Determination of the removal efficiency of a decentralized stormwater treatment system

Abstract

Due to the predicted impacts of climate change on the frequency of storm events, water managers are challenged to improve and adapt the current urban infrastructure. Cities need to be able to deal with the adverse effects due to more frequent and heavier rainfall. This implies the need for sustainable urban drainage systems (SUDS) that can deal with these challenges. SUDS can be seen as designs that can improve both, the quantitative as well as qualitative characteristics of stormwater. The study area in the south of Amsterdam will be equipped with a SUDS, namely a SediSubstrator L by Fränkische Rohrwerke. The objective is to remove particles from stormwater discharge to prevent clogging of the infiltration facility (AquaBASE) installed in sequence. Additionally, it can remove harmful pollutants which are adsorbed to the fine particles. This research aims to investigate the characteristics of the sediments in stormwater in theory and develop an appropriate sampling strategy to monitor the relevant parameters in the field. The long-term objective hereby is to determine the removal efficiency of the SediSubstrator L using a finite amount of parameters. To do so, a threefold approach composed by a detailed literature review, stormwater sediment sampling as well as a model simulation was used. Based on the identified parameters to be assessed, a specific monitoring and sampling setup is proposed to determine the removal efficiency of stormwater sediments in the study area. Along with the efficiency, the volume of sediments that are caught by the facility can be approximated. The latter needs to be removed once the entire storage volume of the sedimentation facility is filled. It is recommended that Waternet uses measured stormwater data to improve the model, while monitoring the efficiency under field conditions. Once the results of the model and the actual field measurements match, the model can be used to simulate the sediment removal for long time series. In this way the cleaning intervals can be predicted, while avoiding expensive long-term measurements to do so.