Two tornadic thunderstorms in ostensibly weak deep layer shear environments in southeastern Colorado: cyclic supercells of May 25 (Kiowa County) and May 31 (Baca County) 2010

Similar strong low-level directional and weak deep bulk layer shear with favorable buoyancy environments occurred in southeastern Colorado on 25 and 31 May 2010. The Kiowa County storm of 25 May produced at least three supercell tornadoes as it moved into western Kansas northwest of Tribune. The Baca County storm of 31 May produced two tornadoes near Pritchett and eventually moved into the Oklahoma Panhandle and produced at least two additional major long-lived tornadoes. These storms occurred during a week long period characterized by a major anchor trough in the Four Corners region of the Southwest. Strongest flow in the middle and upper troposphere extended, generally, across the western portions of Colorado, into eastern Wyoming and then northeastward, with only a fringe of 25-35 knot winds impinging on the western High Plains of Colorado, extreme western Kansas and the northwestern Panhandles.

While the formation of rotating storms was anticipated, the weak deep layer shear environments (exemplified by winds initially 25 knots or less at 500 mb) suggested that the rotating phase of such storms would be brief because precipitation would overwhelm the updraft areas before storms could take advantage of the low-level shear environment,. The latter was characterized by robust clockwise loops in the low-level hodograph, in both cases nearly a 360 degree turning of the shear vectors in the 0-3 km layer. However, bulk shear magnitudes in the 0-1 km layer were not strong in either case.

Storm motion estimation algorithms (such as the Bunkers technique) are strongly influenced by middle and upper tropospheric wind fields as an estimate of mean storm motion vectors and, as such, did not estimate actual initial storm motions accurately. The actual storm motions (towards the northwest at 7 knots on 25 May and south at 7 knots on 31 May) appear to have been dictated by updraft propagation related to the robust turning of the shear vectors in the lowest 3 km. Both of these storm motions yielded updraft propagation along boundaries, a dry line-outflow-synoptic scale boundary on 25 May and a strong outflow boundary on 31 May, that we hypothesized resulted in tilting of horizontal vorticity along those boundaries into the updraft, contributing to much stronger low level shear than existed in the ambient synoptic scale environment. The strongly deviate storm motions had the following additional effects: (a) storm relative flow both at mid and anvil levels became favorable both for rotating updrafts and for adequate storm ventilation, yielding an environment favorable for classic supercells; and (b) measures of low level shear, including storm relative helicity in the 0-1 and 0-3 km layers, indicate that the the low level shear environment then became favorable for mesocyclone-induced tornadogenesis.

It is interesting to note that in both cases the initial storm motion favored the tilting of horizontal vorticity probably generated along the boundaries into the initially-strongly growing towers, resulting in at least four non-mesocyclone (landspout) tornadoes for the Kiowa County case and multiple high-based funnel reports for the Baca County case. Later on in this phase, both storms apparently proceeded through the supercell cascade and produced one or more supercell tornadoes. Finally, by late afternoon the deep layer shear environment and hodograph characteristics had evolved to such a degree that both storms began moving in a more conventional deviate fashion (i.e., as predicted by the Bunkers technique). Both storms then produced additional supercell tornadoes at that time. The case studies presented here suggest that storm motion predictions by algorithms be evaluated more critically before judging the potential for tornadic convection. Also, while rapid decision techniques based upon “rules of thumb” for the lower Great Plains failed in these cases, neither case should be considered an “anomaly”, and the values of parameters evaluated in the storm-relative sense were indeed consistent with those for classic, tornadic supercells. The estimation of the true storm motion appeared to be key in both cases.