Lightning is the most beautiful as well as dangerous phenomenon appearing in nature. As this is a natural event, it cannot be stopped but some attempts are necessary which may enable us to minimize its disastrous effects. The observational site of this study is BIT Mesra, Ranchi, Jharkhand which lies in the eastern part of India. In my Ph.D. thesis, the behaviours of Atmospheric Surface Layer (ASL) parameters during lightning conditions have been observed. Flux analysis and spectral analysis of these parameters have done during lightning conditions. Based on this, an attempt has been made to develop a model to forecast the lightning occurrence during pre- monsoon season over this region. Artificial Neural Network (ANN) is used to forecast lightning. The results showed that the ANN can be used as an efficient tool for lightning forecasting over this region during pre-monsoon season. Apart from ASL parameters, vertical structure of lower atmosphere is also evaluated using radiosonde data on lightning days and a comparison has been made with the observations during clear days.