Weather prediction plays a challenging role in the climate systems. This study concerned with the data acquisition from the high resolution of earth observation by satellite system on Sentinel-2. For classifying the prediction events, the Support Vector Machine (SVM) model has yielded 89.67% accuracy. Long Short-Term Memory Model (LSTM) and Linear Regression (LR) Analysis help to quantify the weather prediction to improve the accuracy level of 95.5%. General Circulation Models of Earth’s Climate System for Weather Forecasting has used to forecast the temperature and humidity over a 1-week period. By utilizing historical weather data, including temperature and humidity to analyze the temporal dependencies effectively. The model's performance is evaluated using metrics such as Mean Absolute Error (MAE) of 0.236 and Root Mean Squared Error (RMSE) of 0.189. Results indicate that the LSTM model significantly improves prediction accuracy and compared with traditional methods. This approach not only enhances our understanding of atmospheric patterns but also provides a robust framework for future research in meteorological predictions.