1. The Polar Wind Product Suite
Jeff Key*, Dave Santek+, Richard Dworak+, Chris Velden+,
Jaime Daniels#, Andrew
*NOAA/National Environmental Satellite, Data, and Information Service Madison, Wisconsin USA
+ Cooperative Institute for Meteorological Satellite Studies, University of Wisconsin Madison, Wisconsin USA
#NOAA/National Environmental Satellite, Data, and Information Service Camp Springs, Maryland USA
@IMSG, Inc.
10th International Winds Workshop, Tokyo, February 2010

2. The Polar Wind Product Suite
VIIRS
S-NPP (GOES-R algorithm)
NOAA-20
Direct broadcast at Fairbanks and Sokankylä (heritage algorithm)
Davis, Antarctica (under development) EW
AVHRR
Global Area Coverage (GAC) for NOAA-15, -18, -19
Metop-A, -B
HRPT (High Resolution Picture Transmission = direct readout) at
Barrow, Alaska, NOAA-18, -19
Rothera, Antarctica, NOAA-18, -19
Historical GAC winds, Two satellites throughout most of the time series.
MODIS
Aqua and Terra separately
Direct broadcast (DB) at
McMurdo, Antarctica (Terra, Aqua)
Casey, Antarctica (under development)
LEO-GEO
Combination of may geostationary and polar-orbiting imagers. Fills the degree latitude gap.
See the Notes section for additional information.
Operational
Operational
Operational
Operational
Notes:
“Operational” means official NESDIS operational product.
NOAA-17 was decommissioned on 10 April AVHRR unusable since March 2010.
NOAA-16 was decommissioned on 9 June 2014.
MODIS winds at Tromsø, Norway (DB) were discontinued in September After a hardware (computer) failure, Ksat said that they did not want to support the project anymore.
5. MODIS winds at Fairbanks, AK were discontinued in 2014 and replaced by VIIRS winds. The MODIS data delivery became sporadic.

3. Polar Winds Product History
10-day MODIS winds case study made available
Real-time AVHRR winds
MODIS DB winds at Casey
MODIS DB winds at Fairbanks
LEO-GEO winds
MODIS DB winds at McMurdo
Mixed-satellite MODIS winds
VIIRS DB winds at Fairbanks, Sodankylä
Real-time Terra and Aqua MODIS winds
AVHRR historical winds (v2)
Metop AVHRR winds
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
Daily AVHRR winds
AVHRR HRPT winds from Rothera
NESDIS MODIS winds on GTS
MODIS DB winds at Tromsø
VIIRS winds
MODIS DB winds at Sodankylä
ECMWF and NASA DAO demonstrate positive impact
MODIS, AVHRR winds with nested tracking algorithm
Terra MODIS winds in ECMWF operational system
DB winds distributed via EUMETCast
AVHRR HRPT winds from Barrow

4. MODIS Winds: Single Satellite
Aqua and Terra winds are generated separately
Data from the NOAA Real-Time System (aka “bent pipe”), composites of two or three 5-min granules.
1 km MODIS product (MOD021KM) is remapped to 2 km
Cloud-track and water vapor winds
NCEP’s GFS is used as the background
Pros: Complete polar coverage
Cons: Some MODIS granules arrive (very) late

5. AVHRR GAC and Metop Winds
Six satellites: NOAA-15, -16, -17, -18, -19, and Metop
4 km for GAC; 2 km for Metop
Cloud-track winds
NCEP’s GFS is used as the background.
NOAA-19
Pros: Complete polar coverage; excellent temporal sampling with all satellites. Good preparation for NPOESS VIIRS.
Cons: No water vapor clear winds (no clear sky). Lower spatial resolution (GAC) yields fewer vectors. Height assignment uncertainty for thin clouds.
Metop

6. MODIS Polar Winds 24-hr Loop
Routine production of MODIS winds began in 2002

7. AVHRR Polar Winds NOAA-11 August 5, 1993 1800 UTC
AVHRR winds are being generated with historical AVHRR ( ) and in real-time.
Yellow: Below 700 hPa
Light Blue: hPa
Magenta: Above 400 hPa
NOAA-11 August 5, UTC
NOAA-14 August 14, UTC

8. Direct Broadcast Sites for Polar Winds and Other Products
Fairbanks
(Terra MODIS)
McMurdo
(Terra/Aqua MODIS)
Rothera
(HRPT AVHRR)
Barrow
(HRPT AVHRR)
Tromsø/Svalbard
(Terra MODIS)
Sodankylä
(Terra MODIS)
Barrow
McMurdo
Fairbanks
Svalbard
Sodankylä
Rothera

9. Historical AVHRR Polar Winds
Generally two satellites at any given time, NOAA-7 through -19
Global Area Coverage (GAC) data gridded at 5 km.
Cloud-track winds using IR channel only (no water vapor channel).
NCEP Reanalysis is background.
Pros: An essential product for reanalysis projects.
Cons: Low resolution. Height assignment uncertainty for thin clouds.

10. MODIS Winds in NWP Current Operational Users:
European Centre for Medium-Range Weather Forecasts (ECMWF) - since Jan Testing DB winds.
NASA Global Modeling and Assimilation Office (GMAO) - since early 2003.
Deutscher Wetterdienst (DWD) – since Nov Testing DB winds.
Japan Meteorological Agency (JMA), Arctic only - since May Testing DB winds.
Canadian Meteorological Centre (CMC) – since Sept 2004.
US Navy, Fleet Numerical Meteorology and Oceanography Center (FNMOC) –since Oct DB winds since Apr AVHRR GAC and MetOp since Nov 2007.
UK Met Office – since Feb DB winds (experimentally) since Jun AVHRR GAC since .
National Centers for Environmental Prediction (NCEP) and the Joint Center for Satellite Data Assimilation - since Nov 2005.
MeteoFrance - since Jun MODIS, AVHRR, and all DB winds.
National Center for Atmospheric Research (NCAR), Antarctic Mesoscale Model (AMPS) – since Oct DB winds (McMurdo)
Australian Bureau of Meteorology - since Sept (?) 2007
Hydrological and Meteorological Centre of Russia (Hydrometcenter) – since mid DB winds.

11. Model impact studies have shown that the polar winds have a positive impact on weather forecasts not just in the polar regions, but globally.
Figure: Anomaly correlations as a function of forecast range for the 500 hPa geopotential over the Northern Hemisphere extratropics (north of 20 degrees latitude). The study period is 5-29 March Including the MODIS winds in the model (red line) extends the 5-day forecast at a given accuracy by 3-6 hrs. (Figure courtesy of ECMWF)
Forecast scores (anomaly correlations) are the correlation between the forecast geopotential height anomalies, with and without the MODIS winds, and their own analyses.
From ECMWF

12. Forecast Busts (NASA GMAO)
While the overall impact of the MODIS winds is positive, the impact at any given time may be positive, negative (occasionally), or neutral. What’s important, however, is that the MODIS winds significantly reduce the likelihood and magnitude of forecast “busts”, as shown below in time series of anomaly correlations.
Southern Hemisphere Extratropics
Arctic
Blue is forecast with MODIS winds; red is control run

13. Impact in Tropics: GFS Model
The Joint Center for Satellite Data Assimilation found that MODIS winds also impact hurricane track forecasts.
Average hurricane track errors (nm)
Frequency of superior hurricane performance
48.9  
44.8
39.6
29.4
Cntrl
51.1
55.2
60.4
70.6
Cntrl + MODIS 74 64 52 34
Cases (#)
00-h
24-h
48-h
120-h
Time
13.2 
66.5
102.8
301.1
Cntrl
11.4
60.4
89.0
252.0
Cntrl + MODIS 74 64 52 34
Cases (#)
00-h
24-h
48-h
120-h
Time
In both tables, the forecast time is the bottom row. The control run (Cntrl) did not assimilate the MODIS winds.
Percent of cases where the specified run had a more accurate hurricane position than the other run. Note: These cases are for hurricanes in the subtropics during 2004.