From ERA-15 to ERA-40 and ERA-Interim · ERA group 8 8 6 3 1 0 0 0 0 0-12-11-8-2 0 0 0 0 0 1 75°S...

Sakari Uppala

ECMWF

With contributions by colleagues

From ERA-15 to ERA-40 and ERA-Interim

CONTENTS

Operations Reanalysis Climate

ERA-15 ERA-40

Towards the ERA-Interim

Future reanalysis plans at ECMWF

ReanalysisOperations

To produce high quality forecasts with early delivery to customers, once only

To produce, with regular intervals, time series of climate quality synoptic analyses in good physical and dynamical balance over a long period

Aim

Proven observations used. All observations monitored, new observing systems introduced passively.

Manual control through blacklist. Changes made based on daily monitoring

Observations, including reprocessed and recovered, with the experience from previous reanalyses and operations

Blacklist defined for the whole reanalysis period, based on the previous reanalyses

Use of observations

High resolution system, updated frequently, 3-4 times/ year

Operational system configured in lower resolution and kept unchanged as far as possible

Data assimilation system

ReanalysisOperations

Operations performed one day per day using GTS observations and data from satellite producers

Reanalysis performed ~ 10 days per day using historical pre-prepared observations and boundary fields as input

Reanalysis team << Operations team

Therefore reanalysis has a greater need to develop automated processes to identify when assimilation has a problem and adaptive algorithms to handle errors

Operational forecast performance 1980-2006

monthlymoving averageNorthern Hemisphere

500 hPa geopotentialANC reaching 60%

Southern Hemisphere

500 hPa geopotentialANC reaching 60%

Tropics

850 hPa wind vectorABC reaching 70%

ERA-15 ERA-40

Extract from the ECMWF amended convention

.

.

2. The objectives of the Centre shall be:

a) to develop, and operate on a regular basis, global models and data-assimilation systems for the dynamics, thermodynamics and composition of the Earth's fluid envelope and interacting parts of the Earth-system, with a view to:

i. preparing forecasts by means of numerical methods;ii. providing initial conditions for the forecasts; andiii. contributing to monitoring the relevant parts of the Earth-system;

b) to carry out scientific and technical research directed towards improving the quality of these forecasts;

c) to collect and store appropriate data;

.

.

ReanalysisObservational

Single source: radiosondetemperatures, satellite radiances, T2m ,Ps, precipitation, snow, …etc

Multiple sources: Temperature, wind, humidity, pressure, radiance data. Model background used as an extra observation. Physical relationships and error statistics play important role.

Information used in the analysis

Quality controlled and bias corrected mean observations or observations interpolated into fields, often monthly, for use in climate assessments and climate model validations

Synoptic analyses, integrals of physical processes, quality controlled observations and their departures from the background. Huge application potential including data sparse polar regions.

Products

CLIMATE

Univariate Multivariate

Analysis

Used as a separate product Through physical parameterization of the assimilating model SST and ICE have a large influence on the products and also affect the bias corrections and quality control of data

Sea surface temperature and Sea ice dataset

Excellent mean agreement by construction Optimize the fit to all observation types simultaneously constrained by physical knowledge

Analysis agreement with observations

ReanalysisObservationalCLIMATE

Results published and distributed widely

Gridded data, indices and some observational data available on-line

ECMWF reanalyses and products available online at lower resolution and in full resolution for the member states

NCEP observations and reanalyses available worldwide

Dissemination

ReanalysisObservationalCLIMATE

Observation biases corrected using nearby observations

Outside observed areas biases due to interpolation method

Model background used

-To correct known observational biases

-To tune satellite radiances

Possible model biases affect the analyses, but are marginal over data dense areas

Products & biases

1000

100

10000

100000

10

Number of ERA-15 users

Num

ber o

f ERA

-40

user

s

Reanalysis “team” resources

20061997

ECMWF reanalyses

ERA-40 1957-2002

ERA-15 1979-1993

Improved data assimilation system- Assimilating model T106L31 T159L60 - OI 3D-Var FGAT- Analysis of O3

Greatly extended use of satellite dataERA-15 experience ERA-40 blacklistMore comprehensive use of conventional observationsUse of Meteosat reprocessed winds, CSR passiveImproved SST & ICE datasetOcean wave height analysis

Model levels

ERA-15/ L31

ERA-40/ L60

65 km

32 km

SST anomaly, HADISST until 1981 November, NCEP 2d-Var then on

monthly(ship, buoy) weekly (satellite, ship, buoy)

Global wave climatology atlasS. Caires, A. Sterl, G. Komen and V. Swail

http://www.knmi.nl/onderzk/oceano/waves/era40/atlas.html

1971 - 2000 90th percentile of significant wave height February

m

VTPRVTPR1973

TOMS/ SBUVTOMS/ SBUV1979

1979TOVS:TOVS:

HIRS/ MSU/ SSUHIRS/ MSU/ SSUCloud Motion WindsCloud Motion Winds

1987SSM/ISSM/I 1991

ERSERS--111995ERSERS--22

1998ATOVS:ATOVS:AMSUAMSU--AA

METEOSAT METEOSAT ReprocessedReprocessed

Cloud Motion Cloud Motion WindsWinds

1982 1988

CONVENTIONAL SURFACE AND UPPERAIR OBSERVATIONSNCAR/ NCEP, ECMWF, JMA, US Navy, Twerle, GATE, FGGE, TOGA, TAO, COADS, …

AIRCRAFT DATAAIRCRAFT DATA1973

1957 2002 Observing Systems in ERA-40

METEOSAT Reprocessed Winds

Use of atmospheric satellite data in reanalysesUse of atmospheric satellite data in reanalyses

1c

1c

1c

-

NESDIS operational T & q retrievals

SSU SSM/IAMSUMSUHIRSVTPR

GEODMSPVTPR/ TOVS/ ATOVS

Oper+reprocessedAMWs, CSR

passively

1c radiances and 1D-Var retrievals of

rainy radiances 1c1c1cNA

ERA-Interim

1989

Oper+reprocessedAMWs

JMA retrievals of TCWV1c1c1cNA

JRA-25

1979

Oper+reprocessedAMWs, CSR

passively

1D-Var retrievals of TCWV & wind

speed1c1c1c1c

ERA-40

1957-2002

Oper AMWs-NA1D-Var retrievals of T & q

using CCR. Above 100hPa NESDIS retrievals.

NAERA-15

1979-1993

Oper AMWs-NCEP

1948

Observation biases

Data assimilation assumes observation errors to be unbiased

Bias correction of radiosonde temperatures

Bias tuning of satellite radiances

Conventional•Surface•Upperair•Aircraft•Satellite products

TOVS radiances

SSMI radiances

AIRS radiances

O3 data

Bias

mod

ules

NWP systemNWP system

Observations and observation equivalents

from the model and analysis

in database

Quality anddeparture

information

Input observations How the observations were used in the analysis

(feedback)

Conventional feedback•Surface•Upperair•Aircraft•Satellite products

TOVS feedback

SSMI feedback

AIRS feedback

O3 feedback

(Input and feedback observations in BUFR code)

Radiosonde temperature bias OB-FG(1994, South West Canada)

Withoutcorrection

Withcorrection

(4 solar elevation angle intervals and the mean)

analysis

background

VTPR FGGE

ERA-40 winds validated against rocketsonde winds at Ascension Island

25km

30km

35km

45km

Need to homogenize radiosonde biases in time

(example: Haimberger, 2005 using ERA-40 feedback data)

SAIGON / TAN-SON-NHUT 00UTC 200hPa temperature (Background – Observation)

Corrected

Russian sonde VaisalaFrench sonde

SAIGON / TAN-SON-NHUT 00UTC 200hPa temperature (Background – Observation)

Corrected

Russian sonde VaisalaFrench sonde

ERA-40 1957-2002

ERA-Interim 1989 to continue as CDAS

Experimental program to decide the DA configuration

- Period August 1999 December 20003D-Var FGAT6 hour 4D-Var12 hour 4D-VarStatic <-> Adaptive radiance bias tuning (Dick Dee)Model version T159L60 with significant upgrades since ERA-40

- Technical development of the monitoring environment- Preliminary runs 1989

passive radiance assimilation,1989 1997 …

ERA-40 1957-2002


Several performance measures

- The Hydrological cycle (Per Kållberg)

- The Age of air (Beatriz Monge-Sanz, University of Leeds)

- Forecast performance

- Time series of Observations-Background departures (Dick Dee)

- Detection of Tropical cyclones

- Analysis increments

Population: 366,366,366,366,366,366,366,366,366,366 (averaged)Mean calculation method: standard

Date: 20000101 12UTC to 20001231 12UTCN.hem Lat 20.0 to 90.0 Lon -180.0 to 180.0

Anomaly correlation forecast500hPa Geopotential

Mean curves

1 2 3 4 5 6 7 8 9 10

Forecast Day0

10

20

30

40

50

60

70

80

90

100

0387

0453

0455

0457

oper

ERA-403D-Var

4D-Var 12h

4D-Var 6h

OPER


Date: 20000101 12UTC to 20001231 12UTCS.hem Lat -90.0 to -20.0 Lon -180.0 to 180.0


Mean curves

1 2 3 4 5 6 7 8 9 10

Forecast Day0

10

20

30

40

50

60

70

80

90

100

0387

0453

0455

0457

oper

ERA-403D-Var

4D-Var 12h

4D-Var 6h

OPER




Mean curves

1 2 3 4 5 6 7 8 9 10

Forecast Day0

10

20

30

40

50

60

70

80

90

100

0387

0469

0470

ERA-404D-Var 12h Static

4D-Var 12h Adaptive

StaticAdaptive

Averaged Zonal Temperature

0.01

0.02

0.03

0.050.070.1

0.2

0.3

0.50.7

1

2

3

57

10

20

30

5070

100

200

300

500700

1000

Pres

sure

(hPa

)

60 levels 91 levels

1

2

3

4

5

6

78910

12

14

161820

25

30

3540455055606570

91

Leve

l num

ber

ERA-40/ L60

65 km

ERA-Interim/ L91

81 kmModel levels

ERA-15/ L31

ERA-40/ L60

65 km

32 km

Data assimilation system CY31R1• 12 hour 4D-Var • T255L91• Wavelet Jb• New humidity analysis• Improved model physics

ERA-40 1957-2002


Use of observations• ERA-40 and ECMWF operational observations the basic input source

• Satellite level-1c radiances

Better RTTOV and improved use of radiances especially IR

Assimilation of clear radiances and 1d-retrievals of rain affected radiances from SSM/I

Adaptive bias correction

• Improved use of radiosondes

Bias correction and homogenization based on ERA-40 and experimental runs (Leopold Haimberger)

• Correction of SHIP/ SYNOP surface pressure biases

• Use of reprocessed Meteosat winds

• Use of GOME profile data from RAL

• New set of Altimeter wave height data 1991 (Jean Bidlot)

ERA-40 1957-2002


Mean analysisMean backgroundforecasts

Mean of Analysis increment (Analysis – Background) Standard Deviation of analysis increment

Mean analysisMean backgroundforecasts

Mean of Analysis increment (Analysis – Background) Standard Deviation of analysis increment new formulation of humidity analysis

Total Column Water Vapour (kgm-2)ERA-40 versus the new humidity analysis

Tropical oceans

Mean differences between ERA-40 and ERA-15

T2m (K) July 1989 10m wind speed January 1989

Contour interval 0.5ms-1

Yellow/red indicates ERA-40 windier than ERA-15

Contour interval 2KYellow/red indicates ERA-40

warmer than ERA-15

2

2

2

2

1

1

1

0

-3

-3

-2 -1

-1

-1

-1

02

2

2

2

1

1

1

0

-3

-3

-2 -1

-1

-1

-1

0

30°S30°S

15°S 15°S

0°0°

15°N 15°N

30°N30°N

135°E

135°E 150°E

150°E 165°E

165°E 180°

180° 165°W

165°W 150°W

150°W 135°W

135°W 120°W

120°W 105°W

105°W

ERA40 monthly/daily climate mean 198901differences to1001 monthly/daily mean 198901screen level windspeed

m/s

-5

-4

-3.5

-3

-2.5

-2

-1.5

-1

-0.50.5

1

1.5

2

2.5

3

3.5

4

5

ERA group

8

8

6

3

1

0

0

0

0

0

-12

-11

-8

-2

0

0

0

0

0

8

8

6

3

1

0

0

0

0

0

-12

-11

-8

-2

0

0

0

0

0

75°S

60°S

180°165°W150°W135°W

30°W 15°W 0° 15°E 30°E 45°E

60°E

75°E

90°E

105°E

120°E

135°E

150°E165°E

ERA40 monthly/daily climate mean 198906differences to1001 monthly/daily mean 198906screen level temperature

°C

-24-22-20-18-16-14-12-10-8-6-4-224681012141618202224

group

T2m (K) June 1989 10m wind speed January 1989

Contour interval 0.5ms-1

Yellow/red indicates ERA-Interim windier than ERA-40

Contour interval 2KYellow/red indicates ERA-

Interim warmer than ERA-40

Mean differences between ERA-Interim and ERA-40




Mean curves

1 2 3 4 5 6 7 8 9 10Forecast Day

0

20

40

60

80

100

0256

OPER

a0060

ERA

ERA-40

ERA-Interim test

ERA-15




Mean curves

1 2 3 4 5 6 7 8 9 10

Forecast Day0

20

40

60

80

100

0256

OPER

a0060

ERA

ERA-40

ERA-Interim test

ERA-15

Population: 188,279,279,279,279,279,279,279,279,279,279 (averaged)Mean calculation method: standard

Date: 19890101 12UTC to 19891231 12UTCTropics Lat -20.0 to 20.0 Lon -180.0 to 180.0

Root mean square error forecast850hPa Vector Wind

Mean curves

0 1 2 3 4 5 6 7 8 9 10

Forecast Day0

1

2

3

4

5

6

0256

OPER

a0060

ERA

ERA-40

ERA-Interim test

ERA-15

ERA-Interim testERA-40

Time series of daily D+7 forecast anomaly correlationsSouthern Hemisphere

Could start in 2010 depending on resources

~ 1940

Important components

Recovery, organization and homogenization of observations

Improved SST & ICE dataset

Variational analysis technique aimed for reanalysis

Comprehensive adaptive bias handling

Handling of model biases

ERA-InterimERA-70?

!!

0

10

20

30

40

50

60

USA UKGer

many

Swizerla

ndFran

ce

Netherl

andsItaly

Austra

liaDen

mark

Japan

Norway

Austri

aCan

ada

China

IndiaSwed

enFinlan

d

South

Africa

Spain

Banglad

esh

Icelan

dTaiw

an

Other

Europe

anOth

er

User feedbackWeb search

Survey on the use of ERA-40 from webhttp://www.ecmwf.int/research/era/era40survey/

Currently more than6500 unique users have accessed the

dataserver

ERA-40 Atlas Top 20: Feb-June 2006

0

100

200

300

400

500

600

700

800

900

2m temperature

Mean sea level pressure

Total precipitation

10m vector wind with isotachs

Evaporation minus precipitation

OrographyVector wind with isotachs at 200hPa

Net surface fluxes of solar radiation - positive downwards

Vector wind with isotachs at 850hPa

Net surface fluxes of heat - positive downwards

Zonal mean zonal wind - tropospheric perspective

Land Sea Mask

Potential temperature and vector wind at 2pvu

Column integrated water vapour

Column integrated fluxes of water vapour with their convergence

Zonal mean zonal wind - stratospheric perspective

Column integrated heating

Vertical velocity [omega] at 500hPa

The NAO Index.

Surface fluxes of latent heat - positive downwards

Number of unique users requesting images : 1406

From ERA-15 to ERA-40 and ERA-Interim · ERA group 8 8 6 3 1 0 0 0 0 0-12-11-8-2 0 0 0 0 0 1 75°S...

Documents

Transcript of From ERA-15 to ERA-40 and ERA-Interim · ERA group 8 8 6 3 1 0 0 0 0 0-12-11-8-2 0 0 0 0 0 1 75°S...