Rain fall is one of the most complex and difficult element for carrying out weather forecast. Understanding and modeling of rainfall is a highly complex activity due to uncertain nature of atmospheric process that generates rain fall, as it is marred with large variation in scales of time and space. In this book, simplest type of neural networks viz Single Neural Network (SNN) is used in an independent sense for predicting various rain fall parameters. Combinations of these SNNs in an ensemble fashion resulting into Ensemble Neural Networks (ENN) with independent and inter dependent, single layer or multilayer has been studied using back propagation algorithm for training. Predictability of each model has been compared with the recorded actual data and the performance is gauged by five different error metrics. The performance of these ENN models have been found to be remarkable and have shown significant rise in prediction accuracy and speed.