SCOR is inviting to contribute to their “Literature Review Spreadsheet” here.
This is a structured repository designed to catalogue and analyse research on AI-based sea ice models. It aims to provide a collaborative resource for tracking the evolution of machine learning approaches, including model architectures, training and validation datasets, and evaluation frameworks.
The document is organised into two main sections:
– Data Entry: This sheet is for recording information on AI sea ice models. It includes:
- Publication Metadata: Authors, year, title, and journal information.
- Model Specifications: Architecture (e.g., CNN, U-Net), physical constraints, and operational status.
- Data & Methodology: Training/validation datasets, input variables, spatial/temporal resolutions, and forecast lead times.
- Evaluation & Insights: Quantitative performance, limitations, data gaps, and suggested future work.
– Data Dictionary: A reference guide that defines each column in the Data Entry sheet, including expected data types and categories.
The Data Dictionary and the Column headings on the Data Entry sheet are locked for editing.
Please contact us if you have suggestions for modifying these.
We encourage you to contribute by adding new studies or refining existing entries.
Your input will help build a comprehensive, up-to-date resource for the community.
The spreadsheet is an open resource for the community. If you use the data in a publication or presentation, please acknowledge ORCAS, a PCAPS Task Team and SCOR working group 173.
Please feel free to share the spreadsheet with colleagues who may be interested in contributing. Also, anyone interested in joining the ORCAS community mailing list can sign up here.
If you are unable to access Google Sheets, please contact clare.eayrs@nyu.edu, and we will be happy to provide an alternative.
Find out more here.