Tornadogenesis

Did you know that most tornadoes actually form from the bottom up, not from the top down?

Rapid X-band Polarimetric Radar mounted on truck.
Image source: https://arrc.ou.edu/radar_raxpol.html

A topic that has always intrigued me is the process and evolution of tornado formation, referred to as tornadogenesis. I used rapid-scan radar observations of the 24 May 2011, El Reno Oklahoma tornado to initially investigate this topic for my PhD dissertation. I returned to this idea several years later and added seven more tornadogenesis cases with improved near ground observations using high resolution, rapid scanning radar observations from the RaXPol, radar out of the ARRC at the University of Oklahoma in Norman.

Bottom-up evolution of tornadogenesis. Rotation is concentrated at the ground, is accelerated upward and in the process, it strengthens and rotation builds quickly from near the ground upward.

I was motivated to pursue this later study when I discovered a video of a tornado from a storm that I collected data that was time stamped several minutes earlier than what the radar observed and what my team labeled as the tornado start time. Intrigued, I investigated why the radar observed start time was later than the video showing the tornado, which had been meticulously time corrected (https://journals.ametsoc.org/view/journals/bams/97/11/bams-d-15-00174.1.xml – The link can be shortened to say : Seimon et al. 2016). I discovered that the radar-based time used in the formal study describing this event to determine tornadogenesis was based off of the time when there was tornado strength rotation through a 3-Km + layer of the lower atmosphere. But; there was indeed tornado strength rotation in the radar data at the lowest height data of radar observations (about 30 m above the ground) for nearly 2 minutes prior to the onset of the deeper vortex. This motivated me to look at additional cases to see if more tornadoes followed this same trend. My analyses led me to conclude that 6/7 tornadoes in the study clearly formed from either the bottom up, or through sudden strengthening of rotation nearly simultaneously over the depth of the layer between the ground and the lower part of the storm (1-3 Km above the ground). I also concluded that the process of “spinning up” a tornado happens incredibly quickly: over times as short as 30-60 seconds! This is much faster than the traditional hypothesis where tornadoes gradually descend from higher in the storm toward the ground over the course of 5-10 minutes. Only one case may have indicated a rapid top-down process, again occurring on timescales as small as 60 s. Interestingly, the vertical resolution of this outlier case was sensitive to the definition of tornado strength rotation. (See the study at: https://journals.ametsoc.org/view/journals/mwre/150/7/MWR-D-21-0227.1.xml) Subsequent preliminary investigation of 8 additional tornadogenesis cases that have yet to be published further supported this conclusion, with one 1/8 forming in a non-descending manner.

Non-descending evolution of tornadogenesis where a broad region of rotation exists over in the layer between the cloud and the ground. When the rotation is collocated below a region of upward motion like the low-leve mescyclone, convergence of this broad rotation amplifies rotation nearly simultaneously over the depth of the rotating column.

So: The spatio-temporal evolution of rotation during tornado formation based off of rapid-scan radar observations suggests that many, and perhaps most tornadoes do not form from the cloud down to the ground, but rather from the ground up for simultaneously between the ground and the storm above.

So why does the funnel cloud look like it is forming from the cloud to the ground?

Because it is! However, the funnel cloud is not a true indicator of the tornado. Some tornadoes do not even have a funnel cloud, and many have a funnel cloud that only partially extends to the ground.

The presence of a funnel cloud is a function not only of rotation, but also the vertical profile of temperature and moisture: That is to say, the rate at which temperature and moisture change with height. In most situations when severe storms form, temperature decreases rapidly with height in the environment. If air from the surface is lifted, it too will become colder. This lifted surface air generally conserves its original moisture content, so as this surface air is lifted and cools, it will reach a height where it becomes saturated and a cloud forms. This height represents the bottom of the cloud. If all environmental conditions remain constant with time and height, then the cloud base remains flat and does not change height. However, rotation like in the case of a rotating thunderstorm that could produce a tornado, will change things. Small-scale rotation (like thunderstorm scale) will induce localized low pressure. This lower pressure will allow condensation to occur for lower moisture values than if there were no rotation present. This means that two columns of air having the same values of temperature and moisture will condense at different times if their pressures are different such that the air with lower pressure will condense first. In the case of a funnel cloud, air close to cloud base is nearly saturated. Just a small amount of rotation will reduce the air pressure enough to cause condensation to occur close to cloud base. But this same exact amount of rotation will not be associated with a cloud closer to the ground because the air temperature is generally warmer at the ground, which means the air at the ground is less saturated than the air above. In the case of a developing or strengthening tornado, as rotation strengthens, the pressure deficit in the rotating column of air increases (i.e. the localized pressure inside the developing tornado falls). This means that condensation can occur at progressively lower relative humidity values, or in other words at lower heights. In some instances, the funnel cloud makes it all the way to the ground. Other times it does not.

Example of an ongoing tornado without a funnel cloud (right) and corresponding radar images from the same time. The damage survey (not shown) indicated that damage began just prior to this photo being taken. Radar imagery shows (from right to left): reflectivity with a weak echo hole, radial velocity with a tornadic vortex signature, and correlation coefficient with a tornadic debris signature all of which indicate the presence of a tornado. Yet, visually there is no condensation funnel. © Jana Houser

It is important to note that you can have tornado-strength rotation over the full depth of air between the cloud and the ground, and a funnel cloud might only be present near the cloud base. Thus, damage can be occurring at the ground without the funnel cloud, reaching the ground. Our eyes cue into the progression of the funnel cloud because this is what we see. The rotating air is essentially invisible to our eyes. But if you see a funnel cloud, there is quite possibly already a tornado occurring well before the funnel cloud touches the ground. As such, it is not appropriate to use terminologies, such as the tornado touched down, or the tornado is on the ground. By definition, a tornado requires ground-based rotation, to begin with!

A tornado with a funnel cloud part-way to the ground, although near-ground radar-observed
rotation was tornadic in intensity. From Goshen County, WY (5 May, 2009)

 

How exactly to tornadoes form? What do we know about this process?

Tornado formation is complex, and it is likely that multiple configurations of various factors can ultimately lead to the same outcome. It is kind of like making four different cakes, using different ingredients, but at the end of the day, you still get four cakes.

Many tornadoes form from rotating thunderstorms called supercells (See figures below). These storms have strong, upward moving channels of air called updrafts. They traditionally occur in environments with strong instability and strong vertical wind shear. The physical configuration of precipitation around these storms with respect to areas of upward and downward motion make them favorable for tornado production. The majority of precipitation falls in a region to the right, or typically east and northeast of the main storm updraft. This region of moderate to heavy precipitation is referred to as the forward flank region. This area is usually accompanied by relatively cool, downward, moving air that is generated by evaporational cooling, from the falling precipitation. A gradient in precipitation will typically form from the updraft and extend north eastward. Along this gradient, the downward moving air on the north side is adjacent to buoyant, warm and environmental air that is ascending on the south side. The downward motion next to the upward motion promotes rotation oriented like a bicycle tire at low levels of the storm, (mostly below 1 km above ground level). Additionally, when there are strong changes in wind speed and direction in the lowest ~500 m or so of the atmosphere ahead of the storm, this low-level vertical wind shear contributes to the horizontally-oriented rotation (known as horizontal vorticity) available to the storm. The horizontal vorticity gets transported toward the updraft, (usually westward), as a combined result of the storm inflow, and the storm motion eastward. Once this horizontal rotation is near the updraft, it can either be ingested by the updraft right away and re-oriented into the vertical, forming what is referred to as the low level mesocyclone, or it is transported around the western rear side of the updraft. In the latter case, it may encounter another region of descending air called the rear flank downdraft. At this point, the rotating air is transported toward the ground, and in the process it is tilted into the vertical, especially between the interface of downward moving air in the RFD and upward moving air in the storms primary updraft. The combination of downward movement and tilting of the rotation into the vertical leads to the formation of near ground vertically-oriented rotation (this is the proper orientation for a tornado to form). Near ground rotation might additionally form from vertical changes in wind speeds associated with frictional processes behind the rear flank boundary. Simply, friction reduces the wind speed at and near the ground, but the speed changes very quickly with height, again creating horizontal vorticity. Higher up, the strong rotation associated with the low-level mesocyclone drives enhanced upward acceleration. When near-ground rotation is located underneath this low-level mesocyclone, the upward motion into the low level mesocyclone can accelerate the rotation, inward and upward, much like an ice skater, pulling in her arms, and a tornado forms!

 

 

 

Citations and Further Reading:

Coffer, B. E., and M. D. Parker, 2017: Simulated Supercells in Nontornadic and Tornadic VORTEX2 Environments. Mon. Wea. Rev., 145 (1), 149–180, https://doi.org/10.1175/367mwr-d-16-0226.1

Coffer, B. E., M. D. Parker, J. M. Peters, and A. R. Wade, 2023: Supercell low-level mesocyclones: Origins of inflow and vorticity. Monthly Weather Review, https://doi.org/https://doi.org/10.1175/MWR-D-22-0269.1, URL https://journals.ametsoc.org/view/journals/mwre/aop/MWR-D-22-0269.1/MWR-D-22-0269.1.xml

Dahl, J. M. L., 2017: Tilting of Horizontal Shear Vorticity and the Development of Updraft Rotation in Supercell Thunderstorms. J. Atmos. Sci., 74 (9), 2997–3020, https://doi.org/10.1175/jas-d-17-0091.1

Dahl, J. M. L., and J. Fischer, 2023: On the Origins of Vorticity in a Simulated Tornado-Like Vortex. Journal of the Atmospheric Sciences, https://doi.org/https://doi.org/10.1175/JAS-D-22-0145.1

Davies-Jones, R., 1984: Streamwise Vorticity: The Origin of Updraft Rotation in Supercell Storms. Atmos. Sci., 41 (20), 2991–3006.

Davies-Jones, R., 2015: A review of supercell and tornado dynamics. Atmos. Res., 158-159, 274–291, https://doi.org/10.1016/j.atmosres.2014.04.007

Davies-Jones, R., and H. Brooks, 1993: Mesocyclogenesis from a Theoretical Perspective. The Tornado: Its Structure, Dynamics, Prediction, and Hazards, 79, 105–114.

Klemp, J. B., and R. Rotunno, 1983: A study of the tornadic region within a supercell thunderstorm. Journal of Atmospheric Sciences, 40 (2), 359 – 377.

Klemp, J. B., and R. B. Wilhelmson, 1978: The simulation of three-dimensional convective storm

dynamics. 1070–1096 pp.

Murdzek, S. S., P. M. Markowski, Y. P. Richardson, and R. L. Tanamachi, 2020: Processes preventing the development of a significant tornado in a Colorado supercell on 26 May 2010. Mon. Wea. Rev., 148 (5), 1753–1778, https://doi.org/10.1175/MWR-D-19-0288.1

Orf, L., R.Wilhelmson, B. Lee, C. Finley, and A. Houston, 2017: Evolution of a long-track violent tornado within a simulated supercell. Bull. Amer. Meteor. Soc., 98 (1), 45–68, https://doi.org/10.1175/BAMS-D-15-00073.1

Parker, M. D., 2023: How well must surface vorticity be organized for tornadogenesis? Journal of the Atmospheric Sciences, https://doi.org/https://doi.org/10.1175/JAS-D-22-0195.1 URL https://journals.ametsoc.org/view/journals/atsc/aop/JAS-D-22-0195.1/JAS-D-22-0195.1.xml

Parker, M. D., and J. M. L. Dahl, 2015: Production of Near-Surface Vertical Vorticity by Idealized Downdrafts. Mon. Wea. Rev., 143 (7), 2795–2816, https://doi.org/10.1175/mwr-d-14-00310.1

Peters, J. M., B. E. Coffer, M. D. Parker, C. J. Nowotarski, J. P. Mulholland, C. J. Nixon, and J. T. Allen, 2023: Disentangling the influences of storm-relative flow and horizontal streamwise vorticity on low-level mesocyclones in supercells. Journal of the Atmospheric Sciences, 129–149, https://doi.org/10.1175/jas-d-22-0114.1

Roberts, B., M.Xue, and D. T. Dawson, 2020: The Effect of Surface Drag Strength on Mesocyclone Intensification and Tornadogenesis in Idealized Supercell Simulations. J. Atmos. Sci., 77 (5), 1699–1721, https://doi.org/10.1175/jas-d-19-0109.1

Rotunno, R., 1981: On the evolution of thunderstorm rotation. Monthly Weather Review, 109 (3),  577–586.

Rotunno, R., and J. B. Klemp, 1985: On the rotation and propagation of simulated supercell thunderstorms. J. Atmos. Sci., 42 (3), 271–292, https://doi.org/10.1175/1520-0469(1985)042⟨0271:OTRAPO⟩2.0.CO

Schenkman, A. D., M. Xue, and M. Hu, 2014: Tornadogenesis in a High-Resolution Simulation of the 8 May 2003 Oklahoma City Supercell. J. Atmos. Sci., 71 (1), 130–154, https://doi.org/10.1175/jas-d-13-073.1

Wicker, L. J., and R. B. Wilhelmson, 1993: Numerical Simulation of Tornadogenesis Within a Supercell Thunderstorm. The Tornado: Its Structure, Dynamics, Prediction, and Hazards, 79, 75–88.