THE BARON TORNADO INDEX (BTI)

1
THE BARON TORNADO INDEX (BTI)

• Presented by Matt Havin

2
WHAT IS THE BTI?
The Baron Tornado Index

• An integration of the latest available mesoscale
  model data combined with single-site radar data,
  used for advanced analysis of storms during
  potentially tornadic events

3
How does it work?
The Baron Tornado Index

• The BTI ranks each storm for tornadic potential
  in the continental U.S. on a 1-10 scale, all in
  real-time.
• Weak storms are rated a 1, and the strongest
  storms with a likely tornado are rated a 10.

4
What meteorological data is used to produce the
BTI?
The Baron Tornado Index

• Data is taken in real-time from every NWS radar
  site within the continental United States (CONUS)
• From this radar data, all storms in CONUS are
  identified and ranked based upon a variety of
  attributes, including
• Reflectivity
• Velocity
• VIL (hail cores)
• Baron shear markers
• Additional radar attributes

5
What meteorological data is used to produce the
BTI?
The Baron Tornado Index

• After ranking the storms from the radar data, we
  then look at numerical model data to determine if
  the mesoscale environment is conducive for
  tornadoes
• We currently use the operational Rapid Update
  Cycle (RUC) from NCEP for model data, and will
  switch to the Rapid Refresh (RR) in 2010
• To match up the real-time radar data of the storm
  to that of a numerical model, we do an
  interpolation of the model data down to 1 minute
  increments

6
What meteorological data is used to produce the
BTI?
The Baron Tornado Index

• When looking at the RUC model data, there are a
  few numerical model products that we focus a lot
  of attention on when looking for tornadoes
• Research performed by meteorologists at the Storm
  Prediction Center, the NWS, and in the private
  sector, has shown that the majority of tornadoes
  occur with CAPE between 1500-3500 J/kg, 0-1km SRH
  values between 150-300 m2/s2, and LCL heights
  below 1300 meters

7
What meteorological data is used to produce the
BTI?
The Baron Tornado Index

• After processing both the radar (NEXRAD) and
  model data (RUC), we then produce a BTI number
  for all the storms in the continental U.S.
• We are able to do this using algorithms developed
  from performing hundreds of tornado case studies
  that looked at all these pieces of data

8
How long has it been operational? What successes
has it had since its release?
The Baron Tornado Index

• The BTI has been used in the broadcast industry
  since Jan. 2008. The first tornadic event that
  the BTI was used in was the Super Tuesday tornado
  outbreak on Feb. 5th, 2008, where it pinpointed
  numerous deadly tornadoes across much of the
  South.

9
GEORGIA TORNADOES FEB. 18th
The Baron Tornado Index

• The BTI hit 7 on several different tornadic
  storms on the night of February 18th. The EF-2
  tornado that passed south of Peachtree City had a
  7.4 BTI before touchdown, and tracked for 20
  miles.

1/21/2014
9
Matt Havin
10
MURFREESBORO, TENNESSEE APRIL 10TH
The Baron Tornado Index

• On Good Friday, a fatal tornado rated EF-4 struck
  Murfreesboro. The BTI hit 9.4 more than 10
  minutes before the tornado first touched down.
  The tornado tracked over 23 miles and reached ½
  mile wide.

1/21/2014
10
Matt Havin
11
LAKE GUNTERSVILLE, ALABAMA APRIL 10TH
The Baron Tornado Index

• A long-track tornado rated EF-3 crossed Lake
  Guntersville just after 3 PM on Good Friday. The
  tornado caused extensive damage to lake houses on
  Preston Island during the 28 mile track.

1/21/2014
11
Matt Havin
12
LAKE GUNTERSVILLE, ALABAMA APRIL 10TH
The Baron Tornado Index

• A long-track tornado rated EF-3 crossed Lake
  Guntersville just after 3 PM on Good Friday. The
  tornado caused extensive damage to lake houses on
  Preston Island during the 28 mile track.

1/21/2014
12
Matt Havin
13
GOSHEN COUNTY, WYOMING JUNE 5TH
The Baron Tornado Index

• An EF-2 tornado was tracked by the VORTEX2 team
  on the afternoon of June 5th. Doppler on Wheels
  measured wind speeds of 130 mph near the ground.
  BTI hit 7.1 at the time tornado was first spotted.

14
GOSHEN COUNTY, WYOMING JUNE 5TH
The Baron Tornado Index

• An EF-2 tornado was tracked by the VORTEX2 team
  on the afternoon of June 5th. Doppler on Wheels
  measured wind speeds of 130 mph near the ground.
  BTI hit 7.1 at the time tornado was first spotted.

1/21/2014
14
Matt Havin
15
Strategies for using the BTI effectively on the
air
The Baron Tornado Index

• Look for high BTI values (7-10), especially on an
  isolated storm or supercell.
• Example May 10th, 2008 - Stuttgart, AR storm
  with a BTI of 10 produced 20 mile long-tracked
  tornado that produced EF-3 damage
• Comparison of all BTI values in your DMA (one
  storm has 5 BTI, ten other storms have a 2 BTI)
• Example April 10th, 2008 - Southeast Iowa
  stacked low pressure event. Many tornadoes with
  BTI of 4-6, and all others storms with BTI of 1-3
  and no tornado
• Increasing trend in BTI (storm going from BTI of
  3 to 5 to 7 within 15 minutes)
• Example March 14th, 2008 downtown Atlanta
  tornado rapidly increased from 3 BTI to 7 BTI and
  produced the tornado soon afterwards

16
WHEN TO EXERCISE CAUTION WHEN USING THE BTI
The Baron Tornado Index

• Always keep in mind that the BTI is an automated
  tornado detection algorithm. Like anything else
  in the weather world, it is far from perfect.
  Here are common situations where the BTI may not
  detect tornadoes

• Tropical storms/hurricanesTornadic features
  often too small/atypical to be detected by NWS
  NEXRAD radarwhich lowers BTI.
• Cold-core low pressure systemsMarginal CAPE
  (lt500J/kg) and low-topped mini-supercells make
  the overall tornadic potential look less likely,
  which lowers the BTI. However, somewhat useful
  when looking for relative BTI values between 2-5.

17
Leading the next generation of tornado detection
improvements
The Baron Tornado Index

• The BTI was introduced to the broadcast industry
  nationwide in January 2008, and has already
  proven to be highly effective at detecting a
  storm ripe for tornado development with lead
  times of 15-30 minutes on numerous occasions
• Upcoming model improvements, such as the
  replacement of the RUC with the Rapid Refresh
  model in 2010, should help meteorologists and the
  BTI be more effective at identifying potentially
  tornadic events
• Improvements in radar technology, such as the
  upcoming dual-polarimetric NEXRAD upgrade by L-3
  Communications and Baron Services, should help
  lower false alarm rates and improve lead times
  beyond the limits of currently available
  technology