, 2002
Model improvements
• too many to name them all
• Here is just one example – the role of the
stratosphere
high index
The two states of the North
Atlantic oscillation (NAO)
low index
The coupling between ocean-atmosphere-stratosphere
Pattern: AO & AAO
Feature: Midwinter warmings
Forcing: QBO
ozone
Pattern: NAO & PNA
solar-cycle
ENSO & PDO
S
T
R
A
T
O
S
P
H
E
R
E
T
R
O
P
O
S
P
H
E
R
E
Feature: Blockings over Pacific and Atlantic
Forcing: Aerosols, gravity waves
Pattern: THC & GC
Feature: SST-Anomalies
Forcing: Seaice
O
C
E
A
N
C
o
u
p
li
n
g
Baldwin,'02
Kodera,'00
Shindell,'99
?
C
o
u
p
li
n
g
Hurrell,'01
Schlesinger,'0
Blessmann
The coupled ocean-tropospherestratosphere model EGMAM
39
STRATOSPHERE
(MESOSPHERE)
19
TROPOSPHERE
●
atmosphere: ECHAM4
●
ocean:
HOPE - G
---------------------------------------
(STRATOSPHERE)
●
coupled:
ECHO - G
OCEAN
20Level
Blessmann
NAO
observations
with stratosphere
without stratosphere
The power spectrum of the 19 level and the 39 level coupled
ocean-atmosphere model for the NAO-index
Blessmann
Ultimate Goal
Technical infrastructure
• Earth simulator - hardware (Japan)
• ESMF (Earth system modelling facility) –
software (USA)
• PRISM (PRogramme for Integrated earth
System Modelling) – software – European
Union
The science :
The users:
- General principles
- Standard physical interfaces
- GUI interface
- Configuration editor
- Diagnostics outputs
PRISM System
The technical developments:
- System architecture
- Coupler and I/O
- Software management
- Vizualisation and diagnostics
The participating models
- Atmosphere
- Atmos. Chemistry
- Ocean
- Ocean biogeochemistry
- Sea-ice
- Land surface
On going PRISM / ESMF collaboration
Earth System Model
Running environment
Coupling
infrastructure
ESMF
User code
Supporting software
PRISM
Scientific projects
• ENSEMBLES (EU-Project) 70+ partners
– Workpackage RT2A: climate change
experiments as suggested by IPCC
• Climateprediction.com (NERC-project,
UK)
– Climate change experiments on home PC‘s,
similar to yeti@home
futurissimo
• Comprehensive simulation of the Holocene
• Simulation of the last glacial-interglacial
• Paleo-data assimilation