Unicorn: U-Net for Sea Ice Forecasting with Convolutional Neural Ordinary Differential Equations

TL;DR

Unicorn is a novel neural network architecture combining U-Net and neural ODEs to improve weekly sea ice forecasting by capturing complex spatiotemporal dynamics, outperforming existing models.

Contribution

The paper introduces Unicorn, a deep learning model integrating multiple time series images and neural ODEs within U-Net for enhanced sea ice forecasting accuracy.

Findings

Achieves 12% lower MAE in sea ice concentration prediction.

Improves sea ice extent classification by approximately 18%.

Demonstrates superior performance over state-of-the-art models.

Abstract

Sea ice at the North Pole is vital to global climate dynamics. However, accurately forecasting sea ice poses a significant challenge due to the intricate interaction among multiple variables. Leveraging the capability to integrate multiple inputs and powerful performances seamlessly, many studies have turned to neural networks for sea ice forecasting. This paper introduces a novel deep architecture named Unicorn, designed to forecast weekly sea ice. Our model integrates multiple time series images within its architecture to enhance its forecasting performance. Moreover, we incorporate a bottleneck layer within the U-Net architecture, serving as neural ordinary differential equations with convolution operations, to capture the spatiotemporal dynamics of latent variables. Through real data analysis with datasets spanning from 1998 to 2021, our proposed model demonstrates significant improvements over state-of-the-art models in the sea ice concentration forecasting task. It achieves an average MAE improvement of 12% compared to benchmark models. Additionally, our method outperforms existing approaches in sea ice extent forecasting, achieving a classification performance improvement of approximately 18%. These experimental results show the superiority of our proposed model.