Abstract: Data mining is the process of analyzing large sets of data and then extracting useful and relevant data. Data mining has many tools for predicting the behaviour allowing systems to make proper decisions. It can answer questions that are strong to resolve. Therefore, they can be used to predict meteorological data which is called as weather prediction. Weather forecasting is an important application in meteorology and has been one of the most challenging problem around the world. Predicting the weather is important to help preparing for the best and the worst climate. Clustering is the common data mining technique for finding hidden patterns in data. Clustering tries to group a set of objects and find the similarity between those objects. In this study, we are going to apply clustering algorithms on spatial datasets to group together climatic data for weather analysis. We will measure the performance of various clustering algorithms and record their drawbacks. We will propose a model that will try to overcome these drawbacks thus giving effective results.