Material quantities in building structures and their

Material quantities in building structures and their environmental impact

by

Catherine De Wolf

B.Sc., M.Sc. in Civil Architectural Engineering Vrije Universiteit Brussel and Université Libre de Bruxelles, 2012

Submitted to the Department of Architecture

in Partial Fulfillment of the Requirements for the Degree of

Master of Science in Building Technology

at the

Massachusetts Institute of Technology

June 2014

© 2014 Massachusetts Institute of Technology. All rights reserved.

Signature of Author: Department of Architecture

May 9, 2014

Certified by: John A. Ochsendorf

Professor of Architecture and Civil and Environmental Engineering Thesis Supervisor

Accepted by: Takehiko Nagakura

Associate Professor of Design and Computation Chair of the Department Committee on Graduate Students

John E. Fernández

Professor of Architecture, Building Technology, and Engineering Systems Head, Building Technology Program

Co-director, International Design Center, MIT Thesis Reader

Frances Yang

Structures and Sustainability Specialist at Arup Thesis Reader

“It is […] important to remember that unlike operational carbon emissions the embodied

carbon cannot be reversed”

Craig Jones, Circular Ecology

Material quantities in building structures and their environmental impact by Catherine De Wolf

Submitted to the Department of Architecture in Partial Fulfillment of the Requirements for the Degree of Master of Science in Building Technology on May 9, 2014.

Thesis Supervisor: John Ochsendorf Title Supervisor: Professor of Architecture and Civil and Environmental Engineering

Abstract Improved operational energy efficiency has increased the percentage of embodied energy in the total life cycle of building structures. Despite a growing interest in this field, practitioners lack a comprehensive survey of material quantities and embodied carbon in building structures. This thesis answers the key question: “What is the embodied carbon of different structures?” Three primary techniques are used: (1) a review of existing tools and literature; (2) a collaboration with a worldwide network of design firms through conversations with experts and (3) the creation of a growing interactive database containing the material efficiency and embodied carbon of thousands of buildings. The first contribution of this thesis is to define challenges and opportunities in estimating greenhouse gas emissions of structures, expressed in carbon dioxide equivalent (CO2e). Two key variables are analyzed: material quantities (kgmaterial/m2 or kgm/m2) and Embodied Carbon Coefficients (ECC, expressed in kgCO2e/kgm). The main challenges consist of creating incentives for sharing data, identifying accurate ECCs and resolving transparency while protecting intellectual ownership. The main opportunities include using Building Information Models to generate data, proposing regional ECCs and outlining a unified carbon assessment method.

The second contribution is the development of an interactive online tool, called deQo (database of embodied Quantity outputs), to provide reliable data about the Global Warming Potential of buildings (GWP, measured in kgCO2e/m2 and obtained by multiplying the two key variables). Given the need for a long-term initiative, a framework is offered to create an interactive, growing online database allowing architects, engineers and researchers to input and compare their projects. The third contribution is the survey of 200 existing buildings obtained through deQo. Two general conclusions result from this survey of building structures: material quantities typically range from 500 to 1500 kg/m2 and the GWP typically ranges between 200 and 700 kgCO2e/m2. Conclusions from this survey include that healthcare buildings use more materials whereas office buildings have a lower impact. Additionally, specific case studies on stadia, bridges and skyscrapers demonstrate that the design approach can have a significant impact on the embodied carbon of building structures. Ultimately, this thesis enables benchmarking of the environmental impact of building structures. Key words: Embodied Carbon/Energy; Life Cycle Analysis; Database; Materials; Structures

Acknowledgments First, I am very grateful to my advisor, Prof. John Ochsendorf for his perception, guidance and enthusiasm during the development of this research and his support and friendship during my Master studies at MIT. I am also thankful to my readers, Prof. John Fernandez and Frances Yang (Arup), for their constructive suggestions during the development of this thesis. Their willingness to give their time so generously has been very much appreciated. My special gratitude goes to Frances Yang, Andrea Charlson and Kristian Steele (Arup) for supervising my visiting research at Arup and offering me the opportunities to collaborate with practitioners worldwide on the topic of this thesis. Also, I am especially thankful to Wolfgang Werner (Thornton Tomasetti) whose collaboration was essential for the development of the database. The thoughtful contributions of these talented engineers brought new perspectives to my thesis. For their help in the work towards this thesis, I am thankful in particular to the following researchers and friends: Ornella Iuorio for her study of the literature of embodied carbon coefficients; Mayce El Mostafa and Virginie Arnaud for their help in calculating the embodied carbon of stadia and tall office buildings; Julia Hogroian for her work on the material quantities in stadia; Eleanor Pence for her help coding the interactive database; Caitlin Mueller for her valuable feedback and finally the Fall 2013 students in the “Building structural systems II” class for their projects on skyscrapers. Furthermore, I am very thankful to Kathleen Ross who always supported me and helped me out with administrative tasks and travels. My highest appreciation goes to John Ochsendorf and Frances Yang for giving me the chance to discuss my work with world-class experts in structural engineering. The conversations with leading experts were inspirational and brought more depth to this thesis. I am honored to have met Edward Allen, Jörg Schlaich, William Baker, David Shook, Jim D’Aloisio, Kate Simonen, Patrick McCafferty, Craig Jones, Roderick Bates, Alice Moncaster, Julian Allwood, the SEAONC and SEI committees and many more. I would like to thank the Arup and Thornton Tomasetti for their assistance with the collection of my data. This research was sponsored by the Belgian American Education Foundation, the KuMIT project, the Harold Horowitz Award and Arup. Finally, I thank my family and friends for always being there for me.

7

Contents

ABSTRACT 5

ACKNOWLEDGMENTS 6

CONTENTS 7

1. INTRODUCTION 9

1.1. Motivation 9 1.2. Definitions of concepts 10 1.3. Problem statement 12 1.4. Organization of thesis 13

2. STATE OF THE ART 15 2.1. Material quantities 16 2.2. Embodied Carbon Coefficients (ECCs) 17 2.3. Examples of existing implementations 19 2.4. Summary 20

3. METHODOLOGY 23

3.1. Personal conversations with practitioners 23 3.2. Assessment of existing literature and tools 23 3.3. A new interactive database 24 3.4. Summary 25

4. CHALLENGES AND OPPORTUNITIES IN OBTAINING EMBODIED QUANTITIES 27

4.1. Introduction 27 4.2. Getting material quantities 27 4.3. Accurate ECCs 28 4.3.1. Literature on ECCs 28 4.3.2. Applied ranges for ECCs 30 4.4. Implementation of a unified method 32 4.5. Summary 34

5. FRAMEWORK FOR A DATABASE 35

5.1. Introduction 35 5.2. Database specifications 36 5.2.1. General information 36 5.2.2. Structural information 39 5.2.3. Default versus entered ECCs 42 5.2.4. Contribution of a new database 42 5.3. Relational database 44 5.4. Web-based interface 44 5.5. Collaboration with industry 46 5.5.1. Revit plug-in developers 46 5.5.2. Industry and research database 47 5.6. Summary 48

8

6. SURVEY OF EXISTING BUILDINGS 49

6.1. Introduction 49 6.2. Existing building structures from the industry 49 6.3. Case Study I: Analysis of stadia 54 6.3.1. Description 54 6.3.2. Material quantities in stadia 55 6.3.3. Embodied carbon of stadia 56 6.3.4. Discussion of results 57 6.4. Case study II: Lessons from historic bridges 57 6.4.1. Description 58 6.4.2. Material quantities in historic bridges 59 6.4.3. Embodied carbon of historic bridges 60 6.4.4. Comparison Roman arch and Inca suspension bridge 61 6.4.5. Comparison with recent bridge designs 62 6.5. Case study III: Comparing tall buildings 63 6.5.1. Description 63 6.5.2. Material quantities in tall buildings 63 6.5.3. Embodied carbon of tall buildings 64 6.5.4. Discussion of the results 66 6.6. Summary 66

7. CONCLUSIONS 69

7.1. Discussion of results 69 7.2. Summary of contributions 71 7.3. Future research 72

BIBLIOGRAPHY AND REFERENCES 75

Part 1: General bibliography and references 75 Part 2: Stadia references 80 Part 3: Historic bridges references 81 Part 4: Tall building references 82

APPENDICES 85

Appendix A: Nomenclature 85 A.1. List of acronyms 85 A.2. Lexicon 86 Appendix B: Tables 88 B.1. Ten first projects in deQo 88 B.2. Analysis of stadia 89 B.3. Analysis of historic bridges 90 B.4. Analysis of tall buildings 92

9

1. Introduction

1.1. Motivation

Life cycle energy in buildings includes operational energy for heating, cooling, hot water, ventilation, lighting on one hand and embodied energy for material supply, production, transport, construction and disassembly on the other. The terms “embodied carbon” and “Global Warming Potential” (GWP) refer to the equivalent in carbon dioxide of all lifecycle greenhouse gas emissions and is expressed in weight of carbon dioxide equivalents (CO2e). Many leading structural engineering and design firms are currently developing in-house embodied carbon estimators to answer the following question: what is the embodied carbon for different structures? Craig Jones, co-founder of the Inventory of Carbon & Energy (Hammond & Jones, 2010), highlights the importance “to remember that unlike operational carbon emissions the embodied carbon cannot be reversed” (Circular Ecology, 2014). The key question of this thesis examines embodied carbon for several reasons. First, carbon reduction is needed now. As mentioned in the latest Intergovernmental Panel on Climate Change (IPCC) report, substantial carbon reductions need to occur in the near future in order to avoid extreme climate disruptions (IPCC, 2014). Embodied carbon savings in building structures are an obvious opportunity to reduce the impacts in the short term. Next, some buildings have a short lifetime, which results in a high percentage of embodied carbon in the total environmental impact of a building. This is true for a diverse range of buildings, even those most prized and lauded as architectural exemplars at the time of their construction. For example, the Folk Art museum (Figure 1.1), built in 2001, will be torn down for an extension of the MOMA after a lifespan of less than 15 years (New York Times, 2014).

Figure 1.1: Folk Art Museum and MOMA, New York City (New York Times, 2014)

Moreover, with the effective reduction of the operational carbon of buildings in future decades, embodied carbon will become a more significant percentage of greenhouse gas emissions caused by buildings. Furthermore, research in this field will help structural engineers and architects to understand how to lower embodied carbon and fill this gap in literature (Dixit et al., 2012). Finally, rating schemes such as LEED (USGBC, 2013) have begun including embodied carbon in their credit system, though without defining baselines for benchmarking (Yang, 2014).

10

It should be noted that this thesis is limited to structural material quantities. Cladding and other non-structural materials are not considered for two reasons. First, structure accounts for the greatest weight in buildings and contributes to about half of the total carbon emissions due to materials (Webster et al., 2012). With a breakdown of embodied carbon for the different elements in offices, hospitals and schools, Kaethner and Burridge (2012) demonstrate that the super- and substructure together represent more than 50% of the total embodied carbon emissions of buildings (Figure 1.2). Second, this helps to focus attention on a well-defined quantity while still having a significant impact.

Figure 1.2: Average breakdown in building elements of embodied carbon in offices, hospitals and

schools, after (Kaethner and Burridge, 2012)

1.2. Definitions of concepts Several key concepts are used regularly throughout this thesis. Since misunderstandings exist around concepts such as “carbon”, “Embodied Carbon Coefficients” or “carbon equivalent”, this section defines how they are interpreted in this work. Precise definitions can be found in the nomenclature in appendix A. The value for “embodied energy” is not the same as the value for “embodied carbon”. The same amount of embodied energy can emit different intensities of greenhouse gases (GHG) depending on the fuel used and the carbon emitted or absorbed by the materials processed. For example, emissions occur in the chemical processing of cement, whereas carbon is sequestered in wood. It is useful to account in terms of embodied carbon instead of embodied energy, as CO2 contributes considerably to climate change. Also, measuring specifically in carbon helps to compare the embodied with operational emissions. Carbon accounting therefore facilitates the assessment of the whole life cycle impacts of buildings (Kaethner and Burridge, 2012). The total embodied carbon of a building is often referred to as “Global Warming Potential” (GWP). The GWP is expressed in kilograms of carbon dioxide equivalent per functional unit. The carbon dioxide equivalent is noted “CO2e” and represents the equivalent in carbon dioxide of the GHGs produced during the manufacturing and transportation of materials. The other GHGs such as CH4, N2O, SF6, PFC and HFC can hence be converted to CO2 using conversion factors in order to obtain a common unit for the environmental impact (IPCC, 2014), i.e. the “carbon dioxide equivalent”.

71% Structure & Construction 29% Others

8% Finishes

4% Internal planning

4% R

oof 13%

External cladding

16% C

onstruction 13%

Substructure 42%

Superstructure

11

A “functional unit” is a specified metric used to normalize the carbon footprint of buildings, in order to compare ‘apples to apples’. For example, the useable floor area can be the functional unit. The GWP is then measured in kgCO2e/m2. The number of occupants is another example of a functional unit. The GWP is then measured in kgCO2e/occupant. Two key variables are analyzed in this thesis: Structural Material Quantities (SMQ), expressed in kg of material (kgmaterial or kgm) per functional unit (often m2), and Embodied Carbon Coefficients (ECC), expressed in kg of CO2e (kgCO2e) per kg of material (kgm). Presently, there is no clear standard for accurate ECC values and information on SMQ values for buildings is scarce. As illustrated in equation [1], the GWP (kgCO2e/m2) is obtained by multiplying the two key variables: SMQ (kgm/m2) and ECC (kgCO2e/kgm), together.

GWP (kgCO2e/m2) = SMQ (kgm/m2) ! ECC (kgCO2e/kgm) [1] Figure 1.3 illustrates how the GWP of different materials can be added up for building projects to define the embodied carbon.

Figure 1.3: GWP divided into the composing materials of different building projects

In this thesis, “structure” refers to the structure of a building and implies the loadbearing part of something built or constructed. When talking about a pattern, relation or organization, other words, such as “framework”, are used.

0"

1"

2"

3"

4"

5"

6"

7"

Proj

ect A

Proj

ect B

Proj

ect C

Proj

ect D

Proj

ect E

Proj

ect F

Em

bodi

ed C

arbo

n or

GW

P

(kg C

O2e

/m2 )

Reinforcement

Steel

Concrete

12

1.3. Problem statement This thesis aims to build literacy on the embodied carbon for different structures. Data is collected in collaboration with a network of worldwide design firms. The research is conducted in three primary streams:

1) Collect data from recently published literature on Life Cycle Assessments (LCA); 2) Collaborate with design and construction firms to build a useful database of their

projects; 3) Create an interactive growing database where participants can input projects to

include thousands of buildings worldwide. Two necessary variables are introduced in section 1.2: SMQ (kgm/m2 ) conveys the amounts of steel, concrete, wood, etc. and ECC (kgCO2e/kgm) expresses the carbon equivalent of these materials. The former is sought from industry participants; the latter is computed with existing LCA software and current literature. Developing confidence in the numbers for embodied carbon of structures (GWP in kgCO2e/m2) requires a large amount of data. No framework or reliable database exists yet assessing material efficiency and embodied carbon in structures. This lack of information results in buildings using materials in a wasteful way with impunity. However, to be able to rationally compare and assess projects, a wider set of input parameters is needed beyond the two key variables. The first set of input parameters consists of general project information such as the building location, typology and geometry. The second are structural parameters such as material choices and quantities, structural systems, soil types and loads. Outputs include ranges of GWP, illustrated in graphics comparing building projects or categories (Figure 1.3). An intermediate result is better information on material efficiency. This thesis has three main contributions. The first contribution is to define the challenges and opportunities in obtaining the material quantities and estimating the embodied carbon of structural materials. To obtain the key variables SMQ and ECC, many challenges occur in existing tools and literature. However, by listing the challenges and offering possible solutions and opportunities, this thesis clarifies which paths can be followed to reduce the embodied carbon of building structures. The second contribution of this thesis is the development of an interactive online tool to give confidence in the GWP measurement of buildings. A framework is presented for a relational database collecting both the embodied carbon of buildings, and their material quantities. The idea is to create a worldwide online tool, which allows architects, engineers and researchers to input their projects and compare it with others. The third contribution is the survey of two hundred actual buildings taken from the database. Additionally, a new unified embodied carbon assessment method is applied to specific case studies, such as stadia, historic bridges and recent skyscrapers in order to critically analyze and refine the process. Ultimately, this will enable benchmarking and comparisons of the environmental impact of a wide range of projects.

13

1.4. Organization of thesis The following chapter (Chapter 2) discusses the state of the art on material quantities, ECC values and implementing the calculation of the GWP. Chapter 3 elaborates on the methodology. The Chapters 4 to 6 expand on the contributions and results of this thesis. First, Chapter 4 reviews the challenges and opportunities in obtaining material quantities and ECCs. Chapter 5 then develops a framework for an interactive database collecting data on existing building projects. Finally, Chapter 6 surveys 200 actual buildings from industry and applies the defined approach for estimating the embodied carbon of structures to three case studies: sport stadia, historic bridges and tall office buildings. Finally, Chapter 7 discusses the key contributions, takes conclusions and proposes paths for future research.

14

15

2. State of the art

This chapter summarizes the state of the art on material quantities, ECCs and environmental impact analysis tools. Gaps and challenges are highlighted in this chapter and will be addressed in the following chapters. Chapter 2 is divided into three parts. The first two parts discuss the literature on the two key variables: SMQ and ECC. The last part addresses existing tools to implement the calculation of the GWP of buildings. Figure 2.1 shows the wide range of results for embodied energy and carbon, illustrating the need for an agreement on an accurate way to estimate the embodied energy and embodied carbon of buildings. Cole and Kernan (1996) estimated the embodied energy is 4 to 9% of a 50 year life-cycle energy, whereas the Athena Institute of Sustainable Materials (2009) mentions 9 to 12% of a 60 year lifespan and Webster et al (2012) talk about 2 to 22% over 50 year. For the embodied carbon, the numbers vary widely as well, from 11 to 80% depending on the source (Simpson et al., 2010).

Figure 2.1: Wide-ranging results for embodied carbon, after (Simpson et al., 2010)

Cole & Kernan

1996

EE = 4-9% of 50 yr life-cycle energy

Eaton & Amato

1998

EC = 37-43% of 60 yr life-cycle carbon emissions

Build Carbon Neutral

2007


Simon Group et al

2008

EC = up to 80 % of life-cycle carbon emissions

Athena

2009


Arup (Simpson et al)

2010


Webster et al

2012


16

2.1. Material quantities In the 1890s, a tower design competition in London asked for material weight as one of the criteria (Figure 2.2) and in the 1920s, Buckminster Fuller asked: “how much does your house weigh?” (Lynde, 1890; Braham and Hale, 2013). However, material efficiency does not always drive building design today. Indeed, as no framework yet exists to assess the embodied carbon of a structure, many buildings may be consuming materials in a wasteful way with impunity.

Figure 2.2: Particulars of the design competition for the London tower in 1890 (Lynde, 1890).

Recently, other studies have attempted to map material efficiency of tall buildings considering the number of floors and structural systems (Cho et al., 2004; Elnimeiri and Gupta, 2009; Ali and Moon, 2007). Data on material quantities from leading structural design firms such as Arup and Thornton Tomasetti have been collected in the material quantity and the database of embodied Quantity outputs (deQo), developed at the Structural Design Lab within the Building Technology program at MIT (De Wolf and Ochsendorf, 2014; deQo, 2014). Cho et al. (2004) express unit material quantities in volume (m3 per m2) for concrete and in mass (kg per m2) for rebar and steel. In Figure 2.3Figure 2.3, they compare the structural steel quantities for various story heights. As Elnimeiri and Gupta (2009) express it, “good structural engineering revolves around achieving efficiency and minimization of material”.

Figure 2.3: Structural steel quantities by number of stories (Cho et al., 2004)

CTBUH 2004 October 10~13, Seoul, Korea 665

Table 3. Unit direct construction costs for tall composite building structures Unit Cost per Material Unit Cost for Structural Members

Project Material *Won/m2 % Str. Mem. *Won/m2 %

Unit Cost Comparison

Concrete 28,497 15.7 Core + Link 56,223 30.9 Reinf. 32,748 18.0 Column 37,023 20.4

Str. Steel 73,533 40.4 Beam/Girder 39,078 21.5 Deck 15,102 8.3 Outrigger 7,953 4.4

Precast 4,093 2.3 Slab 32,985 18.1 Formwork 27,866 15.3 Foundation 8,557 4.7

B(66F)

Total 181,839 100 Total 181,839 100

104.6

Concrete 28,189 15.9. Core + Link 41,958 23.7 Reinf. 20,361 11.5 Column 34,997 19.3

Str. Steel 88,953 50.3 Beam/Girder 57,351 32.4 Deck 14,667 8.3 Outrigger 5,653 3.2

Formwork 24,740 14.0 Slab 34,048 19.2 Foundation 2,905 1.6

E(55F)

Total 176,911 100 Total 176,911 100

101.8

Concrete 34,674 19.9 Core + Link 48,995 28.2 Reinf. 41,284 23.7 Column 36,476 21.0

Str. Steel 53,785 30.9 Beam/Girder 33,871 19.5 Deck 11,759 6.8 Belt Wall 13,973 8.0

Formwork 32,360 18.6 Slab 31,941 18.4 Foundation 8,605 4.9

G(69F)

Total 173,862 100 Total 173,862 100

100.0

* $1 1,200 Won

In order to compare the construction costs of three buildings, the structural quantities of Table 2 are transferred to direct construction costs for building structures based on the price as of June, 2004. Formwork cost and different costs for different concrete strengths are also taken into account in the process. Table 3 shows that the tallest building, G, has the smallest cost of 173,862 won/m2, while B and E have 181,839 and

176,911 won/m2, respectively, giving differences of 4.6 and 1.8 %. The tallest G costs even less than the fourteen story smaller E. The G utilized relatively more quantities of concrete and reinforcement with smaller structural steel amount, which appears to be the reason for the cost saving. This fact shows the importance of selected structural materials for the economical construction of tall buildings.

Fig. 1. Unit structural steel quantities4)

Formwork costs occupy 14 to 18.6 % among total structural expenses; horizontal structural members,

beams and slab, occupy 29.6, 51.6 and 37.9 % for B, E and G, respectively. The high value of E occurs due

17

Fazlur Khan (1980) introduced the notion of “premium for height” (Figure 2.4). With a growing number of stories, the weight of steel in kg/m2 is not only increasing due to columns and bracing walls, but also due to the increasing lateral wind loads. Ali and Moon (2007) and Rizk (2010) give weights of steel in pounds per square feet (psf) versus the number of floors for steel framed tall buildings. The Council on Tall Buildings and Urban Habitat (CTBUH) points out it is often difficult to compare these material quantities, as some studies may include or exclude the foundations, mezzanine floors, etc. (CTBUH Journal, 2010).

Figure 2.4: Weights of steel per numbers of stories, after (Khan, 1980; Ali and Moon, 2007)

The scope of current studies tends to be narrowed down to a specific type of structure (e.g. skyscrapers) or program (e.g. offices) (Amato and Eaton, 1998). Literature is lacking about the ranges of material weights in typical buildings.

2.2. Embodied Carbon Coefficients (ECCs) Similar uncertainty and gaps in data also characterize the literature that examines Embodied Energy and Embodied Carbon. Moncaster and Symons (2013) describe the general lack of data in the field of embodied carbon. Alcorn (1996) discusses the “Embodied Energy Coefficients” (EEC, in MJ/kgmaterial) of building materials, while Dias and Pooliyadda (2004) define “Embodied Carbon Coefficients” (ECC, in kgCO2e/kgmaterial). The phrase “Carbon Intensity Factors” (CIF) is sometimes used interchangeably for the “Embodied Carbon Coefficients” (ECC). Various reports have analyzed the environmental impact of concrete (Vares and Häkkinen,1998; Lagerblad, 2005; Collins, 2010; Struble and Godfrey, 1999; NRMCA, 2014), as well as the impact of cement (Young, Turnbul and Russel, 2002). Other articles describe the embodied energy of metals (Chapman and Roberts, 1983) and in particular steel (ISSF, 2013; IISI, 2013; Eurofer, 2000; Stubbles, 2007; World Steel, 2014). Next to concrete and steel, discussions exist on the embodied energy and carbon of other construction materials such as timber (Pullen, 2000; Corrim, 2014). However, there is a significant variability in the results for both EEC and ECC values.

Stru

ctur

e co

st

per

unit

flo

or a

rea

Extra cost for lateral resistance

Columns and bracing walls

Floor framing

Number of stories A

vera

ge w

eigh

t of

stee

l (kg

/m2 )

350

300

250

200

150

100

50 0 10 20 30 40 50 60 70 80 90 100

Premium for height

320 kg/m2

120 kg/m2

18

There is a need for material manufacturers, such as the concrete, steel and timber industry, to make the ECC data of their products more transparent and consistent (The Concrete Centre, 2013; Moynihan et al., 2012; Weight, 2011). However, each manufacturer puts forth a set of assumptions, such as which life cycle stages to include. For example, taking into account the carbon sequestration may be beneficial for the ECC value of timber, while taking into account the end of life recycled content may be beneficial for the ECC value of steel (Weight, 2011). Efforts exist to summarize the ECCs of common construction materials. In the United Kingdom, the Inventory of Carbon and Energy (ICE) report from the University of Bath is one of the most complete open source databases for ECCs (Hammond and Jones, 2010). Figure 2.5 illustrates the variability of the available data on the embodied energy of steel. The wide range of available EEC/ECC values undermines useful comparisons of the environmental impact of different buildings. The ICE report selects the best available embodied energy and carbon data. However, there is still a need for values per country or region. The same concrete mix used in a big city in China or a small town in the United States will not have the same coefficients if factors such as transport are included (Ochsendorf, et al., 2011). In addition, the embodied carbon in the ICE report is often underestimated by considering only the initial stage (‘cradle to gate’) instead of the whole life cycle (‘cradle to grave’ or ‘cradle to cradle’). The life cycle stages are defined in the TC350 European standards (Moncaster and Symons, 2013). In the Unites States, the Carbon Working Group discusses the embodied carbon of common construction materials. However, the group highlights the uncertainty and variability of available data quality (Webster et al., 2012).

Figure 2.5: Variability of available embodied energy data of Steel (Hammond and Jones, 2010)

Several Life Cycle Inventory (LCI) and LCA tools exist to calculate impacts of single projects or materials. The commercial software Gabi, EcoInvent and SimaPro are common practice for LCA calculations, but can lack transparency due to protection of intellectual property (GaBi, 2013; EcoInvent, 2013; SimaPro, 2013). Furthermore, the Athena Institute of Sustainable Materials has integrated LCI data at the building scale (Athena, 2009).

Year of data

Em

bodi

ed E

nerg

y (E

E)

(MJ/

kg) 90

80

70

60

50

40

30

20

10

0 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

EE Scatter Graph - Steel

19

2.3. Examples of existing implementations The following paragraphs illustrate several existing databases and tools for estimating embodied carbon. The Athena Institute is a non-profit organization based in Canada that has integrated LCI data into building industry specific tools: the Athena Eco Calculator (free) and the Athena Impact Estimator (Athena, 2009). Also, EcoInvent provides thousands of LCI datasets in various fields from agriculture to electronics (EcoInvent, 2013) Furthermore, various companies have developed in-house tools focused on estimating the embodied carbon of their projects. In 2013, Kieran Timberlake and PE International released the Tally environmental impact tool (Tally, 2013) extracting data from Revit models (Revit, 2014). The SOM Environmental Analysis tool is a user-friendly embodied carbon calculator for design projects (SOM, 2013). In the United Kingdom, the non-profit Waste Reduction Action Program (WRAP) recently launched a project-based database of embodied carbon (WRAP, 2014). WRAP asks users of the web-interface to clearly mark building life cycle stages and to reference the LCA software used, without asking specifically for material quantities (Charlson, 2012). The Royal Institution of Chartered Surveyors (RICS) has started developing a range of benchmarks based on the available data on embodied carbon in buildings (RICS, 2014). Many leading structural engineering firms have also started the development of in-house databases of structural material quantities or embodied carbon for their own projects. One thoroughly developed example is the Arup Project Embodied Carbon and Energy Database (PECD) mainly consisting of Arup buildings or projects from literature (PECD, 2013; Kaethner and Burridge, 2012; Yang, 2013). Although PECD contains approximately 600 projects, the data scarcity, scattering and wide ranges of data still hinders the definition of a baseline. Other companies, such as Thornton Tomasetti, have also developed a database of the material quantities, extracted via a Revit plug-in, and the embodied carbon in their projects. Studies demonstrate the need to inform not only designers and clients, but also governments and the public of the need to develop solutions for carbon savings (Clark, 2013). Table 2.1 gives a brief overview of companies and institutes working on similar problems. As can be discerned from this table, the existing initiatives only look at one or two of the parameters to estimate the embodied carbon of buildings. This thesis aims to bridge the gap between the various influences.

20

Mat

eria

l Qua

ntiti

es

EC

C

Impl

emen

tatio

n

Reports Inventory of Carbon & Energy (ICE) ✓ Structure and Carbon (Carbon working group) ✓ Cole and Kernan, 1996 ✓ Eaton and Amato, 1998 ✓ Council of Tall Buildings and Urban Habitat (CTBUH) ✓ Software Athena ✓ ✓ GaBi, SimaPro ✓ SOM Environmental Analysis Tool ✓ ✓ Tally beta tool ✓ ✓ Build Carbon Neutral, 2007 ✓ Databases Arup PECD ✓ ✓ Thornton Tomasetti ✓ ✓ WRAP ✓ ✓

Table 2.1: Leading efforts in material quantity collection, ECC values, database implementation

2.4. Summary Despite a growing interest in this field, designers and clients still lack a global embodied carbon estimator (De Wolf et al., 2014). Studies have attempted to map material efficiency of tall buildings considering the number of floors and structural systems (Khan, 1980; Cho et al., 2004; Ali and Moon, 2007; Elnimeiri and Gupta, 2009). Leading structural design firms such as Arup and Thornton Tomasetti collected data on material quantities in their projects in-house. Other studies are assembling data on ECCs, such as the ICE database of the University of Bath in the United Kingdom (Hammond and Jones, 2010) or the Carbon Working Group in the United States of America (Webster et al., 2013). There is a need for material manufacturers, such as the concrete, steel and timber industry, to make the ECC data of their products more transparent and consistent. Several LCI/LCA software exist on the material and product scale, such as GaBi (GaBi, 2013), EcoInvent (EcoInvent, 2013) and SimaPro (SimaPro, 2013), or on the building scale, such as the Athena Institute of Sustainable Materials (Athena, 2009). The WRAP embodied carbon

21

database collects total carbon footprints of buildings, without asking for the material quantities in the projects still (De Wolf and Ochsendorf, 2014; WRAP, 2014). The RICS hopes to develop new benchmarks for embodied carbon based on these new data-points (RICS, 2014). Overall, estimating the embodied carbon of building structures is data-intensive. However, the available data on material quantities or ECCs are scarce and often unreliable. A general lack of transparency hinders the usage ECC values available in the literature. Also, no baseline has been defined for benchmarking material efficiency and embodied carbon in buildings. This thesis will address these issues, by looking at the opportunities within the challenges. A framework for an interactive database a survey of several hundreds of existing buildings will follow.

22

23

3. Methodology The study of publicly available carbon data in Chapter 2 (Figure 2.1) illustrates the variability in assumptions from different organizations. The existing literature is synthesized and the published numbers on both key variables – the material quantities and the ECCs – are compared in order to advise a unified method of estimating the GWP of buildings. The methodology for estimating material quantities and the environmental impact of building structures is threefold. First, one-on-one conversations with leading structural engineering and design firms leads to a description of the current common practice and existing challenges in estimating embodied carbon of their building projects. Further, existing literature and tools for estimating the environmental impact of buildings are reviewed. Finally, the implementation of an interactive interface and a relational database will unify the collection and processing of data.

3.1. Personal conversations with practitioners To start, conferences and discussions with experts provide feedback from the professional sectors of building design, engineering and construction. These interactions are held in the form of personal conversations, presentations at conferences or open discussions in offices. The design firms and experienced practitioners give a critical review of what should be queried in a new database in order to legitimately compare the environmental impact of building structures. These conversations with experts pave the way to accurate baselines for embodied carbon while keeping the results easily accessible for the stakeholders. Next to this critical review of the field, these interactions with practitioners also play a crucial part in the gathering of data. These contacts with professionals deliver a significant amount of data on material quantities in building projects. Leading structural engineering firms such as Arup and Thornton Tomasetti have shared hundreds of projects with this research to analyze, compare and validate the embodied carbon assessment method in their offices.

3.2. Assessment of existing literature and tools As mentioned above, the available data are variable and often unreliable. It is therefore important to thoroughly and critically review the challenges and opportunities arising from the literature and existing tools. On one hand, (scarce) published numbers on material quantities in buildings are listed. On the other hand, the different numbers for ECCs are compared and evaluated. Finally, the current implementation of the actual estimation of the embodied carbon of whole building structures is assessed. In particular, this thesis uses and compares various existing embodied carbon estimating tools. Various software or databases have been assessed (Table 3.1): the Athena Impact Estimator for Buildings, the GaBi LCA software, the Tally tool developed by Kieran Timberlake and PE

24

International, the SOM Environmental Analysis Tool, the WRAP embodied carbon database, etc. Very little has been written on the comparison of the available tools.

Company EC estimating tool Reference Athena Sustainable Materials Institute Impact estimater Athena, 2009

PE International GaBi Software GaBi, 2013 Kieran Timberlake & PE international Tally Tally, 2013

Skidmore, Owings & Merrill SOM Environmental Analysis Tool SOM, 2013 WRAP WRAP Embodied Carbon Database WRAP, 2014

Table 3.1: Existing reviewed embodied carbon estimating tools

3.3. A new interactive database Recognizing the need for a uniform method of assessing embodied carbon in buildings, this research undertakes a long-term initiative to create an interactive, growing database of building projects. The online interface allows architects, engineers and researchers to input data on the material quantities and embodied impact of their projects. The interactive web-based interface, where practitioners from industry can input data on their projects from all around the world in a transparent way, is connected to a growing relational database. While controlling the quality of the entered data, contribution of all industries is encouraged. A “Wiki” approach with multiple participants facilitates the communication with companies. The relational database is coded in MySQL, the interactive web-based interfaced in html and php scripts make the connection between both. The following example illustrates the working of the interactive database. A user inputs the structural material quantities of an office building in London. Then, the participant can opt for the default ECC value offered by the database or enter his/her own customized value. The result will give the GWP of the project compared to other similar structures. Figure 3.1 shows the current interface and database developed for the purpose of this thesis. A transparent database composed of thousands of projects, will give participants a greater confidence in the numbers for the GWP of buildings. Visualizing and comparing results will increase the understanding of the environmental impact of various building types and structural systems. Ultimately, the analysis of material quantities and embodied carbon will direct architects and engineers towards designs with a lower embodied carbon.

25

Figure 3.1: Part of the interactive interface of the database

The following chapters will expand on the three main contributions of this thesis. Chapter 4 reviews the challenges and opportunities in the field of embodied carbon. Chapter 5 proposes a framework for the interactive database. Chapter 6 illustrates the survey of over 200 existing buildings. The results will be summarized, discussed and will lead to conclusions in Chapter 7.

3.4. Summary The methodology can be summarized in three techniques: personal conversations with leading experts, a review of the existing tools and literature and the development of a new interactive database. The experts are structural engineers, architects, researchers and consultants. The existing tools are databases such as WRAP, assessment tools such as Athena or LCA software such as GaBi. The interactive database is composed of a web-based interface and a relational database and will be discussed in more detail in Chapter 5. As a whole, these three techniques led to the development of the challenges and opportunities in obtaining material quantities and ECCs (Chapter 4), the framework for a new database (Chapter 5) and the survey of over 200 existing buildings (Chapter 6).

Structural System & Material Choice

Included Building Components

Natural Hazard & Climate Zone

Building Life Cycle Stages

Database for Embodied Quantity Outputs Search My Projects New Project Settings Logout deQo

PROBLEM NEXT WHAT? HOW? HOW? RESEARCH

26

27

4. Challenges and opportunities in obtaining embodied quantities

4.1. Introduction

While individual companies and researchers are developing their own in-house databases, it is important to understand the challenges arising from carbon accounting. For an accurate, complete and reliable database, it is essential to know the obstacles before it is possible to overcome them. The thesis therefore addresses the lack of methodology or regulation divided among three topics:

1. The first task is to collect material quantity data and assess their quality. This can allow for comparisons across building types and structural systems.

2. The second goal is to propose standards for ECCs that are reasonably accurate but at the same time do not require a complicated calculation from a complete LCA for each material and each project.

3. The third topic addresses the implementation of the database. It includes unifying the different methodologies in a transparent way while respecting intellectual property.

For each of the three problems identified, this thesis defines opportunities as well as challenges. The opportunities illustrate the possibilities of the environmental impact database and can solve part of the tasks in each topic, but challenges remain to be undertaken.

4.2. Getting material quantities The first challenges consist in obtaining accurate material quantities and generating as much data as possible. For accurate data, the scope of what is included should be very well defined (for example, should the sub-structure be included?). Generating large amounts of data requires incentives for architecture, engineering and construction firms to share their project information. Indeed, hundreds of projects are needed to create a representative sample pool. The opportunity in getting structural material quantities arise from Building Information Models (BIM) such as Revit (Revit, 2014). Connecting Revit models to a database of material quantities will generate data quickly in an automated way. Architecture and structural design firms now almost always use BIM models for their projects. These 3D model based data consequently makes a significant amount of data on material quantities in buildings available. New design projects have a digital version of the amount of steel, concrete, timber or other materials used in the building structure. With the right incentives, these design firms can share the information on their project in a fast and automated way with BIM, facilitating the collection of data on material choice and material quantities in the building projects. Accordingly, a considerable amount of quantitative data is already available in design firms and a user-friendly interactive web-interface should help populate the database.

28

4.3. Accurate ECCs

4.3.1. Literature on ECCs

The definition of ECCs is an important and complex matter. Typically these data are obtained from Life Cycle Inventory (LCI) databases. Many databases are available such as Bath University’s ICE report (Hammond and Jones, 2010), Athena (Athena, 2009) and EcoInvent (EcoInvent, 2013). In particular, the ICE presents ‘cradle-to-gate’ data for carbon and energy impacts of building materials mainly in the United Kingdom. Athena integrates average transportation distances (results customizable for different regions of the United States of America) as well as impacts from construction, maintenance and demolition (Athena, 2009). The tool helps designers evaluate environmental impacts without developing detailed material quantity takeoffs. The EcoInvent datasets are based on industrial data. The tool is compatible with most LCA and eco-design software tools (EcoInvent, 2013). In order to understand the variability of ECCs among different databases, the values for cement and concrete obtained by ICE and Athena are reported in Table 4.1. For concrete in particular, the ECC can vary in a significant way depending on strength (Table 4.1, Figure 4.1 and Figure 4.2), cement quantity, percentage of fly ash (Figure 4.1) and blast furnace content. Moreover, in the case of reinforced concrete the environmental impact of reinforcing bar (rebar) has to be considered: the ICE suggests adding 0.77 for each 100kg of rebar per m3. With different concrete strengths and rebar contents, the results vary widely per m3 (240 kg versus 788 kg, Figure 4.2). A critical study is needed to interpret this extreme variability to establish reliability in the available concrete ECCs.

ICE Athena Cement 0.74 0.776 Concrete 16/20 0.100 0.091 Concrete 25/30 0.113 0.128

Table 4.1: ECC for cement and concrete, data after (Hammond and Jones, 2010) and (Athena, 2009)

Figure 4.1: ECC for concrete, varying strengths and percentages of fly ash,

data after (Hammond and Jones, 2010)

0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20

RC 20/25 (20/25 MPa)

RC 25/30 (25/30 MPa)

RC 28/35 (28/35 MPa)

RC 32/40 (32/40 MPa)

RC 40/50 (40/50 MPa)

0% fly ash 0.132 0.140 0.148 0.163 0.188

15% fly ash 0.122 0.130 0.138 0.152 0.174

30% fly ash 0.108 0.115 0.124 0.136 0.155

EC

C fo

r co

ncre

te

(kg C

O2e

/kg m

)

29

Figure 4.2: Embodied carbon for a 1m3 of concrete at varying strengths and percentages of rebar, data

after (Hammond and Jones, 2010) Next to concrete, the ECC values for steel products also present an important variation, as illustrated in Figure 4.3 for various steel products and different recycled contents.

Figure 4.3: ECC for steel products at varying recycling contents, data after (Hammond and Jones, 2010)

0

100

200

300

400

500

600

700

800

16/20 MPa 20/25 MPa 25/30 MPa 28/35 MPa 32/40 MPa 40/50 MPa

0% rebar 240 257 271 288 317 362

1% rebar 425 442 456 473 502 547

2% rebar 535 552 567 583 612 658

3% rebar 665 682 697 713 742 788

kgC

O2e

fo

r 1

m3 o

f co

ncre

te

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

UK Typical - EU 59% Recycled

R.O.W. Typical - 35.5% Recycled

World Typical - 39% Recycled

Primary - 100% hypothetical virgin

General 1.46 2.03 1.95 2.89

Bar & Rod 1.40 1.95 1.86 2.77

Coil (Sheet) 1.38 1.92 1.85 2.74

Coil (Sheet Galv.) 1.54 2.12 2.03 3.01

EC

C fo

r st

eel

(kg C

O2e

/kg m

)

30

Table 4.2 proposes average ECC values after the ICE database (Hammond and Jones, 2010) and EcoInvent (EcoInvent, 2013). The opportunity lies in the definitions of average numbers per location. If the database can propose an agreement for ECC standards, it can allow practitioners to use both average and more sophisticated customized values.

Material ECC in kgCO2e/kg

Concrete1 Standard 0.11* High Strength 0.13*

Steel2

Sections (beams, columns) 1.1 Sheeting 2.6 Studs 1.2 Plates 2.5

Rebar3 65% recycled content 1.7 *The Carbon coefficients for concrete can vary in a significant way depending on strength, cement quantity,

percentage of fly ash and blast furnace content, in this table two values are provided that can be applied respectively for normal concrete (C20/25 - C28/35 and 30% fly ash) and high strength concrete (C32/40 -

C40/50 and 30% fly ash). This does not include reinforcing steel in the concrete. 1after (Hammond and Jones, 2010);

2after (EcoInvent, 2014). 3after (GaBi, 2013)

Table 4.2: Evaluation of ECCs for a set of structural materials, in collaboration with Ornella Iuorio

4.3.2. Applied ranges for ECCs This section discusses a choice of default ECCs, based on a critical review of databases, software and design scheme guides of leading companies. Figure 4.4 illustrates the results for unreinforced concrete, from different sources such as the ICE database, GaBi, Athena, The Concrete Centre. The deQo database follows the most reliable source or an average of the most reliable numbers.

Figure 4.4: ECC of low, medium and high strength concrete from various sources

0.05

0.1

0.15

0.2

0.25

20 30 50

EC

C o

f co

ncre

te

(kg C

O2e

/kg m

)

Low Medium High Concrete Strengths (MPa)

ICE Bath University

GaBi 2006

The Concrete Centre

Athena 2003

deQo

31

Not only do these amounts depend on location (transport from extraction site to construction site), but also vary as a function of material composition. For example, for an ECC of reinforced concrete, numbers depend significantly on the mixes. Indeed, different cement contents, fly ash replacements, or rebar contents will have a considerable impact on the value of the ECC. However, to simplify the approach to estimating the GWP of buildings, an average ECC range can be determined. Figure 4.5 proposes average values for the ECC of reinforced concrete from low to high strength concrete and from 0 to 5% of rebar.

Figure 4.5: Average ECC of reinforced concrete

The same analysis is conducted for steel values (Figure 4.6), differentiating hotrolled steel and rebar. The ICE database gives the value for steel in the United Kingdom; Athena is applied to North America; SimaPro gives a general value for both hotrolled and rebar steel; and GaBi incorporates different global values. As can be discerned from Figure 4.6, the values are very different from one source to another.

Figure 4.6: ECC for hotrolled and rebar steel from various sources

0.1

0.15

0.2

0.25

0.3

20 30 40 50

EC

C o

f co

ncre

te

(kg C

O2e

/kg m

)

Concrete Strength (MPa)

0% rebar

2% rebar

5% rebar

0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00

ICE Bath University

Athena 2003 GaBi 2006 SimaPro deQo

EC

C o

f st

eel

(kg C

O2e

/kg m

)

Hotrolled Steel

Rebar

32

A similar comparison was established for timber as a structural material. Figure 4.7 gives a summary of the ECC values for timber (Arup, 2008).

Figure 4.7: ECC of glued laminated timber from various sources

Having made this overview of material quantities and ECCs, a more critical evaluation is possible for the GWP results for building projects in section 4.4.

4.4. Implementation of a unified method The last topic in this chapter is the implementation of a database with material quantities and embodied carbon. To define the challenges in this field, the first step is to analyze existing embodied carbon estimator tools (Table 4.3). Current developments are mainly design-oriented. Usually, tools do not give a specific indication of a baseline for benchmarking. A database-oriented tool is needed to shift away from the design process towards disclosure of projects after construction, in order to give an idea how the embodied carbon of a new design will compare to a typical building of the same typology and structural system.

Tools Advantages Disadvantages

TALLY beta version • Uses BIM output • Design oriented

• Control what BIM includes? • Transparency

SOM Environmental Analysis Tool

• Basic default assumptions • Option for customized details

• No Normalized metrics (kg/m2) • Linear modeling of ECCs

Athena Carbon Estimator

• Established estimator • No LCA expertise required

• No BIM connection • Transparency

ARUP PECD • Uses BIM output • Hundreds of projects available

• Control what BIM includes? • Transparency

Table 4.3: Review of used and tested tools Two important aspects contradict one another: the need for transparency and the protection of intellectual property. Indeed, if a comparative database divulges all projects, the risk exists that companies will use it to demonstrate “lower” embodied carbon compared to their competitors, which could result in skewed data-points. Also ownership of data should be taken into account. An option would be to allow anonymous data input, which then undermines the transparency.

0 0.2 0.4 0.6 0.8

1

UK Germany Australia Canada Global Global

v2.0 2006 2007 2003 2014 2014

Bath ICE GaBi RMIT Athena Arup study deQo

EC

C o

f gl

ulam

(k

g CO

2e/k

g m)

33

Also, the implementation has to make sure the input method is unified, so that apples are compared with apples. Therefore, a uniform method is necessary. If the same two variables (SMQ in kgm/m2 and ECC in kgCO2e/kgm) are used for all project entries, with a clear definition of what is included (sub/superstructure), it will be possible to compare similar building types (office, residential, healthcare, stadia, etc.), structural systems, or locations. A unified and transparent database, with clear standards on ECCs and populated with thousands of projects, would give a greater confidence in the Global Warming Potential (kgCO2e/m2) of building structures and result in a baseline for comparison in the field of material weight and embodied carbon. Examples of potential outcomes of the database are given in Figure 4.8 and Figure 4.9. These express the preliminary results of the GWP of real projects. The user can filter by different parameters, such as building type (Figure 4.8) or structural system (Figure 4.9). The preliminary results of available data have shown that average values range from 200 to 500 kgCO2e/m2.

Figure 4.8: Example of graphic showing GWP ranges for varying building types

Figure 4.9: Example of graphic showing GWP ranges for varying structural systems

Building type 1 Building type 2 Building type 3

GW

P (k

g CO

2e/m

2 )

900

800

700

600

500

400

300

200

100

0

185

164

143

123

102

82

61

41

21

0

GW

P (lbsC

O2e /ft 2)

13

composite structure.

Figure 5: Example of GWP graphic showing range for different building types

Figure 6: Example of GWP graphic showing range for different structural systems

CONCLUSIONS AND FUTURE WORK Synthesizing challenges & opportunities The contributions are

• … • … • …

0

100

200

300

400

500

600

700

800

900

building(type(1( building(type(2( building(type(3(

GWP (kg CO2e/m2 )

Global Warming Potential (GWP) per building type (kg CO2e/m2)

0(

200(

400(

600(

800(

1,000(

1,200(

structural system 1 structural system 2

GWP (kg CO2e/m2)

Global Warming Potential (GWP) per structural systems (kg CO2e/m2)

185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0

GWP (lbs CO2e / ft2)


0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )

Global Warming Potential (GWP) per building type (CO2e/m2)

0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


Maximum Standard deviation Median Standard deviation Minimum

0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )

Global Warming Potential (GWP) per building type (CO2e/m2) 13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


200

400

600

800

1,000

1,200

Structural system 1 Structural system 2

GWP (kgCO2e/m2)

13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


0

200

400

600

800

1,000

1,200


GWP (lbsCO2e/ft2)

13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


Structural System 1 Structural System 2

GW

P (k

g CO

2e/m

2 )

1,200

1,000

800

600

400

200

0

GW

P (lbsC

O2e /ft 2)

246

204

164

123

82

41

0

13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )






• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


200

400

600

800

1,000

1,200


GWP (kgCO2e/m2)

13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


0

200

400

600

800

1,000

1,200


GWP (lbsCO2e/ft2)

13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



13





• … • … • …

0

100

200

300

400

500

600

700

800

900


GWP (kg CO2e/m2 )


0(

200(

400(

600(

800(

1,000(

1,200(


GWP (kg CO2e/m2)


185

164

143

123

102

82

61

41

21

0

246

204

164

123

82

41

0



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )



0

100

200

300

400

500

600

700

800

900


GWP (CO2e/m2 )


34

4.5. Summary The main challenges discussed in this chapter are the following.

• The creation of incentives for companies to share data on their building projects. • The identification of accurate default ECC values considering various locations. • The resolution of data transparency while protecting intellectual ownership.

Nevertheless, these challenges also lead to following opportunities. • The compatibility of data collection with Building Information Modeling tools allows

for the generation of hundreds to thousands of datapoints. • The proposal for an agreement on accurate ECC values will facilitate the calculation of

the embodied carbon of buildings. • A unified methodology for calculating the GWP of buildings will define reference

buildings for assessing the embodied carbon of building structures. Moreover, this chapter offered default ECCs based on the existing literature. Results for concrete range between 0.1 and 0.2 kgCO2e/kg. With 2% of rebar, it can range between 0.13 and 0.23. With 2% of reinforcement, the value ranges between 0.18 and 0.28. For steel, a distinction is made between rebar (around 1.7 kgCO2e/kg) and hotrolled steel (around 0.8 kgCO2e/kg). Note that these values are used mostly in projects in the United States, but can vary with the location or material specification. Glued laminated timber has an ECC ranging between 0.1 and 0.8 kgCO2e/kg.

35

5. Framework for a database

5.1. Introduction The second contribution of this thesis is to create a framework for a database of building projects. Indeed, efforts exist to create an embodied carbon database of materials, such as the Inventory of Carbon & Energy (Hammond and Jones, 2010). However, until very recently, no database of buildings existed. This research proposes a framework for a worldwide, transparent and interactive database where architects, engineers and other stakeholders can input data about their building projects, more precisely about the material quantities and embodied carbon in their building structures. In 2014, WRAP launched a database collecting embodied carbon in buildings (WRAP, 2014). However, they do not collect material quantities. This method requires a priori knowledge of the embodied carbon in a building project. Also, the collected carbon results in the WRAP database originate from various studies, making different assumptions. The embodied carbon calculated with different tools can therefore not always be compared equally. Therefore, this chapter will propose a framework for a complete database including material quantities together with the embodied carbon of building structures. The database elaborated in this Chapter is named “deQo” or “database of embodied Quantity outputs”. The input and output parameters should be compatible with other databases such as WRAP, Project Embodied Carbon Database (PECD), etc. and use existing, international listings, classes and standards. The database should contain a significant amount of data entered with comparable assumptions. Figure 5.1 illustrates the framework for an interactive database. On one side (Figure 5.1, left), the web-based interface is created to collect data on material quantities in building projects. Architects and engineers can input data on their projects on this interactive interface (deQo, 2014). The user can access this part online and can make different queries.

Figure 5.1: Framework for an interactive database

On the other side (Figure 5.1, right), a relational database, inaccessible to the user, stores the project data. This database in MySQL is connected to the HTML web-interface through a PHP script (Figure 5.1, center), processing the data back and forth.

HTML PHP MySQL web-interface connection database

Platform 1 COLLECT

Platform 2 PROCESS

Platform 3 STORE

accessible by user inaccessible by user inaccessible by user

36

The main aims of the relational database are the following: • Build literacy on typical embodied carbon of structures; • Offer a large data population beyond a single company; • Compare project options (material choice, structural system, etc.) transparently; • Shape a baseline for benchmarking in embodied carbon; • Ultimately allow designers, industry and education to optimize design solutions.

This chapter is contains six sections and a summary. In section 5.2, the database specifications are elaborated: what should be included in the in- and output parameters? In addition, section 5.3 details a framework for the relational database itself. In section 5.4, the interactive web-interface and tool to query results are illustrated. Furthermore, section 5.5 suggests the collaboration with industry and possibilities for connections with Building Information Models (BIM) plug-ins. Finally section 5.6 summarizes the contributions of this database.

5.2. Database specifications This section explains the general framework of the database and expands on the features and options that the database should include in order to be useful for the industry. The in- and out-put parameters of the database will be discussed. The collected information is divided in two groups (Table 5.1): general and structural information. A future integration of operational energy, maintenance and financial cost is possible. 1. General Information

Credits, location, program, geometry

2. Structural Information Material choice, structural systems, material quantities

BIM, BOQ, etc.

Embodied Carbon Coefficients Default or entered by user

Results Material Quantities or GWP ranges

Comparative charts

Table 5.1: Framework for a database

5.2.1. General information The database input asking for general information (Figure 5.2 and Figure 5.3) contains the following input parameters: project name and source information (which can be held anonymous); the date of design, construction, completion and/or publication; building stage; location specifying region, country and town; program; architect, engineer and contractor; accredited rating scheme and applied building code. These parameters are selected carefully after reviewing existing tools and assessing the need for a universal tool in the industry. The main drivers for this relatively concise interface were transparency, the ability to normalize by different functional units, the accurate estimation of ECCs, the integration of material quantities in carbon assessments, the compatibility with other existing databases, the capability to compare projects legitimately and the aim to take conclusions on low-carbon design strategies.

37

General information (1)

Date

Location

Upload photo

Building type (choose all that apply)

Figure 5.2: Interface, “General information” section, part 1

38

General information (2)

Geometry

Figure 5.3: Interface, “General information” section, part 2 The project name has to be entered, even if the contributor can choose to stay anonymous. The source has to be clearly noted, in order to be able to post-verify the data. In case the contributors wish to highlight their project, they are able to upload an image of their building and must specify the source of the image for later publication purposes. The program or type of building, i.e. residential, office, recreational, healthcare etc., are an important factor. Indeed, a hospital has different requirements and therefore different material quantities than an office. Geometry analysis includes aspects such as height and number of floors. The geometry has a mandatory entry: the total useable net floor area (m2 or sf). Indeed, to be able to normalize the data, a functional unit should divide the absolute values of material quantities (kg or lbs) and embodied carbon (kgCO2e) in structures. However, another functional unit should be used if more appropriate, for example the number of seats in stadia or the number of fulltime occupants for schools. The gross floor area is asked when the total useable net floor area is unknown. A percentage of this value can be used as an approximation of the net floor area.

39

5.2.2. Structural information

The second set of input parameters contains the structural information (Figure 5.4) of the building projects. The structural system is divided into vertical, horizontal and lateral systems. Then, the material choices and material quantities are requested. Furthermore, the user needs to specify which building components are included. Resilience towards earthquake and other natural hazards are taken into account. Alongside this information, factors such as the climate zone are entered. Figure 5.4 illustrates the interface asking about the structural information of a newly entered building project.

Structural information (1)

Structural system

Material choice

Buidling components included

Natural hazard and climate zone

Figure 5.4: Interface, “Structural information” section, part 1 A main material is selected for vertical, horizontal and lateral loads before specifying the structural system. Table 5.2 illustrates the available structural systems the user can pick from.

40

Struc tura l Sys t em Material Vertical Horizontal Lateral Concrete In-situ wall In-situ 1-way spanning Shear wall

In-situ column In-situ 2-way spanning Rigid frame

Precast column In-situ flat slab Infill wall

Other Ribbed and waffle slab Other

Post-tensioned band beams (long span)

Post-tensioned flat slab

Precast hollow core (composite)

Precast hollow core (non-composite)

Precast long span (composite)

Other Steel Studs & panel wall Composite metal decking Bracing

Steel column Non-composite metal decking Diagrid

Diagrid Beam & decking Other

Truss & decking

Diagrid

Other

Timber Studs & panel wall Solid timber slab Shear wall

Solid column Timber joists & decking Bracing

Glue laminated column Solid beams & girders Other

Solid stacked wall Glue laminated beams & girders

Other Timber truss & girders

Other

Masonry Brick Wall Vaults Shear wall

Concrete Block Wall Other Other

Other

Other

Table 5.2: Main Materials and corresponding structural systems Table 5.3 illustrates the list of building components. The user selects which components are included in the analysis. Non-structural building components are added in order to make the database expandable to non-structural material quantities in the future.

Building components included Struc tura l Non-s t ruc tura l Foundations ✓ Basement retaining walls ✓ Lowest floor construction ✓ Frame ✓ Upper floors ✓ Roof ✓ Stairs and ramps ✓ External walls ✓ ✓ Windows and external doors ✓ Internal walls and partitions ✓ ✓ Wall finishes ✓ Floor finishes ✓ Ceiling finishes ✓ Fittings, furnishings and equipment ✓ Services ✓ Roads, paths and paving ✓ External drainage ✓ External services ✓

Table 5.3: Structural and non-structural building components included

41

Before the material quantities are entered, the source of these data is specified (dwgs, BIM, bill of quantities, etc.). For the material quantities, only the materials that have been selected in ‘material choice’ and ‘structural system’ will be displayed (Figure 5.5). They can either be entered in absolute values (in kg) or relative values (kg per appropriate functional unit, usally m2). If only the absolute value is given, this value is normalized. The absolute value is divided by the functional unit. In most cases, the functional unit will be the useable floor area, given in the general information section. The relative material quantities are consequently expressed in kilograms per square meter. Also, the applied life cycle stages entered in order to make ‘apples-to-apples’ comparisons for the GWP results.

Structural information (2) Material quantities

Embodied Carbon Coefficients

Life Cycle Stages

Figure 5.5: Interface, “Structural information” section, part 2

42

Additional, optional information can be added such as the structural grid, typical span, the foundation and ground types (Table 5.4). If information is known on these types, it is useful to compare material quantities of structures with similar grounds/foundations. Also, the loads can be entered: dead and live load in kN/m2 as well as the wind, snow and seismic loads. The database will ask whether the dead load is superimposed or total. Instead of entering a number for the live load, a use type can also be specified.

Table 5.4: Ground types

5.2.3. Default versus entered ECCs The scope of the project location is worldwide. The goal is to include hundreds to thousands of structures globally and to define default ECCs for location zones, such as countries and regions. ECCs are computed considering the location of the building. However, the user can enter his/her own ECC calculations when citing the source clearly. The GWP can then be calculated by multiplying the SMQs with the ECCs. It is important to clearly state the assumptions users make when entering their own project data. For example, if they enter the ECCs, they should state clearly which life cycle stages are included and reference the source of their calculations (GaBi, SimaPro, etc.). The quality assessment of the data will depend on these assumption specifications.

5.2.4. Contribution of a new database The proposed framework for a new database results from the review of existing carbon estimating tools and conversations with experts. Table 5.5 summarizes the contribution of the parameters collected deQO and compares the parameters to those in the existing tools reviewed in Chapter 4. The reasons for incorporating these parameters in the database, summarized in the second column of Table 5.5, are the following: data should be as transparent as possible (“transparency”) while protecting proprietary rights (“intellectual property”); data should be compatible with other tools (“compatibility”); data should allow accurate ECC estimation (“ECC”); data should be comparable (“comparison”) and therefore normalized

A Rock or other rock-like geological formation, including at most 5 m of weaker material at the surface.

B Deposits of very dense sand, gravel, or very stiff clay, at least several tens of meters in thickness, characterized by a gradual increase of mechanical properties with depth.

C Deep deposits of dense or medium dense sand, gravel or stiff clay with thickness from several tens to many hundreds of meters.

D Deposits of loose-to-medium cohesion less soil (with or without some soft cohesive layers), or of predominantly soft-to-firm cohesive soil.

E A soil profile consisting of a surface alluvium layer with v s values of type C or D and thickness varying between about 5 m and 20 m, underlain by stiffer material with v s > 800 m/s.

S1 Deposits consisting, or containing a layer at least 10 m thick, of soft clays/silts with a high plasticity index (PI 40 or more) and high water content

S2 Deposits of liquefiable soils, of sensitive clays, or any other soil profile not included in types A – E or S 1

43

(“normalization”); and data should say something about design strategies (“design”). The parameters are selected carefully in order to have a limited number of questions in the interface while enabling a complete analysis of material quantities and embodied carbon in buildings.

deQo Why is this parameter important? W

RA

P

SOM

env

ironm

enta

l ana

lysi

s

Tal

ly e

nviro

nmen

tal i

mpa

ct

PEC

D

Tho

rnto

n T

omas

etti

data

base

Ath

ena

GaB

i

Sim

aPro

Eco

Inve

nt

ICE

General Information Project name Transparency ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ Metric and imperial units Compatibility ✓ ✓ ✓ Architect Transparency ✓ Engineer Transparency ✓ ✓ ✓ Contractor Transparency ✓ Client Transparency ✓ Publication Transparency ✓ ✓ Anonymity Intellectual property ✓ Date Design ✓ ✓ ✓ ✓ Location ECC & Design ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ Program Design ✓ ✓ ✓ ✓ ✓ Volume Compatibility ✓ ✓ ✓ ✓ Total useable floor area Normalization ✓ ✓ ✓ ✓ ✓ ✓ Total gross floor area Normalization ✓ ✓ ✓ ✓ ✓ ✓ Height Normalization ✓ ✓ ✓ ✓ ✓ ✓ Number of stories Normalization ✓ ✓ ✓ ✓ ✓ Story height Design ✓ ✓ ✓ ✓ Number of occupants Normalization Wall to floor ratio Design ✓ ✓ ✓ Plan shape Design ✓ ✓ ✓ Structural Information Structural system Design & Normalization ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ Material choice ECC ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ Building components Comparison ✓ ✓ ✓ ✓ Natural hazard zone Comparison ✓ ✓ Climate zone Comparison ✓ Material quantities Normalization ✓ ✓ ✓ ✓ ECCs Comparison ✓ ✓ ✓ ✓ ✓ ✓ ✓ Life cycle stages Comparison ✓ ✓ ✓ ✓ ✓ Additional notes Comparison Other aspects Work with BIM Compatibility ✓ ✓ ✓ Projects globally Comparison ✓ ✓ ✓ ✓ ✓ ✓ ✓ Projects across firms Comparison ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ Building scale Design ✓ ✓ ✓ ✓ ✓ ✓

Table 5.5: Motivation and comparison of deQo parameters with existing tools

44

5.3. Relational database

A relational database has been created to collect all the entered information. This means that all the data are set in tables related to each other. Through a php script, the MySQL database and html interface are connected. Figure 5.6 illustrates how the data are stored and how they can be viewed, edited and exported.

Figure 5.6: Relational database, accessed through phpMyAdmin

All the input parameters discussed above, such as the material quantities, embodied carbon coefficients and total floor area are related to each other and can be exported to spreadsheets or used directly within the MySQL database. The simple multiplication, illustrated by equation 2, is performed within the relational database and stored for comparison with other projects.

GWP (kgCO2e/m2) = Relative SMQ (kgm/m2) ! ECC (kgCO2e/kgm) [2]

5.4. Web-based interface The web-interface is aiming at clarity and transparency. As the database becomes more useful with a higher quantity of data, the access to it should be easy and the input parameters should be well defined. The list of input parameters should therefore be short (yet complete) so that the input process is fast.

45

The access to the database through the web-interface is granted after an (open-source) registration (in order to insure data quality). Although the name of the participants will be published, the individual projects can either be kept anonymous or highlighted following the request of the contributor. The first page of the interface is shown in Figure 5.7.

Figure 5.7: Register for deQo (deQo, 2014)

In the interactive web-interface, the user can click on the ‘Search’ button to illustrate the ranges of results already entered in the database (Figure 5.8). On the y-axis of the search tool, either “Material Quantities” or “Global Warming Potential” can be shown.

Figure 5.8: Search tool

Implementing Embodied Carbon & BIM models

Urban Modeling Interface

PROBLEM NEXT WHAT? HOW? HOW? RESEARCH

3

DATABASE FOR EMBODIED QUANTITY OUTPUTS

deQo

embodiedco2.scripts.mit.edu

urbanmodellinginterface.ning.com

46

The relational database should report results separately by building type, building component, material, structural system, life cycle stage, location, loading, average floor height, span, total height and number of floors. Finally, the rating scheme a building received can also be listed as another filtering factor to give an idea how well low embodied carbon projects are rewarded. A specific approach needs to be followed for protecting the anonymity of building projects. For example, if a data-point has a height of 800 meters and a location set in Dubai, it is publicly known to be the Burj Khalifa (Figure 5.9, right). If another project is 1776 feet tall and located in New York City, confidential data on the Freedom tower (Figure 5.9, left) will be given away. It is therefore important to give the results only in ranges (of material quantities, embodied carbon or even height).

Figure 5.9: Freedom tower (Dunlap, 2014), Burj Khalifa (photograph by the author)

In this thesis, all input parameters are expressed in metric units. However, clicking on ‘imperial’ instead of ‘metric’ allows all data entries and results to show in the appropriate unit for the user.

5.5. Collaboration with industry

5.5.1. Revit plug-in developers In order to collect a considerable amount of data on material quantities, collaboration with plug-in developers for Building Information Models (BIM) is necessary. An example of such a tool is the recently developed Revit plug-in for estimating the environmental impact of buildings. By using Revit models of their projects, Arup and Thornton Tomasetti were able to contribute hundreds of projects to the current deQo database (Figure 5.10).

Tally tool (Tally, 2014) PECD (PECD, 2013) (Schumaker et al., 2014)

Figure 5.10: Databases and tools using Revit for collecting material quantities

20

WHY? PROBLEM LITERATURE NEXT PROBLEM RESULTS

3. Implementation: calculating the GWP of structures

Challenge: Resolve data transparency while protecting intellectual ownership

“Location: New York City; Height: 1776 ft”

“Location: Dubai; Height: 2,722 ft”

17


1. Getting Material Quantities

Opportunity: Using BIM models to generate material quantity data

17




17




47

5.5.2. Industry and research database The growing database is an important step towards an agreement for benchmarking. Indeed, the database will define baselines for building structures in order to measure a project’s efficiency in terms of material use or carbon emissions. To create such a database, this research relies on voluntary data input from the leading firms in the industry, such as engineering offices and contractors. Furthermore, any data to feed into the database will have to be generated according to a universal set of rules (e.g. on boundary conditions and embodied carbon coefficients) to allow for meaningful comparisons and statistical analysis. Figure 5.11 illustrates the collaboration between MIT and WRAP as well as participation opportunities for other companies. The deQo tool collects not only embodied carbon of building structures, but also their material quantities. With an agreement on corresponding ECCs, a connection can occur with the WRAP database, that is collecting solely the embodied carbon of buildings. Industry can participate by adding to the ‘Bill of Materials Database’ as well as to the ‘Project Embodied Carbon Database’. The more industry participates to both initiatives, the more accurate conclusions will be on material efficiency in structures and benchmarking of embodied carbon. A way to create an incentive to add more data to the database is to reward the carbon accounting in rating systems such as LEED. Also, a simple ‘release of information’ can be signed by a company to allow its employees to add data about their projects. The companies’ logos can be added to the ‘contributors’ section on the web-interface.

Figure 5.11: Collaboration deQo & WRAP databases and participation opportunities

Material Quantities

Embodied Carbon Coefficients (ECC)

Embodied Carbon

x

=

Bill of Materials Database

Project Embodied Carbon Database

Materials Efficiency

Benchmarking

ECC

deQo

48

5.6. Summary

In this chapter, a framework is developed for a database collecting material quantities and the environmental impact of building projects. The database developed for the purpose of this thesis focuses on embodied carbon of building structures, but is designed with possibilities to expand to operational carbon and non-structural elements. While working with existing tools and databases such as WRAP, Tally or the SOM environmental analysis tool, the database for embodied quantity outputs (deQo) is a framework for future development. As the field of embodied carbon needs to mature in the coming decade, the various existing tools will merge in one way or another to compare and combine the available data. The deQo tool proposes in this chapter aims to help the industry to develop new benchmarks for embodied carbon in their structural projects. In comparison with other available tools, deQo attempts to offer a succinct but complete interface to allow meaningful comparisons of the environmental impact of buildings. By also looking at the material quantities, deQo can lead to more material efficiency in design.

49

6. Survey of existing buildings

6.1. Introduction This chapter summarizes the results of the data collected through deQo. A survey of the 200 existing building projects will be presented. Different normalizations, filters and presentations of the material quantities and embodied carbon will lead to conclusions. Additionally, the developed approach is applied on case studies. The first case study compares nine stadia to assess the impact of their design strategies on the environment. The second case study compares two historic bridges with opposite sustainability strategies in order to learn lessons from history. The third case study compares skyscrapers worldwide.

6.2. Existing building structures from the industry This section shows the results of the material quantities and embodied carbon of real existing buildings from the industry, all fully or nearly completed. Some projects are collected through literature review, but most projects are shared by leading design firms such as Arup, Thornton Tomasetti through the deQo interface. Figure 6.1 illustrates the concrete, hotrolled and rebar steel for ten projects entered in the deQo database. A wide range appears already. An obvious trend is that healthcare buildings tend to have the highest amount of material. Only the government building has a greater number. In fact, this government project is an imposing city hall containing a high amount of concrete. The lowest material weights are in office buildings.

Figure 6.1: Material quantities in the ten first projects entered in deQo by program

0

200

400

600

800

1000

1200

1400

1600

1800

2000

SMQ

(k

g/m

2 ) rebar steel

hotrolled steel

concrete

50

Figure 6.1 shows that in some cases iconic design (such as the government city hall) can have a significant impact on the material efficiency. The companies also entered ECCs, verified with the appropriate coefficient values of the corresponding location. The results for the GWP of the building projects can be obtained by multiplying these coefficients with the material quantities, as shown in Figure 6.2.

Figure 6.2: Embodied carbon in the ten first projects of deQo

Steel having a higher ECC than concrete, the contribution of the different materials to the total number shifts compared to the material quantity analysis. Hence, these charts give a better understanding of the environmental impacts of the structures. Figure 6.3 gives an overview of 44 individual projects, ranked again by program. Healthcare buildings are still using a considerable amount of materials, but a wider variety of office buildings appear. Indeed, the embodied carbon of office buildings can vary by height (skyscraper versus low-rise), material use (steel, timber, concrete, composite) or location (city center, distance to material extraction). A new type is the residential building, where much higher quantities are observed. This study also counts in the partitioning walls, which explains this jump. A warehouse is mainly driven by structural efficiency, as the partitioning doesn’t have to be as thorough as in residential buildings, which explains the lower impacts.

0

50

100

150

200

250

300

350

400

450

500

GW

P

(kg C

O2e

/m2 )

rebar steel

hotrolled steel

concrete

51

Figure 6.3: Total embodied carbon per program of 44 individual building projects

In the studies above, the projects were sorted by program. When continuing this analysis on a greater number of building structures, the projects are moreover sorted by structural system, height, number of occupants, location, etc. Figure 6.4 shows the same building projects ranked by structural systems. The results show that concrete structures vary from 100 till 450 kgCO2e/m2, masonry from 225 till 575 kgCO2e/m2, steel from 80 till 475 kgCO2e/m2 and timber from 70 till 490 kgCO2e/m2. The most common values are 275 kgCO2e/m2 for masonry and 175 kgCO2e/m2 for timber. Much higher variations exist for concrete, steel and composite structures.

Figure 6.4: Total embodied carbon per structural system of 44 individual building projects

0

100

200

300

400

500

600

Car

Par

king

C

entr

al U

tility

Pla

nt

Gov

ernm

ent

Hea

lthca

re

Hea

lthca

re

Hea

lthca

re +

Off

ice

Off

ice

Off

ice

Off

ice

Off

ice

Off

ice

Off

ice

Off

ice

Off

ice

Off

ice

Off

ice

Off

ice

Off

ice

Off

ice

Off

ice

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

Res

iden

tial

War

ehou

se

War

ehou

se

GW

P

(kg C

O2e

/m2 )

0

100

200

300

400

500

600

Com

posi

te

Com

posi

te

Com

posi

te

Con

cret

e C

oncr

ete

Con

cret

e C

oncr

ete

Con

cret

e C

oncr

ete

Con

cret

e C

oncr

ete

Con

cret

e M

ason

ry

Mas

onry

M

ason

ry

Mas

onry

M

ason

ry

Mas

onry

M

ason

ry

Mas

onry

M

ason

ry

Mas

onry

M

ason

ry

Mas

onry

M

ason

ry

Mas

onry

M

ason

ry

Stee

l St

eel

Stee

l St

eel,

diag

rid

Tim

ber

Tim

ber

Tim

ber

Tim

ber

Tim

ber

Tim

ber

Tim

ber

Tim

ber

Tim

ber

Tim

ber

Tim

ber

GW

P

(kg C

O2e

/m2 )

52

The same graphical results can be obtained for other parameters, such as location, height, date, number of floors, number of occupants. However, due to intellectual property rights, the following results, including 200 projects from industry, will be shown in ranges, rather than individual projects. Moreover, the box-and-whiskers graphical representation facilitates the visualization of the ranges, outliers, minimum and maximum. For example, Figure 6.5 gives an overview of different programs and the corresponding material quantity ranges. The boxes give the standard ranges with a line indicating the median. The whiskers indicate the minimum and maximum, where the crosses indicate the outliers, excluded from the analysis.

Figure 6.5: Ranges of material quantities for 200 real projects per program category

Figure 6.6 illustrates another way of organizing the data, i.e. by structural system. The 200 actual projects are divided in concrete, steel and composite structures.

Figure 6.6: Ranges of material quantities for real projects per structural system

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

Com

mer

cial

Hos

pita

bilit

y

Edu

catio

n

Gov

ernm

ent

Hea

lthca

re

Res

iden

tial

Spor

ts &

E

nter

tain

men

t

Mix

ed u

se

SMQ

(k

g/m

2 )

0

1000

2000

3000

4000

5000

6000

7000

8000

Concrete Steel Composite

SMQ

(k

g/m

2 )

53

The same graphical representation of the GWP demonstrates the ranges of the embodied carbon of buildings by program (Figure 6.7) or structural system (Figure 6.8).

Figure 6.7: Real projects per program category, ranges of embodied carbon

As expected, the healthcare ranges are slightly higher, as are cultural and hospitability (hotels, which are similar to residential buildings). All buildings range on average between 250 and 700 kgCO2e/m2. These ranges are a first step towards benchmarking of building projects for their environmental impact.

Figure 6.8 divides the 200 projects in concrete, steel, timber and composite structures. Next to a few outliers, the ranges are visualized with the boxes. A wider variety of results exist for the steel buildings. As expected, the timber structures have a lower GWP than other structural types. It is important to note that these numbers consider the ‘cradle to construction’ stages and not the end of life.

Figure 6.8: Real projects per structural system, ranges of embodied carbon

0

500

1000

1500

2000

2500

3000

3500

4000

Com

mer

cial

Hos

pita

bilit

y

Edu

catio

n

Gov

ernm

ent

Hea

lthca

re

Res

iden

tial

Spor

ts &

E

nter

tain

men

t

Mix

ed u

se

GW

P

(kg C

O2e

/m2 )

0

500

1000

1500

2000

2500

3000

3500

4000

Concrete Steel Timber Composite

GW

P

(kg C

O2e

/m2 )

54

In the following sections, several case studies will be analyzed in more depth. The first part discusses the analysis of stadia, the second part historic bridges and the third part tall office buildings.

6.3. Case Study I: Analysis of stadia The first section introduces the studied sport stadia. The second section presents the material quantities used in this case study. The third section calculates the embodied carbon of the projects. Finally, the results are discussed.

6.3.1. Description Most stadia in this study are partially covered, while some have a retractable roof. All stadia were constructed between 2000 and 2011. Table 6.1 illustrates the stadia in this case study. Stadium Locat ion Date Struc tura l Engineer Millennium Stadium Cardiff, Wales 2007 WS Atkins Allianz Stadium Munich, Germany 2005 Arup Joao Havelange Olympic St. Rio de Janeiro, Brazil 2007 Andrade Rezende London Olympic Stadium London, England 2011 Buro Happold Wembley Stadium London, England 2007 Mott Stadium Consortion Aviva Stadium Dublin, Ireland 2010 Buro Happold Australia Sydney Stadium Sydney, Australia 2000 Sinclair Knight Merz Emirates Stadium London, England 2006 Buro Happold Beijing Olympic Stadium Beijing, China 2008 Arup Jaber Al Ahmad Stadium Kuwait, Kuwait 2010 Schlaich Bergermann & Partner

Table 6.1: List of sport stadia with location, date of completion and structural engineer

The material quantities and embodied carbon vary with the sizes of the sport stadia. It is therefore necessary to normalize the results. Two methods are used for normalizing: per area or per seat. Also, when comparing the final results, the presence of a (partial or retractable) roof should be taken into account. Table 6.2 gives the number of seats and type of roof. The ‘area per seat ratio’ is given in order to comprehensively interpret the results. Also, the sources for material quantities are given in Table 6.2 and can be found in part 2 of the references. Stadium Area (m2) # Seats Area/Seat Roof References Millennium Stadium 40,000 74,500 0.54 Retractable [1], [2]

Allianz Arena 37,600 68,000 0.55 Partially covered [3]

Joao Havelange Olympic St. 34,250 45,000 0.76 Partially covered [4]

London Olympic Stadium 61,575 80,000 0.77 Partially covered [5]–[8]+[14]

Wembley Stadium 79,578 90,000 0.88 Retractable [9], [10]

Aviva Stadium 66,460 51,700 1.29 Partially covered [11]–[13]

Australia Sydney Stadium 160,000 110,000 1.45 Partially covered [14]

Emirates Stadium 122,000 60,355 2.02 Partially covered [15], [16]

Beijing Olympic Stadium 254,600 91,000 2.80 Partially covered [17]–[19]+[14]

Jaber Al Ahmad Stadium 400,000 65,000 6.15 Partially covered [20]–[22]

Table 6.2: List of sport stadia with area/seat ratio, number of seats, roof type and references

55

6.3.2. Material quantities in stadia

A comparison of the material quantities in steel and concrete is conducted primarily through a thorough literature review (references can be found in Table 6.2). Figure 6.9.a indicates the material quantities in kg normalized per area in m2. The results demonstrate that most stadia range from 800 kg/m2 to 3000 kg/m2. However, both the Beijing Olympic and the Allianz Stadium have relatively high concrete and steel quantities, compared to the other stadia (up to ten times higher). Nonetheless, the ‘area/seat’ ratio reveals a significant difference: the Allianz stadium with a ratio of 0.55 m2/seat can receive the same number of spectators on a much smaller area than the Beijing stadium with a ratio of 2.80 m2/seat. This means that the comparison per square meter is inequitable: the total material quantity of the Allianz is divided by an area that is 5.68 times smaller than the Beijing stadium for the same amount of spectators. This results in a much higher normalized material quantity. Therefore, it may be more useful to consider the material quantities normalized by the number of seats, which gives a better evaluation of the fulfillment of the function. Figure 6.9.b illustrates the material quantities divided by the number of seats. The material quantities range from 500 to 4000 kg per seat, except for the Beijing Olympic Stadium or “Bird’s Nest.” The material quantity per seat in the Beijing Olympic stadium is approximately ten times higher than in the London Olympic Stadium.

Figure 6.9: Material Quantities in kg: (a) normalized per square meter; (b) normalized per seat

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

Mill

eniu

m

Alli

anz

Are

na

Joao

Hav

elan

ge

Lond

on O

lym

pic

Wem

bley

Avi

va

Aus

tral

ia S

ydne

y

Em

irate

s

Bei

jing

Oly

mpi

c

Jabe

r Al A

hmad

SMQ

(kg/

m2 )

Material Quantities per Area

concrete per area

rebar per area

steel per area

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

Mill

eniu

m

Alli

anz

Are

na

Joao

Hav

elan

ge

Lond

on O

lym

pic

Wem

bley

Avi

va

Aus

tral

ia S

ydne

y

Em

irate

s

Bei

jing

Oly

mpi

c

Jabe

r Al A

hmad

SMQ

(kg/

seat

)

Material Quantities per Seat

concrete per seat

rebar per seat

steel per seat

56

6.3.3. Embodied carbon of stadia

Figure 6.10 illustrates the embodied carbon or GWP of the sport stadia. For projects based in the United Kingdom and Australia, the values of the ICE database are followed. For projects in America, the values of the Carbon Working Group are used. For the Beijing, Munich and Kuwait projects, an LCA of the materials is performed using GaBi (Gabi, 2013). The numbers are illustrated in Table 6.3.

Steel has a higher ECC than concrete. For the same material weights, steel will therefore have a higher impact in the total embodied carbon. Consequently, these embodied carbon charts give a better understanding of the environmental impacts of the stadia, compared to the charts of material quantities. Nevertheless, similar trends appear. The high use of steel and concrete results in a high embodied carbon in the stadia.

Figure 6.10: GWP of stadia: (a) normalized per square meter; (b) normalized per seat

0

200

400

600

800

1,000

1,200

1,400

Mill

eniu

m

Alli

anz

Are

na

Joao

Hav

elan

ge

Lond

on O

lym

pic

Wem

bley

Avi

va

Aus

tral

ia S

ydne

y

Em

irate

s

Bei

jing

Oly

mpi

c

Jabe

r Al A

hmad

GW

P (k

g CO

2e/m

2 )

Embodied Carbon per Area

EC concrete per area

EC rebar per area

EC steel per area

0

500

1,000

1,500

2,000

2,500

3,000

3,500 M

illen

ium

Alli

anz

Are

na

Joao

Hav

elan

ge

Lon

don

Wem

bley

Avi

va

Aus

tral

ia

Em

irat

es

Bei

jing

Jabe

r Al A

hmad

GW

P (k

g CO

2e/m

2 )

Embodied Carbon per Seat

EC concrete per seat

EC rebar per seat

EC steel per seat

ECC United Kingdom Brazil Kuwait China Germany Australia Concrete 0.10-0.13 0.13 0.15 0.17 0.13 0.15 Hotrolled steel 0.88-0.89 0.88 0.71 0.88 0.88 0.89 Rebar steel 1.71 1.71

Table 6.3: ECCs used for this study

57

6.3.4. Discussion of results

The method of normalization (by area or by seat) has an important influence on the interpretation on the results. A stadium showing both high material usage and a high embodied carbon is the Beijing Olympic Stadium. By comparing both types of normalization, a better understanding of the numbers for GWP emerges. Nevertheless, as the results by seat give a better representation of the function of stadia, this method is preferred. Both key variables result in clear trends in the different stadium design strategies. Comparing material quantities and embodied carbon in stadia illustrates the difficulty of this field. Indeed, a simple shift in functional unit from kgCO2e per m2 to kgCO2e per seat changes radically the comparison. Also, the lack of available data on both variables, the material quantities and the embodied carbon coefficients, can question the reliability of the results. While this research is very data-intensive, the case studies of the stadia demonstrate the simplicity and transparency of the proposed approach in equation [1]. The results show that design strategies for stadium structures have a significant impact on their embodied carbon footprint. For example, the “Bird’s Nest” had an esthetic design pattern inspired by Chinese-style 'crazed pottery' (National Stadium, 2014), which resulted in a high material use and embodied carbon.

6.4. Case study II: Lessons from historic bridges This section presents two complete opposite design solutions for sustainable historic: Inca suspension bridges and Roman arch bridges (Figure 6.11). Both bridges are made from local materials, but the design approach is fundamentally different: where Roman stone arch bridges are created to be permanent and have a life time of several thousands of years, the Peruvian communities rebuild the suspension grass bridges every year. This comparative analysis will illustrate the trade-off between lightweight temporary bridges versus heavy permanent bridges in terms of embodied carbon. The key question is: “What can historic bridges teach us on sustainability?” Moreover, the comparison between the permanent and temporary structures can raise the question on the lifetime engineers and architects should consider for their designs. Indeed, temporary and reusable structures could compensate a less durable approach. The chosen case studies are the Roman stone arch “Pont Saint Martin” in Aosta, Italy, and the last remaining Inca suspension grassbridge “Keswha-Chaka” in Huinchiri, Peru. They both have a single span around 30 meters. It is important to note that this analysis is meant to be comparative. The historic analysis require assumptions on the materials and construction methods, so that it is not accurate to talk about absolute numbers. The absolute values are not as relevant as the comparison between the cases, so that a conclusion can follow around the differences between permanent and temporary bridge structures.

58

6.4.1. Description The Roman stone arch “Pont Saint Martin” crosses the Lys river in Aosta, Northern Italy and was built between 27 B.C. and A.D. 14 (Figure 6.11, left). With a span around 35.6 meters, it is the largest known span of Roman bridges. The single span arch is mostly made of gneiss, a local stone material and filled with soil (Blake, 1947). The Inca suspension bridge “Keshwa-chaka” (which translates as grass-bridge) between the villages of Huinchiri and Quehue crosses a 30 meters wide canyon over the Apurimac River in Peru (Figure 6.11, right). The lightweight construction is composed of braided cables for the floor and handrails, connected with vertical grass ropes and fixed at both sides with stone abutments. The villages at both sides of the rivers come together in a yearly three-day festival to rebuild the grass bridge after they throwing the old grass bridge into the river (Ochsendorf, 1996).

> 2000 years

High initial cost Low maintenance cost

High quality stone End of Life reuse

Local material

1 year Low initial cost High maintenance Low quality plants Washed away yearly Local material

Images after [8] Information after [6], [7]

Images after [12] Information after [10]

Figure 6.11: Pont Saint Martin (left) and Keshwa-Chaka (right) The two bridges have a single span with similar length. However, the Roman arch Pont Saint Martin is working in compression and designed for a permanent lifetime, where the Inca Bridge Keswha-Chaka is working in tension and designed for a lifespan of one year. The Roman arch uses high quality stone that could be reused at the end of life of the structure as opposed to the low quality grass used in the Inca Bridge, which is washed away every time the old bridge is thrown in the river. Nevertheless, both bridge designs have a sustainable approach that current bridge engineers and architects can learn from: they use local material and offer efficient solutions to the design problem (Ochsendorf, 2004). 59

Ca

se S

tud

y

1 year Low initial cost High maintenance Low quality plants (washed away) Local material

> 1000 years High initial cost

Low maintenance High quality stone (EoL reuse)

Local material

“What is the difference in embodied carbon and material weight between historic bridges and recent bridge designs?”

The case studies chosen are the Roman stone arch Pont Saint Martin in Aosta, Italy, and the last remaining Inca suspension grassbridge Keswha-Chaka in Huinchiri, Peru. They both have a single span around 30 meters. It is important to note that this analysis is meant to be comparative. The historic bridges require a certain number of assumptions on the materials and construction methods, so that it is not accurate to talk about absolute numbers. The absolute values are not as relevant as the comparison between the cases, so that a conclusion can follow around the differences between permanent and temporary bridge structures.

6.5.1. Description

Span 35.6 m 27 B.C. – A.D. 14

Lys Aosta, Italy

Segmental stone arch

Span 30 m 15th century - today Apurimac Huinchiri, Peru Suspension grass bridge

59

Ca

se S

tud

y




Local material



6.5.1. Description

Span 35.6 m 27 B.C. – A.D. 14

Lys Aosta, Italy




Low maintenance cost High quality stone End of Life reuse

Local material


Figure 39: Pont Saint Martin (left) and Keshwa-Chaka (right)1 2 3 4

1 Aosta, Wikipedia, May 2013, http://nl.wikipedia.org/wiki/Aosta

59

Ca

se S

tud

y




Local material



6.5.1. Description

Span 35.6 m 27 B.C. – A.D. 14

Lys Aosta, Italy





Local material




59

Ca

se S

tud

y 1 year Low initial cost High maintenance Low quality plants (washed away) Local material



Local material



6.5.1. Description

Span 35.6 m 27 B.C. – A.D. 14

Lys Aosta, Italy





Local material




Pont Saint Martin Span 35.6 m

27 B.C. – A.D. 14 Lys

Aosta, Italy Segmental stone arch

Keshwa-Chaka Span 30 m 15th century - today Apurimac Huinchiri, Peru Suspension grass bridge

59

6.4.2. Material quantities in historic bridges First, the material quantities are estimated, based on geometry descriptions in publications, drawing and construction video’s. The sources for the two bridges can be found in part 3 of the references. The ECCs are estimated based on the material descriptions and the ICE database (Hammond and Jones, 2010). The calculations are summarized in (Figure 6.12).

PONT SAINT MARTIN KESHWA-CHAKA

Vault, spandrel walls,

pavimentum & parapets 686 m3 – 1790 t

Local gneiss – ECC 0.017 Filling

780 m3 – 1248 t Soil – ECC 0.023

Abutments – 9070 kg Stone – ECC 0.056 Vertical ropes, floor & handrails – 3630 kg Twisted cords, diameter 5 cm for ropes Braided cables, 6 x 45 m for floor and rails Puna grass – ECC 0.01 Deck – 225kg Sticks & matted reeds

Figure 6.12: Weight/ECC Pont Saint Martin & Keshwa-Chaka, images after [4], [5], [9], [10], [11] The results of the calculation of the material quantities in the Pont Saint Martin are illustrated in Table 6.4. An extended calculation can be found in appendix B.4.a.

Parts

Volume m3

Material

Density kg/m3

ECC kgCO2e/kg

Weight kg

Weight per meter kg/m

Vault 270 Gneiss 2610 0.017 704700 22443 Spandrel walls 390 Gneiss 2610 0.017 1017900 32417 Pavimentum 42 Limestone 2500 0.017 105000 3344 Parapets 26 Gneiss 2610 0.017 67860 2161 Filling 780 Soil 1600 0.023 1248000 39745

Table 6.4: Calculation weights and ECC for Pont Saint Martin The results of the calculation of the material quantities in the Keshwa-Chaka bridge are illustrated in Table 6.5. An extended calculation can be found in appendix B.4.b.

55


Vault, spandrel walls, pavimentum & parapets

686 m3 – 1790 t Local gneiss – ECC 0.017

Filling 780 m3 – 1248 t

Soil – ECC 0.023



Parts

Volume m3

Material

Density kg/m3

ECC kgCO2e/kg

Weight kg




Parts

Volume m3

Material

Density kg/m3

ECC kgCO2e/kg

Weight kg


Abutments / Stone 2500 0.056 9070 302 Cables 1.98 Puna grass 395 0.01

Vertical Ropes 0.6 Puna grass 395 0.01 3630 121 Deck 0.45 Sticks 500 0.01

Table 6.5: Calculations sheet weights and ECC for Keshwa-Chaka

Ca

se S

tud

y Vault, spandrel walls, pavimentum & parapets


Filling 780 m3 – 1248 t

Soil – ECC 0.023

Abutments 9070 kg Stone – ECC 0.056

Vertical ropes Twisted cords, � 5 cm Floor/handrails Braided cables, 6 x 45 m 3630 kg Puna grass – ECC 0.01 Deck 225 kg

Sticks/matted reeds

Cas

e St

udy Vault, spandrel walls,



780 m3 – 1248 t Soil – ECC 0.023



Sticks/matted reeds

Cas

e St




780 m3 – 1248 t Soil – ECC 0.023



Sticks/matted reeds

55


Vault, spandrel walls, pavimentum & parapets


Filling 780 m3 – 1248 t

Soil – ECC 0.023



Parts

Volume m3

Material

Density kg/m3

ECC kgCO2e/kg

Weight kg




Parts

Volume m3

Material

Density kg/m3

ECC kgCO2e/kg

Weight kg





Ca

se S

tud

y Vault, spandrel walls, pavimentum & parapets


Filling 780 m3 – 1248 t

Soil – ECC 0.023



Sticks/matted reeds

Cas

e St




780 m3 – 1248 t Soil – ECC 0.023



Sticks/matted reeds

Cas

e St




780 m3 – 1248 t Soil – ECC 0.023



Sticks/matted reeds

60

Parts

Volume m3

Material

Density kg/m3

ECC kgCO2e/kg

Weight kg





6.4.3. Embodied carbon of historic bridges Now that the weights and ECCs are determined, the Global Warming Potential (GWP) is given by multiplication. The total weight of the different parts have been divided by the span, in order to obtain a “linear” GWP, i.e. the GWP per meter of length of the bridge for the sake of comparison. Considering the Roman arch has a lifespan of more than 2000 years, where the Inca suspension bridge is less permanent, the lifetime of the bridges should be taken into account while looking at the Global Warming Potential. The “linear” GWP is divided by the number of years for which it has been standing to obtain an “annual” GWP. The results are shown in Table 6.6 for Pont Saint Martin and in Table 6.7 for the Keshwa-Chaka bridge.

Parts Material Weight / meter GWP / meter GWP / meter /year (kg/m) (kgCO2/m) (kgCO2/m/y) Vault Gneiss 22442.68 381.53 0.19 Spandrel walls Gneiss 32417.20 551.09 0.28 Pavimentum Limestone 3343.95 56.85 0.03 Parapets Gneiss 2161.15 36.74 0.02 Filling Soil 39745.22 914.14 0.46

Total 1940.34 0.97

Table 6.6: GWP of Pont Saint Martin For the Keshwa-Chaka, the bridge is divided in a permanent and a temporary part. Indeed, the abutments are not rebuilt every year. This is the permanent part, which is divided by the number of years that the bridge has been in use (600 years). The suspended part, the grass bridge itself, is divided by one, as it requires the whole construction and material supply again every year. The “linear” and “annual” GWP are given in Table 6.7.

Parts Material Weight / meter GWP / meter GWP / meter / year (kg/m) (kgCO2/m) (kgCO2/m/y) Abutments Stone 302.33 16.93 0.03 Cables Puna grass

Vertical Ropes Puna grass 121.00 1.21 1.21 Deck Sticks, matted reeds

Total 18.14 1.24

Table 6.7: GWP of Keshwa-Chaka

61

If comparing the “linear” GWP without taking into account the lifespan, the Roman arch has an embodied carbon (1940 kgCO2/m) more than 100 times higher than the Inca suspension bridge (18 kgCO2/m). However, if dividing the permanent parts by the years that the bridge has been in use, the Inca Bridge has a higher embodied carbon (1.24 kgCO2/m/y) than the Roman arch (0.97 kgCO2/m/y).

6.4.4. Comparison Roman arch and Inca suspension bridge In Table 6.8, the comparison of both bridges is illustrated, with both the “linear” and the “annual” GWP compared.

Pont Saint Martin Keshwa-Chaka



Local material – Gneiss & Soil

1 year Low initial cost High maintenance Low quality plants Washed away yearly Local material – Stone & Puna grass

Embodied Carbon . 1940.34 kgCO2/m 0.97 kgCO2/m/y

18.14 kgCO2/m 1.24 kgCO2/m/y

Table 6.8: Comparison Pont Saint Martin and Keshwa-Chaka, images from (Foer, 2013; O’Connor, 1993; The Last Handwoven bridge, 2013)

The first number shows that the Roman arch Pont Saint Martin has an embodied carbon of 1940 kgCO2/m or a factor hundred more than the Inca suspension bridge Keshwa-Chaka with an embodied carbon of 18 kgCO2/m. When taking the lifespans into account, the Roman arch has a slightly lower embodied carbon than the Inca Bridge. When comparing both bridges, other aspects than embodied carbon, span and lifespans would need to be taken into account. The load capacity of the Roman arch is much higher (car traffic instead of pedestrian loads). Moreover, the Inca suspension bridges are also subject to a lot of lateral movements, especially during consequent wind loads. The Inca suspension bridge is part of a cultural heritage and community energy and vitality. Additionally, the bridge would probably last longer, at least up to two years, than the one-year lifespan it is given due to the annual festivities. In both cases, the (linear or annual) GWP value is very low in general. The use of local materials, traditional manufacturing and efficient design has a significant impact on the embodied carbon. These low environmental impacts can teach current engineers and architects about efficient design.

62

6.4.5. Comparison with recent bridge designs A comparison with recent bridges puts previous results in perspective. Figure 6.13 illustrates the material quantities of recent bridges with various spans (Clune, 2013). However, recent bridges are road bridges with multiple lanes carrying much heavier live loads than the historic counterparts discussed in this case study, so that the comparison is only theoretical.

Figure 6.13: Recent Bridges, material quantities (Clune, 2013)

The bridge with the most similar span (around 45 m) has a mass of 500 kg/m2. This bridge has the lowest embodied carbon (best case!) among the recent bridges. With a width similar to Pont Saint Martin (which allows traffic), i.e. 6 m, the mass becomes 83 kg/m. An ECC for steel of 1.77 gives an embodied carbon of 147.5 kg CO2/m. If we consider a lifetime of 50 years, the “annual” GWP becomes 2.95 kg CO2/m/y, which is three times more than the two historic bridges. The upper bound of the range, a cable-stayed bridge, has a “linear” GWP of 590 kg CO2/m and an “annual” GWP of 11.8 kg CO2/m/y. These numbers do not include non-structural materials or maintenance. Indeed, recent bridges often require a new layer of asphalt every twenty years or a new layer of protective paint. Though paving materials and maintenance are not included in results for recent bridges, the GWP is still ten times higher than for the two historic bridges (Table 6.9). The Oakland Bay Bridge, San Francisco, (Figure 6.14) has a mass of 2100 kg/m2 and a width of 17.5m. The “linear” mass is of 120 kg/m. Multiplying with the ECC gives an embodied carbon of 212.4 kg CO2/m, and dividing with the lifetime of the bridge of 76 years, gives 2.8 kg CO2/m/y. Pont Saint Martin 0.97 kgCO2/m/y

Keshwa-Chaka 1.24 kgCO2/m/y

Oakland Baybridge 2.8 kgCO2/m/y

Recent bridge range 2.8-11.8 kgCO2/m/y

Table 6.9: Comparison of “annual” Global Warming Potential

Figure 6.14: Oakland Bay Bridge, San Francisco, United States (photograph by the author)

Oakland Bay Bridge

Sydney Harbor Bridge

63

6.5. Case study III: Comparing tall buildings This section discusses the material quantities and embodied carbon in tall buildings. A comparison of tall landmark buildings worldwide is illustrated by ten case studies.

6.5.1. Description The ten skyscrapers considered in this study are listed in Table 6.10, with their location, date of completion, structural engineer, structural system, height and sources (see part 4 in references). Name Location Date Structural Engineer Structure Height (m) Refs. 30 St Mary Axe London 2004 Arup Steel Diagrid 180 [1] – [4]

The United Tower Kuwait 2011 WSP Cantor Seinuk Reinf. Concrete 240 [5] – [10]

The Shard London 2012 WSP Cantor Seinuk Reinf. Concrete 306 [11] – [16]

Al Hamra Firdous Kuwait 2011 SOM Reinf. Concrete 413 [17] – [20]

Willis Tower Chicago 1974 SOM Steel 443 [21]

World Financial Center Shanghai 2008 Leslie E. Robertson Trusses & columns

492 [22] – [24]

Taipei 101 Taipei 2004 Thronton Tomasetti; Evergreen Engineering

Composite 508 [25] – [26]

One World Trade Center New York 2014 WSP Cantor Seinuk Hybrid 546 [27]

Shanghai Tower Shanghai 2014 Thornton Tomasetti Concrete 632 [28] – [33]

Burj Khalifa Dubai 2010 SOM Buttressed Core 828 [34] – [41]

Table 6.10: Case studies for tall buildings The height of the skyscrapers varies from 180 m for 30 St Mary Axe or “Gherkin” in London till 828 m for the Burj Khalifa in Dubai. As demonstrated by Khan (see section 2.1), the height of tall buildings influences the material quantities. Therefore, the results are always shown in graphics with increasing building height from left to right (or from top to bottom in case of Table 6.10). All buildings have been completed between 2004 and 2014, except for the Willis Tower in Chicago, in order to see how recent buildings relate to older skyscrapers.

6.5.2. Material quantities in tall buildings Figure 6.15 gives the material quantities normalized by floor area. As opposed to the previous calculations in this thesis, the gross floor area has been used, as the internal floor area was not always available for the ten towers. The sources (Bibliography and references, part 4) are mostly publications, bill of quantities or design office documents. Although the sources used to collect the data were fairly reliable, the accuracy could be improved. For example, The United Tower and The Shanghai Tower were both constructed with reinforced concrete, which has a certain percentage of steel. Therefore to obtain more accurate results the exact percentage of reinforcement bars in the concrete should be known. When this percentage was not known for the different cases, a general average of steel percentage in the reinforced concrete was taken.

64

Figure 6.15: Material quantities normalized per gross floor area

6.5.3. Embodied carbon of tall buildings To have a better understanding of the environmental impact of the tall buildings, the used ECCs were adapted to each region: the United Kingdom, the United States, the Middle East and China. Based on the ICE coefficients, LCA calculations adapted the ECCs to include transportation for the different countries. The modification consisted in the multiplication of the original transportation input by a ratio taking into account the surface area of the country. Then, a review of the concrete mix proportions used in every country was performed. The collection of data represented many challenges during this step. For example, no reliable source on the exact concrete mix of the World Financial Center, Shanghai, was found. Therefore, published information on high strength concrete was used, since this is the type of concrete used in the tower. The common water/cement ratio used for concrete lies between 0.45 and 0.6. However in the case of the high strength concrete, the water/cement ratio lies between 0.3 and 0.4. Consequently, the average value of 0.35 was used with a gravel-content of 950 kg/m³. Moreover, the typical cement content for this type of concrete varies between 350 to 500 kg/m³ of cement. Hence, a value of 450 kg/m³ of cement was selected and by applying the water cement ratio chosen the amount of water was calculated to be 157 kg. The water content is slightly lower than the usual water portion found in regular concrete due to the addition of plasticizers.

0

500

1,000

1,500

2,000

2,500

3,000

SMQ

(kg/

m2 )

concrete per area

rebar per area

steel per area

65

In the case of the United Tower in Kuwait, the main inputs used in the Life Cycle Assessment were extracted from the Bill of Quantities (BOQ) obtained from the client, United Real Estates. All information on the construction and the structure of the tower in Kuwait is confidential, and therefore it is difficult to obtain accurate information. From the BOQ, the main reinforced concrete quantities were extracted. However, the additional information regarding the structural system was obtained from construction magazines published in the Gulf. The only information on the exact concrete mix in the BOQ available was the Specification 03300. Therefore, technical papers published in Kuwait were used to adjust the cement content and water/cement ratio in the LCA sheet. Moreover, The BOQ does not specify the proportion of reinforcement in the concrete. Therefore it was assumed that steel reinforcement represents 0.5 percent of the reinforced concrete. One of the challenges of this research is to be very transparent about the data used and the assumptions made. Therefore, the case studies are analyzed with different ECCs, given in Table 6.11. The values of the various sources are comparable in order of magnitude. The recycling and the reuse phases of the materials were not taken into account. Consequently, to assure the accuracy of the numbers found, a comparison is performed with the values from the ICE database and the Carbon Working Group. The results for the embodied carbon of tall buildings are shown in Figure 6.12.

Figure 6.16: Embodied carbon per area of tall buildings

0

50

100

150

200

250

300

350

400

450

500

GW

P (k

g CO

2e/m

2 )

EC concrete per area

EC rebar per area

EC steel per area

ECC United Kingdom Middle East China United States Concrete 0.16 0.15 0.17 0.13 Hotrolled steel 0.89 0.71 0.88 0.88 Rebar steel 1.7 1.7 1.7 1.7

Table 6.11: ECCs used for this study

66

The results show the highest impact from concrete and reinforcement for the Al Hamra Firdous Tower and the United Tower, both in Kuwait. The Taipei 101 tower also represents a higher impact due to concrete quantities used for the construction. The 30 St Mary Axe or “Gherkin” in London has the lowest embodied carbon impact. It used a steel diagrid to enhance the structural efficiency of the tower. It should be noted that, though it was constructed in 1974, the Willis tower has the second lowest impact.

6.5.4. Discussion of the results The high emissions for the Kuwait towers can be linked to the structural system of the skyscrapers. The architectural shape of the Al Hamra Tower (Figure 6.17.a.) requires more concrete than the other tall buildings. Also, the reinforced concrete structure with the concrete core in the United Tower (Figure 6.17.b.) was designed to accommodate the architectural wave feature of the building. The architectural design did not take into account material reduction.

Figure 6.17: (a) Al Hamra (photograph by the author), (b) United Tower (KPF, 2012), (c) Gherkin

(Munro, 2006)

The design of the steel diagrid in the 30 St Mary Axe Tower or “Gherkin” (Figure 6.17.c.) demonstrates that integrating sustainable design and structural engineering starting from the concept stage influences the embodied carbon emissions of the tall buildings and that iconic design is not incompatible with embodied carbon reductions.

6.6. Summary This chapter surveys over 200 existing buildings as well as historic and recent case studies. The survey of material quantities in buildings designed by leading engineering firms was conducted and the results of the ECCs in Chapter 4 lead to a range of embodied carbon values for various building types and different structural systems. Then, three case studies illustrated the process in sport stadia, historic bridges and tall buildings.

39NR 3 • 2004 • NYHETER OM STÅLBYGGNAD S T Å L B Y G G N A D S P R O J E K T

introduced, this increases the change incolumn angle and with it the spreadingeffect of vertical column loads.

In the Swiss Re building all these ho-rizontal forces are carried by perimeterhoops at each node level, which alsoprovide equilibrium for any asymmetricor horizontal loading conditions. Thecombination of these geometrical actionsresults in compression in the hoops atthe top of the building, where the co-lumns are more steeply angled and ligh-ter loaded, to very significant tension

The real thing!

Structural plan near mid-height ofbuilding (showingarrangement ofclear-span radialfloor beams aligning with perimeter columnpositions and lightwell edges).

Plan of the 18th storey,with denotation of the grid

of the raised floor.

Office division (note: showing possiblevariations of office planning layout).

The 3D- model pro-ved to be indispensa-ble in the communi-cation. The structuralengineer made the initial coordinationmodel with centre-lines and sizing, the contractor andsubcontractors used it for detailing andinterfaces with cladding and MEPservices.

The shape of the tower is influenced by the

physical environment ofthe city. The smooth

flow of wind around thebuilding was one of the

main considerations.➤

s 36-43 Swiss NY 04-10-06 16.03 Sida 39

67

The survey of existing buildings demonstrated that the highest weight of material come from concrete, where the highest impacts are mainly due to concrete and hotrolled steel. The survey reveals that healthcare buildings have the highest amount of material (range between 1000 and 2300 kg/m2), followed by residential buidlings (range between 1500 and 1800 kg/m2), whereas office or government buildings have the lowest amount of material (range between 500 and 1500 kg/m2). The GWP shows similar trends with typical values between 100 and 1000 kgCO2e/m2. The various structural systems, such as concrete, steel, timber and composite, were compared. Timber had the lowest cradle-to-site impact. The case studies of the stadia illustrated an important aspect of estimating the GWP of buildings: the choice of functional unit. Dividing the total amount of materials by area or by seat resulted in different conclusions. When normalizing by seat, the Beijing Olympic Stadium was shown to have an impact ten times higher than the London Olympic Stadium. The case studies of the historic bridges showed that recent bridge design can take lessons from history. The Keshwa Chaka Inca bridge illustrates temporary low-carbon design as opposed to the Roman arch Pont Saint Martin using a permanent low-carbon design. Finally, the case studies of the tall buildings illustrate the impact of structural design on the material quantities and embodied carbon in structures. The results range from 250 to 2500 kg/m2 for the material quantities and 50 to 450 kgCO2e/m2 for the GWP. “How much does your building weigh Mr. Foster?” – An Art Commissioners film This chapter is the first attempt in research in answering this question asked by an Art Commissioners film and by many structural engineers and architects.

68

69

7. Conclusions

“ There’s only one creature capable of leaving a footprint that size…” From King Kong movie

7.1. Discussion of results

This thesis has three types of results: (1) a critical review, (2) a new database and (3) a survey analysis. First, this research synthesizes the challenges and opportunities for collecting material quantities and embodied carbon. Second, to collect this information, it develops a framework for a database. Third, a survey of 200 real building projects formulates an understanding of the material efficiency and life cycle impact of actual buildings.

(1) The first part of this thesis discusses the challenges and opportunities in estimating the embodied carbon in structures. In order to obtain the Global Warming Potential (GWP), Structural Material Quantities (SMQ) are multiplied by Embodied Carbon Coefficients (ECC). For each of these three parameters, one main challenge and one main opportunity are defined.

For the material quantities, the industry needs incentives to share enough data on their building projects. Confidence in the results on material efficiency relies on voluntary data input from the leading firms in the industry, such as design offices, engineering firms and contractors. Motivations are necessary for stakeholders to spend additional time entering the material amounts in the database. However, BIM models such as Revit offer the opportunity to collect thousands of projects in an automated way. Linking Revit models to the database will generate thousands of data on material quantities in buildings. Another challenge is to identify accurate ECC values. The available numbers vary widely, due in part to different life cycle stages included. Most currently available ECCs are cradle-to-gate. Therefore, parts of the life cycle such as transportation, construction or demolition are omitted. However, there is an opportunity to define default ECCs, based on the existing literature. For unreinforced concrete, ECCs range between 0.1 and 0.2 kgCO2e/kg depending on the strength. For reinforced concrete, the values lie between 0.13 and 0.23 kgCO2e/kg for 2% rebar and between 0.18 and 0.28 kgCO2e/kg for 5% rebar. Average values for steel are 1.7 kgCO2e/kg for rebar and 0.8 kgCO2e/kg for hotrolled steel. Values shift according to varying location, manufacturing, transportation, fly ash replacement, recycled content, etc. After obtaining these two key variables, calculating the GWP itself presents challenges. Indeed, many firms request that their data be kept confidential. However, protecting the intellectual ownership of companies critically compromises comprehensive transparency. Nonetheless, a unified methodology will facilitate benchmarking of embodied carbon in structures. The opportunity lies in defining a universal assessment of the GWP, by obtaining the two key variables, SMQ and ECC, in a transparent way. Using the database developed in part (2) and the survey in part (3), a greater confidence in the numbers on embodied carbon in building structures will identify reference buildings for comparison.

70

(2) The second part of this thesis synthesizes the opportunities developed in part (1) into a new database, called “deQo” or “database for embodied Quantity outputs”. This thesis establishes a framework for this database collecting material quantities and embodied carbon of building projects.

A relational database is developed in MySQL connected to an html web-interface through php scripts. The deQo tool collects not only the embodied carbon, in addition to the work from the WRAP embodied carbon database (WRAP, 2014), but also the amounts of steel, concrete, timber and other materials in building structures. The interface targets structural materials, but the database was designed to expand to all building materials. Many leading engineers are developing methods for estimating the embodied carbon of their projects, such as the SOM environmental analysis tool (SOM, 2013) or the Tally environmental impact tool (Tally, 2014). The review of these existing tools and conversations with leading experts have defined the parameters in deQo. The field of life cycle impacts of buildings is maturing and the existing efforts will likely soon merge into a comprehensive approach. This database is a first step towards an agreement on ranges for material weights and embodied carbon in typical buildings. In the long term, deQo will offer the industry a basis for benchmarks on material efficiency as well as low-carbon buildings.

(3) The third part of this thesis analyzes a survey of over 200 buildings collected through deQo developed in part (2). A unified methodology (GWP = SMQ ! ECC) is applied to the data from the survey and to three case studies in more detail.

A survey of 200 existing buildings reveals the material efficiency and the carbon impact of structures. As can be expected, healthcare buildings have the highest material amounts between 1000 and 2300 kg/m2 and the highest embodied carbon between 300 and 800 kgCO2e/m2. Residential buildings also have high impacts, with material quantities ranging from 1100 to 1900 kg/m2 and GWP values ranging from 250 to 750 kgCO2e/m2. The lowest impacts can be found in office and government buildings with material quantities between 500 and 1500 kg/m2 and GWP between 200 and 1000 kgCO2e/m2. In general, the typical embodied carbon in structures ranges from 100 to 1000 kgCO2e/m2. Additionally, the different structural systems are compared, with timber having the lowest impact when taking into account cradle-to-site life cycle stages. Furthermore, detailed case studies illustrate three important aspects on comparing embodied carbon and material efficiency in buildings. The first case study analyzes stadia. The study illustrates the importance of an appropriate functional unit. Indeed, when the material quantities are normalized by floor area, stadia have a lower impact compared to other building types in the survey. However, for a more comprehensive view on the GWP of stadia, the material quantities can be divided by number of seats as a functional unit. The Beijing Olympic Stadium or “Bird’s Nest” shows the highest amount of materials and impact, up to ten times higher than the London Olympic Stadium. Results therefore debate whether the floor area is the best choice as a functional unit. Another option for comparing the embodied carbon across building types is the number of total occupants, focusing on the service of the building rather than the geometry.

71

The second case study is the comparison of two historic bridges: the Inca grass bridge “Keshwa Chaka” and the Roman stone arch “Pont Saint Martin”. The trade-offs between temporary and permanent structures raise the question of what the total lifespan of buildings should be. Lessons learned from these historic structures are the following: using local materials and manufacturing and using force paths to shape structures. These strategies are followed in both the temporary and permanent options. The last case study consists of tall buildings. The material quantities range widely from 250 to 2500 kg/m2 as does the embodied carbon from 130 to 450 kgCO2e/m2. The structural system revealed itself crucial to lower the impact on the environment.

Taken as a whole, the three contributions (1), (2) and (3) offer a new understanding of material efficiency and the life cycle impact of building structures.

The critical review of the current challenges in part (1) has helped to develop a transparent, interactive, growing database in part (2). This database is crucial in the comparison of existing buildings in part (3). The combination of all results provides confidence in the numbers for material quantities and GWP of building structures.

7.2. Summary of contributions This thesis answers the following key question: “What is the embodied carbon for different structures?” The answer to this question is important for multiple reasons. First, unlike operational carbon, embodied carbon is permanent. Lowering the latter therefore leads to immediate savings in carbon emissions. Moreover, the percentage of embodied carbon in the whole building life cycle is becoming significant with increasing operational energy efficiency and shortening building lifespans. Finally, structural engineers and architects need a transparent way of comparing the life cycle impact of their projects against reference buildings, especially since rating schemes started incorporating embodied carbon in their credits. The major contribution of this thesis is to pave the way to a more unified method for collecting material quantities, defining accurate ECC ranges and calculating the GWP of building structures. The results discussed in Section 7.1 are a first attempt to estimate the material efficiency and environmental impact of buildings in a transparent way. An understanding of the emissions of buildings will become as intuitive as the CO2 emissions of cars. For comparison, driving from Philadelphia to Boston (480km) would generate 104 kg of carbon, whereas the construction of the One World Trade Center or “Freedom Tower” generated 100,000,000 kg of carbon. First, the challenges and opportunities encountered in existing literature and tools are analyzed. Three aspects are studied: the two key variables, SMQ and ECC, and an equation calculating the GWP. Separating the current practice in three topics develops a transparent approach. With a critical review of current literature and tools, this thesis creates a basis for a more unified and transparent methodology.

72

Second, the synthesis of the opportunities rising from the challenges led to a framework for a new database. Indeed, a new interactive, growing database was created to collect thousands of building projects from leading structural design firms. The database of embodied Quantity outputs or “deQo” incorporates the simple method multiplying two key variables to obtain the GWP. Third, deQo already collected over 200 existing projects from worldwide architecture or structure companies. The survey of these projects contributed to a new understanding of material efficiency and environmental impact of structures. The results for building structures are normalized material weights ranging on average between 500 and 2500 kg/m2 and normalized embodied carbon ranging on average between 250 and 750 kgCO2e/m2. This thesis lays the groundwork that industry and research needs for benchmarking embodied carbon in buildings. All of this work is motivated by the prospect that ultimately, designers will incorporate Global Warming Potential as one of the factors to take into account in their design process.

7.3. Future research Based on the approach developed in this thesis, the following paths are open for exploration:

• Integrate Building Information Modeling (BIM) in carbon estimating tools

While this thesis used BIM such as Revit through the data collection from practitioners, ultimately, there is a need for integrating BIM models directly into the database in order to compute thousands of projects at the time. Plugging in Revit software will allow benchmarking through comparison of thousands of projects across companies.

• Dynamic alternative to the GWP The measure of the GWP, i.e. “carbon dioxide equivalent”, is a static way of measuring the environmental impact of buildings. However, Kendall (2014) demonstrates this static measurement could skew the results, because methane tends to disappear with the years as opposed to carbon dioxide emissions. Converting gases such as methane to the equivalent in carbon dioxide is consequently a static simplification of a dynamic phenomenon. Though this thesis used GWP as the most feasible, currently available assessment method, future research can explore how to integrate complex dynamic parameters such as the instantaneous climate impact (ICI) or the cumulative climate impact (CCI). These factors would better reflect the dynamic aspect of greenhouse gases.

• Expanding to non-structural materials and to operational carbon The thesis is limited to structural materials only and the embodied life cycle stages. The method can be extrapolated to non-structural materials such as cladding and services. Also, combining the embodied and operational carbon will give a complete view of the whole life cycle impact of buildings.

73

• Extrapolating on the urban scale

This thesis looks at the embodied carbon on the material and building scale. However, current research efforts exist to implement embodied carbon at the urban scale. The “Urban Modeling Interface” or “umi” (umi, 2014), developed in the building technology program at MIT, simulates the life cycle impact of neighborhoods. A buildings’ massing model is associated with selected materials from a database. The result is a visualization of the LCA impacts per year (Figure 7.1).

Figure 7.1: Massing model and yearly impact visualization in “umi” after (umi, 2014; Cerezo, 2013)

• Incorporate embodied carbon in design, starting at the initial conceptual scheme stage

Architecture answers a combination of many different boundary conditions. Embodied carbon cannot be analyzed in isolation from other design factors. This thesis offers an objective analysis of the GWP of buildings, the baselines for benchmarking and the survey of case studies. Further research can develop a multi-scale synthesis (Figure 7.2) to integrate embodied carbon in architecture, starting at the initial concept scheme. First, a study is necessary on the followed strategies in low impact buildings in the database. By exploring the successful case studies, themes will appear in the design process. Second, an analysis should determine how embodied carbon correlates with financial cost. Balancing material and construction cost with financial advantages due to accreditation will pave the way to a cost efficient strategy for low impact buildings. Third, it is crucial to take into account how strategies for decreasing embodied carbon in buildings interact with other variables in the design process. Indeed, a method that lowers the embodied carbon of a building could in some cases undermine the operational energy, the structural efficiency, the functionality or the spatial experience of the architecture. Further research can explore how life cycle impacts relate to these other parameters. The relationship of embodied carbon to other criteria will identify its importance in the design process.

D

0

T

w

th

p

to

p

b

m

st

d

fa

e

EN

ER

GY

ESIGNING THE

C

0 10 20 30

he relation b

which the ope

he life of the

roduction an

otal lifecycle

erformance

oth present

making desig

tructures and

esigned to

actor. Theref

nergy withou

Figure 15: Emlifec

Y

E URBAN ENER

CARLOS CERE

40 50 60 70

between the

eration comp

e building, th

nd maintena

profile which

of two total

the same v

gn decision

d short lifecy

last a very

ore, it is not

ut defining a

mbodied and opecycle time functio

EMB

RGY LIFECYCL

EZO DAVILA - M

80 90 100 1

two can be

ponent overc

he more sig

nce become

h defines a b

energy pro

value. This

s about the

ycles, mater

long time, o

possible to

specific tem

ration energy ons

TIM

BODIED ENERGY

E

MDesS Thesis (

0 10110

e characteriz

comes the e

nificant the

es. The addit

building or a

files is defin

‘crossing po

e ideal lifes

rials energy

operation en

evaluate the

mporal framew

E

EN

ER

GY

Energy and En

0 20 30 40

zed by the a

embodied co

energy asso

tion of both

a project ene

ned by the p

oint’ has to

span of a p

has a bigge

nergy tends

e relative imp

work.

Figure 16: Life hypothe

vironment)

50 60 70 80

analysis of th

omponent. T

ociated with

functions ge

ergetically. Th

point in time

be conside

project. In t

er impact. In

to be a de

portance of e

cycle energy proetical envelope

51

0 90 100 110

he year in

he later in

materials

enerates a

he relative

e in which

ered when

temporary

buildings

eterminant

embodied

ofiles

TIME

Y

74

Figure 7.2: Further research

The objective analysis in this thesis combined with a future multi-scale synthesis will lead to a new design approach in architecture. The qualitative and quantitative assessment of embodied life cycle impacts of buildings will develop new design criteria. While leaving all design options open to architects and engineers, the expanded knowledge will inform designers and their clients on the impact of their choices. These new guidelines will create opportunities for reducing the embodied carbon in the built environment.

• Making the business case Finally, research on embodied carbon alone is not enough to incorporate change in design. Not only structural engineers and architects should be informed of low-carbon design strategies, but also the government and the public. It is important to make the business case for lowering the embodied carbon of structure, either by informing, rewarding through rating schemes such as LEED and BREEAM or by legislation. “Informing both government and the media of the need for the right legislation, and the long-term cost benefits of this, is probably as important in the delivery of significant change as developing the actual solutions to deliver energy and carbon savings in buildings.” – David Clark (CLARK, 2013)

! !

Des

ign

Cas

e S

tudy

B

rigi

ttin

en

- Project 1 - Project 2 - Project 3 - Project 4 - Project 5 - … - Project 10 - … - Project 100 - … - Growing

DATABASE

1. GWP

Matl Quants

ECC Values

2. B

ench

mar

ks

3. C

ase

Stud

ies

A)

Ob

jec

tiv

e

An

aly

sis

B)

Mu

lti-

sca

le

Sy

nth

esi

s

1.  Patterns in low

embodied carbon design

2. vs. financial cost

3. vs. other design

parameters

1_Research_Objectives

75

Bibliography and references The references and bibliography of this thesis can be found hereunder. Part 1 gives the general bibliography and references in alphabetical order. Part 2, 3 and 4 give the references specific to the case studies. A detailed list of references for the material quantities in the stadia, historic bridges and tall buildings enhances the transparency of the applied method.

Part 1: General bibliography and references Alcorn, A. (1996). “Embodied Energy Coefficients of Building Materials.” Centre for Building Performance Research, Victoria University of Wellington, Wellington. Ali, M. M. and Moon, K. S. (2007) “Structural Developments in Tall Buildings: Current Trends and Future Prospects.” Architectural Science Review, 50 (3): pp. 205–223. Amato, A. and Eaton, K. (1998) “A comparative Life Cycle Assessment of Modern Office Buildings.” The Steel Construction Institute, Berkshire. Arup (2008) “Structural Scheme Design Guide.” Arup, Internal report, London. Athena Sustainable Materials Institute. (2009) “Impact Estimator for Buildings.” Accessed November 15, 2013. http://www.athenasmi.org Baker, William, Skidmore, Owings and Merrill, personal conversation with John Ochsendorf and Catherine De Wolf, Cambridge (MA), October 23, 2013. Blake, M.E. (1947) Ancient Roman Construction in Italy forom the Prehistoric Period to Augustus. Washington: Carnegie Institution, 421p. Braham, W. W. and Hale, J. A. (2013) “1929 Richard Buckminster Fuller. 4D Time Lock.” in: Rethinking Architectural Technology, London: Routledge, pp. 42-46. Build Carbon Neutral. (2007) “Estimate the embodied CO2 of a whole construction project.” Accessed December 1, 2013. http://buildcarbonneutral.org Cerezo, C. (2013) “Designing The Urban Energy Life Cycle. A framework and a tool for embodied energy accounting and conservation.” Masters in Design Studies thesis, advisor: Christoph Reinhart, Harvard University, Cambridge (MA). Chapman, P. F. and Roberts, F. (1983) Metal Resources and Energy. Oxford: Butterworth-Heinemann, 256p. Cho, H.-W., Roh, S.-G., Byun, Y.-M. and Yom, K.-S. (2004) “Structural Quantity Analysis of Tall Buildings.” CTBUH Conference: Tall Buildings in Historical Cities – Culture & Technology for Sustainable Cities, Seoul, October 10–13, pp. 662-668. Clark, D.H. (2013) What color is your building? Measuring and reducing the energy and carbon footprint of buildings. London: RIBA Publishing, 263p.

76

Clune, R. (2013) “Algorithm Selection in Structural Optimization.” PhD thesis, advisor: John Ochsendorf, Massachusetts Institute of Technology, Cambridge (MA). Cole, R. J. and Kernan, P. C. (1996) “Life-Cycle Energy Use in Office Buildings.” Building and Environment, 31 (4), pp. 307-317. Collins, F. (2010) “Inclusion of carbonation during the life cycle of built and recycled concrete: influence on their carbon footprint.” International Journal of Life Cycle Assessment, 15 (6), pp. 549-546. CTBUH (2010) “Tall Buildings, Structural Systems and Materials.” CTBUH Journal: Tall Buildings in Numbers (2), pp. 40-41. Darling, D. (2004) The Universal Book of Mathematics, New York City: Wiley, 90p. De Wolf, C., Iuorio, O. and Ochsendorf, J. (2014) “Structural Material Quantities and Embodied Carbon Coefficients: Challenges and Opportunities.” Sustainable Structures Symposium, Corey Griffin (ed.), Portland: Portland State University, April 17-18, pp. 309-328. De Wolf, C. and Ochsendorf, J. (2014) “Participating in an Embodied Carbon Database. Connecting structural material quantities with environmental impact.” The Structural Engineer, 2 (February), pp. 30-31. deQo (2014) “database for embodied Quantity outputs.” available at embodiedco2.scripts.mit.edu Dias, W. P. and Pooliyadda, S. P. (2004) “Quality based energy contents and carbon coefficients for building materials: A System approach.” Energy 29, pp. 561-580. Dixit M.K., Fernández-Solís J.L., Lavy S. and Culp C.H. (2012) “Need for an embodied energy measurement protocol for buildings: A review paper.” Renewable and Sustainable Energy Reviews, 16 (6), pp. 3730-3743. Dunlap, D.W. (2014) “1 World Trade Center Is a Growing Presence, and a Changed One.” The New York Times, Accessed June 12, 2014. http://cityroom.blogs.nytimes.com/2012/06/12/1-world-trade-center-is-a-growing-presence-and-a-changed-one/ Eaton, K. J. and Amato, A. (1998) “A Comparative Life Cycle Assessment of Steel and Concrete Framed Office Buildings.” Journal of Constructional Steel Research, 46, 1-3. EcoInvent (2013) “EcoInvent.” Swiss Centre for Life Cycle Inventories, Accessed December 13, 2013. http://www.ecoinvent.ch Elnimeiri, M. and Gupta, P. (2009) “Sustainable Structure of Tall and Special Buildings.” CTBUH 2nd Annual Special Edition, Tall Sustainability, Wiley, 17 (5), pp 881-894. Eurofer, (2000) “European Steel Industry and Climate Change.” European Confederation of Iron and Steel Industry. GaBi (2013) “GaBi 4 extension database III: steel module and GaBi 4 extension database XIV: construction materials module.” PE International, Accessed December 1, 2013. www.gabi-software.com Hammond, G. and Jones, C. (2010). “Inventory of Carbon & Energy (ICE), Version 1.6a.” Sustainable Energy Research Team (SERT), Department of Mechanical Engineering, Bath: University of Bath.

77

IISI (2013) “The measure of our sustainability, Report of the world steel industry 2004.” International Iron and Steel Institute (IISI), Accessed September 12. http://www.worldsteel.org/dms/internetDocumentList/bookshop/Sustainability_Report2004/document/2004%20Sustainability%20Report.pdf IPCC (2014), “Climate Change 2014: Mitigation of Climate Change.” International Panel for Climate Change (IPCC), Geneva, Accessed April 15, 2014. www.ipcc.ch. ISSF (2013) “ISSF - Stainless steel LCI spreadsheet.” Institute of Stainless Steel Forum (ISSF), Brussels, Accessed September 12, 2013. http://www.worldstainless.org/Files/issf/non-image-files/PDF/GeneralIntroductiontostainlesssteelLCI1.pdf Iuorio, O., Landolfo, R. and Ochsendorf, J. (2013) “Embodied Carbon of Lightweight Steel Structures.” Proceeding of XXIV Congresso C.T.A.: The Italian steel days, pp. 333-340. Kaethner, S. and Burridge, J. (2012) “Embodied CO2 of structural frames.” The Structural Engineer, May, pp. 33-40. Kendall A. (2014) “Climate change mitigation: Deposing global warming potentials.” Nature Climate Change, 4(5), Nature Publishing Group, a division of Macmillan Publishers Limited, pp. 331-332. Khan F. R. and Rankine J. (1981) Tall building systems and concepts. New York City: American Society of Civil Engineering, 1337p. KPF (2012) “United Tower.” Accessed November 11, 2014. www.kpf.com/project.asp?S=1&ID=246 Lagerblad, B. (2005) “Carbon Dioxide Uptake During Concrete Life Cycle - State of the Art.” Swedish Cement and Concrete Research Institute, CBI, Stockholm. Lynde, F. C. (1890) Descriptive Illustrated Catalogue of the sixty-eight competitive designs for the Great Tower for London. London: The Tower Company, St. Stephen's Chambers. Marmon, J., Naugle, M. and Werner, W. (2014) “Buildings instead of Landfills: recycled plastic waste in concrete structures.” Sustainable Structures Symposium, Corey Griffin (ed.), Portland: Portland State University, April 17-18, pp. 247-260. Moncaster, A.M. and Symons, K.E. (2013) “A method and tool for ‘cradle to grave’ embodied carbon and energy impacts of UK buildings in compliance with the new TC350.” Energy and Buildings 66, Cambridge (UK), pp. 514-523. Moynihan M.C. and Allwood J.M. (2012) “The flow of steel into the construction sector.” in: Resources, Conservation and Recycling 68, Cambridge: University of Cambridge, pp. 88-95. Munro, D. (2006) “Swiss Re’s Building, London.” Arup, London, Accessed November 11, 2012, 8p. www.epab.bme.hu/oktatas/2009-2010-2/v-CA-B-Ms/FreeForm/Examples/SwissRe.pdf National Stadium (2014) “National Stadium (Bird’s Nest).” Arup, Accessed on April 30, 2014. www.arup.com/projects/chinese_national_stadium.aspx

78

New York Times (2014), “Moma to raze ex American folk art museum building.” New York Times, Accessed on April 16, 2014. www.nytimes.com/2013/04/11/arts/design/moma-to-raze-ex-american-folk-art-museum-building.html?pagewanted=all&_r=0 Ochsendorf, J., Norford, L., Brown, D., Durschlag, H., Hsu, S. L. and Love, A. (2011). “Methods, Impacts and Opportunities in the Concrete Building Life Cycle, Research Report R11-01.” Concrete Sustainability Hub, Department of Civil and Environmental Engineering. Cambridge (MA): Massachusetts Institute of Technology

Ochsendorf, J.A. (2004) “Sustainable Structural Design: Lessons from History.” Structural Engineering International, 3, pp. 192-194. Ochsendorf J.A. (1996) “Inca Suspension Bridges.” Department Report 96-8, Department of Civil and Environmental Engineering, Cornell University. PECD (2013) “Project Embodied Carbon Database.” Arup, In-house used tool. Pullen S. (2000) “Estimating the Embodied Energy of Timber Building Products.” Journal of the Institute of Wood Science, 15 No. 3 (87), pp. 147-151. Reap, J., Roman, F., Duncan, S. and Bras, B. (2008) “A survey of unresolved problems in life cycle assessment, Part 2: impact assessment and interpretation.” International Journal of Life Cycle Assessment, 13(4), pp. 290-300. Revit (2014) “Building design and construction software.” Autodesk, Accessed April 1, 2014. www.autodesk.com/products/autodesk-revit-family/overview RICS (2014) “Methodology to calculate embodied carbon in a building’s construction life cycle.” Royal Institution of Chartered Surveyors, data courtesy of Atkins, London. Rizk, A. S. (2010) “Structural Design of Reinforced Concrete Tall Buildings.” CTBUH Journal, (1), pp. 34-41. Sauven, Edward, Buro Happold, personal conversation with John Ochsendorf and Catherine De Wolf, London, November 15, 2012. Schlaich, Jörg, Schlaich Bergermann & Partner, personal conversation with John Ochsendorf and Catherine De Wolf, Cambridge (MA), October 4, 2012. Schumaker, J. (2014) “Embodied Carbon Efficiency Tool.” Thornton Tomasetti, Accessed on April 19, 2014. www.thorntontomasetti.com/blog/post/20-Embodied-Carbon-Efficiency-Tool SEAONC Sustainable Design Committee and SEI Sustainability Committee (2013) “For practicing structural engineers to provide comments to improve MIT Database Inputs Template.” Report by Frances Yang, San Francisco. SimaPro (2013) “The world’s most widely used LCA software.” SimaPro UK Ltd., Accessed November 12, 2013. www.simapro.co.uk Simpson, S., Cousins, F., Ayaz, E. and Yang, F. (2010) “Zero Carbon Isn’t Really Zero: Why Embodied Carbon in Materials Cannot be Ignored.” Design Intelligence. Arup, Accessed December 1,

79

2013. www.slideshare.net/enginayaz/zero-carbon-isnt-really-zero-why-embodied-carbon-in-materials-cant-be-ignored Smith, B.P. and Feldson (2008) “Whole Life Carbon Footprinting”, Simons Developments, The Structural Engineer, 86 (6) pp. 15-16. SOM (2013) “Environmental Analysis Tool™.” Skidmore, Owings & Merrill, Accessed August 30, 2013. www.som.com/publication/environmental-analysis-tool-tm Struble, L. and Godfrey, J. (1999) “How Sustainable is Concrete? International workshop on sustainable development and concrete technology.” Chicago: University of Illinois. Stubbles, J. (2007) “Carbon 'Footprints' in U.S. Steelmaking.” Steel Manufacturers Association (SMA), Accessed August 30, 2013. http://66.39.14.41/archive/20071120.htm Tally (2013) “Revit add-in Tally™ (beta version). Real Time Environmental Impact tool. revised Tally Revit Application.” Kieran Timberlake Research Group, Accessed December 30, 2013. www.kierantimberlake.com/pages/view/95/tally/parent:4 The Concrete Centre (2013) “Embodied carbon dioxide (CO2e) of concretes used in buildings.” mpa The Concrete Centre on behalf of the Sustainable Concrete Forum, London. Umi (2014) “Urban Modeling Interface.” Accessed on May 3, 2014. urbanmodellinginterface.ning.com USGBC (2013) “LEED v4 for Building Design And Construction.” U.S. Green Building Council or USGBC, Washington D.C., 154p. Vares, S. and Häkkinen, T. (1998) “Environmental burdens of concrete and concrete products.” Technical Research Centre of Finland, VTT Building Technology, Espoo, Accessed July 20, 2013. http://www.tekna.no/ikbViewer/Content/739021/doc-21-10.pdf Webster M. D., Meryman H., Slivers A., Rodriguez-Nikl T., Lemay L. and Simonen K. (2012) “Strucutre and Carbon – How Materials Affect the Climate.” SEI Sustainability Committee, Carbon Working Group, American Society of Civil Engineers or ASCE, Reston. Weight D.H. (2011) “Embodied through-life carbon dioxide equivalent assessment for timber products.” Institution of Civil Engineers Energy 164, Issue EN4, pp. 167–182. Wise, C., Pawlyn, M. and Braungart, M. (2013) “Living in a materials world.” Eco-engineering, Nature, 494, pp. 172-175. Wolfgang, Werner and Dugum, Hussam, Thornton Tomasetti, personal conversation with John Ochsendorf and Catherine De Wolf, Cambridge (MA), July 29, 2013. WRAP (2014) “Embodied Carbon Database.” UK Green Building Council or UKGBC, Accessed April 14, 2014. http://ecdb.wrap.org.uk Yang, F. (2014) “Benchmarking Embodied Impact Performance of Structure.” Sustainable Structures Symposium, Corey Griffin (ed.), Portland: Portland State University, April 17-18, pp. 131-146. Yang, Frances, Arup, personal conversation with Catherine De Wolf, San Francisco, July 11, 2013.

80

Yang, F., Shook, D., Williams, P. and De Wolf, C. (2013) “Embodied carbon: a meeting of the minds and collaboration opportunities.” Arup, SOM, Webcor, MIT, Conference, July 1, San Francisco. Young, S. B., Shannon T. and Russell A. (2002) What LCA can tell us about the cement industry. Five winds international; World business council for sustainable development, 117p.

Part 2: Stadia references This section gives the references for the stadia. The material quantities were cross-referenced with the following sources. They are classified per stadium. Millenium Stadium, Cardiff

[1] Design Build Network (2014) “Millennium Stadium, Wales.” Accessed April 2, 2014. www.designbuild-network.com/projects/millennium-stadium-wales/

[2] Millennium Stadium (2014) “Facts and Figures.” Accessed April 2, 2014. www.millenniumstadium.com/information/facts_and_figures.php.

Allianz Arena, Munich

[3] Allianz Arena (2014) “Nuts and Bolts.” Accessed April 15, 2014. www.allianz-arena.de/en/fakten/detaillierte-zahlen/

Joao Havelange Olympic Stadium

[4] Design Build Network (2014) “João Havelange Olympic Stadium, Brazil.” Accessed April 2, 2014. www.designbuild-network.com/projects/joao-havelange/.

London Olympic Stadium

[5] Buro Happold (2013) “London 2012 Olympic Stadium.” Accessed April 20, 2013. www.burohappold.com/projects/project/london-2012-olympic-stadium-132/.

[6] Detail Das Architecturportal (2012) “London 2012 – Olympic Stadium.” www.detail-online.com/architecture/news/london-2012-olympic-stadium-019389.html, Detail Magazine 7+8/2012, 2012.

[7] National Geographic Channel (2012) “London’s Olympic Stadium.” Accessed August 20, 2012. natgeotv.com/uk/londons-olympic-stadium.

[8] Sauven, Edward, engineer at Buro Happold, personal conversation with John Ochsendorf and Catherine De Wolf, London, November 13, 2012.

Wembley Stadium

[9] Design Build Network (2014) “Wembley Stadium, London, United Kingdom.” Accessed April 2, 2014. www.designbuild-network.com/projects/wembley/.

[10] Wembley (2014) “Stats and Facts.” Accessed April 2, 2014. www.wembleystadium.com/Press/Presspack/Stats-and-Facts.

Aviva Stadium

[11] Design Build Network (2014) “Aviva Stadium, Ireland.” Accessed April 2, 2014. www.designbuild-network.com/projects/aviva-stadium/.

81

[12] Kilsaran Build “Case Study: Architectural Concrete at the Aviva Stadium.” Report. [13] McGuirk, Laura, engineer at Kilsaran Build, in personal conversation with Julia Hogroian and

Catherine De Wolf, March 18, 2014. Australia Sydney Stadium

[14] Cullen J., Carruth M.A., Moynihan M., Allwood J.M. and Epstein D. (2011) “Learning legacy Lessons learned from the London 2012 Games construction project.” Cambridge University, Olympic Delivery Authority, 031.

Emirates Stadium

[15] Association for Project Management (2014) “Emirates Stadium: More Than a World Class Stadium.” Arcadis, AYHplc, Accessed April 2, 2014. www.apm.org.uk/sites/default/files/Arcadis_Presentation_(Resized).pdf.

[16] Design Build Network (2014) “Emirates Stadium, United Kingdom.” Accessed April 2, 2014. www.designbuild-network.com/projects/ashburton /.

Beijing National Stadium

[17] Arup (2012) “Chinese National Stadium.” Accessed August 20, 2012. www.arup.com/Home/Projects/Chinese_National_Stadium.aspx

[18] Chinese Architect (2012) “Beijing National Stadium.” Accessed August 20, 2012. www.chinesearchitecture.info/OLYMPICS/OL-001.htm.

[19] Design Build Network (2014) “Beijing National Stadium, ‘The Bird’s Nest’, China.” Accessed April 2, 2014. www.designbuild-network.com/projects/national_stadium/.

Jaber Al-Ahmad Stadium

[20] Kharafi Group (2012) “Al-Sheikh Jaber Al-Ahmad International Sports Stadium, Kuwait.” Accessed August 20, 2012. www.makharafi.net/buildings.html.

[21] Kuwait City (2012) “Jaber Al-Ahmad International Stadium (60,000).” Accessed August 20, 2012. skyscrapercity.com/showthread.php?t=283227.

[22] Stadiony (2012) “Jaber Al-Ahmad International Stadium.” Accessed August 20, 2012. stadiony.net/projekty/kuw/jaber_al_ahmad_stadium.

Part 3: Historic bridges references This section gives the references for the historic bridges. The material quantities were cross-referenced with the following sources. They are classified per bridge. [1] Clune, R. (2013) “Algorithm Selection in Structural Optimization.” PhD thesis, Massachusetts

Institute of Technology, Cambridge (MA). [2] Ochsendorf, J.A. (2004) “Sustainable Structural Design: Lessons from History.” Structural

Engineering International, 3, pp. 192-194.

82

Pont Saint Martin [3] Blake, M.E. (1947) “Ancient Roman Construction in Italy forom the Prehistoric Period to

Augustus”, Washington. [4] Frunzio G. and Monaco M. (1998) “An approach to the structural model for masonry arch

bridges: Pont Saint Martin as a case study.” Centro Interdipartimentale di Ingegneria per I Beni Culturali, Università Federico II, Napoli, paper in book (Sinopoli [9])

[5] Gneiss rockpack, Geology superstore, Accessed May 10, 2013. www.geologysuperstore.com/product/hornblende-gneiss-rock-pack-1772

[6] O’Connor, C. (1993) Roman Bridges. Cambridge University Press, 235p. [7] O’Connor, C. (1993b) Development in Roman stone arch bridges. University of Queensland, Australia. [8] “Pont Saint Martin”, Locali dautore, Accessed May 2, 2013.

http://www.localidautore.com/paesi/pont-saint-martin-1678.aspx [9] Sinopoli, A. (1998) “Arch Bridges: History, Analysis, Assessment, Maintenance and Repair”, 2nd

International Conference, Venice, 6-9 October 1998, pp. 231-238. Keshwa-Chaka bridge [10] Foer J. (2013) “The Last Incan Grass Bridge. The sixth hidden wonder of South America.” Slate,

Accessed May 13, 2013. www.slate.com/articles/life/world_of_wonders/2011/02/the_last_incan_grass_bridge.html

[11] Ochsendorf J.A. (1996) “Inca Suspension Bridges.” Department Report 96-8, Department of Civil and Environmental Engineering, Cornell University.

[12] “The Last Handwoven bridge.” Accessed May 2, 2013. www.atlasobscura.com/places/last-handwoven-bridge

Part 4: Tall building references Alia, M. M. and Moona, K. S. (2007) “Structural Developments in Tall Buildings: Current Trends and Future Prospects.” Architectural Science Review, 50(3), pp. 205-223, Accessed October 22. www.tandfonline.com/doi/pdf/10.3763/asre.2007.5027 30 St Mary Axe

[1] Aldersey-Williams, H. (2011) “Towards biomimetic architecture.” Accessed October 15, 2012. www.nature.com/nmat/journal/v3/n5/full/nmat1119.html

[2] Arup (2012) “30 St Mary Axe.” Accessed October 15, 2012. www.arup.com/Home/Projects/30_St_Mary_Axe/

[3] Munro, D. (2006) “Swiss Re’s Building, London.” Arup, London, 8p., Accessed November 11, 2012. www.epab.bme.hu/oktatas/2009-2010-2/v-CA-B-Ms/FreeForm/Examples/SwissRe.pdf

[4] Freiberger, M. (2007) “Perfect buildings: the maths of modern architecture.” Plus Magazine, March, Accessed November 1, 2012. www.rigb.org/assets/uploads/docs/PLUS_MAGAZINE_Perfect%20buildings.pdf

The United Tower

[5] KPF (2012) “United Tower.” Accessed November 11, 2012. www.kpf.com/project.asp?S=1&ID=246

[6] El Mostafa, Hakam, KIPCO Group, U.R.E., Personal conversation on October 20, 2012. [7] United Tower (2012) “We love the clouds.” 48p., Accessed November 2, 2012.

http://www.unitedtowers.net/AxCMSwebLive/upload/utbrochure1_1172.pdf [8] Unitedtowers.net (2012), Accessed October 20, 2012. www.unitedtowers.net/

83

[9] Perime (2012) “United Tower”, Sharq, Kuwait, Accessed November 2, 2012. www.perime.com/projects.cfm

[10] UN Data (2012) Accessed November 2, 2012. data.un.org/CountryProfile.aspx?crName=KUWAIT#Environment

The Shard

[11] Ferguson, H. (2012) "Building the Shard." Ingenia, September, 24-30. [12] Kennett, S. (2013) "The Shard: Foot of the Mountain." Building.co.uk, Accessed September 11,

2013. www.building.co.uk/the-shard-foot-of-the-mountain/3162661.article [13] Moazami, K. (2008) and Ahmad Rahimian. "Shard at London Bridge Tower."

STRUCTUREmag, June, Accessed September 10, 2013. www.structuremag.org/article.aspx?articleID=694

[14] Moazami, K., Parker, J., and Giannini, R. (2010) “Unique Hybrids at London Tallest.” CTBUH 8th World Congress: Steel-Concrete-Steel, Dubai.

[15] Thompson, P. (2010) "Shard's Giant Core Shoots up." Steel Construction News 7187: 1-3. [16] WSP (2013) “The Shard.” London: WSP.

Al Hamra Firdous Tower

[17] Agarwal, R., Noor, A., Lindsay, H., Neville, M., Mazeika, A., and Sarkisian, M. (2007) “Sculpted High-rise, The Al Hamra Tower.” SOM, Council on Tall Buildings and Urban Habitats: Structural Engineers World Congress, Bangalore, India.

[18] Gonchar, J. (2012) "Al Hamra Firdous Tower." Architectural Record, May. [19] Haney, G. (2009) "Al Hamra Firdous Tower." Architectural Design, Mar.-Apr. 2009, pp. 38-41. [20] Sarkisian, M., Aybars, A., Neville, M., and Mazeika A. (2012) "Sculpting a Skyscarper."

Civil Engineering—ASCE 82.9, pp. 52-61. Business Source Complete. Willis Tower

[21] SkycraperPage (2014) “Willis Tower.” SkyscraperPage.com, Accessed May 1, 2014. http://skyscraperpage.com/cities/?buildingID=5

World Financial Center

[22] Vertigo (1999) “The strange new world of the contemporary city.” Rowan Moore (ed.), Laurence King, Glasgow.

[23] Construction Technology (2012) “Tall Buildings-Shanghai World Finance Center, Ahmedabad.” Cept University.

[24] Arcelor Mittal (2012) “Shanghai World Financial Center.” Accessed October 10, 2012. www.constructalia.com/english/case_studies/china/shanghai_world_financial_center

Taipei 101

[25] Skyscrapercenter (2014) “Taipei 101.” Skycsrapercenter.com, Accessed May 1, 2014. www.skyscrapercenter.com/taipei/taipei-101/117/

[26] Concretepumping (2014) “World Record Pumping Height Achieved on Tallest Building 2004”, Accessed May 1. http://concretepumping.com/dictionary/index.php/WORLD_RECORD_PUMPING_HEIGHT_ACHIEVED_ON_TALLEST_BUILDING_2004

84

One World Trade Center [27] Tweeten, L. and Maltby, E. (2014) “Inside the tower, The top of America Exclusive: the

inside story of building one world trade center.” Time Magazine, March 17. Shanghai Tower

[28] CTBUH YPC (2013) Presentation – Shanghai Tower. Presentation by the Young Professionals Committee of the Council on Tall Buildings on May 22, 2013 at the Chicago Architecture Foundation, Chicago, IL, USA. (www.youtube.com/watch?v=riRVkoUEBP4)

[29] Jacobson, C. (2012) “A New Twist on Supertall: An American firm approaches the design of its 121-story, mixed use tower now rising in Shanghai as a vertical collection of neighborhoods.” Architectural Record, May, Accessed October 1, 2012. archrecord.construction.com/projects/portfolio/2012/05/Shanghai-Tower.asp

[30] Xia, J., Poon, D. and Mass, D.C. (2010) “Case Study: Shanghai Tower.” CTBUH Technical Paper.

[31] Gensler Design Update (2010) “Shanghai Tower.” Gensler Publications, Accessed October 1, 2012. gensler.com.

[32] Gu, J.P. (2012) “Shanghai Tower: Re-Thinking the Vertical City.” CTBUH Technical Paper. [33] Zhu, Y., Poon, D., Zuo, S. and Fu, G. (2012) “Structural Design Challenges of Shanghai

Tower.” CTBUH Technical Paper.

Burj Khalifa [34] Baker, W.F., and Pawlikowski, J.J. (2012) “Higher and Higher: The Evolution of the

Buttressed Core.” [35] Baker, W.F. (2012) “The World’s Tallest Building - The Burj Khalifa” [36] Baker, W.F. (2012) “The Burj Khalifa Triumphs - Engineering an Idea: The Realization of the

Burj Khalifa” [37] Abdelrazaq, A. “Design and Construction Planning of the Burj Khalifa.” Dubai, UAE [38] Abdelrazaq, A. “Validating the Structural Behavior and Response of Burj Khalifa: Synopsis of

the Full Scale Structural Health Monitoring Programs.” [39] Aldred, J. “Burj Khalifa - A new high for high performance concrete.” [40] Baker, W.F., Skidmore, Owings & Merrill LLP, “Tall Buildings and the Burj Khalifa” Interview,

Accessed October 20, 2012. http://chicagowindowexpert.com/2010/04/22/bill-baker-of-skidmore-owings-merrill-llp-tall-buildings-and-the-burj-khalifa/

[41] SOM (2012) “Burj Khalifa - ARCHITECT OFFICIAL PROJECT PAGE”, Accessed October 20, 2012. www.som.com/project/burj-khalifa-formerly-burj-dubai

85

Appendices

Appendix A: Nomenclature

A.1. List of acronyms BIM Building Information Models BOM Bill of Materials BOQ Bill of Quantities BREEAM Building Research Establishment Environmental Assessment Methodology,

available at: http://www.breeam.org CCI Cumulative Climate Impact CIF Carbon Intensity Factor, synonym of ECC CO2e Carbon dioxide equivalent deQo database of embodied Quantity outputs,

available at: embodiedco2.scripts.mit.edu ECC Embodied Carbon Coefficient EEC Embodied Energy Coefficient GHG Greenhouse Gases GWP Global Warming Potential html HyperText Markup Language ICE Inventory of Carbon & Energy (Hammond and Jones, 2010) ICI Instantaneous Climate Impact IPCC Intergovernmental Panel on Climate Change KPF Kohn Pedersen Fox Associates,

available at: www.kpf.com LCA Life Cycle Assessment LCI Life Cycle Inventory LEED Leadership in Energy and Environmental Design,

available at: http://www.usgbc.org/leed MIT Massachusetts Institute of Technology MySQL Open-source relational database management system,

named after co-founder Michael Widenius's daughter, My and SQL stands for Structured Query Language

PECD Project Embodied Carbon Database, developed by Arup and Climate Earth (Yang, 2014)

php Hypertext PreProcessor (backronym), a server-side scripting language RICS Royal Institution of Chartered Surveyors,

available at: www.rics.org SOM Skidmore, Owings & Merrill LLP,

available at: www.som.com SQM Structural Material Quantities umi urban modeling interface WRAP Waste Reduction Action Program, developed the wrap embodied carbon database,

available at: ecdb.wrap.org.uk

86

A.2. Lexicon Carbon Dioxide Equivalent (CO2e)

The “equivalent” in carbon dioxide of all emitted greenhouse gases (GHGs). Carbon Dioxide (Only the CO2) ≠ Carbon Dioxide Equivalent (CO2e), which converts all GHGs to the “equivalent” in CO2. For example, 1 lbs of methane (CH4) is equivalent to 21 lbs of carbon dioxide (CO2).

Cradle to Gate The life cycle stages from the material extraction till the finished product.

Cradle to Site The life cycle stages from the material extraction till the product arrives on site.

Cradle to Grave All the life cycle stages: material extraction, processing, product manufacturing, transport to site, construction, use, demolition and either reuse/recycling/landfill. All the life cycle stages of a building are illustrated in Figure A.1.

Operational energy Energy used for the heating, cooling, hot water,

ventilation and lighting of a building. Embodied energy

Energy used for the material extraction, material processing and product manufacturing, the transport of products to the construction site, the construction of a building and the demolition at the end of life. It is all the life cycle energy excluding the operational energy.

PROBLEM WHY?

CRADLE GATE SITE GRAVE

Based on [CLARK, What Colour is your Building?]

PROBLEM LITERATURE RESULTS NEXT

4

Why embodied?

Figure A.1: Life cycle stages of a building, based on (CLARK, 2013)

87

Whole life cycle energy Whole life cycle energy = operational energy + embodied energy (definition a) Sometimes, the transport of the building occupants (≠ transport of product), for example commuting of employees for an office building, and their consumption are also considered. The definition of whole life cycle energy then also contains these aspects. Whole life cycle energy = operational energy + embodied energy + transportation energy + consumption energy (definition b) In this thesis, when talking about the whole life cycle energy, definition (a) is implied.

Embodied carbon Embodied Energy ≠ Embodied Carbon Embodied carbon corresponds to the emitted greenhouse gases (GHGs) to produce the embodied energy. ‘Carbon’ is not the same as ‘energy’, as the GHG emissions depend on the fuel used and the carbon emitted or absorbed by the materials processed too. For example, processing cement can emit CO2, where timber sequestrates CO2.

Functional unit A specified metric used to normalize the carbon footprint of buildings in order to compare ‘apples to apples’. For example, the useable floor area can be the functional unit, so that the total embodied carbon is divided by the total useable floor area. The GWP is consequently measured in kgCO2e/m2. The number of occupants is another example of a functional unit. The GWP is then measured in kgCO2e/occupant.

Global Warming Potential Measure of embodied carbon of a building project, expressed in kg of carbon dioxide equivalent (kgCO2e) per functional unit (usually m2).

88

Appendix B: Tables

B.1. Ten first projects in deQo

Program Location Structural System Area (m2) Office, high rise London, UK Steel, diagrid 47,850 Government New Port Beach, USA Composite 8,705 Healthcare San Francisco, USA Concrete 46,755 Healthcare San Francisco, USA Concrete 4,849 Healthcare + Office LA, USA Concrete 6,821 Office CA & Mexico Concrete 7,870 Office, high rise USA Composite 8,367 Office, mid rise Seattle, USA Concrete 89,354 Central Utility Plant LA, USA Composite 6,538 Car Parking USA Concrete 181,951

Table B.1: Ten projects entered through deQo, name made anonymous Table B.2 gives the material quantities divided into sub- and superstructure and divided by material: concrete (C), hotrolled steel (S) and reinforcement bar (R). Substructure Superstructure Total Program kg/m2 kg/m2 kg/m2

C S R C S R C S R Office, high rise 195 175 4 195 175 4 Government 262 0 9 317 112 10 579 112 19 Healthcare 507 0 9 488 66 11 995 66 21 Healthcare 827 0 70 404 153 13 1231 153 83 Healthcare + Office 627 0 24 305 75 9 932 75 33 Office 236 0 38 112 71 13 348 71 51 Office, high rise 287 0 10 111 25 4 398 25 13 Office, mid rise 776 0 14 215 45 5 991 45 19 Central Utility Plant 979 0 26 645 179 21 1624 179 47 Car Parking 210 0 7 518 16 24 728 16 31

Table B.2: Material quantities of the 10 projects in Table B.1.1. Source ECC GWPper material GWPtotal

Program ECC kgCO2e/kg kgCO2e/m2 kgCO2e/m2 C S R C S R Office, high rise LCA 0.16 0.89 0.89 31 156 4 190 Government Arup 0.14 1.41 0.57 83 157 11 251 Healthcare Arup 0.10 0.56 0.57 101 37 12 150 Healthcare Arup 0.15 1.40 0.57 185 215 47 447 Healthcare + Office Arup 0.15 1.42 0.57 139 106 19 263 Office Arup 0.15 1.38 0.56 52 98 29 179 Office, high rise Arup 0.11 1.49 0.57 42 37 8 87 Office, mid rise Arup 0.12 1.48 0.57 124 66 11 201 Central Utility Plant Arup 0.08 1.04 0.43 130 187 20 337 Car Parking Arup 0.15 1.88 0.57 107 30 17 155

Table B.3: ECCs used for the materials and GWP of the 10 projects in Table B.1.1.

89

B.2. Analysis of stadia unit Millenium Allianz Joao H. London Wembley location Cardiff Munich Rio de J. London London floor area m2 40,000 37,600 34,250 61,575 79,578 number of seats 74,500 68,000 45,000 80,000 90,000 area/seat m2/seat 0.54 0.55 0.76 0.77 0.88 total steel kg 12,000,000 20,000,000 2,700,000 10,700,000 23,000,000 total rebar kg 4,000,000 total concrete kg 40,000,000 256,800,000 96,000,000 102,000,000 216,000,000 steel per area kg/m2 300 532 79 174 289 rebar per area kg/m3 100 0 0 0 0 concrete per area kg/m2 1,000 6,830 2,803 1,657 2,714 steel per seat kg/seat 161 294 60 134 256 rebar per seat kg/seat 54 0 0 0 0 concrete per seat kg/seat 537 3,776 2,133 1,275 2,400 ECC steel kgCO2e/kg 0.88 0.88 0.88 0.89 0.88 ECC rebar kgCO2e/kg 1.71 1.71 1.71 1.71 ECC concrete kgCO2e/kg 0.13 0.13 0.13 0.10 0.13 EC steel per area kgCO2e/m2 264 468 69 155 254 EC rebar per area kgCO2e/m2 171 0 0 0 0 EC concrete per area kgCO2e/m2 130 888 364 159 353 EC steel per seat kgCO2e/seat 142 259 53 119 225 EC rebar per seat kgCO2e/seat 92 0 0 0 0 EC concrete per seat kgCO2e/seat 70 491 277 122 312

Table B.4: Data on stadia, five first stadia unit Aviva Australia Emirates Beijing Jaber Al A. location Dublin Sydney London Beijing Kuwait floor area m2 66,460 160,000 122,000 254,600 400,000 number of seats 51,700 110,000 60,355 91,000 65,000 area/seat m2/seat 1.29 1.45 2.02 2.80 6.15 total steel kg 5,000,000 12,000,000 3,000,000 110,000,000 1,965,000 total rebar kg 10,000,000 10,000,000 total concrete kg 84,000,000 210,000,000 144,000,000 1,300,000,000 288,000,000 steel per area kg/m2 75 75 25 432 5 rebar per area kg/m3 0 63 82 0 0 concrete per area kg/m2 1,264 1,313 1,180 5,106 720 steel per seat kg/seat 97 109 50 1,209 30 rebar per seat kg/seat 0 91 166 0 0 concrete per seat kg/seat 1,625 1,909 2,386 14,286 4,431 ECC steel kgCO2e/kg 0.88 0.89 0.88 0.88 0.71 ECC rebar kgCO2e/kg 1.71 1.71 1.71 1.71 1.71 ECC concrete kgCO2e/kg 0.11 0.15 0.13 0.17 0.19 EC steel per area kgCO2e/m2 66 67 22 380 3 EC rebar per area kgCO2e/m2 0 107 140 0 0 EC concrete per area kgCO2e/m2 139 197 153 868 137 EC steel per seat kgCO2e/seat 85 97 44 1,064 21 EC rebar per seat kgCO2e/seat 0 155 283 0 0 EC concrete per seat kgCO2e/seat 179 286 310 2,429 842

Table B.5: Data on stadia, five last stadia

90

B.3. Analysis of historic bridges

B.3.a. Calculations Pont Saint Martin

Legend Span drawing 31.4 m other Life time literature 2000 years

Material ECC Density

Gneiss 0.017 2610 http://geology.about.com/cs/rock_types/a/aarockspecgrav.htm Limest. 0.017 2500 http://geology.about.com/cs/rock_types/a/aarockspecgrav.htm

Soil 0.023 1600 http://soils.usda.gov/sqi/assessment/files/bulk_density_sq_physical_indicator_sheet.pdf

Parts

Volume m3

Material

Density kg/m3

ECC kgCO2e/kg

Weight kg

Weight kg/m

GWP kgCO2e/m

GWP kgCO2e/m/y

Vault 270 Gneiss 2610 0.017 704700 22443 382 0.19 Spandrel 390 Gneiss 2610 0.017 1017900 32417 551 0.28

Pavi. 42 Limestone 2500 0.017 105000 3344 57 0.03

Parapets 26 Gneiss 2610 0.017 67860 2161 37 0.02 Filling 780 Soil 1600 0.023 1248000 39745 914 0.46

Total 1940 0.97

Table B.6: Calculations Pont Saint Martin The Pont Saint Martin has a span of approximately 35.6 m with a rise of one-third of the span. The width of the bridge is 5.8m (4.6m between parapets) and the height is 12m (13.6m when counting the parapet). The main material used in the bridge is blocks of local gneiss (O’Connor, 1993). Based on the drawings and descriptions in (Frunzio and Monaco, 1998; Sinopoli, 1998]), the volumes of different parts of the Roman arch could be determined. The detailed calculations are shown in attachment. The volume of the vault, spandrel walls, pavimentum and parapets, which are made of stone, is 686 m3. The densities of the stones (Alden) are set at 2610 and 2500 kg/m3. This part weights 1790 tonnes. The interior part of the bridge, which is filled by 780 m3 of soil, with a density of 1600 kg/m3 (Soil, 2008), weights 1248 tonnes. The ECC factors given by (Hammond and Jones, 2010) for gneiss and limestone are 0.017 kgCO2/kg and 0.023 kgCO2/kg for soil.

91

B.3.b. Calculations Keshwa-Chaka

Legend Span drawing 30 m other Life time literature 600 years

Material ECC Density

Stone 0.056 2500 Puna 0.01 395 Sticks 0.01 500

Parts

Volume m3

Material

Density kg/m3

ECC kgCO2e/kg

Weight kg

Weight kg/m

GWP kgCO2e/m

GWP kgCO2e/m/y

Abutments / Stone 2500 0.056 9070 302 16.93 0.03 Cables 1.98 Puna 395 0.01

Ropes 0.6 Puna 395 0.01 3630 121 1.21 1.21 Deck 0.45 Sticks 500 0.01

Total 18.14 1.24

Table B.7: Calculations Keshwa-Chaka To calculate the embodied carbon of the 30 meter long Keshwa-chaka, a description of the construction components is necessary. The strands of grass used to braid are 45 centimeters (18 inch) long. Over 15240 meters (50000 feet) of grass cord are produced every year. The grass comes from the local mountainside and is manufactured by twisting it by hand. The twenty-four cords are one centimeter (3/8 inch) thick and almost 60 meters (200 feet) long. These cords are twisted into ropes with a diameter of five centimeters (two inches). Next, these ropes are themselves braided in six 45 meters long cables. These are the main structural elements of the grass bridge itself. This first part takes one day and the help of all villagers to accomplish (Ochsendorf, 1996). The second day of the festival, only adult male villagers work on the cutting of the old bridge and the installation of the cables by minimizing the sag in the middle and wrapping and tying them around the stone abutments. The stone abutments are masonry of 9070 kilogram (20000 pounds). The last day, the Chaka-Camayoc or bridge-keeper ties the floor cables together and adds the vertical cords between the floor and the handrails. He adds sticks and matted reeds to create a deck on the floor (Ochsendorf, 1996). The weight of one cable and decking is 907 kg (4100 pounds – 14*150 pounds = 2000 pounds). The total weight of the bridge is 8000 pounds, which is 3630 kg. The stone abutments are 9070 kg (Ochsendorf, 1996). The grass used for the Keshwa-chaka is cut of the local mountainsides and is typically the Puna grass found in the Andes. The ECC is similar to that of straw, 0.01 kgCO2/kg (Hammond and Jones, 2010).

92

B.4. Analysis of tall buildings unit Mary Axe United T. The Shard Al Hamra Willis Location London Kuwait London Kuwait Chicago Date 2004 2011 2012 2011 1974 Structural Engineer Arup WSP WSP SOM SOM

Structural System Steel

Diagrid Reinforced

Concrete Hybrid Reinforced

Concrete Steel Floor area m2 74,300 98,000 126,712 195,000 416,000 Height m 180 240 306 413 443 Number of floors 41 59 87 80 111 Total steel kg 8,358,000 752,781 12,671,200 50,000 76,000 Total rebar kg 187,035 2,936,146 9,800,000 Total concrete kg 9,351,736 146,807,304 77,040,896 490,000,000 132,115,200 Steel per area kg/m2 112 8 100 0 0 Rebar per area kg/m2 3 30 50 0 Concrete per area kg/m2 126 1,498 608 2,513 318 ECC steel kgCO2e/kg 0.89 0.71 0.89 0.71 0.89 ECC rebar kgCO2e/kg 1.70 1.70 1.70 1.70 1.70 ECC concrete kgCO2e/kg 0.16 0.15 0.16 0.15 0.16 EC steel/area kgCO2e/m2 100 5 89 0 0 EC rebar/area kgCO2e/m2 4 51 0 85 0 EC concrete/area kgCO2e/m2 20 225 97 377 51

Table B.8: Data on stadia, five last stadia unit WFC Taipei 101 1WTC Shanghai Burj Khalifa

Location Shangai TaiPei New York

City Shanghai Dubai Date 2008 2004 2014 2014 2010

Structural Engineer

Leslie E. Robertson Associates

Thronton T.; Evergreen

Engineering WSP Thornton T. SOM

Structural System Trusses &

columns Composite Hybrid Concrete Buttressed

Core Floor area m2 381,600 193,400 325,279 521,000 334,000 Height m 492 508 546 632 828 Number of floors 101 106 82 128 162 Total steel kg 13,861,826 48,000,000 28,390,000 Total rebar kg 4,748,813 Total concrete kg 237,440,637 489,600,000 381,600,000 739,820,000 574,480,000 Steel per area kg/m2 36 148 85 Rebar per area kg/m2 12 0 Concrete per area kg/m2 622 2,532 1,173 1,420 1,720 ECC steel kgCO2e/kg 0.88 0.88 0.88 0.88 0.71 ECC rebar kgCO2e/kg 1.70 1.70 1.70 1.70 1.70 ECC concrete kgCO2e/kg 0.17 0.17 0.13 0.17 0.15 EC steel/area kgCO2e/m2 32 0 130 0 60 EC rebar/area kgCO2e/m2 21 0 0 0 0 EC concrete/area kgCO2e/m2 106 430 153 241 258

Table B.9: Data on stadia, five last stadia

Documents

Material quantities in building structures and their