Uncoupled urban canopy models replace the full-fledged atmospheric representation with a computationally light equivalent that can be used in a design optimization context. This type of model can be forced by boundary conditions that are directly measured or derived from an off-line meso-scale simulation. More often, it is driven, indirectly, by conditions measured at a nearby rural weather station (e.g., airport) via a hyper-simplified representation of the momentum/energy exchanges between the urban canopy layer and the surrounding atmosphere.
Within the urban canopy model, we replace the real urban geometry with an equivalent regularized arrangement of buildings and streets that is based on the average morphological parameters of the actual domain. We use a dynamic thermal network lumped-parameter model to represent the thermal interactions between buildings and the urban environment. These interactions are mostly mediated by the urban canyon air, itself in contact with the paved surfaces as well as the free atmosphere above the canopy. This model implements sophisticated aerodynamic and evapotranspiration schemes. The simulation runs for every hour of a standard year. The main outcomes are average building energy use and urban heat island intensity.
Two main categories of uncoupled urban canopy models exist: single-layer and multilayer. The single-layer scheme focuses on the overall exchange of heat, momentum, and moisture with the atmosphere right above the urban canopy. Air temperature and humidity are assumed to be uniform in the canyon. The multi-layer scheme is typically parameterized in terms of the horizontally averaged flow and scalar transport. This approach allows a higher resolution of atmospheric processes, but it incurs a higher computational cost.