Empirical orthogonal function analysis of the diurnal cycle of precipitation in a multi‐scale climate model

Long‐term variability in the hydrologic cycle is poorly simulated by current generation global climate models (GCMs), partly due to known climatological biases at shorter timescales. We demonstrate that a prototype Multi‐scale Modeling Framework (MMF) provides a superior representation of the spatial and temporal structure of precipitation at diurnal timescales than a GCM. Results from empirical orthogonal function (EOF) decomposition of the boreal summer climatological composite diurnal cycle of precipitation in an MMF are compared to a GCM and satellite data from the Tropical Rainfall Measuring Mission. The eigenspectrum, principal component time series, and the spatial structure of leading EOFs in an eigenmode decomposition of the MMF composite day are a much better match to observations than the GCM. Regional deficiencies in the MMF diurnal cycle are manifest as localized anomalies in the spatial structures of the first two leading EOFs.