Transcript Climate models and assessment of climate change

Introduction to
climate modeling
Peter Guttorp
University of Washington
[email protected]
http://www.stat.washington.edu/peter
Acknowledgements
ASA climate consensus workshop
Kevin Trenberth
Ben Santer
Myles Allen
IPCC Fourth Assessment Reports
Steve Sain
NCAR IMAGe/GSP
Weather and climate
Climate is
–average weather
WMO 30 years (1961-1990)
–marginal distribution of weather
temperature
wind
precipitation
–classification of weather type
state of the climate system
Weather is
–current activity in troposphere
Models of climate
and weather
Numerical weather prediction:
–Initial state is critical
–Don’t care about entire distribution,
just most likely event
–Need not conserve mass and energy
Climate models:
–Independent of initial state
–Need to get distribution of weather
right
–Critical to conserve mass and energy
The heat engine
Greenhouse effect
A simple climate model
What comes in
Sr 2 (1 a)
Solar constant
1367 W/m2
Earth’s albedo
0.3
must go out
4r2 T4
Effective emissivity
(greenhouse, clouds)
0.64
Stefan’s constant
5.67×10-8 W/(K4·m2)
Solution
Average earth temperature is T=285K
(12°C)
One degree Celsius change in average
earth temperature is obtained by
changing
solar constant by 1.4%
Earth’s albedo by 3.3%
effective emissivity by 1.4%
But in reality…
The solar constant is not constant
The albedo changes with land use
changes, ice melting and cloudiness
The emissivity changes with
greenhouse gas changes and
cloudiness
Need to model the three-dimensional
(at least) atmosphere
But the atmosphere interacts with land
surfaces…
…and with oceans!
Historically
mid 70s Atmosphere models
mid-80s Interactions with land
early 90s Coupled with sea & ice
late 90s Added sulphur aerosols
2000 Other aerosols and carbon cycle
2005 Dynamic vegetation and
atmospheric chemistry
The climate engine I
If Earth did not rotate:
tropics get higher solar radiation
hot air rises, reducing surface pressure
and increasing pressure higher up
forces air towards poles
lower surface pressure at poles makes
air sink
moves back towards tropics
The climate engine II
Since earth does rotate, air packets do not
follow longitude lines (Coriolis effect)
Speed of rotation highest at equator
Winds travelling polewards get a bigger
and bigger westerly speed (jet streams)
Air becomes unstable
Waves develop in the westerly flow (low
pressure systems over Northern Europe)
Mixes warm tropical air with cold polar air
Net transport of heat polewards
Modeling the atmosphere
Coupled partial differential equations
describing
Conservation of mass
Conservation of momentum
Conservation of water
Thermodynamics
Hydrostatic equilibrium
Boundary values
Radiative forcings
The effect of gridding
Parameterization
Some important processes happen on
scales below the discretization
Typically expressed in terms of resolved
processes (statistically) or data
Examples:
dry and moist convection
cloud amount/cloud optical properties
radiative transfer
planetary boundary layer transports
surface energy exchanges
horizontal and vertical dissipation processes
Can data force
parametrizations?
Experiment with simple climate model
Realistic priors on forcings
Using several data sets on
hemispheric annual mean temperature
oceanic heat content
Markov chain Monte Carlo analysis
Goal: Estimate climate sensitivity
(temperature response to CO2 doubling)
Hemispheric model
Schlesinger, Jiang & Charlson 1992
NH mixed layer
SH atmosphere
SH mixed layer
NH interior
ocean
SH interior
ocean
NH bottom
SH bottom
Vertical heat transport by upwelling and
diffusion
Atmosphere in equilibrium with ocean
SH polar ocean
NH polar ocean
NH atmosphere
Stochastic model
Observation Y
Model output M(,)
Truth Z
parameters
SOI E
forcings
 , Y    Y Z   Z M,,,E    
Missing data treated as additional
parameters to be estimated
Mixed layer
Upwelling
velocity
Vertical heat
diffusivity
Air-ocean
exchange
Polar parameter
Ocean
hemispheric
exchange
SOI coeff, SH
SOI coeff, NH
Comparison of Mean
Simulation Properties
Simulated
Land Temp
Difference:
Sim- Observed
Sources of uncertainty
Forcings
Sea surface temperature is uncertain,
especially for early years
Greenhouse gases vague estimates for
early part
Data
Global mean temperature is not
measured
Uncertainty in estimates may be as big
as 1°C
Greenhouse
gases
Anthropogenic CO2
from fossil fuel and
land use change
Methane from
agriculture and fossil
fuels
1/3 of NOx from
agricultural sources
Historical data
Sensitivity
Reasonable climate models must
reproduce
El Niño
Pacific Decadal Oscillation
Dust bowl, Sahel drought etc.
El Niño simulations
El Niño simulations
“obs”
temp
precip
slp
simulations
Cloud (OLR) Anomalies and ENSO
Observed
Simulated
Hack (1998)
More Cloud
Less Cloud
Regional models
Dynamic downscaling: Higher
resolution models driven by lower
resolution global models
Statistical downscaling: Regression
model using global model, terrain etc.
Stochastic downscaling: Stochastic
model for subgridscale processes
driven by global model
Dynamic downscaling
of a GCM
Comparing RCM to data
Regional climate model RCM3 from
SMHI
Forced by ERA40
Need to compare distributions
Data observed minimum daily
temperatures at Stockholm
Observatory
How well does the climate
model reproduce data?
Resolution in a regional
climate model
50 x 50 km
Where is the problem?
Regional model corresponds to
grid square average
average over land cover type
3 hr resolution
Data correspond to
point measurement
open air
continuous time
Model
problems with cloud representation
constrain to lower resolution model?
Data issues
Need for high quality climate data
repository (Exeter workshop)
Reanalysis not only needed for met
data
Lots of satellites are deteriorating–
many are not being replaced
Some countries will not make data
available to the international
community
Homogenization
Historical SST data issues
Ocean surface temperature record
Data from buoys, ships, satellites, floats
Arctic ice pack