Folie 1 Climate models, downscaling and uncertainties Hans von Storch, GKSS Research Centre, Geesthacht, and KlimaCampus clisap, University of Hamburg.

Folie 1

Climate models, downscaling and uncertainties

Hans von Storch, GKSS Research Centre, Geesthacht, and

KlimaCampus „clisap“, University of Hamburg Germany

Folie 2

Who is this?

Hans von Storch (born 1949)

Diploma in mathematics,PhD in meteorology

Director of Institute for Coastal Research, GKSS Research Center, near Hamburg,

Professor at the Meteorological Institute, KlimaCampus, University of Hamburg

Works also with social and cultural scientists.

Folie 3

Overview:1. Quasi-realistic climate models

(„surrogate reality“)

2. Free simulations and forced simulations for reconstruction of historical climate

3. Climate change simulations

4. Downscaling - Regional climate modelling

5. Regional scenarios

Folie 4

Models as surrogate reality• dynamical, process-based models, • experimentation tool (test of hypotheses) • design of scenario • sensitivity analysis • dynamically consistent interpretation and extrapolation of observations in space and time (“data assimilation”) • forecast of detailed development (e.g. weather forecast)

characteristics: complexity quasi-realistic mathematical/mechanisticengineering approach

Folie 5

Components of the climate system. (Hasselmann, 1995)

Folie 6

Quasi-realistic climate models …

… are dynamical models, featuring discretized equations of the type

)(dt

dΨ, k

iki

k P with state variables Ψk and processes Pi,k.

The state variables are typically temperature of the air or the ocean, salinity and humidity, wind and current.

… because of the limited resolution, the equations are not closed but must be closed by “parameterizations”, which represent educated estimates of the expected effect of non-described processes on the resolved dynamics, conditioned by the resolved state.

Folie 7

atmosphere

Folie 8

Dynamical processes in the atmosphere

Folie 9

Dynamical processes in a global atmospheric general circulation model

Folie 10

Results of a survey among climate modellers in 1996, 2003 and 2008

Bra

y an

d vo

n S

torc

h, 2

010

Folie 11

Modell

Beobachtet

Klimazonen

Klassifikation nach KoeppenErich Roeckner, pers. Mitteilung

Folie 12

Observed Simulated Winter(DJF)

Erich Roeckner, pers. Mitteilung

Zyklogenese

Sturmbahn-dichten

Folie 13

Precipitation in IPCC AR4 models

Erich Roeckner, pers. Mitteilung

Folie 14

Free and forced simulations for reconstruction of historical climate

Free and forced simulations for reconstruction of historical climate

Folie 15

.... tions"reconstrucdriven Data"

egetationor v

sheets) ice (e.g.,opography or t

ion)configurat orbital (incl.output solar or

ionsconcentrat aerosolor

ionsconcentrat gas greenhouse ith w

) ;F( :"Simulation Forced"

)F( :"Simulation Free"

t

tt1t

t1t

Different ways of running the model

Folie 16

Free Simulation: 1000 yearsno solar variability, no changes in greenhouse gas concentrations, no orbital

forcing

Tem

pera

ture

(at

2m

) de

viat

ions

fr

om 1

000

year

ave

rage

[K

]

Zor

ita,

200

1

Folie 17

1000-2000 simulation

Changing solar forcing and

time variable volcanic aerosol load;

greenhouse gases

1000-2000 simulation

Changing solar forcing and

time variable volcanic aerosol load;

greenhouse gases

Forced SimulationForced Simulation

Folie 18

Folie 19

1675-1710vs. 1550-1800

Reconstruction from historical evidence, from Luterbacher et al.

Late Maunder Minimum

validation

Model-based reconstuction

Folie 20

Global 1675-1710 temperature anomaly

Folie 21

Climate change simulationsClimate change simulations

Folie 22

Folie 23

Scenarios of what?

• Climate = the statistics of weather, usually described by probability density functions, in particular by- their moments (e.g., mean, std deviation, covariances), - percentiles and return values,- spatial characteristics (e.g., EOFs), - temporal characteristics (autocovariance function, spectra)

Folie 24

Scenario building

• Construction of scenarios of emissions. • Construction of scenarios of concentrations of

radiatively active substances in the atmosphere.

• (Ok – not quite exact; aerosols …)• Simulation of climate as constrained by

presence of radiatively active substances in the atmosphere (“prediction” of conditional statistics).

Folie 25

“SRES” ScenariosSRES = IPCC Special Report on

Emissions Scenarios

A world of rapid economic growth and rapidintroduction of new and more efficient technology.

A very heterogeneous world with an emphasis onfamily values and local traditions.

A world of “dematerialization” and introduction of clean technologies.

A world with an emphasis on local solutions toeconomic and environmental sustainability.

“ business as usual ” scenario (1992).

A1

A2

B1

B2

IS92a

IPC

C, 2

001

Folie 26

Scenario building

1. Simulation with global models, which describe several compartments of the global earth system – relatively coarse spatial grid resolution (e.g., 200 km)

2. Simulation with regional models, often with only one or a few compartments (mostly atmosphere) – relatively high spatial grid resolution (e.g., 50 km)

3. Simulation with impact models – a large variety of different systems, e.g., storm surges or ocean waves.

Folie 27

Scenario A2

Scenario B2

Danmarks Meteorologiske Institut

Annual temperature changes [°C]

(2071–2100) –(1961–1990)

Folie 28Agreement among 7 out of a total of 9 simulations

precipitation

IPCC (2001) „regional development“ scenarios A2 and B2.

IPCC (2001) „regional development“ scenarios A2 and B2.

Gio

rgi et

al., 20

01

Folie 29

Typical “global” atmospheric model grid resolution with corresponding land mask.

T42 used in global models. (courtesy: Ole Bøssing-Christensen)

Folie 30

global model

Well resolved

Insufficiently resolved

Spatial scales

vari

ance

Folie 31

DownscalingDownscaling

Folie 32

Globale development(NCEP)

Dynamical DownscalingCLM

Simulation with barotropicmodel of North Sea

Empirical Downscaling

Tide gauge St. Pauli

Cooperation with a variety of governmental agencies and with a number of private companies

Regional and local conditions – in the recent past and next

century

Folie 33

Typical regional atmospheric

model grid resolutions with

corresponding land masks.

50 km grid used in regional

models (courtesy: Ole

Bøssing-Christensen)

Folie 34

Well resolved

Insufficiently resolved

Spatial scales

vari

ance

regional model

Added value

Folie 35

.operator suitable a

)(

:nintegratio Forward

modeln observatio

model dynamical

errorsn observatio and model , h wit

equation n Observatio

equation space State

1111

*1

*1

1

1

Kwith

)dK(dΨΨ

Gd

);ηF(ΨΨ

G

F

δ) G(Ψd

ε) ;ηF(ΨΨ

t*t

*t t

tt

tt*t

tt

ttt

tttt

Concept of Dynamical DownscalingRCM Physiographic detail

3-d vector of state

Known large scale state

projection of full state on large-scale scale

Large-scale (spectral) nudging

Folie 36

Example Extreme Events (Wind & Waves)

2 24.38 25.17 25.96 24.05 25.21 26.37 7.12 7.49 7.86 6.41 6.77 7.135 25.86 27.28 28.70 25.75 27.64 29.53 7.84 8.44 9.04 6.93 7.54 8.15

25 28.44 31.33 34.22 28.09 32.77 37.45 8.99 10.35 11.71 7.52 9.21 10.902 22.50 23.16 23.82 23.16 24.03 24.90 5.89 6.15 6.41 5.52 5.84 6.165 23.76 24.82 25.88 24.33 25.94 27.55 6.34 6.83 7.32 5.89 6.46 7.03

25 25.67 28.00 30.33 26.43 29.75 33.07 6.90 8.20 9.50 5.99 7.88 9.772 23.29 24.15 25.01 23.11 24.03 24.95 6.78 7.06 7.34 5.60 5.84 6.085 24.89 26.32 27.75 24.15 25.94 27.73 7.37 7.79 8.21 5.97 6.46 6.95

25 26.68 30.70 34.72 26.42 29.75 33.08 8.04 9.03 10.02 6.34 7.88 9.42

EU

RS

ON

Yea

rs

Wind [m/s]

K13

Waves [m]Hipocas Observed Hipocas Observed

90rx 90

rx 90rx 90

rx 90rx 90

rx 90rxrx rx rx rx90

rx

2, 5, and 25-year return values with 90% confidence limits based on 10.000 Monte Carlo simulations each.

(Weisse and Günther. 2006)

Folie 37

A set of model data of recent, ongoing and possible future coastal climate(hindcasts 1948-2008, reconstructions and scenarios for the future, e.g., 2070-2100)

Based on experiences and activities in a number of national andinternational projects (e.g. WASA, HIPOCAS, STOWASUS, PRUDENCE)

Presently contains atmospheric and oceanographic parameter(e.g. near-surface winds, pressure, temperature and humidity; upper air meteorological data such as geopotential height, cloud cover, temperature and humidity; oceanographic datasuch as sea states (wave heights, periods, directions, spectra) or water levels (tides and surges) and depth averaged currents, ocean temperatures)

Covers different geographical regions(presently mainly the North Sea and parts of the Northeast Atlantic; other areas such as the Baltic Sea, subarctic regions or E-Asia are to be included)

http://www.coastdat.de, contact: Ralf Weisse ([email protected])

What is coastDat?

Folie 38

- Ship design- Navigational safety- Offshore wind- Oils spill risk- Interpretation of measurements- Chronic Oil Pollution- Ocean Energy

Wave Energy Flux [kW/m]

Currents Power [W/m2]

Some applications of

Folie 39

Scenarios for Northern Germany

Folie 40

RCAO HIRHAM

A2 - CTL: changes in 99 % - iles of wind speed (6 hourly, DJF): west wind sector selected (247.5 to 292.5 deg)

Sce

nari

os

for

207

0-2

10

0

Wo

t h,

pe

rso

na

l co

mm

un

ica

ti on

Folie 41

North German Climate Office@GKSS

An institution set up to enable communication between science and stakeholders

• that is: making sure that science understands the questions and concerns of a variety of stakeholders

• that is: making sure that the stakeholders understand the scientific assessments and their limits.

Typical stakeholders: Coastal defense, agriculture, off-shore activities (energy), tourism, water management, fisheries, urban planning

Folie 42

Online-Atlas „Klimawandel Norddeutschland“

Darstellung unterschiedlicher Größen zum Klimawandel in Norddeutschland für die Zeiträume 2011-2040, 2041-2070 und 2071-2100.

Darstellung der Differenz zu dem Kontrollzeitraum 1961-1990

Darstellung unterschiedlicher Treibhausgasszenarien (nach dem IPCC) gerechnet mit verschiedenen regionalen Modellen

http://www.norddeutscher-klimaatlas.de/

Folie 43

Conclusions

• Global climate modeling allows the representation of global, continental and sub-continental scales. Global models fail on the regional and local scale.

• Global climates is varying because of both internal dynamics as well as external forcing.

• Scenarios of future climate change hinge on the validity of economic scenarios.

• Simulation of regional climate is a downscaling problem and not a boundary value problem.

• Marine weather (winds, waves) have been successfully reconstructed for the years 1958-97 with a 1-hourly resolution. (CoastDat@GKSS)

Folie 44

Background information on this issue:

von Storch, H., S. Güss und M. Heimann, 1999: Das Klimasystem und seine Modellierung. Eine Einführung. Springer Verlag ISBN 3-540-65830-0, 255 pp

von Storch, H., and G. Flöser (Eds.), 2001: Models in Environmental Research. Proceedings of the Second GKSS School on Environmental Research, Springer Verlag ISBN 3-540-67862, 254 pp.

Müller, P., and H. von Storch, 2004: Computer Modelling in Atmospheric and Oceanic Sciences - Building Knowledge. Springer Verlag Berlin - Heidelberg - New York, 304pp, ISN 1437-028X

Folie 45

http://coast.gkss.de/staff/[email protected]

Weblog KLIMAZWIEBELhttp://klimazwiebel.blogspot.com/