User:Isentropiclift/Mars General Circulation Model

A Mars General Circulation Model (MGCM) is a type of computer program used to simulate atmospheric conditions and atmospheric-related surface processes on Mars, either during the present period of Martian exploration with unmanned spacecraft or in the Martian past. They are similar to the Global Climate Models used to simulate the Earth's atmosphere and often share program architecture with their terrestrial counterparts.

Differences Between Mars General Circulation Models and Other Forms of Atmospheric Simulations

MGCMs have two important features that distinguish them from other types of atmospheric computer simulations. They at least are two-dimensional and planetary in scale, so that they can represent the features of the time-averaged planetary circulation, such as the Hadley cell, distinguishing them from simple one-dimensional models and far more complex mesoscale models often used to simulate local radiative and boundary layer processes. MGCMs in general are capable of simulations in one, two, or three dimensions. MGCMs also should be distinguished from numerical weather prediction models such as those used on the Earth for weather forecasting. Numerical weather prediction models are regularly re-initialized to a base state using observational data that is a realistic representation of the true atmospheric state and then allowed to evolve a relatively short time so as to be predictive of future atmospheric conditions. MGCMs, however, are "spun up" from an idealized and unrealistic base state and then allowed to relax to more physically reasonable conditions. They then are evolved for long periods of time and thus cannot be regarded as predictors of future day-to-day weather conditions but as a representation of the average atmospheric conditions and their variability over the duration of the simulation.

History of Mars General Circulation Modeling

The first generally acknowledged MGCM was due to Conrad B. Leovy and Yale Mintz, who in the late 1960s modified an early terrestrial global climate model developed at UCLA based on very limited knowledge of the Martian atmosphere. Leovy and Mintz were successful in representing atmospheric condensation of carbon dioxide at the Martian poles and cyclonic activity in the mid-latitudes during the winter. In the wake of the atmospheric observations of the Mariner 9 and Viking missions, NASA funded the development of a MGCM at NASA Ames Research Center in Mountain View, CA, the Ames Mars General Circulation Model. In the following decades, the number of MGCMs has exploded as spacecraft observations of the atmosphere have multiplied. Operational MGCMs have been developed by NOAA's Geophysical Fluid Dynamics Lsboratory and the California Institute of Technology in the United States; Oxford University, Laboratoire de Meteorologie Dynamique du CNRS, and Instituto de Astrofísica de Andalucía in Europe (as a joint effort); and York University in Canada.

Basic Structure of a MGCM

At a minimum, a MGCM produces time-evolving numerical solutions to an appropriate approximation of the primitive equations of fluid dynamics, transporting heat and momentum and any other optional tracer through a space divided vertically and horizontally into a grid. The program architecture that performs this basic task is called the dynamical core. While the dynamical core may appear to perform relatively simple tasks, their implementation is numerically complex and the potential for instability in a dynamical core is high. The other key part of a MGCM is often called "the physics" or physical parameterizations. Physical parameterizations exist for two reasons. First, the approximated dynamical equations contain in-situ heating and cooling terms due to radiation, evaporation, and condensation. A MGCM thus needs some way to represent how the atmospheric gases and aerosols absorb, radiate, and scatter incoming and outgoing radiation, often simply called radiative transfer. An especially Martian problem in representing radiative transfer processes is atmospheric dust, whose concentration is highly time-variable and connected to transient dust devil and dust storm activity and simultaneously affects and is affected by the characteristics of the surface. Another challenge for physical parameterizations are condensation and sublimation of carbon dioxide, which is a significant term in the total Martian energy budget and also feeds into radiative transfer through the albedo of the polar caps. In fact, accurate representation of the high amplitude seasonal cycle of pressure that results from these processes is a key basic verification test for a MGCM. Physical parameterizations also are necessary to represent proceses that the dynamical core could represent if the grid were infinitesimally small. Since such a grid is computationally possible, various physical parameterizations are included in a MGCM in order to include the effects of these "sub-grid scale processes." Stephen Lewis has noted that a variety of global atmospheric models have been developed for Mars that have greatly simplified physical parameterizations for specific applications but MGCM dynamical cores. Thus, what is and what is not considered an MGCM may be a matter of the flexibility of its physical parameterizations so that they are as appropriate for a variety of applications, e.g., thermal tide behavior in the middle atmosphere and dust lifting from the surface.

Different MGCMs usually have different dynamical cores and different sets of physical parameterizations. Often some of these features are legacy from terrestrial models. For instance, York University's Global Mars Multiscale Model (GM3) borrowed much of its dynamical core from the Global Environmental Multiscale Model (GEM) developed for terrestrial weather and climate applications by the Meteorological Service of Canada. In at least one case, this type of borrowing has occurred in reverse. The California Institute of Technology's generalized planetary general circulation model, the Planetary Weather Research and Forecasting Model (planetWRF) has a dynamical core borrowed from NCAR's originally mesoscale Weather Research and Forecasting Model (WRF) but re-written for global simulations. WRF now has been expanded into a global model for terrestrial applications using planetWRF's dynamical core.

Specific Applications of MGCMs

Future Prospects in Mars General Circulation Modeling

List of Mars General Circulation Models