UNCERTAINTY IN SEWER SEDIMENT DEPOSIT MODELLING: DETAILED VS SIMPLIFIED MODELLING APPROACHES

Abstract

The paper presents the results of a study in which the uncertainty associated with different sewer sediment modelling approaches: a detailed and simplified/parsimonious approach have been compared. Infoworks CS model coupled with a sediment transport code and a simplified sewer water quality model have been employed for the detailed and simplified modelling approach, respectively. The two approaches have been applied and compared to a single case study. The case study was selected as moderate storm events had occurred during the monitoring period. Flooding had been previously observed and this was thought to be caused by significant solids accumulation in the sewer network. The field campaign and the model simulations lasted a period of six months. The uncertainty of both modelling approaches were based on Monte Carlo based computational methods. This was a limitation for the detailed approach with regards to computational time. The simplified model approach was not affected by such a computational burden and enabled a more straightforward assessment of the uncertainty. The simplified approach, due to its structure, only provided a temporal estimate of uncertainty at the final section of the catchment. The detailed approach enabled an assessment of uncertainty at an individual pipe scale but only at the final time in the simulation period. When possible, comparison of the uncertainty estimations from both methods indicated comparable values of predicted uncertainty. Therefore a complementary use of both approaches would allow estimations of levels of uncertainty at both a spatial and temporal scale. The use of both model approaches may constitute a useful decision making tool for sewer system management.