What is a mesocyclone?

The rotating updraft at the heart of a supercell: how it forms, how it looks in velocity data, and why most never produce a tornado.

A mesocyclone is a cyclonically rotating vortex, roughly 2 to 10 kilometers across, inside a convective storm — in practice, the rotating updraft that makes a thunderstorm a supercell. That is the American Meteorological Society’s Glossary of Meteorology definition, which adds that the vertical vorticity involved is typically on the order of 10⁻² s⁻¹ or greater.

The supercell guide covers where the mesocyclone sits among the storm’s other structures. This page goes one level down: where the rotation comes from, the difference between the midlevel and low-level circulations, and how to read one honestly in Doppler data.

Where the rotation comes from

A mesocyclone does not start as vertical spin. It starts as horizontal spin in the environment, and the storm turns it upright.

When wind speed or direction changes with height — vertical wind shear — the air carries horizontal vorticity: picture a rolling tube of air lying on its side. A strong updraft rising through that tube lifts the middle of it, tilting the horizontal vorticity into the vertical. Stretching by the accelerating updraft then concentrates it.

What makes the rotation coincide with the updraft, rather than sit beside it, is the orientation of that vorticity relative to the air flowing into the storm. Davies-Jones (1984, Journal of the Atmospheric Sciences) showed that when the storm-relative inflow carries streamwise vorticity — spin whose axis points along the direction the air is moving, like a thrown football — the updraft itself acquires net cyclonic rotation. This is the physical basis of storm-relative helicity, the forecast parameter that measures how much streamwise vorticity a storm will ingest.

Midlevel and low-level mesocyclones

These are related but not the same thing, and the distinction matters more than any other on this page.

The midlevel mesocyclone forms first, a few kilometers above the ground, from environmental horizontal vorticity tilted by the updraft. It is what defines the storm as a supercell. It can persist for long periods without anything happening at the surface.
The low-level mesocyclone develops later, if at all, within about the lowest kilometer or two. Its vorticity source is an active research question. Rotunno and Klemp (1985) linked it to horizontal vorticity generated along the cool outflow boundary of the storm’s forward flank, then tilted upward. More recent simulations by Coffer, Parker, Peters and Wade (2023, Monthly Weather Review) found the inflow to the low-level mesocyclone came mostly from the undisturbed near-ground environment, with environmental vorticity the dominant source.

The practical reason to care: tornadoes grow out of low-level rotation, not midlevel rotation. A storm can carry a deep, persistent midlevel mesocyclone all afternoon and never develop the near-ground circulation that matters most.

How it appears in velocity

Doppler radar measures only the radial component of motion — toward or away from the antenna. A rotating column seen side-on therefore shows up as a couplet: inbound on one side of the circulation, outbound on the other, with the boundary between them running roughly along the beam. For a cyclonic (counter-clockwise) mesocyclone, the outbound side sits to the right when looking outward from the radar.

Three properties separate a mesocyclone from a random patch of shear:

Scale. The couplet spans a few kilometers, consistent with the 2–10 km definition. Something much tighter is a smaller, more intense circulation — see the tornado signatures guide.
Depth. The rotation appears in several consecutive elevation cuts, stacked over the same area.
Persistence. It survives scan after scan and moves with the storm.

A common way to express strength is rotational velocity: the maximum inbound and maximum outbound speeds across the couplet, added together and halved. It is a useful number, but read it alongside distance between those two maxima, because the same speed difference spread over a wider span is weaker rotation.

Why storm-relative velocity helps

Base velocity includes the storm’s own motion. A supercell moving at highway speed adds the radial part of that motion to every gate, which can make one side of a couplet look overwhelming and the other barely present. The WSR-88D storm-relative velocity product subtracts a storm motion from the radial velocities so the circulation appears more balanced, and the rotation stands out from the translation.

The catch is that the subtracted motion is an estimate. If it is wrong for the storm you are examining — a right-moving supercell often travels differently from the cells around it — the correction is wrong too. Check base velocity alongside it rather than instead of it.

A mesocyclone is not a tornado

The best-documented number comes from Trapp, Stumpf and Manross (2005, Weather and Forecasting), who examined 5,322 mesocyclone detections that met the WSR-88D Mesocyclone Detection Algorithm’s criteria. Only 26% were associated with tornadoes. Height mattered sharply: about 15% of midaltitude detections were tornadic, against more than 40% of low-altitude detections with bases at or below 1,000 m above radar level.

So a mesocyclone raises the odds, and a low one raises them a great deal, but the default outcome for any given mesocyclone is no tornado.

“Radar indicated rotation”

That phrase appears on the SOURCE line of National Weather Service tornado warnings. It means forecasters issued the warning from radar evidence of rotation, not a confirmed report. In the NWS impact-based warning format, the corresponding “Radar Indicated” tag is used when evidence on radar and in the near-storm environment supports a tornado but none has been confirmed. An “Observed” tag means a tornado has been reported.

A radar-indicated warning is not a lesser warning. It is issued so that lead time exists before confirmation does. Act on National Weather Service warnings, whichever tag they carry.

What radar cannot tell you

What is happening under the beam. The beam rises with distance because the earth curves away beneath it. Under the standard 4/3-earth-radius refraction model, the centerline of the lowest 0.5° tilt is about 1.5 km above radar level at 100 km range and about 4 km at 200 km. Beyond roughly 80 km, the lowest tilt is already above the 1,000 m base height that Trapp et al. associated with the most tornadic detections. A distant storm can show a midlevel mesocyclone and reveal nothing about low-level rotation.
Fine detail at range. The WSR-88D’s beam is about 0.95° wide, roughly 1.7 km across at 100 km. A small circulation at that range is averaged across the beam and looks broader and weaker than it is.
Motion across the beam. Doppler sees only radial motion. Flow perpendicular to the beam is invisible, which is why a couplet only appears on the two sides of a circulation facing toward and away from the radar.
Whether it will produce a tornado. The processes that decide tornadogenesis happen near the ground, often below the lowest scan and at scales smaller than a radar gate. No mesocyclone appearance guarantees or rules out a tornado.

Related: the hook echo, the reflectivity feature a mesocyclone often drags into shape, and the wall cloud, the visible lowering beneath the updraft that spotters watch for.

WeatherStory Radar reads full-resolution Level II velocity and storm-relative velocity, every gate, from NOAA’s public archive.