Documentation of the NASA/Ames Legacy Mars Global Climate Model: Simulations of the present seasonal water cycle

International audienceWe describe and document the physics packages in the legacy NASA/Ames Mars Global Climate Model, present simulations of the seasonal water cycle and how it compares with observations, assess the role of radiatively active clouds on the water cycle and planetary eddies, and discuss the strengths and weakness of the model and the implication for future efforts. The physics packages we describe include the treatment of surface properties, the ground temperature model, planetary boundary layer scheme, sublimation physics, cloud microphysics, the use of a moment method for tracer transport, a semi-interactive dust tracking scheme, and a two-stream radiative transfer code based on correlated-k's. With virtually no tuning of the water cycle and assuming the north polar residual water ice cap is the only source of water we find the model gives a reasonably good simulation of the present seasonal water cycle. No persistent clouds form over the residual cap, seasonal variations in column vapor abundances are similar to those observed, the aphelion cloud belt has about the right opacity, and surface and air temperatures are in reasonably good agreement with observations. The radiative effect of clouds does not significantly alter the seasonal and spatial variation of the moisture fields, though the clouds are thicker and the atmosphere somewhat wetter. As others have found cloud radiative forcing amplifies the mean meridional circulation, transient baroclinic eddies, and global thermal tides. However, it also changes the characteristics of forced stationary waves in ways that are not straightforward to understand. The main weakness of the model, we believe, is sluggish vertical mixing. Water is not transported high enough in the model and as a consequence the water cycle is too dry, the aphelion cloud belt is too low, and the mean meridional circulation is too shallow. These, we feel, could be remedied by some combination of non-local mixing, deep mountain-induced circulations, better horizontal and vertical resolution, and/or gravity wave drag. Efforts are now underway to study these issues as we are transitioning away from our legacy code to one with a more modern dynamical core