The atmosphere and climate involve a number of nonlinear processes evolving on a wide range of space and time scales which are subjected to the property of sensitivity to initial conditions. In addition, modeling their evolution implies the use of inherent simplifications (e.g. parameterizations) for processes that are believed to play a secondary role. These approximations are in turn a source of uncertainty that could affect the dynamical properties of the system under investigation. Modeling is even further complicated when these parameters are time dependent like in climate prediction (e.g. CO2 increase). Nowadays, there is an increasing interest in investigating the dynamical properties of the combination of inherent initial condition and model errors, as well as in developing techniques that can improve the quality of the forecasts (both deterministic and probabilistic).
In this session, we invite papers analyzing the sensitivity (and predictability) of atmospheric and climate simulations to initial condition errors, model parameterizations and parameter variations (structural stability, model error dynamics…) in low-order, intermediate order or operational systems. We also invite papers dealing with forecasts corrections based on post-processing techniques or built-in stochastic schemes. Both theoretical and practical studies are welcome.