Title: Predicting Supercell Motion Using A New Hodograph Technique
1Predicting SupercellMotion Using Hodograph
Techniques
Matthew J. Bunkers WFO Rapid City, SD Last
Updated 2/4/2002
2Based onPredicting Supercell Motion Using A New
Hodograph Technique
Matthew J. Bunkers, UNR Brian A. Klimowski,
UNR Jon W. Zeitler, HGX Richard L. Thompson,
SPC Morris L. Weisman, NCAR
by
3Objectives of Study
- Develop a dynamically based method that
consistently predicts the motion of both right-
and left-moving supercells (using only a
hodograph) - Compare the new method with existing methods of
predicting supercell motion - Recommend a preferred method for predicting
supercell motion
4Supercell Motion Myths
- All supercells move to the right of the mean wind
(not truecan move to the left of the mean wind!) - If a storm is moving to the right of the mean
wind, it is a supercell (not truecould just be a
multicell storm)
5Justification forthis Study
- Some currently used methods fail under certain
situations (because they are not Galilean
invariant) - Most supercells (gt 90) produce severe weather
(i.e., hail, flooding, winds, tornadoes) - Nearly all strong or violent tornadoes are
produced by supercells
6Justification (Continued)
- Supercell motion is needed to evaluate
storm-relative helicityhelping to discern
tornadic potential - Anvil-level storm-relative flow may be important
in distinguishing among HP, CL, and LP supercells - Most methods do not address the motion of
left-moving supercells
7Importance of Galilean Invariance
- Next, idealized hodographs are used to illustrate
how Galilean invariance applies to predicting
supercell motion methods based on the mean wind
are not Galilean invariant - 1st slide cyclonic supercell moves slower and
to the right of the mean wind (typical) - 2nd slide cyclonic supercell moves faster and
to the right of the mean wind (northwest flow) - 3rd slide cyclonic supercell moves slower and
to the left of the mean wind (rare)
8Upper-Right Quadrant
9Lower-Right Quadrant
10Upper-Left Quadrant
11Supercell Motion Prediction Methods
- Maddox (1976)30R75
- Colquhoun (1980)inflow outflow
- Davies and Johns (1993)30R75 and 20R85the JDL
method - Weisman (1996)COMET Program module
- Davies (1998)modification of DJ93 above
- Rasmussen and Blanchard (1998)offset from 0-4 km
AGL shear - Bunkers et al. (1998, 2000)this study
12Our Method
- A modification of Weisman (1996) and Weisman and
Klemp (1986) - Based on the internal dynamics of the
supercellcalled the ID method - Galilean invariant and shear-relative
- Observationally, dynamically, and theoretically
based on studies from the 1940s to present
(consistent pattern to supercell motion)
13The ID Method
- Uses the following physical concepts
- Advection of the storm by the mean wind
- Interaction of the convective updraft with the
sheared environment to promote rotation and
propagation - Other external factors, including atmospheric
boundaries and orography, are not accounted for
14The ID Method
- Following is a graphical depiction
- Plot the hodograph
- Plot the mean wind
- Draw the vertical wind shear
- Draw a line perpendicular to the vertical wind
shear that passes through the mean wind - Locate storm motion
15(No Transcript)
16(No Transcript)
17(No Transcript)
18(No Transcript)
19(No Transcript)
20Equation for theCyclonic Supercell
21Data Used in Study
- 260 right-moving (cyclonic) supercells
- 30 left-moving (anticyclonic) supercells
- Data gathered from previous studies and the
northern High Plainsprimary sources include - Davies and Johns (1993)
- Brown (1993)
- Thompson (1998)
22Data Used (Continued)
- Most data gathered from 3 hours from 0000 UTC
using radiosondes - Some cases utilized WSR-88D, profilers, and
averaged soundings - Atypical hodographs were defined as those with
- 0-6 km AGL mean wind lt 10 m/s, or
- a surface wind with a northerly component and gt 5
m/s
23Optimizing the ID Method
- Several iterations were performed to minimize the
error in predicting supercell motion, with the
final results being - 0-6 km AGL non-pressure weighted mean wind
- 7.5 m/s deviation from the mean wind
- 0-0.5 km to 5.5-6 km mean shear vector
24Results (260 hodographs)
ID Method compared individually to others
25Typical Hodograph
26Results (148 Typical Hodographs)
ID Method compared individually to others
27Atypical Hodograph
28Results (77 Atypical Hodographs)
ID Method compared individually to others
29Australian Hodograph
30Importance of Storm Motion
- Research and operational studies have focused on
Storm Relative Helicity (SRH) as a measure of
supercell rotation and tornadic potential - To determine SRH, Storm Motion must be known, or
estimated (by definition) - Following are some examples illustrating the
variability of SRH, and the stability of the
0-6km vertical wind shear
31(No Transcript)
32(No Transcript)
33Supercell-Helicity Relationship
34(No Transcript)
35(No Transcript)
36Supercell-Shear Relationship
37(No Transcript)
38Summary
- The ID method, which is based on the theory for
supercell propagation, is superior to the other
proposed methods evaluated for all hodographs in
this study (by 1 m/s) - This method offers even more improvement in
anticipating supercell motion and storm-relative
parameters for atypical hodographs
39Summary (continued)
- The ID method allows for the prediction of
left-moving supercells (unlike most other
methods) - When the 06-km vertical wind shear exceeds 30
m/s, supercells become more likely (assuming
convective initiation) - The Eta model changed on April 21, 2000 to use
the Bunkers et al. (2000) method for supercell
motion input to SRH calculations
40Complications in Predicting Storm Motion
- Cold pool/shear interactions (internal)
- storm acceleration with time
- Boundaries, merging storms (external)
- e.g, drylines, fronts, outflows
- Orographic influences (external)
- Deeper or shallower storms (internal)
- e.g, mini-supercells, supercells over higher
terrain, elevated supercells
41Complications in Predicting Storm Motion
(continued)
- If the shear is confined to the low levels, the
supercell may become outflow-dominated - stronger gust-front lifting less ventilation
aloft - If the shear is marginal and the CAPE is large,
erratic movement may occur - watch for boundaries/convergence zones
- new cell growth can dominate storm motion
42Complications in Predicting Storm Motion
- If the shear is exceptionally large, significant
deviations from the mean wind may occur
43Bunkers and Zeitler (2000)Highly Deviant
Supercells, 20thSLS
- Even the ID Method fails to accurately predict
the motion of some supercells (i.e., error gt 5
m/s) - A number of factors could account for these
highly deviant supercells - unrepresentative wind profile
- inappropriate mean wind layer
- exceptionally strong vertical wind shear
- weak mid-level vertical wind shear
44Bunkers and Zeitler (2000)Highly Deviant
Supercells, 20thSLS
- Focused on exceptionally strong vertical wind
shear and weak mid-level vertical wind shear - Expanded the dataset to 339 cases
- 245 (72) predictions had a mean absolute error
of 2.7 m/s (Dataset 1) - 94 (28) predictions had a mean absolute error of
7.3 m/s (Dataset 2)
45Bunkers and Zeitler (2000)Highly Deviant
Supercells, 20thSLS
- Dataset 2 was split into 3 partitions
- Weak 08-km vertical wind shear
- Stronger gust front lifting (outflow dominated)
- Strong 08-km vertical wind shear
- Updraftshear interactions more important
(supercell processes dominated) - Strong 03-km shear/Weak 48-km shear
- Combination of gust front lifting and
updraftshear interactions
46(No Transcript)
47(No Transcript)
48Recommendations
- Use the ID method as a starting point to predict
supercell motion - Determine if a shallower or deeper mean wind than
0-6 km is warranted - Identify boundaries and orography that may
influence supercell motion - Understand that the supercell motion will change
with time - Examine the distribution of the vertical wind
shear - Be aware of your environment!