Modeling storage capacity in Amsterdam’s sewer system


Courtesy of Innovyze

About Waternet

Waternet is a joint executive within the Amsterdam City Council and the regional management board Amstel, Gooi en Vecht, undertaking strategy, policy, design, planning, projects and administration for wastewater, groundwater and potable water. The organization controls 12 wastewater treatment plants, 3,500km of stormwater and wastewater pipelines, 500km of pressure pipes, and around 350 major pumping stations, with the system treating over 130M.m3 of wastewater annually. A team of four is responsible for dimensioning and monitoring of the sewerage and drainage systems.

Project background

The city’s main wastewater treatment works, Amsterdam West, was constructed in 2000, but the pipe network within the plant has suffered several failures, some that have been anticipated and remediated, and others that have been sudden and unanticipated.

As a result, the organization had to plan an optimum scenario to stop key pumping stations so that wastewater could be stored within the network for the duration of the repairs.

The city’s sewer system is complex – the drainage districts are generally small, and connect either in series to the treatment works or in parallel via a number of intermediate pumping stations. The city also has four booster pumping stations that can pump over greater distances and carry flows from the main upstream pumping stations.

The network itself is a mix of foul water, stormwater and combined sewers, the latter comprising 18% of the total. Five of the drainage districts are connected via a series of sewers and effectively form one larger district.

Wastewater from the 19 outlying drainage districts is input directly into the pressure system – during the repairs, this will be restricted to one pumping station entry point. The city has a policy of not storing this wastewater in its central pipes for any period of time because sedimentation could increase cleaning costs or cause polluted overflows if a storm were to occur. However, because of the frequent pipe breaks, this situation arises about once a year.

Using InfoWorks CS

As a result, Waternet decided to combine five all-pipe InfoWorks CS models to produce an emergency scenario that would determine the amount of storage capacity in the system, utilizing the information from the flow meters at the wastewater treatment plants and main pumping stations.

Using this data, the city center wastewater network’s storage capacity could be determined, which would enable the organization to establish the maximum duration of the wastewater treatment works closure.

There was some uncertainty about the storage levels – graphing from the closure point suggested that after 18 hours all of the sewers would be full, with the city center pipes and storage basins filling two hours later.

However the data from the city center measurement gauges provided a different picture, suggesting that the storage capacity in this critical area would not be evenly used throughout the scenario.

It was also clear that there was significantly more storage capacity during the night than during the day because of the greater amount of dynamic filling in daytime. The uncertainties determined that both night-time and daytime scenarios be modeled.

To achieve this, the 18,697 node combined model was used to determine storage in the system under special dry weather flow conditions. These models contained 17 internal weirs located between the storage basins and the wastewater system, with 234 external weirs that were omitted to simplify the model.

Modeling showed that in the night-time scenario some drainage areas, particularly those to the north of the city, did not fill because of the distance from the central system and the fact that the gravity sewers were at too high an elevation to allow filling.

The modeling also revealed the existence of some temporary pumping stations, which had been installed because of major construction work in the city center that had also temporarily decommissioned a major sewer. These alterations meant that some parts of the network could not fill, further affecting the ability of the system to store wastewater evenly.

Closure of the temporary pumping station near the construction works was also simulated in InfoWorks CS. This determined that at the pumping station where the 19 districts would inject their wastewater into the network at night, the internal weir level was reached in 17.5 hours.

A similar scenario in which a different pumping station was shut down produced a much slower fill rate, providing an extra 1.5 hours in which repairs could be undertaken and making better use of storage in the various districts.

Lessons learned

The modeling exercise revealed that under dry weather flow conditions it was possible to model storage volumes between two levels by integrating the additional pumped storage volume, enabling the organization to determine the effect of switching off various pumping stations to arrive at an optimum solution.

The emergency scenario provided valuable information on the storage capacity of the city’s assets, knowledge that the city is using to improve its models, design the network to better store dry weather flows and cope with flooding and other environmental issues.

This article is based on a presentation by Bas de Nijs of Waternet to the MWHSoft 2010 European Water and Flood Modeling Conference in September.

Customer comments

No comments were found for Modeling storage capacity in Amsterdam’s sewer system. Be the first to comment!