1 Use of Ocean Surface Winds in NCEP’s Global Data Assimilation System Stephen J. Lord Director NCEP Environmental Modeling Center NCEP: “where America’s.

1

Use of Ocean SurfaceWinds in NCEP’s

Global Data AssimilationSystem Stephen J. Lord

DirectorNCEP Environmental Modeling Center

NCEP: “where America’s climate, weather, and ocean services begin”

2

Overview

• Satellite data used in NWP and NWP applications

• The NASA-NOAA-DOD Joint Center for Satellite Data Assimilation– JCSDA-sponsored data impact studies– Impact of QuikSCAT and Windsat (L. Bi et al,

U. Wisconsin and JCSDA)– Improved use of surface wind observations

3

The Environmental Forecast Process

Observations

Analysis

Model Forecast

Post-processed Model Data

Forecaster

User (public, industry…)

NumericalForecastSystem

Data Assimilation

4

Satellite data used in NCEP’soperational data assimilation systems

• HIRS sounder radiances• AMSU-A sounder radiances• AMSU-B sounder radiances• GOES sounder radiances• GOES, Meteosat, GMS

winds• GOES precipitation rate• SSM/I precipitation rates• TRMM precipitation rates• SSM/I ocean surface wind

speeds• ERS-2 ocean surface wind

vectors

• Quikscat ocean surface wind vectors

• AVHRR SST• AVHRR vegetation fraction• AVHRR surface type• Multi-satellite snow cover• Multi-satellite sea ice• SBUV/2 ozone profile and

total ozone• AIRS• MODIS Winds • Altimeter sea level

observations (ocean data assimilation and wave data assimilation system)

5

POES Data Delivery00Z Average 1B Data Counts

0

0.5

1

1.5

2

0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00

(Mill

ion

s)

Dump Time - Cycle Time (Hours)

Ave

rag

e R

epo

rt C

ou

nt

HIRS-2 (1bhrs2) HIRS-3 (1bhrs3) AMSU-A (1bamua) AMSU-B (1bamub) MSU (1bmsu)

Loca

tions

Rec

eive

d (M

)

GFSData

Cutoff

NAMData

Cutoff

Next-generationSatellite Data Delivery

6

NCEP Forecast Systems andMission Applications

• Forecast Systems– Global Forecast System (GFS) & GDAS– North American Model (NAM) & RDAS– Rapid Update Cycle (RUC)– Global Ensemble System (GEN)– Short-Range Ensemble Forecast System (SREF)– Air Quality Forecast System– Hurricane System (HUR)*– Real Time Ocean Forecast System (RTOFS)– Global and Regional Wave System (WAV)– Ice Drift System (ICE)– Climate Forecast System (CFS)

* System does not have associated data assimilation system

7Five Order of Magnitude Increase in Satellite Data Over Next Ten YearsFive Order of Magnitude Increase in Satellite Data Over Next Ten YearsFive Order of Magnitude Increase in Satellite Data Over Next Ten YearsFive Order of Magnitude Increase in Satellite Data Over Next Ten Years

Count

(Mill

ions)

Daily Satellite & Radar Observation Count

20001990 2010 2010-10%of obs

2002 100 M obs

NPOESS Era Data Volume

2003-4 125 M obs

Level 2 radar data 2 B

2005 210 M obs

8

NASA-NOAA-DOD Joint Center for Satellite Data Assimilation

(JCSDA)– NOAA, NASA, DOD partnership– Mission

• Accelerate and improve the quantitative use of research and operational satellite data in weather and climate prediction models

– Current generation data– Prepare for next-generation (NPOESS, METOP,

research) instruments

– Supports applied research• Partners• University, Government and Commercial Labs

9

• February 2001– SSM/I precipitation estimates in physical initialization (preparation for

TRMM data)• May 2001

– Inclusion of cloud liquid water in model and data assimilation• October 2001

– TRMM TMI precipitation estimates added to physical initialization• January 2002

– QuikSCAT data added (3-8% improvement in 10 m winds vs. mid-latitude deep ocean buoys at 24-96 h; 7-17% improvement for MSLP)

• October 2002– Preparation for AIRS (upgraded OPTRAN, cloud detection, data

thinning algorithms)• June 2005

– AIRS data added (center spot, reduced channels)• November 2005

– MODIS winds added

Research Data Added to NCEP Operational Atmospheric Data Assimilation

2001-2005

10

Data Assimilation Impacts in the NCEP GDAS

(cont)

AMSU and “All Conventional” data provide nearly the same amount of improvement to the Northern Hemisphere.

N. Hemisphere 500 mb AC Z 20N - 80N Waves 1-20

15 Jan - 15 Feb '03

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Forecast [days]

An

om

aly

Co

rrel

atio

n '

control

no amsu

no conv

11

N. Hemisphere 1000 mb htanomaly correlation

N. Hemisphere 1000 mb AC Z 20N - 80N Waves 1-20

15 Jan - 15 Feb '03

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Forecast [days]

An

om

aly

Co

rrel

atio

n

'

control

no amsu

N. Hemisphere 1000 mb AC Z 20N - 80N Waves 1-20

15 Jan - 15 Feb '03

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Forecast [days]

An

om

aly

Co

rrel

atio

n '

control

no hirs

AMSU: 0.5 day improvement at 5 days

12

Jung and Zapotocny

JCSDAFunded by

NPOESS IPO

Satellite data ~ 10-15% impact

Impact of Removing AMSU, HIRS, GOES Wind, Quikscat Surface Wind Data on Hurricane Track Forecasts in the Atlantic Basin - 2003 (34 cases)

-20.0

-15.0

-10.0

-5.0

0.0

5.0

10.0

15.0

12 24 36 48 72 96 120

Forecast Hour

% Im

prov

emen

t NOAMSU

NOHIRS

NOGOESW

NOQuikscat

Impact of Removing AMSU, HIRS, GOES Wind, Quikscat Surface Wind Data on Hurricane Track Forecasts in the East Pacific Basin - 2003 (24 cases)

-60.0

-50.0

-40.0

-30.0

-20.0

-10.0

0.0

10.0

20.0

30.0

12 24 36 48 72

Forecast Hour

% Im

pro

vem

ent

NOAMSU

NOHIRS

NOGOESW

NOQuikscat

Better

Worse

Worse

Better

EPAC

ATL

Impact ofREMOVING

SatelliteData

NOTStatisticallySignificant

13

Assimilating and determining the impact of sea surface winds measured by WindSat/Coriolis data in the Global

Forecast System

Li Bi

Tom Zapotocny

John Le Marshall

Michael Morgan

James Jung

31 May 2006

14

Goals of the Study

• Run GFS with QuikSCAT (cntrl254)

• Run GFS without QuikSCAT (noqscat254)

• Run GFS with Windsat & QuikSCAT

• Study statistical properties of QuikSCAT and Windsat products

15

N. Hemisphere 500 hPa AC Z 20N - 80N Waves 1-20

1 Jan - 15 Feb '04

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

0 1 2 3 4 5 6 7

Forecast [days]

An

om

aly

Co

rrel

atio

n '

Control Noqscat

S. Hemisphere 500 hPa AC Z 20S - 80S Waves 1-20

1 Jan - 15 Feb '04

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

0 1 2 3 4 5 6 7

Forecast [day]

An

om

aly

Co

rrel

atio

n '

Control Noqscat

16

N. Hemisphere 500 hPa AC Z 20N - 80N Waves 1-20

1 Jan - 15 Feb '04

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

0 1 2 3 4 5 6 7

Forecast [days]

An

om

aly

Co

rrel

atio

n '

Control Windsat

S. Hemisphere 500 hPa AC Z 20S - 80S Waves 1-20

1 Jan - 15 Feb '04

0.6

0.65

0.7

0.75

0.8

0.85

0.9

0.95

1

0 1 2 3 4 5 6 7

Forecast [day]

An

om

aly

Co

rrel

atio

n '

Control Windsat

17

Tropical WindsTropics 850 hPa AC V 20N - 20S Waves 1-20

1 Jan - 15 Feb '04

0.4

0.5

0.6

0.7

0.8

0.9

1

0 1 2 3 4 5 6 7

Forecast [day]

An

om

aly

Co

rrel

atio

n '

Control Windsat & QuikSCAT

18

WindSat and QuikSCAT Wind FieldsWindSat QuikSCAT

http://www.npoess.noaa.gov/polarmax

19

JCSDA Community Radiative Transfer Model (CRTM)

• Upgrades for 2006-2007– Major Science upgrades available for immediate testing & further development

• Scattering by clouds• Surface optics• Multiple stream (impacts surface emissivity and reflection)• Aerosol absorption and scattering (Weaver, JCSDA AO)• Preparation for advanced instruments (IASI, CrIS, etc)

– Laying foundation for advanced applications (2006-2010)• Begin assimilation of cloudy radiances• Requires major

– Computing and human resources for complete evaluation of impact – Evaluation and upgrades to forecast model (for forecast cloud properties)– Bias correction and QC development (partial cloudiness, etc)

– Code structure and performance• Execution efficiency, memory footprint• To prepare for new absorption models (e.g. OSS) from JCSDA AO investigators (e.g. AER)• Increase flexibility for future changes• Establish and refine testing procedures (offline and in GSI)

20

Final Comments

• Surface Vector Winds (SVWs) are not a major driver of NWP skill

• Nonetheless, SVWs provide a useful supplement to sounding data and other wind retrievals for specific ocean phenomena (e.g. hurricanes)

• Forecaster use of SVWs is a major consideration

• Preliminary results QuikSCAT appears to be a better instrument than Windsat

• Future SVW capability should match QuikSCAT capability

21

ThanksQuestions?

22

Planning for FY09Integration and Testing of New

Observations1. Data Access (routine, real time) 3 months2. Formatting and establishing operational data base 1 month3. Extraction from data base 1 month4. Analysis development (I) 6-18 months5. Preliminary evaluation 2 months6. Quality control 3 months7. Analysis development (II) 6-18 months8. Assimilation testing and forecast evaluation 1 month9. Operational implementation 6 months10. Maintain system* 1 person “till death do us part”

* Scientific improvements, monitoring and quality assurance

Total Effort: 29-53 person months per instrument

23

Facilitating Steps• Continue to increase support for

– Computing– Community-based data assimilation and model advances at NOAA and

NASA• Begin to support

– Altimetry and surface wind instruments and data assimilation– Quality control (Operations)– Use of imagery and tracers (e.g. ozone) as proxy for direct wind

observations– “Coupled” data assimilation of atmosphere, land, ocean

• Increase prioritization and planning efforts for FY09– Understanding observing system impacts– New instrument classes

• Unique measurements• Cover under observed aspects of atmosphere, ocean, land

– Atmospheric winds– Coastal ocean data assimilation– Air quality & atmospheric monitoring– Land Surface data assimilation with direct use of radiances

– Example• Wind lidar

24

Doppler Wind Lidar (DWL) Impact

Conv Only

Conv. + TOVS

Conv + TOVS + DWL(best)

Conv + DWL(non-scan)

Conv + DWL(PBL )

Conv + TOVS + DWL(non-scan)

Conv +DWL(Best)

Conv + DWL(Upper)

V at 200 hPa

V at 850 hPa

4

-4

4

-4

8

8

00

0

Forecast hour

%

%Wind Anomaly Correlation

Differences from Conv. Data Only

TOVS +Best DWL

TOVSonly

TOVS +Best DWL

TOVSonly

25

500mb 5 Day Global Forecasts

40

45

50

55

60

65

70

75

80

85

90

1980 1990 2000 2010

Year

An

om

aly

Co

rre

lati

on

NH GFS

SH GFS

NH Reanalysis

SH Reanalysis

Impact of Observations and Numerical Forecast SystemTechnology Growth on Global Forecasts

Obsonly

NFSTech

Growth+ Obs

NFS TechGrowth:

ComputingData Assim.

ModelsEnsembles

26

ECMWFImprovement in medium-range forecast skill

NFSTech

Growth+ Obs

12-month running mean anomaly correlation (%) of 500hPa height forecasts

Obsonly

27

Current Satellite Data Assimilation Development (cont)

• Improved use of satellite data for SST analysis– Improved AVHRR QC and bias correction (Xu Li, A. Harris)– Addition of simplified ocean mixed-layer model (EMC-MMAB, GMAO)– Use of microwave instruments (e.g. AMSR-E)

• Upgrades to ozone assimilation– GOME and current NASA, NOAA instruments (CPC, JSDI; Stajner, GMAO, AO)

• Land surface data assimilation– Use of GMAO Catchment model as multi-Land Surface Model (LSM) system

(together with Noah, VIC and Sacramento LSMs)– Collaboration on advanced Ensemble Kalman Filter (EKF) techniques

• Ocean data assimilation– Use of altimeter data (EMC, Behringer)– Impacts on S/I forecasting (EMC, Behringer)– GMAO uses Poseidon isopycnal model but will test developments in MOM-4

• Observing system design and impacts– Analysis adjoint diagnostic tools– Observing System Simulation Experiments (OSSEs) for

• Understanding interaction between observing system and DA system • Defining potential impact of and preparing for future instruments

28

Global Data AssimilationObservations Processing

• Definitions– Received: The number of observations received operationally

per day from providers (NESDIS, NASA, Japan, Europeans and others) and maintained by NCEP’s Central Operations. Counted observations are those which could potentially be assimilated operationally in NCEP’s data assimilation system. Observations from malfunctioning instruments are excluded.

– Selected: Number of observations that is selected to be considered for use by the analysis (data numbers are reduced because the intelligent data selection identifies the best observations to use). Number excludes observations that cannot be used due to science deficiencies.

– Assimilated: Number of observations that are actually used by the analysis (additional reduction occurs because of quality control procedures which remove data contaminated by clouds and those affected by surface emissivity problems, as well as other quality control decisions)

29

Global Data AssimilationObservations Processing (cont)

2002 July2005

Notes November 2005

Operations

Received 123 M 169.0M Nov. 2005 increase attributed to additional AIRS, MODIS winds,

NOAA-18 and NOAA-17 SBUV data

236.1 M

Selected 19 M 23.6 M 26.9 M

Assimilated 6 M 6.7 M 8.1 M

30

Current Satellite Data Assimilation Development at the JCSDA

• Community Radiative Transfer Model (CRTM)– NESDIS/ORA leads scientific development– EMC transitions development to operations & maintains operational codes– GMAO focuses on applications to NASA instruments used in research DA

systems– Examples of CTRM applications

• AIRS• MODIS• WindSat• SSM/IS• AMSR• OMI• ATMS• IASI• CrIS• OMPS

• JCSDA partnership for COSMIC – Project management (NESDIS)– Data delivery, formatting (UCAR, NCEP Central Ops)– Scientific algorithms and QC (JCSDA, NESDIS, UCAR)– Testing with CHAMP data prior to launch with DA system (JCSDA, EMC, UCAR)