Climate Modeling Activity in the MRI

Akio Kitoh

Climate Research Department

Meteorological Research Institute

2010.10.19, WGNE26, Tokyo

MRI

Technical Support

JMA WMO, IOC, etc.

MRI’s role

Forecast Research Climate Research

Typhoon Research

Physical Meteorology Research

Atmospheric Environment and Applied Meteorology Research

Meteorological Satellite and Observation System Research

Seismology and Volcanology Research

Oceanographic Research Geochemical Research

Contribution

Research Programsdisaster mitigation Programs

Reports

etc.

1. Earth System Model (MRI-ESM; MRI-CGCM3)

2. Super-high-resolution Atmospheric Models (AGCM-20km and NHRCM-5km)

2nd Phase (FY1995-1999) 3rd Phase (FY2000-2004) 4th Phase (FY2005-2009)

JMA’s Global Warming Research Project

RCM （ 40km ）

Vol 4（ 2001 ）

Vol 1-3（ 2006-2009 ）

RCM （ 20km ）

Ver 1（ 2003 ）

CGCM2.0A: T42 L30O: 2°×2.5° L23

Vol 5（ 2003 ）

Vol 6（ 2005 ）

Ver 2（ 2004 ）

Vol 7（ 2007 ）

Global A-O Coupled Model

CGCM1A: 4°×5° L15O: 2°×2.5° L21

Climate Model Development at MRI

Earth System Model

Global Warming Projection

ReportJapanese Standard Climate Scenario

IPCC2nd Report

1995

CGCM2.3A: T42 L30O: 2°×2.5° L23

3rd Report2001

4th Report2007

CRCM （ 20km ） NHRCM （ 4km ）

CGCM3.0A: TL95 L48O: 0.5°×1° L50

CGCM3.1A: TL159 L48O: 0.5°×1° L51

RCM （ 20km ）

MASINGAR MRI-CCM

Chemical Transport Model

Regional Climate

Now in the 5th Phase

AR5 2013

OceanOcean

Sea iceSea ice Ocean Biogeochemical Carbon Cycle

Ocean Biogeochemical Carbon Cycle

AtmosphereAtmosphere

Land/VegetationLand/Vegetation

Snow on Ice sheetSnow on Ice sheet RiverRiver

LakeLake

Cloud Microphysics

interaction with aerosols

Cloud Microphysics

interaction with aerosols

MRI Earth System Model

AGCM

OGCM

Land Ecosystem

Carbon Cycle

Land Ecosystem

Carbon Cycle

Coupler ‘Scup’ MRI.COMMRI.COM

GSMUVGSMUV

TL159 (~120km)L48(0.01hPa)

Tripolar1°× 0.5°

Atmos. ChemicalClimate Model

Atmos. ChemicalClimate Model

MRI-CCM2MRI-CCM2

Ozone (Strat. + Tropo.)

Aerosol ChemicalTransport Model

Aerosol ChemicalTransport Model

MASINGARMASINGAR

Sulfate, BC, OC, Dust, Sea-salt

CO2

CO2

MRI-CGCM3MRI-CGCM3

Each component can be coupled with different resolutions

TL95 (~180km)T42 (~280km)

obs model

Surface ozone (ppbv)

2007.05.09 06Z

Tropospheric ozone

Total ozone seasonal change

Ozone: reproducibility and future projectionOzone: reproducibility and future projection

Change in ozone hole area

Ozo

ne h

ole

are

a (

m2 ×

1

0-1

2)

TOMSPRM5

Aerosol optical thicknessAerosol optical thickness

MODIS retrieval: 2001 ～ 2006 mean

Model simulation:2001 ～ 2006 mean

MRI model resolutionsMRI model resolutions

MRI-CGCM2.3(CMIP3: IPCC-AR4)

MRI-CGCM3.1(CMIP5: IPCC-AR5)

MRI-CGCM3.xNext version

AGCMhorizontal resolution

T42 (280km ） TL159 (120km ） TL959 (20km ）

AGCMvertical resolution

L30(top: 0.4hPa)

L48(top: 0.01hPa)

L60(top: 0.01hPa)

OGCMhorizontal resolution

2°(0.5)×2.5°Lat-lon

0.5°×1.0°Tripolar

1/8°×1/12° (10km）

Tripolar

OGCMvertical resolution

L23 L51 ?

Schedule of the CMIP5 simulations with the MRI-CGCM3 and MRI-ESM1

1. Earth System Model (MRI-ESM; MRI-CGCM3)

2. Super-high-resolution Atmospheric Models (AGCM-20km and NHRCM-5km)

Needs for high resolution models for adaptation studies

• representation of topography depends on resolution (land-sea distribution, mountain height, snow-rain threshold, …)

• low resolution models often fail to reproduce precipitation systems such as tropical cyclones, stationary front systems and blocking

• high resolution models have better mean climate

JMA Numerical Analysis and Prediction System

20-km mesh AGCM as the highest resolution climate model

← Climate models for IPCC AR4 (2007) →

JMA extended fcst

JMA seasonal fcst

← Climate models for IPCC AR5 (2013) →

Real topography

HighRes for AR5HighRes climate

model

Future change in NH blocking frequency (JJA)

The higher horizontal resolution is required to accurately simulateEuro-Atlantic blocking. The Euro-Atlantic blocking frequency is predicted to show a significant decrease in the future.

20km

120km 180km

60kmMatsueda

and Palmer

CMIP3 models

wet

dry

CMIP3 models project wet (dry) conditions over north (south)Europe. Robust signals from CMIP3 models.

MRI AGCM

Much weaker signals at high resolution

Precipitation change over Europe (JJA)

180km(climate)

20km(NWP)

similar

robust

less blocking

Matsueda and Palmer

Unreliable?

Indian summer monsoon rainfall

20-km modelIMD observation

Orographic rainfall is

successfully reproduced

Rajendran and Kitoh (2008) Current Science

Time-slice experiments: 20km/60km

• JMA : Operational global NWP model from Nov 2007• MRI : Next generation climate model• Resolution: TL959(20km)/TL319(60km) with 60 layers• Time integration: Semi-Lagrangian Scheme (Yoshimura,

2004)• Cumulus convection: Prognostic Arakawa-Schubert• Three time periods

– Present (1979-2003), Near future (2015-2039), Future (2075-2099)

detrended observed SST (1979-2003)

CGCMs C20C experiments (-2000)

1979 2003 2075 2099

linear trend in CGCMs(2075-2099)

CGCMsSRES A1B experiments (2000-2100) ɢT

linear trend in CGCMs

Å{ Å{É¢T ÅÅ

future change in CGCMs

observed SST for AMIP (1979-2003)

present-day climate future climate

([2075-2099]-[1979-2003])

For 60-km model, ensemble runs• Four different SST anomalies• Three I.C. ensembles each

How to prescribe future SSTs

Use CMIP3 multi-model SST changes

near future

end 21c

HWDI 60-km vs 20-km 20-km model

60-km model ensemble

HWDI: heat wave duration

near future

end 21c

R5d 60-km vs 20-km 20-km model

60-km model ensemble

R5d: greatest 5-day total precipitation no significance test for 20-km model

There is model resolution dependency; 20-km model projects larger increase in

RX5D

Heavy precipitation increases even in near-future but not statistically significant; it is

significant almost everywhere in Asia at the end of the 21st

century

Tropical cyclones

It is likely that future tropical cyclones will become more intense, with larger peak wind speeds and more heavy precipitation associated with ongoing increases of tropical sea surface temperatures.

There is less confidence in projections of a global decrease in numbers of tropical cyclones. [IPCC AR4]

MEXT Kyo-sei Project (FY2002-2006)

and KAKUSHIN Program (FY2007-2011)

using the Earth Simulator by the MRI group

Wind Profile Change （ at max wind speed ）

Ver

tical

p le

vel

Distance from center

・ Large increase in strong wind radius at mid level of troposphere

・ Large change of inner-core wind velocity

Sample Num=1035 Sample Num=937

＃ gray color=no significant difference

unit ： m/s

Streamflow

Decrease 　　　　　　　　　　　Increase

• Increasing flood risk over the Parana river basin

• Decreasing streamflow in the Andean mountains

Future-Present20-km model: Present

Cooperation activities of the MRI groupCooperation activities of the MRI group (by Earth Simulator Earth Simulator computed model outputscomputed model outputs for adaptation studiesfor adaptation studies)

Adaptation study in Coastal Zones of Caribbean countries: Barbados(one, 2005), Belize (one, 2005)

Adaptation studies in Colombian coastal areas, high mountain ecosystems: Colombia (two, 2005; two, 2009)

Adaptation to Climate Impacts in the Coastal Wetlands of the Gulf of Mexico: Mexico (two, 2006)

Adaptation to Rapid Glacier Retreat in the Tropical Andes: Peru (one, 2006; 2010?), Ecuador (one, 2006; one, 2009), Bolivia (one, 2006)

Amazon Dieback: Brazil (two, 2008)

Cooperation under the Cooperation under the JICAJICA (Japan International Cooperation Agency)(Japan International Cooperation Agency) fundsfunds Adaptation studies in agriculture in Argentina: Argentina (three, 2008) Adaptation studies in monsoon Asia: Bangladesh, Indonesia,

Philippines, Thailand, Vietnam (one each, 2008 & 2009) Adaptation studies in Yucatan wetland: Mexico (three, 2009) Adaptation studies in South America: Argentina, Bolivia, Paraguay,

Uruguay (2010 & 2011)

Cooperation under the Cooperation under the World BankWorld Bank funds funds

Other collaborations with India, Korea, Thailand, USA, Spain, …

SST

O

CMIP3 AOGCMs

A

20km,60km AGCM

NHM5km

Time slice experiments

Prediction of regional climate by one-way nested NHM

Lower B.C.

A

Kakushin Team-Extremes Time-Slice ExperimentsKakushin Team-Extremes Time-Slice Experiments

Regional Climate ModelNHM2/1km

Projected SSTNested in the NHM5km

AGCM/NHM are climate model versions of the JMA operational NWP models

Nested in the AGCM20km

JMA NHMHorizontal resolution NHM5km NHM2km

Grid points 669x527x50 767x562 x50

Cumulus parameterization Yes (Kain-Fritsch) No

Cloud physics 1-moment 3-ice bulk scheme 2-moment 3-ice bulk scheme

The spectral boundary coupling (SBC ） method Yes No

Execution period-Main experiment-The preliminary experiment

17th May – 30th Oct. for each yearfor 25 years x 3 series.for 2002-2006 for the perfect boundary experiment using regional analysis data.

1st Jun. – 30th Oct. for each yearfor 25 years x 3 series.Selected months for 2002-2006 using the NHM5km perfect boundary experiment.

Time step 24s 12s

Output (2D) 1 hourly 1 hourly

Output (3D) 3 hourly 3 hourly

Changes in pdf/cdf of precipitation (Jun-Oct)

pdf/cdf of daily precipitation (all Japan) pdf/cdf of hourly precipitation (all Japan)

End 21c

Present

End 21c

Present

Near-future: No change in daily precipitation　　　　　　 Increase of strong hourly precipitation　　　　　　　　 → increase of short-term strong rainEnd 21c: Increase of strong both of daily and hourly precipitation　　　　　　　　　　　　 daily 40% increase for > 150mm/day　　　　　　　　　　　　 hourly 60% increase for > 50mm/hour

Near-futureNear-future

Upward motion (top 3): NHM5km

Mean 3rd Updraft (m/s)

Mean Top3Updraft (m/s)

PF

P F

a)

b)

c)

LCL ：463→468mLCL-LFC ：76→83hPaCAPE: 1139→1538J/kgCIN: 26→30J/kg

DRY

S Nharder to occur

taller convection

Pacific Ocean

Pacific Ocean

Japan Sea

Japan Sea

Japan

Japan

Present

Future

Schematic of changing characteristics

Jet stream will shift southward.

Higher SSTSupply of high e air mass at the low level

Dry

Delay of the northward movement of the Baiu front

Larger convective instability

More intense convections

Increment of intense

precipitation

Relationship between GW and ENSOPrecip Future–Present (Jul) Precip El Niño–La Niña (Jul)

MRI-CGCM2

Comparisons of locations and values of heavy rainfall events

>1000mm/day >500mm/day >300mm/day

Obs

NHM5km

PresentNHM5km

Future

July Precipitation Characteristics: precip ・ wet days ・ SDIIOBS(APHRO) AGCM-20km RCM-5km RCM-2km

# all data are averaged in 20-km grid

We find that RCM-5km is applicable for projections of future changes in daily precipitation, while RCM-2km for hourly precipitation