solar variability and climate changesolar.physics.montana.edu/svecse2008/pdf/cubasch_svecse.pdf ·...

1

Solar variability and ClimateChange

Ulrich Cubasch

Freie Universität, Berlin

Joint work with Fidel Gonzalez-Rouco, Tom Crowley, Gabi Hegerl, Stefanie Legutke, Jürg Luterbacher, Ulrich Schlese, Hans von Storch, Reinhard Voss, Eduardo Zorita U. Langematz, K.Matthes, S. Schimanke, P. Sinigoj,

T. Spangehl, S. Wagner

Joint work with Fidel Gonzalez-Rouco, Tom Crowley, Gabi Hegerl, Stefanie Legutke, Jürg Luterbacher, Ulrich Schlese, Hans von Storch, Reinhard Voss, Eduardo Zorita U. Langematz, K.Matthes, S. Schimanke, P. Sinigoj,

T. Spangehl, S. Wagner

Sponsored by SPP CAWSES, DFG Germany

2

outline

• Introduction• Orbital influences• Direct influences• Indirect influences• Summary

3

4

5

6

Introduction

7The climate system

External ForcingAnthropogenic Forcing

8

Change of solar inputby orbital parameters

obliquity�tilt of the earth axis

excentricityprecession

~100 ky

~ 23 und 19ky

~41 ky

precessionexcentricity

excentricity

precession

obliquity

9

Sun

UV

Earth

Ozone T, U

Indirect Influence

Dynamical impact

trop.

Strat.

Visible

Earth

Direct Influence

Radiativeimpact

trop.

Strat.

Sun

?

Possible Ways for Solar Influence on Climate

10

The influence ofcosmic rays onclouds

Possible Ways for Solar Influence on Climate

Not yet quantified and not yet modelled!

11A climate model

12

The land-sea mask of a typical climatemodel used in long simulations

(ca. 3.75 x 3.75 deg)

13

Orbital forcing

14

Simulations of 125 ky BP and 115 ky BP (Eem and its end)

known: orbital parameterknown: greenhouse gas

concentration

15125 ka bp

115 ka bp

You are here

16

Periodicity of the orbital parameter115 ky bp

125 ky bp

17

125 ky BP 115 ky BP

The radiation anomaly compared to present day

18

125 ka bp 115 ka bp

The near-surface temperature

19

January temperatures of the early Eemian(Corylus-phase)

(difference to observed present-day values)

N. Kühl

modelled reconstructed

20

Transition to the cold phase

21

Permanently snow-covered areas in thesimulation for 115 kyr BP

22

Sea level changes

-1.4

-1.2

-1

-0.8

-0.6

-0.4

-0.2

0

0 500 1000 1500 2000

Years

Sea

leve

l cha

nge

[m]

Snow accumulation over North America

Oceanic density changes

Oceanic density changes with removedtrend of preindustrial simulation

snow accum.

thermal expansion

total

23

Simulations for the Holoceneclimate optimum

24

7 ka BP-present 6 ka BP-present 5 ka BP-present

Forcings

Orbital Forcing (Berger, 1978)

S. Wagner

25

Solar forcing: based on 14C production rate (Solanki et al., 2004)Scaling corresponds to 0.30% insolation difference present day- Maunder Minimum S. Wagner

Selected time interval

today

26

ForcingsGreenhouse gas forcings: Antarctic ice cores (Flueckiger et al., 2002)

CO2

CH4

N2O

S. Wagner

27

DJF

Time series filtered with 50 year Hamming low pass filter

JJA

„Europe“:0 - 30°E45°N - 70°N

S. Wagner

only orbital

only orbital

orbital, GG, TSI

orbital, GG, TSI

European temperatures

28

Direct influence

29

How was the climate variabilitybefore industrialization?

30

+

=

Volcanisms

solar radiation

Forcing

31

solar variability

• 11-year cycle (Schwalbe-cycle): ca. 0.1 % ~ 0.24 W/m2 at the surface• 80-year cycle (Gleissberg-cycle): ca.

0.24 - 0.3 % 0.5-0.75 W/m2 at thesurface

• many more.....?

32

Simulations of the climateof the last century

Volcanism

Solar variability

Greenhouse Gases

Temperaturelittleice age

medivalwarm period

IPCC, 2007

33

Simulated near surface-temperature anomaly1675-1710

34

1675-1710vs. 1550-1800

Reconstruction from historicalevidence , from Luterbacher et al.

Model-Simulation

Late Maunder Minimum

35

What part of the climatevariability is anthropogenic, what

part natural?

36

The observed near-surface temperature change,averaged over the northern hemisphere

37

The observed and modelled (solar only) near-surface tem perature change,averaged over the northern hemisphere

38

The observed and modelled (solar only, volcanoes only) near-surface temperature change, averaged over the north ern hemisphere

39

The observed and the modelled (solar, volcanic, solar&v olcanic) near-surfacetemperature change, averaged over the northern hemispher e

40

The observed and the modelled (greenhouse gases) near-su rfacetemperature change, averaged over the northern hemispher e

41

The observed and the modelled (greenhouse gases&solar) n ear-surfacetemperature change, averaged over the northern hemispher e

42

The observed and the modelled (greenhouse gases&volcanoe s)near-surface temperature change, averaged over the north ern hemisphere

43

The observed and the modelled (greenhouse gases in connec tion withsolar, volcanic, solar&volcanic) near-surface

temperature change, averaged over the northern hemispher e

44

The observed and the modelled (greenhouse gases, dto. in connection withsolar, volcanic, solar&volcanic) near-surfacetemperature change , averaged over the northern hemispher e

observations

45

The AR4 modelling results

natural and anthropogenic forcing

natural forcing

observations

46

Can we see the 11-year solar cycle in the troposphere/ocean?

47

NCEP-reanalysis

GPRC Xie & Arkin

The precipitation anomalyin the tropical Pacificduring solar maximum

van Loon et al, 2006

48

scheme for solar max•solar forcing is spatial heterogeneous, strongest in the cloudless subtropics• increased solar radiation enhances the evaporation in the subtropics• intensified moisture transport reinforces the main precipition regimes• increased subsidence decreases cloud cover in the subtropics• stronger trade winds enforce upwelling and trigger a La Nina like situation

48

Meehl et al. (in press)

49

The precipitationanomaly duringsolar maximum

observed

van Loon et al, 2006

When I know the state of the solar cycle, I canpredict ENSO and with it the rain in California for

the next 10 years!? …..butENSO has no 11-year cycle!

50

The idealized forcing and the response of the near surface temperature

51

Precipitation patternObserved: J/F mean max year anomaly (van Loon et al. 2006)Modelled: January/February max-

min composite of 60 years

51

• precipitation declines along the equator with pos. changes northward and southward during solar maximum

52

• The 11-yr signal in Sun’s irradiance appears to excite a damped resonant response (i.e. the QDO) in the Earth’s climate system. (White + Liu, 2008) with a lag of 3-5 years.

• The direct forcing appears to be insufficient to explain the size of the response (Meehl et al, in press), therefore both, the direct and the indirect response have to work together.

53

Indirect influence

54

Solar effect in the stratosphere

• Solar heating modulates temperature gradients in the upper stratosphere/ lower mesosphere• reinforcement of zonal winds deflects planetary waves, which creates positive divergence of Eliassen-Palm-flux• this results a weakening of the Brewer-Dobson-circulation aligned with positive temperature anomalies in the lower tropical stratosphere

54

Kodera and Kuroda 2002

55

Forcing of 110 year runComposite (1953 to 2004) + repeat (1962-2004)

1953 20041962

f10.7 flux

U (m/s) @50 hPa

from observations

(NOAA/SEL f10.7 dataset)

from observations

(M. Giorgetta’s extended QBO dataset)

repeat

repeat

repeat

repeat

56

Solar Signal in Ozone (%/100 f10.7)WACCM vs. Observations

modelled observed

57

Solar Signal in Temperature (K/100 f10.7)modelled observed

58

59

Summary• The sun influences the climate on various time-scales

by orbital variations, TSI and UV-shift• However, its influence is not sufficient to explain the

recent warming.• The 11-year solar signal in the upper stratosphere

leads to circulation changes in the lower stratosphereand down to the troposphere at polar and equatoriallatitudes.

• There are indications that it has some influence in thePacific region, which might give some potential fordecadal weather prediction.

• The interplay between direct and indirect effect is notclear.

• The interstellar partical flux and its influence on climate are not yet quantified and can therefore notbe modellled.