Skip to content
DraconiticResearch
Search Sources API
194,220 studies31 sourcesUpdated 43 min ago
Peer-reviewedCC BYJournal of Geophysical Research Machine Learning and Computation

OceanForecastBench: A Benchmark Data Set for Data‐Driven Global Ocean Forecasting

  • Yi Han1
  • Haoming Jia2,1
  • Xiang Wang1
  • Huizan Wang1
  • Wei Wu2
  • Jianming Zheng3
  • Peikun Xiao1
  1. 1College of Meteorology and Oceanography National University of Defense Technology Changsha China
  2. 2College of Computer Science and Technology National University of Defense Technology Changsha China
  3. 3State Key Laboratory of Mathematical Engineering and Advanced Computing Wuxi China
Open at sourceDOICite

Abstract

Global ocean forecasting aims to predict key ocean variables such as temperature, salinity, and currents, which is essential for understanding and describing oceanic phenomena. In recent years, data‐driven deep learning‐based ocean forecast models, such as XiHe, WenHai, LangYa and AI‐GOMS, have demonstrated significant potential in capturing complex ocean dynamics and improving forecasting efficiency. Despite these advancements, the absence of open‐source, standardized benchmarks has led to inconsistent data usage and evaluation methods. This gap hinders efficient model development, impedes fair performance comparison, and constrains interdisciplinary collaboration. To address this challenge, we propose OceanForecastBench, a benchmark offering three core contributions: (a) High‐quality global ocean reanalysis data over 28 years for model training, including 4 ocean variables across 23 depth levels and 4 sea surface variables. (b) High‐reliability satellite and in situ observations for model evaluation, covering approximately 100 million locations in the global ocean. (c) An evaluation pipeline and a comprehensive benchmark with 5 typical baseline models, leveraging observations to evaluate model performance from multiple perspectives. OceanForecastBench represents the most comprehensive benchmarking framework currently available for data‐driven ocean forecasting, offering an open‐source platform for model development, evaluation, and comparison. The data set and code are publicly available at: https://github.com/Ocean‐Intelligent‐Forecasting/OceanForecastBench .

Related

  1. LangYa: a large AI model for global ocean forecasting
    Nan Yang, Chong Wang, Zimeng Zhao +5
    中国科学通报：英文版Peer-reviewedCC BY-NCFull text22 citedOceanography & cryosphereOpenAlex
  2. A deep learning global ocean forecasting model with sub-daily and eddy-resolving resolution
    Yuan Niu, Qiusheng Huang, Xiaohui Zhong +10
    npj Climate and Atmospheric SciencePeer-reviewedCC BYFull text8 citedOceanography & cryosphereOpenAlex
  3. HourGlass: A probabilistic data-driven temporal downscaler for global and regional weather forecasting
    Magnus Sikora Ingstad, Mariana Clare, Olav Ersland +16
    arXiv (Cornell University)PreprintCC BYFull text7 citedAtmospheric science & meteorologyOpenAlex
  4. Njord: A Probabilistic Graph Neural Network for Ensemble Ocean Forecasting
    Daniel Holmberg, Joel Oskarsson, Erik Wikingsson +2
    arXivPreprintCC BY 4.0Full text5 citedAtmospheric science & meteorologyarXivOpenAlex
  5. Random analog prediction provides a benchmark for seasonal Arctic sea ice extent forecasting
    Faiq Raees, Francesco Paparella
    Scientific ReportsPeer-reviewedCC BY 4.0Full textAtmospheric science & meteorologyCrossref

Cite

BibTeXRIS
BibTeX
@article{han2026oceanforecastbench,
  title = {{OceanForecastBench: A Benchmark Data Set for Data‐Driven Global Ocean Forecasting}},
  author = {Yi Han and Haoming Jia and Xiang Wang and Huizan Wang and Wei Wu and Jianming Zheng and Peikun Xiao},
  journal = {Journal of Geophysical Research Machine Learning and Computation},
  year = {2026},
  month = jul,
  doi = {10.1029/2025jh000838},
  url = {https://doi.org/10.1029/2025jh000838},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - OceanForecastBench: A Benchmark Data Set for Data‐Driven Global Ocean Forecasting
AU  - Yi Han
AU  - Haoming Jia
AU  - Xiang Wang
AU  - Huizan Wang
AU  - Wei Wu
AU  - Jianming Zheng
AU  - Peikun Xiao
PY  - 2026
DA  - 2026/07/28
JO  - Journal of Geophysical Research Machine Learning and Computation
DO  - 10.1029/2025jh000838
UR  - https://doi.org/10.1029/2025jh000838
AB  - Global ocean forecasting aims to predict key ocean variables such as temperature, salinity, and currents, which is essential for understanding and describing oceanic phenomena. In recent years, data‐driven deep learning‐based ocean forecast models, such as XiHe, WenHai, LangYa and AI‐GOMS, have demonstrated significant potential in capturing complex ocean dynamics and improving forecasting efficiency. Despite these advancements, the absence of open‐source, standardized benchmarks has led to inconsistent data usage and evaluation methods. This gap hinders efficient model development, impedes fair performance comparison, and constrains interdisciplinary collaboration. To address this challenge, we propose OceanForecastBench, a benchmark offering three core contributions: (a) High‐quality global ocean reanalysis data over 28 years for model training, including 4 ocean variables across 23 depth levels and 4 sea surface variables. (b) High‐reliability satellite and in situ observations for model evaluation, covering approximately 100 million locations in the global ocean. (c) An evaluation pipeline and a comprehensive benchmark with 5 typical baseline models, leveraging observations to evaluate model performance from multiple perspectives. OceanForecastBench represents the most comprehensive benchmarking framework currently available for data‐driven ocean forecasting, offering an open‐source platform for model development, evaluation, and comparison. The data set and code are publicly available at: https://github.com/Ocean‐Intelligent‐Forecasting/OceanForecastBench .
SN  - 2993-5210
ER  -