These items provide a performance comparison of machine learning stacking models and individual CMIP6 and GWTVDI predictions across three main scenarios: SSP1-2.6, SSP2-4.5, and SSP5-8.5.