Technological evolution survey allowed the broad use of 2D mathematical models for flood simulation. However, it is possible to happen that the answer required for a given problem does not need a 2D approximation or even does not configure a 2D surface solution. Urban flood simulations may fall in this second case, since urban structures may interact with flood flows and introduce discontinuities in the surface solution. This study aims to highlight the discussion about the physical interpretation and the modeller role as key elements in the interpretation and representation of physical systems. To support this proposal, the MODCEL, a Quasi-2D flow-cell model, was used in two different ways: in a detailed raster approach, similarly to usual 2D model uses; and in an alternative conceptual and interpretive way, using larger cells representing homogeneous portions of the territory. The modelling results showed that equivalent responses can be obtained. Although other models can be used and could offer different absolute results, the relative analysis offered sufficient support to the research hypothesis that a physical-based process conducted by a conscient modeller is crucial for model reliability and optimization.
Interpreting the physical reality and recognizing model simplifications are key points to guarantee simulation quality.
How much a detailed model is really needed to produce proper answers?
The use of Quasi-2D flow models built in a conceptual and interpretive way can be an alternative to the use of more complex and data demanding 2D models.