Britain's deadliest tornado-The 1913 outbreak.

[WeatherArc]

Hey guys, while we are in a dip between any potential heat and storms next week I decided to do some research into the deadliest and one of the strongest tornado outbreaks on record in the UK, the 1913 outbreak that affected parts of Wales and Western England. This outbreak would record 5 strong tornadoes, 3 rated F3/T7 and 2 rated F2/T5. Atleast 6 people would die and over 150 were injured.

An interesting thing about this outbreak was when it occurred, October the 28th. We usually associate strong thunderstorms/tornadoes with the summer heat and instability however this isnt always the case in the Uk. Looking at the data most tornadoes occur within the Autumn with only a few notable exceptions (Birmingham 2005) 

Going back through reanalysis, a large trough with an attendant area of low pressure was situated to the south west of the country, this positioning allowed for direct southerlys to pump heat northwards creating some unusually warm conditions for the end of the month, temperatures widely over 21 °C were recorded with some spots even higher. The 850mb map shows most of the country under the 8 degree isotherm, with parts of the south east under 9. This unusual warmth combined with Atlantic weather fronts to create some unusual Autumn instability.  I would expect instability to be in the 800-2000 j/kg range, given the steep lapse rates and 20 degree plus surface temperatures. 

The devastating South Wales tornado of October 27th, 1913

WWW.GEOLOGYWALES.CO.UK

This paper by geology wales details the event in great depth and i would recommend people take a read if they want to research this event further.

The first reports of a storm came at around 16:00, with a weather spotter reporting several flashes of lightning followed by  "a terrific hailstorm such as the oldest inhabitant never remembered having seen" The storm lasted around 10 minutes before moving off to the North East, hitting areas of Somerset before crossing the Bristol Channel and hitting areas in the Vale of Glamorgan. It was this storm that went on to produce the deadliest tornado in the Uk, killing 6 and injuring 100 in and around the community of Merthyr Tydfil. The exact wind speeds of the tornado are unknown however by using the damage an estimate of over 160 mph was made, this would equal a high end EF3 on the Enhanced Fujita scale. The storm itself was in my opinion was an extremely cyclic supercell, producing multiple strong tornadoes. All reports of tornadoes on this day came from this cell, it likely produced over 10 tornadoes as it tracked over Wales and North-West England. Unfortunately i cant find any photos of this supercell or any of the tornadoes it produced. After Merthyr Tydfil this cell would produce another EF3 in Abercynon, this one was likely as strong as the previous tornado, A high-end EF2 in the Taff valley, and another EF2 (low-end)that struck the outskirts of Shrewsbury, the final strong tornado this cell would produce would be in the Runcorn area at around 20:30, this  EF3 would cause large amounts of damage as it struck the town.  This cell also  produced 10 other weak tornadoes throughout its life, all of these were rated EFU. 

My (badly drawn) guess of the track of the storm going off reports. 

Some interesting meteorological points, the low pressure system off the coast was reported to be rapidly deepening, a pressure of 990 millibars was reported at midnight of the 27th, by 18:00-as the storm was near the Bristol Channel, the pressure had dropped to 975 millibars, I wonder if this rapid deepening of the low helped accelerate the 500mb winds, helping increase Bulk Shear values and therefore increasing favorability for Supercells with all significant hazards. Something else i find interesting was that a South-east wind was reported at the surface. South East  winds at the surface and southerly winds at 800mb would allow for ample directional shear, especially if combined with the speed shear from the strengthening jet coming out of the West/South West. This shear at the surface would allow for significant veering and backing of the low level winds. 

For anyone who doesn't know about veering or backing and how it can increase tornado likelihood, l will try to explain it using some old charts pulled from the US. 

 

This chart to above shows wind barbs, one long line on the barb indicates 10 knots, half a line is 5 knots and a triangle is 50 knots, for further info the NWS does a very good job at explaining them.

Information about wind barbs

WWW.WEATHER.GOV

To the right of the barbs there's some horizontal lines in different colours, red indicates the winds from 0 (surface) to 3km up in our atmosphere, green indicates 3-6km, yellow 6-9km and blue 9km plus.

Notice at the bottom of the chart the wind barbs are pointed to the north west, while at the start of the green lines they are pointed to the North East, this change of direction at the low levels  (veering) dramatically increases tornado probabilities, especially if the winds right at the surface are backed (where the wind comes from a southerly or southeasterly direction) 

 

You can see here from this chart taken from the USA that the turning of the winds in the low levels allows for a large, clockwise turning hodograph supportive of violent tornadoes

So if surface winds were out of the south east, and the winds at 500mb were from the South west we likely had some decent backing at the surface and veering aloft. However, I believe that the tornadoes in South Wales were dramatically aided by the valleys, we know from tornado outbreaks in the Appalachians that valleys can act as localised sources of backing, as they funnel winds through them. The fact that the strongest tornadoes occured within the valleys makes me suspect this increased backing aided in the storms tornadogenesis. 

The tornado at  Merthyr Tydfil was reported to be 50 yards wide when it first touched down near Efail Isaf, as it tracked northwest it strengthened and considerably widened to over 300 yards wide near the community of Treharris. The forward speed of the tornado was around 35mph. 

This is an account by a meteorologist living in the community of Edwardsville, near Treharris- "At 4.30 p.m. the wind, a gentle breeze, was from S. by E and by 5 p.m. was due S. At 5.15 p.m. a dead calm set in. The sky from 430p.m. was heavily mottled with dark patches of cumuli in the South, but these clouds presented straight edges in the SSE. It was a "troubled" sky. To the South, as the twilight advanced, a dark sullen sky was noted. The cumulus form had disappeared except in the East, where a heavy mottled appearance was still observable until darkness came swiftly. During about half-an-hour of calm the atmosphere was oppressive, giving one a sense of great uneasiness, and a remark was made that rain would probably ease the tenseness.Rain began to fall at 5.20 p.m. The first flash of lightning (red) was noted at 5.25 p.m., coming from a dark cloudbank in the South The rain ceased in about 10 minutes, and the red-coloured was followed by intensely blue lightning, flashing at frequent intervals. Very few peals of thunder were heard, these being sharp, with an absence of reverberation, and from about 5 40 p m the thunder was not evident. The blue lightning was appalling. When the flash occurred, which was most frequent, there seemed to be three or four interweaving flashes, all of a deep blue, and, what was strange, the waves of blue fire seemed to be rolling on the ground. A few seconds before 5.50 p.m. we heard a noise resembling the hissing of an express locomotive. The sound grew rapidly in volume, at last resembling the rushing speed of many road lorries racing along. The oppressiveness that had been previously noticed increased, and the heat and air-pressure were pronounced during the rushing noise. We endeavoured to move out of the room to the passage for greater safety, because a hurried remark was made that the engines of these supposed passing loaded steam lorries had collided before the house, and were about to burst, when the panes of our windows were broken by stones, tiles, slates, dried cement, and splintered timber. The missiles broke the Venetian blinds and struck the opposite walls. We made for the rear of the house, but all the windows were being bombarded also by small material and corrugated iron sheets. We could distinctly hear the chimney-pots fall on the roof, and the material sliding off being dashed on the pavement and doorstep. We could see the kitchen clock from the hall passage. It had stopped at 5.51 p.m. It was not struck by any object. It is the largest clock in the house. Two smaller clocks in other rooms were not affected. After this crash had ceased (this only lasted from 60 to 90 seconds), rain fell in torrents. The lightning set fire to the tar which had been sprayed some three weeks previously on the main Cardiff and Merthyr road, some 12 yards from our house door. A distinct smell of sulphur pervaded the air. The lightning continued fitfully and much less intense for about 5 minutes after the climax of the storm had passed, and thunder occurred during the rushing, crashing, roaring noise." 

This plus some other interesting accounts can all be found on the geology wales link I posted above, would recommend a read of the accounts by people after the tornado past. 

One thing i'm surprised by is the lack of fatalities in the third and final F3 tornado that hit Runcorn at around 8:30-9, nighttime tornadoes are known to record higher levels of fatalities, with notable ones such as Greensburg in 2007 and Mayfield in 2021 causing catastrophic fatalities, even a tornadoes of similar strength such as the Bowling Green EF3 in 2021 killed 16 people. The fact that a strong tornado hit a major population area and only injured a few is a miracle. 

Some damage photos from the outbreak.  

 

 

 

Tornado

WWW.ALANGEORGE.CO.UK

 There's quite a few photos on this site of the damage. 

I'll try to find a few more articles, if anyone has any charts or accounts from this then please feel free to post them.

The fatalities from this outbreak are very inconsistent, geology wales says 3 while wikipedia says 6.

Sorry about any spelling mistakes, i've had about 2 hours of sleep so there's probably plenty in there  

Edited August 12, 2023 by WeatherArc

[Rufus Butterfield]

Very well done explanation and a good read. Hopefully nothing like that happens in near future, because as much as we like storms, severe damage and loss of life obviously isn’t good. 
I saw this video recently on this event which was an interesting watch.

 

[WeatherArc]

32 minutes ago, Rufus Butterfield said:

Very well done explanation and a good read. Hopefully nothing like that happens in near future, because as much as we like storms, severe damage and loss of life obviously isn’t good. 
I saw this video recently on this event which was an interesting watch.

 

Thank you , I didn't think to check on youtube for any videos, very informative and interesting. The report of a secondary storm forming before the main supercell produced the tornado makes me think a constructive cell merger occurred, increasing near storm low level SRH and greatly aiding tornadogenesis. The same thing happened on the 28th June 2012 near Sleaford, and to a more extreme example in Joplin 2011 

Metweather on twitter made an absolutely amazing radar breakdown of the Sleaford tornado, can see how that merger helped the supercell produce by increasing near environment helicity, wonder if the same thing happened here? 

[WeatherArc]

Does anyone know if there are any temp/dewpoint/wind or any other surface charts for this day? Im guessing skew-t charts weren't around then. Would be very interesting to view.

Also I quite like going back and looking at /analysing past weather events so if anyone has suggestions of any tornado/convective themes it would be great  to hear them

Edited August 12, 2023 by WeatherArc

[viking_smb]

amazing the pictures up there back then, all the people in the photos are pushing the daisies up,