Since the mid-20th Century, the Arctic is warming about two times faster than the rest of the world. The urgent necessity of identifying the drivers for climate change and understanding the Arctic climate system motivates this study, which aims to provide more physical insights into the Arctic climate and improve weather forecasts in the Arctic. This thesis, I mainly focus on the variations and forecasting of the Arctic sea ice. Given the close relation between the Arctic energy budget and sea ice variability, an emphasis is placed on the meridional energy transport. The study continues in the direction of sea ice forecasts and a series of experiments of extended range sea ice forecasts with novel deep neural networks are presented.