Using aggregation/skeletonization network models for water quality simulations in epidemiologic studies

Abstract

In recent years, epidemiologic studies and vulnerability assessments have come to rely on water distribution system models for refined estimates of street‐level exposure scenarios in conducting health studies and assessing potential contamination occurrences. A water distribution system network analysis at the street level can result in thousands to tens of thousands of hydraulic appurtenances such as pipelines, hydrants, and valves, making model calibration, verification, and system operation simulations time‐consuming and complex. Methods of aggregation, skeletonization, and network simplification have been used to reduce the complexity of the networks being analyzed. The issue then is whether network aggregation based on hydraulic characteristics and controls conserves mass so that the simplified network provides reliable results for the water quality analyses needed for epidemiologic studies and contamination assessment. In this study, a network aggregation methodology based on both hydraulic and water quality aggregation of an all‐pipes network was applied to a complex allpipes network described in the literature and used for an epidemiologic study. Simulation results obtained using the quality aggregation method, closely resembled both hydraulic (i.e., pressures) and water quality (i.e., concentrations) behavior of the original full‐sized (all‐pipes) system, while reducing system size by almost half.