Cosmic rays and climate

1

Jasper Kirkby /CERNCERN Colloquium, 4 June 2009

1. Present climate change

2

1880 1900 1920 1940 1960 1980 2000

-.4

-.2

.0

.2

.4

.6

Annual Mean5-year Mean

Global Land-Ocean Temperature Index

Tem

per

ature

Anom

aly (°C

)

Climate forcings (IPCC 2007)

• 0.7oC rise since 1900 (not uniform)

• IPCC findings:

‣ Total anthropogenic 1.6 W/m2

(≅ 1 candle per 25 m2)

‣ Negligible natural (solar) contribution: 0.12 W/m2

‣ Clouds poorly understood3

GISS

4

Why clouds are important for climate change

• Clouds cover ~65% of globe, annual average

• Net cooling of 30 W/m2

• c.f. 1.6 W/m2 total anthropogenic

John Constable, Cloud study, 1821

NASA CERES satellite

• Data from CERES satellite (Clouds and Earth’s Radiant Energy System)

• Clouds (and oceans) are poorly simulated in climate models(finest grid sizes ~100 km x 100 km)

5

II. Evidence for pre-industrial solar-climate variability

• Numerous palaeoclimatic reconstructions suggest that solar/GCR variability has an important influence on climate

• However, there is no established physical mechanism, and so solar-climate variability is:‣ Controversial subject‣ Not included in current climate models

6

7

Little Ice Age and the sunspot record

0

100

200

1850 1900 1950 20000

100

200

Sun

sp

ot

nu

mb

er

Year

1600 1650 1700 1750 1800

Maunder Minimum

DaltonMinimum

Little Ice Age

• Inactive sun (low sunspot peak, long cycle length) ⇒ cold climate

• Active sun (high sunspot peak, short cycle length) ⇒ warm climate

The frozen Thames, 1677

CO2(ppm)

CO2320

290

!18O

(‰)GCR/

"14C

(‰)

!18O

-8.0

-7.5

-7.0

-10

0

10

20500 1000 1500 2000

Year (AD)

Mangini et al., EPSL 235 (2005)

GCR

1oC

Global climate - last 2000yr

• Little Ice Age and Medieval Warm Period

• Global observations

8

0

3

6

La

ke

Mu

cu

ba

ji m

ag

ne

tic

su

sce

ptib

ility

(S

I x1

0-6

)

gla

cia

l a

dva

nce

incre

asin

g G

CR

Little Ice Age

1000 1200 1400 1600 1800 2000

1000 1200 1400 1600 1800 2000

Year (AD)

hockey stick

Te

mp

era

ture

an

om

aly

(oC

)

0.4

0

-0.4

-0.8

Ca

rbo

n-1

4 a

no

ma

ly (

x 1

0-3

)

20

0

-20

a) Northern hemisphere temperature

b) Galactic cosmic rays

c) Tropical Andes glaciers

instrumental

14C (rh scale)10Be (Greenland)

(arb. scale)

multi-proxy(tree rings, pollen,

shells, stalagmites, etc.)

Medieval Warm

worldwide boreholes

10Be (South Pole)

(lh scale)

40

30

20

10

0

-10

GC

R c

ha

ng

e (

% f

rom

19

50

va

lue)

-3

2

G

RIP

-3

1

-20

.6

Dye

3

-1

9

Gre

en

lan

d s

urf

ace

te

mp

era

ture

(o

C)

Greenland boreholes (rhs)

high GCR flux cool climate

low GCR flux warm climate

Austrian speleothem:

Siberian climate - last 700 yr

• Correlation recently reported between solar/GCR variability and temperature in Siberia from glacial ice core

• 30 yr lag (ie. ocean currents may be part of response)

9

2

1

0

-1Sib

eria

te

mp

era

ture

(o

C)

So

lar

mo

du

latio

n

(! u

nits)

4

2

0

-2

1300 1400 1500 1600 1700 1800 1900 2000

Year

temperature(lhs)

10Be

14C

solar/GCR (rhs):

Eichler et al. GRL 36 (2009)

N. Atlantic ice rafted debris - last 10 kyr (Holocene)

10

Bond et al, Science 294, 2001

12 10 8 6 4 2

Calendar age (kyr BP)

-4

0

4

-4

0

4

-0.4

0.0

0.4

0.1

0.0

-0.1

ice-rafted debris

ice-rafted

debris

0

Little Ice Age

Medieval Warm

GCR

(10Be)

GCR

(14C)

Low cosmic

ray flux

High cosmic

ray flux

Low cosmic

ray flux

High cosmic

ray flux

Less

ice

More

ice

Less

ice

More

ice

Cosmic rays: Ice-rafted

debris (%):

• LIA is merely the most recent of around 10 such events in Holocene

quartz grainshaemetite-stained grains

Icelandic glass grains

0.12 mm

foraminifera

W1

W2

W3W4

D5

D6

D7D8

D9

D11

D12D10 D13

W0 Equator

July ITCZ

January ITCZ

GCR influence on ITCZ/monsoon in Little Ice Age?

11

Drier in LIAWetter in LIA

ITCZdisplacementduring LIA

high GCR flux southerly ITCZ shift

low GCR flux northerly ITCZ shift

Indian Ocean monsoon - 6.5-9.5 kyr ago

• Solar/GCR forcing of Indian Ocean monsoons (ITCZ migration) on centennial—even decadel—timescales

12

High pres s ure

IT C Z

Jun-Aug

monsoons

510 ± 301400 ± 303010 ± 60

3740 ± 130

3790 ± 140

4380 ± 260

4370 ± 100

4900 ± 140

5750 ± 100

6540 ± 180

6960 ± 50

7910 ± 210

8880 ± 908870 ± 230

9760 ± 15010,060 ± 210

10,470 ± 17010 c

m

6.500 7.000 7.500 8.000 8.500 9.000 9.500

–15

–10

–5

0

5

10

14C

(‰

)

18O

(‰

VP

DB

)

Age (kyr BP)

8.0007.900 8.100 8.200 8.300

Age (kyr BP)

a

–5.5

–5.0

–4.5

–4.0

–3.5

–5.5

–5.0

–4.5

–4.0

18O (rainfall)

14C (GCR intensity)

–20

–15

–10

–5

0

5

10b 14C (GCR intensity)

18O (rainfall)

U. Neff et al. (Nature 411, 2001)

14C

(‰

)

18O

(‰

VP

DB

)

high

GCR

low

GCR

high

GCR

dry

wet

dry

wet

low

GCR

Solar-climate mechanisms

• Candidates forsolar-climate variability:‣ direct effect:

✦ total solar irradiance✦ solar UV

‣ indirect effect:✦ galactic cosmic rays (GCRs) /ionising particles

• Can be resolved in principle since GCRs are, in addition to 11-year solar cycle, modulated by:‣ solar magnetic disturbances (esp. high latitude effects)‣ geomagnetic field (low latitude effects)‣ galactic environment

13

Glo

bal m

ean tem

pera

ture

(0C

)

0

-0.05

-0.10

0.05

0.10

0.15

Lean et al. (1995)

Lean et al. (2002)

solar irradiance

estimated forcing

(MAGICC GCM)

1600 1650 1700 1750 1800

Year

1850 1900 1950 2000

360

380

400

420

440

18

O (‰

, V

PD

B)

–7

–8

–9

–10

–11

–12

Mo

nso

on

inte

nsity

East Asian monsoon instensity; Wang et al., Nature 451, 2008

Sanbao/Hulu caves, China

Age (kyr BP)

0 20 40 60 80 100 120 140 160 180 200 220

N. H

em

isphere

sum

mer

insola

tion (

W m

-2)

2! U-Th dating errors

40

100

80

60

120

Ge

om

ag

ne

tic d

ipo

le m

om

en

t (1

02

1A

m2

)

18

O (

‰ V

PD

B)

–9.0

–7.8

–8.2

–8.6

–7.4

–7.0

00050 400030002000 00090001 800070006000

Age (yr B.P.) Age (yr B.P.)

r(0–5000 BP)

= 0.86

480

470

460

490

Su

mm

er

inso

latio

n,

30

N (

W m

-2)geomagnetic field

(± 2! band)

Asian monsoon

(Dongge Cave, China)

summer

insolation

Knudsen & Riisager, GSA 2009

00050 4000300020001000–30

20

10

0

–10

–20

–0.2

–0.4

–0.6

0.0

0.2

0.4

r (0–5000 BP)

= 0.71

18O

(‰

VP

DB

)

dip

ole

mo

me

nt

(10

21

Am

2)

geomagnetic field

Asian monsoonresiduals for last 5000yr

mo

nso

on

in

ten

sity

mo

nso

on

in

ten

sity

Asian monsoon and geomagnetic field

14

• Asian monsoon controlled by orbital insolation - with strong millennial-scale variability

• Possible influence of geomagnetic field on Asian monsoon?

• Opposite sign of effect: lower B field → increased GCR → increased monsoon intensity, which could result from latitudinal differences of solar- and geomagnetic modulations

15

• Orbital period of Sun/Earth around Milky Way ~550Myr

• High GCR flux in spiral arms => 140 Myr period

• Same period and phase found in benthic sea temperature (4oC amplitude) and ice age epochs (icehouse/greenhouse)

Galactic modulation of climate? - 500 Myr

Shaviv & Veizer,GSA Today, 2003

GCRCO2

icehousegreenhouse

III. Solar variability in the 20th century

16

Sun (photosphere) seen in visible (677nm) at solar max (2001)

17

NASA/ESASOHO

Sun (corona) seen with extreme UV eyes (20nm)

18

NASA/ESASOHO

0

5

10

15

So

lar

ma

gn

etic f

lux (

x1

01

4 W

b)

Year

0

0.4

0.8

1.2

1.6

10B

e c

on

ce

ntr

atio

n (

x1

04 g

-1)

1700 1750 1800 1850 1900 1950 2000

10Be (GCR)

solar magnetic fluxgalactic cosmic

rays

cycle

19

cycle

20

cycle

21

cycle

22

cycle

23

Moscow

Mirny

Murmansk

Alma-AtaC

osm

ic r

ay in

ten

sity (

cm

-2s

-1)

3.5

3.0

2.5

2.0

1.5

200

100

0

Su

np

ot

nu

mb

er

1960 1970 1980 1990 2000

Year

b)

a)

Cosmic ray changes during 20th century

19

• Solar open magnetic flux increased by x2.3 in 20th century

• GCR net decrease by ~20% (mostly in 1st half of century)

• Largely only solar cycle variations of GCR flux in 2nd half

Lebedev balloon GCR data:

weaker solarmodulation at low latitudes,due to geomagnetic shielding

1920 19401900 1960 1980 2000

Year

Cu

mu

lative

se

a le

ve

l ch

an

ge

(m

m)

150

100

50

0

Holgate, GRL (2007)

1920 1940 1960 1980 2000

Year (decade mid-point)

Rate

of sea level change (

mm

/yr)

-4

-2

0

2

4

6

8sea-level rate:

9 stations

177 stations <TOPEX

satellite>

Sea-level change in 20th century

• Steady rise of sea-level; mean rate = 1.7 mm/yr• No increase in rate during recent decades• Thermal expansion of oceans (mainly) + land ice melting• Rate of sea-level rise is strongly modulated

20

1920 1940 1960 1980 2000

Year (decade mid-point)

Rate

of sea level change (

mm

/yr)

-4

-2

0

2

4

6

8sea-level rate:

9 stations

177 stations <TOPEX

satellite>

sunspot number

0

50

100

150

200

Sunspot num

ber

(sm

ooth

ed)

1920 1940 1960 1980 2000

Year

⇒ solar modulation? (but solar irradiance variation is too small)

Sea-level positive feedback

21

Recent global temperatures - last 30 yr

• Most of ~2 yr fluctuations due toEl Niño-Southern Oscillation (eg 1997-98)+ volcanoes (El Chichon 1982, Pinatubo 1991)

• Satellite and radiosonde (tropospheric) data show

‣ less warming than thermometer measurements of surface

‣ no enhanced upper tropospheric warming, expected from GHG

• Mean global temperatures flat over last 8 years. Cause not known but CO2 increasing, so must be natural forcing

22

surface temperatures; thermometers

(Hadley and GISS)

tropospheric temperatures;

satellite microwaves (UAH)

1.0

0.8

0.6

0.4

0.2

0

-0.2

-0.4

-0.6

Tem

pera

ture

anom

aly

(oC

)

Year

1980 1985 1990 1995 2000 2005

1.0

0.8

0.6

0.4

0.2

0

-0.2

-0.4

-0.6

Pacific Decadel Oscillation

• PDO (Hare 1996) is similar to ENSO (temperature anomalies and surface winds), except for long (30 yr) periodicity and primary effect on Pacific NW

• PDO transitions coincide with gradient changes of global temperatures

• PDO may be shifting to negative phase23

No

rma

lise

d c

en

tra

l d

ep

th

0.60.4 0.8 1.0 1.2 1.4 1.6

! Wavelength (+1564 nm)

0.0

0.1

0.2

0.3

0.4

0.5

0.6

Fe I 1564.8

CNOH

OH

OH 1565.3

Jan 2002

Jun 1991

Fe 1564.8 magnetic spliting

minimum B field for visible sunspots

2000 2005 2010 2015

2000 2005 2010 2015

Year

Year

2014

2015

(Su

nsp

ot

/ p

ho

tosp

he

ric)

inte

nsity

Sunspot contrast

(1.0 = invisible)

1.0

0.9

0.8

0.7

0.6

Su

nsp

ot

ma

gn

etic f

ield

(G

)

3000

2500

2000

1500

Livingston and Penn, 2005 example measurements:

Sunspot weakening

‣ Temperature-sensitive molecular lines‣ Zeeman splitting of Fe I line‣ Continuum brightness of sunspot umbrae

• Sunspot umbrae warming at 45K /yr

• Sunspot magnetic fields decreasing at 77 G/yr

• Independent of sunspot cycle

• Linear extrapolation ⇒ sunspots vanish after

2015 (like Maunder Minimum)24

Livingston and Penn, National Solar Observatory, Tucson, AZ"Sunspots may vanish by 2015", submitted 2005 (rejected for publication)

1364

1366

1368

Sola

r Ir

radia

nce (

Wm

!2)

1364

1366

1368

!2)

0.1%cycle 21 cycle 22 cycle 23

Total solar irradiance (World Radiation Center, Davos)

Galactic cosmic rays (Oulu neutron monitor)

Year

19801975 1985 1990 1995 2000 2005 2010

Year

1980 1990 2000 2010

10

0

-10

-20Counting r

ate

change (

%)

Current very low solar activity

• Currently one sunspot on Sun, and GCRs high

• Next sunspot cycle 24 is very late

• Mean sunspot cycle length is 11.1 yr

• Length of cycle 23 is now >13.1 yr

• Last time such a long cycle occurred was cycle 4 (1784-1798) just before the Dalton minimum - coincided with notable cold period of few decades

• We live in interesting times for the Sun...(hopefully a blessing not a curse)

25

0

100

200

1850 1900 1950 20000

100

200

Su

nsp

ot

nu

mb

er

Year

1600 1650 1700 1750 1800

Maunder Minimum

DaltonMinimum

Little Ice Age

IV. Physical mechanism

• GCRs/ionising radiation may affect cloud amount via:‣ CCN number concentration, and/or‣ ice particle formation in clouds

26

2. increased cloud lifetime

(drizzle suppression)

1. increased cloud

albedo

unperturbed cloudscattering and

absorption of

solar radiation

“direct effect” “indirect effect”

(more cloud droplets)

cloud droplet

(CCN)aerosol

27

• All cloud droplets form on aerosol “seeds” known as cloud condensation nuclei - CCN

• Cloud properties are sensitive to number of droplets

• More aerosols/CCN => brighter clouds, with longer lifetimes

Radiative forcing from aerosols

cloud condensation

nucleus (CCN)

(~100 nm)

cloud droplet

(~10-20 m)

• Aerosol particles = condensation seeds

• Charged particles = condensation seeds (at very high supersaturations)

• Can cosmic rays, under natural conditions, influence aerosols, clouds and climate?

Seeds for cloud formation

28

+

+ ++

+

--

-- -

contrails

bubble chamber

ship tracks off N.America

critical cluster

neutral

cluster

charged cluster

spontaneous growth

sponta

neous g

row

th

evaporation

nucleation

barrier

! G

ibb

s f

ree

en

erg

y

0 5 10No. H2SO4 molecules

0

Possible mechanism• Important source of cloud

condensation nuclei is gas-to-particle conversion: trace gas → CN → CCN

• Ion-induced nucleation pathway is energetically favoured but limited by the ion production rate and ion lifetime

29

aerosolparticle,

CN

cloud condensationnucleus, CCN

cloud droplet

H2OH2SO4

H2O

N2+

galacticcosmic

rays

clusterion

criticalcluster

HSO4¯

ion pairs

O2¯

NO3¯ - -

neutralcluster

critical cluster

-H2O

H2SO4 H2SO4

H2O H2O H2SO4

SO42-

H3O+

H2O

H2SO4

H3O+

0.3 nm 1 nm 100 nm > 10 µm

F. Yu et al., ACP 2008

Ratio of ion-induced nucleation rates to all primay aerosol sources

(dust, sea salt, black carbon, organic carbon) - for lowest 3km altitude

.001 .01 .1 1 3 10 30 100 300 1000 10000

Is ion-induced nucleation globally important?

• Modeling studies:‣ Kazil et al.

ACP 2006: “No”‣ Pierce & Adams

GRL 2009: “No”‣ Yu et al.,

ACP 2008: “Yes”

• All modeling studies depend on uncertain experimental parameters

• Atmospheric observations over land (boreal forest) suggest ~10-20% of new particles are ion-induced (Laakso et al, 2007), but alternative model interpretation of same dataset (Yu and Turco, 2008) find much higher fraction, ~80%

• Ion-induced nucleation likely to be more important over oceans and at high altitudes (lower background aerosols, trace gas concentrations and temperatures) - but few measurements exist

30

highest SPE intensity sites (6 stars)

Ae

roso

l In

de

x (

no

rm.

va

ria

tio

n)

Aerosol production by solar cosmic rays

• Solar proton event 20 Jan 2005 (GLE)

• TOMS satellite measurement of optical depth/Aerosol Index (AI)

• Increase of sulphate/nitrate aerosol

• Further satellite/LIDAR observations have been made of increased aerosol production in atmosphere due to ionising particles

31

Mironova et al, GRL 2008

Cloud observations

• Original GCR-cloud correlation made by Svensmark & Friis-Christensen, 1997

• Many studies since then supporting or disputing solar/GCR - cloud correlation

• Not independent - most use the same ISCCP satellite cloud dataset

• No firm conclusion yet - requires more data - but, if there is an effect, it is likely to be restricted to certain regions of globe and at certain altitudes & conditions

• Eg. correlation (>90% sig.) of low cloud amount and solar UV/GCR,1984-2004:

32

Usoskin et al, GRL 2006

+ + + + + + + + + + + + + + + + +

+ + + + +

fair weather

current

(J ~ 2.7 pA/m2)

ionosphere (>90 km altitude)

surface

thunderstorm

current generators

(40 flashes /s)

~1400A

~105 !

(fair weather

value)

~200 !

+250 kV

0 V

0.7 F

" = RC ~2min

33

• Cosmic rays ionise atmosphere and control Earth-ionosphere conductivity

• Large aerosol charges at cloud boundaries => unipolar space charge region

• Can be entrained inside clouds and may affect:‣ Rate of aerosol accretion by cloud droplets‣ Ice particle formation‣ Atmospheric dynamics

• Largest ionisation in polar regions; many observations of cloud and T/P changes caused by Forbush decreases, solar disturbances, magnetic sector crossings...

Global electrical circuit

++ + +++

++ + ++ ++

+++

+

++ +

++

+++

++ +++

__

_

_

_

__

_

_

++

+

+

+

+

+

__ _

_ _ __

__

_

__ _

___

__ _+

+

+

_+ region of net -

positive charge

region of net -

negative charge

cloud layer

(low conductivity)charge accumulation

bipolar ion-

environment

aerosol

scavenging

bipolar ion-

environment

Electric field

fair weather

field

cloud-perturbed

field

25

0 m

fair weather current

~ 2 pA m -2

highly charged

aerosols

V. CLOUD experiment at CERN

34

CLOUD Collaboration31 May 2009

Austria:University of Innsbruck, Institute of Ion Physics and Applied PhysicsUniversity of Vienna, Institute for Experimental Physics

Bulgaria:Institute for Nuclear Research and Nuclear Energy, Sofia

Estonia:University of Tartu, Department of Environmental Physics

Finland:Helsinki Institute of Physics and University of Helsinki, Department of PhysicsFinnish Meteorological Institute, HelsinkiUniversity of Kuopio, Department of PhysicsTampere University of Technology, Department of Physics

Germany:Goethe-University of Frankfurt, Institute for Atmospheric and Environmental SciencesLeibniz Institute for Tropospheric Research, Leipzig

Portugal:University of Lisbon, Department of Physics

Russia:Lebedev Physical Institute, Solar and Cosmic Ray Research Laboratory, Moscow

Switzerland:CERN, Physics DepartmentFachhochschule Nordwestschweiz (FHNW), Institute of Aerosol and Sensor Technology, BruggPaul Scherrer Institute, Laboratory of Atmospheric Chemistry

United Kingdom:University of Leeds, School of Earth and EnvironmentUniversity of Reading, Department of MeteorologyRutherford Appleton Laboratory, Space Science Department

United States:California Institute of Technology, Division of Chemistry and Chemical Engineering

cloud

CLOUD collaboration

35

• 19 institutes from Europe, Russia and USA

• 14 atmospheric institutes+ 5 space/CR/particle physics

• CLOUD-ITN network of 10 Marie Curie fellows: 8 PhD students + 2 postdocs

1 nm 1 µm 1 mm 1 m 1 km 1000 km

cloud system

cloud scale

microphysical scale

laboratory

ground/aircraft

satellite

observational scale

molecule

/ion

raindropCCN cloud parcel

cloud droplet

/ice particle

critical

cluster

Cloud observational scales

36

CLOUD

CLOUD method

• aerosol chamber + state-of-the-art analysing instruments in CERN PS beamline

• laboratory expts. under precisely controlled conditions (T, trace gases, aerosols, ions)

• study aerosol nucleation & growth; and cloud droplet & ice particle microphysics - with and without beam

37

P. Minginette

humidifier

3.5 GeV/c !+

electrostatic

precipitator

4kV +HV (0-20kV)

-HV (0-20kV)

beam hodoscope

UV lamp

array (254nm)

UV collimator

sampling

probes

03 generatorultrapure air system

2m aerosol chamber

field

cage

HV electrode

air

mixing

station

liq.N2dewar

(500 l)

liq.O2dewar

(500 l)SO2

air

mixing

station

liq.N2dewar

(500 l)

liq.O2dewar

(500 l)SO2

chemical ionisation mass

spectrometer (CIMS;H2SO4)

atmospheric ion

spectrometer(AIS)

condensation particle

counter (CPC) battery

Gerdien condenser

SO2, O3 analysers

T, P, UV, H20

measurements

scanning mobility

particle sizer (SMPS)

38

CLOUD-06

• Beam tests of pilot CLOUD experiment at CERN PS in Oct/Nov 2006• Aims:‣ Technical input for CLOUD design‣ First physics measurements (H2SO4 ion-induced nucleation)

aerosol no.

concentration

(3-6 nm diameter)

H2SO4 vapour

concentration

90 min

delay

Fiedler, Arnold, Kulmala et al, ACP 2005 (Hyytiälä, Finland)

0.1 pptv

H2SO4

2000 /cm3

121086420Time [hr]

100008000600040002000

0

Conc. [c

m-3

]

1

10

Mobili

ty d

iam

ete

r [n

m]

1

10

AIS (-)

AIS (+)

3.5

3.0

2.5

2.0

1.5

1.0

log(d

N/d

log(D

p ) [cm

-3])

CPC threshold 3 nm 3 nm 5 nm 7 nm 9 nm

Aerosol bursts

• Bursts of aerosol particle production growing to CCN size in few hours observed throughout troposphere

• Associated with H2S04

production, but at very low concentrations

• Not yet understood:

‣ Extra vapours (NH3, VOC)?

‣ Ion-induced nucleation?

39

CLOUD-06 measurements:

Kulmala et al:

6000

5000

4000

3000

2000

1000

0Part

icle

concentr

ation [cm

-3]

840-4-8Time [h]

100

200

0

Beam

inte

nsity

[kH

z]

33

5

7

9

1

8

2 3

45 6

7

run 35

CP

C thre

shold

[nm

]

10-5

10-4

10-3

10-2

10-1

100

Nucle

ation r

ate

, J

3 [cm

-3 s

-1]

250200150100500

Beam intensity [kHz]

SO2 [ppb] 0.3-0.6 1.5 6

CLOUD-06 results

• Results of pilot CLOUD run:

‣ validated the basic experimental concept of CLOUD

‣ suggestive evidence for ion-induced nucleation of H2SO4-H2O under atmospheric conditions

‣ provided important technical input for CLOUD design

40

CLOUD-09 design requirements• Large chamber:‣ Diffusion lifetime of aerosols/trace gases to walls ~L2

‣ Dilution lifetime of makeup gases ~L3

=> 3m chamber has typically 5-10 hr lifetimes

• Ultra-clean conditions:‣ Condensable vapours, eg. [H2SO4] ~0.1 pptv‣ Ultrapure air supply from cryogenic liquids‣ UHV procedures for inner surfaces, no plastics

• Temperature stability and wide T range‣ 0.1oC stability‣ Fibre-optic UV system for photochemistry‣ -90C → +100C range

• Field cage up to 30 kV/m:‣ Zero residual field

• Particle beam‣ Wide beam for ~uniform exposure

• Comprehensive analysers (measure “everything”, as for collider detectors...)‣ Mass spectrometers for H2SO4, organics, aerosol composition

41

100

80

60

40

20

0

CP

C 3

02

5 [#

cm

-3]

Time [hrs]

50

40

30

20

10

0

CP

C 3

01

0 [

# c

m-3

]

27

26

25

Te

mp

era

ture

[°C]

24201612840Time [hrs]

1.2

0.8

0.4

0.0

SO

2 [

pp

b]

40

30

20

10

0

O3 [

pp

b]

40

20

0

RH

[%]

3 nm CPC 3025 7 nm CPC 3010

CLOUD-09 chamber at CERN

42

P. Minginette

CLOUD plans• 2009:‣ commission CLOUD-09‣ study H2SO4-H2O nucleation with and without beam‣ reproducibility of nucleation events‣ PTR-Mass Spect. to measure organics at 10 pptv level‣ new ion-TOF Mass Spect. for ion characterisation

• 2010:‣ commission thermal system (-90C → +100C)‣ study H2SO4/water + volatile organic compounds with

and without beam‣ temperature dependence (effect of altitude)

• 2011-2013:‣ extend studies to other trace vapours, and to cloud

droplets & ice particles (adiabatic expansions in chamber)

43

Conclusions

• Climate has continually varied in the past, and the causes are not well understood - especially on the 100 year timescale relevant for today’s climate change

• Strong evidence for solar-climate variability, but no established mechanism. A cosmic ray influence on clouds is a leading candidate

• CLOUD at CERN aims to study and quantify the cosmic ray-cloud mechanism in a controlled laboratory experiment

• The question of whether - and to what extent - the climate is influenced by solar/cosmic ray variability remains central to our understanding of anthropogenic climate change

44