Experimental Observations and Theoretical Modeling of Lightning Interaction with Tall Objects

Detta är en avhandling från Stockholm : KTH Royal Institute of Technology

Sammanfattning: This thesis investigates various discharge processes associated with upward lightning initiated from tall objects, both experimentally and theoretically, trying to bridge the knowledge-gap in understanding the mechanisms of the phenomena.Upward lightning initiation has attracted a great deal of interest during recent years due to the proliferation of tall communication towers and wind turbines.  Simultaneous lightning current waveforms, close electric field changes, and lightning location system data associated with upward lightning measured at the instrumented Gaisberg Tower (GBT) near Salzburg in Austria from 2005 to 2009 is studied. It was observed  that a majority of upward flashes (87%) were self-initiated, namely they were initiated at the tower top without any nearby preceding discharge activity, whereas only  26 (13%) upward flashes were nearby-lightning-triggered, namely they were triggered by nearby preceding lightning discharges occurring within 100 milliseconds or so. This observation is different from the study conducted in Rapid City in South Dakota of USA where they observed the majority of upward flashes (80/81) initiated from ten tall towers were nearby-lightning-triggered. The possible reasons for self-initiated upward flashes dominating at the GBT could be (1) the field enhancement due to the shape of Gaisberg Mountain which is more than 800 m above the surrounding terrain of the city of Salzburg and (2) low altitude of charge region in the cloud during non-convective season (September to March) in Austria.Three modes of charge transfer had previously been identified in cloud-to-ground lightning, namely that happens during leader-return stroke sequence, during continuing current and during M-components. On the basis of the analysis of the simultaneous current, electric field changes, and high-speed video images measurements, a mixed mode of charge transfer to ground for ICC pulses and M-components in tower-initiated (upward) lightning is proposed. Compared to classical rocket-triggered lightning, occurrence of multiple branches of upward lightning from tall towers is very common. A newly illuminated or re-illuminated branch connecting to the already luminous channel attached to the tower during initial continuous current and continuing current following return strokes is frequently observed. A mixed mode of charge transfer to ground is composed of a new or decayed branch involving the leader/return-stroke mode of charge transfer to ground, and superimposing on the old channel with continuous (continuing) current mode of charge transfer to ground. This mode can explain ICC pulses exhibiting shorter risetimes, larger peaks, and shorter half-peak widths than “classical” M-components as reported in previous studies.Charge density expressions for the tall object and lightning channel when lightning strikes a tall object are derived based on continuity equation and current expressions given by the series point current source model representation of the tall object and lightning channel. Validation of charge density expressions was performed by comparing the total vertical electric field at ground level calculated by the traditional dipole technique with that calculated by the monopole technique using charge density expressions derived above. Distribution of charge density versus height along the tall object and along the lightning channel by considering the different values of current reflection coefficient at strike object base and the return stroke front speed in the lightning channel has provided an alternative way to explain the polarity inversion of vertical electrical field at very close range from strike object base.Characteristics of upward positive and bipolar lightning flashes observed from the GBT during 2000-2009 are studied in detail. A total of 26 upward positive flashes and 21 upward bipolar flashes were identified, which only accounted for 4% and 3%, respectively, of the total 652 flashes measured at the GBT during the 10-year observation period. There are only very few previous studies of upward positive and bipolar flashes. Bipolar flashes lower both negative and positive charges during different phases with the same flash. In the positive flashes, median values for flash peak current, flash duration, flash charge transfer, and flash action integral were determined as 5.2 kA, 82 ms, 58 C and 0.16×103 A2s, respectively. From simultaneous current and high-speed video measurements of one bipolar flash it is inferred that sub-branches are connected to one branch transferring both positive and negative charges successively, running into charge source regions of opposite polarity in the thundercloud.