NOAA 29th Annual Climate Diagnostic & Prediction Workshop, 18 - 22 October 2004, Madison, USA Operational Climate Monitoring from Space The Satellite Application.

NOAA 29th Annual Climate Diagnostic & Prediction Workshop, 18 - 22 October 2004, Madison, USA

Operational Climate Monitoring from SpaceThe Satellite Application Facility on

Climate Monitoring (CM-SAF)

presented by Jörg Schulz, Deutscher Wetterdienst

contributions from:Peter Albert, Steven Dewitte, Annegret Gratzki, Rainer Hollmann,Karl-Göran Karlsson, Terhikki Manninen, Richard Müller, Rob Roebeling, and Werner Thomas

SAF on Climate Monitoring: Visions


Outline

1. Network of Satellite Application Facilities and the CM-SAF

2. Meteosat Second Generation and MetOp

3. Products and Validation

4. Product Usage within Climate Sciences

5. Future Prospects


EUMETSAT's SAF network


The focus on climate aspects requires:

• a long term commitment to ascertain (EUMETSAT) and process

(CM-SAF) data in an

• operational and reliable environment / mode

Data processing and product generation with focus on climate aspects

means:

• generate homogeneous data sets (space and time)

• carefully apply specific verification and validation procedures

in order to achieve:

• quality controlled and

• quality assured products with

• appropriate condensation in space and time

Why a dedicated CM-SAF?


CM-SAF has the mandate to generate thematic climate data records in an operational off-line environment. It requires calibrated and cross calibrated radiance data sets from different satellite operators.

The products come in three major groups:

Cloud parameters (Cloud Fractional Cover, Cloud Top Height, Cloud Type,

Cloud Top Temperature, Cloud Phase, Cloud Optical Thickness, Cloud Water Path)

Radiation budget parameters at the surface and TOA (Surface: Incoming Short Wave Radiation, Net Short Wave Radiation, Outgoing

Long Wave Radiation, Downward Long Wave Radiation, Net Long Wave Radiation, Radiation Budget, Albedo (weekly); TOA: Incoming solar radiative flux, Reflected solar radiative flux, Emitted thermal radiative flux)

Water vapour in the atmosphere (Total and layered precipitable water, temperature, and relative humidity)

Product Groups


I2004

I2007

I2005

I2006

Initial Operations Phase

V2

V3 MSG + EPS

Area Extension

Merging

All products

NOAA

MSG

HCPV1NOAACloudsRadiation

CM- SAF: Schedule & Versioning


Meteosat Second GenerationMetOp as part of the Initial Joint Polar System


- Operational since 29 January 2004

- 12 spectral channels

- Full disk imagery every 15 minutes

- Meteosat-9: launch in 2nd quarter of 2005

- Two more satellites to follow

<== for details see paper in BAMS, 2002

Courtesy of Jo Schmetz

Meteosat-8(first of Meteosat Second Generation)


Chl 4:3.9 µm

Chl 5:6.2 µm

Chl 6:7.3 µm

Chl 7:8.7 µm

Chl 8:9.7 µm

Chl 9:10.8 µm

Chl 10:12.0 µm

Chl 11:13. µm


Eight channels in thermal infrared spectrum


Calibration validation for thermal IR channels

0.2 K

0.4 K 0.3 K

1.5 K

-0.3 K

-0.7 K

0.9 K

-2.00

-1.50

-1.00

-0.50

0.00

0.50

1.00

1.50

2.00

2.50

IR 3.9 IR 6.2 IR 7.3 IR 9.7 IR 10.8 IR 12.0 IR 13.4

Ke

lvin

Comparison to HIRS measurements



MSG-1 GERB first image - 12 December 2002

Total channel Short wavelength channel


Instrument Payload of the Metop Satellites


Products and Validation


Examples - Cloud Fractional Cover

feb04 mar04

apr04 may04


Examples - CFC partially cloudy pixels

feb04 mar04

apr04 may04


Comparison and resulting problems

European Cloud ClimatologyCourtesy of R. Meerkötter, DLR

Main results:• higher cloudiness over water surfaces• satellite results systematically lower than corresponding results from synop. stations - even lower over land

(Possible) reasons:• (much) higher contrast over water • scenery effect () ?• observation rules ? • threshold algorithms exhausted ?

Way out:• (more) physical retrieval ?• further tuning ?• new methods using temporal evolution ?


Examples - CTY - May 6th, 2004

00:45

06:45

12:45

18:45


Cloud Optical Thickness & Cloud Liquid Water Path

Cloud Properties for Cabauw, The Netherlands, 19 April 2004

MSGNOAA

Instantaneous average CLWP

Meteosat-8 (10:00 utc) = 15 gm-2

Noaa-17 (10:08 utc) = 22 gm-2

Daily CLWP Meteosat-8

Average = 73 g.m-2

Std = 78 g.m-2

MSGNOAA


Reflected solar flux March 04 Emitted thermal flux March 04

Examples - Reflected & Emitted TOA flux


Top of atmosphere emitted flux

March 2004

Monthly mean diurnal cycle


Examples - Surface Albedo

July04


Examples - Surface Downward Flux

Feb 04 Mar 04

Apr 04 May 04


HCP processing chain (V3)

ATOVSIASIAIRS

SEVIRITPW3 layer PW

TPW5 layer PW, RH, T6 level T,q

Create Daily average3 km2 sinusoidal projection

Create Daily TPW, LPW, RHfor each source in x km2

sinusoidal projection

Temporal sampling correction MERGER IIsuperimpose mergedvertical structure andTPW onto SEVIRIspatial sampling

AMSU/BMHSSSM/ISSMIS

TPW5 layer PW, T

GRASTPWn layer PWn level T, q

MERGER IOptimum interpolationof daily fields on 45 km grid

Merged products

High resolutionproducts

Q/A

Q/A

Q/A

Q/AQ/A

Q/A

Q/A

Q/A


ATOVS, June 8, 2004

TPW

Temperature @ 850 hPa - 700 hPa Rel. Humidity @ 850 hPa - 700 hPa

LPW @ 850 hPa - 700 hPa

Cloud Fraction (15 (km)2)


SEVIRI only products


Operational meteorological networks of the NMHSs

Special networks as e.g. BSRN, EUREF, GUAN

Research sites as e.g. Cabauw, Sodankylä, Lindenberg, Valencia

Measurement campaigns (Sodankylä April 2004,BBC,BBC2, CNN I & II,Vapic, Lautlos, AMMA)

Product validation data sources


• Vertical albedo profiles 0 – 1000 m above various land use classes

• Surface albedo of three about 10 km * 10 km areas scanned with 1 km spacing with a flight altitude of 450 m and a flight speed of 50 km/h

• Calibration using albedo measurements at Sodankylä mast

aapa mire

boreal forest


Sodankylä, June 2003

16.723.2

15.122.0

13.511.019.1 15.9

55.7

16.4 13.4

24.015.9 15.414.917.016.4

21.216.513.816.315.2

0.0010.0020.0030.0040.00

50.0060.00

70.0080.00

90.00100.00

0.00 5.00 10.00 15.00 20.00 25.00 30.00

UTC time (Days)

Alb

edo

(%

)Product validation - examples

R.Roebeling, KNMIK.-G. Karlsson, SMHI

T. Manninnen, FMI

R. Hollmann, R. Müller, A. Gratzki, DWD


Data sets for water vapor validation

Level 1 radiances:• Use NWP monitoring systems, e.g. ECMWF data reception statistic;

Level 2 instantaneous and level 3 products:• ground based networks and reference sites via co-location, e.g., GPS,

PMW, and radiosonde;• field campaign data, e.g., BBC2 at Cabauw, VAPIC at SIRTA site in

Palaiseau, France, 2004, AMMA (West Africa);• other satellite data or algorithms, e.g, MODIS data or other SEVIRI

algorithm;• cross comparison to models.

Q/A


Product Usage in Climate Sciences


Seasonal 2-D distribution of cloudiness for the entire SCANDIA area (Scandinavia) in the period 1991-2000

SCANDIA RCA2 ERA-40

WINTER

SPRING

SUMMER

AUTUMN

SCANDIA: AVHRR-based cloud climatology (Karlsson, 2002, Int. J. Climatol., 23, 1023-1044)

RCA2: SMHI Rossby Centre regional climate simulation model version 2(Jones et al, 2004, Ambio, 33, 199-220)

ERA-40: ECMWF Re-Analysis cloud dataset (Uppala, 2001, ECMWF Workshop Proc.)


New RCA3 results investigating the effect of optically very thin clouds

WINTER

SPRING

SUMMER

AUTUMN

SCANDIA RCA3 original RCA3 filtered

Purpose for filtering:

SCANDIA cloud detection limit close to optical thickness of 1.0

Conclusion:

New RCA3 version shows an unrealistic seasonal cycle with too large contribution from optically very thin clouds!


New RCA3 results investigating the effect of cloud overlap assumptions

SCANDIA RCA3 Max-Random RCA3 Maximum

WINTER

SPRING

SUMMER

AUTUMN

Purpose:

Modelled cloudiness sensitive to way of combining clouds in different vertical grid layers.

Conclusion:

Maximum overlap gives less cloudiness (as expected).However, different overlap assumptions cannot explain basic problem with seasonal cycle!


Future Prospects


• Some CM-SAF products will be global others not;

• Extension of the time series backwards;

• CM-SAF will have full re-processing capability;

• Existing products will be improved, mostly by providing error estimates;

• New parameters will be added, likely candidates are aerosols and

precipitation;

• CM-SAF is looking for international partnership with international bodies,

satellite operators, and science institutions to be part of a global

integrated climate monitoring network.

Key issues for the CM-SAF development


www.cmsaf.dwd.de

Updated CM-SAF internet pages October 04

Data search and ordering through Web User Interface ( ~ Jan 05)


Upcoming event: USER Workshop 2005

NOAA 29th Annual Climate Diagnostic & Prediction Workshop, 18 - 22 October 2004, Madison, USA Operational Climate Monitoring from Space The Satellite Application.

Documents

saf data

dedicated saf

climate aspects

eigth saf

nwc saf

eumetsat saf network

thematic climate data

cloud parameters