Lightning presents a significant danger to electrical and electronic equipment, buildings and monuments, people and animals. Although our knowledge about lightning has significantly improved over last century, the development of new composite materials and the construction of an increasing number of tall structures present new challenges to the lightning protection techniques. Electric power generation based on wind energy is an extremely fast growing sector of renewable energy production. Lightning is the major cause of damage to wind turbines and their components. While modern wind turbines measure 150 m in height, future wind power generation units will be characterized by even taller turbines, as a result, it is expected that wind turbines will be more exposed to lightning strikes. Within this context, three key problems related to lightning protection of wind turbines and other tall structures were identified and studied within this thesis. First, the problem of the initiation of upward flashes from tall structures is addressed. Two techniques are commonly used to study this process in real scale: direct measurements of lightning currents using tall towers and rocket-triggered lightning experiments. An advanced analytical model was developed to study the relation between the parameters of ground and wire corona during triggered lightning experiments. The influence of several previously disregarded parameters was studied, which is a significant improvement over formerly used methods. An analysis of the initiation of upward lightning flashes for the Gaisberg and the Säntis Towers is presented that is based on the data from the European lightning location system (EUCLID) and direct lightning current measurements at the towers. Secondly, a new method for the estimation of the number of upward flashes from a tall structure is developed. The method is based on the analysis of the data provided by lightning location systems and it can therefore be applied to any tall tower or wind turbine covered by a lightning location system. Unlike available methods, it can be applied for structures in various locations with complex terrain. The method is applied to various structures around Europe. It is also shown that the method used in the current standards to esti-mate the risks due to lightning can result in underestimated values of lightning incidence and therefore in the choice of an inade-quate lightning protection level. Using the proposed approach, the proportion of upward flashes observed at different research towers used for lightning measure-ments is analysed. The obtained results confirm the validity of the parameters of downward flashes obtained at this tower and used in current standards on lightning protection. The application of the proposed method to evaluate lightning incidence to a group of tall structures is also presented in the thesis. The case of the Mont-Crosin wind turbine park was used to illustrate it, for which the lightning incidence is evaluated over several time periods corresponding to the construction periods of the wind turbine park. Finally, the last question discussed in this thesis is the distribution of lightning flashes over large areas. Specifically, the distribution of lightning flash density as a function of terrain elevation is analysed. Two improved methods were proposed. The lightning flash density distributions over the Alps and the Pyrenees are then evaluated using the proposed methods and the obtained results suggest that the lightning flash density grows over the whole altitude range.