Climatic Signal and Climatic Noise in Lorenz's Low Order Model of the Atmospheric Circulation

The climate state may be defined by a set of statistics computed over a very large number of replicas of the Atmosphere (ensemble), each replica evolving independently of each other, but all replicas subject to the same boundary conditions (external forcing). Such dynamic approach of the climate state is particulary adequate to define the concepts of climatic signal (linked to external forcing) and climatic noise (associated to distinct events of the same climatic state). Lorenz's low-order model of the general circulation is a useful tool to study the atmospheric signal and noise when the system is subject to prescribed forcing (e.g. seasonal forcing). This is due to the richness of the model as a dynamical system together with its low computacional cost that allows building ensembles with a large number of replicas. In our study we analyse the dynamical behavior of the atmospheric circulation based on a large set of numerical integrations (≈10000). The climate state is analysed and particular attention is devoted to the problem of separating the climatic signal from the climatic noise for different types of seasonal forcing.