Another step along the path toward radically open modeling is the ability to refine an existing model through disaggregation and the ability to swap out modules from one simulation to replace a module that performs the same function in another model.

Examples of these processes include:

• Geographic disaggregation
  A model of the world’s economy at a global level can be disaggregated into multiple economic regions, such as North America, Europe, Latin America, Middle East, Africa, Asia.

• Sectoral disaggregation
  A model that simulates the energy sector at the macro level of total energy supply and demand can be disaggregated to separate out the various fuels that supply energy, such as oil, coal, natural gas, and renewables; or the various uses that are the basis for energy demand, such as electric power, transportation, home heating and cooling, industrial and commercial.

• Swapping out modules
  Numerous carbon cycle models are currently used in IAMs. One could imagine a user wanting to test the results of a particular IAM if it made use of a different carbon cycle model than the one it originally used. If carbon cycle modules could be swapped in and out easily from IAMs, undertaking such a test could be feasible, and a single IAM could be tested with multiple carbon cycle models.

To be able to swap in and out model modules, the interfaces between modules must be defined by explicit specification of what inputs and outputs the module must feature.

Return to Steps toward radically open modeling