This work presents the first simultaneous X-ray measurement and high-speed video observation of the propagation of a lightning leader producing X-rays. As a result, the three-dimensional leader distance from the X-ray measurement and, for the first time, the conditions of the preexisting channel during the leader propagation were observed. Although four leaders in this seven-stroke flash followed the same path to ground, X-rays were only observed during the leader before the return stroke with the highest peak current. The fact that the other three leaders following the same path to ground did not produce detectable X-rays confirms the hypothesis that leader line charge density is an important factor that determines X-ray production. The fact that X-rays was recorded only when the leader tip was at a certain portion of the lightning channel confirms that the orientation of the leader plays an important role in the detection of X-rays. Plain Language SummaryIt was known that lightning can produce X-rays. However, in this study, thanks to the use of a high-speed video camera it was possible to determine when lightning produces X-rays, how far it was, how it was oriented when the detection of X-rays, and what the conditions of the preexisting channel were during the leader propagation. The observations of the present work allow for new insights, confirmation of some hypotheses, and comparison with past studies. The results presented help to understand why X-rays are sometimes detected and sometimes not. It is shown that the amount of charge transferred by the discharge plays a crucial role. This study also confirms that the orientation of the descending leader plays an important role in the detection of X-rays.

In this paper we analyze electric‐field and current measurements of upward leaders induced by a downward negative lightning flash that struck a residential building. The attachment process was recorded by two high‐speed cameras running at 37,800 and 70,000 images per second and the current measured in two lightning rods. Differently from previous works, here we show, for the first time, current measurements of multiple upward leaders that after initiation propagate to connect the negative downward moving leader. At the beginning of the propagation of the leaders that initiate on the instrumented lightning rods, current pulses appear superimposed to a steadily increasing DC current. The upward leader current pulses increase with the approach of the downward leader and are not synchronized but present an alternating pattern. All 2D leader speeds are approximately constant. The upward leaders are slower than the downward leader speed. The average time interval between current pulses in upward leaders is close to the interstep time interval found by optical or electric field sensors for negative cloud‐to‐ground stepped leaders. The upward leaders respond to different downward propagating branches and, as the branches alternate in propagation and intensity, so do the leaders accordingly. Right before the attachment process the alternating pattern of the leaders ceases, all downward leader branches intensify, and consequently upward leaders synchronize and pulse together. The average linear densities for upward leaders (49 and 82 μC/m) were obtained for the first time for natural lightning.

In this paper we analyze electric-field and current measurements of competing upward leaders induced by a downward negative lightning flash that struck a residential building. The attachment process was recorded by two high-speed cameras running at 37,800 and 70,000 images per second and the current measured in two lightning rods. Differently from previous works, here we show, for the first time, the behavior of multiple upward leaders that after initiation compete to connect the negative downward moving leader. At the beginning of the propagation of the leaders that initiate on the instrumented lightning rods, current pulses appear superimposed to a steadily increasing DC current. The upward leader current pulses increase with the approach of the downward leader and are not synchronized but present an alternating pattern. All leader speeds are constant. The upward leaders are slower than the downward leader speed. The average time interval between current pulses in upward leaders is close to the interstep time interval found by optical or electric field sensors for negative cloud-to-ground stepped leaders. The upward leaders respond to different downward propagating branches and, as the branches alternate in propagation and intensity, so do the leaders accordingly. Right before the attachment process the alternating pattern of the leaders ceases, all downward leader branches intensify, and consequently upward leaders synchronize and pulse together. The average linear densities for upward leaders (49 and 82 μC/m) were obtained for the first time for natural lightning.