217

Appendix 1 CMIP3 Models Used in

Climate Projections

Blowholes, Tongatapu

218 Climate Change in the Pacific: Scientific Assessment and New Research | Volume 1: Regional Overview

For many of the climate projections presented in Chapters 6 and 7, CMIP3 data were either analysed directly (e.g. temperature, rainfall) or used to calculate a climate variable (e.g. surface relative humidity, potential evapotranspiration) or diagnostic (e.g. Genesis Potential Index, aragonite saturation state) of interest (Table A1). However, many research institutions participating in CMIP3 did not provide data for all the model variables and SRES emission scenarios requested by the project organisers. As such, data availability is inconsistent for different variables, emission scenarios (i.e. A2, A1B and B1) and output frequencies (i.e. daily or monthly).

Table A1. CMIP3 model variables required to provide information on each PCCSP variable / climate diagnostic.

PCCSP variable / climate diagnostic Required model output (CMIP3 naming convention)

Surface air temperature Surface air temperature (tas)

Daily maximum surface air temperature Daily maximum surface air temperature (tasmax)

Daily minimum surface air temperature Daily minimum surface air temperature (tasmin)

Rainfall Precipitation (pr)

Surface wind Calculated from zonal surface wind speed (uas) and meridional surface wind speed (vas)

Solar radiation Surface downwelling shortwave radiation (rsds)

Surface relative humidity Calculated from surface specific humidity (huss), sea level pressure (psl), surface altitude (orog) and surface air temperature (tas) data

Potential evapotranspiration Calculated from surface downwelling shortwave radiation (rsds), surface air temperature (tas), surface altitude (orog), (calculated) surface relative humidity and precipitation (pr) data (Morton, 1983)

Sea surface temperature Sea surface temperature (tos)

Sea surface salinity Salinity (so)

Ocean temperature Sea water potential temperature (thetao)

Ocean salinity Salinity (so)

Ocean density Calculated from potential temperature (thetao) and salinity (so) data

Ocean circulation Calculated from eastward seawater velocity (uo) and northward seawater velocity (vo)

Buoyancy frequency squared Calculated from potential temperature (thetao) and salinity (so) data

Steric sea level Calculated from potential temperature (thetao) and salinity (so) data

Dynamic component of sea level Sea surface elevation (zos)

Genesis potential index Calculated from sea surface temperature (tos), sea level pressure (psl), air temperature (ta), specific humidity (hus), zonal wind speed (ua) and meridional wind speed (va) data

Curvature vorticity parameter Calculated from air temperature (ta), specific humidity (hus), zonal wind speed (ua) and meridional wind speed (va) data

CSIRO direct detection method Requires sea level pressure (psl), air temperature (ta), zonal wind speed (ua), meridional wind speed (va) data

Aragonite saturation state (Ωar) Requires sea surface temperature (tos) and sea surface salinity (so) data

Information on the CMIP3 model data used in providing information on each PCCSP climate variable / diagnostic is therefore provided here (Table A2, Table A3, Table A4). It should be noted that:

• For any given variable and timescale, where data are available for one or more of the A2, A1B or B1 scenarios, data are also available for the corresponding ‘Climate of the Twentieth Century’ (20c3m) simulation.

• Due to high data archiving costs, daily GCM data are typically only available for the periods 1980–1999, 2046–2065 and 2080–2099. No such limitations apply for monthly data.

• For certain variables and emission scenarios, modelling groups provided data for multiple simulations of the same model, using slightly different initial conditions. The information outlined here indicates the models for which at least one simulation is available.

• The CMIP3 data are freely available, along with detailed documentation, from the Program for Climate Model Diagnosis and Intercomparison at Lawrence Livermore National Laboratory (www-pcmdi.llnl.gov).

219Appendix 1: CMIP3 Models Used in Climate Projections

Table A2. CMIP3 models used in calculating atmospheric projections requiring daily timescale data.

Temperaturea Rainfall Tropical cyclonesb

A2 A1B B1 A2 A1B B1 CVP CDD GPI

1 BCCR-BCM2.0

2 CCSM3

3 CGCM3.1 (T47)

4 CGCM3.1 (T63)

5 CNRM-CM3

6 CSIRO-Mk3.0

7 CSIRO-Mk3.5

8 ECHAM5/MPI-OM

9 ECHO-G

10 FGOALS-g1.0

11 GFDL-CM2.0

12 GFDL-CM2.1

13 GISS-AOM

14 GISS-EH

15 GISS-ER

16 INGV-SXG

17 INM-CM3.0

18 IPSL-CM4

19 MIROC3.2 (hires)

20 MIROC3.2 (medres)

21 MRI-CGCM2.3.2

22 PCM

23 UKMO-HadCM3

24 UKMO-HadGEM1

Total models used 11 15 15 14 16 16 4 10 14

Pale blue cells denote those models and scenarios used in calculating projections. Italicised model names indicate those models excluded due to unacceptable performance in simulating the current climate (Section 5.5.1)aRefers to both the daily maximum and minimum surface air temperaturebThe Curvature Vorticity Parameter (CVP), CSIRO Direct Detection (CCD) scheme and Genesis Potential Index (GPI) were only calculated for the A2 emission scenario.

220 Climate Change in the Pacific: Scientific Assessment and New Research | Volume 1: Regional Overview

Table A3. CMIP3 models used in calculating atmospheric projections requiring monthly timescale data.

Model nameSurface air

temperature, rainfallSurface wind Solar radiation

Surface relative humidity, potential evapotranspiration

A2 A1B B1 A2 A1B B1 A2 A1B B1 A2 A1B B1

1 BCCR-BCM2.0

2 CCSM3

3 CGCM3.1 (T47)

4 CGCM3.1 (T63)

5 CNRM-CM3

6 CSIRO-Mk3.0

7 CSIRO-Mk3.5

8 ECHAM5/MPI-OM

9 ECHO-G

10 FGOALS-g1.0

11 GFDL-CM2.0

12 GFDL-CM2.1

13 GISS-AOM

14 GISS-EH

15 GISS-ER

16 INGV-SXG

17 INM-CM3.0

18 IPSL-CM4

19 MIROC3.2 (hires)

20 MIROC3.2 (medres)

21 MRI-CGCM2.3.2

22 PCM

23 UKMO-HadCM3

24 UKMO-HadGEM1

Total models used 15 18 17 13 16 15 15 18 16 9 11 10

Pale blue cells denote those models and scenarios used in calculating projections. Italicised model names indicate those models excluded due to unacceptable performance in simulating the current climate (Section 5.5.1).

221Appendix 1: CMIP3 Models Used in Climate Projections

Tab

le A

4. C

MIP

3 m

odel

s us

ed in

cal

cula

ting

ocea

nic

pro

ject

ions

req

uirin

g m

onth

ly t

imes

cale

dat

a.

Mo

del

nam

eS

ea s

urfa

ce

tem

per

atur

eS

ea s

urfa

ce s

alin

itya

T, S

, den

sity

, b

uoya

ncy,

st

eric

sea

leve

lb

Dyn

amic

sea

leve

lC

ircul

atio

nΩ

ar

A2

A1B

B1

A2

A1B

B1

A1B

A1B

A1B

Allc

1B

CC

R-B

CM

2.0

2C

CS

M3

3C

GC

M3.

1 (T

47)

4C

GC

M3.

1 (T

63)

5C

NR

M-C

M3

6C

SIR

O-M

k3.0

7C

SIR

O-M

k3.5

8E

CH

AM

5/M

PI-

OM

9E

CH

O-G

10FG

OA

LS-g

1.0

11G

FDL-

CM

2.0

12G

FDL-

CM

2.1

13G

ISS

-AO

M

14G

ISS

-EH

15G

ISS

-ER

16IN

GV-

SXG

17IN

M-C

M3.

0

18IP

SL-

CM

4

19M

IRO

C3.

2 (h

ires)

20M

IRO

C3.

2 (m

edre

s)

21M

RI-

CG

CM

2.3.

2

22P

CM

23U

KM

O-H

adC

M3

24U

KM

O-H

adG

EM

1

Tota

l mo

del

s us

ed14

1616

1115

1311

1315

6

Pal

e bl

ue c

ells

den

ote

thos

e m

odel

s an

d sc

enar

ios

used

in c

alcu

latin

g pr

ojec

tions

. Ita

licis

ed m

odel

nam

es in

dica

te th

ose

mod

els

excl

uded

due

to u

nacc

epta

ble

perfo

rman

ce in

sim

ulat

ing

the

curr

ent c

limat

e (S

ectio

n 5.

5.1)

a Mod

el n

umbe

rs a

re g

reat

er fo

r se

a su

rface

sal

inity

than

for

salin

ity th

roug

h th

e de

pth

of th

e oc

ean,

as

PC

CS

P re

sear

cher

s ap

proa

ched

CM

IP3

mod

ellin

g gr

oups

requ

estin

g ad

ditio

nal

surfa

ce s

alin

ity d

ata.

b Ref

ers

to th

e po

tent

ial t

empe

ratu

re (T

), sa

linity

(S),

pote

ntia

l den

sity

and

buo

yanc

y fre

quen

cy s

quar

ed th

roug

h th

e de

pth

of th

e oc

ean.

c The

sam

e m

odel

s w

ere

used

for

the

A2,

A1B

and

B1

scen

ario

s.Ω

ar =

Ara

goni

te s

atur

atio

n st

ate.