Tornadogenesis in supercell storms—what we know and what we don't know

This paper discusses current knowledge and mysteries of tornadogenesis in supercell thunderstorms. The following questions will be addressed.

High-resolution numerical cloud models now produce tornado-like vortices within simulated supercells in non-rotating atmospheres. The vorticity at the ground is typically baroclinic in origin. This result is due in part to excessive evaporative cooling in models that use Kessler microphysics schemes. Unexpectedly, in-situ field observations have failed to detect a baroclinic zone at the surface near several strong and violent tornadoes. Is there a baroclinic zone aloft through which parcels that enter the tornado pass slowly while descending to near the ground?

Or is there a barotropic tornadogenesis mechanism operating? A ‘bare-bones' axisymmetric model of a mesocyclone without thermodynamics indicates that precipitation falling through a mid-level mesocyclone can drag angular momentum down to the ground. Some of this angular momentum advects inwards. A tornado forms barotropically and then decays owing to the absence of a ‘buoyant cork' aloft.

Is shearing instability at the gust front important? In some simulations, vortices appear to form by this means with two or three of them merging to form a tornado-like vortex. Presumably, the vertical vorticity at the gust front originates ultimately from tilting of baroclinic vorticity.

Why do some imminent tornadoes never develop when strong warning signs are present? Possible explanations are as follows. An axisymmetric model with thermodynamics shows that a tornado will not form if the air at the ground is too negatively buoyant. Presumably a tornado also will not form (at the axis) if strong centrifugal forces prevent convergence.

Do microbursts sometimes trigger tornadoes? Radar observations and damage surveys occasionally reveal a microburst prior to the touchdown of a tornado.

Do some tornadoes form aloft first and descend by a dynamic pipe effect (DPE) or do all tornadoes form throughout a depth that spans the lowest few kilometers? Is the apparent DPE an artifact of low radar resolution?

Is the tornado a contraction of an entire tornado cyclone or is there sometimes an identifiable parent tornado cyclone after the tornado forms?