Member JLDN (Japanese Lightning Detection Network) is a large-scale lightning location system, which covers from Hokkaido to Southwest Islands of Japan. It observes LEMP (lightning electromagnetic impulse) in the frequency ranges from VLF to LF, and estimates location and current amplitudes of lighting return strokes. The present study evaluates its ability to discriminate LEMP from return-strokes by comparing its output with the observation of a fastantenna network constructed on Fukui Plain. It turns out that about 90% of the sources of LEMP, interpreted by JLDN as positive return strokes in the 10 kA range in summer, are not cloud-to-ground strokes but in-cloud discharges. The number of termination points on ground produced by negative cloud-to-ground strokes in summer is about 1.3 to 1.4 times of the number of negative flashes observed by JLDN.

No abstract

No abstract

No abstract

Two types of lightning flashes, which are upward lightning and downward lightning, have been observed at Tokyo Skytree. Relationship between upward lightning occurrence and meteorological conditions has been investigated with the data obtained by Japan Meteorological Agency. It has been found that when the altitude of −10 • C is lower than 6000 m, most of lightning flashes are upward and vice versa.

Waveforms of vertical electric field observed at distances of 27, 57, and 101 km from the 634‐m tall Tokyo Skytree (Japan) struck by lightning have been reproduced using the 3‐D finite difference time domain method. Optically observed 3‐D lightning channels and currents directly measured in the tower were used. Distribution of current along the lightning channel was represented using the transmission line model, assuming a constant propagation speed v = 0.5c. Simulations with commonly assumed vertical channel did not allow us to reproduce the observed early (within 5 μs or so) field zero crossing when the 1‐D speed was set to be constant and equal to the 3‐D speed (v = 0.5c) used for the actual‐channel‐geometry case. It was, however, possible to achieve a good match for a vertical channel when the vertical component of 3‐D speed was evaluated for each nonvertical channel segment and used in the corresponding range of heights.

No abstract

No abstract