SIX INTERNATIONAL WORKSHOP ON TROPICAL CYCLONES 2006

1
TOPIC 0.1 QUANTITATIVE FORECASTS OF TROPICAL
CYCLONES LANDFALL IN RELATION TO AN EFFECTIVE
WARNING SYSTEM TRACK FORECASTS LIXION A. AVILA
RSMC MIAMI
SIX INTERNATIONAL WORKSHOP ON TROPICAL CYCLONES
2006 IWTC-VI SAN JOSE, COSTA RICA NOVEMBER 2006
2
TOPIC 0.1 WORKING GROUP PHILLIPE CAROFF, JEFF
CALLAGHAN, JAMES FRANKLIN and MARK DeMARIA
SIX INTERNATIONAL WORKSHOP ON TROPICAL CYCLONES
2006 IWTC-VI SAN JOSE, COSTA RICA NOVEMBER 2006
3
(No Transcript)
4
Average track forecast errors (nm) for the 1 to
5 day period before landfall (2001-2005)
5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
Average Lead Time of NHC Hurricane Warnings By
Decade
9
Katrina Track Forecasts1200 UTC 24 August
10
Katrina Track Forecasts0000 UTC 27 August
11
Rita Track Forecasts 1200 UTC 21 September
Severe left bias in track models
12
Rita Track Forecasts1200 UTC 22 September
Remarkable improvement in track guidance Likely
the impact of surveillance data from the NOAA
G-IV jet?
13
Ophelia Track Forecasts (nightmare) 1200 UTC 9
September
14
Wilma Track Models10/21/05 18z
U.K. Met. Office 72 h
GFS 72 h
NCEP GFDL 72 h
NOGAPS 72 h
Navy GFDL 72 h
Verifying position 10/24/05 18z
15
Excellent example of GUNA consensus HURRICANE
ISABEL, 1200 UTC 11 SEP 2003
16
Not-so-excellent example of GUNA consensus
HURRICANE KATE, 1800 UTC 29 SEP 2003 This is a
case where forming a selective consensus can be
effective.
17
New Tropical Cyclone Wind Speed Probability
Products from NHC
June 2006
Richard Knabb, Chris Landsea, Edward Rappaport,
Michelle Mainelli, Chris Lauer, Alison
Krautkramer, James Franklin, and Jamie
Rhome National Hurricane Center Scott Kiser and
Tim Schott - NWS Headquarters Mark DeMaria
NESDIS and John Knaff - CIRA
18
Existing TPC/NHC Products Used to Convey
Uncertainty
Watch/Warning GraphicIndicates forecast track
and long-term mean error
19
Why a new probability product?

• Need for improved means of conveying tropical
  cyclone forecast uncertainty to various types of
  users
• New product is about a weather event at any
  specific location
• Conveys chances of wind speeds of at least
  particular thresholds
• 34 kt (tropical storm force)
• 50 kt
• 64 kt (hurricane force)
• Accounts for combined uncertainty in track,
  intensity, and size
• Extends to 5 days
• Includes inland locations
• Replaced strike probabilities in 2006

20
Challenges and Ongoing Work

• Significant training and outreach needed
• Lessen the focus on exact track forecast
• Do users understand probabilities?
• Getting familiar with the probability values
• Small probabilities of an extreme event
• Enhancements to graphical products
• Faster creation of grids, and the graphics
  created from them
• Objective guidance for watch/warning breakpoints
• Eventual use by NWS forecast offices in their
  products
• Verification
• Gridded products for NDFD

21

• Discussion