Comparison of Radar Echo Top Heights Producing Severe Weather in Summer and Spring

1
Comparison of Radar Echo Top Heights Producing
Severe Weather in Summer and Spring

• Kevin A. Sullivan

Mentor W. A. Gallus
December 8, 2003
2
Objective

• Relationship radar echo top heights and severe
  weather.
• Criteria for echo tops predicting severe weather
  between spring and summer.
• Identify outliers and determine when and why the
  outliers occurred.

3
Outline

• 1) Data Collection
• 2) Analysis
• 3) Results
• 4) Conclusions

4
Defined Echo Tops

• Composed of 16 data levels.
• Determined by highest elevation angle radar beams
  are deflected.
• Minimum reflectivity detected is 18 dBZ.

5
Data Collection

• Severe Storm Data
• National Climatic Data Center
• Severe Storm Prediction Center

• Non-Severe Storm Data
• National Center for Atmospheric Research

• Radar Echo Top Data
• Iowa Environmental Mesonet (ISU)

6
Severe Storm Data

• Severe Storm Criteria
• Winds 58 mph
• Hail ¾ inch diameter
• Tornado
• 1376 Severe Weather Reports (April 2002-Present)
• Eliminate Biases (Averaging)

7
Non-Severe Data

• Obtained from National Center for Atmospheric
  Research.
• Data obtained only on non-severe storm days.
• 61 days of non-severe thunderstorms were obtained
  for comparison.

8
Radar Echo Top Data

• Iowa Environmental Mesonet
• Gempak used to view files

9
Radar Range Limitations
National Weather Service, 2003
10
Radar Limitations

• Echo tops missed within 25 miles of the radar
  site.
• Echo top resolution was bad far from radar.
• Radar images were only available every 5 minutes.

11
Study Focus

• Study conducted with radar and storm reports only
  in Iowa.

• WSR-88D sites
• Des Moines (DMX)
• Omaha (OAX)
• Sioux Falls (FSD)
• La Crosse (ARX)
• Davenport (DVN)

12
Analysis

• T-test Significance Testing for Difference of
  Means.

• Side by side box plots were used to compare
  distributions in the spread of radar echo tops.

13
Results Spring vs. Summer

• Echo Top Mean during the summer much higher than
  spring.
• T-statistic supports the difference of means.

14
Severe Wx Echo Tops Plot 2002
15
Severe Wx Echo Tops Plot 2003
16
Box Plots of Seasonal Echo Tops
17
Outliers (Spring)

• April 4, 2002 and May 13, 2002
• High low-level instability
• Surface to 850 mb
• Temp gradient larger than 20 degrees C
• Most likely surface based storms
• Low echo tops
• Reports were associated with tornadoes

18
Outliers (Summer)

• August 20, 2002 and June 27, 2003
• Spring time scenario
• Storms occurred on a warm front.
• Abnormally cool temperatures for summer.
• 70s south of front, 60s to the north
• Dew Points were also much lower than typically
  found in summer.
• Lower tropopause level

19
Results Severe vs. Non-severe

• Severe weather echo tops are much higher than
  non-severe echo tops.
• The t-statistic supports a significant difference
  of means.

20
Severe vs. Non-severe (Spring)
21
Severe vs. Non-severe (Summer)
22
Conclusions

• Echo top means were significantly lower in spring
  than summer.
• Same criterion for severe weather based on echo
  top heights should not be used over entire year.
• Echo top can be a good predictor of seasonal
  severe weather.

23
Conclusions (cont.)

• Echo tops should be considered with other severe
  weather indicators
• VIL
• Wind Shear
• Instability
• Low level instability or wind shear can cause
  severe weather to be missed by radar echo tops.

24
Final Remarks

• This study could be useful in the short term
  prediction of severe weather.
• It might be interesting to compare radar echo
  tops with actual warnings issued by the NWS.
• Completing this study over a longer period of
  time might be useful.

25
Acknowledgements

• W. A. Gallus
• Daryl Herzmann

Questions???