The expected effects of climate change on increased and more frequent rainfall events ask for more innovative solutions to manage urban stormwater. Sustainable Urban Drainage Systems (SuDS) offer an eco-friendly method to disconnect stormwater from the sewer system. The Municipality of Rotterdam, the Netherlands, has integrated multiple SuDS into its drainage network, including bioswales. Bioswales are vegetated areas that slow down, collect, and filter (storm) runoff. However, uncertainty exists regarding their performance under different conditions. This thesis aims to answer the following research question: How do bioswales perform under various conditions, as evaluated by a hydrological groundwater model?

The bioswale groundwater model used in this thesis, developed by Deltares, utilizes the Unsaturated Zone Flow (UZF) package of MODFLOW to simulate the hydrological response of bioswales. The model was calibrated and validated using existing monitoring data, and a one-at-a-time sensitivity analysis was performed to identify the most influential factors affecting bioswale performance. The case-study bioswale was tested under design storms reflecting current and 2050 summer and winter conditions, as well as prolonged wet winter rainfall. Two design scenarios were proposed to improve bioswale performance.

The case-study calibration results showed that the model could realistically simulate water levels and discharges. However, the existence of preferential flow in the unsaturated zone, not accounted for by the UZF package, led to a time-lag in modelled drain discharge. The sensitivity analysis indicated that infiltration parameters strongly influence emptying time, peak discharge, and time-lag in the model. The performance assessment showed that the case-study bioswale met the emptying time criterion, but the peak discharge limit was exceeded during summer events. Simulating prolonged wet winter rainfall showed that consecutive rainfall events could be more critical regarding winter bioswale performance, compared to a single winter design storm. The bioswale design improvements demonstrated that relocating the drain from the centre to the side of the bioswale, thereby increasing the distance water needs to travel, significantly reduced peak discharge, though at the cost of longer emptying times. Widening the bioswale increased storage volume; therefore, the connected paved surface area could be increased, but the effect of adding additional drains on bioswale performance was limited.

To increase the understanding of bioswale performance, further empirical research on vegetation, macropores, and preferential flow is recommended, along with improvements to the modelling of these processes. In terms of model application, the bioswale groundwater model, with some adjustments, can be applied to other SuDS types that might be less sensitive to the natural influences of vegetation change and macropores. Combining the modelling of individual bioswales and SuDS, as done in this study, with urban-scale modelling could significantly improve Rotterdam’s climate resilience.

In recent years, Sustainable Urban Drainage Systems (SUDS) have gained popularity for managing stormwater in urban areas. However, effective asset management of these systems remains challenging due to the widespread reliance on reactive maintenance. This thesis examines the condition of dry swales in Utrecht, Netherlands, and their role in managing stormwater under the municipality’s current maintenance framework.

A comprehensive visual inspection of 210 dry swales was conducted, alongside an evaluation of the area characteristics influencing component failures. The results indicate that only 67% of the swales function properly, with overflow failures being the most common issue, particularly due to clogging; 5% of swales were found to be in a failure state. Conversely, the vegetation layer emerged as the component requiring the most continuous maintenance, with only half of the swales sustaining a well-functioning vegetation layer. Notably, complete clogging issues in swale overflows were linked to a high percentage of impermeable areas within the catchment and a smaller length-to-width scale.

The analysis further revealed that filter basins exhibited the highest percentage of functioning components, emphasising the critical role of the infiltration process. However, even minor deterioration in these basins could significantly impact overall swale performance. The study also identified that a larger impervious area within a swale’s catchment correlates with increased failure likelihood in various components. Factors such as tree density in a catchment area, impervious area, and the age of the swales were shown to contribute to issues like standing water and sediment accumulation in filter basins.

The investigation into sediment accumulation at overflows indicated that sediment tends to build up more in catchment areas with higher percentages of impermeable surfaces, although this phenomenon is influenced by multiple factors that warrant further research.

The findings underscore the necessity of improving the current reactive asset management of swales. By establishing a comprehensive database of inspection data, future research can better inform predictive asset management applications, incorporating parameters such as impervious area percentage, tree density, and swale age into machine learning models to predict swale failures.

Ultimately, the research emphasizes the importance of effective asset management for SUDS, not only to enhance swale efficiency but also to mitigate urban flooding risks. Regular inspection and monitoring are essential to understanding swales’ functionality and degradation over time, informing design improvements and promoting the adoption of sustainable urban drainage systems. This study contributes to the broader goal of creating more resilient and sustainable urban environments through robust asset management of SUDS.

This research has investigated how Sewer Water Harvesting (SWH) can be applied to provide a climate-proof fresh water source to support Urban Green Spaces (UGS) in Amsterdam. SWH is the process of extracting raw municipal sewage from the sewer and locally treating this to provide fit-for-purpose water in a dense urban environment while treatment residuals are discharged back into the sewer. SWH can help to meet the increasing water demand of UGS in Amsterdam, which experiences exacerbated dry periods as a result of climate change, while conventional water sources are unlikely to meet this demand.

The overall aim was to provide a conceptual design example of how SWH could be applied in the Amsterdam context to uncover what kind of impact can be achieved and advise on how SWH can be implemented From an analysis of potential applications, irrigation of UGS during dry periods was selected for the focus of the study. Suitable locations were identified, from which the Vondelpark was selected as study area for this research. Quality requirements for irrigation water and discharge of treatment residuals were determined. The water demand of the study area was determined by modelling the soil moisture balance using transformed weather data, taking into account climate change. Based on these requirements, a conceptual design of an SWH-unit comprised of fine screening, MF, NF and UV steps. To evaluate this potential impact for Amsterdam as a whole, the findings from the study area were extrapolated. The cost of SWH were compared to alternative water sources and the potential direct economic benefits. This demonstrated that costs of SWH are acceptable and can be further decreased. Furthermore, the potential impact on plant and soil health was
evaluated. Interviews with stakeholders identified barriers and opportunities of SWH and resulted in some recommendations for larger scale implementation.

The results of this research indicate that SWH can provide a new and reliable water source during dry periods to support UGS. SWH-units can be designed as mobile and modular units that can for a large part be operated and monitored remotely. The results further demonstrate that potential negative environmental effects can be prevented or mitigated and SWH can even improve the plant and soil health of UGS. From an engineering perspective, challenges related to the water quality are unlikely to be insurmountable. However, three aspects still require a significant amount of time and investment before SWH can be implemented on a larger scale. These are: (1) the lack of regulatory framework, (2) the unresolved responsibility for operation and (3) extensive water quality testing and environmental impact assessment. To accelerate innovation it is recommended to start as soon as possible with addressing these remaining issues. Commercial operation of SWH can provide an interesting opportunity, all the more so because SWH can also be used for household or industrial applications. The involvement of a wider variety of stakeholders can further help to overcome the remaining barriers.

Urban wastewater systems can impact the urban ecology by untreated wastewater discharges through combined sewer overflow (CSO) events, or by partial treatment of the wastewater at the wastewater treatment plant (WWTP). CSO events can cause oxygen depletion, eutrophication, and the discharge of pathogens. The partial treatment of the wastewater causes an increased concentration of ammonium in the WWTP effluent, which can lead to toxic ammonium levels in the receiving river water. These problems can be (partially) mitigated by optimizing the existing infrastructure. Optimization of the available storage will help handle the increased pressure on the urban drainage system (UDS) and the stricter environmental regulations simultaneously. A method to control the dynamic performance of the combined sewer system is Real-Time Control (RTC). A RTC strategy controls the combined sewer system dynamically based on real-time information about the system state. This research aims to develop a RTC strategy that decreases the negative ecological impact of the combined sewer system on the river by optimizing the available in-sewer volume. By doing so, the objectives to reduce the total amount of spilled CSO volume and to decrease the ammonium peaks towards the WWTP should be met. This research is applied to the case study of Geldrop-Mierlo, this is a municipality located in the UDS of Eindhoven. The trade-off between those two objectives was explored in the Wastewater Process simulator WEST. Rainfall events with a maximum intensity of 3.1 mm/hr and higher or rainfall events with maximum intensity < 3 mm/hr and total rainfall depth of > 4.8 mm, were found to be more likely to cause DO dips. The objective function which is used in the optimization process is dependent on the forecasted rainfall and the trade-off described above. The UDS is modeled in a full-hydrodynamic (FH) model and a simplified conceptual model. The conceptual model is made to reduce the computation time. The catchment of the FH model is split up into 3 different catchments, and each is modeled as a reservoir in the conceptual model. The characteristics of each reservoir are dependent on the characteristics of these catchments in the FH model. The characteristics that are included are storage curve, outflow dynamics, and CSO dynamics. Both models are calibrated and validated. The UDS is controlled with the Model Predictive Control (MPC) methodology using a Genetic Algorithm (GA) to find the optimal solution to minimize the negative ecological impact of the UDS on the river. Both the FH model and simplified model are used in the MPC optimization. Based on the analysis of the case study, the optimization results show that the impact of the MPC procedure on the receiving river is not significant. The reasons for this are location specific, but the main findings are that 1) the hydraulic constraints of the catchments restrict the MPC procedure from working, 2) although the calibration results of the conceptual model indicated accurate results, this does not guarantee that the model is also accurate enough to use in the MPC procedure.

Water utilities face many challenges, including pipe bursts that cause significant non-revenue water losses. Detecting those bursts early is important for the water sector in its path to achieve sustainable water resource management. This study presents a scalable data-driven methodology for burst detection in water distribution systems that is based on Long Short-Term Memory (LSTM)-based neural networks (NNs) and includes two stages: prediction and classification. Time-series of hydraulic (flow and pressure) signals are fed to the LSTM, whereas domain (time) features of the next time step are fed independently to regular neurons. These two streams of information are then concatenated to predict the values of the hydraulic features of the next time step. The model is trained on normal conditions only, so that when fed with data corresponding to a burst, the predictions will mismatch the observations. Comparison of the predictions to the observations is quantified though an error function, which is then used for classification. Specifically, a variable error threshold that corresponds to a pre-defined extreme percentile of the error distribution is used to discern bursts from normal conditions. The methodology is corroborated on two different types of bursts: (a) real bursts in district metered areas (DMAs) in the United Kingdom and (b) simulated fire hydrant leak tests in the same DMAs. For the real bursts, sensitivity analysis of the algorithm is performed to assess how data resolution and error threshold affect the performance. The flexibility of the method is studied for the simulated fire hydrant leaks, where additional information streams from new sensors are incorporated in the model by means of applying transfer learning and fine-tuning. The results obtained demonstrate that this scalable LSTM-based methodology works reasonably well in real-life settings and can successfully identify burst events, both real and simulated, even in DMAs with a small number of installed sensors. Furthermore, it is assessed how the flexibility of the LSTM neurons is pivotal for burst detection when utilizing a varying number of sensors.

From 2018 to 2020, the city of Breda (NL) faced drought, with significant economic losses and water scarcity problems. The water board within which the city is located began to consider new possible water management practices, due to the unsustainability of the current ones. This study aims to investigate how a change of the current paradigm can improve the current approach in water management, shifting from a linear to a water management circular approach, and from command-and-control to adaptive. Starting from the case scenario of Breda, the study considers how through the reuse of management wastewater treatment plant effluent and stormwater, and the application of sewer mining units, it may affect the circularity and adaptability of the system. Then, based on Breda's experience, an attempt is made to investigate the concepts of circularity and adaptability. \\To be able to understand the current water situation and approach in the city of Breda, a water balance for the year 2020 was drafted. The water balance for 2020 for the city of Breda shows a positive value of 50 mm/year. The definitions of circularity and adaptability for a water management system were determined, and for each of the concepts a framework has been developed, to assess the level of circularity and adaptability before and after the interventions. Then, a workshop with local experts in the field was organized. Different strategies of interventions were proposed to them, in order to evaluate their suitability. Then, a final strategy was formulated, where wastewater effluent is employed as irrigation water for agriculture in the south of the municipality, stormwater discharge into wetlands to recharge aquifers and three sewer mining units in the rural area of Breda. These interventions showed a positive effect on the water system, since the volume of water stored in the system has increased, giving an overall better performance of the circularity and adaptability indicators. Moreover, the final strategy shows a better water balance of 100 mm/year. This research aims at creating a better understanding of concepts like circularity and adaptability. The study shows that there is a potential for re-using water to enhance the overall performance of the water system under certain conditions.

Leakage is the main source of water loss in water distribution networks (WDNs). Therefore, leak detection and localization technology is a major concern for water utilities to save water and meet the ever-growing water demand. This study presents two methodologies for leak localization in District Metered Areas (DMAs): (1) a model-based method using the hydraulic model, flow and pressure measurements, as well as leak flow; (2) a data-driven method that relies on graph-based interpolation. The performance of the model-based method is proved to be negatively affected by model errors and limited sensors. To solve these two problems, on the one hand, flows and residuals between observed and model-simulated data in the non-leak situation are used to develop a residual model to calculate offset values for model output correction. On the other hand, graph-based interpolation is introduced to create ‘virtual’ sensor measurements in the presence of a limited number of sensors. The data-driven method proposed in this work uses graph-based interpolation to estimate the head signals at the nodes without sensors and subsequently create pressure maps. Leak localization is achieved by comparing pressure maps in the non-leak and leak situations. In this process, this methodology does not require a well-calibrated model and leak flow information. Both two methodologies are tested on fire hydrant leak tests in DMAs in the United Kingdom. Results obtained by using the model-based method illustrate the positive impact of model output correction on localization results and the performance of this method under different conditions such as different times of the leak and different sizes of leak flow. The data-driven method performs fairly well in DMAs with a higher spatial density of sensors. Furthermore, the results of both two methods are compared to demonstrate the suitability of the methods in different cases.

Plastic pollution is one of the most challenging global environmental problems. Currently, more than 1000 rivers transport approximately 80% of the plastic influx into the oceans. Naturally, more and more companies are interested in tackling this problem. One of them is Noria Sustainable Innovators, a company based in Delft (Netherlands). It is focussed on the detection, removal, and reuse of plastic from Dutch waterways. The company has the ambition to automate the detection of plastic for a wide range of applications. The quantification of plastic and understanding its spatiotemporal variability are crucial for the mitigation of plastic pollution. Current monitoring methods (e.g., visual counting) are tedious, time-consuming, and labour-intensive. Furthermore, the detection of different plastic debris objects could provide more insight about the source of plastic pollution. This thesis explores the feasibility of automating plastic detection in waterways using modern deep learning (DL) algorithms named convolutional neural networks (CNNs) with image classification and object detection techniques. To train these models, a large dataset is required. Due to the unavailability of data, images were gathered in a controlled environment with two GoPros and a Huawei P30. The data was aggregated during sunny and cloudy conditions, different camera heights (2.7m and 4.0m) and angles (0 and 45 degrees). For the simplest case (2.7m/0 degrees), a maximum accuracy of 87.6% was obtained for the multiclass classification of plastic debris in images, using the DenseNet121 model. By applying a majority vote for the three best performing models (DenseNet121, ResNet50 and InceptionV3), the accuracy could be increased to 91%. A qualitative and quantitative analysis found that the following factors influence the model performance negatively: presence of organic material, wind, transparent objects, submerged objects, small objects, overlapping and occluding plastic debris, sun glint and reflection of other objects on the water surface. Sunny conditions yielded a lower accuracy (79%) than cloudy conditions (90%), explained by the presence of sun glint. By applying object detection, the error sources influencing the model performance could be reduced. For training and testing data from 2.7m/0 degrees on one class (‘plastic debris’), the YOLOv4 model yielded an accuracy of 95.61% (GoPro). For four classes (e.g., plastic bottles, other plastic, paper, metal tins) an average accuracy of 66.04% was found, indicating that the model experienced difficulties distinguishing different floating debris in water. Furthermore, it was also shown, that the use of a different image source (Huawei P30), does not have a negative effect on the accuracy (96.63%) compared to the original image source (GoPro). Furthermore, due to height differences, discrepancy in object sizes and different camera settings, the trained model had large difficulties generalizing to a dataset from Indonesia (12.23%). On the other hand, training on the dataset from Indonesia and testing on the dataset from 2.7m/0 degrees achieved an accuracy of 63.51%. Although the error sources could be reduced, the model was still negatively impacted by small, transparent objects, submerged objects and the presence of sun glint. This study clearly showed that Deep Learning-based computer vision can detect floating plastic debris with a high accuracy and have the potential to automate the process of plastic detection in the future. Future work would comprise the following aspects: sensor improvements (polarising filter for sun glint and multispectral sensor for continuous monitoring), data collection from the natural environment and different image sources, implementation of guidelines for Citizen Science platforms, addition of an object tracking module for monitoring (YOLOv4) and focussing on the detection of specific plastic debris objects after the removal from waterways.

The current intensification of the hydrologic cycle, in combination with expanding settlements in flood prone areas, makes an increasing share of the global population exposed to flood risks. Many parts of the world are, however, still lacking the data needed for flood risk management and risk reduction. The recent development of information and communication technologies has remarkably lowered the costs to collect data for flood resilience, which has accommodated the rise of community mapping projects to fill data gaps in resource-strained environments. This thesis utilises drainage data collected by the Ramani Huria community mapping project in Dar es Salaam, Tanzania, to investigate if community mapped drainage data can improve flood predictions on neighbourhood scale. A coupled 1D-2D hydrodynamic model is developed of Kijitonyama ward, and is run with and without Ramani Huria’s drainage data implemented in the 1D schematisation. The simulated flood depth for the scenarios is validated with citizen’s observations on flood depth during a rain event on 3 March, 2019. The developed model is then applied to investigate the impact of solid waste accumulation in the drainage system on floods, by closing the drainage segments that were recorded as blocked in Kijitonyama ward by Ramani Huria staff the morning after the simulated event. An experimental scenario is also run, to examine the impact of blocked culverts. The results show that community mapped drainage data indeed can enhance the performance of hydrodynamic models, as the model output corresponds better with the validation data when implementing Ramani Huria’s drainage data in the 1D schematisation, compared with a scenario run with only a 2D schematisation. The scenarios run with solid waste blockages do not influence the model output when comparing with citizen’s observations, but increase the water level in the drainage segments located upstream of the blockages.