Long Term Trends in National Hurricane Center

1
Long Term Trends in National Hurricane Center
Watches and Warnings
Mark DeMaria, NOAA/NESDIS/StAR, Fort Collins, CO

and James L. Franklin, NOAA/NCEP/TPC, Miami, FL
Presented at the Interdepartmental Hurricane Con
ference
March 8, 2007
2
Outline

• Introduction
• Data and Verification Methods
• 1963-2006 sample
• Verification and Long Term Trends
• Conclusions

3
Hurricane Watches and Warnings

• Watch hurricane conditions are possible within
  36 hours
  
• Warning hurricane conditions are expected
  within 24 hours
  
• Watches and Warnings (WWs) re-evaluated every 6
  hours
  
• Can be updated at any time in special advisories
  
  
• End points of WWs selected from coastal
  breakpoints

4
(No Transcript)
5
(No Transcript)
6
Input to Watches and Warnings

• Major considerations
• Forecast track, intensity, wind structure
• Forecast uncertainty
• Larger area included than receives hurricane
  winds
  
• Other factors
• Time of day
• 5 PM warnings preferred over 11 PM
• Storm surge and evacuation lead times
• Input from coordination call

7
Data and Verification Methods

• All WW breakpoints for contiguous U.S. digitized
  for 1965-2006
  
• Cindy 1963 and Cleo 1964 also included
• 183 coastal breakpoints used to define coastline
• Supplemented with 157 intermediate points
• Brownsville, Texas to Eastport, Maine 3600 nmi
• Observed hurricane winds from NHC data
• NHC best track position, intensity
• Operational 64 kt wind radii
• Best track radii since 2004

8
Breakpoint 2
Breakpoint 1
Observed Hurricane Wind Region
9
WW Length and Lead Time

• WW Length
• Size of WW for an individual advisory
• Total length with WW at any time in storm
  lifetime
  
• WW Lead Time
• Time between first issuance of WW and arrival of
  hurricane winds
  
• Valid only for points that received hurricane
  winds
  

10
Lengths of Hurricane Warnings from Individual
Advisories1963-2006
N890
11
Top Five Warning Lengths

• Gloria 1985 Sep 27 1053 nmi
• Floyd 1999 Sep 16 1045 nmi
• Bob 1991 Aug 18 984 nmi
• Charley 2004 Aug 13 906 nmi
• Belle 1976 Aug 10 838 nmi

12
Hurricane Warning Length Decadal Averages
13
Storm-Average Warning Lead Time(time from when
warning was first issued at a location to
arrival of hurricane winds)
14
Hurricane Warning Lead Time Decadal Averages
15
Hurricane Watch Length and Lead TimeDecadal
Averages
Length
Lead Time
16
Warning and Watch Statistics(2000-2006)

• Average individual warning length
  290 nmi
  
• Average storm-total warning length
  362 nmi
  
• Average storm-total length w\ hurricane winds
  89 nmi
  
• Prob. of warned point receiving hurricane winds
  25
  
• Lead time of hurricane warning
  34 hr
  
• Average individual watch length
  210 nmi
  
• Average storm-total watch length
  425 nmi
  
• Average storm-total length w\ hurricane winds
  85 nmi
  
• Prob. of watch point receiving hurricane winds
  20
  
• Lead time of hurricane watch
  50 hr
  

17
Conclusions

• The average length of NHC hurricane warnings has
  decreased in the 2000s reversing a 35 year trend
  of increases
  
• Average warning length is 290 nmi
• The average lead times of warnings and watches
  are now 34 hr and 50 hr
  
• Long-term averages are 27 hr and 40 hr
• Warned location has 1 in 4 chance of hurricane
  winds
  
• Watch location has 1 in 5 chance of hurricane
  winds