No abstract

Recently fractal theory has been used to model the observed paths of lightning discharges. This paper extends previous work by using a fractal approach to describe the effects of tortuosity and branching of the lightning channel. In particular, we use the model to make predictions of the probability of lightning strikes to practical structures. Some of the specific estimates include the probability of strikes as a function of interception angle, predictions of the strike points on structures, and the probability of ''side strikes'' to tall structures. Significant polarity and geometric effects are shown. The shielding effects of nearby taller structures, shielding failure or '' breakthrough'' probabilities of so-called protected objects, and the effects of ground topography are also examined. The results are discussed in terms of the implications for the protection of structures against lightning.

A theoretical determination based upon a critical charge criterion and ionization processes is presented of the striking distance r s at which the upward leader is initiated from a struck object. The striking distance is shown to be leader-current dependent in the form r s ∝ i 2/3 . Comparisons are made with earlier, more numerically based models, which used an alternative equivalent-radius approach. The altitude dependence of the striking distance is also investigated. Since pressure, absolute humidity and temperature decrease with increasing altitude, the variation of streamer properties with altitude can be inferred from laboratory experiments. Calculations on this basis show that the striking distance increases significantly with increasing altitude, causing a corresponding increase in the risk of lightning strikes for given structures in mountain or aviation environments.

This paper presents the results of an analysis of the observed incidence of lightning strikes to a wide range of earthed structures in Hong Kong. The data were obtained for 161 structures and span the period 1988{1996. These structures were protected with lightning air terminals, placed in the optimum locations on each structure using a new positioning method which is an improvement of the simple`electrogeometric model'. The aims of the paper are fourfold, namely to assess (i) the e¬ectiveness of this new lightning protection positioning method, (ii) the validity of various attractive radius models, especially since many of the structures are located in high-density city blocks, (iii) the validity of a Poisson model for describing the strike frequency, and(iv) the e¬ect of any assumed height dependence of the median peak current on the attractive radius models.The results show a signi cant positive correlation between expected and actual strike frequency and con rm that the interception e¯ciency of the lightning protection system is in accord with the speci ed protection levels. In general, attractive radius and Poisson strike models for`free-standing' structures are found to be valid except for structures of height less than 100 m in dense city blocks. The results also show an in®uence due to positive ®ashes and a decrease in attractive radius if the peak lightning stroke current is assumed to be dependent on the structure height.

Diffraction of light of a visible spectral range by subwavelength metal gratings has been investigated experimentally and theoretically using rigorous electromagnetic calculations. It has been demonstrated that an effect of surface plasmon resonance (SPR), at which total absorption of light by metal grating can be observed, occurs under certain conditions. Large positive and negative Goos-Hänchen (GH) shifts occur near the SPR. It has been shown that the reflected beam is split into two parts, the relative powers of which depend on the incident beam width and the grating depth. The dependence of the GH shifts on the grating period and grating depth has been investigated for different incident beam widths. The high sensitivity of the GH shift on the incident angle of a light beam near the SPR has been demonstrated.