SAWDIS Share Buttons

SAWDIS Share Buttons:

14 February 2010

"Qumbu Tornado" Storm Analysis (3 February 2010)

NOTE: The aim of this analysis is to see which synoptic, mesoscale, model, radar and real time observations can be collected and identified to indicate severe weather or tornado possibilities. Knowledge of these indicators can help new and existing SAWDIS Weather Observers and hopefully the SA Weather Service to know what to look for in future severe weather events in order to assist in further research and ultimately to hopefully forecast these events with greater accuracy. Please note that the SAWDIS do no have enough information and data to describe this severe storm as a tornadic event. This is surely also not the purpose of this article.

Introduction to Tornado Events

A tornado is defined as a violently rotating column of air with small diameter extending from the thunderstorm to the ground. Tornadoes are small scale by-products of thunderstorms, with less then 1% of thunderstorms producing tornadoes. Until the middle of 1998 a general public perception prevailed hat tornadoes do not occur in South Africa, but occur mostly in the USA. This perception seems to have changed during the 1998 - 1999 summer season when three damaging tornadoes occurred. These were named the Harrismith tornado (15/11/1998; F2 - F3), Mount Ayliff tornado (18/1/1999; F4) and the Umtata tornado. ( 15/12/1998; F2) Severe storms associated with extensive wind damage can, however, be erroneously reported as tornadic storms. An important point to remember is that a tornado's size is not necessarily an indication of its intensity. Large tornadoes can be weak and small tornadoes can be violent and vice versa. The life cycle of the tornado should also be taken into consideration. Damage varies from F0 to F5 with increasing numbers indicating increasing damage and therefor increasing intensity. The official estimate scale of a tornado is only made after the tornado has passed. The FP scale is, however, subjective and varies according to the degree of experience of the surveyor.

Note: Click on all images for larger view.

Image: Vusi Mgobozi (Picture of the storm in the Tabankulu area (3/2/2010)

The Qumbu "Tornado" Storm on 3 February 2010.

The SA Weather Service issued the following warning at 04h00 well prior to the development of the severe thunderstorm.

"Heavy falls of rain (>50mm in 24hours) are possible in places in the north-eastern parts of the Eastern Cape from tonight."

On the afternoon of 3 February 2010, a severe thunderstorm developed over areas of the North Eastern Cape. It started developing to the west of Maclear and traveled in an easterly direction. The image below show the path of the thunderstorm. (Large circle - area of development)

The storm started at 14h21 (all times in SAST) and travelled in a east/north eastward direction until about 16h23. This weather system produced a severe thunderstorm which was described by residents in the area of Qumbu and Ntabankulu as a tornado.

The event received only minor media attention, and was referred to as a possible tornado. There were no reports of human loss although damage to property, trees and livestock were reported.

The "tornado" developed in a "tornado prone" area of South Africa only a distance away from where the F4 Mount Ayliff tornado occurred killing 25 and injured 500 people on the 18 January 1999 at 16h30.

Weather conditions preceding the Storm

The SAWDIS have very little information of weather conditions preceding the Storm. An aspect that will be addressed in future which could assist in better analyzing the development of the storm. The SAWDIS had no private weather stations or observers in the path of the storm except for a report received from an observer in the Thabankulu area. I am however of the opinion that Eta Models, Cape, Lifted Index, Showlater Index and SWEAT Index, Supercell Environment, Wind Shear, Moisture Flux etc. can be obtained but at a substantial cost. The analyzing of conditions conducive to severe storm and tornado outbreaks and methods identifying these conditions is best left to meteorologists of the SA Weather Service.

The SAWDIS however do have the synoptic chart prediction for South Africa on the 3 February 2010.
Image: SA Weather Service (Click on image for larger view.)

A low pressure system was present over the interior of the country with another low pressure system just west of East London. A high pressure system was present over the south east in the Indian Ocean. Temperatures of more than 30 °C were forecasted over the central interior, while cooler conditions (25°C and less) were forecasted over the south eastern parts of the country.

The SAWDIS Real Time-Line of the storm was however very helpful. The storm was in the beginning tracked via the East London Radar. However the Durban Radar was used to record the time-line on SAWDIS Twitter. Durban Radar information:

Durban 30.0117S, 30.9300E 0114m EEC WSR 88 D 5 cm - C-band 1o SAWS

14h21 Thunderstorms currently developing/active in the North Eastern Cape, Free State and Kwazulu Natal - SAWDIS

(Note: Between 14h30 and 14h52 the storm moved over the Qumbu area but no tornadic activity is visible on the Durban or East London Radars. This however does not mean that there was no strong wind in the area of Qumbu.)

15h01 Intense thunderstorm currently active just south of Mount Frere. This storm is moving towards Ntabankulu. Monitor this storm!! - SAWDIS

15h05 Storm south of Mount Frere increased in intensity - SAWDIS

15h12 Mount Frere storm changed direction. Now moving eastwards towards Flagstaff. - SAWDIS (The storm was now close to Ntabankulu)

15h17 The Storm just west of Flagstaff increased in size. This storm is moving fast and might have damaging potential. Strong Wind? - SAWDIS

15h22 The storm west of Flagstaff increased again in intensity. This is now a big storm. - SAWDIS

15h27 Storm west of Flagstaff now stationary. - SAWDIS

15h33 Storm now active in the Flagstaff area. Intensity decreased but still 45dBz in places. Heavy rain and hail possible. - SAWDIS

15h42 The Flagstaff storm is ever so changing. This could be a mean storm! - SAWDIS

15h43 Would love to have a SAWDIS Observer on the ground at Flagstaff for a real-time update. - SAWDIS

15h46 The Flagstaff storm changed direction again and increased again in intensity- SAWDIS

15h48 The storm seem to be moving towards Bizana - SAWDIS

15h58 Storm west of Bizana increased in intensity. 55 dBz in areas. - SAWDIS

16h02 Intense thunderstorm now active in Bizana. This storm has damaging potential. Monitor this storm! - SAWDIS

16h18 Bizana storm moving NE-wards. Intensity and size decreased. Monitor this storm for re-development. - SAWDIS

Description of the damage and impact on human lives.
According to Vusi Mgobozi the so called tornado also hit the Ntabankulu area.

"11 villages were affected. Houses have been destroyed, livestock killed, trees uprooted. A Hyundai bakkie was pushed by the force of the storm and fell about 200m down a slope where it hit a house. Schools were damaged and church houses completely destroyed. The storm hit us at about 14H20 and when I watched it, it had an underdeveloped funnel, strong winds, less rain and lightning. It left a trail of destruction up to Mbizana. "

"..... it had two hanging clouds, the air was so cold and the clouds were moving all over the place. I also noticed that the clouds were moving in an anti clockwise direction, if you were to look the cloud from under. It took about 5 to 10 minutes no rain but just strong winds and then it started to rain but still not much rain."

Vusi also provided several images depicting the damage caused by the storm.

Images: Vusi Mgobozi (Note all these images were taken in the Tabankulu area.)

Summary and Conclusions

The prediction, detection and monitoring of severe storms/tornadoes are some of the most important services provided by the meteorological profession. Throughout the world the timely issue of warnings for severe weather is critical to the protection of life and property, even more so for tornadoes. Nevertheless timely warnings remains very difficult. Severe weather events such as tornadoes can only be accurately predicted in situations where a storm already exists. Real time data such as radar, satellite, upper-air soundings and visual observations are of utmost importance once thunderstorm development has started.

Radar reflectivity values and the shape of the more intense cells can say much about the possibility for tornadoes to develop. Improved and timeous warnings of tornadoes can only be made if real-time observations and quality of radar data are available. This is where the SA Weather and Disaster Information Service can play a very important roll in providing real-time observations to the SA Weather Service. With an observer in the area of the storm valuable information can be gathered from real time observations to analyze the intensity and damage the storm caused.

However much has been learned from information gathered of the severe storm that struck the Qumbu/Tabankulu areas. The information gather will assist the SAWDIS in compiling a quick-reference guide which will contain guidelines to assist real time observers to identify important 'evidence" before, during and after a severe storm or tornado.

With the current information at hand the SAWDIS is not in a position to confirm the fact that a tornado did indeed strike the area of Qumbu or Ntabankulu on the 3 February 2010.


The SAWDIS would like to thank the SA Weather Service, The Media, Mr Vusi Mgobozi who provided valuable information used in this short analysis. This article would have suffered with out the meteorological data, photographs, newspaper clippings and eye-witness reports they provided.

Article: J. Terblanche ZS1I, SAWDIS, Mossel Bay
13 February 2010


SA Weather Service
Report: Tornadic thunderstorm events during the 1998 - 1999 South African summer E de Coning and BF Adam.
Mr Vusi Mgobozi
Tornado FAQ, Rodger Edwards