An Examination of Tornado-Production Tendencies of Southern-Most Supercells Compared to Adjacent Linearly Oriented Supercells

When storm chasers are faced with a situation where multiple supercells are present, oriented along a linear axis and they must decide on a storm to target, frequently they choose the southern-most supercell.  It has been shown in numerical simulations that for most configurations of the shear vector in the northern hemisphere, the southern-most storm is favored for longevity and isolation from other storms (Bluestein and Weisman 2000).  Often, it is assumed that this storm also will be more favorable for tornado production than supercells farther north (or downshear), presumably owing to the factors mentioned above.  This work examines multiple case studies in which a large number of tornadic and nontornadic supercells are configured along a linear axis (mostly north-south oriented), in order to determine whether or not southern-most supercells are statistically more likely to produce tornadoes than other supercells in the line.  In order for a case to be included in the dataset, at least one supercell within a line must have produced a tornado.  The total number of supercells within the linearly-oriented convection is tracked, as are all the individual supercells that produce tornadoes.  Those that produce tornadoes are categorized as tornadic, while those that do not are put in a separate non-tornado category.  A chi-square test is performed to determine the statistical dependence between tornado production and supercell location relative to the southern-most storm of the line. 

Bluestein, H. B., and M. Weisman, 2000: The interaction of numerically simulated supercells initiated along lines. Mon. Wea. Rev., 128, 3128-3149.