Urban design decision support tools aimed at achieving desired outcomes – such as reduction of greenhouse gas emissions – must respond to the inherent complexity of urban systems, and the inherent uncertainties within measurement and inventory methods.  Moreover, they must accommodate the epistemological limitations of all models, arising from their dynamic relationship with the often self-modifying phenomena they are intended to model. Drawing on methodologies from other fields, we present here the outline of a methodology that meets that requirement, exploiting the capacity for iteration, empirical evaluation, and collaborative refinement over time.  We show how this methodology is suitable for application in a new generation of decision support tools for urban design.

Keywords: modeling methodologies; carbon reduction; greenhouse gas emissions.