Clustering and synchronization of lightning flashes in adjacent thunderstorm cells from lightning location networks data

[1] We analyzed sequences of lightning flashes in several thunderstorms on the basis of data from various ground-based lightning location systems. We identified patterns of clustering and synchronicity of flashes in separate thunderstorm cells, distanced by tens to hundreds of kilometers from each other. This is in-line with our early findings of lightning synchronicity based on space shuttle images (Yair et al., 2006), hinting at a possible mutual electromagnetic coupling of remote thunderstorms. We developed a theoretical model that is based on the leaky integrate-and-fire concept commonly used in models of neural activity, in order to simulate the flashing behavior of a coupled network of thunderstorm cells. In this type of network, the intensity of the electric field E_i within a specific region of thunderstorm (i) grows with time until it reaches the critical breakdown value and generates a lightning flash while its electric field drops to zero, simultaneously adding a delta E to the intensity of the internal electric field in all thundercloud cells (E_j.k.l⋯) that are linked to it. The value of ΔE is inversely proportional to the distance between the "firing" cell i and its neighbors j, k, 1; we assumed that thunderstorm cells are not identical and occupy a grid with random spacing and organization. Several topologies of the thunderstorm network were tested with varying degrees of coupling, assuming a predetermined probability of links between active cells. The results suggest that when the group coupling in the network is higher than a certain threshold value, all thunderstorm cells will flash in a synchronized manner.