Common EOFs: a tool for multi-model comparison and evaluation

Abstract With the increase in volume of climate model simulations for past, present and future climate, from various institutions across globe, there is a need efficient robust methods comparison and/or evaluation. This manuscript discusses common empirical orthogonal function analysis with step-wise algorithm, which can be used above objective. The method looks simultaneous diagonalisation several covariance matrices fashion ensuring thus monotonic decrease eigenvalues all groups, allowing therefore dimension reduction. applied to number tropospheric stratospheric fields main four reanalysis products, also historical CMIP6, Coupled Model Intercomparison Project (Phase 6). Monthly means as well winter daily gridded data are considered over Northern Hemisphere. shows consistency between mass mid-tropospheric reanalyses, but reveals significant differences 2 m surface-air temperature terms explained variance. CMIP6 models, on other hand, show reflected percentage variance leading EOFs inter-group variation ranging 5–10% troposphere about 25% stratosphere. Higher order statistics within modes variability, addition further merits discussed.