Climate change and urbanization are converging to challenge city drainage infrastructure due to their adverse impacts on precipitation extremes and the environment of urban areas. Sustainable drainage systems have gained growing public interest in recent years, as a result of its positive effects on water quality and quantity issues and additional recreational amenities perceived in the urban landscape. This paper reviews recent progress in sustainable drainage development based on literature across different disciplinary fields. After presenting the key elements and criteria of sustainable drainage design, various devices and examples of sustainable drainage systems are introduced. The state-of-the-art model approaches and decision-aid tools for assessing the sustainable alternatives are discussed and compared. The paper further explores some limitations and difficulties in the application of the innovative solutions and suggests an integrated and trans-disciplinary approach for sustainable drainage design.

Estimating the expected annual damage (EAD) due to flooding in an urban area is of great interest for urban water managers and other stakeholders. It is a strong indicator for a given area showing how vulnerable it is to flood risk and how much can be gained by implementing e.g., climate change adaptation measures. This study identifies and compares three different methods for estimating the EAD based on unit costs of flooding of urban assets. One of these methods was used in previous studies and calculates the EAD based on a few extreme events by assuming a log-linear relationship between cost of an event and the corresponding return period. This method is compared to methods that are either more complicated or require more calculations. The choice of method by which the EAD is calculated appears to be of minor importance. At all three case study areas it seems more important that there is a shift in the damage costs as a function of the return period. The shift occurs approximately at the 10 year return period and can perhaps be related to the design criteria for sewer systems. Further, it was tested if the EAD estimation could be simplified by assuming a single unit cost per flooded area. The results indicate that within each catchment this may be a feasible approach. However the unit costs varies substantially between different case study areas. Hence it is not feasible to develop unit costs that can be used to calculate EAD, most likely because the urban landscape is too heterogeneous.

An accurate and rapid urban flood prediction model is essential to support decision-making for flood management. This study developed a deep-learning-technique-based data-driven model for flood predictions in both temporal and spatial dimensions, based on an integration of long short-term memory (LSTM) network, Bayesian optimization, and transfer learning techniques. A case study in northern China was applied to test the model performance, and the results clearly showed that the model can accurately predict the maximum water depths and flood time series for various hyetograph inputs, with substantial improvements in the computation time. The model predicted flood maps 19 585 times faster than the physically based hydrodynamic model and achieved a mean relative error of 9.5 %. For retrieving the spatial patterns of water depths, the degree of similarity of the flood maps was very high. In a best case scenario, the difference between the ground truth and model prediction was only 0.76 %, and the spatial distributions of inundated paths and areas were almost identical. With the adoption of transfer learning, the proposed model was well applied to a new case study and showed robust compatibility and generalization ability. Our model was further compared with two baseline prediction algorithms (artificial neural network and convolutional neural network) to validate the model superiority. The proposed model can potentially replace and/or complement the conventional hydrodynamic model for urban flood assessment and management, particularly in applications of real-time control, optimization, and emergency design and planning.

Automated detection and classification of sewer defects can complement the conventional labor-intensive sewer inspection process by providing an essential tool to classify sewer defects in a more efficient, accurate, and consistent way. This paper presents a convolutional neural networks (CNNs)–based model to automatically detect and classify six most commonly observed sewer defects (i.e., cracks, disjoints, obstacles, residential walls, tree roots, and normal categories) obtained from multisource CCTV images under various circumstances. Data augmentation techniques (including geometric and color transformations) are applied to enhance the model performance. The proposed CNN model is further compared with a state-of-the-art solution (retraining the SqueezeNet using defect images) by adopting transfer learning technique. An average prediction accuracy of 90% is achieved, indicating that the investigated defects can be well recognized by the model without any expert knowledge of sewer detection. There is a higher degree of confidence in predicting tree roots and disjoints, followed by residential walls and cracks. Results show that the prediction accuracy has increased by 15% thanks to data augmentation. Despite the transferred SqueezeNet model achieved a higher accuracy (95%), it cost almost 13 times the computation time of the CNN model. The study demonstrates the feasibility of the deep learning technology in the automated classification of sewer defects and advances the knowledge in the research field.