TORNADOES

Objectives:

A tornado can produce the most intensely violent weather conditions on Earth. Its extreme winds can take lives and cause considerable property damage. Wind speed in a tornado, estimated from the extent of property damage, is the basis for rating these systems from 0 (weakest) to 5 (strongest) on the Enhanced Fujita scale (EF-scale). Dr. Theodore Fujita of the University of Chicago developed the original scale in 1971. It was updated to the enhanced scale based on additional research. Most tornadoes are spawned by and travel with severe thunderstorms. In the Northern Hemisphere the rotating vortex generally exhibits a counterclockwise circulation as seen from above. The special weather pattern required for tornadic thunderstorms to develop is most common in spring and summer in the central United States.

After completing this investigation, you should be able to:

Introduction:

The year 1973 was an exceptional year for tornadoes. That year over 1100 tornadoes were reported. But that was just the prelude, as April 3 and 4, 1974, saw the greatest outbreak of tornado death and destruction to that date in the United States. During a sixteen-hour period, 148 tornadoes swept across thirteen states. These storms proved very strong, long-lived and deadly. There were 330 deaths and almost 5,500 injuries. At a commemoration event twenty-five years later, a National Weather Service official noted, “We want the public to be aware that deadly storms such as the 1974 outbreak can and will happen again, and we want people to be prepared.” (http://www.publicaffairs.noaa.gov/stories/sir58.html ()) Unfortunately, the tornado episodes of spring 2011 demonstrated that these deadly tornado threats will always be with us.

The NOAA Storm Prediction Center (SPC) is one of NOAA’s National Centers for Environmental Prediction. Their website is http://www.spc.noaa.gov/ (). The SPC provides severe thunderstorm and tornado watches for the contiguous U.S. A severe weather watch, e.g. thunderstorm or tornado watch, is issued when meteorological conditions are favorable for that weather situation to occur. (The warning of an impending or a virtually occurring event is issued by your local National Weather Service office.)

Figure 1 from the SPC shows the monthly totals of April tornadoes from 1950, when reliable numbers became available, through 2010. The spring month of April is traditionally among the most active (with May typically being the peak of the U.S. tornado season) as storm systems become energized with contrasts of lingering winter cold and invading spring warmth clashing during thunderstorm formation accompanying midlatitude cyclones.

Fig11B-1.png

Figure 1. Monthly Totals of April tornadoes from 1950 to 2010 from NOAA’s SPC.

1. The 1974 outbreak along with additional tornadoes that month brought the greatest number of tornadoes reported in April that year to [(124)(163)(267)]. That single year drove the 1970’s decadal yearly average to 124.

2. Overall the decadal averages have been increasing with the exception of the 1980s. The green line shows the linear trend of tornado numbers implying that April 2011 would be expected to have about [(150)(170)(210)] tornadoes. The increasing numbers of tornadoes being reported, as noted in the figure text, likely implies greater population density combined with ubiquity of personal communications, cell phones, text messages, etc.

An April for the Record Books

April 2011 proved to be a very active month for severe weather from thunderstorms accompanying several midlatitude cyclonic systems generating strong winds (50 knots, 58 mph, or greater), hail (one-inch, 2.54 cm, diameter or more) and tornadoes. Totals for April 2011 were 3324 reports of damaging winds with 68 of those being for 65 kt (75 mph) or more, 2085 reports of hail with 326 being for large hail (2 in. or greater) and 875 tornadoes. Final tornado numbers, eliminating multiple reports of the same storms, are 758 which still set a record for April. By any standard, this month was exceptionally destructive. There were 363 deaths due to these April tornadoes. The number of April 2011 tornadoes set a record for the highest number in any month; the previous record was 542 in May 2003.

And Then There Was May

While May 2011 did not exhibit the widespread tornado outbreaks of April, the month was very active with one exceptionally deadly tornado occurring on 22 May. Joplin, Missouri was crushed by an EF-5 tornado causing at least 157 deaths and over 750 injuries. This tornado was the most deadly single tornado since modern record-keeping began in 1950 (http://www.crh.noaa.gov/sgf/?n=event_2011may22_summary ()).

A stationary front was stretched across Missouri with a strong jet stream overhead providing the uplift needed for explosive thunderstorm development. Figure 2 is the GOES East visible image of the central U.S. at 2345Z. The figure shows supercell thunderstorms along the front.

The NOAA Storm Prediction Center (SPC) tallies storm reports for the U.S. This frontal system spawned 75 tornadoes, 409 reports of hail greater than 1-inch diameter (50 of > 2”) and 359 incidents of damage due to winds of 50 knots or greater (5 from 65 kts or more) according to SPC statistics for the 24-hour period ending 12Z on 23 May 2011. (To retrieve reports of significant storms for a specified date, visit the SPC’s website. Under Weather Information on the left, click on “Storm Reports”.)

Fig11B-2.jpg

Figure 2. GOES visible view at 2345Z 22 May 2011 of thunderstorms along a stationary front near Joplin, Missouri, labeled with a red dot in center.

One supercell thunderstorm in this extensive storm system generated several tornadoes and caused wind damage across southeastern Kansas to southwestern Missouri including the EF-5 tornado that devastated Joplin, MO.

Figure 3 is a track map of the Joplin tornado showing the damage path from the survey conducted by the NWS Office in Springfield, MO. Numbers within color-coded triangles along the path denote the level of damage on the Enhanced Fujita scale at that location. The EF categories are given in Table 1 below. The center of the tornado’s path is shown by the red line while the brown shading denotes the relative width of destruction along the track. The tornado’s initial touchdown was at the “i” in the left of the image. This was to the west of Joplin.

Fig11B-3.gif

Figure 3. Track map of the Joplin tornado damage with Enhanced Fujita values. [NWS Office in Springfield, MO].

Table 1. Enhanced Fujita (EF) Scale Wind Speed Ranges

(For a description of the EF scale, see http://www.spc.ncep.noaa.gov/efscale/ ())

EF Scale

3-Second Gust Speed (mph)

EF 0

65 - 85

EF 1

86 - 109

EF 2

110 - 137

EF 3

138 - 167

EF 4

168 - 199

EF 5

> 200

3. The track map indicates that the tornado moved generally toward the [(south then southwest)(northwest then north)(north then northeast)(east then southeast)]. This direction resulted from the winds in the lower troposphere and the thunderstorm development.

4. The most intense tornado damage classification category is an EF-5. This was indicated at [(1)(2)(3)(4)] points along the path.

5. These locations [(did)(did not)] coincide with the metropolitan area of the city of Joplin (generally from Oakland Park to Shoal Creek Estates and W 7th St. to Highway 249). This damage path led to the extreme death and destruction associated with this tornado.

The information from the damage report survey associated with this tornado was:

Start Time:

5:34 PM CDT

End Time:

6:12 PM CDT

EF scale rating:

EF5

Est. Path Width:

¾ to 1 mi.

Path length:

22.1 mi.

Fatalities:

159

6. According to the EF scale levels, maximum wind speeds in the tornado were likely [(65 - 85)(86 - 109)(110 - 137)(138 – 167)(168 – 199)(greater than 200)] mph.

7. From the elapsed time on the ground and the length of the damage path, the speed of advance of the tornado was about [(10)(20)(35)(60)] mph.

The SPC provides severe thunderstorm and tornado watches for the contiguous U.S. (The warning of an actual impending or occurring event is issued by the local National Weather Service office. For NOAA weather watch/warning definitions, see http://www.nws.noaa.gov/glossary/index.php?letter=w ().)

8. The NWS issued a tornado warning for the Joplin area at 5:17 PM CDT. This provided [(2)(10)(17)] minutes of lead time prior to the tornado’s “Start Time”. Even with the warning time, the speed of travel, power of the tornado and urban path resulted in the extreme death and destruction that occurred.

Figure 4 is the composite views of the radar reflectivity on the left showing precipitation intensities and the storm relative radial velocities on the right from the Springfield NWS office at 2243Z (5:43 PM CDT) on 22 May 2011. The location of Joplin at the time of the radar images is depicted by the white asterisk. The location of the radar at Springfield, MO, is off each of the images to its right (east).

Fig11B-4.gif

Figure 4. Composite reflectivity (left) and storm relative radial velocities (right) from the Springfield NWS office at 2243Z (5:43 PM CDT) on 22 May 2011.

9. The shadings in the left reflectivity view show the intense precipitation (and perhaps some debris) associated with the supercell thunderstorm circulation as a magenta semicircle about Joplin. Such a comma-shaped structure about Joplin is called a “hook echo”. The hook shape [(would)(would not)] alert a meteorologist to the likelihood of tornadic activity near that location.

10. The right radial velocity view in Figure 4 displays the tornadic vortex signature (TVS) of bright red/orange and green/blue colors adjacent to each other at Joplin. Recall from Investigation 7B that red/orange denotes Doppler velocities away from the radar site (located to the east) and green/blue are toward. Draw a short arrow away from the radar site across the lightest orange group of the pixels in the indicated area. Also draw a short arrow toward the radar site across the bright green pixels adjacent to the orange area. These arrows located to either side of Joplin represent the radial velocities away and toward the radar’s location, respectively. Your pattern of arrows suggests a circulation that is [(clockwise)(counterclockwise)].

The Springfield NWS webpage also provides radar reflectivity and relative velocity views for several other times along with an animation of reflectivities for that event. There are impressive damage photos from the ground and a link to a sequence of aerial “NOAA High Resolution Imagery of the Joplin Tornado Damage”. By clicking the Survey link along the top menu line, a detailed, street-by-street review of the awesome destruction is provided. This event clearly dispels the myth that tornadoes do not hit cities!

The (Tornado) Year That Was

For 2011 there were 1691 tornadoes with 553 fatalities compared to the average over the past decade of 1274 tornadoes. These numbers ranked 2011 as the 4th deadliest tornado year in U.S. history! [http://www.noaanews.noaa.gov/2011_tornado_information.html ()]

Additional Links

The following series of websites contain information from the individual storm reports of the April tornado events and of the entire outbreak:

Huntsville NWS report: http://www.srh.noaa.gov/hun/?n=april27_anniversary ().

Birmingham NWS report: http://www.srh.noaa.gov/bmx/?n=event_04272011 ().

NASA satellite views: http://www.nasa.gov/topics/earth/features/harvest_tornado.html ().

NASA Tuscaloosa image: http://earthobservatory.nasa.gov/IOTD/view.php?id=50434 ().

NOAA NSSL national rotation image: http://www.norman.noaa.gov/2011/04/nssl-product-captures-april-27-tornado-outbreak-storm-rotation-tracks/ ().

Review of storm elements and causes of such destruction: http://www2.ucar.edu/currents/recipe-calamity-ingredients-horrific-tornado-outbreak ().

More information on the Enhanced Fujita Scale is available at: http://www.spc.noaa.gov/efscale/ (). The first occurrence of a documented EF-5 tornado was in Greensburg, KS in May 2007 (http://www.crh.noaa.gov/news/display_cmsstory.php?wfo=ddc&storyid=9475&source=2 () and http://www.crh.noaa.gov/images/ddc/News/Greensburg/Greensburg_1year_later.pdf ()).

Finally, for an account of the historic Super Tornado Outbreak of 1974, see: http://www.publicaffairs.noaa.gov/storms/ (). For more on the infamous Tri-State Tornado of 1925, see: http://www.crh.noaa.gov/pah/1925/ (). Your local NOAA/NWS office websites may have links to notable severe weather episodes in your area.

Last, but not least, a site to answer (almost?) all of your tornado questions: http://www.spc.noaa.gov/faq/tornado/ ().

As directed by your course instructor, complete this investigation by either:

  1. Going to the Current Weather Studies link on the course website, or
  2. Continuing the Applications section for this investigation that immediately follows.

Investigation 11B: Applications

The 2013 tornado season began quicker and earlier than normal with an active January, almost double the annual average of the previous three years. February was about average and March and April were much below normal. The late spring and early summer became more active with another lag until October. November, especially Sunday, November 17th’s outbreak, was abnormally strong for the month.

The Storm Prediction Center (SPC) is one of NOAA’s National Centers for Environmental Prediction. Its website is http://www.spc.noaa.gov/ (). SPC issues national watches for organized thunderstorms and their related activity. Warnings for specific severe weather events are issued by individual NWS offices. The SPC also tallies storm reports for the U.S. Figure 1 of Investigation 11A showed a preliminary total of 136 tornadoes on Sunday. The most devastating tornado of that day struck Washington, in central Illinois. (To retrieve reports of significant storms for a specified date, visit SPC’s website. On the blue menu bar above the Watches map, click on “Storm Reports”.)

Figure 5 is a map of central Illinois with the damage paths of seven tornadoes reported in the region that day. Track number two, to the upper left shown in purple near Peoria, was the path of damage attributed to the tornado that passed through Washington. A map scale and direction arrowhead is provided in the extreme lower right corner. The Lincoln (Central Illinois) NWS webpage report of the tornadoes is at http://www.crh.noaa.gov/ilx/?n=17nov13 (). Clicking on the Washington tornado in the summary listing table below the tracks map (http://www.crh.noaa.gov/ilx/?n=17nov13-tor2 ()) provides more information on that tornado along with radar imagery and damage photographs.

Fig11B-5.png

Figure 5. Central Illinois tornado tracks on 17 November 2013.

11. According to the image legend, the most intense tornado damage classification category (rating) of the Washington tornado was EF- [(1)(2)(3)(4)(5)]. The EF categories are given in Table 1 above.

12. According to the EF scale levels, maximum wind speeds in the tornado were in the range of [(65 - 85)(86 - 109)(110 - 137)(138 – 167)(168 – 199)(greater than 200)] mph.

13. The tornado initially touched down southeast of Peoria. The track map indicates that the tornado moved generally toward the [(southwest)(northwest)(northeast)(southeast)]. This direction was generally aligned with the direction of mid-tropospheric winds. [The upper tropospheric winds were seen on the Monday Image 2 map.]

The reported summary details on this tornado from the Lincoln, IL NWS webpage are:

Rating

EF – 4

Estimated max. wind

190 mph

Casualties

1 death, 125 injuries

Damage path length

46.2 miles

Max. path width

0.5 miles

Approx. start point/time

2.4 mi. SE of East Peoria, 10:59 am

Approx. end point/time

2 mi. E of Long Point, 11:47 am

14. From the elapsed time on the ground (48 minutes) and the length of the damage path, the speed of advance of the tornado was approximately [(10)(20)(40)(60)] mph.

Figure 6 is a composite map of the base reflectivity images for the Washington tornado from the Lincoln and Chicago NWS radars (http://www.crh.noaa.gov/news/display_cmsstory.php?wfo=lot&storyid=98187&source=0#radar ()). A yellow arrow near the low left portion of the reflectivity track denotes the approximate location of Washington, IL.

Image997.GIF

Figure 6. Composite of reflectivity images for the Washington tornado.

15. The shadings at the successive positions show the intense precipitation (reflectivities) in reds associated with the supercell thunderstorm. Individual white pixels are the most intense reflectivity amount, typically resulting from large debris in the air. Observe the same relative southwest side position in each reflectivity view. The red curl-shaped area denotes the “hook echo”. A meteorologist [(should)(should not)] be alert to the likelihood of tornadic activity associated with a hook-shaped echo.

Figure 7 is adapted from the composite image showing views of radar reflectivity on the upper left, differential reflectivity on the upper right, and correlation coefficient on the lower right from the Lincoln NWS radar at 11:07 am 17 November 2013. These views are Doppler radar products resulting from the latest technology upgrade using dual polarization of the transmitted beam. The radar transmission is alternately sent in vertical and then horizontal polarization directions. The resulting radar returns are compared by the computer for these display products. This principle is similar to visible light behavior with polarized sunglasses.

Fig11B-7.gif

Figure 7. Views of radar reflectivity (upper left), differential reflectivity (upper right), and correlation coefficient (lower right) from the Lincoln NWS radar at 11:07 am 17 November 2013.

16. The upper left reflectivity view shows the reflectivity values in pink and white pixels associated with the hook-echo region at Washington at image time. These intense reflectivity values are only seen in tornadic situations, so they [(would)(would not)] likely be related to objects detected by the radar energy that were larger than raindrops. This distinctive radar reflectivity pattern accompanying strong tornadic activity is called a “debris ball”. The large range of reflectivity energies detected at that location is the differential reflectivity as seen in the upper right view.

The correlation coefficient is the statistical relationship between the horizontally and vertically polarized radar energy returns. Equal values indicate almost round precipitation such as small droplets. Some differences indicate flatter large raindrops while even lower values are created by the large size variations in debris caught in tornadic rotations.

17. The lower right correlation coefficient view shows relatively low correlation values by the blue pixels at the same point where strong targets were shown by reflectivities. This lack of correlation [(would)(would not)] be consistent with a mix of objects such as flying debris.

For more information on the details of dual polarization radar images and some examples, one site is http://www.meteor.iastate.edu/classes/mt407/powerpoint/dualpol.pdf ().

As noted by the pictures and news reports, Washington was extensively damaged yet relatively few fatalities were recorded in the area. This was attributed to the radar-detection lead times and the warnings provided to the public, often by social media from friends and neighbors.

Suggestions for further activities: Investigate the tornado websites given in this investigation. Examine the types of radar imagery available to forecasters to use in issuing severe weather and tornado warnings. Current radar imagery, including single station views (NEXRAD), is available from the “NWS Radar Page” link on the course website. Selecting a location on the interactive map will allow one to see regional views and select individual station reports. Views of reflectivity in lowest level scan (“base”) and greatest intensity of any level (“composite”) are available along with base and storm relative velocities and 1-hour and storm total precipitation amounts. These can also be animated.

Investigation 11B: