Determination of key sensor locations for non-point pollutant sources management in sewer network

Optimal selection of monitoring sites in cities for SARS-CoV-2 surveillance in sewage networks

Selecting sampling points to monitor traces of SARS-CoV-2 in sewage at the intra-urban scale is no trivial task given the complexity of the networks and the multiple technical, economic and socio-environmental constraints involved. This paper proposes two algorithms for the automatic selection of sampling locations in sewage networks. The first algorithm, is for the optimal selection of a predefined number of sampling locations ensuring maximum coverage of inhabitants and minimum overlapping amongst selected sites (static approach). The second is for establishing a strategy of iterations of sample&analysis to identify patient zero and hot spots of COVID-19 infected inhabitants in cities (dynamic approach). The algorithms are based on graph-theory and are coupled to a greedy optimization algorithm. The usefulness of the algorithms is illustrated in the case study of Girona (NE Iberian Peninsula, 148,504 inhabitants). The results show that the algorithms are able to automatically propose locations for a given number of stations. In the case of Girona, always covering more than 60% of the manholes and with less than 3% of them overlapping amongst stations. Deploying 5, 6 or 7 stations results in more than 80% coverage in manholes and more than 85% of the inhabitants. For the dynamic sensor placement, we demonstrate that assigning infection probabilities to each manhole as a function of the number of inhabitants connected reduces the number of iterations required to detect the zero patient and the hot spot areas.

Is Clustering Time-Series Water Depth Useful? An Exploratory Study for Flooding Detection in Urban Drainage Systems

As sensor measurements emerge in urban water systems, data-driven unsupervised machine learning algorithms have drawn tremendous interest in event detection and hydraulic water level and flow prediction recently. However, most of them are applied in water distribution systems and few studies consider using unsupervised cluster analysis to group the time-series hydraulic-hydrologic data in stormwater urban drainage systems. To improve the understanding of how cluster analysis contributes to flooding location detection, this study compared the performance of K-means clustering, agglomerative clustering, and spectral clustering in uncovering time-series water depth dissimilarity. In this work, the water depth datasets are simulated by an urban drainage model and then formatted for a clustering problem. Three standard performance evaluation metrics, namely the silhouette coefficient index, Calinski–Harabasz index, and Davies–Bouldin index are employed to assess the clustering performance in flooding detection under various storms. The results show that silhouette coefficient index and Davies–Bouldin index are more suitable for assessing the performance of K-means and agglomerative clustering, while the Calinski–Harabasz index only works for spectral clustering, indicating these clustering algorithms are metric-dependent flooding indicators. The results also reveal that the agglomerative clustering performs better in detecting short-duration events while K-means and spectral clustering behave better in detecting long-duration floods. The findings of these investigations can be employed in urban stormwater flood detection at the specific junction-level sites by using the occurrence of anomalous changes in water level of correlated clusters as flood early warning for the local neighborhoods.