Polygon Warnings

1
Polygon Warnings

• Mike Coyne
• National Weather Service
• Southern Region

2
Overview

• Background Information
• The advantages of polygon warnings
• What are the issues concerning the polygon?
• What is currently being done
• Future Plans

3
Background

• National Weather Service issues four types of
  short-fused warnings
• Tornado Warnings (TOR)
• Severe Thunderstorm Warnings (SVR)
• Flash Flood Warnings (FFW)
• Special Marine Warnings (SMW)
• Issued normally for one county at a time
• Can lead to a large false-alarm area, especially
  in the case of large or irregular-sized counties

4
Background

• Verification has been strictly county-based
• Legacy dissemination methods all tied to
  counties
• NOAA All-Hazards/Weather Radio
• Uses SAME alert code based on counties
• Internet displays again, based on counties

5
What is the Polygon?
6
What is the Polygon?

• Allows forecaster to graphically highlight the
  area of greatest threat
• Creates a template text product for the warning,
  including proper headers, issue/expiration times,
  county UGC codes
• ALSO includes latitude/longitude vertex points
  for the pathcast, or polygon warning --- LITTLE
  USED by NWS, partners, and public

7
What is the Polygon?
8
Polygon Advantages

• Shows specifically where the threat is located
• More accurately shows warning area on systems
  displaying warnings graphically
• Reduction of risk area to public
• Better graphical description capabilities
• Wider local distribution via cell phones, PDAs,
  etc.
• Increase NWS role in the confirmation part of the
  warning process
• Private sector starting to turn to polygons
• Allows NWS to refine warnings to true threat area
• Allows us to track and set goals for false alarm
  area
• Better warning quality
• Keeps NWS in technological step

9
Polygon Advantages
Polygon Eliminates Area False Alarmed
10
One WFOs Study (2004)

• In 2004, issued tornado warnings that covered
• 31,990 square miles (494 individual towns)
• Utilizing the polygon approach, the polygons
  covered
• 9,500 square miles (152 individual towns)
• If the polygon approach were in place
• Would have reduced our total warning area by
  22,490 square miles
• Would have unnecessarily warned 242 fewer towns
• That's a reduction of 70!

11
Better Warning Information
12
Better Warning Information
13
Better Warning Information
14
Polygon Issues

• How should NWS treat these warnings?
• Is the polygon the area of maximum threat?
• Is the polygon the warning area?
• What does our customers/partners wish to see?
• What are the internal and external education
  issues with polygon warnings?
• Do we need to reexamine the entire warning
  process?

15
Other Issues

• Training
• Software (e.g. Warngen, WWA, Storm Data)
• How to train at the operation interface
  (scientific, technical, cultural)
• Dissemination
• Education/public outreach
• Interoperability of dissemination systems
• How to handle cross-CWA boundary issues (polygons
  crossing from one CWA to another)?
• Customer concerns over potential increase in
  number of warnings (broadcast interruptions,
  etc.)
• Verification
• How to measure skill w/respect to polygon
  warnings?
• Consider and understand potential impact on GPRA
  goals
• Operations
• Workload (keeping track of multiple polygons
  multiple small polygons, or few large polygons?
  situation dependent?)
• Do we restrict to short-fuse warnings or extend
  to long-fused warnings

16
No Issue is Trivial
2 maps of all Polygon Warnings Issued in 2003
Can you see the counties?
17
Polygon Team

• NWS Polygon Team formed in 2004
• Purpose Define issues associated with the
  polygon approach to the warning process and
  recommend solutions.
• Test will be run in 2005 with some NWS forecast
  offices
• De-emphasis on county borders
• Focus on location of severe threat
• Test alternate measures of verification that are
  polygon-based
• Will open the door for dissemination using
  current technologies (e.g., cell phones, pagers,
  GPS-enabled devices)

18
Team Members

• Mike Looney, CRH (facilitator)
• Mike Coyne, SRH
• Steve Naglic, WCM WFO Columbia SC (ER)
• Pete Wolf, SOO WFO Jacksonville, FL
• Jeff Lorens, WRH
• Ken Waters, PRH
• Brent MacAloney (OCWWS)
• Rich Okulski (OCWWS)
• Noreen Schwein, CRH (Hydrology)
• Joe Schaefer, SPC

19
Questions

• Mike.Coyne_at_noaa.gov
• [phone] x. 153