Arctic and Antarctic Sea Ice Extent Forecasting using Hybrid LSTM Technique

Abstract

Recent changes in global climate patterns have triggered the accelerated melting of polar sea ice, especially in Arctic regions. A much faster rate of decrease in the sea ice extent (SIE) is observed at than previously expected. The Antarctic region, on the other hand, has shown a stable sea ice pattern throughout the last few decades. However, the southern polar region is not completely unaffected. Recent studies of the Bellingshausen and Amundsen Seas have shown a downward trend in sea ice. The SIE is crucial for regulating global climate patterns, ocean circulation, and human activities, including shipping and fishing. Hence, forecasting sea ice extent is vital for global economic planning and climatological studies. In this proposed study, time-series forecasting of five Antarctic and Arctic regions are evaluated using a hybrid convolutional long short-term memory (ConvLSTM) and a bidirectional long short-term memory (BiLSTM) and compared with a standalone long short-term memory (LSTM). This study uses regional sea ice extent data rather than considering the extent across entire hemispheres. With lower and stable RMSE scores across all lead times, the proposed hybrid BiLSTM model shows better performance in regional sea ice forecasting than does the standalone and ConvLSTM. The study also indicated that the climatic conditions of a particular region play a crucial role in forecasting efficiency, especially at longer lead times.