Enhanced Ensemble Modeling in Spatial and Temporal Data for Climate Change Prediction

Precise climate prediction is important for comprehensively tackling urgent ecological problems, but typical models still battle with completely catching extraordinarily detailed trends. This research work thoroughly presents an ensemble-based methodology incorporating LSTM for each temporal datum in conjunction with CNN for each spatial attribute, thus heightening predictive correctness. The proposed model shows high precision by optimizing predictive accuracy extensively with ensemble techniques like stacking and increasing, achieving a Root Mean Square Error (RMSE) of 0.78°C, Mean Absolute Error (MAE) of 0.57°C, and a Mean Absolute Percentage Error (MAPE) of 1.99%. Also, an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score of 0.9063 points out the strong explanatory power. The Future work plans to extend the model to predict climate variables like precipitation and humidity, broadening the effect on climate forecasting.