THE INFLUENCE OF HELICITY ON NUMERICALLY SIMULATED CONVECTIVE STORMS

A three-dimensional numerical cloud model is used to investigate the i nfluence of storm-relative environmental helicity (SREH) on convective storm structure and evolution, with a particular emphasis on the iden tification of ambient shear profiles that are conducive to the develop ment of long-lived, strongly rotating storms. Eleven numerical simulat ions are made in which the depth and turning angle of the ambient vert ical shear vector are varied systematically while maintaining a consta nt magnitude of the shear in the shear layer. In this manner, an attem pt is made to isolate the effects of different environmental helicitie s on storm morphology and show that the SREH and bulk Richardson numbe r, rather than the mean shear in the low levels, determine the rotatio nal characteristics and morphology of deep convection. The results dem onstrate that storms forming in environments characterized by large SR EH are longer-lived than those in less helical surroundings. Further, it appears that the storm-relative winds in the layer 0-3 km must, on average, exceed 10 m s-1 over most of the lifetime of a convective eve nt to obtain supercell storms. The correlation coefficient between ver tical vorticity zeta and vertical velocity w, which (according to line ar theory of dry convection) should be proportional to the product of the normalized helicity density, NHD (i.e., relative helicity), and a function involving the storm-relative wind speed, has the largest peak values (in time) in those simulated storms exhibiting large SREH and strong storm-relative winds in the low levels. Even when the vorticity is predominantly streamwise in the storm-relative framework, giving a normalized helicity density near unity (as is the case in many of the se simulations), significant updraft rotation and large w-zeta correla tion coefficients do not develop and persist unless the storm-relative winds are sufficiently strong. The correlation coefficient between w and zeta based on linear theory is found to be a significantly better predictor of net updraft rotation than the bulk Richardson number (BRN ) or the BRN shear, and slightly better than the 0-3-km SREH. Both the theoretical correlation coefficient and the SREH are based on the mot ion of the initial storm after its initially rapid growth. Linear theo ry also predicts correctly the relative locations of the buoyancy, ver tical velocity, and vertical vorticity extrema within the storms after allowance is made for the effects of vertical advection. In predictin g the maximum vertical vorticity both above and below 1.14 km, rather than the actual w-zeta correlation, the 0-3-km SREH performs slightly worse than the BRN. The correlation coefficient, SREH, and BRN all do a credible job of predicting storm type. Thus, it is recommended that operational forecasters use the BRN to predict storm type because it i s independent of storm motion, and the SREH to characterize the rotati onal properties of storms once their motions can be established. Final ly, the ability of the NHD to characterize storm type and rotational p roperties is examined. Computed using the storm-relative winds, the NH D shows little ability to predict storm rotation (i.e., maximum w-zeta correlation and maximum vertical vorticity), because it neglects the magnitudes of the vorticity and storm-relative wind vectors. Histogram s of the disturbance NHD show a distinct bias toward positive values n ear unity for supercell storms, indicating an extraction of helicity f rom the mean flow by the disturbance, and only a slight bias for multi cell storms.