Electrification and lightning are simulated for a small continental multicell storm. The results are consistent with observations and thus provide additional understanding of the charging processes and evolution of this storm. The first six observed lightning flashes, all negative cloud-to-ground (CG) flashes, indicated at least an inverted dipole charge structure (negative charge above positive). Negative CG flashes should be energetically favorable only when the negative charge region contains appreciably more charge than the lower positive region. The simulations support the hypothesis that the negative charge is enhanced by noninductive charge separation higher in the storm that also causes development of an upper positive charge region, resulting in a "bottom-heavy" tripole charge structure. The two-moment microphysics scheme used for this study can predict mass mixing ratio and number concentration of cloud droplets, rain, ice crystals, snow, graupel, and hail. (Hail was not needed for the present study.) Essential details of the scheme are presented. Bulk particle density of graupel and hail can also be predicted, which increases diversity in fall speeds. The prediction of hydrometeor number concentration is critical for effective charge separation at higher temperatures (−5 < T < −15) in the mixed-phase region, where ice crystals are produced by rime fracturing (Hallett-Mossop process) and by splintering of freezing drops.

[1] A three-dimensional dynamic cloud model is used to investigate electrification of the full life cycle of an idealized continental multicell storm. Five laboratory-based parameterizations of noninductive graupel-ice charge separation are compared. Inductive (i.e., electric field-dependent) charge separation is tested for rebounding graupel-droplet collisions. Each noninductive graupel-ice parameterization is combined with variations in the effectiveness of inductive charging (off, moderate, and strong) and in the minimum ice crystal concentration (10 or 50 L À1 ). Small atmospheric ion processes such as hydrometeor attachment and point discharge at the ground are treated explicitly. Three of the noninductive schemes readily produced a normal polarity charge structure, consisting of a main negative charge region with an upper main positive charge region and a lower positive charge region. Negative polarity cloud-to-ground (CG) flashes occurred when the lower positive charge (LPC) region had sufficient charge density to cause high electric fields. Two of the three also produced one or more +CG flashes. The other two noninductive charging schemes, which are dependent on the graupel rime accretion rate, tended to produce an initially inverted polarity charge structure and +CG flashes. The model results suggest that inductive graupel-droplet charge separation could play a role in the development of lower charge regions. Noninductive charging, on the other hand, was also found to be capable of producing strong lower charge regions in the tests with a minimum ice crystal concentration of 50 L À1 .Citation: Mansell, E. R., D. R. MacGorman, C. L. Ziegler, and J. M. Straka (2005), Charge structure and lightning sensitivity in a simulated multicell thunderstorm,

[1] Lightning discharges are simulated by using a stochastic dielectric breakdown model within a numerical thunderstorm model with extensive parameterizations of electrification mechanisms. The lightning model simulates the macroscopic bidirectional extension of discharges as a step-by-step stochastic process. Discharge channels are propagated on a uniform grid, and the direction of propagation (including diagonals) for a particular step is chosen randomly, with the probability for choosing a particular direction depending on the net electric field. After each propagation step the electric fields are recomputed via Poisson's equation to account for the effect of the conducting channel. The lightning parameterization produces realistic looking, three-dimensional, branched lightning discharges. A variety of lightning types have been produced, including intracloud discharges, negative cloud-to-ground (CG) lightning, and positive CG lightning. The model simulations support the hypothesis that negative CG flashes occur only when a region of positive charge exists below the main negative charge region. Similarly, simulated positive CG flashes were found to occur only in regions of storms where the two significant charge layers closest to ground had roughly a ''normal dipole'' structure (i.e., positive charge above negative).