Andrew BurtonBureau of Meteorology, Perth, Australia
Use of Scatterometer Winds in TC Forecasting
Tropical Cyclone Warning Centre Perth
Application of Scatterometer to Tropical Cyclone Forecasting
• Formation• Size (radius of gales)• Wind distribution• Not for absolute intensity (winds saturate at >60 90? knots)
Where to Get Scatterometer Data
• NRL Monterey http://www.nrlmry.navy.mil/tc_pages/tc_home.html
• NOAA/NESDIS QuikSCAT http://manati.wwb.noaa.gov/quikscat
Storms page – includes ambiguities: http://manati.wwb.noaa.gov/cgi-bin/
qscat_storm.pl Alternative NOAA site, with SSMI wind speeds:
http://polar.wwb.noaa.gov/winds/globdata.html• FNMOC
http://152.80.49.210/PUBLIC/SCAT or http://www.fnmoc.navy.mil/PUBLIC/SCAT• Remote Sensing Systems
http://www.ssmi.com
NRL Monterey http://www.nrlmry.navy.mil/tc_pages/tc_home.html
NOAA/NESDIS QuikSCAT http://manati.wwb.noaa.gov/quikscat
FNMOChttp://152.80.49.210/PUBLIC/SCAT/
Near Real-Time Ambiguity Removal
FNMOC Ambiguity Removal over SSMI
Remote Sensing Systemshttp://www.ssmi.com
Scatterometer Coverage
QuikSCAT: SeaWinds Measurements
From Dr. M. Freilich, Oregon State University
V-pol H-pol
SeaWinds: Swath Geometry
From Dr. M. Freilich, Oregon State University
Red = V-polBlue = H-polEdge
View (2 solns) V-pol only
Subtrack View (4 solns,but small angle var)
Ideal View(4 solns, 90
deg var)
Forward Look
Backward Look
Scatterometry: 2-Look Solutions
From Dr. M. Freilich, Oregon State University
1 2 3 4
Solution: wind ~10m/s at ?? degFrom Dr. M. Freilich, Oregon State University
Scatterometry: 4-Look Solution(s)
From Dr. M. Freilich, Oregon State University
1 2
3 4
Most-likely solution: 10m/s at 40deg
From Dr. M. Freilich, Oregon State University
Ambiguities: The Chicken-Scratch Diagram
Location, Intensity, Wind Distribution
Source: Tropical storms discussion group
25 Jan 1800Z 26 Jan 0517Z 26 Jan 1800Z
27 Jan 1712Z 28 Jan 0427Z
XX
X
X
O O
O
O
O
20S170W170W
170W170W
160W 160W
Small system (X) analysed for 3 days --no help from NWP model
• Edge of swath (~ 7 wind vector cells)
• Rain effects
• Sensitivity to errors in NWP• Practical wind regime 3-45? m/s
(problems with both very light and very strong winds)
• Resolution (25km) – impact in tight gradients
• Ambiguity Removal Process and rain flag process can affect final solution
Interpretation Challenges
Edge Problems
Along the whole edge… or small portion…
FNMOC DISPLAY
•Position using the curvature outside ‘rain block’ region.
•Look for good north-south winds.
Rain Effects – “tear drop”
Streamlines
X
Look for non-rain flagged winds
Beware of winds perpendicular to the swath, even when they are not flagged
TC Chris 03/02/02 0914Z
Isotachs
X
Look for min speed near centreTC Chris 04/02/02 1002Z
Errors in NWP
Wrong Model
Position?
TC Guillaume
19/02/02 1341Z
Where is TC HUDAH?No circulation!
?
Max Winds 95 knots
Try to fix in trough equator-ward of the strongest winds
?
? cc
Max Wind 55 KTS
(Light winds?) -----low skill
AVN 19/12Z tau 2420/2356Z
10S10S
20S
20S
In this case, poor model initialization combined with a lower skill nadir position, picks proper wind speed, but NO circulation center
TC Paul
Model initialization errors
Comparing Different Solutions
FNMOC-NOGAPSFNMOC-NRT
NRCS imageryNormalised Radar Cross Section
14 S
10 S
18 S
92 E 88 E 84 E 92 E 88 E 84 E
Scatterometer winds give wrong estimate for centre
Comparison between Quikscat solution from NESDIS 30/11/2001 at 0023Z and fair LLCC seen in SSMI near 14.4S 89.1E 30/11/2001 at 0218Z
Microwave Imagery vs Quikscat
Analysis Methods - Summary
• Ignore the bad - streamline the good
• Tear-drop – curved end
• TC’s – equatorward side of max wind
• Compare different solutions
• Isotach method – ignore direction
Conclusions
• Provides coverage over data sparse areas
• Wind speeds generally good – useful for areas of gales etc
• Use the data if it makes sense
• Be aware of low skill areas and different ambiguity removal processes (compare!)
• Do not use in isolation
