Aqua-planet configuration

of the UM

One of the really good things about aqua-planet integrations is that they don't need to run for very long in order to establish robust results. This makes the aqua-planet ideal for sensitivity tests. Neale and Hoskins recommend integrations of 18 months in length, run from initial conditions from a full GCM experiment. They then discard the first 6 months as the model adjusts to the ocean-covered globe. If running from initial conditions from a previous aqua-planet experiment, this adjustment time could be reduced to 1-3 months, depending on how different the new experiment is from that from which the initial conditions are taken.