Multi-Task Deep Learning Based Spatiotemporal Arctic Sea Ice Forecasting

Thumbnail Image

Files

team-1.zip (45.56 KB)
2021-Multi-Task Deep Learning Based Spatiotemporal Arctic Sea Ice Forecasting-(Big Data 2021).pdf (2.14 MB)

Links to Files

https://ieeexplore.ieee.org/document/9671491

Permanent Link

https://doi.org/10.1109/BigData52589.2021.9671491
 http://hdl.handle.net/11603/25881

Collections

UMBC Information Systems Department
UMBC Faculty Collection
UMBC Student Collection

Author/Creator

Author/Creator ORCID

https://orcid.org/0000-0002-9933-1170

Date

2022-01-13

Type of Work

conference papers and proceedings
preprints
computer code
Text

Department

Program

Citation of Original Publication

E. Kim et al., "Multi-Task Deep Learning Based Spatiotemporal Arctic Sea Ice Forecasting," 2021 IEEE International Conference on Big Data (Big Data), 2021, pp. 1847-1857, doi: 10.1109/BigData52589.2021.9671491.

Rights

© 2021 IEEE.  Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

Subjects

UMBC Big Data Analytics Lab
UMBC High Performance Computing Facility (HPCF)

Abstract

Critical natural resources and processes in the Arctic depend heavily on sea ice. Thus, accurate and timely predictions of Arctic sea ice changes is important. Arctic sea ice forecasting involves two connected tasks: predicting sea ice concentration (SIC) at each pixel and predicting overall sea ice extent (SIE). Instead of having two separate models for these two forecasting tasks, in this paper we study how to use multi- task learning techniques and leverage the connections between ice concentration and ice extent to improve accuracy for both forecasting tasks. Because of the spatiotemporal nature of the data, we designed two novel multi-task learning models based on the CNN and ConvLSTM, respectively. Further, in conjunction with multi-task models, we developed custom loss functions which train the models to ignore land pixels and optimize for both concentration and extent when making predictions. Our experiments show that multi-task models provide better accuracy for a 1-month lead time than models that predict sea ice extent and concentration separately. Our accuracies are better than or comparable to results in related state-of-the-art studies. Our best model in SIC prediction outperformed the best existing SIC prediction model in the literature with 1.78% less error, and our best model in SIE prediction outperformed the best existing SIE prediction model with 0.283 million km2 less error.