Modélisation de la Terre : la vue d'ensemble[en] 

Steffen defines the Earth System as "the suite of interacting physical, chemical, biological and socio-economic global-scale processes and cycles, driven by energy, which provide conditions necessary for sustaining life on the planet". 

The author describes three different Earth System models in detail: 

• The extension of global climate models (GCMs), evolving from atmosphere-dominated, numerical weather prediction models "towards more holistic Earth System models" including ocean and sea-ice dynamics, atmospheric chemistry and carbon cycles. Earth System models based on GCMs offer "a more complete and detailed representation of the atmosphere". 
• Earth System models of Intermediate Complexity (EMICs), involving all critical biophysical processes in the appropriate balance. EMICs emphasise the interactions between components as much as the processes within them. 
• Integrated models that fully include the human dimension and the biophysical dynamics of the Earth System. 

Steffen explains that GCMs are the preferred Earth System models for simulating changes in climate over a century or less, whereas EMICs are more suited to exploring long-term dynamics, such as ice ages. 

One advantage that EMICs have over GCMs is "their ability to simulate abrupt changes and threshold effects", such as the transition of the Sahara savanna ecosystem into a desert 6000 years ago, says Steffen. 

However, neither EMICs nor GCMs have yet tackled "the most daunting challenge facing earth system modelling" – including human activity in the Earth System model. 

Steffen says that "modelling human dynamics in the context of the Earth System presents significant challenges that are yet to be solved", including the degree to which biophysical and human processes can and should be integrated, and the difficulty of selecting a suitable temporal range. 

He concludes that a single approach to building integrated Earth System models will not be the optimum strategy, and favours "comparing and synthesising results from a range of different approaches" instead.