WFO Omaha/Valley Nebraska (OAX) Warning Ways, Woes, and Wishes

1
WFO Omaha/Valley Nebraska (OAX)Warning Ways,
Woes, and Wishes

• Daniel Nietfeld
• Science and Operations Officer

2
Number of Warnings by WFO 1995 - 2004
3
We use all 3 Pieces to the Warning Puzzle
Ground Truth (Spotters)
Radar
Environment
4
Warning Operations

• Two People Per Sector
• Radar Meteorologist
• Product Meteorologist
• Separate Mesoscale Analyst
• Sector 1, Sector 2, and Meso workstations in a
  circle within the circle

5
OAX Operations Area Layout
6
Warning Sector 1
Mesoanalyst
Products
COMMS (NAWAS / Phones)
RADAR
Products
RADAR
Warning Sector 2
7
HAMS
Warning Sector 1
RADAR
Warning Sector 2
Products
Mesoanalyst
COMMS (NAWAS / Phones)
8
RADAR
Products
9
2 People Per Sector

• Eyes Never off RADAR
• Detail and QC can be added to products
• RADAR person can ask for second opinion on how
  does this look to you?
• Product person can ask, How does this sound?
• Allows for multiple SVSs per warning without
  compromising RADAR storm interrogation

10
2 People Per Sector

• Requires more people in the office

11
OAX Radar Methodology

• Storm-Structure Mentality
• 4-Panel and All-Tilts Methodology to see slices
  of the storm
• Use Base Products as opposed to Volume
  Products
• Most real-time information
• We do not depend on Algorithms
  (use as a guidance products to alert us)

12
Would like to see better All-Tilts Navigation
  2257 2301 2305 2310
Loop 19.5 19.5 19.5 19.5
Loop 15.6 15.6 15.6 15.6
Loop 12.5 12.5 12.5 12.5
Loop 10 10 10 10
Loop 8 8 8 8
Loop 6.4 6.4 6.4 6.4
Loop 5.1 5.1 5.1 5.1
Loop 4 4 4 4
Loop 3.2 3.2 3.2 3.2
Loop 2.4 2.4 2.4 2.4
Loop 1.8 1.8 1.8 1.8
Loop 1.3 1.3 1.3 1.3
Loop 0.9 0.9 0.9 0.9
Loop 0.5 0.5 0.5 0.5
  Loop Loop Loop Loop
         
Current slice
13
(No Transcript)
14
A Comprehensive Hail Warning Technique Developed
Across the Central Plains

• Rod Donavon
• National Weather Service
• Des Moines, IA (DMX)

Correlation between large hail and the height of
the 50 dbZ echo as a function of the freezing
level
15
50 dBZ Echo Height Warning Criterion

• FZLVL(AGL) Warning
  Criteria (AGL)
• 6000 ft 15000 ft
  
• 6500 ft 16000 ft
• 7000 ft 17000 ft
  
• 7500 ft 18000 ft
• 8000 ft 19000 ft
  
• 8500 ft 20000 ft
• 9000 ft 21000 ft
  
• 9500 ft 22000 ft
• 10000 ft 23000 ft
• 10500 ft 24500 ft
• 11000 ft 26000 ft
• 11500 ft 28000 ft
• 12000 ft 30000 ft
  
• 12500 ft 32000 ft
• 13000 ft 34000 ft
  
• 13500 ft 36000 ft
• 14000 ft 38000 ft
• 14500 ft 40000 ft

16
Some Use of

• VIL and DVIL
• Storm Top Divergence
• LRM mid and high

17
SCAN and FFMP

• Very limited use of SCAN
• First Echo
• Trends of core height, max dbZ
• EXTENSIVE use of FFMP
• Wonderful program

18
Ground Truth

• Becoming the 1 problem with the warning process,
  mainly contributing to FAR
• Supercell storm
• mid level weak-moderate rotation
• Hook echo
• LCL 2500 meters, with LFC 3200 meters
• Law Enforcement reports a tornado, but is really
  just a lot of dust. We have no choice but to
  issue a tornado warning!

19
May 17, 2005 Arnold Nebraska Dry Microburst
20
May 10, 2005 Near Grand Island, NE
LFC Ht.
21
May 10, 2005 Grand Island NE Supercell
22
WarnGenWe Basically Like it, but

• Slowing down too much
• Buggy, glitchy, too many errors
• System froze up this spring at OAX
• Difficult to match SVS to SVR/TOR
• Suggest color coding warnings to
  polygons/statements
• Storm Motion Limitations
• Only allows for a linear storm motion

23
WarnGen

• Template syntax is too code-oriented
• Difficult to ingest GIS data into warnings
• Difficult getting watch data into warnings
• Too much manual tweaking of polygons (minor
  issue)

24
WarnGen PolygonsWarnings Gone Wild
25
Warning Software

• We use a different program to issue warnings for
• Tornadoes/Severe Thunderstorms
• River floods
• Winter Storms
• Airport Weather Warnings
• Short Term Forecast applications?
• Too many paper/written logs/forms, etc

26
Warning NWP Assistance ?

• Run a simulation of a particular storm to see the
  NWP solution of that storms evolution in the
  next 1 to 2 hours
• 4-D Assimilation with RADAR data, mesonet data,
  etc
• Run in-house at convective process scales
• Run time in a few minutes (perhaps a new run with
  each new volume scan?)

You told us to think big !
27
Any New Software MUST Be

• Simple
• Fast
• NOT Cumbersome
• Well Tested in the field

28
May 22, 2004(Were there 6 tornadoes at 915 PM?)
From a recent Media RADAR Seminar
Who Spilled the box of Cheerios ?!
29
May 22, 2004(Were there 9 tornadoes at 957 PM?)
From a recent Media RADAR Seminar