Potential dangers of simplifying combined sewer hydrologic/hydraulic models

In large combined- and storm-sewer systems it is impractical to model every pipe, manhole, and component of the system. Time and budget constraints, combined with a lack of relevant input data, compel modelers to make assumptions in a quest to create a simplified model that adequately represents the hydrologic and hydraulic behavior of the system. Two of the most commonly applied simplification techniques are conduit skeletonization and subcatchment aggregation. This paper aims to highlight the potential dangers of making such simplifications, allowing modelers to make qualified simplifying assumptions. A base model and four simplified models were tested for a small (5.2ha) catchment in Chicago using the simulation packages ILLUDAS, HEC-HMS, and InfoSWMM. In addition, a base model and five simplified models were compared for a 341ha catchment using InfoSWMM. The effects of conduit skeletonization and subcatchment aggregation were found to be dependent on the simulation package being used and sensitive to the degree of simplification of the system. In using simplified models there is a danger that the user may not correctly predict the magnitude, timing, and shape of the outfall hydrograph.