One decade of temperature change at the North Pole and South Pole using ANN backpropagation algorithm

Abstract

Global warming is an event in which the average temperature of the atmosphere, ocean and land rises. Changes in atmospheric temperature cause changes in the physical conditions of the atmosphere to become unstable, causing anomalies in weather parameters that cause climate change. In the last decade there have been temperature changes in the North Pole (Greenland) and the South Pole (Antarctica). The impacts caused by temperature changes are melting ice sheets, rising sea levels, extinction of species in large numbers, people living on the coast are threatened by tidal floods, while small islands can sink. To anticipate temperature changes, a model is needed that can forecast air or temperature conditions using Artificial Neural Network with Backpropagation method using Matlab software media. Data obtained from Data Access Viewer of NASA. After training and testing the data, the average value of the prediction results for the south polar region (Antarctica) is -47.910C with a Mean Square Error (MSE) value of 0.0015. Meanwhile, for the North Pole region (Greenland), the average value of the prediction results is -22.410C with an MSE value of 0.0069. By looking at the average value of the prediction results, it can be concluded that the temperature change for the South Pole region (Antarctica) is higher than the temperature change for the North Pole region (Greenland).