21ST CENTURY CHANGES IN THE NORDIC CLIMATE: UNCERTAINTIES DERIVED FROM …/21st century... · 2011-03-02 · uncertainties derived from an ensemble of regional climate model simulations

21ST CENTURY CHANGES IN THE NORDIC CLIMATE: UNCERTAINTIES DERIVED FROM AN ENSEMBLE OF

REGIONAL CLIMATE MODEL SIMULATIONS

Seasonal mean temperature, precipitation and wind speed over the Nordic

region are analysed in an ensemble of 16 regional climate model (RCM)

simulations for 1961-2100. To construct the ensemble we have used the

Rossby Centre RCM with boundary conditions from seven global climate

models (GCMs) under four emission scenarios. Most of the simulations were

downscaling experiments of GCMs forced by the emission scenario SRES

A1B (Nakićenović and Swart, 2000). One of the GCMs was run three times

under A1B differing only in initial conditions. The Rossby Centre ensemble

and the more large-scale European climate change signal has earlier been

documented by Kjellström et al. (2010).

Erik Kjellström*, Grigory Nikulin, Ulf Hansson, Gustav Strandberg and Anders UllerstigSwedish Meteorological and Hydrological Institute, S-601 76 Norrköping, Sweden,

*e-mail: [email protected]

50A1BIPSL-CM4 (IPSL, France)20

25low (Q3)19

50high (Q16)18

50low (Q3)17

50A1Bref (Q0)HadCM3 (Hadley Centre, UK)

16

50B115

12.514

2513

5012

5011

50A1B10

50A2ECHAM5 (MPI-met, Germany)9

50B28


50B26

50A1B5

50A2CCSM3 (NCAR, USA)4

253

50A1BBCM (NERSC, Norway)2

50A1BArpège (CNRM, France)1

Horisontal resolution (km)Emission scenarioAOGCM (Institute, country)

No

50A1BIPSL-CM4 (IPSL, France)20

25low (Q3)19

50high (Q16)18

50low (Q3)17

50A1Bref (Q0)HadCM3 (Hadley Centre, UK)

16

50B115

12.514

2513

5012

5011

50A1B10


50B28


50B26

50A1B5

50A2CCSM3 (NCAR, USA)4

253

50A1BBCM (NERSC, Norway)2

50A1BArpège (CNRM, France)1

Horisontal resolution (km)Emission scenarioAOGCM (Institute, country)

No

Different AOGCMs

Different initial conditions

Different modelformulation (GCM)

Different emissionscenarios

Different horizontalresolution

All simulations on the ENSEMBLES grid

with RCA3

The Baltic Sea Iceland

0 1 2 3 4 5 6−2

0

2

4

6

8

10

12

14

A1 B1C1

E1

F1

G1

H1

I1

J1K1

L1

M1N1

O1

P1

A2

B2C2

E2

F2

G2

H2

I2

J2

K2

L2

M2

N2

O2

P2

A3

B3C3

E3

F3

G3

H3

I3 J3

K3

L3

M3

N3

O3

P3

A −ECHAM5−r1B −ECHAM5−r2C −ECHAM5−r3D −HADCM3−lowE −HADCM3−refF −HADCM3−highG −BCMH −CNRMI −CCSM3J −IPSLK −CCSM3L −ECHAM4M −ECHAM5−r1N −CCSM3O −ECHAM4P −ECHAM5−r1

Temperature change (oC)

Win

d sp

eed

chan

ge (

%)

0 1 2 3 4 5 6−8

−6

−4

−2

0

2

4

A1B1

C1

E1

F1

G1

H1

I1

J1

K1

L1

M1N1

O1

P1

A2B2

C2 E2

F2

G2

H2

I2

J2

K2

L2

M2

N2

O2

P2

A3

B3

C3 E3

F3

G3

H3

I3

J3

K3

L3

M3

N3

O3

P3



Win

d sp

eed

chan

ge (

%)

0 1 2 3 4 5 6−4

−2

0

2

4

6

A1

B1

C1

E1

F1

G1

H1I1

J1

K1

L1

M1

N1

O1

P1

A2

B2C2

E2F2

G2

H2

I2

J2

K2

L2

M2

N2

O2

P2

A3B3C3

E3

F3

G3

H3I3

J3

K3

L3

M3N3

O3

P3



Win

d sp

eed

chan

ge (

%)

Nordic mainland

Figure 3. Change in annual mean wind speed versus change in T2m relative to 1961-1990. The letters

indicate which driving GCM that has been used. The numbers for which time period the change is calculated:

(1) 2011-2040, (2) 2041-2070 and (3) 2071-2100. Colours indicate emission scenarios (A2-red, A1B-black,

B2-blue, B1-green). The grey line is a least-square fit to the data.

A changing wind climate?

Figure 1. Change in annual mean precipitation versus change in T2m relative to 1961-1990. The letters indicate

which driving GCM that has been used. The numbers for which time period the change is calculated: (1) 2011-

2040, (2) 2041-2070 and (3) 2071-2100. Colours indicate emission scenarios (A2-red, A1B-black, B2-blue, B1-

green). The grey line is a least-square fit to the data.

Changing precipitation

Annual mean Summer (JJA)Winter (DJF)

Figure 2. Change in P-E versus change in T2m relative to 1961-1990 for the Nordic mainland. The letters

indicate which driving GCM that has been used. The numbers for which time period the change is calculated:

(1) 2011-2040, (2) 2041-2070 and (3) 2071-2100. Colours indicate emission scenarios (A2-red, A1B-black,

B2-blue, B1-green). The grey line is a least-square fit to the data.

A wetter climate?

0 1 2 3 4 5 6

0

5

10

15

20

25

30

A1

B1

C1

D1

E1F1

H1

I1

J1

M1

O1

P1

A2

B2

C2

D2

E2

F2

H2

I2

J2

M2

O2

P2

A3

B3

C3

D3

E3

F3

H3

I3

J3

M3

O3

P3


P−

E c

hang

e (%

)

A A1B−ECHAM5−r1B A1B−ECHAM5−r2C A1B−ECHAM5−r3D A1B−HADCM3−lowE A1B−HADCM3−refF A1B−HADCM3−high

H A1B−CNRMI A1B−CCSM3J A1B−IPSL

M A2−ECHAM5−r1

O B2−ECHAM4P B1−ECHAM5−r1

0 2 4 6 8 10

0

5

10

15

20

25

30

35

A1

B1

C1

D1

E1

F1H1

I1

J1

M1

O1

P1

A2

B2

C2

D2E2F2

H2

I2

J2M2

O2

P2

A3

B3

C3

D3E3

F3

H3

I3 J3

M3

O3

P3


P−

E c

hang

e (%

)



M A2−ECHAM5−r1


0 1 2 3 4

0

20

40

60

80

100

120

140

160

A1

B1

C1

D1E1

F1H1

I1

J1

M1O1

P1A2

B2C2

D2

E2

F2H2

I2

J2

M2O2

P2

A3B3C3

D3

E3F3

H3

I3

J3

M3

O3

P3


P−

E c

hang

e (%

)



M A2−ECHAM5−r1


0 1 2 3 4 5 6−5

0

5

10

15

20

25

30

A1

B1

C1

D1E1F1

G1H1

I1

J1

K1 L1

M1

N1

O1

P1

A2

B2

C2

D2E2

F2

G2

H2

I2

J2

K2

L2

M2N2

O2

P2

A3 B3

C3

D3E3

F3

G3

H3

I3

J3

K3

L3M3

N3

O3

P3


Pre

cipi

tatio

n ch

ange

(%

) A A1B−ECHAM5−r1B A1B−ECHAM5−r2C A1B−ECHAM5−r3D A1B−HADCM3−lowE A1B−HADCM3−refF A1B−HADCM3−highG A1B−BCMH A1B−CNRMI A1B−CCSM3J A1B−IPSLK A2−CCSM3L A2−ECHAM4M A2−ECHAM5−r1N B2−CCSM3O B2−ECHAM4P B1−ECHAM5−r1

The Baltic Sea IcelandNordic mainland

0 1 2 3 4 5 6−5

0

5

10

15

20

25

30

A1B1C1D1

E1

F1

G1

H1

I1

J1

K1

L1

M1

N1

O1

P1

A2B2

C2

D2

E2

F2

G2H2

I2

J2

K2

L2

M2

N2

O2

P2

A3B3

C3

D3

E3

F3

G3

H3

I3

J3

L3

M3

N3O3

P3


Pre

cipi

tatio

n ch

ange

(%


0 1 2 3 4 5 6−5

0

5

10

15

20

25

30

A1

B1

C1

D1E1F1

G1H1

I1

J1

K1

L1

M1

N1

O1

P1

A2

B2

C2

D2

E2

F2G2

H2

I2

J2

K2

L2

M2N2

O2

P2

A3

B3

C3

D3

E3

F3

G3

H3

I3

J3

K3

L3

M3

N3

O3

P3


Pre

cipi

tatio

n ch

ange

(%


The Rossby Centre ensemble

AcknowledgementsPart of the simulations and subsequent analysis work has been funded by the European ENSEMBLES (GOCE-CT-2003-505539) project, the Nordic

Climate and Energy Systems (CES) project and the Swedish Mistra-SWECIA programme funded by Mistra (the Foundation for Strategic

Environmental Research). The model simulations were made on the climate computing resource Tornado funded with a grant from the Knut and Alice

Wallenberg foundation. The institutes providing the global model data used as boundary conditions are kindly acknowledged. We are grateful for the

E-OBS dataset from the EU-FP6 project ENSEMBLES (http://www.ensembles-eu.org) and the data providers in the ECA&D project

(http://eca.knmi.nl).

ReferencesKjellström, E., Nikulin, G., Hansson, U., Strandberg, G. and Ullerstig, A., 2011. 21st century changes in the European climate: uncertainties derived

from an ensemble of regional climate model simulations. Tellus, 63A(1), 24-40. DOI: 10.1111/j.1600-0870.2010.00475.x

Nakićenović, N., and R. Swart (eds.), 2000. Special Report on Emissions Scenarios. A Special Report of Working Group III of the Intergovernmental

Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 599 pp.

The ensemble is used to; i) evaluate the simulated Nordic climate against observed

climatologies for 1961-1990, ii) assess future climate change and iii) illustrate

uncertainties in future climate change related to natural variability, boundary

conditions and emissions.

� Biases in temperature and precipitation in the 1961-1990 period are strongly

related to errors in the large-scale circulation in the GCMs (Fig. 5).

� Statistically significant increases in temperature are seen for all of the Nordic

region already in the next decades (Fig. 4).

� Precipitation increases in northern Europe and most so in winter (Figs. 1 & 4).

� Generally wetter conditions in the future (Fig. 2).

� Wind speed changes are generally small. Increase in wind speed are found in

parts of the northern seas and decreases in many other areas (Figs. 3 & 4).

� Uncertainty largely depends on choice of GCM. This strong dependency is

related to their representation of changes in the large-scale circulation.

� The uncertainty related to forcing (i.e. emission scenario) is most important by

the end of the century while natural variability sometimes dominates the uncertainty

on local to regional scales in the nearest few decades.

Summary

CTL

-20-16-12

-8-4048

1216

oC

Tem

pera

ture

SCN1 - CTL SCN2 - CTL SCN3 - CTL

0 1 2 3 4 5 6 7 8

oC

RCA3(6GCMs) DJF | CTL: 1961-1990 | SCN1: 2011-2040 | SCN2: 2041-2070 | SCN3: 2071-2100

0153045607590

105120150200300

Pre

cipi

tatio

n

mmmonth

-60 -40 -20 0 20 40 60 %

23456789

1011

m/s

Win

d

100210061010

10141018

1022

-1-1

1

1

1

22

-1 -0.5 0 0.5 1 1.5 2 m/s

CTL

05

10

15

20

25

3035

oC

Tem

pera

ture

SCN1 - CTL SCN2 - CTL SCN3 - CTL

0 1 2 3 4 5 6

oC

RCA3(6GCMs) JJA | CTL: 1961-1990 | SCN1: 2011-2040 | SCN2: 2041-2070 | SCN3: 2071-2100

0153045607590

105120150200300

Pre

cipi

tatio

n

mmmonth

-60 -40 -20 0 20 40 60 %

12

3

4

5

6

78

m/s

Win

d

10131013

1013

10151017

-0.5

-0.5

-0.50.5

-1 -0.5 0 0.5 1 1.5 2 m/s

Figure 4. Six-member A1B ensemble (marked in orange in the table) means of T2m (top), P (middle), and W10m and

MSLP (lower) for 1961-1990 (left) and the respective changes in 2011-2040, 2041-2070 and 2071-2100 compared

to 1961-1990 in the three rightmost columns. Only differences significant at the 5% significance level are shown.

For MSLP all changes are shown but statistically significant differences are indicated with thick lines.

Winter (DJF) Summer (JJA)

Gradual changes with 6 different AOGCMs under A1B

Figure 5. (top, left) Mean T2m (colour, °C, E-OBS) and MSLP (hPa, ERA40) and biase s of T2m and MSLP in the

RCA3 simulations driven by (top, middle) the ERA40 reanalysis, (middle and bottom rows) six different AOGCMs

and (top, right) their ensemble mean. Only T2m differences significant at the 5% significance level are shown.

Correspondingly, statistically significant differences for MSLP are indicated with thick lines.

E-OBS

10061010

1014

1018

1018

-20-16-12

-8-4048

1216

oC

2m temperature Winter (DJF) 1961-1990RCA3 (ERA40)

-2

-2

2

2

RCA3 (6 GCMs)

-2

-2

2

2

2

-2-2

2 2

RCA3 (ECHAM5)

-2

-2

-2

-2

-2

-2

2

2

2

2

RCA3 (HADCM3)

-2

-2

2

2

-2-2

-2

RCA3 (BCM)

-10

-6

-2

-2

-2

-2

2

2

2

2

6 6

6

-10

-6

-2

6

RCA3 (CCSM3)

-10

-6

-6

-2

-2

2

6

6

6

-10-6

-6

-26

RCA3 (CNRM)

-2

-2

2

2

2

2

2

2 2 2

2

2

2

6

6

2

2

22

6RCA3 (IPSL)

-2

-2

2

2

2

6

6

6

6

-2

26

6

-6 -4 -2 0 2 4 6

oC RCA3(ERA40, GCMs) - E-OBS

E-OBS

101210141016

0

6

12

18

24

30

oC

2m temperature Summer (JJA) 1961-1990RCA3 (ERA40)

2

2

2 2

2

2

2

2

2

2

2

RCA3 (6 GCMs)

-2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

RCA3 (ECHAM5)

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2

2 2

2

2RCA3 (HADCM3)

-6

-2

-2

-2

-2

-2

-2

-2 -2

-2

-2

-2 -2

RCA3 (BCM)

-2

-2

-2

-2

-2

-2

-2

-2

-2

-2

-2

2

2

6

6

6

-2

-2

2

6

RCA3 (CCSM3)

-2

2

2

2

2

2

2

2

2

2

2

2

2

22

2RCA3 (CNRM)

2

2

2

2

2

2

2

2

2

2

2

6

6

6

6

6

6

6

22

2 6

RCA3 (IPSL)

2

2

6

10

10

2

2

6

-6 -4 -2 0 2 4 6

oC RCA3(ERA40, GCMs) - E-OBS

Biases in the control periodWinter (DJF) Summer (JJA)

Documents

21ST CENTURY CHANGES IN THE NORDIC CLIMATE: UNCERTAINTIES DERIVED FROM …/21st century... · 2011-03-02 · uncertainties derived from an ensemble of regional climate model simulations