The Ensign Tornado - 29th May 2018
A poor night’s sleep in Goodland, Kansas led the day starting rather slowly, as almost continuous thunderstorms developed close to the town overnight, with several very close lightning strikes and incredibly loud thunder to match. The hotel we’d stayed at suffered badly, with the ceiling in the breakfast room having partially caved in by the following morning (presumably from the torrential rain for much of the night) - we still had breakfast underneath the ceiling which looked perilously close to falling down beside us. 

The heavy rains overnight (estimating 50+mm) leaked into the car, so some hasty bailing of water and drying of mats was needed before we could set off. By the time we did eventually hit the road, it was about 12:20 CDT - much later than we had hoped to leave. The plan, like almost all other chasers, was to head south towards NW Oklahoma, where there was a threat of discrete supercells quickly upscaling into a MCS (with a small tornado risk too). Our initial chase target was Alva, Oklahoma, and we would head south through Dodge City, Kansas. 
Surface weather map, valid for 16:00 (Central Time) on the 29th May 2018. A modifying outflow boundary remained in place at the surface across N Oklahoma/ S Kansas (drawn on this analysis as a trough), while a progressive shortwave trough was expected to move over the region during peak solar heating.
Credit: NOAA/Weather Prediction Center.​

SPC convective outlook valid for 15:00 CDT. An enhanced risk was in place across NW Oklahoma/ S Kansas; in the vicinity of the surface outflow boundary, and where convective-allowing models were consistently developing thunderstorms. Credit: Storm Prediction Center.
SPC tornado outlook valid for 15:00 CDT. A 5% tornado threat was in place across the target area. Ironically, 2 and 5% tornado threats are often considered better chasing conditions than a large threat of violent tornadoes conincident with a major outbreak - a) because the atmosphere is far less volatile, making it much safer to chase without worrying about all the usual hazards, but on steriod. And b), becuase the storms are unlikely to upscale so quickly into an MCS mess (which often happens when there is a greater threat for severe weather), or that the storms are often high-precipitation, making it impossible to see an approaching tornado. 
Credit: Storm Prediction Center. 
Before we left, we noted an outflow boundary (from the previous night’s convection) draped north-south close to Dodge City and the surrounding towns. As we drove south, we noted a discrete line of towering cumulus developing along this outflow boundary, but ignored it initially to continue  with the initial chase target in Oklahoma. However, a north-south oriented line of discrete supercells had developed on this outflow, with backed SE’ly winds re-developing quickly once the outflow passed through. We soon watched as the northern-most storm developed a pronounced hook with strong rotation on radar. With two other discrete storms in development to the south of this initial storm, we noted that these may become the most healthy storms of the day - especially since only one storm had developed in NW Oklahoma - and was still a 90 minute drive away.

We stopped briefly to top up on fuel in Meade, to the SW of Dodge City, and then made the decision to drive north on Highway 23 to catch these developing supercells on the outflow boundary. There were three storms in development - the northernmost storm being the aforementioned supercell to the north of Dodge City. We could make out a distinct wall cloud developing on the middle of the three storms from about 15 miles away, which also began to look increasingly like a tornado-producing supercell on radar. We then stopped briefly on State Highway 56 - about 8 miles south-west of Ensign, Kansas, and watched as the wall cloud rapidly rotated. A funnel then very quickly developed, almost reaching the ground straight-away, before lifting halfway into the clouds. For about a minute or so, the funnel remained halfway to the ground, before it touched down and quickly became a large stovepipe tornado. 

The tornado was on the ground for around 10 minutes, and tracked ENE over open farmland close to the small town of Ensign. We tracked the tornado north-east along the road for a bit, stopping again about 3 miles to the south to get more pictures and finally watch the tornado dissipate. With this storm then quickly looking like becoming outflow dominated, both on radar and visually, we made a unanimous decision to drop south to intercept the southernmost developing supercell, which was also developing a rotating wall cloud by this point. 

A lot of driving on poor dirt roads ensued, while we struggled to keep up with the eastward-moving storm. We did see several funnel clouds drop from the increasingly menacing-looking wall cloud, and tracked the storm towards the community of Bloom. Here we paused for a while, assessing the situation - the storm was heading SSE, about to track across the main road a few miles ahead - bringing torrential rain, large hail and the threat of an increasingly-rain wrapped mesocyclone with it. We made the decision to drive east on the road, before diving south on some dirt roads to escape the hail core. Whilst undertaking this manoeuvre, what appeared to be a newly developing mesocyclone appeared about a mile to our north, and a brief funnel seemed to appear from the swirling clouds. Luckily we managed to avoid this threat, and after several questionable dirt roads - some with dead ends that we only knew about when seeing the signs, we managed to get far south enough to photograph, by this stage, the rapidly downdraft-dominated storm. 

The storm soon upscaled into a MCS mess, and so we ended the day in Coldwater, S Kansas, where we watched the thunderstorms continue to grow from a distance, and a huge anvil, covered in mammatus, glowing red, orange and purple in the evening sunlight. What a day!

A damage survey would later give the tornado a conclusive EF0 rating - entirely due to the fact that the tornado had not hit anything whatsoever - not even any farm machinary - certainly a good thing! 
The wall cloud, just prior to producing the funnel.
The funnel in its early stages. It developed very quickly from the wall cloud, in just a few seconds. 
An enhanced image showing the elephant trunk tornado just after touch-down. 
Ensign Tornado 1.
Ensign Tornado 2.
A shelf cloud that developed from a separate storm - that exploded to the south of the initial Ensign Tornado producing supercell. 
The pink, purple and orange shades of a decaying thunderstorm anvil in the late evening sunlight. 
An evening thunderstorm and photogenic horses - I think they just wanted to have a photoshoot! 
Animation of the Ensign tornadic supercell (middle storm). A velocity couplet and hook echo developed very quickly on the storm, and dissipated just as fast. Credit: Radarscope.