Identification of El Niņo's dynamical regime in the Climate Forecast System

The atmospheric component of the NCEP Climate Forecast System (CFS) system currently consists of a global spectral model at horizontal resolution of triangular truncation of 62 waves (T62) coupled to an ocean model. The next upgrade to the CFS will involve higher horizontal resolution for both the atmosphere and ocean. High resolution allows a model to generate processes that are sufficiently chaotic to act as stochastic forcing of a more slowly varying dynamical system. We show that this is the case in the CFS by comparing ENSO behavior in the T62 model versus that in the T126 model. The comparison is quantified using Linear Inverse Modeling, which also allows us to compare the dynamical regime in the model with that represented in observations.