Laboratory studies have shown that thunderstorm charging caused by the interactions of ice crystals and graupel pellets is affected in sign and magnitude by temperature and cloud liquid water content; the presence of water droplets is a requirement for substantial charge transfer. Relationships showing the dependence of charge transfer on ice crystal size and velocity have previously been reported and now, in a continuation of the laboratory studies, the effect of liquid water content on the charge transfer has been investigated. The experiments show that positive graupel charging occurs at temperatures above a “charge sign reversal temperature” with negative charging at lower temperatures. The reversal temperature moves to lower temperatures when the liquid water content is increased. However, at low values of liquid water content, the sign of the graupel charging is inverted being positive at low temperatures and negative at higher temperatures. A discussion is presented of the various charge transfer theories. The results are consistent with the idea of two competing mechanisms whose relative success depends on the temperature and liquid water content. Positive graupel charging occurs when the graupel surface grows from the vapor and the crystals interact with a negative surface charge caused by a temperature gradient across the rime ice surface layer. Negative graupel charging occurs when the surface growth effect is swamped by freezing droplets which create either a pseudo contact potential with which the crystals interact, or a positive surface charge, due to dislocations in the rime ice, which is removed during glancing crystal interactions. Relationships between charge transfer, liquid water content, temperature, ice crystal size, and velocity have been determined and the equations may be used in numerical models of the development of thunderstorm electric fields. A one‐dimensional model indicates that the charge separation rates noted here are adequate to account for thunderstorm electrification. Use of the equations with cloud parameter values obtained in a thunderstorm research flight, leads to a predicted charge reversal level around −13°C, which is in agreement with the analysis of the electric field in the thunderstorm studied.

Abstract. The process of thunderstorm electrification by charge transfers between ice crystals and riming graupel pellets (the noninductive process) has been the subject of extensive study in the laboratory in Manchester. Quantitative dependencies of the sign and magnitude of charge transfer have previously been determined as functions of ice crystal size, graupel/crystal relative velocity, temperature, and the effective liquid water content (EW) in the cloud experienced by the riming graupel pellets. We now present results of laboratory studies of thunderstorm charging in terms of the rime accretion rate (RAR = EW x V), which combines into one variable the velocity and EW dependence of the sign of graupel charging on temperature. The magnitude of the charge transfer can be determined from its dependence on the crystal size and graupel velocity, while the sign of the timer charging can now be determined from a new figure showing the dependence of the charge sign on RAR and temperature. This figure may be used to compare charge transfer results from other laboratories obtained over a range of graupel/crystal velocities. These new experiments extend the temperature range of the previous studies and indicate that negative charging of graupel can occur at temperatures as high as -2øC in conditions of low RAR, while at temperatures below-30øC, more positive graupel charging is noted than in the earlier work.

Measuring systems for atmospheric ice nuclei are undergoing development anew and are beginning to meet the needs for studies of aerosol effects on ice-containing clouds. U nderstanding and predicting the formation of ice in clouds and its possible relation to the changing state of atmospheric composition (aerosols and gas phase) remain enigmatic. Such knowledge and capabilities are critical to quantifying the role of aerosols and their changing compositions on clouds, precipitation, and climate (Denman et al. 2007;Levin and Cotton 2009). This challenge is a major motivation for renewed attempts to measure ice nucleation processes in general, and to design and deploy new portable systems for measuring ice nuclei (IN), the particles that are considered the only means for initiation of the ice phase at temperatures warmer than about −36°C in the atmosphere. The fundamental desire to understand ice nucleation remains the same as when such research began in earnest more than 60 yr ago. The search to identify atmospheric ice nuclei lapsed during the 1970s-80s

A laboratory investigation of electric charge transfer during the impact of vapour-grown ice crystals and supercooled water droplets upon a simulated soft-hailstone target has shown that the magnitude of the charge transferred to the riming surface when crystals separate from it is a function of temperature, crystal dimension, relative velocity, liquid water content, and impurity content of the water droplets and hence the impurity content of the riming target. The sign of the charge transfer depends on temperature, liquid water content and droplet and rime impurity content.In the absence of crystals, no charge transfer was detected during riming. In the absence of supercooled water droplets, crystals impacting at 10m s-on an evaporating rime target produced a small negative charge on the rime of less than -0.25fC per separating crystal. When the target surface grew by vapour diffusion it gained a small positive charge during such interactions. Much larger charges and completely different charge transfer behaviour was noted during riming. The target became positively charged at high liquid water contents and temperatures above a critical value, but negatively charged at lower temperatures or with lower liquid water contents. The critical sign reversal temperature at a liquid water content of 1 g m -3 was about -20°C. At -10°C with a liquid water content of 2gm-3, a 125pm crystal impacting at 3ms-I charged the target by + lOfC upon separation. The charge transfer increased sharply with impact speed and crystal size. Warming the positively charging rime to cause it to evaporate failed to reverse the sign of the charge transfer. Experiments with impurities showed that the sign reversal temperature increased if the droplets contained contaminants at concentrations found in cloud water.It is suggested that there are two distinct charge transfer processes during crystal interactions with an ice target, the dominant one requiring the presence of supercooled water droplets. Careful control and knowledge of the microphysical properties of the clouds used in these experimental simulations has permitted an examination of charge transfer under many of the conditions used in previous studies. The results provide an understanding of the differences and a reconciliation between some of the previously disparate findings in terms of the two distinct charge transfer regimes.

SUMMARYCollisions between vapour-grown ice crystals and a riming target, representing a graupel pellet falling in a thunderstorm, were shown by Reynolds, Brook and Gourley to transfer substantial charge, which they showed to be adequate to account for the development of charge centres leading to lightning in thunderstorms. Related experiments by Takahashi and Jayaratne et al. determined that the sign of charge transferred is dependent on the cloud liquid water content and on cloud temperature. There are marked differences between the results of Takahashi and Jayaratne in the details of the dependence they noted of the sign of graupel charging on cloud water and temperature. More recently, Pereyra et al. have shown that results somewhat similar in form to those of Takahashi are obtained by modifying the experimental technique used to prepare the clouds of ice crystals and supercooled water droplets used in the experiments.In order to help resolve the reason for the differences in charge transfer results in various studies, work has continued in the Manchester laboratory with a modified cloud chamber in which the cloud conditions of the crystals and droplets may be controlled independently. Results indicate a profound effect on the charge sign of the particle growth conditions in the two clouds involved. For example, by suitable adjustments to the water contents of the two clouds, graupel is charged negatively by rebounding ice crystal collisions at higher cloud water contents than have been noted previously. It is suggested that the most important influence on charge sign is the relative diffusional growth rate of the two ice surfaces at the moment of impact and that this is affected by an increase in cloud supersaturation experienced by the ice crystals during the cloud mixing process just prior to collision. A range of cloud conditions is used in the present work in order to help determine the reasons for the various results reported previously.Examination of some thunderstorm observations in the context of the present results points to the importance of mixing on the sign of the charge transferred during particle collisions when two cloud regions of different histories mix together.