Datasets for "Advancing global sea ice prediction capabilities using a fully-coupled climate model with integrated machine learning"

This zip file contains all data presented in the manuscript "Advancing global sea ice prediction capabilities using a fully-coupled climate model with integrated machine learning" by Gregory et al, 2025. These include monthly-mean reforecast data for coupled model forecast experiments corresponding to SPEAR, Hybrid_IO and Hybrid_CPL. Additional data include ML training data, network weights, and normalization statistics for Hybrid_CPL (see 10.5281/zenodo.7818178 for respective Hybrid_IO data). A notebook is also included to recreate the plots from the manuscript. The corresponding Fortran source code which performs the sea ice bias correction has been saved as a version 1.0 release of the SIS2 code, provided here https://github.com/William-gregory/SIS2/releases. It is also provided in this Zenodo repository under SIS2.zip