Deliverable 4.3.5 Real-time monitoring and forecast platform to support early warning of faecal contamination in recreational waters

Abstract

Controlling urban floods and managing direct discharges of effluents into receiving waters from combined sewer overflows (CSO) are two major challenges faced by urban water management utilities. Discharges from large cities can have significant environmental impacts on marginal water bodies, affecting the quality of life in general, and recreational activities in particular (Marsalek and Rochfort, 2004; David and Matos, 2005; Passerat et al., 2011). These impacts can be exacerbated by climate change. First, the growing magnitude and frequency of extreme precipitation events (Groisman et al., 2005; Frei et al., 2006) will increase the number and severity of the discharges. Secondly, sea level rise and the resulting increase of salinity intrusion into the sewers can degrade the performance of wastewater infrastructures, affecting gate and pump operations and advanced biological wastewater treatment procedures. The ability to jointly manage an entire urban drainage and treatment system, towards an efficient and environmental-friendly operation of these infrastructures in a climate-change context, is often limited by the lack of reliable real-time information. Existing information systems are frequently devoted to specific parts of the network, lacking synoptic and cross-domain data. In addition, data and prediction tools are usually focused on physical variables alone. Water quality information is, at best, supported by very simple modelling approaches and limited sensors. More often, this information is sparse and not organized to provide efficient command and control procedures, taking into account climate change effects in the various domains. Timely prediction and monitoring of environmental conditions, as well as anticipation of hazardous events, are essential parts of recreational waters management. Monitoring and forecasting platforms can provide the necessary information for safe and efficient economic activities, and the protection of valuable natural assets, including the preservation of ecosystems and recreational areas. To this end, an innovative, real-time, coupled urban and estuarine platform was developed to support the integrated water quality management of wastewater systems, from the upstream catchment to the receiving waters. The platform efficiently integrates the monitoring and modelling of the different physical and water quality processes from the catchment to the receiving waters, at the appropriate spatial and temporal scales. It provides real-time web access to on-line hydrodynamic and water quality monitoring networks and short-term model predictions, based on a coupled modelling system that includes relevant interactions between the urban drainage system and the receiving waters, automatically compared with available on-line network data. This innovative decision support tool for urban drainage systems management is organized to provide tailor-made, automatic services to support the major operation tasks, drilled-down to the necessary details for decision support.The forecasting engine behind the platform provides hydrodynamic and faecal contamination predictions in all components of the systems (drainage network, wastewater treatment plant and estuary), accounting for all interactions between them. Prediction models are forced by regional forecasts whenever possible, and by real-time data otherwise. The accuracy of the predictions is verified through continuous, automatic comparison with data from the innovative on-line monitoring network, including both physical and water quality sensors (Rodrigues et al., 2014). Based on the platform’s data and model forecasts, an early-warning system is being proposed, supported by alert triggers both on the sewer network information and estuarine conditions. The system is being applied to the Lisbon demo, accounting for the impact of the combined sewage outfall from the Alcântara catchment on the Tagus estuary (David et al., 2014).