Department of Civil and Environmental Engineering Division of Water Environment Technology CHALMERS UNIVERSITY OF TECHNOLOGY Gothenburg, Sweden 2017 Challenges in planning for sustainable stormwater management in French cities A case study of the Grand Lyon MARIE GRAND-CLEMENT II III MASTER’S THESIS BOMX02-16-157 Challenges in planning for sustainable stormwater management in French cities A case study of the Grand Lyon MARIE GRAND-CLEMENT Department of Civil and Environmental Engineering Division of Water Environment Technology CHALMERS UNIVERSITY OF TECHNOLOGY Gothenburg, Sweden 2017 IV Challenges in planning for sustainable stormwater management in French Cities A case study of the Grand Lyon MARIE GRAND-CLEMENT © MARIE GRAND-CLEMENT, 2017 Master’s Thesis no BOMX02-16-157 Department of Civil and Environmental Engineering Division of Water Environment Technology Chalmers University of Technology SE-412 96 Gothenburg Sweden Telephone +46 (0)31-772 1000 Cover: View from the Porte des Alpes technological park | May 2016 © Thierry Fournier, Métropole de Lyon Gothenburg, Sweden 2017 V Challenges in planning for sustainable stormwater management in French cities A case study of the Grand Lyon MARIE GRAND-CLEMENT Department of Civil and Environmental Engineering Division of Water Environment Technology Chalmers University of Technology ABSTRACT Because of climate change and expanding urbanization, issues concerning the management of stormwater in cities are arising. The increased imperviousness of urban areas cause problems of pollution and environmental protection, and traditional methods of combined or separated sewers are no longer sufficient to handle stormwater flows. Scientists are therefore promoting the sustainable management of stormwater through the use of source control techniques that should be integrated early on in urban planning, but these new solutions are still difficult to implement because of their multi-functionalities. This thesis focuses on the practices and challenges faced by French stormwater stakeholders, with the aim to find solutions to address their specific needs. A literature study was therefore conducted to explain the general context of urban stormwater challenges in France. It was followed by a case study on the metropolitan area of the Grand Lyon, where interviews were carried out with targeted stakeholders in the field of stormwater management. The research shows that while legislative instruments are present, challenges still exist in the collaboration between the different services and companies involved, where responsibilities are not clearly distributed. There also seems to be a lack of knowledge at several levels in the planning process and especially for maintenance operations, which is emphasized by a lack of follow-up on existing solutions that are forgotten or not correctly used. To face these challenges, several solutions are proposed, including decision-support tools, experience banks, online or on-site information portals, maintenance sheets, and training courses. Keywords: stormwater, stormwater management, sustainable stormwater management, techniques alternatives, urban planning VI ACKNOWLEDGEMENTS This Master’s Thesis has been carried out under the supervision of researchers at the Department of Civil and Environmental Engineering, Division of Water Environment Technology between October 2016 and February 2017. I would first like to thank my supervisor Sebastien Rauch for the freedom and support he gave me during the course of the thesis, as well as my examiner Leonardo Rosado. Special thanks to Cecilia Burzio who helped me get started, gave me updates on the Climate-KIC project, and proofread my report, and for her availability for friendly chats whenever I stopped by her office. Thanks to Niloufar Tabesh for her remarks and feedback on my work I would also like to particularly thank the different interviewees who took time out of their days to answer my questions and who were all available when I had further requests. Finally, I would like to thank my French girls and my landlords for supporting me during this time with love and amazing food! VII TABLE OF CONTENTS ABSTRACT ..................................................................................................................... V ACKNOWLEDGEMENTS ............................................................................................... VI TABLE OF CONTENTS ................................................................................................ VII LIST OF FIGURES ......................................................................................................... IX LIST OF ABBREVIATIONS ............................................................................................. X 1 Introduction ................................................................................................................ 1 1.1 Background .................................................................................................................... 1 1.2 Aims and objectives ....................................................................................................... 2 1.3 Limitations ...................................................................................................................... 3 1.4 Outline of the thesis ....................................................................................................... 3 2 Methodology .............................................................................................................. 4 2.1 Theory on qualitative research interviewing ................................................................... 4 2.1.1 Definition of qualitative interviews ........................................................................... 4 2.1.2 Conducting the interview ........................................................................................ 4 2.1.3 Recording and analyzing data ................................................................................ 5 2.1.4 Biases and issues with the method ........................................................................ 6 2.2 The interviews in practice .............................................................................................. 6 3 Theory ........................................................................................................................ 8 3.1 Urban stormwater challenges ........................................................................................ 8 3.1.1 Urbanization and stormwater flows ......................................................................... 8 3.1.2 Polluted runoff ....................................................................................................... 11 3.1.3 Summary and impacts .......................................................................................... 13 3.2 Sustainable stormwater management and techniques alternatives ............................. 14 3.2.1 Principles of sustainable stormwater management .............................................. 14 3.2.2 Examples of techniques alternatives .................................................................... 16 3.3 History of French stormwater management ................................................................. 19 3.3.1 Technical innovation and the hygienist movement ............................................... 19 3.3.2 Urbanization and rise of environmental awareness .............................................. 21 3.4 Legislation framework and responsibilities .................................................................. 22 3.4.1 European and national legislation ......................................................................... 22 3.4.2 Water management tools ...................................................................................... 22 3.4.3 Urban planning tools ............................................................................................. 23 3.4.4 Conclusion on stormwater management and legislative tools .............................. 24 3.5 Decision and communication-support tools ................................................................. 25 3.5.1 Interactive decision-support tools ......................................................................... 25 3.5.2 Guides and fact sheets ......................................................................................... 27 3.5.3 The issue of maintenance ..................................................................................... 28 4 Case study ............................................................................................................... 30 4.1 The Grand Lyon metropolitan area .............................................................................. 30 4.1.1 Geographical context ............................................................................................ 30 4.1.2 Administrative status ............................................................................................. 31 VIII 4.2 Stormwater in the Grand Lyon ..................................................................................... 32 4.2.1 Organization and responsibilities .......................................................................... 32 4.2.2 Current practices and issues ................................................................................ 32 4.2.3 Related projects and examples ............................................................................ 35 4.3 Interviews ..................................................................................................................... 38 4.3.1 Interview A - Grand Lyon water department ......................................................... 38 4.3.2 Interview B - Researcher at the DEEP Laboratory ............................................... 40 4.3.3 Interview C - Engineer at a consulting firm in urban planning .............................. 42 5 Discussion ............................................................................................................... 43 5.1 Challenges faced by urban stormwater stakeholders .................................................. 43 5.2 Available solutions ....................................................................................................... 45 5.3 The Climate-KIC STORMAN pathfinder ...................................................................... 48 5.4 Limitations and future work .......................................................................................... 50 6 Conclusion ............................................................................................................... 51 7 References .............................................................................................................. 52 APPENDIX I – Notes for Interview A ................................................................................. i APPENDIX II – Transcription of Interview B .................................................................... iv APPENDIX III – Written answers of Interview C ............................................................. ix IX LIST OF FIGURES Figure 3.1: Transformation of the water cycle with urbanization. ................................................. 9 Figure 3.2: Influence of urbanization on kinetics and runoff flows for a watershed after a storm event. ........................................................................................................................ 10 Figure 3.3: Comparison of the average concentrations (C) per site for streets and parking lots with limit values (LV) for water streams (ratio C/LV). ................................................... 12 Figure 3.4: Comparison of the average concentrations (C) per site for different types of rooftops with limit values (LV) for water streams (ratio C/LV). ............................................ 12 Figure 3.5: Impacts of urbanization on the water cycle. ............................................................. 13 Figure 3.6: Unprotected storage area where the material can pollute stormwater and directly reach the sewers. ................................................................................................................ 15 Figure 3.7: Schematic diagram of a swale. ................................................................................ 16 Figure 3.8: Schematic diagram of an infiltration trench. ............................................................. 17 Figure 3.9: Schematic diagram of an infiltration well. ................................................................. 18 Figure 3.10: Different types of porous pavements: porous concrete; concrete paving with drainage openings; turf paving. ........................................................................................... 18 Figure 3.11: Wet basin in a park and dry basin designed as a sports court. .............................. 19 Figure 3.12: Different levels of legislation in French stormwater management. ......................... 25 Figure 3.13: Screen components of the Adaptation Support Tool. ............................................ 26 Figure 3.14: Spatial representation of urban runoff risk (a) and urban flooding risk (b) in the Greater Lyon. ...................................................................................................................... 27 Figure 3.15: Example of a summary sheet intended for maintenance technicians. ................... 29 Figure 4.1: Location of the Grand Lyon in France. ..................................................................... 30 Figure 4.2: Distribution of wastewater and stormwater flows in the Grand Lyon. ...................... 34 Figure 4.3: Treated water flows and untreated water flows rejected to the natural environment (model results). ............................................................................................... 35 Figure 4.4: Map of the stormwater management techniques at Porte des Alpes. ...................... 36 Figure 4.5: Views of the moat in the La Buire park. ................................................................... 37 Figure 5.1: Main groups of people involved with stormwater in the Grand Lyon and challenges they face ........................................................................................................... 45 Figure 5.2: Tools to help solve challenges in stormwater management on the three areas of concern ............................................................................................................................... 46 Figure 5.3: Distribution of expressed needs in the different planning stages from the Climate- KIC pathfinder. .................................................................................................................... 49 X LIST OF ABBREVIATIONS ADOPTA: Association pour le Développement Opérationnel et la Promotion des Techniques Alternatives, or “Association for the Operational Development and Promotion of Techniques Alternatives” ASTEE: Association Scientifique et Technique pour l’Eau et l’Environnement, or “Technical and Scientific Association for Water and the Environment” BMP: Best Management Practices CEPRI: Centre Européen de Prévention du Risque Inondation, or “Flooding Risk Prevention European Center” CITERES: Unité Mixte de Recherche Cités, TERritoires, Environnement et Sociétés, or “Mixed Research Entity on Cities, Territories, Environment, and Societies” COD: Chemical Oxygen Demand CSO: Combined Sewers Overflows DEEP Laboratory: Déchets, Eau, Environnement, Pollution, or “Waste, Water, Environment, Pollution” DTA: Directive Territoriale d’Aménagement, or “Territorial Planning Directive” EPCI: Etablissement Public de Coopération Intercommunale, or “Public Establishment for Intercommunal Cooperation” EVS: Unité Mixte de Recherche Environnement, Ville et Societe, or “Mixed Research Entity on Environment, City and Society” GIS: Geographical Information Software GRAIE: Groupe de Recherche Rhône-Alpes sur les Infrastructures et l’Eau, or “Rhône-Alpes Research Group on Infrastructures and Water” LID: Low Impact Development LOD: Lokalt Omhändertagande av Dagvatten, or “Local Disposal of Stormwater” MAPTAM Law: Loi de modernisation de l’action publique territoriale et d’affirmation des métropoles, or “Law on modernization of public territorial action and metropolis affirmation” NOTRe Law: Loi portant nouvelle organisation territoriale de la République, or “Law on the new territorial organization of the Republic” OTHU: Observatoire de Terrain en Hydrologie Urbaine, or “Urban Hydrology Observatory” PAH: Polycyclic Aromatic Hydrocarbons PLU: Plan Local d’Urbanisme, or “Local Urbanism Plan” PPRN: Plan de Prévention des Risques Naturels, or “Natural Risks Prevention Plan” PPRNI: Plan de Prévention des Risques Naturels d’Inondations, or “Natural Risks Flooding Prevention Plan” SAGE: Schéma d’Aménagement et de Gestion des Eaux, or “Planning and Water Management Scheme” SCoT: Schéma de Cohérence Territoriale, or “Territorial Coherence Scheme” SDAGE: Schéma Directeur d’Aménagement et de Gestion des Eaux, or “Planning and Water Management Master Scheme” SUDS: Sustainable Urban Drainage Systems Techniques Alternatives: Alternative techniques to traditional sewers TSS: Total Suspended Solids VRD: Voiries et Réseaux Divers, or “Streets and Other Networks” ZAC: Zone d’Aménagement Concerté, or “Joint Development Zone” CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 1 1 Introduction Questions of sustainable development and climate change have become more and more urgent since the United Nations Conference of 1992 in Rio de Janeiro, and scientists are predicting rises in temperatures from 2 to 6.C by 2100, depending on the efforts that are made (Riebeek, 2010). An important consequence of this is the change in the water cycles of the planet, with increasing precipitation and more frequent violent events in European countries like France or Sweden (Füssel, Jol and Hildén, 2012). In addition to that, urbanization trends all over the world modify runoff patterns and increase pollutant concentrations in stormwater (Barbosa, Fernandes and David, 2012), creating areas that are more vulnerable to flooding and health-related risks. Therefore, sustainable stormwater management is a growing field of research that municipalities and other affected stakeholders must consider closely to find the best solutions for their cities (Barbosa, Fernandes and David, 2012). Alternative techniques to conventional sewer systems are hence being developed, but there are still several obstacles to their expansion (Stahre and Geldof, 2003). 1.1 Background In 1992, the United Nations Conference on Environment and Development in Rio de Janeiro, Brazil created the first important, global resolutions for sustainable development with the Agenda 21 (United Nations, 1992). Since then, countries, organizations and people have all tried to work towards the common objective of “development that meets the needs of the present without compromising the ability of future generations to meet their own needs”, such as it is defined in the Brundtland Report of 1987 (United Nations, 1987). And in 2015, the United Nations met again to devise a set of 17 new goals for sustainable development by 2030 (United Nations, 2015). These objectives are vast and encompass the three “pillars” of sustainable development: economic, social, and of course, environmental. Several of these goals are connected to urban stormwater issues (United Nations, 2015): - #06: Clean water and sanitation - #11: Sustainable cities and communities - #13: Climate action - #14: Life below water Managing urban stormwater in a sustainable manner is therefore a very relevant concern today, and new solutions are gaining ground all over the world as alternatives to traditional combined sewers. Stormwater, or eaux pluviales in French, is defined as rainwater once it has touched the ground (or any other surface, like rooftops) and flowed on it. It also includes water coming from snow melt, hail, and ice falling or forming on the ground, along with infiltration water (RPDE, no date b). The French term assainissement, which can be translated to “sanitation”, usually means the treatment of wastewater but can also include stormwater in some CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 2 legislative texts or guides (RPDE, no date a); it will be used as such in this report. Traditionally, stormwater and wastewater in cities are handled by sewer networks, that can be either combined (oldest techniques, in dense city centers, where stormwater and wastewater are conveyed through the same sewers) or separated (new developments, stormwater and wastewater do not travel through the same pipes). However, with increased risks of flooding, environmental pollution, and rising costs, French authorities have started to promote the use of techniques alternatives (alternative solutions) to the conventional sewers in the 1990s (Chocat, 1993), with the support of newly created specialized associations (GRAIE in 1985 or ADOPTA in 1997) (GRAIE, 2015; ADOPTA, 2016) and public water agencies (Agence de l’Eau Rhône Méditerranée Corse, no date). Techniques alternatives have hence developed exponentially since then, but because of their new features and multi-functionalities, there are still several obstacles to their expansion. Lack of knowledge and problems in collaboration have led several research teams to work on decision-aid and planning support tools to help stakeholders design and manage these new infrastructures (Baptista et al., 2007; Billger, Thuvander and Wästberg, 2016; van de Ven et al., 2016). The Grand Lyon, an urban area in the south-east of France, is seen along with the cities of Bordeaux or Douai as one of the pioneers in the development of techniques alternatives (Chocat, Sibeud, et al., 2014); this is partly due to the presence of important research laboratories and water associations (Martin, Ruperd and Legret, 2007). In addition, the metropolis has recently developed a new general sanitation plan which includes several prescriptions on stormwater, and has started a project called Ville Permeable (“Pervious city”) aimed at overcoming obstacles to the expansion of sustainable stormwater management (Direction de l’Eau du Grand Lyon, 2015b). 1.2 Aims and objectives With that in mind, the aim of this thesis is therefore to get a better understanding of the current practices and challenges faced by French urban planners in terms of stormwater management, and to find solutions that could address their specific needs. This goal will be achieved through interviews of relevant stakeholders and a case study of the metropolis of Lyon, where significant progress has already been made even though problems may still arise. The specific objectives linked to this thesis are therefore: - To describe the main challenges for sustainable stormwater management in French urban areas; - To describe the current practices for stormwater management in terms of: - Legislation; - Responsibilities (at the national, regional and municipal level); - Decision-making and planning processes; - Implementation and maintenance; - To assess the tools developed for stormwater management, including decision- support tools; - And to assess the needs and demands for improving stormwater management, based on the case study. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 3 The obtained results will also be briefly compared to the findings of the Climate-KIC Climate- Smart Stormwater Management pathfinder carried out at the Water Environment Technology division of Chalmers University of Technology. This project aims at understand the planning processes and challenges faced by stormwater stakeholders in the Gothenburg region. 1.3 Limitations There are some obvious biases linked to using interviews as method in a research project - these will be developed in more detail in Chapter 2. However, some other limitations include the scope of the thesis, which focuses only on the case study of the Grand Lyon metropolitan area. Some references to other French cities are made (for example by some of the interviewees) but they are not considered in depth; the conclusions will therefore be based on one example only. Due to a lack of time, only three people from different organizations were interviewed, reducing the number of points of views. Complications also arise because of the different definitions linked to sustainable stormwater management. What is considered sustainable in one place might not be in another, and practices or recommended solutions may vary from one place to another. Sustainable stormwater management solutions also have different names depending on the country and the context: Best Management Practices (BMPs) or Low Impact Development (LID) in the US, Sustainable Urban Drainage Systems (SUDS) in the UK, Lokalt Omhändertagande av Dagvatten (LOD) in Sweden, and finally techniques alternatives (“alternative techniques” to conventional sewer systems) in France (Barbosa, Fernandes and David, 2012). As France is the focus of this thesis, this term will be used throughout the report in its French translation, and will be defined more thoroughly in a later part. In addition, a lot of legislative and regulatory terms are also country-specific, so in order to diminish this obstacle, a glossary of acronyms was drafted. Another limitation comes from the amount of data in French used for the study. Translations from French to English had to be executed, but some language subtleties or nuances might be unintentionally left out or misinterpreted because of missing vocabulary from the researcher, or because no exact translation exists. This is especially true for the interviews, where informal language and spoken speech might be more difficult to translate. 1.4 Outline of the thesis This thesis is structured in five chapters, with Chapter 1 being the previous introduction. Chapter 2 details the methodology used for the interviews. Chapter 3 is a theoretical review, aimed at giving the reader the background information needed in terms of urban stormwater challenges, French current practices and legislation, and tools to support the development of techniques alternatives. The case study is then presented in Chapter 4, with an introduction on the city of Lyon and its metropolitan area, and the outcome of the interviews. Finally, Chapter 5 presents a discussion on the results and proposes solutions tailored at the needs of the different stormwater management decision-makers in Grand Lyon and in France, as well as a comparison with the challenges faced by stormwater stakeholders in Gothenburg. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 4 2 Methodology In order to meet the specified objectives, a literature study was first conducted on sustainable stormwater management and French practices. Information was collected about the challenges posed by urban stormwater and about available technical solutions such as techniques alternatives. The history of French stormwater management was also reviewed along with current French legislation and distribution of the responsibilities that bind the different stakeholders, with a focus on the Grand Lyon area. Finally, a study was done to get a broader view of the support tools available for decision-makers. The second part of the thesis consisted in carrying out qualitative interviews with different people involved in stormwater management in Lyon. There are many different kinds of interviews, but research interviews aim at obtaining information from knowledgeable people about issues linked to the research project (Gillham, 2000); this definition will hence be applied throughout this thesis. 2.1 Theory on qualitative research interviewing 2.1.1 Definition of qualitative interviews As Taylor, Bogdan and DeVault (2015) state in Introduction to qualitative research methods: a guidebook and resource, there is no perfect research method but one should be chosen taking into consideration research interests, context, and time constraints; for instance, interviews may be useful when research interests are clearly defined and to reach knowledge on events or problems not otherwise accessible. Qualitative, in-depth interviews are mostly used in research instead of closed-questions surveys when a deep understanding is necessary. They can vary in structure, but usually follow three main criteria (Gillham, 2005): - The questions are open, allowing the interviewee to choose their own answers; - There is interaction between interviewee and interviewer, who can ask for clarification; - The interviewer has a purpose which gives the interview its structure. Taylor, Bogdan and DeVault (2015) also oppose qualitative interviews to structured interviews: the former are flexible and dynamic while the latter are more standardized, with the same questions being asked to all subjects. Qualitative, in-depth interviews must therefore face several constraints, because of their informal nature. People might not be available for a long time, they might have some apprehension to talk about certain subjects, there might be some political or ethical restrictions on what can be addressed… (Gillham, 2005) And finally face-to-face interviews might be impossible because of geographical constraints, as was the case for this study. 2.1.2 Conducting the interview A main question is therefore how to select informants. Kvale (1996) points out that one should “interview as many subjects as necessary to find out what you need to know”. Hence, CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 5 there is no fixed minimum number of informants to carry out a good research; in addition, there seems to be an inverse correlation between the number of subjects and the depth of the interviews: the more people are interviewed, the less detailed the interviews are (Taylor, Bogdan and DeVault, 2015). The approaches used to select informants are varied, but most will try to aim at the maximum diversity in the subjects pool. Time and willingness to participate should also be considered. During the interview, the first step is to quickly establish a positive relationship between the research and the interviewee, in order to create a comfortable environment in which the interviewee can talk freely about personal experiences and opinions (Dicicco-Bloom and Crabtree, 2006). The aim is to try to approach as much as possible a natural and everyday exchange (Taylor, Bogdan and DeVault, 2015). Therefore, the first questions should be general and open, aimed at starting a conversation. The researcher can then follow-up with unplanned questions, trying to keep them as non-directive as possible so the interviewee can choose their own answers. For example, instead of saying “Didn’t this make you feel strange?” the researcher should ask “How did this make you feel?” (Dicicco-Bloom and Crabtree, 2006). Taylor, Bogdan and DeVault (2015) also encourage the use of an interview guide, consisting of a list of themes to address with all informants. Questions can then be asked spontaneously during each interview, but this gives a structure to keep the interviewer in the right track for his or her research. Another way to obtain information from an interviewee, when he or she has little time available, is the e-mail interview. This medium, like a face-to-face interview, can give quality personal data because of its informal nature, is much faster to implement, and allows the interviewee to answer at his convenience (Gillham, 2005). It also provides the researcher with a “ready-made” transcription of the respondent’s answers. In addition, the necessity to build a positive relationship is not as strong as for a face-to-face exchange, even if some courtesy rules still have to be followed (Gillham, 2005). 2.1.3 Recording and analyzing data Answers to qualitative interviews can be recorded in several ways, the most frequent of which are audiotape recording and note-taking (Dicicco-Bloom and Crabtree, 2006) (for email interviews, the written answers of the informant directly act as a transcription of the interview; this will not be dealt with in this paragraph). Technical issues with recording can compromise the results; the researcher should therefore make sure the equipment is correctly working beforehand in order to avoid background noise or inaudible speech. Informants should also be made aware of the recording and need to give explicit consent for it before the interview (Taylor, Bogdan and DeVault, 2015). Note-taking is less time- consuming and allows for direct analysis, with the interviewer already categorizing responses during the interview (Muswazi and Nhamo, 2013). Finally, once the interview is over the data has to be analyzed. For this, the researcher must go over the recording or the notes taken during the interview in order to highlight the most relevant key points and sort them into categories (Gillham, 2000); one way to do this might be to transcribe the entire interview, but this is very tedious (a one-hour interview needs about five hours to be properly transcribed (Introduction to Research: Handling Qualitative CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 6 Research Data, no date)). Once the pertinent data has been extracted, it can be coupled with literature information in order to generate an understanding of the intended research questions. 2.1.4 Biases and issues with the method However, as with any research methods, there are a lot of concerns with qualitative interviewing. They can be technical, ethical or linked to the participants’ reliability. The anonymity of the respondents must be maintained when there is a possibility that what they are sharing can jeopardize his or her position in a system (Dicicco-Bloom and Crabtree, 2006), otherwise the interviewee might not willing to share certain information. In addition, the fact that the interview is carried out almost like a conversation means that it can be made of the same distortions which happen in everyday life (Taylor, Bogdan and DeVault, 2015), and the researcher’s analysis inevitably involves some kind of subjective interpretation (Gillham, 2005). Issues can also come from the medium with which the interview was recorded; whether it is, as mentioned before, due to technical problems with the recorder, or because some information was missed out during note-taking. As such, note-taking by itself is often not recommended because of its inability to capture a certain level of detail (Muswazi and Nhamo, 2013) and because of the increased risk of subjectivity from the researcher (Introduction to Research: Handling Qualitative Research Data, no date). However, with any kind of recording, participants may feel reluctant to express their true opinions (Muswazi and Nhamo, 2013), and an important level of trust has to be built between the researcher and the interviewee in order to minimize this. Finally, in the specific case of email interviews, answers might either be too formal or too colloquial. Respondents might also have a tendency to express themselves in a “note” or “list” format, very different from human speech, and not develop fully what them mean (Gillham, 2005). It is therefore important to ask for clarification with follow-up emails if the answers are not deemed detailed enough. 2.2 The interviews in practice For this thesis, in-depth interviews were carried out with three different people involved in stormwater management in Lyon. Here the interviews were done as a way to complement the literature study and to get personal insights on the problems stormwater stakeholders are currently facing. A member of the Grand Lyon water department was therefore interviewed (interview A), followed by a researcher at the DEEP laboratory (interview B) and a member of an engineering firm working on urban development projects (interview C). Because of geographical constraints, none of the interviews were effectively “face-to-face”; however, videoconferences were used for two of them in order to approach this technique as much as possible. They will therefore be referred to as “face-to-face” interviews in the following part. As the thesis went by, the method was tested and the format of the interviews varied in order to try to achieve the most scientific results. Indeed, as was explained before, there are several possible biases linked to this research method. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 7 For example, the first interview with the Grand Lyon employee was carried out face to face through Skype: the advantages of this is that is more personal and allows for more exchange between the research and the interviewee. It gives the possibility to easily deepen the subject in order to follow up some relevant answers, or to skip questions that might not be of interest in the end. However, the conversation was in this case not recorded, and the transcription only comes from note-taking throughout the interview (Appendix I). One problem of this is that there is a big influence from the researcher in the results, as: - Not everything is always written down because of a lack of time, - Some things might not be comprehensible when re-read and taken out of context, - The interviewer might (intentionally or not) miss out some aspects of the conversation when taking notes and only write down what matters to him or her. Therefore, in order to minimize the bias coming from the interviewer, the other two interviews were recorded fully (through tape-recording or written answers) and can be found in Appendix II and III. The exchange with the researcher was also carried out through video- conference, and allowed for a more two-sided conversation. However, the email format was used for the interview with the engineer, which then resembled more a survey than an actual interview. This was due to a limited availability from his part which only allowed him to answer a few questions. This has the advantage of giving a true image of the interviewee’s ideas, unaltered by the researcher’s point of view, and of having more structured and clear responses to the actual questions, but does not allow for an extended discussion, and gives much shorter answers. Overall, such qualitative interviews are bound to be biased in some way as they are necessarily subjective. It is also possible that some people omit certain topics, either because they are afraid of the consequences, do not think it is relevant, or simply forget about it. The relationship between the researcher and the interviewee should also be considered; if they know each other, the exchange will not be the same as if they have just met. However, it is still important to have such points of view (even if there are few), especially when the objective of the research is to identify problems and offer solutions. Table 1 summarizes the setups for each interview. Table 1: Interview setups Interview Company/Organization Type of interview Length Transcription A Grand Lyon water department Videoconference 60 min Notes B DEEP Laboratory Videoconference 30 min Full transcription C Artelia engineering firm Email 6 questions Written answers To finish, the results from the interviews were analyzed to find the main challenges faced by urban stormwater stakeholders in Lyon. The main problems were identified and solutions were then proposed based on the literature study conducted previously. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 8 3 Theory In this part, the studied literature will be reviewed in order to understand the main concepts linked to stormwater and sustainable stormwater management. First the challenges created by increased precipitation and its impact on cities will be developed, followed by a presentation of available sustainable solutions, the so-called techniques alternatives. A French history of stormwater management will then be introduced, followed by a description of the current legislation and responsibilities of concerned stakeholders. Finally, some decision-support tools and management aids will be presented. 3.1 Urban stormwater challenges In terms of stormwater, the main problems that arise from urbanization are problems of quantity and of quality. Unpredictable consequences of climate change will cause flooding and contamination issues, especially in less developed countries more vulnerable to such risks (United Nations World Water Assessment Programme, 2015); it is therefore important to assess these risks and understand where they come from. 3.1.1 Urbanization and stormwater flows Over the last century, urban developments have changed completely the aspect and typology of our cities. While in 1904 93% of United States roads were unpaved, with the development of automobiles, areas with human presence have become almost entirely impervious, especially in cities (Arnold and Gibbons, 1996). Streets, rooftops, but also patios and even compacted soils alter the hydrologic cycle of the whole system, degrading the water resources: infiltration of rainwater through the soil is prevented, and surface runoff increases (EPA, 1999), while the groundwater table is not renewed fast enough (Arnold and Gibbons, 1996). Figure 3.1 shows the relationship between impervious land coverage and water cycle transformation. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 9 Figure 3.1: Transformation of the water cycle with urbanization. Adapted from (Arnold and Gibbons, 1996) One consequence of this is that peak flow rates happen sooner and are higher than in natural environments, with a greater runoff volume. The frequency of events surpassing a certain flow rate also increases, with a higher probability that the network’s capacity will be surpassed (CEPRI, 2014). In Figure 3.2, the evolution of the flow at the outlet of a watershed is shown against time. In light blue, before urbanization: the flow rate is smaller all along the event and the total volume is smaller. In dark blue, after urbanization: this time the peak flow rate is much more important since the runoff is not slowed by the ground typology, and the total volume to be treated is greater because of the lack of infiltration. In addition, there is an important volume that cannot be handled by the network, which has to overflow somewhere (most often into nearby streams). CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 10 Figure 3.2: Influence of urbanization on kinetics and runoff flows for a watershed after a storm event. Adapted from (CEPRI, 2014) Increased stormwater runoff also has a great impact on erosion from construction sites, downstream areas and stream banks (Arnold and Gibbons, 1996). The higher volumes of water, mixed with sediment, can widen stream channels and increase water temperature ranges, causing channel instabilities. Some visible examples of this are exposed stream banks, fallen vegetation, and sedimentation, but this also influences stream habitat as the environment is modified to accommodate more rainfall (EPA, 1999). Another consequence of increased imperviousness is the slow renewal of groundwater tables. Drinking water supply is threatened, as well as water provision from groundwater to local streams, which may diminish if the flow is too low (Arnold and Gibbons, 1996). In addition, the drying up of the soil in cities, due to low rainwater infiltration, may cause differential settlements which may lead to instabilities in the buildings above (Chocat, 2015). The urban heat island effect is also a consequence of urbanization closely linked to stormwater. It is a phenomenon which is observed when temperature in cities are significantly warmer than in surrounding suburbs and rural areas because of a lack of vegetation (to block solar radiation and cool the air with evapotranspiration) and an increase in the amount of materials with higher thermal properties. Because they release so much heat, cities can then alter local atmospheric flows, allowing air to rise and increasing significantly the amount of rainfall (Seyoum, 2012). To take care of this excess water, two techniques are traditionally used (EPA, 1999): CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 11 - Separated sewers, with one network for wastewater and one for stormwater. In this case stormwater is not mixed with wastewater but is usually released untreated in the environment. - Combined sewers, where stormwater and wastewater travel through the same pipes to a wastewater treatment plant. In the case of important storm events, the volume of water often exceeds the capacity of the plant; the excess (a mix of waste and stormwater) is hence released through combined sewers overflows (CSO) to the natural environment. 3.1.2 Polluted runoff In both cases, however, there is untreated stormwater runoff that eventually finds its way to receiving streams and the natural environment. This can be dangerous because stormwater runoff contains several kinds of pollutants which are harmful to aquatic life but also to humans (EPA, 1999). According to the CEPRI (2014), pollutants found in stormwater runoff include: - Solid waste like leaves, plastics, etc. which can obstruct network pipes; - Solid and fine minerals from erosion, construction sites, etc. which can plug pipes and absorb heavy metals; - Atmospheric pollution particles deposited on surfaces; - Heavy metals from industrial activities, traffic, or construction materials; - Hydrocarbons from traffic essentially; - Products linked to land upkeep like road salt or pesticides; - Organic materials, potentially pathogenic. While it can be difficult to assess the level of pollution of stormwater runoff, because of the number of parameters involved (type of surface, type of materials, usage…), the Seine- Normandie Water Agency (Gromaire et al., 2013) has carried out research to compare the concentration of well-known pollutants in runoff to threshold values for ecological quality of streams (Gromaire et al., 2013). Figure 3.3 shows the results for roads and parking lots, while Figure 3.4 focuses on runoff from rooftops. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 12 Figure 3.3: Comparison of the average concentrations (C) per site for streets and parking lots with limit values (LV) for water streams (ratio C/LV). Adapted from (Gromaire et al., 2013) Figure 3.4: Comparison of the average concentrations (C) per site for different types of rooftops with limit values (LV) for water streams (ratio C/LV). Adapted from (Gromaire et al., 2013) They show that for example, Cu and Zn concentration is often ten times higher than the accepted value for natural streams; this increases to one thousand times higher when the CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 13 rooftops are made of copper or zinc, respectively. While TSS and COD concentrations are often below the limit for rooftops runoff, they are more important for streets and parking lots; these figures show clearly that in both cases, pollutant concentration is much higher than what is accepted to ensure ecological quality of natural environments. In addition, during important storm events sewers might become saturated and not be able to treat all of the incoming flow. Since sewers in French cities are often combined (or at least, the separation between wastewater and stormwater is not always perfect), rainy weather overflows are usually a mix of wastewater and stormwater, which is rejected directly to the natural environment. In that case, pollutant concentrations can be two to ten times higher than the norms for wastewater plant discharges (Chocat and GRAIE, 2016). Finally, pollution in stormwater runoff can be present both in particle form and in dissolved form. While particle contamination can be more or less easily treated by slow techniques of decantation (the particles fall to the bottom from their own weight) or filtration (through filters or soils with or without vegetation), dissolved pollutants can pass through those filters and research is still ongoing to find the best solutions to remove these contaminants (especially pesticides) from stormwater runoff (Sibeud and Pourchet, 2013). 3.1.3 Summary and impacts Figure 3.5 therefore summarizes the challenges faced by stormwater management stakeholders and citizens due to urbanization: Figure 3.5: Impacts of urbanization on the water cycle. (Chocat et al., 2007) Unsustainable stormwater management can hence have varied but significant consequences: sudden and violent floods may have severe health impacts on populations as time to evacuate is shortened; exposition to products carried by water and wastewater, humidity, and sludge may also increase the spreading of diseases and contaminations. Economic impacts may also be important as infrastructure damages in dense cities cause CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 14 losses of money due to restoration needs and breaks in activities. Properties and patrimonial heritage also face risks of destruction. Finally, environmental impacts may be considerable as ecological quality and biological diversity are threatened by stormwater pollution (CEPRI, 2014). 3.2 Sustainable stormwater management and techniques alternatives 3.2.1 Principles of sustainable stormwater management In order to face the challenges posed by urban stormwater, new techniques have been implemented as soon as the 1980s to treat pollution and to regulate flow rates. With time, they developed to include more functionalities and to work at smaller scales when source control was promoted (Chocat et al., 2008), and an integrated approach is now encouraged; one which combines hydraulic and technical criteria with ecological and aesthetic considerations (Lindh, 2013). New urban stormwater management techniques must be adapted to local constraints and be an integral part of the city’s urban plan (Lami et al., 2006). In addition to treating flooding and pollution issues, Chocat et al. (2008) highlight three main benefits of sustainable stormwater management: - Urban and landscape aesthetic with the reintroduction of nature in the city; - Promotion of stormwater as a useful resource; - And climate regulation interests; with most techniques combining several of these benefits. Cost optimization and environmental education are also aspects that have to be taken into consideration when implementing sustainable stormwater management (Lami et al., 2006). Sustainable urban stormwater management solutions, or techniques alternatives as they are called in France, can be sorted into two main categories depending on the level at which they handle the stormwater. Source control aims at minimizing the generation and the pollution of stormwater, reducing the needs for costly treatment methods. The objective of treatment control methods, on the other hand, is to clean polluted stormwater and divert it away from traditional sewers (Lindh, 2013). Ideally, both should be associated in order to create a “treatment train” that will reduce pollutants in runoff, reduce runoff volume, and treat the remaining runoff pollutants (Urban Drainage and Flood Control District, 2010). 3.2.1.1 Source control The first strategy of source control is to disrupt the natural water cycle as little as possible; this means that impervious land coverage must be minimized by using porous materials and permeable surfacing, or by keeping natural land (Luchesi, 2008b). Lindh (2013) acknowledges planning and design actions that can be undertaken to reach this objective: - Plan for compact urban development and limit growth to areas best suited to it; - Reduce connection of impervious ground to the stormwater drains; - Include natural areas which can directly capture stormwater (like ponds); - Promote the use of pervious surface materials. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 15 Another important aspect of source control is the decrease of stormwater contamination. Most of the significant pollutants in stormwater appear once it touches the ground (or any other surface); indeed, while rainfall is necessarily polluted, it is usually drinkable before it reaches the ground. However, as runoff flows along streets, it cleans and erodes surface materials. Contamination depends on factors like rain intensity, runoff volume, and type of materials encountered, but the main factor remains the distance: stormwater that is infiltrated right where it fell is likely to be much less polluted than stormwater which has traveled a long distance (Sibeud and Pourchet, 2013). Therefore, source control techniques involve all kinds of disciplines and stakeholders: governments, businesses, and local inhabitants should try to employ water quality friendly materials, build roof runoff control, and prevent soil erosion (California Stormwater Quality Association, 2003). Industrial activities like vehicle washing, waste disposal and outdoor storage should be protected (see Figure 3.6) to prevent their pollutants from reaching the stormwater and structural source controls like coverage or spill containment should be implemented. Procedural source controls can also be used by municipalities and organizations to educate raise awareness throughout the community (Urban Drainage and Flood Control District, 2010). Figure 3.6: Unprotected storage area where the material can pollute stormwater and directly reach the sewers. (Urban Drainage and Flood Control District, 2010) 3.2.1.2 Treatment control Sustainable stormwater treatment solutions use natural processes to efficiently remove pollutants from contaminated stormwater. They can also act as obstacles which slow runoff flow and lessen impacts on sewer networks. Main treatment control techniques include infiltration of runoff to the soil, retention and treatment for later release, slow conveyance of water through vegetated areas, and technological flow-through treatment solutions (California Stormwater Quality Association, 2003). Most contaminants in runoff are polluted particles, which are efficiently dealt with by decantation and filtration through a significant layer of soil; stormwater can then be released into nearby streams or reach groundwater tables without harm (Lami et al., 2006). In an infiltration system, runoff trickles over permeable surfaces and pollutants settle into the ground, where they can be naturally mitigated. These systems can be open or closed and CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 16 are often designed to capture the “first-flush” storm event, paired with detention solutions to reduce peak hydraulic flows. Retention and detention basins are used to capture runoff temporarily and release it at a slower rate; in addition, settlement and physical removal of pollutants ensure the quality of the released water. Aquatic plants and microorganisms in retention basins can also increase biological and biochemical pollutant removal. Biofilters can also be used to transport runoff slowly over vegetation, allowing for filtration of sediments and pollutants through biological activity (California Stormwater Quality Association, 2003). Finally, specific, multi-vocational sites like parking lots, playground and parks can be used as temporary storage areas in the case of violent storm events (Lami et al., 2006). 3.2.2 Examples of techniques alternatives 3.2.2.1 Swales and ditches Swales and ditches are long, shallow, and have gentle slopes on the sides. They receive water from pipes or directly from runoff from nearby surfaces, which is then infiltrated through the ground or evacuated by an outlet (Lami et al., 2006); this is shown in Figure 3.7, for both an infiltration-type swale and a retention-type swale. In addition, they are easy to integrate because of their linear shape, and bring greenery that can serve as wildlife habitat (Luchesi, 2008b). Their design and construction is quite simple, they are easy to maintain, and their implementation costs are low (SyAGE, no date). However if the groundwater table is less than a meter deep, infiltration is not possible as pollutants can still be present (Lami et al., 2006). Figure 3.7: Schematic diagram of a swale. Adapted from (Lami et al., 2006) Swales and ditches need the same type of regular maintenance as urban green spaces: mowing of the grass, picking up of the leaves and garbage… This is especially easy when the slopes are gentle. In addition, special operations need to be undertaken every three to five years to de-compact the soil and ensure infiltration can happen properly (Luchesi, 2008b). 3.2.2.2 Infiltration trenches Infiltration trenches are quite similar to swales except they are filled with porous materials like gravel or pebbles. Stormwater is brought through pipes or by direct runoff, and is then CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 17 temporarily stored in the trench. It can then be evacuated through an outlet to the receiving environment or infiltrated through the soil, while the pollutants are captured in the porous material (Luchesi, 2008b). Infiltration trenches can be covered by vegetation or left open, as shown in Figure 3.8. Figure 3.8: Schematic diagram of an infiltration trench. Adapted from (Lami et al., 2006) Advantages of this technique alternative include an easy and controlled installation, low investment costs, an efficient depollution, and an effortless integration even in dense urban areas without requiring too much space (they can be implemented alongside or even under parking lots, streets, biking lanes… (SyAGE, no date)). Finally, they are easy and safe to maintain since they are filled with material, but need to be cleaned regularly to prevent clogging and stagnation of water (Lami et al., 2006). 3.2.2.3 Infiltration wells Another technique alternative often recommended by local authorities is the infiltration well. They are isolated objects and can be deep or not; similarly to the infiltration trenches, their function is to retain water temporarily and let it infiltrate into the soil (or reach a drain) while catching pollutants (Figure 3.9). However, they do not have a very large capacity and can be quickly saturated during violent storm events; in that case they must be associated with other techniques which can take care of the excess water (Luchesi, 2008b). CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 18 Figure 3.9: Schematic diagram of an infiltration well. Adapted from (Lami et al., 2006) Much like the other techniques alternatives mentioned before, infiltration wells are easy to design, construct, and maintain, with low investment and maintenance costs. Their low demand for space means that they can be used in a variety of contexts, whether it is below a parking lot, a playground, or a private garden. Finally, they need to be accessible so they can be inspected every semester and cleaned from potential clogging (Lami et al., 2006). 3.2.2.4 Porous pavements Porous structures are pavements which allow the rainfall to be infiltrated right where it falls: this is a source control measure that promotes the limitation of impervious land coverage and the reduction of the stormwater runoff travelled distance. Several kinds of pervious materials and techniques exist (see Figure 3.10) and they are used as a replacement for traditional pavements; for that reason they can be used on any simple project like parking lots, pedestrian lanes, and entryways. They can also be easily coupled with other techniques alternatives like swales or trenches. However, costs should be expected to be 10 to 15% higher than for traditional pavements (Luchesi, 2008b). Figure 3.10: Different types of porous pavements: porous concrete; concrete paving with drainage openings; turf paving. (Luchesi, 2008b) Annual maintenance must be undertaken, either with vacuum sweepers or high-pressured water, in order to maintain the porosity of the material. For areas with vegetation, chemical CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 19 fertilizers and weed-killers must be proscribed to prevent contamination of the water (Luchesi, 2008b). 3.2.2.5 Ponds and basins All of the previously described techniques alternatives are small and easily integrated into dense urban areas. Ponds and basins, on the other hand, require more space and more investment, and should be conceived as multifunctional areas (Lami et al., 2006). Sport courts and playgrounds can be used as dry basins which fill up during storm events and retain water temporarily, while wet ponds can be used in decorative gardens with overflows for violent storms, as shown in Figure 3.11. They can work in different manners, whether it is by cleaning the stormwater through the settlement of pollutants (Luchesi, 2008b) or infiltrating it through the soil (Lami et al., 2006). Figure 3.11: Wet basin in a park and dry basin designed as a sports court. (Lami et al., 2006) With such type of infrastructures, several precautions must be taken to ensure the safety of local citizens and the correct maintenance of the infrastructure. Water level should not be allowed to rise over one meter, or in this case, direct access must be prevent to reduce drowning risks (Luchesi, 2008b), and specific maintenance is needed to prevent clogging or water stagnation. This also implies that cleaning operatives must have ecological and environmental knowledge to follow water, fauna and flora quality (Lami et al., 2006). 3.2.2.6 Storing and green roofs Finally, the roofs of buildings should be taken advantage of to reduce stormwater pollution and runoff. Water can be stored temporarily and possibly evaporated and infiltrated if the roof is vegetated, slowing down outflow rates and minimizing contact with harmful materials (Luchesi, 2008b). The main advantages of this technique are its easy adaptation to any type and size of building, while not taking any new space; its relative simplicity; and its thermal functions in the case of green roofs. However, it must be implemented by skilled workers to ensure watertightness and needs to be correctly maintained several times per year; users must also be informed on its functioning and cleaning procedures. 3.3 History of French stormwater management 3.3.1 Technical innovation and the hygienist movement In the Middle Ages, the need for sewers was already important; in Paris they were open urban streams transporting both waste and stormwater and creating several health hazards. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 20 The role of Paris as the capital of an empire meant that technical innovations were often developed in order to create a model city, and they were encouraged by its government. Therefore, starting from the 16th century, the streams were covered and transformed into sewers (Chocat, 1997). In the middle of the 17th century, urban water management is handled by engineers, who start to create massive sanitation systems to take care of stormwater. Indeed, before the expansion of water distribution networks, wastewater was not very voluminous and was mostly reused as fertilizer or for the fabrication of cannon powder (Chocat, 1997). However, the great plagues of 1832 and 1884 pave the way for the creation of the “hygienist movement” and the development of the tout à l’égout (“all in the sewers” in English), with a direct release to nearby streams and rivers. This model meant that the state was now taking care of all of the waste and stormwater in the city; some local governments (like in Lyon and Bordeaux for example) were reluctant to do this because of the cost of the construction of sewers and the gains that were obtained from selling wastewater for agricultural usage. For this reason, Lyon built its combined sewers network in 1961 (Cohen, 1998). This “all in sewers” strategy stopped almost entirely the development of autonomous methods for stormwater management, and rainwater cisterns were becoming obsolete as they gave water of poor quality, easily contaminated and lacking minerals. However, the first attempts that could be related to techniques alternatives were absorption wells, destined at infiltrating waste and stormwater to the ground. Engineers were motivated to develop this kind of technique especially for stormwater, but they were soon forbidden because of the fear of contamination and pollution of the groundwater table. Therefore, alternative solutions for stormwater management were quickly discarded by authorities who saw sewer networks as the only scientific and acceptable solution to managing stormwater. A connection was made by Ward in a speech in 1852, where he compared urban water to blood flowing through an organism; there had to be a continuous movement where water entered pure in the city and left it carrying its wastes (Chocat, 1993). Hiding dirty waters and improving life quality was also an advantage of underground sewers. However, with the first important rural exodus that followed the Second World War combined with the rapid development of private sanitation equipment, the amount of water coming out of cities increased drastically, with a huge impact on the quality of surrounding environments (Chocat, 1993). This did not make the authorities question their reasoning, and they only prescribed new technical rules for the design and calculations of the sewers, based on the Caquot formula, with the CG 1333 memorandum in 1949. This was aimed at creating standard, safe, and cheap networks in all French cities, and only allowed for traditional sewer solutions. Because of this legal stiffness, engineers were closed off from practices coming from other countries, as they needed to apply the rules of the CG 1333 to get public funding (Chouli, 2006). CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 21 3.3.2 Urbanization and rise of environmental awareness Until the beginning of the 1970s, this method was applied everywhere and facilitated the rapid urbanization due to the thirty-year post war boom, the “Glorious Thirty”. This, along with the development of the automobile and the development of private housing and commercial areas, led to a great increase in impervious surfaces (Lami et al., 2006). National action was taken little by little to support and improve urban life, with the planning of “new cities” and the desire to improve social housing and living conditions. However, urban pollution started to become a concern and green areas were introduced slowly into cities, while environmental concerns arose (Chouli, 2006). In 1966, the government realized the shortcomings of the CG 1333 memorandum and started a research program on substitute solutions to handle stormwater. Retention basins were also becoming more popular with the help of international engineers, in “new cities”, which were often situated in places where natural evacuation was difficult. This resulted in a new memorandum in 1977 which gave focus to separated sewers and retention basins, storm basins next to wastewater treatment plants to limit overflows, and new, more precise calculation methods (Chouli, 2006). Pollution concerns were therefore combined with economic and flooding considerations. This new memorandum gave more options to engineers developing stormwater management infrastructures, and regions where flooding risks were important developed these new techniques quickly (like Bordeaux or the Seine-Saint-Denis), but many still chose to keep traditional sewers, because of misconceptions, bad experiences and lack of necessary skills (Dupuy and Knaebel, 1982). Nonetheless, catastrophic storm events like the floods in Narbonne (1988) and Nimes (1989) or the pollution of the Seine in Paris (1990 and 1991) showed the limits of purely technical responses to stormwater management. This was especially true in long- established cities, where there was little available free space (Petrucci, 2012). Since then, an integrated approach, taking the name of “techniques alternatives” has been formalized and promoted by national and local authorities, with three main objectives: decrease flooding risks, limit pollution, and integrate stormwater management into urban planning decisions (Lami et al., 2006). Initially, though, the development of techniques alternatives was slow, and mostly encouraged by a few municipalities helped by research organizations, who published a lot of guidance material and handbooks during the 1990s. Local experiments for urban integration of sustainable stormwater management played an important role in the development of techniques alternatives, but private infrastructures usually gave mixed results, because they were badly conceived or maintained. It also happened that the owners were not aware of the existence of techniques alternatives on their land. Private constructors did not yet see the advantages to going to a more integrated approach to stormwater management (Petrucci, 2012). However, even though issues still exist, the development of techniques alternatives has increased exponentially in recent years. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 22 3.4 Legislation framework and responsibilities 3.4.1 European and national legislation The French national legislation on stormwater is broad and scattered into different codes and regulations. First of all, the Urban Wastewater European Directive, adopted in 1991 to protect the water environment from the effect of wastewater discharges, has allowed for extended funding to be distributed to renew and renovate sewers, disconnect stormwater, and create storm basins (Ministère de l’Environnement de l’Energie et de la Mer, no date). In terms of French legislation, the Civil Code defines the status of stormwater, and is applicable to both private individuals and local governments. It states that stormwater is the propriety of the occupier who receives it on his land, but that he cannot refuse stormwater that naturally flows from higher lands. According to the Public Health Code, municipalities can also define prescriptions for and even refuse connections to public networks (O2D Environnement, no date). In addition, the 1992 Law on Water compels municipalities to define zones where imperviousness must be limited and stormwater managed (Ministère de l’Environnement de l’Energie et de la Mer, no date). The Environmental Code, modified by the 2006 Law on Water also defines the project surfaces which have to be submitted to declaration or authorization: above 1 ha and below 20, a declaration suffices, but above 20 ha, an authorization from the state services is necessary. This surface includes the surface of the catchment basin of which the runoff is intercepted (Luchesi, 2008a). Stormwater management has hence been an optional skill handled by municipalities (or their Public Establishment for Intercommunal Cooperation, or EPCI in French) for a long time. On December 30th, 2006, the law on Water and Aquatic Environments gave municipalities or their EPCI the possibility to create a public administrative service for stormwater, and to collect a tax aimed at giving more resources for sustainable stormwater management, but also at encouraging source control (an abatement was planned in this case) (Garrigues, 2012). However, the tax was quickly removed in 2015 because it did not bring in enough funding; the public service for stormwater management remains but now needs to be included in the general budget (Carré, 2016). This was then followed by the MAPTAM law in 2014, which made stormwater management an obligatory skill for all municipalities (or their EPCI) (Vincent and Grelier, 2016). Finally, the 2015 NOTRe Law (about the new territorial organization of the Republic), states that “water and sanitation” is now a competence that necessarily falls to the EPCI (Ministère de l’Environnement de l’Energie et de la Mer, no date), even if some elected officials did not agree with it as they felt that precise and local knowledge was necessary for successful stormwater management (Vincent and Grelier, 2016). 3.4.2 Water management tools In addition to the legislation that defines the different responsibilities, several regulatory tools are available for public planners and engineers to define their policies for sustainable stormwater management. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 23 The first are the Planning and Water Management Master Schemes (SDAGE in French), established with the 1992 Law on Water. There are 7 in metropolitan France and 5 in the overseas territories which are elaborated at the scale of the general hydrographical basin for a period of six years; the third generation of SDAGE has just been approved for the period 2016-2021. The Water Agencies define their orientations and action plans for the upcoming years along with state services and the National Water Office, but also local governments and citizens (Gest’eau, no date c). The different SDAGE have legal effect: this means that all urban planning and water management documents at lesser scales must be compatible with their corresponding SDAGE (Bacot, Brelot and Valin, 2014). The next level in water management are the optional Planning and Water Management Schemes (SAGE), established at the watershed level since the 1992 Law on Water and reinforced by the 2006 Law on Water. A SAGE must be composed of a sustainable management plan, defining objectives and the means to attain them, supplemented by a set of rules to follow to reach these goals (Bacot, Brelot and Valin, 2014). A SAGE is elaborated collectively by local governments, users (farmers, industrials, landowners…), and the corresponding Water Agency, in order to reconcile regular usage of water with the protection of the environment while taking local characteristics into account. Even though it is not mandatory, it has a legal effect and must be followed by other planning documents and regulations. In 2016, more than 100 SAGE exist and cover 49.1% of the French territory (Gest’eau, no date b). The “milieu contracts” are alternatives that engage stakeholders financially on a voluntary action plan; together with the SAGE, they cover more than ¾ of the country (Gest’eau, no date a). In addition, Natural Risks Prevention Plans (PPRN) can also be drafted to prevent certain risks linked to stormwater (but not only), like flooding (PPRNI) or mudslides (Luchesi, 2008a). They can for example define areas where imperviousness should be reduced or, on the contrary, where infiltration is forbidden due to the nature of the soils, and are prescribed by prefects following inventory and characterization of the possible hazards (Bacot, Brelot and Valin, 2014). According to the article L2224-10 of the General Local Authorities Code (Article L2224-10, 2010), municipalities or their EPCI must also establish a pluvial zoning, delimiting areas where: - “measures should be taken to limit ground imperviousness and ensure control of the flow and runoff of stormwater”; - “it is necessary to plan installations for collection, eventual storage, and if needed, treatment of stormwater when the pollution they bring to the aquatic environment may greatly reduce efficiency of sanitation apparatus”. Finally, the article L2224-12 of the same Code states that municipalities or their EPCI must come up with a set of regulations concerning sanitation (wastewater and stormwater), defining the work handled by the public services and the obligations of users, owners, and operators (Article L2224-12, 2006). 3.4.3 Urban planning tools One of the main characteristics of urban stormwater management is that while it cannot be decoupled from wastewater management in terms of infrastructures, it must also be associated with urbanism and city planning: indeed, management of urban planning documents and programs is necessary to control ground occupation and water flows (Carré CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 24 et al., 2010). For that reason, such documents should always be considered when aiming for sustainable stormwater management. The first level of urban planning documents is the Territorial Planning Directive (DTA). They are created by the state and are not meant to cover the whole territory, but rather for areas with high planning, development, and environmental protection stakes. They can encompass several departments or even regions and today, six exist in France (Berthelot, 2013). A DTA sets national orientations in terms of development and improvement of territories, but also its objectives in terms of the location of important transport infrastructures, equipment, and environmental protection zones (Bacot, Brelot and Valin, 2014). The Territorial Coherence Scheme (SCoT) corresponds to the next level of planning regulations, instituted by the Solidarity and Urban Renewal law of 2000. Again, it is optional and is born from local initiative from municipalities or their EPCI (Bacot, Brelot and Valin, 2014). It is a long term territorial project which must respect the principles of sustainable development in order to find a balance between urban renewal, controlled urban development, rural development, and protection of natural areas, to diversify urban functions and social mix, and to respect the environment. In 2015, 448 were accepted or in project, covering almost 60% of the national territory (Ministère du Logement et de l’Habitat durable, 2016). Finally, the smallest level of planning documents is the Local Urbanism Plan (PLU), established at the municipal or intercommunal level. It must respect the SCoT and be compatible with the SDAGE and the SAGE; in addition, the pluvial zoning can sometimes be directly integrated in the PLU (Bacot, Brelot and Valin, 2014). It defines the planning and sustainable development project of the municipality and contains a set of regulations describing ground affectation, construction heights and exterior aspects, areas to showcase or protect, neighborhoods to develop, energy performances, social mix zones, etc. (Direction Générale des Collectivités Territoriales, no date). 3.4.4 Conclusion on stormwater management and legislative tools Figure 3.12 shows the different levels of French legislation that are related to stormwater management, from European directives to local urban planning documents. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 25 Figure 3.12: Different levels of legislation in French stormwater management. Adapted from (Département de Seine-et-Marne, 2014) 3.5 Decision and communication-support tools The complexity of planning for sustainable stormwater management is evident. The diversity of stakeholders both in public and private sectors, coupled with complicated legislative frameworks and lack of experience may bring about several concerns which may halt the progress of sustainable stormwater management (Chocat et al., 2008; Caradot et al., 2010; Ellis, Lundy and Revitt, 2011; Boudet, Principaud and Maytraud, 2016). To face these issues, a lot of research has been carried out in the development of tools or aids aimed at helping stakeholders choose the right stormwater management method, but also communicate and work together in order to ensure its proper functioning. 3.5.1 Interactive decision-support tools The first category of such aids consists of decision-support tools or methods, intended to facilitate knowledge acquisition and decision-making at the design phase. Many of them are web- or computer-based and rely on multicriteria analysis. One example of this is the tool developed under the DayWater research program between 2002 and 2005. It’s an adaptative tool that can be used at different scales and by people from different backgrounds, and is accessible on the Internet (Thevenot, 2006). Once logged in, the user can find information and fact sheets about different Best Management Practices, urban CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 26 dynamics, pollution, risk management, etc., but also a multicriteria analysis tool where he or she can fill in different weights for different technical, environmental, economic, and social criteria. The best solution, based on predefined site characteristics, can then be proposed. A “guided tour” is also available when the user has a specific planning or design problem (DayWater, no date). Researchers have worked a lot on decision-making tools based on multicriteria analysis because this approach allows the combination of a vast number of objectives, while exposing potential conflicts and reducing the possibility of subjectivity. Another example of such a tool is the AvDren software, presented by Baptista et al. (2007). This tool is based on the definition of a general performance indicator (resulting from the combination of several sub-criteria) and a cost indicator, which can then be calculated for different scenarios through simulation. Another digital decision-support tool is the one developed by a team of Dutch researchers as part of the Climate-KIC “Blue Green Dream” project (van de Ven et al., 2016). They have created two complementary tools that can be used during the initiative and design phases of a planning project to inform, explore, and test different techniques for sustainable stormwater management. The first is the Climate Adaptation App and is available online (Bosch Slabbers Landscape + Urban Design et al., no date); it can, after selecting a number of filters on the target and the site characteristics, give a number of appropriate solutions. They can then be implemented virtually on a local map in the Adaptation Support Tool (see Figure 3.13), where their effect and impact is calculated (storage capacity, groundwater recharge, runoff reduction…) (van de Ven et al., 2016). Figure 3.13: Screen components of the Adaptation Support Tool. (van de Ven et al., 2016) Finally, the expansion of GIS based tools to support stormwater management is also to note. Warwick, Charlesworth and Blackett (2013) have developed maps indicating the best locations for different types of sustainable urban drainage systems based on physical and anthropogenic factors, while Caradot et al. (2010) use GIS software to evaluate and model CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 27 overflow risks in the Grand Lyon and in the city of Mulhouse. Figure 3.14 shows how other researchers go even further by considering human, material and environmental vulnerabilities to map out global vulnerability to hydrologic hazards and link it to run-off or submersible zones (Renard and Chapon, 2010). In all cases, GIS software allows stakeholders to quickly understand priority areas. Figure 3.14: Spatial representation of urban runoff risk (a) and urban flooding risk (b) in the Greater Lyon. (Renard and Chapon, 2010) 3.5.2 Guides and fact sheets In addition to decision-support tools developed by researchers, a lot of associations, municipalities, and other public organizations have drafted guides and fact-sheets, intended to raise awareness and share general information to stakeholders who wish to gain knowledge about sustainable stormwater management. Some are mostly informative and aim at spreading knowledge about techniques alternatives and the stakes involved, but also about general concerns linked to water. This is the case of the Méli-Mélo project started by the GRAIE and Media Pro and destined for a large audience of French citizens (Chocat, Brelot, et al., 2014). The project’s goal is to break popular misconceptions about drinking water, wastewater, and stormwater in cities through the use of humoristic short videos, illustrations, and short texts based on scientific references, which are intended to be “pirated” and shared intensively. In 2016, the GRAIE has also published a set of “true/false” notes answering commonly asked questions and fears about techniques alternatives to clarify knowledge of the different stakeholders involved (Chocat and GRAIE, 2016). Other documents include feedback on urban developments that successfully implemented sustainable stormwater management (Direction de l’Eau du Grand Lyon, 2013), and fact sheets about the most common types of techniques alternatives (Lami et al., 2006; Luchesi, 2008b; Mairie de La Mothe-Achard, 2016). They usually describe the advantages of each technique, in what context it can be implemented, and guidelines for their design and calculations. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 28 Some organizations have however realized that the problems with sustainable stormwater management did not only come from a lack of knowledge about the techniques alternatives, but also from the handling of the project itself. They have therefore created guides with guidance for stormwater project management, encouraging early integration of stormwater issues, collaboration between stakeholders, and monitoring of operations (Gromaire et al., 2013; Van Es et al., 2015). The practical guide of the Communauté d’Agglomération Hénin- Carvin (Van Es et al., 2015) is one of the most extensive with 305 pages, ranging from definitions of techniques alternatives to feedback on successful projects and guide sheets for an integrated project approach. 3.5.3 The issue of maintenance While local authorities already face a lot of issues when designing sustainable stormwater techniques, a specific concern is related to the maintenance (once the infrastructure is in place) because of the diversity of actors involved. This is mentioned in some of the guides introduced before, which reinforce the need to consider the maintenance very early in the process, and suggest regular control to adapt practices if needed (Gromaire et al., 2013). On the other hand, Boudet, Principaud and Maytraud (2016) have studied management issues in Plaine Commune and have devised “maintenance notebooks” to introduce the actors and stakes of each site, to explain its functioning precisely, and to describe the different maintenance interventions needed and how they should be coordinated. A summary sheet intended for maintenance agents was also drafted so that they could easily understand their role and their tasks on each site (Figure 3.15). A similar system was implemented in the Hauts-de-Seine, with summary sheets on specific sites to build a database of existing infrastructure and of their conduct (Conseil départemental des Hauts- de-Seine, 2015). CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 29 Figure 3.15: Example of a summary sheet intended for maintenance technicians. Adapted from (Boudet, Principaud and Maytraud, 2016) CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 30 4 Case study 4.1 The Grand Lyon metropolitan area 4.1.1 Geographical context In order to understand specific problems that could arise when planning for sustainable stormwater management, the French urban area of Grand Lyon was chosen as a case study. Lyon is the third-largest city in France behind Paris and Marseille (Journal du Net, 2014); but its urban area is the second largest French urban area with 2 237 000 inhabitants in 2014 (Théobald, 2016). It is located in the Rhone department (69) which is part of the Auvergne-Rhône-Alpes region, in the south-eastern part of France. Temperatures in Lyon vary from 0 to 6ºC in January to 16 to 28ºC in July on average; the city has a warm and temperate climate. Annual precipitation is around 830 millimeters, with about 105 days with precipitation per year (Météo France, no date). In addition, precipitation is around the same for every month of the year, with a small decrease in the colder months of December to March (Grand Lyon, 2016). Figure 4.1: Location of the Grand Lyon in France. Adapted from (Sting and Wikialine, no date) The object of this study is not only the city of Lyon but the metropolis of Lyon, also called Grand Lyon. Its location in France is shown on Figure 4.1. This area, with a population of 1 281 971 inhabitants in 2014 and covering a surface area of 538 km2, corresponds the CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 31 merging of the city of Lyon and 58 of its suburbs in order to create a unique type of regional government (Grand Lyon, no date a). 4.1.2 Administrative status In France, regional governments (collectivités territoriales) exist at different levels: municipalities, departments, and regions (Direction de l’Information Légale et Administrative, 2016d). In addition, there are 36 783 municipalities in France, 32 000 of which have less than 2000 inhabitants (Centre de Gestion du Doubs, no date); therefore, in order to reduce costs, gather expertise, and boost local development and planning schemes (Direction de l’Information Légale et Administrative, 2016b), Public Establishments for Intercommunal Cooperation (EPCI in French) were created and since 2013, every municipality must belong to an EPCI (Direction de l’Information Légale et Administrative, 2016a). They are public structures, elected by direct universal suffrage, who take on mandatory and optional competences given by law or handed over by the member municipalities, such as economic development, social habitat, sanitation, water, cultural development, etc. (Direction de l’Information Légale et Administrative, 2016c). Depending on their size and their status, EPCI may have different names: community of municipalities, urban communities, etc., but also, for the largest ones, metropolises. There are currently 14 EPCI that have the name metropolis in France: Nice, Bordeaux, Brest, Grenoble, Lille, Montpellier, Nantes, Rennes, Rouen, Strasbourg, Toulouse, Nancy, and Grand Paris and Aix-Marseille-Provence (Baylet and Grelier, 2017). However, even though it bears the same name, the metropolis of Lyon has a very different status as it is a regional government (much like municipalities and departments) with specific status (Baylet and Grelier, 2017). This is a unique status in France, which allows the metropolis government to take on both the responsibilities coming from the Grand Lyon (which was previously an EPCI) and from the Rhone department on all of the municipalities which it covers (Direction de l’Information Légale et Administrative, 2017). As shown, confusion is hence possible since the Grand Lyon used to be an EPCI, just like all of the other French metropolises. However, in this study both terms will be used interchangeably as they now refer to the same entity; the term “Grand Lyon” was kept for coherence and communication matters. The responsibilities which fall to the Grand Lyon today include responsibilities from the previous urban community, from the department, but also periodically and on certain missions, from the region and from the state. It can also delegate the management of certain competences back to the municipalities (Direction de l’Information Légale et Administrative, 2017). Its five main domains of action include economic development; education, culture and leisure; solidarity (handicap, elderly, family support…); living environment; and daily services which include water, waste and transport management (Grand Lyon, no date c). Therefore, all services linked to sustainable stormwater management are handled by the Grand Lyon in theory: water and sanitation, environmental protection, public space cleaning, and urban planning (Grand Lyon, no date c). However, parks & nature services are still usually handled by the individual municipalities. CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 32 4.2 Stormwater in the Grand Lyon 4.2.1 Organization and responsibilities Because of its status, the metropolis of Lyon has control over the entire water cycle on all of its 59 municipalities. While drinking water production and distribution services are managed by an external commissioned company called Eau du Grand Lyon (“Water of Grand Lyon”), a branch of Veolia, the Grand Lyon services have kept control of all of the collection, transport and treatment of waste and stormwater in the metropolis (Direction de l’Eau du Grand Lyon, 2015b). These public services are executed by the Water Department of the Grand Lyon, divided into three branches: project development, maintenance and operations, and support functions. The Grand Lyon employs 608 agents, 90% of which do technical work (the rest is administrative staff) (Direction de l’Eau du Grand Lyon, 2015b). Since 2011, the Water Department has a triple ISO certification for its integrated management system in Quality, Environment and Security (Grand Lyon, no date b). 4.2.2 Current practices and issues Historically, the urban center of Lyon had sewers to transport stormwater outside of the city until the 1950s, when wastewater, which was previously stored in drained pits outside buildings, was connected, creating a combined sewers network. However with the rapid development of the Lyon conurbation between 1960 and 1990, other methods were adopted with separate sewer systems or independent stormwater collection techniques, but with some issues where stormwater sewers were sometimes connected to the wastewater network, creating a pseudo-combined network with a lot of deficiencies (Direction de l’Eau du Grand Lyon, 2015c). To face this, a first (in 1969) and a second (in 1992) general sanitation plans were drafted based on several principles (Direction de l’Eau du Grand Lyon, 2015c): - Improve natural environment quality; - Reach a better compatibility with urban development; - Control flooding and pollution by stormwater; - Optimize usage of existing infrastructure. Therefore, questions of sustainable stormwater management were already being considered more than thirty years ago in the Lyon metropolis. In 2015, a third general sanitation plan was created for orientation until 2027. Under the Code of regional authorities (Code régional des collectivités), any local government of more than 2000 inhabitants must write a plan for its sanitation (which includes wastewater and stormwater) to define general orientations and relevant issues. It sets up a framework for future investments, projects and maintenance operations to improve the community’s sanitation system at the medium and long-term scales, and should include a diagnostic, a definition of objectives, and a schedule for their implementation. It has been elaborated in collaboration with the Rhone-Mediterranée-Corse Water Agency, all of the Grand Lyon departments (Streets, Cleaning, Urban Planning, …), the mayors and elected officials of the 59 municipalities, and other public organizations (Direction de l’Eau du Grand Lyon, 2015c). CHALMERS, Civil and Environmental Engineering, Master’s Thesis BOMX02-16-157 33 The 2015-2027 general sanitation plan of the Grand Lyon puts emphasis on four main concerns and orientations: 1) Act at the source to preserve human health and aquatic environments 2) Design and pilot sanitation systems to reduce impacts on the environment 3) Manage and develop existing infrastructure 4) Be near and see far to assist territorial planning For each of these issues, strategic objectives will be developed in two parts: one for the objectives which involve the heart of the work of water management actors, and one for the objectives that should be built in coordination with all public and private stakeholders involved (Direction de l’Eau du Grand Lyon, 2015c). In order to reach these objectives all around its territory, the Grand Lyon will adapt to the landscape and geological features of each site to develop and extend its network. Indeed, the west of Lyon is an area with rocky terrain where water does not infiltrate well; separated sewer systems should therefore be favored, while in the East, parcel infiltration is a better solution because of permeability of the so