June 29 STEPS Supercell Storm: Relationships between Kinematics, Microphysics and Lightning

1
June 29 STEPS Supercell Storm Relationships
between Kinematics, Microphysics and Lightning

• Kyle C. Wiens, Sarah A. Tessendorf,
• and Steven A. Rutledge
• Dept. Atmos. Science, Colorado State University,
  Fort Collins, CO

This research was supported by the National
Science Foundation Under grant ATM-9912051.
2
Motivation

• Investigate and document relationships between
  severe weather and anomalous lightning behavior.
• Understand these relationships
• Improve ability to forecast severe weather based
  on these relationships.

3
Outline

• Initial lightning activity and structure
• Total flash rates
• Height of max lightning density
• Compare with radar data
• Positive cloud to ground (CG) flash rates
• Identify source region of CGs
• Compare with radar data.

4
Initial Lightning Activity

• 2130 to 2150 Infrequent IC flashes from
  mid-level negative down to lower positive
• 2130 to 2150 No evidence of upper positive
  charge
• Starting at 2154 flash rates increase
  dramatically as upper positive gets involved
• Tripole-plus structure

5
IC Flash at 2136
- Charge
Charge
6
First 30 Minutes of Lightning
-
-


7
First 30 Minutes of Lightning
-
-


8
Two IC Flashes at 2152
-

-

9
CAPPIS at two max source regions.
10
Vertical Cross-section
LMA within 1 km of each CAPPI
11
Lofted Graupel
-

-

12
Total Flash RatesFor each minute of the day

• Sort LMA data into flashes.
• Use default parameters of New Mexico Techs
  algorithm-- flashes are defined as groupings of
  successive points within
• 3 kilometers horizontal distance
• 150 milliseconds
• Disregard flashes with fewer than 5 points.

13
Height of Max Lightning Density

• Bin LMA sources by altitude
• Take fullest height bin for each minute

This gives a time series of the altitude where
lightning is most concentrated.
14
Flash Rates
Sort into number of flashes each minute
Record height of max LMA point density each
minute
15
Are Flash Rates Reasonable?
Flash rates follow trends of total LMA
points. So, this LMA-derived flash rate is a good
indicator of electrical activity.
16
Flash Rates
Averaged over each radar volume scan interval
(6-7 min)
BWERs
Tornado
17
Flash rates and updraft volume trends.
18
Flash rates and graupel volume trends.
19
Height of Max Density after 2154
Height of Max Density prior to 2154
20
Flash height and contoured graupel volume
21
Total Flash Rate Summary

• Trends in total flash rate follow trends in
  updraft and graupel volume.
• Height of max lightning activity follows height
  of graupel for first hour or so, then stays
  constant.
• Bursts of lightning follow bursts in updraft and
  graupel.
• Only one cloud to ground (CG) flash in first two
  hours of storm.

22
Cloud to Grounds (CGs) Are Almost all Positive.
23
Positive CGs And Graupel Aloft
Positive CGs And Hail Aloft
24
Where are CGs coming from?

• Take only those LMA points associated with CG
  flashes.
• Take mean height of LMA sources for each
  flashproblematic.
• Take height of max LMA density immediately
  following ground strike. I use 100 milliseconds.
• Now compare these heights to vertical structure
  of hail and graupel

25
Zoom in
26
Try to isolate the CGs parent discharge from
other lightning.
27

• Finally, compute
• Mean Height
• Height of max density
• within (100 ms)

28
Graupel Volume with CG Mean Height
29
Graupel Volume with CG Height of Max Density
(100ms)
30
Hail Volume with CG Mean Height
31
Hail Volume with CG Height of Max Density
(100ms)
32
CG Flash Rate Summary

• CGs follow updraft bursts and formation of hail.
• Source region of most CGs is mid (low)-level
  positive region, not anvil.
• Concentrated hail and graupel in CG source
  region, but CGs occur only when there is hail
  present.

33
Profile of cloud liquid water content from 2022
UTC MGLASS sounding at Goodland, Kansas
T? -20o C
Height (km) MSL
CG source region.
T? 0o C
Cloud liquid water content (g m-3)
Courtesy L. Jay Miller
34
CG source region
Takahashi (1978)
35
CG Examples
The following 6 slides show horizontal and
vertical radar cross-sections through regions of
the storm which produced CG flashes.
I first show the radar cross-sections, then
overlay the LMA data.
36
CG at 2246
Strike point.
37
CG at 2246
Strike point.
38
CG at 2325
39
CG at 2325
40
CG at 2345
41
CG at 2345
42
More questionsfew answers.

• Is hail the primary charge carrier responsible
  for CGs?
• Is the hail simply a catalystloading and
  bringing the graupel down? If so, why so few
  negative CGs?
• What role do dynamic features play?
• Rotation and shear
• Location of updraft relative to precip core

43
More To Come...

• Add T28 and balloon observations to corroborate
  CLWC, precip and electrical structure
• Use flat plate (slow antenna) data to corroborate
  CG source heights
• How does this storm compare to other storms
  during STEPS? (Non)Severe ones? Positive
  (Negative) CG producers?

44
Acknowledgements

• New Mexico Tech, the LMA folks
• LMA data
• XLMA software
• OU/NSSL Rust, MacGorman
• Balloon sounding

This research was supported by the National
Science Foundation Under grant ATM-9912051.