Numerical Weather Prediction models have been adopted by most of the meteorological services to issue weather forecasts. Despite the improvement in the models, there are various limitations, specifically Sub-Grid scale weather phenomenon that cannot be explicitly resolved with present models and are derived through statistical relationship, which is not a theoretically stable process and there is a strong need for searching alternative tools. This book provides practical applications of data mining for interpretation of weather patterns. Pre-processing of multidimensional weather data using hyper cubes has been demonstrated to speed up storage and retrieval of weather variables. Association of rainfall with movement of LPS has also been done along with rainfall forecasting using Artificial Neural Networks. k-means clustering technique has been applied on the clusters of ensemble of derived weather variables for real life cases of tornado and cloudburst to locate patterns conducive to formation of these events. This approach should help the researchers, meteorologists and practitioners who may be looking forward to understand this unique blend of meteorology and computer science.