State of the art of LCA applied to urban water infrastructure
A recent review revealed that there is now more than 250 research articles on LCA applications to urban water infrastructure (including wastewater, drinking water, stormwater, and integrated urban water systems) since the first applications in the 1990s (Byrne et al., 2017b). It is now clear that LCA is a valuable tool to explore life-cycle environmental burdens across full urban water systems at the city scale, and elucidate the broader implications of design and operational decisions on environmental impacts (Corominas et al., 2013; Guest et al., 2009; Lane et al., 2015; Loubet et al., 2014).
Urban water infrastructure studies should correspond to the primary objectives of urban water systems, which are to manage water quantity and water quality in a way that protects local and regional human and environmental health. To develop this effort, emerging approaches to include spatial considerations, water quantity, public health, and economic and social factors should be considered to better inform decision-making (Figure 4). Some of these opportunities are for advancements within LCA (e.g., apply spatially differentiated characterization factors) while others are for integration of LCA within broader sustainability assessment frameworks (e.g., those which incorporate quantitative risk assessment) (Byrne et al., 2017b).
Stormwater consideration in LCA
While it is well known that growing urbanization with its increasing intermittent discharge (ID) flows due to stormwater runoff is a leading contributor to impairments of the quality of aquatic ecosystems (Gosset et al., 2016) yet there is a dearth of LCA research on this topic. The majority of stormwater LCA studies have focused on the quantitative aspect (runoff volume reduction) in specific stormwater control measures (e.g. gray and green infrastructures)(Petit-Boix et al., 2017, 2015; Spatari et al., 2011), or in CSO control strategies (De Sousa et al., 2012). Only two studies investigated the links between land use and stormwater in the life cycle of UWS in terms of changes in the availability of water for aquatic ecosystems (Berger and Finkbeiner, 2010; Saad et al., 2013). Only one study addressed the links between urban land use and pollution from untreated stormwater by defining a stormwater pollution burden for a given system which results from urban land use called in background processes (Phillips, 2015; Phillips et al., 2018).
Thus, several challenges remain in the evaluation of impacts arising from IDs, which are not taken into account in any of the reviewed papers and for which there are no inventory flows. In line with the findings of Byrne et al. (2017b), the main identified challenges/opportunities in the inventory step are the consideration of the spatial and temporal variability of IDs due to the climatic conditions, urban catchment characteristics (e.g. land use and human activities), and the type of UWS (infrastructure and management).
LCA goal and scope definition
The main goal of this study is to compare relative contributions of pollutants discharged in dry-weather flows and wet-weather flows. Using the LCA framework, this study compares inventories of discharged pollutants as well as their associated impacts to receiving waters. The scope of the study covers discharges generated by an urban catchment with combined and separate sewer systems at two temporal scales, (i) during one year, and (ii) during wet-weather events. The potential impacts will be assessed for three categories relevant to waterborne pollutants, e.g. eutrophication (marine and freshwater) and freshwater ecotoxicity. Recent research carried out in the framework of the OPUR programme has provided the first comprehensive, broad overview of priority pollutant contamination of stormwater and CSOs. Since CSOs are an untreated mixture of raw wastewater and stormwater, the wastewater quality was also investigated.
The conceptual diagram in Figure 8 presents the system boundaries and inventoried flows towards the ecosphere (receiving waters) generated in the urban technosphere (man-made processes and systems) with both types of sewer networks. All flows were assessed in terms of volume and pollutant concentrations, but only flows from the technosphere are converted into potential impacts (LCIA). In Paris, the high-density inner area is drained by a mostly combined sewer system, collecting wastewater and stormwater for treatment in a wastewater treatment plant, sometimes leading to CSO discharges during storm events with a total annual volume of 13.7 million m3 in 2013 (Tabuchi and Penouel, 2014). The surrounding urban developments are serviced by a separate sewer system at the end of which polluted stormwater is discharged into local watercourses (See Figure S1 in Annex A-1). The UWS includes several large storage capacities for combined flows (of wastewater and stormwater) during storm events, as well as an automated management system for these flows which enabled a significant reduction in CSO discharges in this urban catchment. Thus, other cities may have a higher proportion of CSO discharges relative to the total volume in their sewage system.
Life Cycle Inventory analysis: available data regarding flows reaching receiving waters
This section describes the available data to conduct the Life Cycle Inventory (LCI) analysis on the Paris case study.
Classification of wet-weather events
Since the goal and scope of the study is to determine, at the year and event scales, the significance of wet-weather pollutant discharges to receiving waters, three categories of wet-weather events are proposed. The first event category is representative of a dry weather day (T1) where the only flows reaching the river are treated effluents. Wet-weather events (where a minimum of 1mm of rainfall has been recorded) were grouped into the two other categories:
(i) T2, wet-weather with untreated stormwater reaching the river from the separate sewer systems (mild storms); and (ii) T3, wet-weather with untreated stormwater and CSO discharges reaching the river (severe storms). Table 2 summarises the characteristics of the three types of events.
Monitored priority pollutants
The OPUR research programme provided new insights into the stormwater quality from various urban watersheds, including the Greater Paris area (Zgheib et al., 2012, 2011). In order to include specific urban pollutants, the WFD list of priority pollutants has been extended from 33 substances to a total of 88 (80 organic substances and eight metals) chosen for their representativeness in stormwater (Zgheib, 2009; Zgheib et al., 2008). Pharmaceutical compounds were not included in the WFD list, and as the OPUR programme focussed on pollutants relevant to stormwater, these compounds were not monitored. The same methodology was applied to monitor CSO discharges from the largest CSO outfall in the Paris metropolitan area as well as raw wastewater (Gasperi et al. 2012). Of the 88 priority pollutants monitored in wet-weather flows (stormwater and CSOs), 58 were consistently found and 30 were never detected (Figure 9). Total event mean concentrations and the detection frequency are given as well as details in the measurement procedure of Gasperi et al. (2012, 2014) and Zgheib et al. (2009, 2012).
Routine water quality parameters (common pollutants)
Quality parameters measured in the inventoried flows from all WWTPs managed by SIAAP include routine wastewater quality parameters, such as ammonium, phosphates and total oxidized nitrogen (nitrates and nitrites). Data for the characterization of flows from Greater Paris include daily loads for treated effluents from the wastewater treatment plants (summarised in 2014 SIAAP’s annual activity report), concentrations in stormwater from dense urban and sub-urban (separate sewer) catchments for 20 storms between 2008 and 2009 (Zgheib et al., 2012) and 52 measurements of CSO discharges from Greater Paris by Gasperi et al. (2012) in the CSO database. For stormwater, concentrations on classical parameters (ammonium, phosphates and total phosphorus) were measured, however for oxidized nitrogen forms (nitrates and nitrites), concentration mean and standard deviation in stormwater were taken from another study (Taylor et al., 2005). Data for CSO discharges was retrieved from the CSO database with 52 measurements, however there were no measured concentrations on nitrates and nitrites, which are negligible in these flows.
Table of contents :
CHAPTER 1.: GENERAL INTRODUCTION
1.1 METABOLISM OF URBAN WASTEWATER SYSTEMS
1.1.1 Can we truly liken a city to a living organism?
1.1.2 Another metabolic product of cities: urban stormwater runoff
1.2 INTRODUCTION TO LIFE CYCLE ASSESSMENT (LCA)
1.3 TOWARDS A COMPREHENSIVE ASSESSMENT OF UWS INCLUDING INTERMITTENT DISCHARGES
1.3.1 State of the art of LCA applied to urban water infrastructure
1.3.2 Stormwater consideration in LCA
1.4 OBJECTIVES OF THE THESIS
CHAPTER 2.: IMPACTS FROM URBAN WATER SYSTEMS ON RECEIVING WATERS – HOW TO ACCOUNT FOR SEVERE WET-WEATHER EVENTS IN LCA?
2.2 MATERIALS AND METHODS
2.2.1 LCA goal and scope definition
2.2.2 Life Cycle Inventory analysis: available data regarding flows reaching receiving waters
2.2.3 Life cycle impact assessment
2.3.1 Pollutant loads (LCI)
2.3.2 Impact assessment
2.4.1 Discharges from the UWS
2.4.2 Upstream catchment loadings
2.4.3 Life cycle impact assessment results
CHAPTER 3.: A PARSIMONIOUS SITE-DEPENDENT FRAMEWORK LINKING HUMAN ACTIVITIES, LAND USE AND CLIMATE TO DETERMINE LOADS IN INTERMITTENT DISCHARGES TO RECEIVING WATERS FROM URBAN CATCHMENTS FOR LCI PURPOSES
3.2 OBJECTIVES AND SPECIFICATIONS FOR LCI MODELLING OF UWS
3.2.1 Goal and scope of targeted LCAs
3.2.2 Resulting LCI requirements
3.3 OVERVIEW OF THE PROPOSED FRAMEWORK
3.3.1 Key stormwater pollutants
3.3.2 Pollutant loads inventory sub-model
3.4 MODELLING APPROACH AND PRACTICAL IMPLEMENTATION
3.4.1 Emissions inventory sub-model
3.4.2 Water flows inventory sub-model
3.5 DISCUSSION AND CONCLUSIONS
3.5.1 Model limitations
3.5.2 Data availability and uncertainty
CHAPTER 4.: LIFE-CYCLE PERSPECTIVES FOR THE MANAGEMENT OF INTERMITTENT DISCHARGES IN URBAN WASTEWATER SYSTEMS: LESSONS LEARNT FROM A FRENCH SUBURBAN CATCHMENT
4.2 MATERIAL AND METHODS
4.2.1 Case study description
4.2.2 Goal and scope definition
4.2.3 LCI models
4.2.4 Impact assessment methods
4.3.1 Main contributors in a retention scenario (C2)
4.3.2 Scenario comparison
4.4.1 Reliability of LCI models
4.4.2 Consistency of LCIA methods
4.4.1 Expanding the spatiotemporal scope of LCAs of UWS
4.4.2 STOIC framework for decision support
CHAPTER 5.: GENERAL CONCLUSION