Climate change projections for Vietnam from CMIP5 simulations Ramasamy Suppiah 29 November 2012.

Climate change projections for Vietnam from CMIP5 simulations

Ramasamy Suppiah 29 November 2012

Climate Change in Vietnam | Suppiah Ramasamy2 |

Why we need this study?Vietnam’s long coastline, geographic location, and complex topography and monsoonal climates contribute to its being one of the most hazard-prone countries of southeast Asia, with storms and flooding, and drought in particular, responsible for economic and human losses.

It is expected that anthropogenic climate change conditions can change the severity of climatic extremes in the future

How good are the new generation of global climate models (GCMs) simulating the present climate over a broader region centered Vietnam?

What are the projected changes in mean climate (temperature and rainfall) over Vietnam by the end of the century?

Data


Observed global surface air temperature, winds and mean sea level pressure data from ERA-Interim archives. The data set covers land and ocean areas

Observed land only global rainfall from Climate Research Unit (CRU) and ERA-Interim archives for land and ocean areas.

Observed gridded precipitation (Aphrodite)

Simulated mean sea level pressure, surface air temperature and rainfall for historical and future climates for Representative Concentration Pathways (RCP) for RCP8.5 from Coupled Model Intercomparison Project phase 5 (CMIP5) simulations

15 Organisations and 31 simulations.

Historical climate 1979-2004 and also 1979-2001 for CRU and future climate 2080-2099 for RPC8.5.Future temperature and rainfall changes for RCP8.5 are from 24 and 22 models, respectively.


Methods and SeasonsStatistical analyses:Pattern correlation and root mean square (RMS) errors

Four Seasons: Northeast monsoon season: December to March (NEM)First Intermonsoon season: April and May (FIM)Southwest monsoon season: June to September (SWM)Second Intermonsoon season: October and November (SIM)

Pattern correlation Perfect = 1.0, Worst -1.0RMS Error small is good and high is worst


Some thoughts about GCMs

Simulation of the current climatology

Capture the interannual and possibly intra-seasonal variabilityEl Niño-Southern OscillationTropical cyclonesOnset, retreat and duration of the monsoonsLocation of the Inter-tropical Convergence Zone (ITCZ)

GCMs are the best available tool to simulate future climate

Comparison of CO2 concentrations from SRES (A1B, A1FI, A2, B1) and RCPs (3.0, 4.5, 6.0, 8.5) approaches


Source: Meinshausen, M. and Coauthors, 2011

RCP 8.5

RCP 4.5

Winds at 850 hPa level for First Intermonsoon (FIM) and Southwest monsoon (SWM) seasons

Presentation title | Presenter name7 |

FIM (April-May) SWM (June-September)

Winds ms-1

Latit

ude

Latit

ude

Longitude Longitude

Winds at 850 hPa level for Second Intermonsoon (SIM) and Northeast monsoon (NEM) seasons


SIM (October-November) NEM (December-March)

Winds ms-1

Latit

ude

Latit

ude

Longitude Longitude

Mean sea level pressure for First Intermonsoon (FIM) and Southwest monsoon (SWM) seasons



(hPa) (hPa)

Latit

ude

Latit

ude

Mean sea level pressure for Second Intermonsoon (SIM) and Northeast monsoon (NEM) seasons



(hPa) (hPa)

Latit

ude

Latit

ude

Mean rainfall during First Intermonsoon (FIM) and Southwest (SWM) monsoon seasons



mm/day mm/day

Latit

ude

Latit

ude

Mean rainfall during Second Intermonsoon (SIM) and Northeast (NEM) monsoon seasons



mm/day mm/day

Latit

ude

Latit

ude


Pattern correlation and RMS Error for surface air temperature for four seasons based on the larger region

ACCESS1-0ACCESS1-3CanESM2CCSM4CNRM-CM5CSIRO-Mk3-6-0FGOALS-g2FGOALS-s2GFDL-CM3GFDL-ESM2MGISS-E2-HHadCM3HadGM2-CCHadGM2-ESinmcm4IPSL-CM5A-LRIPSL-CM5A-MRMIROC4hMIROC5MIROC-ESM-CHEMMIROC-ESMMPI-ESM MRI-CGCM3NorESM1-M

RMS Error (oC)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

TAS-IMS (Apr-May)

RMS Error (oC)

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

TAS-OCT

RMS Error (oC)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

TAS-NEMS (Nov-Mar)

RMS Error (oC)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

TAS-SWMS (Jun-Sep)

Surface air temperature for Region (90-140°E,0-40°N)

Best

Worst


Pattern correlation and RMS Error for surface air temperature for four seasons based on the larger region

Surface air temperature for Region (60-160°E,-15-50°N)

ACCESS1.0

ACCESS1.3

bcc-csm1-1

CanESM2

CCSM4

CNRM-CM5

CSIRO-Mk3-6-0

FGOALS-g2

FGOALS-s2

GFDL-CM3

GFDL-ESM2G

GFDL-ESM2M

GISS-E2-H

GISS-E2-R

HadGEM2-CC

inmcm4

IPSL-CM5A-LR

IPSL-CM5A-MR

MIROC5

MIROC-ESM-CHEM

MIROC-ESM

MPI-ESM-LR

MRI-CGCM3

NorESM1-M

RMS Error (oC)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

TAS-FIMS (Apr-May)

RMS Error (oC)

1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

TAS-SIMS (Oct-Nov)

RMS Error (oC)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

TAS-NEMS (Dec-Mar)

RMS Error (oC)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

TAS-SWMS (Jun-Sep)

Best

Worst


Pattern correlation and RMS Error for mean sea level pressure for four seasons based on the larger region


Mean sea level pressure for Region (90-140°E,0-40°N)

RMS Error (hPa)

0 1 2 3 4 5 6 7 8 9 10

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PSL-IMS (Apr-May)

RMS Error (hPa)

0 1 2 3 4 5 6 7 8 9 10

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PSL-OCT

RMS Error (hPa)

0 1 2 3 4 5 6 7 8 9 10

Pa

tter

n C

orr

ela

tio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PSL-NEMS (Nov-Mar)

RMS Error (hPa)

0 1 2 3 4 5 6 7 8 9 10

Pa

tter

n C

orr

ela

tio

n

-0.4

-0.3

-0.2

-0.1

0.00.10.20.30.40.50.60.70.80.91.0

PSL-SWMS (Jun-Sep)

Best

Worst


Pattern correlation and RMS Error for mean sea level pressure for four seasons based on a relatively smaller region


Mean sea level pressure for Region (95-135°E,0-25°N)

RMS Error (hPa)

0 1 2 3 4 5 6 7 8 9 10

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PSL-IMS (Apr-May)

RMS Error (hPa)

0 1 2 3 4 5 6 7 8 9 10

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PSL-OCT

RMS Error (hPa)

0 1 2 3 4 5 6 7 8 9 10

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PSL-NEMS (Nov-Mar)

RMS Error (hPa)

0 1 2 3 4 5 6 7 8 9 10

Pat

tern

Co

rrel

atio

n

-0.3

-0.2

-0.1

0.00.10.20.30.40.50.60.70.80.91.0

PSL-SWMS (Jun-Sep)

Best

Worst


Pattern correlation and RMS Error for rainfall for four seasons. Results against ERA-Interim data for the larger region


Rainfall for Region (90-140°E,0-40°N) Land and ocean areas

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-NEMS (Nov-Mar)

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-SWMS (Jun-Sep)

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-IMS (Apr-May)

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-OCT

Best

Worst


Pattern correlation and RMS Error for rainfall for four seasons. Results against ERA-Interim data for the Asian region

Rainfall for Region (60-160°E,-15-50°N) Land and ocean areasACCESS1.0

ACCESS1.3

CanESM2

CCSM4

CNRM-CM5

CSIRO-Mk3-6.0

FGOALS-g2

FGOALS-s2

GFDL-CM3

GFDL-ESM2G

GFDL-ESM2M

GISS-E2-H

HadGEM2-CC

inmcm4

IPSL-CM5A-LR

IPSL-CM5A-MR

MIROC5

MIROC-ESM-CHEM

MIROC-ESM

MPI-ESM-LR

MRI-CGCM3

NorESM1-M

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-FIMS (Apr-May)

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-SIMS (Oct-Nov)

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pa

tter

n C

orr

elat

ion

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-NEMS (Dec-Mar)

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pa

tter

n C

orr

elat

ion

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-SWMS (Jun-Sep)

Best

Worst


Pattern correlation and RMS Error for rainfall for four seasons. Results against CRU data based on the larger region


Rainfall for Region (90-140°E,0-40°N) Land area only

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-NEMS (Nov-Mar)

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-SWMS (Jun-Sep)

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-IMS (Apr-May)

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-OCT

Best

Worst


A comparison of annual results from both ERA-Interim and CRU data

ERA-Interim (Land and ocean) CRU Data (Land only)ACCESS1-0ACCESS1-3CanESM2CCSM4CNRM-CM5CSIRO-Mk3-6-0FGOALS-g2FGOALS-s2GFDL-CM3GFDL-ESM2MGISS-E2-HHadCM3HadGM2-CCHadGM2-ESinmcm4IPSL-CM5A-LRIPSL-CM5A-MRMIROC4hMIROC5MIROC-ESM-CHEMMIROC-ESMMPI-ESM MRI-CGCM3NorESM1-M

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR-ANNUAL (Jan.-Dec.)

RMS Error (mm/day)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

Pat

tern

Co

rrel

atio

n

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

PR ANNUAL (Jan.Dec.)

Best

Worst


Observed and simulated annual surface air temperature over the region

Temperature (°C)


Annual Temperature bias: Model-Obs

Temperature (°C)


Mean seasonal and annual temperature bias from 24 simulations

Temperature °C


Future annual surface air temperature changes by 8 of 24 models by the end of the century

Temperature (°C)



Temperature (°C)



Temperature (°C)

Temperature change by the end of the century (°C) for RPC 8.5 from 24 models


First Intermonsoon

Northeast monsoonSecond Intermonsoon

Southwest monsoon Annual

Temperature change (°C)

Mean and ranges of future temperature changes ( °C) over VietnamSeason Low Mid HighNEM 1.6 3.7 5.8FIM 2.1 4.1 5.6SWM 2.2 3.8 5.2SIM 1.9 3.7 5.1ANNUAL 2.1 3.8 5.1


Observed and simulated annual rainfall over the region

Rainfall (mm/day)


Annual rainfall bias: Model-Observation

Rainfall (mm/day)


Mean model rainfall bias between simulations and observations:Model minus observations

Rainfall bias (mm/day)

First Intermonsoon


Southwest monsoonAnnual

Number of models agreeing on future rainfall increase from 22 GCMs


Increase

First Intermonsoon


Southwest monsoon

Annual

Rainfall change by the end of the century (%) for RPC 8.5 from 22 models


Rainfall change (%)Mean and ranges of future rainfall changes (%)Over Vietnam

Season Low Mid HighNEM -40 +04 +70FIM -50 -04 +45SWM -14 +07 +25 SIM -08 +20 +65ANNUAL -08 +07 +20

First Intermonsoon


Southwest monsoon Annual


Conclusions

CMIP5 models fairly well capture the spatial patterns ofseasonal and annual surface air temperature. However, there are large model-model variations. Most of the models tend to underestimate observed air temperature over the land areas.

Current GCMs capture the observed spatial patterns and seasonality of rainfall. More rainfall over low latitudes and less rainfall over mid and high latitudes.

Models tend to overestimate rainfall over low and high latitudes and underestimate over mid latitudes. However, there are small differences inregional scale.

Conclusions Cont.


Models suggest an increase of annual temperature between 2.0 and 5.0 with the mid value of 4.0°C over Vietnam by the end of this century for RPC 8.5

Rainfall changes are more complex as decreases and increases are simulated by models for future

Annual average rainfall changes between -08 and +20% with a mid value of +07%. However, there are differences among the seasons. The largest mean rainfall increase is simulated for Second Intermonsoon season (+20%) and a small decrease is for First intermonsoon season (-04%).

Most of the models agree on rainfall increases during, Northeast, Southwest and Second Intermonsoon seasons and decreases during the first intermonsoon season over Vietnam by the end of this century for RCP 8.5.

Future plan


Investigate temperature and rainfall changes for RCP4.5 and RCP8.5 with additional model experiments.

Investigate the effects of aerosol on temperature and rainfall changes over Southeast Asia by the end of this century.

Suppiah Ramasamy

CSIRO Marine and Atmospheric Research, Aspendale, Vic. 3195t +61 3 9239 4554e [email protected]

Thank you

Climate change projections for Vietnam from CMIP5 simulations Ramasamy Suppiah 29 November 2012.

Documents