'Storm interaction' is a broad term referring to storm triggering, enhancement or weakening as a result of the proximity of another storm.  This encompasses a range of behavior of nearby storms including altered hydrometeor and inflow trajectories, propagation of one cell along or across the gust front of another, and/or actual merging. 

Our study is focused on the longevity, propagation and rotational characteristics of mature storms as a function of the initial geometry and development time of two nearby cells.  High resolution idealized simulations are being carried out with the WRF model, changing only the location and thermal perturbations of the incipient storms.  Past studies have suggested that small differences in these initial properties can result in large changes in subsequent storm behavior, including intensity and duration of near-surface rotation.  The current work is directed toward an improved understanding of the range of behaviors possible from such cell interactions, including the timing and intensity of mesocyclogenesis resulting from storms developing in proximity versus cells growing in isolation within the same buoyancy and shear environment.


A related study involved an event on 24 April 2002, when a tornado touchdown occurred shortly after a clear storm merger.  A numerical modeling study was carried out and also presented at the 23rd Severe Local Storms Conference (2006); this work was in collaboration with Ron Przybylinski, the Science and Operations Officer of the St. Louis National Weather Service.  A copy of the 2006 extended abstract is here.

This work is supported by the National Science Foundation under grant NSF ATM-0449753 (Robert Wilhelmson, lead PI; Matt Gilmore, co-PI).

For more information, contact Brian Jewett.

Motivation and example: Illinois outbreak

The storm interaction work was first motivated by an outbreak (long ago, now!) of tornadoes in Illinois - more tornadoes in one day than seen in a typical year in the state.  Upon closer examination, a pattern of extensive splitting and merging was found in the early stages of the outbreak. 

The first two figures at right show examples where a broken line of cells split, with left- and right-split cells moving apart along the line.  This occurred often.

The second set of figures on the right are shortly after the first, showing an example where two cells merged; the last of these four images (bottom right in the set) shows two large supercells which were long lived, spawning numerous tornadoes over a several-hour period (some to F-3 intensity).  Questions arising from this work included:

• was the merging necessary, or at least supportive, of tornadogenesis?  Or, was it incidental - such that an isolated storm forming in this environment could be tornado-producing?
• why were the storms so long-lived?  The cloud-bearing shear was nearly normal to the line, a pattern (Rotunno et al. 1988) which is considered unlikely to result in long-lived supercells due to the difficulty in establishing a steady cell as split cells destructively interfere with one another.

A systematic radar study of this case was undertaken by Dr. Bruce Lee (now of WindLogics, Inc.); figures from Lee, Jewett and Wilhelmson (WAF 2006a, 2006b) appear on the right.  The left figure shows the large number of cells initiated on the dryline (drytrough), warm front and dryline-warm front occlusion.  Of those that went on to be mature supercells (right figure), many developed rotation and several were long-lived tornado-producing cells.  An intriguing finding: many of the tornadoes occurred immediately after cell merging.

Example: splitting resulting in MCS decay

The issues mentioned above are relevant to earlier work related squall line longevity to mesoscale forcing (Jewett and Wilhelmson 2006).

In the absence of forcing (e.g. a front) - in this case, a broken line of isolated storms oriented perpendicular to the shear - the cells split and split cells interacted destructively, resulting in the decay of the line.

This is shown in the image at right (click for movie), which depicts surface wind vectors, 2km rainwater mixing ratio (purple contours), areas of updraft (red surfaces) and regions of downdraft (blue surfaces).  The viewpoint is from WSW looking ENE; the line was initially aligned N-S from off the left side of the frame to off the right side.

In the loop, the line of initial cells grow and split, with downdraft gradually replacing and splitting the updraft, with the split cells moving laterally apart, each with an updraft/downdraft pair (as in the figure at right).  When the split cells intersected, the cells decayed and the squall line dissipated.