UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Water quality and lifecycle assessment of green roof systems in semi-arid climate Dabbaghian, Mohammadreza 2014

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2014_spring_dabbaghian_mohammadreza.pdf [ 3.16MB ]
Metadata
JSON: 24-1.0074339.json
JSON-LD: 24-1.0074339-ld.json
RDF/XML (Pretty): 24-1.0074339-rdf.xml
RDF/JSON: 24-1.0074339-rdf.json
Turtle: 24-1.0074339-turtle.txt
N-Triples: 24-1.0074339-rdf-ntriples.txt
Original Record: 24-1.0074339-source.json
Full Text
24-1.0074339-fulltext.txt
Citation
24-1.0074339.ris

Full Text

  WATER QUALITY AND LIFECYCLE ASSESSMENT OF GREEN ROOF SYSTEMS IN SEMI-ARID CLIMATE  by  Mohammadreza Dabbaghian  B.Sc., Mazandaran University, 2005 M.Sc., Amirkabir University of Technology, 2008  A THESIS SUBMITTED IN PARTIAL FULLFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF APPLIED SCIENCE  in  The College of Graduate Studies  (Civil Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA (Okanagan)   April 2014  © Mohammadreza Dabbaghian, 2014 ii  Abstract Non-point source pollution contributes significantly to stormwater contamination in urban areas. Low impact development (LID) techniques and technologies are developed as a response to these challenges. Green buildings incorporate environmentally responsible and resource-efficient technologies to reduce environmental impacts over their life cycle. Green roof systems are broadly recognized as LID practices that may improve urban environmental quality by reducing stormwater runoffs. Potential impact of green roofs on the quality of runoff may be a deterrent to wider application of green roof systems. Organic and inorganic fertilizers in growing media, for example, may contaminate runoff and generate non-point source pollution. Recently, various environmental assessment methods have been developed to assess the environmental performance of green building technologies. Methods developed to date, however, are insufficient for accurate quantitative estimation and evaluation of triple-bottom-line (TBL) sustainability performance objectives (i.e. economic, environmental, and social) in the context of green building technologies. This study has two main objectives. First, it aims to investigate the performance of green roofs in the context of runoff water quality in the semi-arid environment of Kelowna, British Columbia, Canada. An experimental investigation has been conducted to enhance green roof performance by addition of a supplemental filtration layer. Runoff and precipitation samples were analyzed for water quality parameters including pH, nitrate and ammonia. In the next step, a quantitative sustainability evaluation framework for green building technologies was developed. The proposed framework integrates fuzzy-analytical hierarchy process (FAHP) integrated with a ‘cradle-to-grave’ life cycle assessment to address interactions and influence of various TBL criteria. The experiment results showed that the generic green roofs runoff is acceptable for domestic reclaimed water used under Cnadaian guidelines for domestic reclaimed water. The analysis shows that green roofs are able to reduce non-point source nitrate and ammonia concentrations. The installation of extensive green roofs could decrease a large amount of non-point source nitrate and ammonia emissions in an urban area during their lifespan. The utility of the FAHP approach is demonstrated by comparing sustainability performance of two generic green roof systems with a conventional roof. The results show that an ‘extensive’ green roof system is a more desirable option in terms of long-term sustainability performance criteria. iii   Keywords: Green roofs, stormwater runoff quality, low impact development (LID) practices, non-point source pollution, Performance assessment, fuzzy-analytical hierarchy process (FAHP), life cycle assessment (LCA). iv  Preface Parts of  Chapter 2 and  Chapter 3 Chapter 4 have been submitted as follows: Dabbaghian, M., Hewage, K., Reza, B., Culver, K., Sadiq, R. (2013). “Sustainability performance assessment of green roof systems using fuzzy-analytical hierarchy process (FAHP).” International Journal of Sustainable Building Technology and Urban Development, (11/27/2013). Parts of  Chapter 2 and  Chapter 3 have been submitted as follows: Dabbaghian, M., Hewage, K., Reza, B., Culver, K., Sadiq, R. (2014). “Experimental assessment for extensive green roofs runoff quality in semi-arid environment.” The Journal of Environmental Monitoring and Assessment. (02/25/2014). v  Table of Contents Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iv Table of Contents .......................................................................................................................... v List of Tables .............................................................................................................................. viii List of Figures ................................................................................................................................ x Glossary ....................................................................................................................................... xii Acknowledgement ...................................................................................................................... xiii Dedication ................................................................................................................................... xiv Chapter 1 : Introduction .............................................................................................................. 1 1.1 A Brief History of Green Roof Systems ....................................................................................... 1 1.2 Environmental Impacts of Roofing Systems ................................................................................. 2 1.3 Research Motivation ..................................................................................................................... 2 1.4 Objectives ..................................................................................................................................... 3 1.5 Research Methodology Outline .................................................................................................... 4 1.6 Thesis Structure ............................................................................................................................ 5 Chapter 2 : Background ............................................................................................................... 7 2.1 Green Roof Systems...................................................................................................................... 7 2.1.1 Green Roofs Components ......................................................................................... 8 2.1.2 Environmental Benefits of Green Roofs ................................................................. 10 2.1.3 Green Roofs Concerns ............................................................................................ 13 2.2 Purposes of Green Roof Runoff Quality Assessment ................................................................. 14 2.3 Life Cycle Assessment (LCA) for Environmental Impact Analysis ........................................... 15 2.3.1 Goal and Scope Definition ...................................................................................... 16 2.3.2 LCI Analysis ........................................................................................................... 16 2.3.3 Life Cycle Impact Assessment (LCIA) ................................................................... 16 2.3.4 Life Cycle Interpretation ......................................................................................... 17 2.3.5 LCA for Green Roof Systems ................................................................................. 18 2.4 Multi-Criteria Decision Making (MCDM) ................................................................................. 19 vi  2.4.1 Analytical Hierarchy Process (AHP) ...................................................................... 20 2.4.2 Fuzzy-AHP Analysis .............................................................................................. 21 2.4.3 FAHP Calculations ................................................................................................. 21 2.5 Sustainability Assessment Framework ....................................................................................... 23 2.6 Summary ..................................................................................................................................... 24 Chapter 3 : Experimental Investigation of Green Roofs Runoff Water Quality .................. 26 3.1 Materials and Method ................................................................................................................. 26 3.1.1 Study Site and Experiment Pilot Design ................................................................. 26 3.1.2 Rainfall Effect ......................................................................................................... 28 3.1.3 Chemical Analysis .................................................................................................. 29 3.1.4 Design of Experiment (DOE) ................................................................................. 30 3.2 Results ......................................................................................................................................... 30 3.2.1 pH ............................................................................................................................ 31 3.2.2 Nitrate and Ammonia .............................................................................................. 32 3.2.3 Color and Turbidity ................................................................................................. 34 3.2.4 Oxidation Reduction Potential (ORP) .................................................................... 35 3.2.5 Conductivity ............................................................................................................ 37 3.2.6 Appraisal of Green Roofs Runoff Quality .............................................................. 37 3.2.7 Scenario Analysis for Nitrate and Ammonia Removal ........................................... 40 3.3 Discussion ................................................................................................................................... 45 Chapter 4 : Sustainability Assessment Framework for Green Roof Systems ....................... 47 4.1 Sustainability Assessment Framework ....................................................................................... 47 4.2 LCA Study .................................................................................................................................. 48 4.2.1 Identifying Goal, Scope .......................................................................................... 49 4.2.2 Functional Unit and System Boundary ................................................................... 49 4.2.3 Inventory Analysis .................................................................................................. 49 4.3 Results ......................................................................................................................................... 51 vii  4.3.1 Constructing a Hierarchy Structure ........................................................................ 51 4.3.2 Life Cycle Inventory Data ...................................................................................... 52 4.3.3 Life Cycle Impact Analysis .................................................................................... 52 4.3.4 Selection of Sustainability Indicators for the Hierarchy ......................................... 60 4.3.5 Weighting of Sustainability Indicators ................................................................... 62 4.4 Discussion ................................................................................................................................... 66 Chapter 5 : Conclusions and Future Works ............................................................................ 68 5.1 Summary and Conclusions .......................................................................................................... 68 5.2 Limitations .................................................................................................................................. 70 5.3 Research Contributions ............................................................................................................... 71 5.4 Future Research .......................................................................................................................... 71 Bibliography ................................................................................................................................ 73 Appendices ................................................................................................................................... 85 Appendix A: Impact Category Description............................................................................................. 85 Appendix B: Xero flor XF301 Vegetated mat green roof system specifications (Xeroflor America 2013) ....................................................................................................................................................... 87 Appendix C: Sampling and analysis of waters, wastewaters, soils and wastes ...................................... 90 Appendix D: The experiments’ results ................................................................................................... 92 Appendix E: The current experiment’s pictures ................................................................................... 103 Appendix F: ANOVA assumptions validation ..................................................................................... 104  viii  List of Tables Table  3-1: Characteristics of different roof assemblies in the present study ................................ 28 Table  3-2: Oxidation-Reduction potential (ORP) ranges for different activities ......................... 36 Table  3-3: Required fresh water, domestic reclaimed water and green roof runoff quality ......... 38 Table  3-4: Average water quality parameters for each paired roof .............................................. 39 Table  3-5: The probability of rain and average precipitation in each day based on historical records (2013) ............................................................................................................................... 42 Table  3-6: The average estimated volume of runoff produced by 17 ha roofs ............................. 42 Table  3-7: The amount of nitrate removal using XeroFlor extensive green roofs ........................ 43 Table  3-8: The amount of ammonia removal using XeroFlor extensive green roofs ................... 44 Table  4-1: Material types for individual elements of roofing system for unit of area .................. 50 Table  4-2: The pair-wise comparison of TBL criteria for roofing system ................................... 63 Table  4-3: The pair-wise comparison relevant to Social criterion ................................................ 63 Table  4-4: The pair-wise comparison relevant to Economic criterion ......................................... 63 Table  4-5: The pair-wise comparison relevant to Environmental criterion .................................. 63 Table  4-6: Fuzzy local weights of (w) with δ=1 ........................................................................... 65 Table  4-7: Evaluation of final global preference weights (Gk) with δ=1 ..................................... 66 Table  4-8: Ranking of roofing systems ......................................................................................... 66 Table C.1: USEPA sampling process ………………………………………...…………...………………………………… 90 Table D.1: GR1 sample results…………………………………………………...…………...………………………………… 91 Table D.2: GR2 sample results…………………………………………………...…………...………………………………… 91 Table D.3: GR+CF1 sample results………………………………….………...…………...………………………………… 92 Table D.4: GR+CF2 sample results………………………………….………...…………...………………………………… 92 ix  Table D.5: GR+T1 sample results………..…………………………….………...…………...………………………………… 93 Table D.6: GR+T2 sample results………..…………………………….………...…………...………………………………… 93 Table D.7: GR+WB1sample results….....…………………………….………...…………...………………………………… 94 Table D.8: GR+WB2sample results….....…………………………….………...…………...………………………………… 94 Table D.9: EPDM1sample results……......…………………………….………...…………...………………………………… 95 Table D.10: EPDM2 sample results….....…………………………….………...…………...………………………………… 95 Table D.11: GR+TSG1sample results....…………………………….………...…………...………………………………… 96 Table D.12: GR+TSG2sample results....…………………………….………...…………...………………………………… 96 Table D.13: GB1sample results…………...…………………………….………...…………...………………………………… 97 Table D.14: GB2sample results……….......…………………………….………...…………...………………………………… 97 Table D.15: GR+S1sample results….........…………………………….………...…………...………………………………… 98 Table D.16: GR+S2 sample results……......…….…………………….………...…………...………………………………… 98 Table D.17: Acc. Age GRa1 sample results…..……….……………………….………..………………………………… 99 Table D.18: Acc. Age GRa2 sample results…..……….……………………….………..………………………………… 99 Table D.19: Acc. Age GRb1 sample results…..……….……………………….………..……………………………… 100 Table D.20: Acc. Age GRb2 sample results…..……….……………………….………..……………………………… 100      x  List of Figures Figure  1-1: Research Methodology Outline ................................................................................... 4 Figure  1-2: Thesis Structure ............................................................................................................ 6 Figure  2-1: Typical cross section of a generic green roof .............................................................. 8 Figure  2-2: Linguistic definitions in FAHP .................................................................................. 22 Figure  3-1: Green roof pilot experimental setup at University of British Columbia (Okanagan campus) ......................................................................................................................................... 27 Figure  3-2: Runoff samples .......................................................................................................... 29 Figure  3-3: Hach sampling instruments ........................................................................................ 29 Figure  3-4 : Daily precipitation in Kelowna (Canada Climate 2013) ........................................... 31 Figure  3-5: Average pH for the pilot scale events ........................................................................ 32 Figure  3-6: Average Nitrate for the pilot scale events .................................................................. 33 Figure  3-7: Average Ammonia for the pilot scale events ............................................................. 34 Figure  3-8: Average Turbidity ...................................................................................................... 35 Figure  3-9: Average ORP for the pilot scale events ..................................................................... 36 Figure  3-10: Average Conductivity for the pilot scale events ...................................................... 37 Figure  3-11: Selected area of city of Kelowna (created by google map) ..................................... 41 Figure  3-12: Days with light rain, moderate rain and thunderstorm in Kelowna (2013) ............. 41 Figure  4-1: Sustainability Performance Assessment frame work ................................................. 48 Figure  4-2: Hierarchical tree for comparison of roofing systems ................................................. 51 Figure  4-3: Intensive green roof life stages contribution .............................................................. 52 Figure  4-4: Extensive green roof life stages contribution ............................................................. 53 Figure  4-5: Gravel Ballasted roof life stages contribution ............................................................ 54 Figure  4-6: Non-renewable energy consumption of three different roofing alternatives ............. 55 xi  Figure  4-7: Global warming potential of three different roofing alternatives .............................. 56 Figure  4-8: Ozone layer depletion of three different roofing alternatives .................................... 56 Figure  4-9: Acidification potential of three different roofing alternatives ................................... 57 Figure  4-10: Eutrophication potential of three different roofing alternatives ............................... 58 Figure  4-11: Carcinogens emission of three different roofing alternatives .................................. 58 Figure  4-12: Respiratory inorganics emission of three different roofing alternatives .................. 59 Figure  4-13: Respiratory organics emission of three different roofing alternatives ..................... 60 Figure E.1: Snap shot of the experiment pilot…………………………………………………………………………..  103 Figure F.1: Checking the normality assumption for nitrate……………………………………………………… 104 Figure F.2: Checking the normality assumption for ammonia  ……………………………………………...… 104 Figure F.3: Checking the independence residuals assumption for nitrate ………………………………..  105 Figure F.4: Checking the independence residuals assumption for ammonia ……………………………105 Figure F.5: Checking the constant variance assumption for nitrate ………………………………………… 106 Figure F.6: Checking the constant variance assumption for ammonia ……………………………………..106   xii  Glossary Symbol Definition (Unit) AHP Analytical Hierarchy Process Cd Cadmium CF Coconut Fiber CIRS Centre for Interactive Research on Sustainability Cr Chrome EC Electrical Conductivity EPDM Ethylene Propylene Diene Monomer FAHP Fuzzy-AHP Fe Ferrous G Gravel GAC Granular Activated Carbon GB Gravel Ballasted Roof GHG Green House Gas GR Green Roof System GWP Global Warming Potential K Potassium LCA Life Cycle Assessment LCI Life Cycle Inventory LID Low Impact Development Mn Manganese NO3-N Nitrate-Nitrogen NH4-N Ammonia-Nitrogen ORP Oxidation Reduction Potential Pb Lead PO4-P Phosphate-Phosphorous S Sand T Crushed Tile Tot- N Total Nitrogen TSG Crushed Tile +Sand + Gravel WB Wood Bulk Zn Zinc xiii  Acknowledgement First, I would like to extend my deepest thanks to my supervisors, Drs. Rehan Sadiq and Kasun Hewage for their inspiration. They have supported me throughout my thesis with their patience and knowledge. I will be forever grateful for the encouragement and guidance Dr. Sadiq provided me throughout my academic career. I would also like to acknowledge Dr. Keith Culver, the Director of the Okanagan Sustainability Institute (OSI) for technical contributions to this work. I would also like to thank my committee members, Dr. Sumi Siddiqua and Dr. Ahmad Rteil  for their guidance and insightful feedback. I would like to thank PLM laboratory members, particularly Dr. Bahareh Reza at the School of Engineering (SOE), University of British Columbia. I want to acknowledge Mr. Deren Sentesy, owner of En Circle Design Build, and Dr. Karen Liu, the director of R&D at Xero Flor Green solutions, who provided me with their knowledge in all phases of my research. Financial support from the Okanagan Sustainability Institute of the University of British Columbia, the Natural Sciences and Engineering Research Council of Canada’s ENGAGE program, XeroFlor America, and EnCircle Design Build is also acknowledged. Special thanks to my parents and siblings for their unconditional love and support. I have missed them deeply during these years. I offer my final and deepest thanks to Asal Hashemi, who has been at my side through this research. I am so lucky to have her in my life. xiv  Dedication  gÉ `ç _Éä|Çz ctÜxÇàá 1  Chapter 1 : Introduction This chapter highlights the motivation for this thesis. A brief history of green roof systems and the environmental impacts of conventional roofing systems have been described in Sections  1.1 and  1.2, respectively. Following this, the motivation and objectives of the study have been presented in Sections  1.3 and  1.4. The research methodology outline in the context of thesis organization has been provided in Section  1.5. Finally, the thesis structure is demonstrated in Section  1.6. 1.1 A Brief History of Green Roof Systems Green roofs have been used in buildings for many years. The first historical use of green roofs was found in the region of Mesopotamia located between the Tigris and Euphrates rivers around 500 BC (Osmundson 1999). Implementation of green roof as a modern means of architectural design for best management practices started in German-speaking countries 50 years ago (Osmundson 1999). However, green roofs have been used as a stormwater best management practice in North America only in the last decade. By widespread acceptance of “green building” principles symbolized by constructing high profile housing projects called the Hundertwasser-Haus in Vienna, Austria, roof gardens and facade greening became the center of attention for urban landscape architects, building companies, and environmental researchers (Osmundson 1999). The Research Society for Landscape Development and Landscape Design in Germany, or in German Forschungsgesellschaft Landschaftsentwicklung Landschaftsbau (FLL), developed a branch to study various benefits and impacts of green roofs on the buildings and environment in the 1970s (Dunnett and Kingsbury 2008). This organization is responsible for developing guidelines and standards for green roofing systems. The FLL guideline is frequently referenced in North America due to the absence of specified guidelines developed for the US or Canada. Green roof systems can be categorized based on the depth of the growing medium: extensive green roofs and intensive green roofs. Extensive green roofs, also called eco-roofs or performance roofs have the growing medium almost less than 150 mm, whereas intensive green roofs growing medium is 150 mm and higher (Bianchini and Hewage 2012a). 2  In North America, practitioners and building contractors tend to incorporate green roof systems into their projects. Research has shown that green roof system implementation dramatically increased every year e.g. 115% in 2011, and 24% in 2012 (Green roofs for healthy cities 2013a). The US Green Building Council’s Leadership in Energy and Environmental Design (LEED) program, and other green building initiatives and incentives designed for building owners and contractors are the primary motivations for such a dramatic growth of retrofitting and constructing green roof systems. 1.2 Environmental Impacts of Roofing Systems Roofing system depreciation is the most frequent phenomenon in building systems. Roofing elements deteriorate due to harsh conditions in winter and summer seasons. In the context of roofing deterioration, a conventional building in the United States requires roof replacement at least four times during its lifespan, which produces a large amount of solid waste (Coffelt and Hendrickson 2010). This situation could be worse in Canada due to harsher winter seasons and significant variation of temperature during summers. High volume of wastes from roofing systems can greatly increase the environmental impacts of the building industry (Bianchini and Hewage 2012b). Various techniques have been developed to minimize waste generation and environmental impacts and maximize the environmental performance of a roofing system over its lifespan. Green roof systems offer a wide range of environmental and ecological benefits and improve the quality of indoor and outdoor environments. Advantages and disadvantages of green roof systems are discussed in detail in  Chapter 2. 1.3 Research Motivation A comprehensive experimental investigation is required to assess claimed environmental benefits of green roof systems. The results of this experiment can assist architects and designers in comparing different roofing alternatives in the context of a specific project. Although green roof systems are known as best management practices (BMPs) or low impact development (LID) technologies, some aspects of the environmental performance of green roof systems have still not been comprehensively studied. The basic application of green roof systems is for stormwater management. Green roof systems can reduce stormwater volume and delay the peak hour by 3  capturing a portion of precipitation. Previous studies on green roofs runoff quality were controversial. Some studies showed that the water quality of green roofs runoff is lower than that of conventional roofs. Lower water quality of runoff may increase the amount of non-point source pollution in urban areas. Since the type of plants and soil formulation applied in green roof systems vary from one plant to another, it is necessary to conduct runoff quality field sampling for a plant before the implementation of the green roof system. The runoff water quality analysis can be used for further policy making and urban design for providing a plan for non-point source pollutant management in a city. Lack of proper implementation of green roof systems in local construction industries, building codes, and other important regulations and guidelines prevent designers, architects, and engineers from making an informed decision during the design process. Building components last for decades and require a large investment for construction, operation and maintenance (O&M), and disposal (Nelms et al. 2007). Therefore, decisions in this industry are costly and require a wide range of criteria to be considered. As a result, developing a framework for assessing the sustainability of roofing systems is necessary. 1.4 Objectives The focus of this research is developing a decision support tool for green roof systems’ selection based on the sustainability triple bottom line (TBL). The main objective of this study is to experimentally investigate the performance of green roofs on runoff water quality. An extensive green roof system pilot was constructed near the Engineering, Management and Education (EME) building located at the University of British Columbia–Okanagan campus (UBC-O), Kelowna, Canada. The pilot was run from June to December 2012 and the result of the analysis was implemented on developing different scenarios for non-point source nitrate removal in downtown Kelowna. In the next step, a framework for assessing the sustainability of roofing systems was developed based on the existing knowledge base and experimental study results of green roof systems. The framework helped to compare sustainability of extensive and intensive green roofs with gravel ballasted roof systems for the EME building located at UBC-O, Kelowna, Canada. Following are the specific objectives of the current research project:   Eru La  Dco1.5 RThe reseamethodoThe reseThe infortechnicaldevelopinxplore the enoff water qife cycle assgravel ballaevelop a comupled with esearch Merch methodlogy is comparch started mation and reviews. Tg the sustaiffect of addiuality. essment (LCsted roof. prehensiveLCA to estithodology ology to achrehensivelyFigurwith a com data was cohis informatnability assetional filteriA) of an ex sustainabilmate a relatiOutline ieve the ob described ie  1-1: Reseaprehensive lllected baseion was usessment framng materialstensive greeity assessmeve sustainabjectives of tn  Chapter 3 Crch Methoditerature revd on the prd for experework.  added to grn roof and ant framewoility index (he study is hapter 4 anology Outliniew focuseevious studiment desigeen roof layn intensive rk based on RSI) for rooillustrated ind  Chapter 4e d on the greies, buildingn, lifecycleers on the green roof wthe FAHP fing system Figure  1-1.  en roof sys documents assessment4 ith s. . The tems. , and , and 5  The experiment was designed in order to conduct the runoff water quality assessment. The analysis was performed based on natural rain events, and the main water quality characteristics were assessed. The results were compared with fresh water and reclaimed water guidelines.  In the second step, a sustainability assessment framework was developed for assessing the sustainability of roofing systems. The framework was based on the FAHP and LCA. Important criteria influencing the sustainability of a roofing system were identified. The framework evaluated extensive and intensive green roofs and compared the results with conventional roofing systems. 1.6 Thesis Structure The thesis consists of five chapters as shown in Figure  1-2. The research methodology was developed based on the objectives discussed earlier. In  Chapter 2, detailed background information required for this research has been provided. The advantages and disadvantages of green roof systems have been discussed. Issues related to runoff water quality of the green roof systems have also been discussed in detail. The life cycle methodology and its limitations as well as multi-criteria decision making methods are discussed in relation to the current research. In  Chapter 3, an experimental investigation of extensive green roof systems has been provided. The experiment is performed based on natural rain sampling of 2012. While the experiment sampling was completed, the effluent quality was compared with the effluent of control roofs. Moreover, an additional pre-treatment layer was added to green roof systems. The runoff quality of the enhanced green roofs was analyzed and compared with the generic green roof systems. Optimistic and conservative scenarios for retrofitting a part of Kelowna’s downtown buildings with an extensive green roof were performed to estimate the amount of non-point source nitrate and ammonia removal.  In  Chaptof buildiextensiveWhen thTBL critpurpose, were evaFinally, aer 4, a relating technolo and intense LCA emieria. TBL various subluated and a summary ave sustainabgies based ive green rossions werecriteria con-criteria werssessed usinnd conclusioFigure  1-2ility index on the TBLofs was an performed,sist of econe defined fog the Fuzzyn of the curChapter • Background• Research OChapter • Litreture ReChapter • Experimentof Green roChapter • DevelopingAssessmentChapter • Conclusion• Future work: Thesis Str(RSI) was d criteria. Falyzed and  the RSI fromic, envirr each TBL Analytical rent researc1bjectives2view3al Assessment ofs 4 a Sustainability Framework5sucture eveloped foor this purcompared wamework wonmental, criterion, aHierarchy Ph project is  r assessing pose, the enith a graveas constructand social nd then therocess (FAHpresented inthe sustainatire lifecycl ballasted ed based ocriteria. For roofing sysP) method.  Chapter 5.6  bility le of roof. n the  this tems  7  Chapter 2 : Background This chapter provides the background information for this thesis. The literature review covers the following main topics in this chapter:  Green roof systems including their components, types, environmental benefits, disadvantages, and costs.  Life cycle assessment (LCA) definition, steps, limitations, and its application in green roof systems.  Multi-criteria decision making (MCDM) with a specific focus on AHP. 2.1 Green Roof Systems Non-point source pollution in urban areas is responsible for significant water quality deterioration in North America (USEPA 2009a; Brezonik and Stadelmann 2002). Wash-off of impervious surfaces such as roof surfaces, and direct discharge of pollutants, fertilizers, and pesticides are sources of non-point pollution in urban areas (Gregoire and Clausen 2011; Brezonik and Stadelmann 2002; Egodawatta et al. 2009). While impervious roof surfaces coverage is about 12% in residential areas and 21% in commercial areas (Ellis 2013; Chester and Gibbons 1996; Boulanger and Nikolaidis 2003; Gregoire and Clausen 2011), it is necessary to manage the additional emission of these surfaces. Low impact development (LID) technologies have been developed as an appropriate response to non-point source pollution management in urbanized areas (Ellis 2013; Dolowitz et al. 2012). LIDs incorporate land use planning and engineered designs with the natural features of materials to infiltrate, filter, store, and detain runoff close to its source (McHarg 1995). Various LID practices such as bio-retention cells, green roofs, and grassed swales have been developed in recent years (Dietz 2007; Gregoire and Clausen 2011). Green roofs are increasingly used by urban and environmental planners to mitigate different environmental impacts of urban development. These roofs are covered with vegetation and growing medium equipped with a filtration layer. There are two types of green roofs: intensive and extensive. Intensive green roofs have a thick growing medium and may be planted with trees and shrubs, whereas extensive green roofs have thinner growing medium (≤ 10 cm) and are  planted wGregoire2.1.1 GA green drainage cities 20orientatioadditionastandard with gredrainage growing Regardleretrofittelayer of considereith drought and Clausenreen Roofsroof is a roosystem, filt13). Figuren. Water prl weight ofof care in fuen roofs shsupportive medium. Fss of those d on the basgreen roofsd in this res tolerant veg 2011; USE Componenf with addiration layer  2-1 depictoofing mem green roofrnishing anould be descomponentigure  2-1: interactionsic roofing sy will be inearch. etation suchPA 2009b).ts tional, high, a growings a genericbranes uses. Waterprod installationigned to sus are designTypical cros, green roofstem (Greevestigated a as Sedum a   quality wat medium, green rood in green rofing mem of materiapply greened to stands section ofs are modun roof for hnd layers bnd Delospeer proofing and vegetatf and illustoofs are thibranes shouls. Roof dra roof comp the additi a generic grlar in the mealthy citieselow the wrma (Berndand a root ion (Green rates differcker and abld be instainage systemonents. Greonal load o een roof ajority of c 2013). For ater proofitsson et al. 2barrier systeroof for heent layers ile to supporlled with a s in conjunen roof sysf the greenases and cathis reason,ng layer ar8 009; m, a althy n its t the high ction tems  roof n be  each e not 9  i- Vegetation Vegetation is the most important element that distinguishes green roofs from other types of roofing systems. Selection of a proper vegetation type is one of the challenging tasks in green roofs design. Each plant type has a different weight, benefits, and maintenance procedure. There are several requirements for plant type selection including non-invasive roots, not dropping large quantities of leaves or fruits, and not exceeding the load capacity of the structure (Osmundson 1999). Plant types should be resistant to climate conditions (wet or dry) and freezing in winter, and should be compatible with soil used in green roofs (Osmundson 1999). Moreover, plant types should tolerate temperature extremes and high winds, should quickly cover the growing medium, and should self-repair (Dunnett and Kingsbury 2008; FLL Guidelines 2002). Plant selection for extensive green roofs is almost limited to Sedum or grass mixes due to the conservative nature of the building industry. Limited plant types meet the requirements. These plants naturally grow in harsh, rocky environments with shallow soil. However, there is a question whether a broader range of plant types with the potential benefits to local bio-diversity might be appropriate for use in green roofs. Plant types can influence runoff quantity by providing a better evapotranspiration rate and use of supplemental growing medium for a range of plant species (Dunnett et al. 2005). ii- Growing medium The growing medium is the layer that supports plants and provides the most environmental benefits of green roofs. Growing medium porosity and density can impact hydraulic conductivity of green roofs and structure design reinforcement. The growing medium must support the needs of plants; it must be light and provide an optimized balance between water retention and drainage. Growing medium can significantly change the runoff flow and saturated hydraulic conductivity (Poulenard et al. 2001). iii- Filter layer The filter layer in green roofs prevents clogging of both the green roof drainage layer and the roof drainage system. The filter layer prevents washed particles of the growing medium and plant matters from entering the drainage layer. The filter layer should be water permeable, durable, portable, inexpensive, and tough (Osmundson 1999). In most green roofs a semi-permeable propylene fabric is used (Osmundson 1999; Dunnett and Kingsbury 2008). 10  iv- Drainage layer The drainage layer is a porous material that conveys the free water to the roofing drainage system (DeNardo et al. 2003). The drainage layer provides two critical factors for green roof systems: First, green roofs and especially extensive green roofs are planted with drought-tolerant plants. A drainage layer is required to convey the excess water during storms and avoid drowning the roots of these plants. Second, the drainage layer is required to maximize the thermal performance of the insulation layer (Dunnett and Kingsbury 2008). Granular drainage layers are simple and traditional methods, while other lighter materials like spongy materials, plastics or polystyrene modules, and recently recycled construction materials can be used as a drainage layer (Bianchini and Hewage 2012a). v- Root barrier The root barrier sits between the drainage layer and the water proofing layer. Plant roots naturally seek water and may cause roof membrane punctures and leaks. There are two main strategies to protect the waterproof membrane. The first is implementing a roll of PVC or waterproofing membrane as a root-impervious layer. Another strategy is implementing plastic or metal sheets to effectively isolate plant roots from the waterproofing layer (Dunnett and Kingsbury 2008). In addition, there are several other methods for protecting waterproofing methods, such as chemical root inhibitors (Peck and Kuhn 2001). The roof membrane material acts as a root barrier itself (Osmundson 1999). 2.1.2 Environmental Benefits of Green Roofs Green roof systems are among the technologies receiving increased attention for their potential to mitigate negative environmental impacts of the construction industry. Green roof systems may contribute to stormwater management (Berndtsson 2010; Teemusk and Mander 2007; Rajendran, Gambatese, and Behm 2009; City of Toronto 2010), reducing urban heat island effect (Nelms et al. 2007; Newsham et al. 2009; Rosenzweig, Stuart, and Lily 2006; Peck and Kuhn 2001), reducing the system’s energy consumption (Jaffal, Ouldboukhitine, and Belarbi 2012), and decreasing the total cost of systems over their lifespan (Castleton et al. 2010; Rowe 2011). Green roof systems can also improve building aesthetics and the overall building value (Getter and Rowe 2006; Long et al. 2006). Green roofs provide better protection with additional insulating 11  layers and may prolong the roofing system lifespan to at least 40 years, compared to conventional roofing systems with a 20 year lifespan (Kohler et al. 2001; Carter and Keeler 2008). i- Reducing energy consumption Green roofs’ impacts on energy consumption have been investigated widely. The insulation effects of additional materials reduce energy demand for cooling and heating the building during summer and winter (Jaffal et al. 2012; Newsham et al. 2009). Eumorfopoilou and Aravantinos (1998) examined thermal behavior of green roofs by applying mathematical calculations and stated that about 27% of the total solar radiation absorbed by the green roof is reflected and 60% is absorbed by plants. Green roofs can reduce the surface temperature and the temperature fluctuation of the roof. Onmura et al. (2001) conducted an experiment on green roofs’ surface temperature and compared them with white roofs in Japan. The results showed that green roofs reduce the surface temperature to 28-30ºC, while the surface temperature on conventional roofs is about 60ºC. Sonne (2006) studied the surface temperature of a roof with 50% green roof and 50% without green roof. The study showed that green roofs are able to reduce temperature fluctuation on the roofs’ surface. The surface temperature variation on a part without green roof was about 28°C, while the temperature fluctuation on a green roof was about 1.2°C. Experiments on green roofs show that green roofs increase the energy performance of a building. Liu and Baskaran (2003) argued that green roofs can reduce the energy demand for the building to about 75%. Santamouris et al. (2007) studied the energy performance of green roofs installed on a building in Athens, Greece. Santamouris et al. (2007) elucidated that green roofs significantly reduce the energy demand of a building cooling system during summer. This reduction varied from 6-49% for the whole building and 12-87% for the last floor. However, they argued that green roofs’ influence on a building heating load in winter is insignificant. Fioretti et al. (2010) explored green roofs’ impact on energy performance of a building in two different case studies. The results showed that green roofs have a better performance than conventional roofs and reduce daily energy demand for the building. Chan and Chow (2013) 12  simulated the energy performance of green roofs and argued that a green roof with a thicker soil medium and plant height provides a better thermal insulation effect. ii- Stormwater management Green roofs retain precipitation and gradually evapo-transpire it, whereas conventional roofs immediately drain stormwater into the downstream. Runoff peak flow can be reduced by temporary stored water in the vegetation and soil medium, which can extend the “time-of-concentration” and reduce local urban flooding (Gregoire and Clausen 2011). Despite the fact that green roofs have been recognized as a means of reducing the quantity of stormwater runoff, lack of sufficient evidence on the impacts of green roofs on stormwater runoff quality deters sustainable implementation of them. Green roofs can be used as an effective stormwater management tool in urban areas, because they are able to decrease the quantity of stormwater. Green roofs impact stormwater runoff through lowering and delaying the peak runoff. A study conducted in Vancouver, BC showed that a well-designed green roof is able to protect stream health and reduce the risk of flood in urban areas (Graham and Kim 2003). Green roofs are able to reduce runoff up to 100% in warm weather. However, the percentage of retained water in green roofs diminishes when there is not adequate time between each storm event (Moran, Hunt, and Jennings 2004). According to the experimental results, the retention capacity of green roofs is highly dependent on the volume and intensity of precipitation (Moran et al. 2004). Teemusk and Mander (2007) conducted an experiment on stormwater retention potential of green roofs. The results showed that green roofs are able to reduce light rainfall runoff up to 86%. In the case of heavy rainfalls, green roofs can only delay the runoff up to half an hour, and their impact on runoff volume is insignificant. Green roofs are able to reduce the runoff volume up to 18.9% in high density areas (Gill et al. 2007). iii- Urban heat island effect Buildings in high density areas reduce the amount of long wave radiation heat loss at night and increase the ambient temperature; this phenomenon is called the urban heat island effect (Oke 13  1995). Hard surfaces in urban areas prevent rainwater percolation into the soil and decrease evaporation, which may amplify the ambient temperature heating up. Green roofs can reduce this impact by increasing vegetated areas. Energy is used to evaporate water stored in green roof media, thereby reducing the ambient temperature. Quantifying the influence of green roofs in urban heat island reduction is difficult (Köhler and Schmidt 2003). Previous study results declare that by accounting for wind and precipitation, the effect of green roofs is still noticeable and green roofs are able to reduce ambient temperature of building by around 0.24°C (Bass et al. 2002; Pompeii 2010). iv- Improved air quality Different solutions are proposed for decelerating the declining air quality in cities. Green roofs are able to reduce local air pollution by decreasing summer extreme temperatures, and capturing particulates and gases (Rosenzweig et al. 2006). Green roofs reduce the ambient temperature of urban areas, which can directly reduce the reaction of NOx with volatile organic compounds (Rosenfeld et al. 1998). Moreover, Yok and Sia (2005) stated that green roofs reduce sulfure dioxide by 37% and nitrous acid by 21% in the ambient air. However, the overall nitric acid and particulates increased due to green roofs components and materials in the soil medium. 2.1.3 Green Roofs Concerns Although green roofs would bring various benefits to urban areas, there are some barriers that hold planners, developers, and building owners back. These barriers include the following: i- Economic consideration The costs of green roofs can be divided into four main categories: costs of green roof design, structural reinforcement, capital cost of green roof procurement, and operation and maintenance (O&M) costs. The initial costs of green roofs vary significantly. The initial costs of extensive systems in British Columbia, Canada varies from $12/ft2-$15/ft2, while for intensive systems it starts from $50/ft2 (Bianchini and Hewage 2012a). 14  The O&M costs vary significantly by green roof type and materials. Annual O&M costs are estimated to be $0.75-$1.50 per square foot (Bell et al. 2008). Design fees are about 5% to 10% of the green roofs cost. The structural reinforcement costs vary significantly based on green roofs type and weight. While extensive green roofs can be retrofitted on existing buildings without any additional reinforcement, intensive green roofs require complete structural redesign and reinforcement. 2.2 Purposes of Green Roof Runoff Quality Assessment The quality of green roof runoff is an important aspect of the performance of green roofs, especially when a green roof is combined with an open stormwater system (Berndtsson et al. 2009). Since the volume of runoff from green roofs is lower than from conventional roofs, it is generally assumed that green roofs improve the quality of runoff as well. Most of the previous studies emphasized the poor water quality of green roof runoff. For example, Berndtsson et al. (2006) studied heavy metals and nutrients including Cd, Cr, Fe, K, Mn, Pb, Zn, NO3-N, Tot-N, and PO4-P in green roof runoff and demonstrated that green roofs can be a source of contaminants. Similarly, other studies showed that the organic matter and nutrients in green roof runoff are higher than conventional roofs (Vijayaraghavan et al. 2012; Moran et al. 2004). Teemusk and Mander (2007) reported that a greater amount of nitrogen and phosphorus that had accumulated in green roofs washed away during heavy rainfalls and contaminated source water. In addition, utilizing fertilizers, especially on extensive green roofs, can be detrimental to runoff water quality. The mineralized nutrients from the fertilizers can be rapidly leached from the substrates and can impact runoff water quality. Although this effect can be reduced by using controlled-release-fertilizers, the nutrient leakage from green roofs is still higher than that of other roofs (Shaviv 2001). In the past, few studies have been conducted on developing effective media for improving the green roof runoff quality. A study at Pennsylvania State University elucidated that applying an additional filtering medium in green roof systems may improve the runoff quality (Long et al. 2006). In this study, several advanced filtration media such as granular activated carbon (GAC), zeolites, and polymers were used. Long et al. (2006) stated that while applying GAC media in 15  green roof systems might increase capital costs and maintenance expenditures, the runoff quality can be improved, especially in zinc removal. 2.3 Life Cycle Assessment (LCA) for Environmental Impact Analysis Building performance assessment tools have been developed to evaluate the performance of newly designed technologies and unconventional build processes. In general, two types of assessment tools are developed for the building sector. The first group is green building rating systems (GBRS) such as BREEAM, LEED, CASBEE, and SB-Tool. GBRS include tools that mainly focus on alternatives evaluation based on specific criteria (Reza 2013). In GBRS a number of selected criteria are evaluated on a scale ranging between low and high environmental performance. However, GBRS are based on scoring and weighting criteria that are not always efficient, which may lead to unrealistic and subjective results (Ali and Al Nsairat 2009). In addition, GBRS evaluation methods are based on a number of pre-defined criteria and applications of an innovative building design, new materials, and products, which might not confirm their environmental performance. Since GBRS are based on qualitative assessments, results might lead to an overestimated performance assessment of a new technology and thereby misinform the decision maker. The second group of building performance assessment tools consists of tools that use LCA in their methodology, such as BEES, Athena, Beat, EcoQuantum, and KCL Eco. LCA is an environmental technique to assess the “life cycle” environmental impacts of a product. LCA has been applied in a variety of systems and technologies from the 1990s. Based on the ISO 14000 series on environmental management, LCA is a systematic tool for investigating the environmental impacts of a product or service from the extraction of raw material to the end of life (Klöpffer 2005). The most important feature of LCA is that the product or service’s environmental impacts are evaluated over its life cycle, which is usually defined as “cradle-to-grave” analysis. This feature helps decision makers to gain a complete picture and comprehensive description of the environmental impacts of the objective. According to the ISO 14044 (2006) standard, a typical LCA consists of four phases:  Goal and scope definition  Life cycle inventory (LCI) analysis  Life cycle impact assessment (LCIA) 16   Life cycle interpretation 2.3.1 Goal and Scope Definition The goal of the LCA study should be defined in the first step. Based on the goal, the system function and functional unit are defined. After that, the boundary for the LCA analysis is specified. Also, processes studied in the LCA are described. A cradle-to-grave analysis considers manufacturing, transportation, construction, operation, maintenance, and demolition phases of both systems (ISO 14044 2006). 2.3.2 LCI Analysis  In this step, an inventory of materials inflow to the system and outflow back to the environment is analyzed. The inflows to the system are resources, raw materials, and energy used in the system. Outflow of the system is energy and emissions released to the environmental compartments including air, water, and soil media (Rebitzer et al. 2004). The main inventory of alternatives is performed by considering the life cycle phases of alternatives including manufacturing, transportation, operation & maintenance, and end-of-life phases. It should be noted that there is little reliable data available on the life span of building components (Kellenberger and Althaus 2009). 2.3.3 Life Cycle Impact Assessment (LCIA) The inflows and outflows of the roofing system life cycle are simulated by SimaPro1 software. This software is able to utilize various databases regarding different materials’ life cycle inventory from cradle to grave. Raw material extraction/acquisition, material processing, product manufacture, product use, and end-of-life are the life cycle stages considered by SimaPro. Associated environmental impacts are assessed using the IMPACT 2002+ method. The IMPACT 2002+ method considers the mid-point of impacts for modeling the environmental impacts                                                  1 It is a well-known eco-invent database used for applications such as carbon footprint calculation, product design, and eco design. The databases include eco-invent v.2, US LCI, ELCD, US Input Output, EU and Danish Input Output, Dutch Input Output, LCA Food, and Industry data v.2.  17  (Jolliet et al. 2003) and categorizes the environmental impacts to 9 categories including (Appendix A: Impact Category Description):  Carcinogens  Respiratory Inorganics  Ozone Layer Depletion  Respiratory Organics  Land Occupation  Aquatic Acidification  Aquatic Eutrophication  Global Warming Potential  Non-Renewable Energy Consumption After the impact assessment process, a more environmentally friendly roofing system is the one that produces low level of these adverse effects.  2.3.4 Life Cycle Interpretation The LCA results are discussed; the uncertainties and study limitations are identified and analyzed, the implications of the LCA study are established, and recommendations are proposed. The interpretation phase is often after the LCIA, however, it is not only restricted to that level and important conclusions may arise before the study is completed. There are several LCA studies on green roof systems. However, the results of different LCA studies cannot be compared directly with each other due to different goal and scope definitions, system boundaries, data sources, LCI analysis, assumptions, and uncertainties (Reza 2013). Accordingly, the conclusion is inconsistent. Some researchers argue that LCA contains uncertainties as a result of choosing different databases and life cycle impact assessing methods (Steen 1997; Lloyd and Ries 2008). Since LCA results are prone to uncertainty and vagueness (Harwell et al. 1986), deterministic results of LCA-based tools might not be very reliable. LCA results might overestimate or underestimate the environmental impacts of a technology, which is not desirable. Therefore, LCA-based tools 18  can be integrated with multi-criteria decision-making (MCDM) techniques in order to select the most sustainable solution. 2.3.5 LCA for Green Roof Systems The LCA of green roof systems has been comprehensively investigated in previous studies. Saiz et al. (2006) studied the LCA of extensive green roofs and compared their associated environmental impacts with those of standard roofs. Saiz et al. (2006) evaluated the LCA based on the energy consumption of an eight story building by implementing extensive green roofs. They argued that extensive green roofs reduce the environmental impacts by between 1% and 5.3%. Kosareo and Ries (2007) studied the life cycle environmental cost of intensive and extensive green roofs compared with conventional roofs. The LCA was performed based on the different life stages of all three roofing systems including fabrication, transportation, installation, operation, maintenance, and end of life. The study showed that green roofs can significantly reduce the life cycle environmental impacts of a building by decreasing the energy use. Life cycle cost analysis on green roof systems showed that green roofs are not the most economical alternative for the private sector. Some environmental scientists suggest that other environmental benefits of green roof systems should be considered. Blackhurst et al. (2010) argued that since the green roofs are not the best energy saving techniques and the life cycle cost analysis should consider both private and social benefits. Bianchini and Hewage (2012b) analyzed the life cycle cost-benefit of green roof systems based on the probabilistic net present value (NPV). The result showed that the payback period for extensive green roofs is about 4-5 years considering social and private benefits. Moreover, most LCA studies on green roof systems contain uncertainty on the analysis of the environmental impact contribution of the system (Peri et al. 2012). LCA studies ignored the environmental contribution of the small parts of the system without proper justification. Peri et al. (2012) declared that the extensive green roofs substrate, including fertilizers, have an environmental impact contribution during the green roof system lifespan. Green roof systems’ substrate and fertilizer provide NOx and N2O emission rates (Zaman et al. 2008; Shepherd et al. 1991). 19  The LCA of green roof systems’ specific materials or applications are also explored in previous LCA studies. The LCA analysis of low density polyethylene and polypropylene (polymers) materials used in the drainage layer showed that these materials produce higher amounts of NO2, SO2, O3, and PM10 emission during a lengthy green roof lifespan (Bianchini and Hewage 2012a). The additional air pollution due to the polymers’ manufacturing phase requires 13-32 years to be balanced (Bianchini and Hewage 2012a). Wang et al. (2013) highlighted the importance of the system condition and characteristics on the cost and benefits of the system. Their study stated that green roof systems can balance out the additional economic costs through environmental improvements. 2.4 Multi-Criteria Decision Making (MCDM) MCDM is the method of categorizing various non-dominant solutions for approaching a decision-making problem with multiple and conflicting criteria. Different methods of MCDM with various mathematical sensitivity analysis result in different or Pareto solutions for an individual decision-making problem (Bottero et al. 2011). MCDM methods are gaining credibility in sustainability-oriented development and green building technology choice, due to their capacity to support decision making in complex socio-economic systems at their intersection with the multi-faceted concept of sustainability (Wang et al. 2009). MCDM methods enable decision making to navigate through complexity to select most sustainable options. MCDM helps decision makers to resolve uncertainty-inducing conflicts among criteria, and to reconcile multiple objectives and perspectives (Wang et al. 2009; Sarkis and Talluri 2002). MCDM is a useful tool for environmental management as it is able to convert complicated and often conflicting interests and priorities of a decision maker into a more simplified and sequential process (Kholghi 2001). MCDM techniques can be used to balance the demands of Triple-Bottom-Line (Haimes 1992). “Sustainability” is generally considered a vague term in the decision-making process (Muga and Mihelcic 2008). MCDM can be applied to simplify the term “sustainability” into criteria and quantitative indicators (Tesfamariam and Sadiq 2006; “OECD ” 2001; Palme et al. 2005). 20  Reliable decisions can be made by selecting relevant decision criteria and indicators as well as by selecting the most appropriate MCDM methodology (Rosén 2009; Kruijf 2007). There are various MCDM methods developed for decision-making problems and systems, but there is no agreement on the “best” method for solving a particular decision-making problem in different conditions (Brunner and Starkl 2004; Schilling 2010). 2.4.1 Analytical Hierarchy Process (AHP) One of the most popular decision making frameworks is the analytical hierarchy process (AHP) (Dabaghian et al. 2008; Tesfamariam and Sadiq 2006). The AHP method ranks different alternatives based on the pair-wise comparisons to demonstrate the weights for each criterion. The AHP method was initially developed by Thomas L. Saaty in the 1970s (Saaty 1980). AHP has since gained currency in environmental and sustainability decision making such as sustainable energy decision making (Pilavachi et al. 2009; Hobbs and Horn 1997; Aras et al. 2004; Chatzimouratidis and Pilavachi 2009), water and wastewater management (Galal 2013; Dabaghian et al. 2008; Jaber and Mohsen 2001; Chung and Lee 2009), and built environment and technology selection (Wedding and Crawford-Brown 2007; Tupenaite et al. 2010; Reza et al. 2011; Medineckiene et al. 2010; ALwaer and Clements-Croome 2010; Ali and Al Nsairat 2009). The AHP method provides a platform for complex decision-making problems using objective mathematics to express systematically the subjective preferences of an individual or a group of decision makers (Saaty 1980; Mofarrah et al. 2013). The complex problem can be handled by structuring a hierarchy and the pair-wise comparisons are carried out between each two criteria. Normally, the pair-wise comparisons rank from 1 to 9, where 1 represents equal importance and 9 represents the extreme importance of one criterion over another (Tesfamariam and Sadiq 2006; Dabaghian et al. 2008). Once all pair-wise comparisons are obtained, the overall priority of each alternative is obtained by synthesizing the local and final preference weights (Tesfamariam and Sadiq 2006). The discrete scale of comparisons in AHP is simple and easy to use, but it is not able to handle the uncertainty and ambiguity2 present in assigning the ratings of different attributes (Chan and                                                  2 Vagueness is a property of a term or concept whose meaning is so broad that application of the term cannot distinguish legitimate from illegitimate uses of the term. Classic examples of vagueness include the concept of a 21  Kumar 2007). Environmental problems and issues are always containing lack of information, scarcity of data and vagueness (Tesfamariam and Sadiq 2006). It is often difficult to compare different criteria due to scarcity of information. Vagueness type uncertainty can be propagated using fuzzy set theory (Zadeh 1965). 2.4.2 Fuzzy-AHP Analysis Fuzzy-Analytical Hierarchy Process (FAHP) is a compensatory approach for selecting an alternative and justifying the problem. This approach is able to account for data scarcity and vagueness in decision-making problems (Kahraman et al. 2003; Tesfamariam and Sadiq 2006). Due to the complexity of preferences and the fuzzy nature of the comparison process, using interval judgments is more pragmatically reliable than use of fixed value judgments (Kahraman et al. 2003). FAHP is also able to respond systematically to ambiguity, multiplicity of meanings, lack of essential data, and vagueness caused by linguistic content and subjectivity in judgment (Tesfamariam and Sadiq 2006). 2.4.3 FAHP Calculations In order to achieve the goal of the evaluation, pair-wise comparisons are taken by the decision maker. Triangular fuzzy numbers (TFNs) (1෨, 3෨, 5෨, 7෨, 9෨) are used to show the importance or priority of elements in pair-wise comparisons. By applying TFNs in pair-wise comparisons, fuzzy judgment matrixes ܣሚ	ሺܽ௜௝ሻ are constructed. The fuzzy membership can be utilized by using the α-cut value. The decision maker’s level of confidence in his preferences and judgments can be defined by the α-cut value. Interval sets of values for fuzzy numbers can be generated by α-cut value. If ܽ௜௝ = (m1, m2, m3), then m2 is the mid value of ܽ௜௝ and is one of the integers from 1 to 9 used in AHP method. Let us assume that m2 - m1= m3 - m2=δ is constant. If 0= δ, then values are crisp and fuzziness of comparisons is not incorporated in comparisons. If 0< δ<0.5, then TFNs do not have any crossover points and the cognitive fuzziness does not cast completely. If δ is greater than 1, then the degree of confidence decreases and fuzziness                                                                                                                                                              ‘heap’ of sand: the meaning of ‘heap’ is insufficiently determinate to provide application criteria enabling precise specification of the number of grains of sand required to constitute a “heap” as opposed to some other unit such as a “dune” or a “mountain”.  increasesδ value isIf the deevaluatioequal imp(DI) and and accelAfter theconstructwise comfuzzy weWhile thvalue of The diffefuzziness~ ~1(ij iA a~ ~i ijw A . Zhu et al.  practically cision makns are qualiortance (EIextreme imerate the pa pair-wise ced. Since huparison matights ሺݓ෥	ሻ ae fuzzy weicomparisonsrence betw in the comp~ 1/... )ina ~1( ...iA A (1999) suggindicates ther is not atative, lingu), weak impportance (Exir-wise comFiguromparisonsman judgmrices shouldre computed                            ghts are com can be calceen the minarisons. n~ 1)in ested that 0e conflict ofble to specistic variabortance (WI) to TFNsparison proce  2-2: Lingu are accompents are sub be calculat using fuzzy                                      puted, the ulated. Theimum and m.5<δ<1 is m degree of cify the pairles shown iI), strong im, linguistic vess. istic definitlished, pairject to incoed and be h arithmetic                                       range of as mid value aximum vore applicabonfidence an-wise compn Figure  2-2portance (Sariables canions in FAH-wise compnsistency, tigher than 9operations o                                      sociated uncshows the malues showsle to cast thd fuzzinessarisons by can be usI), demonst easily convP arison matrihe consisten0%. The locver ሺܣሚ௜௝ሻs.                                       ertainty andost likely v the range e fuzziness .  using TFNed. By assigrated imporerted into T ces ሺܣሚ௜௝ሻ cacy ratio of al preferenc                                       the most lalues of weiof uncertain22 since s or ning tance FNs n be pair-es or    (1)   (2) ikely ghts. ty or 23  Final preferences of the alternatives are obtained by aggregating the local priorities at each level. This process is carried out from the alternative level to the goal level, therefore the final preferences can be computed as:                                                                                                                                 (3) The final fuzzy score (FAi) of each alternative is the fuzzy arithmetic sum over each global preference for each alternative Ai.                                                                                                                                    (4) RSI can be calculated by defuzzifying the final fuzzy score of each alternative using Chen’s ranking method (Chen 1985). max 11 minmax min 1 1 max min 1 1( )( )1[ 1 ]2 ( ) ( ) ( ) ( )Aic ac ac a b c c aI bRS a                                                                     (5) Using this method, amin is the minimum of the smallest final fuzzy score among all alternatives’ final fuzzy scores, and cmax is the maximum of the biggest final fuzzy score among all alternatives’ final fuzzy scores. The resulting RSI value gives a quantitative measure of the sustainability level of different green building technologies. The alternative with the highest RSI value is the most sustainable technology for implementation. 2.5 Sustainability Assessment Framework In recent years, green building practices have been developed as a way to mitigate the long-term negative environmental impacts of buildings (Yoon and Lee 2003). Integrated design approaches and technologies have been implemented in green buildings to reduce the adverse impacts of buildings and urban development on the ambient environment and its occupants (Ali and Al Nsairat 2009). Although there are various methods for assessing technologies implemented in green buildings, there is a lack of comprehensive and adequately precise framework for integrated evaluation of ~ ~ ~1~ ~1 1. ,k kkG w GG w~1nAi kkF G24  economic costs and benefits, environmental performance, and social aspects of these technologies (Nelms et al. 2007). Expensive mistakes may be made by overestimating the performance of green building technologies, and sustainable building practices may lose credibility. Approaches to decision-making that seek to include environmental impact among reasons for action are challenged by the inherent complexity in the decision matrix, and the multi-disciplinary nature of the problem of inclusion of environmental impact (Gallopin et al. 2001). Moreover, such decision processes are prone to data scarcity and lack of knowledge (Harwell et al. 1986). Even where sufficient data are available, evaluation criteria often permit subjective judgments and contain ill-defined terms, which in turn give rise to uncertainty in the form of vagueness (Tesfamariam and Sadiq 2006). One major drawback of LCA is that the LCA ends up with categorized environmental impacts of the alternatives, which require a MCDM. In addition, linking the environmental impacts with socio-economic preferences of a process is challenging for many organizations as they would have to handle a complex dilemma with ambiguity and conflicting criteria (Chan and Wang 2013). Therefore, a simple and more cost effective framework is required. FAHP can be applied as a complement for LCA shortcomings. FAHP offers the advantages of AHP and most importantly, it is able to handle the uncertainty and ambiguity present in sustainability dilemma and system selection (Chan and Kumar 2007). Few studies tried to integrate FAHP with LCA to evaluate and index green technologies (Kang and Li 2010; Zheng et al. 2011). Chan et al. (2013) employed an extended fuzzy-AHP to evaluate the greenness of a product design. They estimated a green index for a product based on an FAHP evaluation throughout every stage of products’ life cycle. Alternative products were ranked over their lifecycle stages without performing a full LCA. However, performing a comprehensive LCA is essential to consider long-term sustainability performance of a product. Moreover, the evaluation was only based on environmental impacts of the products and socio-economic impacts were not considered in the evaluation. 2.6 Summary In this chapter, the background information of the current research project is comprehensively reviewed. Green roofs are considered as LID practices. Green roof systems and their associated 25  layers are defined. Green roof systems are roofs covered with a layer of vegetation. Green roof systems consist of a root barrier layer, drainage layer, filtration layer, growing medium and vegetation. Green roof systems are categorized into two main groups: extensive green roofs (with a growing medium< 15cm) and intensive green roofs (with a growing medium > 15 cm). Although green roofs are generally developed for stormwater management, they can provide various environmental benefits including energy saving, urban heat island effect reduction, and air pollution reduction. However, green roofs’ additional initial cost, operation and maintenance costs, and leak hazard may undermine their benefits. LCA is a strong method for analyzing the environmental impacts of a product or a technology from cradle to grave. LCA is able to categorize the environmental impacts of a product into various environmental categories; as a result, LCA results in a multi-criteria problem that requires MCDM techniques to be solved. MCDM techniques can be applied to a wide range of decision making problems with various conflicting criteria. MCDM techniques provide a range of non-dominant solutions for a decision-making problem. Applying different MCDM techniques may result in disparate solutions. AHP is a popular MCDM technique that can be easily implemented by pair-wise comparisons of alternatives against each criterion. The final “best” solution is the alternative with the best score. Since AHP comparisons are based on human judgments, the evaluations may contain uncertainties, vagueness, and ambiguity. Fuzzy calculations can be used in AHP to handle these shortcomings. The application of sustainability assessment frameworks for assessing the sustainability of the green built technologies is discussed. FAHP is a strong MCDM framework that can be used with LCA to mitigate the shortcomings and vagueness of LCA.   26  Chapter 3 : Experimental Investigation of Green Roofs Runoff Water Quality The runoff quality of extensive green roofs was experimentally assessed in this chapter. The experiment consisted of conventional roofs, generic extensive green roofs, and extensive green roofs equipped with an additional pre-treatment layer. This chapter comprises extensive green roof runoff quality performance and enhances the performance of extensive green roofs by adding an additional pre-treatment layer to the green roof systems. The questions are whether green roofs can significantly change the runoff water quality, and if applying an additional pre-treatment layer can improve the runoff water quality. To answer these questions, runoff water quality from sixteen green roofs (with or without an additional pre-treatment layer) have been investigated and compared with four conventional roofs. The quality assessment is based on reclaimed water guidelines and fresh water guidelines for Canada. The experiment materials and method are explained in Section  3.1. The result of the experiment and analysis are shown in Section  3.1.4 followed by discussion and limitations in Section  3.3. 3.1 Materials and Method This section discusses the study of the experiment site plan, experiment pilot design, rainfall sampling process, and chemical analysis of samples. 3.1.1 Study Site and Experiment Pilot Design A green roof pilot experimental setup has been established near the EME building of the University of British Columbia–Okanagan campus (Kelowna, BC, Canada) under semi-arid weather conditions (Klock and Mullock 2001). The roof systems have been designed and built with 3 ft x 5 ft multi-plywood assemblies. The study sections are constructed with the same principles of full-scale roofs. All roof tops have been placed on a 3° slope to simulate common roof design. The pilot consists of eight green roofs, a gravel ballasted roof, and a control roof that was layered with EPDM (ethylene propylene diene monomer) (Figure  3-1). The roofs runoffs were collected at the lower end of roof tops. Each roof was divided into two equal, discrete spaces with a median divider. There is a generic green roof with typical layers, 27  which is considered as a control green roof; two generic green roofs with ten times additional simulated rain with local utility (tap) water (to examine the effect of aging on green roof’s runoff water quality); and five green roofs enhanced with an additional pre-treatment layer. Figure  3-1: Green roof pilot experimental setup at University of British Columbia (Okanagan campus) The selection of an additional filtration layer was based on the ability of the filtration material to amplify the performance of green roofs by decreasing the nutrient leakage at a reasonable price. It was assumed that the additional filtration removes turbidity and suspended solids from runoff. Gravel and sand filters are the most efficient filter media for water treatment (USEPA 1999). Moreover, coconut fibre and crushed tile are other examples of common media for physical water treatment (Nkwonta and Ochieng 2009). As a result, a variety of filtration materials (i.e., washed sand, coconut fibre, wood bulk, crushed tile, and a combination of sand plus crushed tile) for pre-treatment of the stormwater treatment has been applied in the green roof layers between the growing medium and filter sheet. A complete list of green roof pilot tests has been summarized in Table  3-1. The growing medium of the green roofs used in this experiment is a mixture of lightweight, mineral based materials. The soil is consist of porous aggregate and organic matter derived from composted plant materials, biosolids, and/or manure compost (Xeroflor America 2013). It is estimated that the mat thickness is 1 1/4” with 5.5 psf field weight and 8.5 psf saturated weight (Xeroflor America 2013). Sedum and Delosperma are used for the green roof vegetation medium, which has been used in most green roof experiments all over the world (Berndtsson 28  2010). The same pre-cultivated XF 301 vegetation mat provided by Xeroflor America was applied for all the green roof assemblies (Appendix B) (Xeroflor America 2013). Xeroflor pre-cultivated mats were planted with a mixture of drought-resistant green roof species such as Sedum and Delosperma (Xeroflor America 2013). Table  3-1: Characteristics of different roof assemblies in the present study Name Insulation Additional Filtration Growing medium & Vegetation Replication Description Accelerated Age GR EPDM - Pre-vegetated Mats 4 Generic green roof with additional simulated rain water GR EPDM - Pre-vegetated Mats 2 Generic green roof GR+S EPDM Sand Pre-vegetated Mats 2 Generic green roof with an additional pre-treatment filtration layer GR+CF EPDM Coconut Fiber Pre-vegetated Mats 2 Generic green roof with an additional pre-treatment filtration layer GR+WB EPDM Wood Bulk Pre-vegetated Mats 2 Generic green roof with an additional pre-treatment filtration layer GR+CT EPDM Crushed Tile Pre-vegetated Mats 2 Generic green roof with additional pre-treatment filtration layer GR+TSG EPDM Crushed Tile+ Sand Pre-vegetated Mats 2 Generic green roof with an additional pre-treatment filtration layer GB EPDM - - 2 Generic gravel ballasted roof EPDM EPDM - - 2 Control roof 3.1.2 Rainfall Effect Field experiments were conducted with natural rainfall events to evaluate the impact of green roofs and an additional filtering layer on runoff water quality. A preliminary study manifested that at least ~30 to 40 mm of rainfall was required for the soil to reach the saturation point and 29  generate runoff water from green roofs. Hence, samples were collected once runoff started from all green roofs. A bulk sample of rain water and green roof runoff samples were collected in a pre-cleaned, 500 ml polyethylene container during each rainfall event and refrigerated (4⁰C) until analysis (Figure  3-2)(USEPA 2007).  Figure  3-2: Runoff samples 3.1.3 Chemical Analysis Rainwater and green roof runoff samples were collected to compare the amount of pollutants in a wet atmospheric deposition and green roof runoff. Rainwater samples were taken into the laboratory for water quality characterization and analysis. The characterization of NO3-N, NH4-N, ORP, EC, pH, color, and turbidity were performed using HACH instruments (Figure  3-3).  Figure  3-3: Hach sampling instruments 30  3.1.4 Design of Experiment (DOE) The experiment is designed to answer which variable is most influential and whether other uncontrollable variables are impact on the experiment. The current experiment is a statically designed experiment based on Fisher’s factorial concept. Fisher’s factorial concept enables the experimenter to use all performed tests and investigate the main effects (Montgomery 2008). Analysis of Variance (ANOVA) is used for determining the significance of the relationship between different treatments. One-way ANOVA with blocking the intensity of the precipitation and post-ANOVA analysis with α level of 0.05 were performed to determine if any significant change was observed between enhanced green roofs, green roofs, control roofs, and precipitation. ANOVA has three assumptions: Normality, constant variance and independence (Montgomery 2008). If any of these assumptions violated, ANOVA is not applicable and other methods should be applied (Montgomery 2008). 3.2 Results The current study considered the mixed effect of atmospheric deposition, green roof’s materials, and fertilizers on runoff quality, as it was impossible to distinguish pollutants and emission load generated by each of those sources. Experiment sampling collection was started in September 2012 and continued until the end of December, however there was no runoff observed in September and precipitation changed to snow in early December. Therefore, the observation and analysis represents sampling from early October to the end of November 2012. Maximum and minimum precipitation that led to runoff was in late October and in the middle of October, respectively. The amount of rainfall event precipitation from September to December 2012 is shown in Figure  3-4. Sampled precipitation events ranged from 2.3 mm to 11.8 mm and included only rain. 31   Figure  3-4 : Daily precipitation in Kelowna (Canada Climate 2013) Results show that the green roof with an additional layer of coconut fibre and wood bulk produced the lowest quantity of runoff before saturation. The runoff retention or delay was due to the combined effect of additional retention capacity of coconut fibre/wood bulk and the possibility of water absorption in such a porous media. It is noticeable that the water retention capacity of the green roof with coconut fibre was about 40% higher than the generic green roof. Analysis showed that none of ANOVA assumptions are violated and ANOVA can be performed for statistical analysis of the current experiment (Appendix F). Overall, the retention capacity of green roof assemblies was strongly dependent on the weather conditions, which may accelerate evapotranspiration phenomena. The results of runoff water quality characterization and experimental analysis are described in the following sections. 3.2.1 pH Green roofs ability to buffer acid rain and pH fluctuations is one of the benefits of green roofs (Berndtsson et al. 2006; USEPA 2009b). The results from the experimental analysis show that runoff from green roofs has a higher average pH level as compared to EPDM or a gravel ballasted roof. The higher level of pH in green roofs runoff is due to the buffering capacity of green roofs during rainwater passage through the green roof media. This is a considerable environmental advantage of green roofs as it can decrease the direct discharge of acidic runoff to natural water recipients (Berndtsson et al. 2009). Figure  3-5 shows the average and range of pH from generic green roofs and the conventional and gravel ballasted roofs.  The statiwater (acsame as water (p-producedperformaGR+WB0.001). TAlthoughthe greenthe gener3.2.2 NThe resu(Figure significangreen roogreen roois signifigeneric g6.006.507.007.508.008.50pHstical analyscelerated agthe pH of ravalue < 0.0 higher pH nce in buff had an unhis aspect  the additio roof, the pHic green rooitrate and Alts show th 3-6 and Fitly higher lf produced f runoff is tcantly lowereen roof pFigure  3is shows thae green rooinwater. Th01). Similathan rainwaering the pHproductive can be expnal coconut  of GR+CFf. mmonia at all 16 ggure  3-7) aevel of nitra90% less nihe same as tr than the aroduced sig-5: Averaget the pH of fs) was stae runoff of r to Berndtster. Howeve fluctuationeffect on thlained by thfiber layer i runoff wasreen roofss comparedtes as comptrates comphe ammoniammonia conificantly l pH for the pthe generic tistically lowGB and EPson et al.'sr, the additis. For instae bufferinge structuren GR+CF d higher thanproduced h to conveared to otherared with th concentratincentration ess nitrates ilot scale evgreen roof wer than othDM roof to(2009) findonal filterinnce, the ad capacity o and bufferid not impro the other roigher amountional roo rooftops (pe GB roof.on in gravelin EPDM and ammonents ith additioner green rops had a lowings, the geg layer decrditional wof the green ing capacityve the buffoftops withnts of NOfs. The GB-value < 0.0The concent ballasted roroof runoff.ia than theal simulatedof types aner pH thanneric greeneased greenod bulk layroof (p-val of wood ering capac the excepti3-N and N roof prod01). The geration in geof runoff, w In addition enhanced g32   rain d the  rain  roof  roof er in ue < bulk. ity of on of H4-N uced neric neric hich , the reen  roofs in concentrathe accelroofs resuBerndtssresult waconcentragreen roonitrate duNH4-N, t(Berndtssof Mendquality wconventio0.005.0010.0015.0020.0025.00Nitrate (mg/l)this experition of nitrerated age glted in the lon et al. (20s similar totion in the f soil mediring water phe amount on et al. 20ez et al. (20here the rnal green roment (p-vaogen nutrienreen roofs owest conce06) reported the output green roof rua (Robertsoassage fromof Tot-N d06). Poor tre11). Mendeoof with tiof. Figure  3-6lue < 0.00ts among this higher thntration of  that greenof this studnoff can ben and tiedje the soil meecreases, watment of nz et al. (20le surface r: Average N1). In conte other grean in rainfaammonia (0 roofs are a y. An incre explained b 1987). In tdium. It is phich may reitrate runof11) studied esulted in itrate for thrast, the Gen roofs. Thll concentra.01 mg/l) insource of Nase in NO3-y the nitrifihis process,ossible thatsult in a sif from GR+the effect oa higher coe pilot scaleR+S genere concentrations. Accel the runoff. O3-N and NN and a decation proc the ammon by increasink of Tot-NT was similf roofing mncentration events ated the hition of nitraerated age gH4-N, and crease in Ness that occuia is oxidizng in NO3-N in green ar to the finaterial on ru of nitrate 33 ghest te in reen their H4-N rs in ed to  and roofs dings noff than   3.2.3 CThe coloThe greemost instRunoffs from all was red. the otherSince thegrowing in green roofs. In contraThe changreen roo0.0000.1000.2000.3000.4000.5000.6000.7000.8000.900Ammonia (ppm)Folor and Tur of runoff in roof runoances the grfrom GR+Wgreen roofs In general, t side, the EP thickness omedium waroofs, the rust, the runofge in the grf media (gr0000000000igure  3-7: rbidity s the most nff was veryeen roof runB, GR+T, aexcept GR+he average DM runoff f the greenshed out aftenoff colourf from the geen roof ruowing soil aAverage Amoticeable co similar in off was cleand GR+TSGCF were ycolour of GRwas colourle roof mats ir huge rain changed toreen roof ennoff colour nd additionmonia for tntrast betwecolour to thr and witho were darkellow to dar+CF was dss and quiten this exper events. By  faded yellohanced withwas due to al layer if aphe pilot scalen green roe leachate ut any turbid yellow or ok orange inarker than th clear. iment is jusdecreasing tw in gener coconut fiinorganic aplicable). Te events of runoff anfrom wetlanity. range colou colour. The other root 1 inch, thhe concentric and accelber was chand organic he intensityd other roofds. Howevered. The rue GR+CF ruftops’ runofe nutrients iation of nutrerated age gnged to darksubstances i of runoff c34  tops. r, in noffs noff f. On n the ients reen  red. n the olour  was depeprecipitatFigure  3-highest leand GR+by fine cOn the oturbidity level andwas margroof. 3.2.4 OORP mesystem (Gin the sys0.0100.0200.0300.0400.0500.0600.0700.0800.0Turbidityndent on thion, the run8 shows thevel among S. The averomponents ather hand, tlevel of acc lower thaninally simixidation Reasurements erardi 2007tem (Gerard000000000e precipitatioff was less turbidity mthe green roage turbiditynd materialhe turbidity elerated age generic grelar to the seduction Pocan be used). Table  3-2i 2007). on duration coloured. easured fromofs, followe of GR+TSs in crushedFigure  3-8:level of acc green roofsen roof. It icond acceletential (OR to determ shows the  and intensi rooftop rud by gravelG was more tile and san Average Tuelerated age was almosts necessary rated age gP) ine what biORP levels ty. During rnoff. The tu ballasted, c than 200 Nd. rbidity  green roof the same asto mention reen roof aological reaand the relaainfall withrbidity of GonventionalTU, whichs was abou the bulk rathat the turbnd lower thction is takted reaction higher inteR+TSG wa roofs, GR+ might be cat 2.32 NTUinwater turbidity of GRan generic ging place i that is occu35 nsive s the WB, used  . The idity +CF reen n the rring  Range o+100 -50 -10 +25 In genera2007). FodisinfectiThe resulThe roofother grerunoff frGR+CF, 252, 255performaThe resulbe transl200250300350400450500ORP (mV)Table  3-2: f ORP (mV)  to +350   to +50 0 to -225  to +250 l, a higher r instance, on procedut of ORP astop covereden roofs. Thom gravel GR+T, and, and 260 nce than thts show thatated into a .0.0.0.0.0.0.0Oxidation-RRBiologBiologilevel of ORvalues abovre. Studied sessment ofFigure  3- with EPDMis outcome ballasted ro GR+TSG rmV, respee generic gr EPDM andhigher wateeduction Poelated ReactNitrificationDenitrificatioical Phosphorucal PhosphoruP is an indice 485 mV green roofs  harvested ra9: Average  and the gshows that ofs and EPunoff had actively. Theen roof wi gravel ballar quality (atential (ORPion  n s release s removal ator of a saare detrimenshowed a sinwater is sORP for theravel ballasthe probabilDM is obv lower avee two acceth the ORPsted roofs pt least in th) ranges foOxidRedumple with htal to aquaubstantial lehown in Fig pilot scale eted rooftopity of chemiously lowerage of ORPlerated age around 290rovide a higis parameter different aDescriptiation of ammoction of ammo-  - igher watertic life and vel of ORPure  3-9. vents  had a highical or micror than in g among th green roo and 283 mher level ofr) and lowctivities on nia to nitratenia to Nitrate quality (Sucan be used (Suslow 2er ORP thabial reactioreen roofs.e other rooffs had a bV, respecti ORP, whicer probabili36 slow  as a 007).  n the ns in  The tops: etter vely. h can ty of  nitrificatinitrificatichemical3.2.5 CThe meanit is in tstandard conductivlayer waAlthoughaverage cEPDM rconductiv3.2.6 ATo exprebe compwith otheConductivity (µS/cm)on. It shoulon decreases in the growonductivity conductivihe typical rgreen roof,ity of the rs still highe the conduconductivityoofs had theity averageFigppraisal ofss the resultared with Cr green roo0100200300400500600y(µ)d be mentios because aing medium ty of rainfalange of am the conducunoff from tr than raintivity of ru was signific lowest con. ure  3-10: A Green Roos of experimanadian guif runoff expned that by  portion of . l bulk collebient rainwtivity of ruhe other grewater and noff from aantly higheductivity thverage Confs Runoff Qental analydelines for deriments. Inincreasing tnutrients ancted during ater (Yaziznoff was sen roofs (Frunoff fromccelerated ar than runofan the otheductivity fouality sis of runofomestic rec this study, he age of grd ammonia the experim et al. 1989ignificantly igure  3-10)  EPDM ange green rof from EPDMr roofs andr the pilot scf water qualaimed watthe overall een roofs, twash off orent was abo; Lee et alhigher thanenhanced wd the graveofs was m and grave they were ale events lity, the obter, other waquality of ghe probabili react with ut 31 µS/cm. 2010). Fo rainwater.ith an additl ballasted ore constanl ballasted rnear the raained resultter standardreen roof ru37 ty of other  and r the  The ional roof. t, the oofs. infall  s can s, or noffs 38  from the experimental analysis were expressed with respect to guidelines and standards for treated wastewater, domestic reclaimed water, and the quality of urban runoff (available in literature). Table  3-3 summarises the Canadian guidelines for domestic reclaimed water, the quality of runoff from previous green roof experiments, and fresh water quality. Table  3-4 summarises the average of current experiment roofs’ runoff water quality. Table  3-3: Required fresh water, domestic reclaimed water, and green roof runoff quality Parameter Unit Fresh water3 Domestic4 reclaimed water Green roof runoff quality  pH  6.5 - 8.5 6.6 - 8.7 7.45 Nitrate mg/L < 0.5  < 0.1–0.8 0.076 Ammonia mg/L < 0.1 < 1.0–25.4 0.084 Turbidity NTU <1 22-200 153 ORP mV 3907 - - Conductivity mS/cm - 325–1140 3203 Based on Health Canada (2010) standards, the pH of all runoffs was in an acceptable range. With the exception of accelerated age green roofs, the pH of green roofs runoff was higher than the average pH of rainfall, EPDM, and the gravel ballasted roof. The pH of accelerated age green roofs runoff was almost the same as the pH of rainfall. In general, the pH of runoff from all rooftops is in a neutral range (6.5-8.5) and acceptable. Since the acid rain event is not observed in the experiment during rainfall events, the ultimate capacity of green roofs to buffer the acid rain before the pH of the growing medium drops below the applicable level of plant growth or water quality guidelines should be investigated.                                                  3 Guidelines for Canadian Drinking water quality, 2012. 4 Canadian guidelines for domestic reclaimed water, 2010. 5 Mendez et al. 2011. 6 Brendtsson et al. 2009. 7 Suslow, T.V., 2007. 39  Table  3-4: Average water quality parameters for each paired roof Roof Type GR GR+CF GR+CT GR+WB EPDM GR+TS GB GR+S Acc. Age GR Rain pH 7.39 7.23 7.15 7.13 7.09 7.11 7.08 7.22 7.28 7.26 Nitrate 3.97 3.44 3.00 3.64 2.92 5.48 5.29 11.59 4.29 0.69 Ammonia 0.016 0.017 0.028 0.018 0.200 0.050 0.016 0.026 0.066 0.090 Turbidity 19.10 15.70 7.22 135.37 13.74 238.66 120.74 117.47 3.21 0.31 ORP 272.64 252.77 258.20 278.30 315.53 279.45 308.86 280.95 275.64 348.92 Conductivity 440.85 459.90 447.85 448.65 37.70 453.90 170.05 451.35 428.05 28.70 According to the USEPA primary drinking water guidelines, turbidity of 95% of samples should be less or equal to 0.3 NTU (USEPA 2004). The turbidity of the runoff from all rooftop systems was higher than standards and needed to be treated for further utilization. According to Health Canada (2010), the median and maximum acceptable turbidity of reclaimed water is less than 2 and 5 NTU, respectively. The result showed that the turbidity of accelerated age green roof runoff is near to the acceptable range for domestic reclaimed water used in toilet and urinal flushing (Health Canada 2010). The concentration of nitrate in all green roofs, EPDM, and the gravel ballasted roof was significantly higher than the concentration of nitrate in the rainfall. The concentration of nitrate was higher than the acceptable range for fresh water or even domestic reclaimed grey water. The average concentration of nitrate of green roof runoff in this experiment was significantly higher than previous studies on green roofs. The higher concentration of nitrate in green roof runoff can be accounted for by addition nutrients in fertilizers used in green roofs. But the source of nitrate in the EPDM and gravel ballasted roof is different and is due to dissolving nitrogen in the air during precipitation. The ammonia concentration in green roof runoff was lower than the rainfall ammonia concentration. The decrease in ammonia concentration in green roof runoff is a result of the nitrification process that occurs during water passage from green roof media. During the nitrification process, a portion of ammonia is oxidized to nitrate. The concentration of all green roofs, except the accelerated age green roof and GR+S, is in the admissible range for fresh water. It is noticeable that as the green roof ages, the nitrification process in green roof soil decreases 40  dramatically. The thin thickness of green roof soil medium in this experiment and accelerating the green roof usage showed that green roofs need maintenance after a period of operation. The turbidity of runoff of generic and enhanced green roofs was higher than rainfall turbidity and was not in an acceptable range for fresh water. It can be noted that the additional GR+WB, GR+TSG, and GR+S decreased the performance of the green roof and increased significantly the turbidity of runoff. The average turbidity of the other green roofs was moderately higher than the EPDM roof’s runoff. Although the accelerated age green roofs had the lowest turbidity among the roofs, the effluent turbidity was higher than the fresh water and average rainfall turbidity. 3.2.7 Scenario Analysis for Nitrate and Ammonia Removal The existing results of the extensive green roof experiment can be applied to optimistic and pessimistic scenarios for estimating the nitrate and ammonia removal in a selected part of the Kelowna downtown (Figure  3-11). The first scenario is an optimistic scenario that assumes that 50% to 75% of all roofs are covered with XeroFlor extensive green roofs. The pessimistic scenario assumes that only 10% to 25% of buildings are retrofitted with extensive green roofs. The total surface area of roofs in the selected part of the Kelowna downtown was estimated to be 65 ha using Google Earth aerial maps. The total roof surface area is about 16-19.5 ha, which is about 25-30% of the total urban roof surface area. 41  Figure  3-11: Selected area of the city of Kelowna (created by Google Map) Based on the historical records from 1993 to 2012 collected by the Kelowna International Airport weather station, the average rainfall precipitation can be categorized into light rain (0.2 mm to 5 mm), moderate rain (5 mm to 10 mm) and thunderstorm (above 10 mm). The days with different types of rainfall and average monthly precipitation is shown in Figure  3-12 (The Weather Network 2013).  Figure  3-12: Days with light rain, moderate rain and thunderstorm in Kelowna (The Weather Network 2013) 0510152025303540450246810121416Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov DecDays with precipitation Light rainModerate rainThunderstormAverage MonthlyPrecipitation (mm)42  The probability of each type of rain event and average precipitation in each day of every month is estimated based on the historical records and shown in Table  3-5. Table  3-5: The probability of rain and average precipitation in each day based on historical records (The Weather Network 2013) Probability of rain % Average precipitation (mm/day) Month Light rain Moderate rain ThunderstormLight rain Moderate rain ThunderstormApr 33 7 0 0.83 0.33 0May 47 3 3 1.17 0.17 0.25Jun 40 10 3 1 0.5 0.25Jul 27 7 3 0.67 0.33 0.25Aug 30 7 3 0.75 0.33 0.25Sep 27 7 3 0.67 0.33 0.25Oct 43 7 3 1.08 0.33 0.25Nov 43 3 0 1.08 0.17 0 The average volume of runoff produced by 17 ha of roofs in downtown Kelowna is estimated and shown in Table  3-6Error! Not a valid bookmark self-reference.. Table  3-6: The average estimated volume of runoff produced by 17 ha roofs  Volume of precipitation (m3) Month Light rain Moderate rain ThunderstormTotal precipitation volume (m3) Apr 141.67 56.67 0 198.33 May 198.33 28.33 42.5 269.17 Jun 170 85 42.5 297.5 Jul 113.33 56.67 42.5 212.5 Aug 127.5 56.67 42.5 226.67 Sep 113.33 56.67 42.5 212.5 Oct 184.17 56.67 42.5 283.33 Nov 184.17 28.33 0 212.5 Previous studies on XeroFlor extensive green roofs estimated that the retaining capacity of these roofs is 6% to 10% (Taylor 2008). The warmer weather increases the evapotranspiration, which accelerates the removal of retained water in green roofs. The retaining capacity of extensive 43  green roofs during warm seasons is considered to be 10%, and during the cold months (Oct., Nov., and Apr.) it is estimated to be 6%. The amount of nitrate and ammonia that can be removed yearly can be estimated by multiplying the retaining capacity of green roofs with the average and probability of each type of rainfall events in each month. It is notable that the winter season (Dec., Jan., Feb., and Mar.) is excluded from the analysis since when green roofs are covered with snow, they are considered the same as conventional roofs. The amount of nitrate removal based on optimistic and pessimistic scenarios is shown in Table  3-7. Table  3-7: The amount of nitrate removal using XeroFlor extensive green roofs Nitrate Removal (g/month)  Optimistic scenario (50% to 75% is retrofitted) Pessimistic scenario (10% to 25% is retrofitted Period 75% 50% 25% 10% Apr 803 536 268 107 May 1817 1211 606 242 Jun 2008 1339 669 268 Jul 1434 956 478 191 Aug 1530 1020 510 204 Sep 1434 956 478 191 Oct 1148 765 383 153 Nov 861 574 287 115 Each year 11035 7357 3678 1471 Entire life (40 years 441405 294270 147135 58854  44  The ANOVA test on the result of ammonia in roof runoff showed that the ammonia concentration in green roof runoff is significantly lower (α<0.05) than conventional roofs. The amount of ammonia removal can be estimated by a summation of runoff volume reduction and reduced ammonia concentration in green roof runoff. The amount of ammonia removal is shown in Table  3-8. Table  3-8: The amount of ammonia removal using XeroFlor extensive green roofs Ammonia Removal (g/month)  Ammonia removal by retaining the precipitation Ammonia removal by reducing the release concentration  Optimistic scenario (50% to 75% is retrofitted) Pessimistic scenario (10% to 25% is retrofitted Optimistic scenario (50% to 75% is retrofitted) Pessimistic scenario (10% to 25% is retrofitted Period 75% 50% 25% 10% 75% 50% 25% 10% Apr 54 36 18 7 755 503 252 101 May 121 81 40 242 981 654 327 131 Jun 134 89 45 268 1084 723 361 145 Jul 96 64 32 191 775 516 258 103 Aug 102 68 34 204 826 551 275 110 Sep 96 64 32 191 775 516 258 103 Oct 77 51 26 153 1079 719 360 144 Nov 57 38 19 115 809 539 270 108 Each year 736 490 245 1371 7084 4722 2361 944 Entire life (40 years 29427 19618 9809 54856 283341 188894 94447 37779 Results show that XeroFlor extensive green roofs are able to reduce the non-point source pollution of nitrate and ammonia without changing the pH. Based on the different scenarios, the nitrate removal can be estimated to be 300-450kg in the optimistic scenario and 60-150kg in the 45  pessimistic scenario during the extensive green roofs’ lifespan. Moreover, extensive green roofs are able to remove 200-300kg of ammonia in the optimistic scenario and 40-100kg in the pessimistic scenario. 3.3 Discussion The runoff quality performance of generic and enhanced green roofs with an additional filtering layer (e.g. sand, tile coconut fiber, and wood bulk) were compared with the runoff of conventional roofing systems including EPDM and gravel ballasted. Nitrate, pH, EC, ammonia, turbidity, ORP, and colour were measured in roof runoffs and harvested rainwater. The results show that green roof runoff quality is lower than that of the other conventional roofs. Although some of the additional preliminary treatment layers improved the quality of runoff, the harvested rainwater from these roofs needs an additional primary and secondary treatment for further drinking water use. In particular, the harvested rainwater from green roofs needs treatment for turbidity, colour, and nitrate. However, with a small portion of dilution, green roof runoffs can be used as domestic, reclaimed grey water and meets the Canadian domestic reclaimed guidelines. The runoff quality of green roofs increases by aging. The additional layer of coconut fiber and crushed tile improved the runoff quality in some directions but the overall quality of harvested rainwater was still poor. The EPDM runoff had a better quality than the other harvested rainwater in this experiment. Bulk rainwater samples had a lower concentration of the contaminants than runoff from green roofs and conventional roofs. Green roofs can improve the quality of runoff with respect to specific water quality characteristics. The overall runoff quality with respect to the determined characteristics can be considered as acceptable for reclaimed water use. The concentrations of NO3-N and ammonia in green roof’s runoff were higher than runoff from conventional rooftops. The performance of green roofs increases by applying a coconut fiber layer. The coconut fiber layer improved the ORP level of runoff and provided more clear water. Moreover, coconut fiber media increased the retaining capacity of the green roof. In addition, extensive green roofs are able to reduce the amount of nitrate and ammonia produced as non-point source pollution in urban areas. Optimistic and pessimistic scenarios for retrofitting 46  extensive green roofs in downtown Kelowna show extensive green roofs are able to significantly reduce the amount of nitrate and ammonia without changing the pH. This amount reduction can benefit the Okanagan Lake environment as it is vulnerable to non-point source nitrate and ammonia emission.   47  Chapter 4 : Sustainability Assessment Framework for Green Roof Systems A sustainability assessment framework for green roof systems is developed in this chapter. The framework is developed based on the FAHP decision support tool and LCA. The framework is able to calculate a relative sustainability index (RSI) score by using the LCA results, available literature, and interviews and discussions with experts in relevant fields. The sustainability assessment framework is described in Section  4.1. The LCA study of roofing systems including identifying goal, scope, system functional, and the system boundary, as well as inventory analysis are discussed in Section  4.2. Finally, results are presented and discussed in Sections  4.3 and  4.4, respectively. 4.1 Sustainability Assessment Framework Recently various experimental studies and environmental assessment methods have been conducted to assess the environmental performance of green building technologies. Methods developed to date are, however, insufficient for accurate quantitative estimation and evaluation of triple-bottom-line (TBL) sustainability performance objectives (i.e. economic, environmental and social) in the context of green building technologies. The main objective of this chapter is to develop a green building sustainability evaluation framework to estimate the sustainability performance of new green building technologies under conditions of uncertainty and lack of sufficient knowledge. The framework provided here utilizes a fuzzy-analytical hierarchy process integrated with a cradle-to-grave life cycle assessment to address interactions and influence of various TBL criteria. The developed framework is implemented for evaluating and comparing sustainability performance of an extensive green roof and a gravel ballasted roof (both located at the Engineering, Management and Education building at UBC’s Okanagan campus) with an intensive green roof (located at the Centre for Interactive Research on Sustainability at UBC’s Vancouver campus). FAHP is aggregated to LCA to help decision-makers to augment and in that way improve the reliability of LCA results. FAHP-LCA employs conventional LCA capabilities, including life cycle inventory (LCI) and life cycle impact assessment (LCIA).  The FAHbe used aexpressinalternativgreen buthe schem            4.2 LIn this sethe systemP-LCA is as a sustainag a final scoe. Easily coilding technatic FAHPFigCA Study ction, the L, and the fuIdentHpplicable tobility assessre value of mparable Rology and e-LCA for geure  4-1: SuCA for roofnctional unDefininifying sustainabiliierarchy construc multi-critement frameeach alternaSI scores cavaluate life neral green stainability ing systems it are discusg the green buildproty criteria tion Data cFuzzy pairwiFuzzy weigFuzzy weightsFinal green buildiria decision-work for asstive as the rn help deciscycle impacbuilding tecPerformancis framed. Tsed.  ing technology evblem Life cyLifeollectionse comparisonht calculation defuzzificationng evaluation indemaking proessment of elative sustion makers tts of each ahnology evae Assessmenhe goal of taluation cle inventory of a cycle impact assexblems. The green buildainability ino select thelternative. Fluation. t Diagram he LCA, sclternatives ssmentFAHP-LCAing technolodex (RSI) o most sustainigure  4-1 sope, bounda48  can gies, f that able hows ry of 49  4.2.1 Identifying Goal, Scope The goal of the current LCA is to analyze the environmental impacts of three different roofing systems. Based on CEN/TC 350 recommendations for sustainability assessment of construction works, three life stages were considered: the manufacturing and construction stage, the use stage, and the end-of-life stage. The transportation phase produces less impact (about 1.5%-2.4% of total emission) than the operation and manufacturing phases (Peuportier 2001). Since the transportation phase emission was less than 5% in the preliminary analysis and the roofing system alternatives were located in different cities, the transportation phase was ignored in the case study. 4.2.2 Functional Unit and System Boundary Since the lifespan of green roofs is longer than other roofing systems, the functional unit for this analysis is defined based on the system that had a longer lifespan. Green roof systems’ lifespan is reported between 40 and 60 years (Carter and Keeler 2008; Kohler et al. 2001). A conservative functional lifespan (40 years) was selected as the functional lifetime of roofing systems in this study. This selection reduces the uncertainties related to the estimation of environmental effects of the systems. During this period, both green roof types require maintenance every year and more thorough rehabilitation every 10 years. By contrast, gravel ballasted roofs require less maintenance than green roofs. The lifespan of a gravel ballasted roof is about 20 years. Thereafter, deterioration of roofing system components may negatively influence the roofing structure and decking components, at which time the gravel ballasted roof should be replaced (Kohler et al. 2001; Carter and Keeler 2008). In order to compare the two roof technologies over the same functional time, it was assumed that two gravel ballasted roofs would be constructed and used in sequence during the functional time. 4.2.3 Inventory Analysis In this step, an inventory of materials inflow to the system and outflow back to the environment are analyzed. The inflow to the system is resources, raw materials, and energy used in the system. Outflow of the system is energy and emissions release to the environmental compartments including air, water, and soil media (Rebitzer et al. 2004). 50  The main inventory data of alternatives is performed by considering the life cycle phases of alternatives including manufacturing, transportation, operation and maintenance, and end-of-life phases. It should be noted that there is little reliable data available on the life span of building components (Kellenberger and Althaus 2009). The LCI analysis was conducted for each life stage of each roofing system alternative. Information about components of green roof systems was collected based on FLL Guidelines (2002), ASTM E 2400-06, E2397 – 11, E2399 – 11 and E2398 − 11 (ASTM 2013a; ASTM 2013b; ASTM 2013c; ASTM 2013d) for green roof systems, and ASTM D7655/D7655M – 12 (ASTM 2013e) and roofing guidelines RCABC (2011) for gravel ballasted roofs. Different components of each roofing system are shown in Table  4-1. The information regarding materials manufacturing and fabrication, energy chains, and transportation was mainly extracted from the SimaPro software databases. Scenarios for end-of-life of products were defined based on the available literature. It is noteworthy that while recycling processes prevent landfilling of recycled materials, the total cradle-to-grave-to-cradle manufacturing and transportation environmental impacts of recycled materials increase due to additional processes required for recycling. Table  4-1: Material types for individual elements of roofing system for unit of area  Intensive green roof Extensive green roof Conventional roof  Material Material Material Structural support/decking Steel Steel Steel Underlayment Concrete Concrete Concrete Root Barrier Non-Rotting Polypropylene Fibers Polypropylene Polypropylene Drainage Layer Recycled Polyethylene, Polystyrene Waffled Panels Polystyrene Waffled Panels Filter Fabric Non-Rotting Thermal Consolidated Polypropylene Micro-Perforated Polypropylene Micro-Perforated Polypropylene Top layer/ Growing Medium Growing Medium For Semi-Intensive Green Roofs Growing Medium For Extensive Green Roofs Gravel  Plan 4.3 RTo evalucriteria. Fwise com4.3.1 CThe hieraand alterAlternatienvironmwere defFigure  4-t Material esults ate RSI, theAHP compparisons, fuonstructingrchy structunatives. Thves are evaental and sined under 2. FiguDr FAHP analrises differezzy results d a Hierarchre from tope goal of thluated and ocio-economthe main re  4-2: Hieought ResistPlants ysis shouldnt steps inclefuzzificatiy Structur to bottom ie hierarchyweighted oic performcriteria. Therarchical treant  be performuding conston, and finae ncludes goa is the evaver the maance. For a structurede for compaSedum ed considerructing a hil ranking asl, main criteluation of gin TBL per better eva hierarchy rison of rooing the TBLerarchy strusessment. ria, sub-critreen buildiformance cluation, varfor this stufing systems-  criteria as cture, fuzzy eria (if exisng technoloriteria incluious sub-crdy is show 51 main pair-ting), gies. ding iteria n in   4.3.2 LThe anallifecycle into sub-on the fu4.3.3 LConduct The LCAmid-poinIMPACTFigure  4-that the emanufactand 60%of non-rresponsibroof lifes%ife Cycle Inysis in this cimpact anacriteria for bzzy compariife Cycle Imof the LCI a results shot impact ca 2002+ imp3 depicts thmission couring phase of respiratoenewable enle for emispan. F020406080100%ventory Dahapter is pelysis are peretter compsons performpact Analynalysis enaw the magntegories. Thact assessme contributintribution o in intensivry inorganicergy durinsion of moreigure  4-3: ta rformed baformed usinarison and aed by grousis bled evaluatitude of imis analysisent method.on of each f the manufe green roos. Moreoverg the inten than 50% Intensive grsed on the Rg SimaPro ssessment. ps of judgmion of the epacts of ea is conductlife stage ofacturing phafs is respon, the manufsive green of global ween roof lifeSI assessmsoftware. TThe RSI asents. nvironmentach roofing sed by Sima the intensivse is highesible for moacturing pharoof lifespaarming gase stages content framewohe TBL crisessment is l impact ofystem alterPro softwae green roor than the ore than 95%se consumen. The ends during thribution EndOpeMark. The LCteria are divconducted b each alternnative in vare based onf. It can bether phases of carcinos more than-of-life stae intensive g‐of‐liferationnufacturing52 I and ided ased ative. rious  the  seen . The gens  60% ge is reen   The contintensiveHoweveralmost 560%, whimpact cagreen roolife stageFigure  4-roofs the95% of cgravel baimpacts. %ribution of  green roo, the contri0%. The noich is lowertegories anf. As with t of the extenF5 shows th manufacturarcinogens llasted roof More than 6020406080100%an extensivf, the manbution of tn-renewable than that od can be exphe intensivesive green rigure  4-4: e contributioing phase prare produceis responsib0% of GHGe green rooufacturing he manufac energy cof the intenslained by th green roof, oof. Extensive grn of life stoduces the hd during thle for almoss are emittef’s life stagphase prodturing phasnsumption iive green roe smaller aalmost 50%een roof lifages of the ighest amoue manufactut 80% of land during thees is shownuces more e in respiran the manuof. This redmount of m of GHGs ae stages congravel ballant of enviroring phase. d occupatio end-of-life  in Figure than 95% tory inorgafacturing puction can aterials usedre emitted dtribution sted roof. Anmental imThe end-ofn and aquatstage. EndOpeMa 4-4. As witof carcinonics emissihase is lessbe seen in  in the exteuring the ens with the pacts. More-life stage oic eutrophic‐of‐liferationnufacturing53 h the gens. on is  than other nsive d-of- other  than f the ation  Althoughof each sother. Thalternativconsumpoperationthan the roof in iconsume%F the environtage, it is nee comparisoes is showtion for all  phase energravel ballats end-of-lis lower amo020406080100%igure  4-5: mental impcessary to cn between tn in Figurroof types ogy consumpsted roof. Tfe phase is unts of non-Gravel ballaact contribuompare the he non-renee  4-6. As ccurs duringtion of bothhe non-renhigher thanrenewable ested roof liftion of eachenvironmenwable energcan be see the produc intensive anewable ener other roofnergy durine stages con lifespan statal impacts y consumptn, most oftion phase. d extensivegy consumps. In generg its lifespantribution  ge can showof roofing syion of three  the non-reDue to bett green roofstion of theal, the exte. EnOpMa the imporstems withdifferent ronewable ener insulation is slightly l gravel ballnsive greend‐of‐lifeerationnufacturing54  tance  each ofing ergy , the ower asted  roof 55   Figure  4-6: Non-renewable energy consumption of three different roofing alternatives The global warming potential of all roofs is depicted in Figure  4-7. Based on the SimaPro software analysis, most of the CO2 emission for all roofs occurs during the end-of-life phase. The gravel ballasted roof emits the highest amount of CO2 equivalent. The extensive green roof is responsible for the lower amount of emission during its lifespan.  01002003004005006007008009001000ManufacturingIntensive roofOperation End of Life Life cycleNon‐renewable energy (MJ)Intensive green roofExtensive green roofGravel ballasted roof020406080100120140ManufacturingIntensive roofOperation End of Life Life cycleGlobal Warming Potential (kg CO2 eq)Intensive green roofExtensive green roofConventional56  Figure  4-7: Global warming potential of three different roofing alternatives The ozone layer depletion potential of all roofs is shown in Figure  4-8. As is illustrated, the CFC-11 equivalent emission of all roofs during the operation phase is negligible. The extensive green roof emits lower amount of CFC-11 equivalent than other roofs during the manufacturing and end-of-life phases.  Figure  4-8: Ozone layer depletion of three different roofing alternatives The acidification potential of all roofs is demonstrated in Figure  4-9. The SO2 equivalent emission of the extensive green roof is lower than other roofs. Manufacturing and operation phases produce 90% of SO2 equivalent emissions over the roof’s lifespan. Although the operation phase emission of all three roofs is almost the same, the extensive green roof produces significantly lower amounts of SO2 equivalent emission during the manufacturing phase. 00.00000050.0000010.00000150.0000020.00000250.000003ManufacturingIntensive roofOperation End of Life Life cycleOzone Layer Depletion (kg CFC‐11 eq )Intensive green roofExtensive green roofGravel ballasted roof57   Figure  4-9: Acidification potential of three different roofing alternatives The eutrophication potential of the three roofing alternatives is shown in Figure  4-10. The eutrophication potential is shown based on the kg PO4 equivalent. As can be seen, the extensive green roof PO4 equivalent emission is nearly half of the emission of the gravel ballasted roof. Most of the PO4 equivalent emission occurs in the end-of-life stage. In contrast, the operation phase emission is negligible. 00.050.10.150.20.250.30.35ManufacturingIntensive roofOperation End of Life Life cycleAcidification Potential (kg SO2 eq)Intensive green roofExtensive green roofGravel ballasted roof58   Figure  4-10: Eutrophication potential of three different roofing alternatives The carcinogen emissions of three roofing systems based on kg C2H3Cl equivalent is shown in Figure  4-11. As it can be seen, the extensive green roof produces lower amount of carcinogenic emission in compare with the other roofing systems.  Figure  4-11: Carcinogens emission of three different roofing alternatives 00.0050.010.0150.020.025ManufacturingIntensive roofOperation End of Life Life cycleEutrophication Potential (kg PO4eq)Intensive green roofExtensive green roofGravel ballasted roof00.511.522.5ManufacturingIntensive roofOperation End of Life Life cycleCarcinogens (kg C2H 3Cl eq)Intensive green roofExtensive green roofGravel ballasted roof59  The respiratory inorganic emissions of roofing systems is estimated based on kg PM2.5 equivalent and shown in Figure  4-12. The extensive green roofs produces lower amount of respiratory inorganic particles over its lifecycle.    Figure  4-12: Respiratory inorganics emission of three different roofing alternatives Figure  4-13 shows the respiratory organic emissions of roofing systems over their lifecycle and each life phase. The extensive green roof produces lower amount of organic emissions. The lifecycle emission of the extensive green roof is about 50% of the emission of the gravel ballasted roof. 00.010.020.030.040.050.06ManufacturingIntensive roofOperation End of Life Life cycleRespiratory inorganics (kg P.M 2.5 eq)Intensive green roofExtensive green roofGravel ballasted roof60   Figure  4-13: Respiratory organics emission of three different roofing alternatives The detailed LCA confirms the outstanding environmental performance of the extensive green roof in this study, with the exception of energy savings and carcinogen chemical emissions. However, selecting a roofing system alternative is not a single-attribute decision-making process and depends on other factors. As a result, LCA needs to be supported by a multi-criteria decision making problem, which requires additional tools to be solved. 4.3.4 Selection of Sustainability Indicators for the Hierarchy The indicators for the objective hierarchy were selected based on information collected from the peer-reviewed literature and public information. The criteria were selected to achieve the goal of the hierarchy, which is the selection of the most sustainable roofing system. TBL criteria are able to connect environment to the society and economy. Therefore, the second level consisted of sustainability TBL criteria (Reza et al. 2011; Lerario and Maiellaro 2001; Ostendorf et al. 2011; Waheed et al. 2009). The sustainability TBL criteria were divided into sub-criteria to increase the clarity and specificity of the hierarchy. The selected sub-criteria should be independent, concise, and complete and satisfy the upper criterion objective. Moreover, sub-criteria should be relevant to 00.0050.010.0150.020.0250.03ManufacturingIntensive roofOperation End of Life Life cycleRespiratory organics (kg C2H4 eq)Intensive green roofExtensive green roofGravel ballasted roof61  the goal. For this purpose, thirteen sub-criteria were selected based on recommendations articulated in the relevant literature (Levett 1998; Lindholm et al. 2007; UNDPCSD 1995). i- Environmental Impacts Environmental impacts of roofing systems in the FAHP hierarchy were subdivided into six groups: climate change, stormwater management, runoff water quality, resource depletion, waste management, and environmental risks. Climate change refers to the current studies of a wide range of indicators showing that climate change is occurring globally due to a gradual warning of the climate system (Canada climate change 2013; B.C. Air quality 2013). Global warming is the consequence of emission of CO2 and a large number of trace gases such as CH4 and NOx (IPCC 2007). Pair-wise comparisons of different roofing alternatives with respect to climate change criterion have been done based on LCA impact assessment results. Stormwater management is an important challenge in urban areas. High intensity thunderstorms increase runoff of precipitation. This runoff is carried by the sewer systems to streams and may result in floods downstream (Environment Canada 2003). Stormwater management goals include retaining a volume of precipitation and delaying the peak runoff. New approaches to urban planning include use of roofing systems to contribute to stormwater management. These approaches also consider the contribution of roof systems to runoff water quality control. Pair-wise comparisons of different roofing system scenarios with respect to stormwater management and runoff quality control were based on available literature on roofing systems such as Berndtsson (2010), Vijayaraghavan et al. (2012), Moran et al. (2004), Zimmerman et al. (2010), and other studies on green roof and conventional roof stormwater management impact and runoff water quality. Resource depletion refers to the use of renewable and non-renewable resources, with particular concern for non-renewable resources and prolongation of their availability via reduced use and use of alternatives. Alternatives that consume less raw materials and energy in their lifespan are preferred. The comparisons of different alternatives with respect to resource depletion were based on LCA impact assessment results and available literature on the energy performance of 62  green roofs and conventional roofs (Liu and Baskaran 2003; Jaffal et al. 2012; Desjarlais et al. 2008). Waste management is an important criterion in environmental impact assessment. This criterion shows raw materials consumption. Pair-wise comparisons were conducted based on the results of the LCA in mineral extraction category. ii- Economic Concerns The economic concern criterion includes three sub-criteria: capital cost, maintenance cost, and renewal cost. The pair-wise comparisons were based on the available literature on green roof and gravel ballasted roof costs, together with direct contact with roofing system manufacturers, maintenance providers, and green roof owners like UBC Okanagan campus, Carter and Keeler (2008), and Bianchini and Hewage (2012) studies. iii- Social Concerns Social concerns criteria are selected based on the most common social concerns about implementing a roofing system, as documented in the literature. The third main TBL criteria were sub-divided into roofing weight, fire safety, durability, and vulnerability of area. Pair-wise comparisons are made based on the available literature such as the Green Roof Guide (2011), Bianchini and Hewage (2012), and a Sutton et al. (2012) study on prairie-based green roofs, guidelines, and expert judgment. 4.3.5 Weighting of Sustainability Indicators Main assessment areas, main criteria, and associated sub-criteria are weighted with respect to their individual importance under the current case study. Data extracted in this paper was compiled through published literature, open ended interviews, and workshops. Data related to economic concerns under TBL performance criteria were collected based on available literature like journal papers and green roof cost reports, building owners, and informal interviews with consulting and manufacturing companies in North America. Other required information and appropriate pair-wise comparisons about roofing systems were collected based on available literature, and results of LCA and UBC-LCA group discussions. The FAHP weightings were calculated using an Excel spread sheet. Table  4-2, Table  4-3, Table  4-4, and Table  4-5 depict the 63  relative pair-wise comparison of TBL criteria and associated sub-criteria in the current FAHP model. The consistency ratio of each judgment was checked to confirm that it is higher than 90%. Table  4-2: The pair-wise comparison of TBL criteria for roofing system   Social Economic Environmental Social 1    1/4  1/3 Economic 4   1   2   Environmental 4    1/2 1   Table  4-3: The pair-wise comparison relevant to Social criterion   Structural Design Force Fire Safety Durability Vulnerability of Area Structural Design Force 1 7 5 8 Fire Safety 1/7 1 1/3 3 Durability 1/5 3 1 5 Vulnerability of Area 1/8 1/3 1/5 1 Table  4-4: The pair-wise comparison relevant to Economic criterion   Initial Cost O&M Replacement cost Initial Cost 1 7 9 O&M 1/7 1 4 Replacement Cost 1/9 1/4 1 Table  4-5: The pair-wise comparison relevant to Environmental criterion   Climate Change Waste Management Runoff Quality Stormwater Management Resource Depletion Environmental Risks Climate Change 1 5 4 5 3 4 Waste Management 1/5 1 1/3 1/4 1/6 2 Runoff Quality 1/4 3 1 1 1/2 3 Stormwater Management 1/5 4 1 1 1/3 3 Resource Depletion 1/3 6 2 3 1 4 Environmental Risks 1/4 1/2 1/3 1/3 1/4 1 In order to demonstrate the application of the proposed FAHP-LCA method, the results were compared under different α-cut levels. For α-cut levels, 0.5 and 1 values are considered and alternatives were scored. The alternatives’ score under different α-cut levels can be considered as 64  a decision support tool since it is able to show the level of confidence and uncertainty in choosing the most sustainable alternative. Fuzzy pair-wise comparisons have been made among different impact categories and their sub-criteria based on available literature, LCA results, and experts’ judgement. Then the local and final fuzzy weights of alternatives and criteria were calculated. Table  4-6 provides the results of fuzzy local weights of alternatives after pair-wise comparisons, and    65  Table  4-7 shows the final fuzzy weights of alternatives and criteria. Table  4-6: Fuzzy local weights of (w෥ ) with δ=1 Level 2 W1 Level 3 W2 Level 4 W31 (Conv. roof) W32 (Extv. green roof) W33 (Int. green roof) Social concerns 0.08 0.12 0.20 Structural design 0.50 0.65 0.84 0.47 0.67 0.93 0.16 0.24 0.37 0.06 0.09 0.13 Fire safety 0.07 0.10 0.14 0.06 0.07 0.10 0.20 0.28 0.40 0.46 0.65 0.89 Durability 0.15 0.21 0.29 0.06 0.08 0.11 0.40 0.46 0.52 0.40 0.46 0.52 Vulnerability of Area 0.04 0.05 0.07 0.07 0.10 0.17 0.09 0.17 0.26 0.55 0.73 0.99 Economic limitations 0.31 0.54 0.88 Initial cost 0.67 0.78 0.90 0.44 0.63 0.88 0.21 0.29 0.42 0.06 0.08 0.11 O & M 0.13 0.16 0.20 0.46 0.65 0.89 0.20 0.28 0.40 0.06 0.07 0.10 Replacement cost 0.05 0.06 0.07 0.48 0.66 0.90 0.20 0.27 0.39 0.05 0.07 0.09 Environmental impacts 0.21 0.34 0.61 Climate Change 0.27 0.46 0.78 0.10 0.13 0.18 0.63 0.78 0.95 0.06 0.08 0.12 Waste Management 0.02 0.04 0.07 0.08 0.09 0.11 0.55 0.64 0.60 0.29 0.27 0.38 Wastewater Quality 0.07 0.11 0.21 0.46 0.64 0.88 0.17 0.26 0.38 0.07 0.10 0.15 Stormwater Management 0.07 0.11 0.18 0.06 0.08 0.10 0.17 0.23 0.32 0.53 0.70 0.91 Resource Depletion 0.13 0.24 0.44 0.35 0.58 0.89 0.22 0.34 0.59 0.06 0.08 0.12 Environmental risk 0.02 0.04 0.07 0.17 0.22 0.31 0.54 0.71 0.92 0.06 0.07 0.09    66  Table  4-7: Evaluation of final global preference weights (Gk) with δ=1  Conventional roof Extensive roof Intensive roof Social concerns Structural design force 0.0182 0.0512 0.1534 0.0063 0.0185 0.0609 0.0025 0.0067 0.0211 Fire safety 0.0003 0.0008 0.0027 0.0011 0.0032 0.0111 0.0025 0.0075 0.0244 Human health 0.0007 0.0019 0.0061 0.0047 0.0113 0.0292 0.0047 0.0113 0.0292 Durability 0.0002 0.0006 0.0022 0.0002 0.0010 0.0034 0.0015 0.0041 0.0130 Flexibility  0.0905 0.2674 0.6945 0.0420 0.1210 0.3299 0.0121 0.0328 0.0843 Economic limitation Initial cost 0.0186 0.0570 0.1586 0.0081 0.0245 0.0721 0.0022 0.0063 0.0173 O & M 0.0075 0.0212 0.0590 0.0031 0.0087 0.0258 0.0008 0.0021 0.0059 Replacement cost 0.0057 0.0212 0.0860 0.0359 0.1232 0.4505 0.0034 0.0128 0.0817 Environmental issues Climate Change 0.0004 0.0011 0.0049 0.0026 0.0087 0.0262 0.0014 0.0058 0.0165 Waste Management 0.0065 0.0248 0.1123 0.0024 0.0101 0.0491 0.0010 0.0041 0.0195 Wastewater quality 0.0008 0.0027 0.0112 0.0023 0.0083 0.0348 0.0072 0.0253 0.1004 Storm water Management 0.0094 0.0476 0.2382 0.0061 0.0282 0.1569 0.0016 0.0067 0.0318 Land use 0.0007 0.0027 0.0137 0.0024 0.0087 0.0411 0.0002 0.0009 0.0042 Environmental risk 0.0182 0.0512 0.1534 0.0063 0.0185 0.0609 0.0025 0.0067 0.0211 As shown in Table  4-8, the extensive green roof system is the most sustainable alternative for both α-cut values. Since the extensive green roof has the highest RSI value, it can be considered the best solution. The intensive green roof is not a sustainable alternative in this area. However, it is noticeable that by decreasing the uncertainty and increasing the confidence, the RSI value of the extensive green roof decreases and the RSI value of the conventional roof increases. Table  4-8: Ranking of roofing systems Alternative  δ=1  Rank  δ=0.5  Rank Conventional roof 0.40 2 0.47 2 Extensive green roof 0.58 1 0.54 1 Intensive green roof 0.16 3 0.13 3  4.4 Discussion Developers, building consultants and other stakeholders are under increasing public pressure to take sustainability issues and green building technologies into consideration. Reliable, evidence-based tools are needed to help these and other decision-makers to choose the most sustainable 67  options among competing green technology alternatives. The purpose of this study was to develop a decision-making framework that can aggregate the results of LCA with multi-criteria decision-making under uncertainty and lack of knowledge for green building technologies. A hierarchical structure was developed addressing concerns of decision makers during selection of the most sustainable technology linked with the implementation of sustainability TBL criteria. This framework generates a sustainability score for different alternatives. Such a quantified sustainability score will be useful to evaluate the comparative sustainability level of alternatives and to guide decision makers in complex sustainability dilemmas. The results have been summarized and compared for both conventional LCA and proposed FAHP-LCA. The developed sustainability index represents the overall sustainability level of a particular green technology. Environmental impacts are derived from LCA. The relative weights and quantified comparisons for other TBL criteria are analyzed through the fuzzy approach to identify the most sustainable green technology. The results could be implemented to support decision-making processes, for example in environmental consulting companies that plan to reduce environmental impacts with acceptable economic efficiency and consistency, with specified client or public values. Moreover, the outcomes can be useful for regulators seeking to adopt or advocate and demonstrate preferred green technologies and practices. This framework aids decision makers to analyze the sustainability of different alternatives in a particular problem. Although the model is developed based on roofing system alternatives, it can be extended to other green building technologies and other industries by tuning the model with the appropriate criteria and desired objectives for the new MCDM process. Compared to existing LCA studies, the proposed approach in this paper is able to aggregate the results of LCA into a hierarchy process. The FAHP model is flexible enough to capture vagueness of uncertainty within LCA, as well as to incorporate subjective considerations, level of confidence, and preferences of decision makers. The proposed FAHP-LCA framework is able to reduce the possibility of selecting an inappropriate building technology/alternative among various current technologies. This framework thus provides a more robust and more reliable decision-making method for sustainability assessment problems.  68  Chapter 5 : Conclusions and Future Works The summary and conclusions of the current research are provided in Section  5.1, followed by the limitations that arose during the research in Section  5.2. Finally, the research contribution and suggested future works are presented in Sections  5.3 and  5.4. 5.1 Summary and Conclusions LID practices have been an appropriate response to non-point pollutant management in urban areas. Green roof systems are one of the LID practices that have been designed and implemented by architects, engineers, and building owners in recent years. This study investigated the performance of extensive green roofs in a semi-arid climate. The quality of extensive green roofs was assessed and the potential of improving green roof runoff quality was explored. Results show that the runoff quality of extensive green roofs examined in this thesis are statistically similar to the runoff quality of conventional roofs. In addition, the current research developed a sustainability assessment framework to assess the sustainability of roofing systems. The important characteristics in sustainability assessments of roofing systems were identified. The results of the sustainability assessment framework showed that extensive green roofs are the most sustainable roofing system among conventional roofs and intensive green roof systems. Results of  Chapter 3 prove that extensive green roof impact on runoff quality is the same as conventional roofs. The runoff quality of sixteen different extensive green roofs was compared with four conventional roofs. Nitrate, ammonia, pH, colour, turbidity, ORP, and EC were measured to determine the performance of each roof. Results in  Chapter 3 can be summarized as follows: The results of the experiment showed that there was no significant difference between the pH of green roof runoff and conventional roofs. The average pH of green roofs was slightly higher than conventional roofs and rain. Moreover, the pH of generic green roofs decreases with age and additional rain events. The pH of green roofs was in an acceptable range for Canadian guidelines for both fresh water and domestic reclaimed water. 69  The generic green roof nitrate concentration was significantly lower than the concentration of nitrate in gravel ballasted roofs’ runoff. However the nitrate concentration in generic green roof runoff was statistically the same as the nitrate concentration of EPDM roof runoff. The experiment showed that the concentration of nitrate in generic green roofs was higher than the nitrate concentration in rainwater. The nitrate concentration in green roofs was higher than the accepted concentration for Canadian guidelines for fresh water or domestic reclaimed water. The ammonia concentration of generic green roofs was the same as the concentration of ammonia in gravel ballasted roof runoff and was significantly lower than the ammonia concentration in EPDM roof runoff. The sample analysis showed that the ammonia concentration in generic green roofs was about 90% lower than the ammonia concentration in EPDM roof runoff. Moreover, the concentration of ammonia in green roof runoff was lower than the ammonia concentration in rainwater and was in the accepted range for Canadian guidelines for fresh water. The sample analysis showed that the green roof runoff was coloured. Although the green roof runoff was clear, the colour and turbidity of green roof runoff was not in an acceptable range for fresh water or reclaimed water. Green roof runoff had a lower ORP level than EPDM roof runoff, which shows that the runoff from EPDM roofs had a higher water quality. The ORP of green roof runoff was around 220 to 290 mV and was not in an acceptable range for fresh water guidelines.  The conductivity of green roof runoff was significantly higher than conductivity of EPDM and gravel ballasted roof runoff and rainwater. The conductivity of aged green roof runoff was more constant than the conductivity of other green roof runoff. Since Okanagan Lake is vulnerable to non-point source nitrate and ammonia release, two optimistic and conservative scenarios were defined for retrofitting part of downtown area of Kelowna with XeroFlor extensive green roofs. The results show that by retrofitting just 50% to 75% of that area, the nitrate removal can be estimated to be 300-450kg. In the conservative scenario and retrofitting just 10% to 25% of that area, the nitrate removal would be 60-150kg over the extensive green roof lifespan. Moreover, extensive green roofs were able to remove 200-300kg of ammonia in the optimistic scenario and 40-100kg in the conservative scenario. 70  In  Chapter 4, a sustainability assessment framework was proposed and developed. Sustainability triple-bottom-line (TBL) criteria were considered for assessing the sustainability of the roofing system. TBL criteria consist of economic, environmental, and social criteria. Each TBL criterion was divided into sub-criteria for better assessment. The framework is developed based on the LCA and F-AHP methodology. Three different roofing systems, including an intensive green roof, an extensive green roof, and a gravel ballasted roof were compared. The analysis was based on the current roofing systems constructed on SOE and CIRS buildings at UBC campuses. The environmental impacts of each roofing system were performed using LCA. The LCA results show that extensive green roof system located at SOE has a lower contribution to non-renewable energy consumption, global warming gas production, ozone layer depletion impact, and other environmental impacts. The environmental impact contribution of the intensive green roof was significantly higher than the extensive green roof. The intensive green roof’s environmental impact contribution was lower than the gravel ballasted roof contribution in some categories but higher in other environmental categories. Green roof systems’ initial cost, and operation and maintenance costs are much higher than other conventional roofing systems. This additional cost influences the sustainability of green roof systems. The framework considered the uncertainty in decision making, LCA and cost analysis. The assessment was based on the LCA results, available literature and experts’ judgments. The results show that the SOE’s extensive green roof is the most sustainable roofing system among other roofing systems. 5.2 Limitations There are a variety of physical and chemical water quality characteristics regulated by environmental agencies, but due to the scope of the experiment only primary water quality characteristics were considered. The effect of temperature drops in winter, heavy rainfall during the spring season, and drought situations during summer in this area was not examined. It is noticeable that the current limitation is correlated with the previous limitation and may change the green roof runoff quality. Since it was impossible to distinguish contaminants and emission loads from pollutants in the air or green roof fertilization, the mixed effect of green roofs on runoff quality were measured. Planting species need several years to be established in the new 71  environment due to being under several extreme heat and cold temperatures, drought situations, and the performance of vegetation changes. Therefore, these experiments should be prepared over a long-term period e.g. 5 years or more. Although the proposed framework has various advantages over existing methods, there are some limitations that need to be taken into consideration. The main challenge in this model is to provide a single index for sustainability to embody the overall sustainability level of implementing a green building technology. All criteria and associated sub criteria should be accounted and aggregated in the hierarchy model. Aggregating the results of LCA is the most complex part. Converting categorized LCA impacts into different sub-criteria requires a solid knowledge of environmental assessment. In addition, FAHP and LCA are both time consuming and may prolong the process of decision-making. Indeed developing a web-based FAHP tool can facilitate the application of the proposed framework. 5.3 Research Contributions The current research is a significant contribution to assessing the sustainability of green building technologies based on TBL criteria. There is no other research using LCA and fuzzy assessment to develop a sustainability assessment framework for green buildings to date. The results of the current research on the green roof runoff show that there is no significant difference between the quality of green roof runoff and conventional roof runoff. This result can be used for updating the building regulations and guidelines. This study evaluated the runoff quality of green roof systems for re-use purposes. The results show that green roof runoff meets the Canadian reclaimed water guidelines. 5.4 Future Research There is a need to run the experiment over longer periods (e.g. five years or more) and with different types (e.g. various soil depth). This result can provide better analysis of green roof runoff quality in a semi-arid environment. Running the experiment over longer periods of time provides more accurate results on green roof runoff quality considering the aging depreciation. 72  The experiment should be conducted with different plant species (e.g. local species). Performance of different plants can be assessed and provide a better understanding about the applicable plants for the green roof system. A full life cycle cost analysis of green roofs should be performed considering social and environmental benefits as well. This will provide a better understanding about the range of green roofs benefits/cost and help the policy makers to update the regulation guidelines and possible incentives for green roof systems. During the current research, developing an inventory of green roof materials was a challenge. It is necessary to develop a specific database for green roof systems with detailed information of layers, materials, vegetation, and physical and chemical properties. This database can be used for future simulations and building studies.  73  Bibliography Ali, Hikmat H., and Saba F. Al Nsairat. 2009. “Developing a Green Building Assessment Tool for Developing Countries – Case of Jordan.” Building and Environment 44 (5): 1053–1064. http://www.sciencedirect.com/science/article/pii/S0360132308001868. ALwaer, H., and D.J. Clements-Croome. 2010. “Key Performance Indicators (KPIs) and Priority Setting in Using the Multi-Attribute Approach for Assessing Sustainable Intelligent Buildings.” Building and Environment 45 (4): 799–807. http://www.sciencedirect.com/science/article/pii/S036013230900225X. Aras, Haydar, Şenol Erdoğmuş, and Eylem Koç. 2004. “Multi-Criteria Selection for a Wind Observation Station Location Using Analytic Hierarchy Process.” Renewable Energy. Vol. 29. http://www.sciencedirect.com/science/article/pii/S0960148103004051. Arnold, Chester L., and C. James Gibbons. 1996. “Impervious Surface Coverage: The Emergence of a Key Environmental Indicator.” Journal of the American Planning Association 62 (2) (June 30): 243–258. doi:10.1080/01944369608975688. http://dx.doi.org/10.1080/01944369608975688. ASTM. 2013a. “Standard Guide for Selection, Installation, and Maintenance of Plants for Green Roof Systems.” ASTM. 2013b. “Standard practice for determination of dead loads and live loads associated with vegetative (green) roof systems.” ASTM. 2013c. “Standard test method for water capture and media retention of geocomposite drain layers for vegetative (green) roof systems.” ASTM. 2013d. “Standard test method for maximum media density for dead load analysis ofvegetative (green) roof systems.” ASTM. 2013e. “Standard classification for size of aggregate used as ballast for membrane roof systemsB.C. Air quality. 2013. “The Earth’s Climate System.” http://www.bcairquality.ca/climate-change/what-is-climate-change.html. Bass, Brad, Scott Krayenhoff, Alberto Martilli, and Roland Stull. 2002. “Mitigating the Urban Heat Island with Green Roof Infrastructure.” Green Roofs Infrastructure Monitor 4 (1). http://www.5dstudios.com/clients/gcca/wp-content/uploads/2012/04/finalpaper_bass.pdf. Berndtsson, Justyna Czemiel. 2010. “Green Roof Performance towards Management of Runoff Water Quantity and Quality: A Review.” Ecological Engineering 36 (4): 351–360. http://www.sciencedirect.com/science/article/pii/S0925857410000029. Berndtsson, Justyna Czemiel, Lars Bengtsson, and Kenji Jinno. 2009. “Runoff Water Quality from Intensive and Extensive Vegetated Roofs.” Ecological Engineering 35 (3): 369–380. http://www.sciencedirect.com/science/article/pii/S0925857408002024. 74  Berndtsson, Justyna Czemiel, Tobias Emilsson, and Lars Bengtsson. 2006. “The Influence of Extensive Vegetated Roofs on Runoff Water Quality.” Science of The Total Environment 355 (1): 48–63. http://www.sciencedirect.com/science/article/pii/S0048969705001713. Bianchini, Fabricio, and Kasun Hewage. 2012a. “How ‘green’ Are the Green Roofs? Lifecycle Analysis of Green Roof Materials.” Building and Environment 48: 57–65. http://www.sciencedirect.com/science/article/pii/S0360132311002629. Bianchini, F., & Hewage, K. 2012b. “Probabilistic social cost-benefit analysis for green roofs: a lifecycle approach.” Building and Environment 58: 152–162. Blackhurst, Michael, Chris Hendrickson, and H. Scott Matthews. 2010. “Cost-Effectiveness of Green Roofs.” Journal of Architectural Engineering 16 (4) (December 5): 136–143. doi:10.1061/(ASCE)AE.1943-5568.0000022. http://ascelibrary.org/doi/abs/10.1061/(ASCE)AE.1943-5568.0000022. Bottero, Marta, Elena Comino, and Vincenzo Riggio. 2011. “Application of the Analytic Hierarchy Process and the Analytic Network Process for the Assessment of Different Wastewater Treatment Systems.” Environmental Modelling & Software 26 (10): 1211–1224. http://www.sciencedirect.com/science/article/pii/S1364815211001009. Boulanger, Bryan, and Nikolaos P. Nikolaidis. 2003. “Mobility and Aquatic Toxicity of Copper in an Urban Watershed.” Journal of the American Water Resources Association 39 (2) (April): 325–336. doi:10.1111/j.1752-1688.2003.tb04387.x. http://doi.wiley.com/10.1111/j.1752-1688.2003.tb04387.x. Brezonik, Patrick L, and Teresa H Stadelmann. 2002. “Analysis and Predictive Models of Stormwater Runoff Volumes, Loads, and Pollutant Concentrations from Watersheds in the Twin Cities Metropolitan Area, Minnesota, USA.” Water Research 36 (7): 1743–1757. http://www.sciencedirect.com/science/article/pii/S004313540100375X. Brunner, Norbert, and Markus Starkl. 2004. “Decision Aid Systems for Evaluating Sustainability: A Critical Survey.” Environmental Impact Assessment Review 24 (4): 441–469. http://www.sciencedirect.com/science/article/pii/S0195925503002063. Canada Climate. 2013. “Daily Data Report for Kelowna, British Columbia, Canada.” http://climate.weather.gc.ca/climateData/dailydata_e.html?StationID=48369. Canada climate change. 2013. “Information on climate change.” . http://www.climatechange.gc.ca/default.asp?lang=En&n=F2DB1FBE-1. (Accessed 12/09/2013). Carter, T., and A. Keeler. 2008. “Life-Cycle Cost–benefit Analysis of Extensive Vegetated Roof Systems.” Journal of Environmental Management 87 (3): 350–363. Castleton, H.F., V. Stovin, S.B.M. Beck, and J.B. Davison. 2010. “Green Roofs; Building Energy Savings and the Potential for Retrofit.” Energy and Buildings 42 (10): 1582–1591. http://www.sciencedirect.com/science/article/pii/S0378778810001453. 75  Chan, A.L.S., and T.T. Chow. 2013. “Energy and Economic Performance of Green Roof System under Future Climatic Conditions in Hong Kong.” Energy and Buildings 64: 182–198. http://www.sciencedirect.com/science/article/pii/S0378778813002910. Chan, Felix, and Niraj Kumar. 2007. “Global Supplier Development Considering Risk Factors Using Fuzzy Extended AHP-Based Approach.” Omega 35 (4): 417–431. http://www.sciencedirect.com/science/article/pii/S030504830500112X. Chan, Hing Kai, and Xiaojun Wang. 2013. Fuzzy Hierarchical Model for Risk Assessment. London: Springer London. doi:10.1007/978-1-4471-5043-5. http://link.springer.com/10.1007/978-1-4471-5043-5. Chan, Hing Kai, Xiaojun Wang, Gareth Reginald Terence White, and Nick Yip. 2013. “An Extended Fuzzy-AHP Approach for the Evaluation of Green Product Designs.” IEEE Transactions on Engineering Management 60 (2) (May): 327–339. doi:10.1109/TEM.2012.2196704. http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=6204329. Chatzimouratidis, Athanasios I., and Petros A. Pilavachi. 2009. “Technological, Economic and Sustainability Evaluation of Power Plants Using the Analytic Hierarchy Process.” Energy Policy 37 (3): 778–787. http://www.sciencedirect.com/science/article/pii/S0301421508005880. Chen, Shan-Huo. 1985. “Ranking Fuzzy Numbers with Maximizing Set and Minimizing Set.” Fuzzy Sets and Systems 17 (2): 113–129. http://www.sciencedirect.com/science/article/pii/0165011485900508. Chung, Eun Sung, and Kil Seong Lee. 2009. “Prioritization of Water Management for Sustainability Using Hydrologic Simulation Model and Multicriteria Decision Making Techniques.” Journal of Environmental Management 90 (3): 1502–1511. http://www.sciencedirect.com/science/article/pii/S0301479708003022. City of Toronto. 2010. “Green Roofs - Zoning & Environmental Planning - City Planning | City of Toronto.” http://www1.toronto.ca/wps/portal/contentonly?vgnextoid=3a7a036318061410VgnVCM10000071d60f89RCRD. Coffelt, Donald P., and Chris T. Hendrickson. 2010. “Life-Cycle Costs of Commercial Roof Systems.” Journal of Architectural Engineering 16 (1) (March 12): 29–36. doi:10.1061/(ASCE)1076-0431(2010)16:1(29). http://ascelibrary.org/doi/abs/10.1061/%28ASCE%291076-0431%282010%2916%3A1%2829%29. Dabaghian, M. R., S. H. Hashemi, T. Ebadi, and R. Maknoon. 2008. “The Best Available Technology for Small Electroplating Plants Applying Analytical Hierarchy Process.” International Journal of Environmental Science & Technology 5 (4) (September 1): 479–484. doi:10.1007/BF03326044. http://link.springer.com/10.1007/BF03326044. DeNardo, J., A. Jarrett, H. Manbeck, D. Beattie, and R. Berghage. 2003. “Stormwater Detention and Retention Abilities of Green Roofs.” In World Water & Environmental Resources Congress, 1–7. ASCE. doi:10.1061/40685(2003)310. http://ascelibrary.org/doi/abs/10.1061/40685(2003)310. 76  Desjarlais, A.O., T.W. Petrie, and J.A Atchley. 2008. “Evaluating the Energy Performance of Ballasted Roof Systems.” http://www.spri.org/pdf/Thermal Performance of Ballast Study Final Report 05 08 .pdf. Dietz, Michael E. 2007. “Low Impact Development Practices: A Review of Current Research and Recommendations for Future Directions.” Water, Air, and Soil Pollution 186 (1-4) (September 5): 351–363. doi:10.1007/s11270-007-9484-z. http://link.springer.com/10.1007/s11270-007-9484-z. Dolowitz, David, Melissa Keeley, and Dale Medearis. 2012. “Stormwater Management: Can We Learn from Others?” Policy Studies 33 (6) (November): 501–521. doi:10.1080/01442872.2012.722289. http://dx.doi.org/10.1080/01442872.2012.722289. Dunnett, Nigel, and Noel Kingsbury. 2008. Planting Green Roofs and Living Walls. Timber Press. http://www.amazon.com/Planting-Green-Roofs-Living-Walls/dp/0881929115. Dunnett, Nigel, Ayako Nagase, Rosemary Booth, and Philip Grime. 2005. “Vegetation Composition and Structure Significantly Influence Green Roof Performance.” In Greening Rooftops for Sustainable Communities, Washington, DC, May 4-6, 2005, 10. Egodawatta, Prasanna, Evan Thomas, and Ashantha Goonetilleke. 2009. “Understanding the Physical Processes of Pollutant Build-up and Wash-off on Roof Surfaces.” Science of The Total Environment 407 (6): 1834–1841. http://www.sciencedirect.com/science/article/pii/S0048969708012916. Ellis, J.B. 2013. “Sustainable Surface Water Management and Green Infrastructure in UK Urban Catchment Planning.” Journal of Environmental Planning and Management 56 (1) (January): 24–41. doi:10.1080/09640568.2011.648752. http://dx.doi.org/10.1080/09640568.2011.648752. Environment Canada. 2003. “Understanding stormwatermanagement:an introduction to stormwater management planning and design.” www.ene.gov.on.ca/stdprodconsume/groups/lr/.../std01_079720.pdf. (Accessed 11/11/2013). Eumorfopoilou, E., and D. Aravantinos. 1998. “The Contribution of a Planted Roof to the Thermal Protection of Buildings in Greece.” Energy and Buildings 27: 26–36. Fioretti, R., A. Palla, L.G. Lanza, and P. Principi. 2010. “Green Roof Energy and Water Related Performance in the Mediterranean Climate.” Building and Environment 45 (8): 1890–1904. http://www.sciencedirect.com/science/article/pii/S0360132310000806. FLL Guidelines. 2002. “Guideline for the Planning, Execution and Upkeep of Green Roof Sites”. Bonn, Germany. http://www.greenroofsouth.co.uk/FLL Guidelines.pdf. Galal, H. S. 2013. “Integrating sustainability in municipal wastewater infrastructure decision-analysis using the analytic hierarchy process”. University of British Columbia. https://circle.ubc.ca/bitstream/handle/2429/44590/ubc_2013_fall_galal_hana.pdf?sequence=5. (Accessed 14/11/2013). Gallopin, G. C., Funtowicz, S., O’Connor, M., & Ravetz, J. 2001. “Science for the twenty-first century: from social contract to the scientific core.” International Social Science Journal 53 (168): 219–229. doi:10.1111/1468-2451.00311. 77  Gerardi, M. H. 2007. “Oxidation reduction portential and wastewater treatment.” http://www.neiwpcc.org/iwr/reductionpotential.asp. (Accessed 11/05/2012). Getter, K. L., & Rowe, D.B. 2006. “The role of extensive green roofs in sustainable development.” HortScience 41 (5): 1276–1285. Gill, S.E., Handley, J.F,. Ennos, A.R., & Pauleit S. 2007. “Adapting cities for climate change: the role of the green infrastructure.” Built Environment 33 (1): 115–133. Graham, P., & Kim, M. 2003. “Evaluating the stormwater management benefits of green roofs through water balance modeling.” In Greening Rooftops for Sustainable Communities Conference. Green roof Guide. 2011. “Green Roof Guide”. Sheffield. http://www.greenroofguide.co.uk/pdfs/. Green roofs for healthy cities. 2013a. “Green roofs statistics.” http://www.greenroofs.org/index.php/about/aboutgreenroofs. (Accesse 02/02/2013). Green roofs for healthy cities. 2013b. “Green roofs for healthy cities.” http://www.greenroofs.org/index.php/about/aboutgreenroofs. (Accesse 02/02/2013). Gregoire, Bruce G., and John C. Clausen. 2011. “Effect of a Modular Extensive Green Roof on Stormwater Runoff and Water Quality.” Ecological Engineering 37 (6): 963–969.  Haimes, Y. 1992. “Sustainable Development: A Holistic Approach to Natural Resource Management.” IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS, 22 (3): 413–417. Harwell, M., C. Harwell, and J. Kelly. 1986. “Regulatory Endpoints, Ecological Uncertainties, and Environmental Decision-Making.” In OCEANS ’86, 993–998. IEEE. doi:10.1109/OCEANS.1986.1160433.  Health Canada. 2010. “Canadian guidelines for domestic reclaimed water.” http://www.hc-sc.gc.ca/ewh-semt/alt_formats/hecs-sesc/pdf/pubs/water-eau/reclaimed_water-eaux_recyclees/reclaimed_water-eaux_recyclees-eng.pdf. (Accessed 07/06/2012). Hobbs, B. F., & Horn, G. T. 1997. “Building public confidence in energy planning: a multimethod MCDM approach to demand-side planning at BC gas.” Energy Policy 25 (3): 357–375. IPCC. 2007. “IPCC fourth assessment report: climate change 2007.” http://www.ipcc.ch/publications_and_data/ar4/wg1/en/ch2s2-10.html. (Accessed 10/10/2013). ISO. 2006. “14044: 2006. Environmental management-life cycle assessment-requirements and guidelines.” http://www.iso.org/iso/catalogue_detail?csnumber=38498. (Accessed 10/10/2012). Jaber, J. O., & Mohsen, M. S. 2001. “Evaluation of Non-Conventional Water Resources Supply in Jordan.” Desalination 136 (1): 83–92. Jaffal, I., Ouldboukhitine, S., & Belarbi, R. 2012. “A comprehensive study of the impact of green roofs on building energy performance.” Renewable Energy 43: 157–164. 78  Jolliet, O. et al. 2003. “IMPACT 2002+: a new life cycle impact assessment methodology.” The International Journal of Life Cycle Assessment 8 (6): 324–330. doi:10.1007/BF02978505. Kahraman, C., Cebeci, U., & Ulukan, Z. 2003. “Multi-criteria supplier selection using fuzzy AHP.” Logistics Information Management 16 (6): 382–394. doi:10.1108/09576050310503367. Kang, Y, & Li, J. 2010. “Green rationality evaluation of degradable packaging based on lca and fuzzy AHP.” In 2010 IEEE 17Th International Conference on Industrial Engineering and Engineering Management, 329–332. IEEE. doi:10.1109/ICIEEM.2010.5646599.  Kellenberger, D., & Althaus, H.J. 2009. “Relevance of simplifications in lca of building components.” Building and Environment 44 (4): 818–825. Kholghi, M. 2001. “Multi-Criterion Decision-Making Tools for Wastewater Planning Management” 3: 281–286. lock, R., & Mullock, J. 2001. “The weather of british columbia.” http://www.navcanada.ca/EN/media/Publications/Local Area Weather Manuals/LAWM-BC-EN.pdf. (Accessed 14/11/2013). Klöpffer, Walter. 2005. “The Role of SETAC in the Development of LCA.” The International Journal of Life Cycle Assessment 11 (S1) (December 6): 116–122. doi:10.1065/lca2006.04.019.  Köhler, M., & Schmidt, M. 2003. “Study of extensive ‘green roofs’ in Berlin. (S. Cacanindin, Trans.). Roofscapes, Inc.: Water Quality Benefits.” Berlin. http://www.roofmeadow.com/water_quality.htm. (Accessed 15/08/2012). Kohler, M., M. Schmidt, F. Grimme, M. Laar, and F. Gusmao. 2001. “Urban Water Retention by Greened Roofs in Temperate and Tropical Climate.” In The 38th IFLA World Congress. Singapore. Kosareo, Lisa, and Robert Ries. 2007. “Comparative Environmental Life Cycle Assessment of Green Roofs.” Building and Environment 42 (7): 2606–2613.  Kruijf, J. de. 2007. “Problem Structuring in Interactive Decision-Making Processes : How Interaction, Problem Perceptions and Knowledge Contribute to a Joint Formulation of a Problem and Solutions.” http://essay.utwente.nl/524/1/scriptie_de_Kruijf.pdf. Lee, Ju Young, Jung-Seok Yang, Mooyoung Han, and Jaeyoung Choi. 2010. “Comparison of the Microbiological and Chemical Characterization of Harvested Rainwater and Reservoir Water as Alternative Water Resources.” The Science of the Total Environment 408 (4) (January 15): 896–905. doi:10.1016/j.scitotenv.2009.11.001. http://www.ncbi.nlm.nih.gov/pubmed/19962177. Lerario, A., and N Maiellaro. 2001. “Support Measures for Sustainable Building. Towards Sustainable Building.” In Towards Sustainable Building, edited by N Maiellaro, 171–200. Kluwer Academic Publishers. Levett, R. 1998. “Sustainability Indicators-Integrating Quality of Life and Environmental Protection.” Journal of the Royal Statistical Society: Series A (Statistics in Society) 161 (3) (October): 291–302. doi:10.1111/1467-985X.00109. http://doi.wiley.com/10.1111/1467-985X.00109. 79  Lindholm, Oddvar, James M. Greatorex, and Adam M. Paruch. 2007. “Comparison of Methods for Calculation of Sustainability Indices for Alternative Sewerage systems—Theoretical and Practical Considerations.” Ecological Indicators 7 (1): 71–78. http://www.sciencedirect.com/science/article/pii/S1470160X05001111. Liu, K., & Baskaran, B. 2003. “Thermal performance of green roofs through field evaluation.” In First North American Green Roof Infrastructure Conference, Awards and Trade Show,, 1–10. Chicago. http://archive.nrc-cnrc.gc.ca/obj/irc/doc/pubs/nrcc46412/nrcc46412.pdf. (Accessed 03/05/2012). Lloyd, Shannon M., and Robert Ries. 2008. “Characterizing, Propagating, and Analyzing Uncertainty in Life-Cycle Assessment: A Survey of Quantitative Approaches.” Journal of Industrial Ecology 11 (1) (October 9): 161–179. doi:10.1162/jiec.2007.1136. http://doi.wiley.com/10.1162/jiec.2007.1136. Long, Brett, Shirley E. Clark, Katherine H. Baker, and Robert Berghage. 2006. “Green Roof Media Selection for the Minimization of Pollutant Loadings in Roof Runoff.” Weftec: 5528–5548. McHarg, Ian L. 1995. Design with Nature (Wiley Series in Sustainable Design). Wiley. http://www.amazon.com/Design-Nature-Wiley-Series-Sustainable/dp/047111460X. Medineckiene, Milena, Zenonas Turskis, and Edmundas Kazimieras Zavadskas. 2010. “Sustainable Construction Taking into Account the Building Impact on the Environment.” Journal of Environmental Engineering and Landscape Management 18 (2) (June): 118–127. doi:10.3846/jeelm.2010.14. Mendez, Carolina B., J. Brandon Klenzendorf, Brigit R. Afshar, Mark T. Simmons, Michael E. Barrett, Kerry A. Kinney, and Mary Jo Kirisits. 2011. “The Effect of Roofing Material on the Quality of Harvested Rainwater.” Water Research 45 (5): 2049–2059.  Mofarrah, Abdullah, Tahir Husain, and Kelly Hawboldt. 2013. “Decision Making for Produced Water Management: An Integrated Multi-Criteria Approach.” International Journal of Environmental Technology and Management (IJETM) 16 (1/2). Montgomery, Douglas C. 2008. Design and Analysis of Experiments. John Wiley & Sons, Inc. Moran, A., Hunt, B., & Jennings, G. 2004. “A north carolina field study to evaluate green roof runoff quantity, runoff quality, and plant growth.” http://www.bae.ncsu.edu/greenroofs/GRHC2004paper.pdf. (Accessed 07/05/2012). Muga, Helen E., and James R. Mihelcic. 2008. “Sustainability of Wastewater Treatment Technologies.” Journal of Environmental Management 88 (3): 437–447.  Nelms, Cheryl E., Alan D. Russell, and Barbara J. Lence. 2007. “Assessing the Performance of Sustainable Technologies: A Framework and Its Application.” Building Research & Information 35 (3) (May): 237–251. doi:10.1080/09613210601058139.  Newsham, G.R., S. Mancini, and B. Birt. 2009. “Do LEED-Certified Buildings Save Energy? Yes, But...” Energy and Buildings 41 (8): 897–905. 80  Nkwonta, Onyeka, and George Ochieng. 2009. “Roughing Filter for Water Pre-Treatment Technology in Developing Countries: A Review.” International Journal of Physical Sciences 4 (9): 455–463. OECD. 2001. “OECD environmental indicators towards sustainable development”. Paris, France. http://www.oecd.org/site/worldforum/33703867.pdf. (Accessed 16/09/2012). Oke, TR. 1995. “The heat island of the urban boundary layer: characteristics, causes and effects.” http://www.citeulike.org/group/15109/article/9351714. (Accessed 03/02/2013). Onmura, S, M Matsumoto, and S Hokoi. 2001. “Study on Evaporative Cooling Effect of Roof Lawn Gardens.” Energy and Buildings 33 (7): 653–666.  Osmundson, Theodore H. 1999. Roof Gardens: History, Design, and Construction. Ney York: W. W. Norton & Company. Ostendorf, Bertram, Maria Luisa Paracchini, Cesare Pacini, M. Laurence M. Jones, and Marta Pérez-Soba. 2011. “An Aggregation Framework to Link Indicators Associated with Multifunctional Land Use to the Stakeholder Evaluation of Policy Options.” Ecological Indicators 11 (1): 71–80.  Palme, Ulrika, Margareta Lundin, Anne-Marie Tillman, and Sverker Molander. 2005. “Sustainable Development Indicators for Wastewater Systems – Researchers and Indicator Users in a Co-Operative Case Study.” Resources, Conservation and Recycling 43 (3): 293–311.  Peck, S., & Kuhn, M. 2001. “Design guidelines for green roof. Ontario association of architects.” http://www.cmhc.ca/en/inpr/bude/himu/coedar/loader.cfm?url=/commonspot/security/getfile.cfm&PageID=70146. (Accessed 02/04/2012). Peri, Giorgia, Marzia Traverso, Matthias Finkbeiner, and Gianfranco Rizzo. 2012. “Embedding ‘substrate’ in Environmental Assessment of Green Roofs Life Cycle: Evidences from an Application to the Whole Chain in a Mediterranean Site.” Journal of Cleaner Production 35: 274–287.  Peuportier, B.L.P. 2001. “Life Cycle Assessment Applied to the Comparative Evaluation of Single Family Houses in the French Context.” Energy and Buildings 33 (5): 443–450.  Pilavachi, Petros A., Stilianos D. Stephanidis, Vasilios A. Pappas, and Naim H. Afgan. 2009. “Multi-Criteria Evaluation of Hydrogen and Natural Gas Fuelled Power Plant Technologies.” Applied Thermal Engineering 29 (11): 2228–2234.  Pompeii, W. C. 2010. “Assessing urban heat island mitigation using green roofs: a hardware scale modeling approach”. Shippensburg University. http://www.ship.edu/uploadedFiles/Ship/Geo-ESS/Graduate/Theses/pompeii_thesis_100419.pdf. (Accessed 24/02/2013). Poulenard, Jérôme, Pascal Podwojewski, Jean-Louis Janeau, and Jean Collinet. 2001. “Runoff and Soil Erosion under Rainfall Simulation of Andisols from the Ecuadorian Páramo: Effect of Tillage and Burning.” CATENA 45 (3) (September): 185–207. doi:10.1016/S0341-8162(01)00148-5.  Rajendran, Sathyanarayanan, John A. Gambatese, and Michael G. Behm. 2009. “Impact of Green Building Design and Construction on Worker Safety and Health.” Journal of Construction Engineering and Management 135 (10): 1058–1066.  81  RCABC. 2011. “Consumer Guide to Roofing a Guide for the Selection of Roofing Services.” Rebitzer, G., T. Ekvall, R. Frischknecht, D. Hunkeler, G. Norris, T. Rydberg, W.-P. Schmidt, S. Suh, B.P. Weidema, and D.W. Pennington. 2004. “Life Cycle Assessment.” Environment International 30 (5): 701–720. http://www.sciencedirect.com/science/article/pii/S0160412003002459. Reza, Bahareh. 2013. “Emergy-Based Life Cycle Assessment (em-Lca) for Sustainability Appraisal of Built Environment.” University of British Columbia. Reza, Bahareh, Rehan Sadiq, and Kasun Hewage. 2011. “Sustainability Assessment of Flooring Systems in the City of Tehran: An AHP-Based Life Cycle Analysis.” Construction and Building Materials 25 (4): 2053–2066. http://www.sciencedirect.com/science/article/pii/S0950061810005714. Robertson, G.P., and J.M. tiedje. 1987. “Nitrous Oxide Sources in Aerobic Soils: Nitrification, Denitrification and Other Biological Processes.” Soil Biology and Biochemistry 19 (2) (January): 187–193. doi:10.1016/0038-0717(87)90080-0. http://www.sciencedirect.com/science/article/pii/0038071787900800. Rosén, N. 2009. “Evaluation methods for procurement of business critical software systems”. Institutionen för kommunikation och information. http://www.diva-portal.org/smash/get/diva2:222953/FULLTEXT01.pdf. (Accessed 04/07/2012). Rosenfeld, Arthur H., Hashem Akbari, Joseph J. Romm, and Melvin Pomerantz. 1998. “Cool Communities: Strategies for Heat Island Mitigation and Smog Reduction.” Energy and Buildings 28 (1): 51–62. Rosenzweig, C., G. Stuart, and P Lily. 2006. “Green Roofs in the New York Metropolitan Region”. New York. Rowe, D. Bradley. 2011. “Green Roofs as a Means of Pollution Abatement.” Environmental Pollution 159 (8): 2100–2110. http://www.sciencedirect.com/science/article/pii/S0269749110004859. Ryan B. et al. 2008. “USEPA 2008. Reducing Urban Heat Islands: Compendium of Strategies.” Saaty, TL. 1980. The Analytic Hierarchy Process. New York: McGraw-Hill. Saiz, Susana, Christopher Kennedy, Brad Bass, and Kim Pressnail. 2006. “Comparative Life Cycle Assessment of Standard and Green Roofs.” Environmental Science and Technology 40 (13): 4312–4316. doi:DOI: 10.1021/es0517522. Santamouris, M., C. Pavlou, P. Doukas, G. Mihalakakou, A. Synnefa, A. Hatzibiros, and P. Patargias. 2007. “Investigating and Analysing the Energy and Environmental Performance of an Experimental Green Roof System Installed in a Nursery School Building in Athens, Greece.” Energy 32 (9): 1781–1788.  Sarkis, Joseph, and Srinivas Talluri. 2002. “A Model for Strategic Supplier Selection.” The Journal of Supply Chain Management 38 (1) (December): 18–28. doi:10.1111/j.1745-493X.2002.tb00117.x.  82  Schilling, J. 2010. “Towards a greener green space planning.” http://www.lumes.lu.se/database/alumni/08.10/thesis/schilling_jasper_thesis_2010.pdf. (Accessed 08/07/2013). Shaviv, Avi. 2001. “Advances in Controlled-Release Fertilizers.” Advances in Agronomy 71: 1–49.  Shepherd, M.F., S. Barzetti, and D.R. Hastie. 1991. “The Production of Atmospheric NOx and N2O from a Fertilized Agricultural Soil.” Atmospheric Environment 25a (9): 1961–1969. Sonne, J. K. 2006. “Energy performance aspects of a florida green roof.” In Fifteenth Symposium on Improving Building Systems in Hot and Humid Climates. Orlando. http://www.fsec.ucf.edu/en/publications/html/FSEC-PF-412-06/. (Accessed 03/05/2012). Steen, Bengt. 1997. “On Uncertainty and Sensitivity of LCA-Based Priority Setting.” Journal of Cleaner Production 5 (4): 255–262. Suslow, T.V. 2007. “Oxidation-Reduction Potential (ORP) for Water Disinfection Monitoring, Control, and Documentation.” ANR Publication. Davis. http://anrcatalog.ucdavis.edu/pdf/8149.pdf. Sutton, R. K. et al. 2012. “Prairie-based green roofs: literature, templates, and analogs.” Journal of Green Building 7 (1) (January 16): 143–172. doi:10.3992/jgb.7.1.143. Taylor, Brian. L. 2008. “The Stormwater Control Potential of Green Roofs in Seattle.” In International Low Impact Development Conference. Seattle. Teemusk, Alar, and Ülo Mander. 2007. “Rainwater Runoff Quantity and Quality Performance from a Greenroof: The Effects of Short-Term Events.” Ecological Engineering 30 (3): 271–277. http://www.sciencedirect.com/science/article/pii/S0925857407000134. Tesfamariam, S., & Sadiq, R. 2006. “Risk-based environmental decision-making using fuzzy analytic hierarchy process (F-AHP).” Stochastic Environmental Residents Risk Assessment 21: 35–50. doi:10.1007/s00477-006-0042-9. The Weather Network. 2013. “Kelowna precipitation statistics.” http://www.theweathernetwork.com/forecasts/statistics/precipitation/cl11239r0. (Accessed 22/09/2013). Tupenaite et al. 2010. “Multiple criteria assessment of alternatives for built and human environment renovation.” Journal of Civil Engineering and Management 16 (2): 257–266. doi:10.3846/jcem.2010.30. UNDPCSD. 1995. “Work programme on indicators for sustainable development.” UNI EN 832. USEPA. 1999. “Storm water technology fact sheet sand filters.” http://water.epa.gov/scitech/wastetech/upload/2002_06_28_mtb_sandfltr.pdf. USEPA. 2004. “Guidelines for water reuse. Washington, D.C. EPA/625/R-04/108”. Washington. http://water.epa.gov/aboutow/owm/upload/Water-Reuse-Guidelines-625r04108.pdf. 83  USEPA. 2007. “Methods for analyses and properties. Chapter 3.” USEPA. 2009a. “National water quality inventory: 2004 Report. EPA-841-R-08-001.” USEPA. 2009b. “Green roofs for stormwater runoff control.” www.epa.gov/ord. Vijayaraghavan, K, U M Joshi, and R Balasubramanian. 2012. “A Field Study to Evaluate Runoff Quality from Green Roofs.” Water Research 46 (4) (March 15): 1337–45. doi:10.1016/j.watres.2011.12.050.  Waheed, Bushra, Faisal Khan, and Brian Veitch. 2009. “Linkage-Based Frameworks for Sustainability Assessment: Making a Case for Driving Force-Pressure-State-Exposure-Effect-Action (DPSEEA) Frameworks.” Sustainability 1 (3) (August 10): 441–463. doi:10.3390/su1030441.  Wang, Jiang-Jiang, You-Yin Jing, Chun-Fa Zhang, and Jun-Hong Zhao. 2009. “Review on Multi-Criteria Decision Analysis Aid in Sustainable Energy Decision-Making.” Renewable and Sustainable Energy Reviews 13 (9): 2263–2278.  Wang, Ranran, Matthew J. Eckelman, and Julie B. Zimmerman. 2013. “Consequential Environmental and Economic Life Cycle Assessment of Green and Gray Stormwater Infrastructures for Combined Sewer Systems.” Environmnetal Science and Technology 47: 11189–11198. doi:dx.doi.org/10.1021/es4026547. Wedding, G. Christopher, and Douglas Crawford-Brown. 2007. “Measuring Site-Level Success in Brownfield Redevelopments: A Focus on Sustainability and Green Building.” Journal of Environmental Management 85 (2): 483–495.  Xeroflor America. 2013. “Xeroflor America.” http://www.xeroflora.com/specs-tech/technical-documents. (Accessed 05/08/2012). Yaziz, M.I., H. Gunting, N. Sapari, and A.W. Ghazali. 1989. “Variations in Rainwater Quality from Roof Catchments.” Water Research 23 (6): 761–765. http://www.sciencedirect.com/science/article/pii/004313548990211X. Yok, T.P., and A Sia. 2005. “A Pilot Green Roof Research Project in Singapore.” In Green Roofs for Healthy Sustainable Cities Conference. Washington. Yoon, So Won, and Dong Kun Lee. 2003. “The Development of the Evaluation Model of Climate Changes and Air Pollution for Sustainability of Cities in Korea.” Landscape and Urban Planning 63 (3): 145–160. Zadeh, L.A. 1965. “Fuzzy Sets.” Information and Control 8 (3): 338–353.  Zaman, M., M.L. Nguyen, J.D. Blennerhassett, and B.F. Quin. 2008. “Reducing NH3, N2O and NO3-N Losses from a Pasture Soil with Urease or Nitrification Inhibitors and Elemental S-Amended Nitrogenous Fertilizers.” Biology and Fertility of Soils 44: 693–705. Zheng, Guozhong, Youyin Jing, Hongxia Huang, and Yuefen Gao. 2011. “Applying LCA and Fuzzy AHP to Evaluate Building Energy Conservation.” Civil Engineering and Environmental Systems 28 (2) (June): 123–141. doi:10.1080/10286608.2010.482655.  84  Zhu, Ke-Jun, Yu Jing, and Da-Yong Chang. 1999. “A Discussion on Extent Analysis Method and Applications of Fuzzy AHP.” European Journal of Operational Research 116 (2): 450–456.  Zimmerman, M.J., Waldron, M.C., Barbaro, J.R., & Sorenson, J.R. 2010. “Effects of low-impact-development (LID) practices on streamflow, runoff quantity, and runoff quality in the Ipswich River Basin, Massachusetts: a summary of field and modeling studies.” U.S. Department of the Interior and U.S. Geological Survey.  85  Appendices Appendix A: Impact Category Description IMPACT 2002+ method considers nine impact categories:  Carcinogens  Respiratory Inorganics  Respiratory Organics  Ozone Layer Depletion  Land Occupation  Aquatic Acidification  Aquatic Eutrophication  Global Warming Potential  Non-Renewable Energy Consumption These impacts are explained as follows: Carcinogens (kg C2H3Cl eq) Carcinogenic materials are materials that may cause adverse health effects on the human body. Carcinogenic materials are emitted during different chemical activities. Complex production processes may produce higher amounts of known carcinogenic materials. Carcinogenic materials are calculated based on the kg C2H3Cl equivalent. Respiratory Inorganics (kg P.M 2.5 eq) Respiratory inorganics have an adverse impact on human health. These materials may cause or amplify human respiratory diseases (e.g. asthma, bronchitis, acute pulmonary disease, etc.). Respiratory Inorganics are calculated based on the kg P.M 2.5 equivalent. Respiratory Organics (kg C2H4 eq) Respiratory organics have an adverse impact on human health. Respiratory organics are calculated based on the kg C2H4 equivalent.   86  Ozone Layer Depletion (kg CFC-11 eq) Emission of ozone depleting substances causes the protective effect of the ozone layer within the stratosphere to diminish, which is called ozone layer depletion. CFCs, HFCs, and halons are ozone depleting substances. The ozone depletion potential is indicated based on kg of equivalent CFC-11. Aquatic Acidification (kg SO2 eq) Aquatic acidification is a regional impact that influences human health. High concentrations of NOx and SO2 cause adverse human health issues. Aquatic acidification is calculated based on the kg SO2 equivalent. Aquatic Eutrophication (kg PO4P-lim) Aquatic eutrophication is a result of adding limited or rare nutrients to a water body. Due to the additional nutrients, aquatic plants grow rapidly and may consume the soluble oxygen. Aquatic eutrophication causes various environmental impacts ranging from odors to the death of fish. Aquatic eutrophication is calculated based on the equivalent kg PO4P-lim. Global Warming Potential (kg CO2 eq) Global warming potential (GWP) is a reference measure for expressing the global warming potential of an activity in CO2 equivalent. In this category, carbon dioxide is the reference standard for GWP and all other greenhouse gases (GHGs) are referred to as having a “CO2 equivalence effect”. Since the reactivity or stability of gases may change over time, GWP has a time horizon. Since GHG emissions are mostly by products of a combustion function, some materials emit GHGs during the processing of a raw material. Non-Renewable Energy Consumption Non-renewable energy consumption is an important indicator for environmental impacts. As non-renewable energy production takes millions of years, the consumption of these sources of energy should be controlled and managed. Processing raw materials consumes a large amount of non-renewable energy. In contrast, additional insulation saves energy for heating and cooling the building, which may reduce the non-renewable energy consumption. 87  Appendix B: Xero flor XF301 Vegetated mat green roof system specifications (Xeroflor America 2013) Part I – General  1.1 Summary  It is intended as a guideline for materials function and assembly instruction. The green roof materials assembly is subject to modification as needed for each specific project.  1.2 Definitions  A. Root Barrier: A flexible, synthetic polymer layer installed below the green roof system that serves as protection against root encroachment into underlying roof components.  B. Drainage Mat: A composite geotextile that creates a free flowing space below the vegetated and retention fleece layers to permit unrestricted movement of excess water to roof drains.  C. Retention Fleece: A non-woven fabric layer to serve as filter fabric against particle erosion and to retain supplemental water for root uptake and plant use. A lightweight fleece is part of the pre-cultivated XF301 vegetation mat (see definition below). One or two additional fleece layer(s) may be used in the green roof system assembly for enhanced water holding capacity.  D. Pre-cultivated Vegetation Mat: An integrated unit of plant material, growing medium, and natural fiber or geotextile carrier. Pre-cultivated mats are harvested fully vegetated from the production field and delivered to the installation site as flat or rolled sheets.  E. Growing Medium: A low-organic / high-mineral composition growing mix composed of composted organic matter and lightweight porous aggregate.  1.3 Deliveries, Storage, and Handling of Material  Xero Flor plant materials shall be delivered in such a manner to preserve the quality of the plants. Truck delivery must protect the vegetation mats from temperature or wind damage during transport, such as use of plant-compatible tarp covers. Closed or open trailers may be used for transport times less than one day. For longer duration transport times, vegetation mats must be delivered in a climate controlled trailer. Upon arrival, the mats shall be immediately off-loaded, 88  plastic wrap removed (if used), and installed within twelve hours. If timely installation is not achievable, then a holding area shall be reserved to unroll and store the mats until installation.  1.4 Vegetation Coverage Guarantee  A. Xero Flor mats shall be delivered with a minimum of 80% vegetation coverage at the time of installation and achieve a minimum of 90% coverage after the second full growing season.   PART II - PRODUCTS  2.1 XF112 Root Barrier  A. A flexible polymer sheet installed on top of the roof membrane and below the other green roof components. The standard Xero Flor XF112 root barrier is a water-impermeable sheet of 20mil low density polyethylene (LDPE), though may be increased to 30mil (XF113) or 40mil (XF114) thickness as specified by the membrane supplier and/or project designer.  2.2 XF108H Drain Mat  A layer of flexible, non-woven, entangled polymeric filaments with a perforated, geotextile filter-fabric bonded to one side.  2.3 XF157 Water-Retention Fleece  A fabric produced from a blend of recycled, synthetic fibers with a saturated weight of not more than 1.5 psf.  2.4 XF301 Pre-cultivated Vegetation Mat  XF301 is a textile-based vegetation carrier of lightweight fleece sown to PA/PP entanglements bonded to geotextile fabric filled with a thin-layer of growing medium and pre-cultivated with an even layer of low-profile, drought-tolerant vegetation. Mat thickness 1 1/4”, field weight 5.5 psf, saturated weight 8.5 psf. 2.5 XeroTerr Growing Medium  89  A proprietary mixture of lightweight, mineral based materials; including porous aggregate and organic matter derived from composted plant materials, biosolids, and/or manure compost.  2.6 Hose Bib / Water Supply  A. A spigot source or other means of supplying water to the roof with sufficient pressure is required. Irrigation must be applied during the plant recovery phase, e.g. first 1-2 weeks, after installation. In order to support mature establishment of the vegetated community, it is highly recommended that periodic irrigation be applied during the hottest months of the 1st and possibly 2nd growing seasons after installation. The method of supplying irrigation may vary with regard to removable or permanent piping, rotary heads, drip irrigation, or other approved irrigation technologies.  PART III - EXECUTION  3.1 General  All green roof system components, including irrigation if specified, are to be installed by certified contractors with demonstrated experience and project references. The various layers shall be installed in such a manner as to not damage or disturb any previously installed roofing components. Installing the system in any manner inconsistent with manufacturer guidelines voids all guarantees and warranties.    90  Appendix C: Sampling and analysis of waters, wastewaters, soils and wastes  Selection and preparing water samples, and test procedures should comply with this appendix based on the USEPA sampling guide AS/NZS 5667.1:1998, USEPA SW8468. The recommended volumes are for a single sample and volume of sampling may varied based on the analytical method. All containers should be clean and free from relevant contamination. Table C.1: USEPA sampling process Analytical parameter Container Typical volume (mL) Sampling and transport Preservation Maximum holding time  Storage  Ammonia Polyethylene, PTFE or glass 500 Transport under ice Filter sample on site (0.45 μm cellulose acetate membrane filter). Acidify with sulfuric acid to pH < 2, or freeze upon receipt by laboratory Analyse within 24 hours Up to 28 days acceptable Refrigerate (< 6°C) Refrigerate (< 6°C) if acidifying, otherwise freeze (- 20 ºC) Colour Polyethylene, PTFE or glass 500 Transport under ice, in dark  48 hours Refrigerate (< 6°C) in dark. Electrical Conductivity Polyethylene or glass 500 Fill container completely to exclude air. Transport under ice   24 hours 28 days if refrigerated Refrigerate (< 6°C)                                                  8 www.epa.gov/epawaste/hazard/testmethods/sw846/online/index.htm#table 91  Table C.1: USEPA sampling process (continue) Nitrate (NO3-) Polyethylene, PTFE or glass 500 Transport under ice Filter on site (0.45 μm cellulose acetate membrane filter) and freeze sample immediately upon collection. Acidify with HCl to pH <2 48 hours without Acidification 7 days with acidification 28 days if frozen Refrigerate (< 6°C) Freeze (-20 oC) Oxygen, dissolved (DO) Glass BOD bottle with top 300 Exclude air from bottle and seal.  Analyse immediately on site (in situ)  pH Polyethylene, PTFE or borosilicate glass 100 Fill bottle to exclude air. Transport under ice  Determine in situ if possible, or upon arrival to laboratory. Analyse immediatelyTurbidity Polyethylene, PTFE or glass 100 Transport under ice, in dark  Up to 48 hours Refrigerate (< 6°C) in dark.      92  Appendix D: The experiments’ results Table D.1: GR1sample results GR 1 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.29 2.52 37.8 243.3 0.0173 450 10/15/2012 7.05 6.6 48.7 253 0.0156 380 10/23/2012 7.8 4.27 121 232.5 0.0179 479 10/27/2012 7.53 5.52 1.93 228 0.0143 460 10/29/2012 7.63 5.74 28.8 245.6 0.0148 510 10/31/2012 7.5 2.03 1.26 233.9 0.0152 486 11/3/2012 7.31 0.83 0.69 237.8 0.0165 396 11/6/2012 7.22 0.44 3.24 359 0.0167 427 11/12/2012 7.34 1.34 2.96 332.2 0.0179 490 11/17/2012 7.45 0.89 1.49 310.1 0.0134 534  Table D.2: GR2 sample results GR 2 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.15 1.78 10.5 254 0.0148 390 10/15/2012 7.31 1.44 5.59 362 0.0164 410 10/23/2012 7.6 32.3 89 234 0.0146 375 10/27/2012 7.49 11.9  1.79 217 0.0169 359 10/29/2012 7.54 8.3 15 241 0.0187 510 10/31/2012 7.44 1.62 3.24 244.2 0.0261 428 11/3/2012 7.31 0.83 0.8 236.4 7.88E-03 476 11/6/2012 7.24 0.52 4.7 340.3 9.80E-03 390 11/12/2012 7.37 1.46 1.39 320 0.0145 452 11/17/2012 7.29 1.01 2.1 328.4 0.0112 415     93  Table D.3: GR+CF1 sample results GR + CF 1 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.21 2.56 17.3 232.4 0.0165 450 10/15/2012 7.27 1.85 5.58 276.5 0.0156 438 10/23/2012 7.45 4.53 112 224.3 0.0143 520 10/27/2012 7.51 15.6 24.4 212.3 0.0165 497 10/29/2012 7.79 9.76 20 219.2 0.0154 416 10/31/2012 6.6 2.28 5.32 269.9 0.0123 529 11/3/2012 6.67 0.92 3.9 262.3 0.011 480 11/6/2012 6.98 0.76 2.4 266.4 0.0121 420 11/12/2012 7.01 1.56 4.2 298.3 0.0149 478 11/17/2012 6.98 1.21 3.82 267.6 0.0156 450  Table D.4: GR+CF2 sample results GR + CF 2 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.1 2.67 9.01 213.4 0.0179 437 10/15/2012 7.26 1.67 9.51 232.3 0.0145 457 10/23/2012 7.39 4.65 76.3 243 0.0139 429 10/27/2012 7.59 4.05 3.37 224.5 0.0156 490 10/29/2012 7.93 2.27 1.43 225.1 0.054 426 10/31/2012 7.16 2.41 3.16 287.2 0.0126 480 11/3/2012 7.07 0.72 2.06 253.5 0.0189 435 11/6/2012 7.08 2.03 4.5 262.4 0.0179 410 11/12/2012 7.12 3.2 2.54 295.4 0.0138 510 11/17/2012 7.39 4.1 3.2 289.3 0.0167 446     94   Table D.5: GR+T1 sample results GR + T 1 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.03 1.48 13.2 232.3 0.0168 460 10/15/2012 7.14 1.83 6.94 223.3 0.0178 426 10/23/2012 7.43 4.12 35 254.3 0.0145 392 10/27/2012 7.32 3.59 3.45 231.6 0.0137 436 10/29/2012 7.21 11.9 32.3 255.4 0.0167 479 10/31/2012 7.12 5.43 1.66 289.7 0.0173 569 11/3/2012 7.07 1.21 0.88 234.8 0.0792 460 11/6/2012 6.97 1.55 3.2 308.5 0.0254 425 11/12/2012 7.01 2.45 1.4 278.2 0.0198 451 11/17/2012 7.29 3.21 2.15 276.9 0.0187 406  Table D.6: GR+T2 sample results GR + T 2 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.11 2.45 9.66 245.3 0.0234 435 10/15/2012 7.17 2.24 1.65 217.3 0.0198 417 10/23/2012 7.35 2.32 18.3 229.4 0.0189 437 10/27/2012 7.37 2.28 2.09 225.4 0.0201 459 10/29/2012 7.48 2.86 0.95 231.2 0.0167 524 10/31/2012 6.99 3.94 3.77 235.6 0.117 478 11/3/2012 6.61 0.79 0.88 339.4 0.0384 453 11/6/2012 7.09 0.91 2.3 322.3 0.0287 397 11/12/2012 7.03 2.17 1.45 265.7 0.0211 417 11/17/2012 7.11 3.35 3.2 267.3 0.0196 436    95  Table D.7: GR+WB1 sample results GR + WB 1 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.11 3.35 360 243.2 0.0187 401 10/15/2012 7.28 2.29 421 254.3 0.0165 453 10/23/2012 6.73 3.47 122 234.2 0.0134 426 10/27/2012 7.04 5.44 18 256.3 0.0156 478 10/29/2012 7.08 7.78 234 249.3 0.0143 423 10/31/2012 7.32 2.47 1.46 261.8 0.0138 491 11/3/2012 7.29 0.46 0.86 361.8 9.73E-03 423 11/6/2012 7.28 0.55 3.4 339.3 0.0112 486 11/12/2012 7.39 2.23 2.25 301.2 0.0132 418 11/17/2012 7.27 3.2 1.76 256.4 0.0123 469  Table D.8: GR+WB2 results GR + WB 2 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 6.98 4.13 597 234.4 0.0237 432 10/15/2012 6.89 6.27 553 267.3 0.0176 478 10/23/2012 6.96 5.43 321 287.3 0.0167 457 10/27/2012 6.89 7.86 18.9 256.4 0.0145 510 10/29/2012 7.03 10.3 46.1 277 0.0198 453 10/31/2012 7.41 2.15 1.4 260.9 0.0219 428 11/3/2012 7.26 0.48 0.78 360.2 0.0432 392 11/6/2012 7.13 0.57 1.32 307.9 0.0324 459 11/12/2012 7.11 1.98 2.21 289.3 0.0201 481 11/17/2012 7.21 2.34 1.03 267.4 0.0176 415     96  Table D.9: EPDM1 results EPDM 1 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 6.78 1.92 17.6 278.6 0.145 32 10/15/2012 7.09 1.69 7.57 287.4 0.198 46 10/23/2012 7.21 1.67 78.3 289.4 0.154 51 10/27/2012 7.28 2.86 3.38 298.4 0.143 35 10/29/2012 6.89 5.97 29.4 296.6 0.121 21 10/31/2012 7.44 2.07 0.99 276.6 0.101 38 11/3/2012 7.54 0.71 0.55 335 0.171 49 11/6/2012 7.32 0.6 3.23 341.9 0.165 42 11/12/2012 7.22 1.24 2.45 321.3 0.154 38 11/17/2012 7.24 1.45 7.56 345.3 0.176 41  Table D.10: EPDM2 results EPDM 2 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 6.4 9.59 21.6 276.4 0.176 27 10/15/2012 7.08 1.4 5.91 289.4 0.189 51 10/23/2012 6.68 5.43 8.68 269.3 0.252 39 10/27/2012 6.68 5.43 8.68 298.3 0.179 37 10/29/2012 6.88 7.91 57.9 289.1 0.156 61 10/31/2012 7.24 2.79 10.2 256.1 0.196 31 11/3/2012 7.25 0.6 1.7 443.7 0.48 27 11/6/2012 7.16 0.59 3.45 390.4 0.346 29 11/12/2012 7.23 2.11 2.45 387.7 0.265 24 11/17/2012 7.11 2.32 3.24 339.7 0.238 35     97  Table D.11: GR+TSG1 results GR + TSG 1 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 6.81 5.52 624 237.4 0.0256 410 10/15/2012 6.87 1.53 413 223.8 0.0239 398 10/23/2012 6.91 2.18 210 235.3 0.0231 453 10/27/2012 6.96 8.78 40.5 264.9 0.0256 476 10/29/2012 7.23 17.2 407 279.1 0.0245 481 10/31/2012 7.28 11.8 12.6 207.3 0.0251 497 11/3/2012 7.41 5.09 1.17 305.8 0.0219 512 11/6/2012 6.99 4.61 5.43 319.9 0.0189 462 11/12/2012 7.12 3.45 3.46 276.4 0.0216 431 11/17/2012 7.18 8.45 8.49 259.4 0.0179 478  Table D.12: GR+TSG2 results GR + TSG 2 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 6.86 5.88 536 275.4 0.0823 436 10/15/2012 6.89 2.64 748 267.3 0.0694 474 10/23/2012 6.98 3.45 456 287.3 0.0498 483 10/27/2012 7.36 3.47 353 298.3 0.0985 513 10/29/2012 7.81 2.68 924 269.2 0.0768 451 10/31/2012 6.96 8.08 19.6 350.1 0.0996 438 11/3/2012 7.25 2.67 0.87 313.2 0.0808 419 11/6/2012 7.02 4.61 4.23 308 0.0876 378 11/12/2012 7.13 3.24 2.19 289.3 0.0675 425 11/17/2012 7.25 4.25 3.56 321.6 0.0587 463     98  Table D.13: GB1 results GB 1 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 6.37 18.4 83.2 276.5 0.0148 156 10/15/2012 6.77 2.05 28.7 258.4 0.0174 183 10/23/2012 6.87 3.86 864 278.4 0.0165 167 10/27/2012 7.23 5.1 96.4 267.5 0.0157 139 10/29/2012 6.77 18.4 69.6 278.2 0.0138 187 10/31/2012 7.64 2.44 10.1 287 7.91E-03 231 11/3/2012 7.59 0.33 10.5 419.6 0.022 169 11/6/2012 7.39 0.66 9.87 342 0.0198 156 11/12/2012 7.32 3.76 5.64 295.4 0.0211 187 11/17/2012 7.42 4.32 8.74 342.4 0.0176 163  Table D.14: GB2 results GB 2 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 6.35 23.3 114 256.3 0.0139 139 10/15/2012 6.72 2.07 94.7 276.6 0.0157 167 10/23/2012 6.83 2.52 674 256.3 0.0164 183 10/27/2012 7.09 4.6 118 256.4 0.0147 148 10/29/2012 6.98 5.39 136 267.9 0.0176 192 10/31/2012 7.4 1.95 38.4 347 9.21E-03 171 11/3/2012 7.35 0.41 18 397.8 0.0154 189 11/6/2012 7.1 0.6 14.3 378.8 0.0178 139 11/12/2012 7.12 3.27 12.3 365.3 0.0156 179 11/17/2012 7.28 2.45 8.43 329.3 0.0167 156     99  Table D.15: GR+S1 results GR + S 1 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.12 6.3 145 278.4 0.0159 462 10/15/2012 6.91 3.03 288 265.3 0.0173 418 10/23/2012 7.32 21.3 347 248.6 0.0168 439 10/27/2012 7.36 13.3 144 239.5 0.0186 458 10/29/2012 7.79 63.9 504 257.8 0.0158 498 10/31/2012 7.33 2.25 2.61 263.3 7.93E-03 532 11/3/2012 7.2 0.46 0.92 340.5 0.0215 437 11/6/2012 7.1 0.63 6.54 351.6 0.0231 427 11/12/2012 7.17 3.48 4.72 299.3 0.0198 437 11/17/2012 7.43 2.51 3.58 278.4 0.0201 451  Table D.16: GR+S2 results GR + S 2 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 6.71 8.64 157 256.3 0.0278 451 10/15/2012 7.03 2.11 62.7 276.6 0.0173 489 10/23/2012 6.98 17.3 154 267.3 0.0213 389 10/27/2012 6.9 13.1 11.8 239.4 0.0289 410 10/29/2012 7.79 63.9 504 257.8 0.0132 436 10/31/2012 7.16 2.36 1.4 241.7 0.0109 427 11/3/2012 7.2 0.34 1.19 348.7 0.11 419 11/6/2012 7.25 0.93 3.25 328.8 0.0764 498 11/12/2012 7.26 3.48 4.35 301.2 0.0219 482 11/17/2012 7.34 2.52 3.42 278.4 0.0208 467     100  Table D.17: Acc.Age GRa1 results Acc. Age GRa 1 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.56 4.32 2.64 2.87 0.0426 410 10/15/2012 7.29 6.57 3.25 267.3 0.0328 405 10/23/2012 7.46 7.63 2.54 256.4 0.0635 378 10/27/2012 7.41 8.73 2.17 298.4 0.0763 429 10/29/2012 7.36 5.64 2.96 278.4 0.0328 439 10/31/2012 7.37 2.65 1.4 263.8 0.0514 418 11/3/2012 7.37 2.65 1.4 263.8 0.0514 498 11/6/2012 7.27 0.53 1.02 335.9 0.0432 421 11/12/2012 7.17 2.23 1.32 293.4 0.0379 426 11/17/2012 7.28 3.21 2.13 325.4 0.0521 411  Table D.18: Acc.Age GRa2 results Acc. Age GRa 2 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.39 3.45 3.25 276.4 0.783 408 10/15/2012 7.27 5.41 4.53 258.4 0.0489 419 10/23/2012 7.31 9.32 3.68 279.7 0.0543 432 10/27/2012 7.27 6.59 2.48 289.1 0.0732 437 10/29/2012 7.23 7.64 3.48 269.5 0.0683 448 10/31/2012 7.02 2.47 17.6 226.3 0.115 436 11/3/2012 7.03 0.42 0.83 353 0.082 481 11/6/2012 7.31 0.44 1.23 329.5 0.0382 439 11/12/2012 7.21 2.39 3.75 325.8 0.0452 417 11/17/2012 7.32 3.27 2.54 319.4 0.0421 409     101   Table D.19: Acc.Age GRb1 results Acc. Age GRb 2 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.34 5.38 2.62 251.9 0.0361 421 10/15/2012 7.24 6.31 3.19 287.7 0.0621 431 10/23/2012 7.28 9.87 3.29 276.4 0.0483 423 10/27/2012 7.38 6.49 4.27 239.4 0.0584 451 10/29/2012 7.17 6.37 2.28 257.6 0.0637 428 10/31/2012 7.26 2.47 1.3 227.8 0.0581 452 11/3/2012 7.17 0.38 0.84 358 0.0269 412 11/6/2012 7.32 0.63 1.49 325.3 0.0342 398 11/12/2012 7.15 2.21 2.21 301.4 0.0427 432 11/17/2012 7.16 1.93 2.69 284.3 0.0379 416  Table D.20: Acc.Age GRb2 results Acc. Age GRb 2 Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.41 5.49 4.39 256.3 0.0341 417 10/15/2012 7.32 6.29 4.28 245.7 0.0247 438 10/23/2012 7.22 8.95 3.17 298.4 0.0337 431 10/27/2012 7.31 6.94 3.79 279.4 0.038 491 10/29/2012 7.27 7.82 6.37 269.4 0.0278 418 10/31/2012 7.2 2.51 6.41 244.5 0.0475 452 11/3/2012 7.04 0.3 0.72 351 0.0437 463 11/6/2012 7.29 0.67 2.39 333.3 0.0453 431 11/12/2012 7.26 2.63 5.38 279.6 0.0564 423 11/17/2012 7.13 2.31 3.28 295.3 0.0348 401   102     Rain Date pH Nitrate Turbidity ORP Ammonia Conductivity 10/13/2012 7.11 0.74 0.28 337.9 0.0875 21 10/15/2012 7.04 0.71 0.31 321.2 0.0764 25 10/23/2012 7.17 0.68 0.35 365.4 0.0969 41 10/27/2012 7.21 0.79 0.27 347.5 0.0824 16 10/29/2012 7.01 0.87 0.32 343.8 0.0921 34 10/31/2012 7.92 0.72 0.29 331.6 0.0977 15 11/3/2012 7.38 0.41 0.36 346.2 0.0981 52 11/6/2012 7.3 0.56 0.25 345.2 0.0895 41 11/12/2012 7.18 0.76 0.32 376.3 0.0899 31 11/17/2012 7.28 0.65 0.38 374.1 0.0956 11  103  Appendix E: The current experiment’s pictures   Figure E.1: Snap shot of the experiment pilot  AppendiThe ANOwere idenAs it canx F: ANOVVA assumptical. FFig be seen, theA assumpttions are evigure F.1: Cure F.2: Ch samples areions validataluated forhecking theecking the n almost lineion  nitrate and  normality aormality asar. Thereforammonia. Tssumption fsumption foe, the normhe results for nitrate r ammonia ality assumpor other sam  tion is satis104 ples fied.   Figure FFigure F..3: Checkin4: Checkingg the indep the indepenendence residence residduals assumuals assump ption for ni tion for ammtrate onia 105  The Figuindependalmost thres F.3 and ency of same same variaFigurFigureF.4 show thapling is satince. e F.5: Chec F.6: Checkt there is nosfied. Figureking the coning the const any relatios F.5 and Fstant variantant variancn between t.6 show thatce assumptie assumptiohe samples  each series  on for nitrat n for ammoand the of samplinge nia 106  has 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0074339/manifest

Comment

Related Items