This event certainly provided us with a busy day - I got to talk on BBC News and Sky News, and Stuart was also on BBC news.
I posted some thoughts about the development of the storm(s) on the MeteoGroup website, in the news section
here:
Obviously it's a website for anyone, and so I haven't used any particularly technical language!
These are the relevant paragraphs concerning reasoning:
Residents of a north-west London suburb were left shocked on Thursday morning, as a tornado ripped through. The tornado affected the Kensal Rise area, and occurred at around 11am. It was spawned from a line of strong thunderstorms which passed across the capital. This line had its humble beginnings across Cornwall, when they spread in from the Atlantic at around 0730. This line raced to the east-northeast, passing Salisbury at around 10am. As it did so, the wind gusted to around 40mph at nearby Larkhill. At the same time, there was a drop in the humidity of the air, and this may have had a bearing on subsequent developments.
As the storms continued to move north-east, the drier air behind the storm is likely to have evaporated some of the precipitation of the storm, and this would have caused the descending air of the storm to become colder, and thus heavier. This caused a surge in atmospheric pressure just behind the line of storms, and this surge appears to have caused the storms to accelerate forward.
A storm’s ability to rotate and possibly produce a tornado is partly derived by the amount of “spin†in the inflow. That is, the more “helicity†the inflowing air has the better chance the storm has of rotating. Helicity is a rather complex notion, but is akin to the air flowing into the storm in spirals, rather like the coiling cable of a telephone handset. The magnitude of the helicity is partly determined by the wind speeds and direction in the lowest 3km of the atmosphere around the storm, and partly by the motion of the storm itself.
In Thursday’s case, it is plausible that the increased motion of the storm as it raced towards London, combined with a slight change in wind direction ahead of the storm, helped rotation to develop. Had the storm moved at a different velocity, the tornado may not have occurred at all. In addition, the downdraught of air at the rear of the storm, called the “rear-flank downdraughtâ€, is seen as very important in tornado formation. This flow of air can concentrate the rotation in the storm, and is often present in tornadic storms across the Great Plains of the USA.